Regulation of the EU Financial Markets: MiFID II and MiFIR (Oxford EU Financial Regulation) [1 ed.] 2016960865, 9780198767671, 9780191080302

Table of contents :
Title Page
Copyright Page
Contents
Table of Cases
Table of Legislation
List of Abbreviations
Author Biographies
I General Aspects
1. Who’s Afraid of MiFID II? An Introduction
I. Introduction
II. Investment Firms and Investment Services
III. Trading
IV. Supervision and Enforcement
V. The Broader View and the Future of MiFID
VI. Final Remarks
II Investment Firms and Investment Services
2. The Scope of MiFID II
I. Introduction
II. Investment Firms under MiFID II
III. Credit Institutions
IV. Investment Firms under the CRR
V. Structured Deposits
VI. General Exemptions to MiFID II Applicability
VII. Trading on Own Account
VIII. Dealings in Emission Allowances
IX. Financial Instruments
X. Insurance Companies and Insurance Intermediaries
XI. Managers of Collective Investment Undertakings
XII. Data-Reporting Service Providers
XIII. Market Operators
XIV. Conclusion
3. Governance of Investment Firms under MiFID II
I. Introduction
II. Governance-related Regulation of Investment Firms between Prudential and Conduct-of-Business Regulation
III. The Technical Framework under MiFID II
IV. Conclusions
4. The Overarching Duty to Act in the Best Interest of the Client in MiFID II
I. Introduction
II. Antecedents of the Investment Firm’s Duty to Act in the Best Interest of the Client
III. The Duty to Act in the Best Interest of the Client in MiFID II
IV. The Duty’s Functions and Contents
V. The Articulation of the Duty with Respect to Individual Services and Activities
VI. A Look Across the Atlantic
VII. Assessment
5. Product Governance and Product Intervention under MiFID II/MiFIR
I. Introduction
II. Product Governance
III. Product Intervention
IV. Conclusion
6. Independent Financial Advice
I. Introduction
II. Economic Background
III. MiFID I
IV. MiFID II
V. Conclusions
7. Conflicts of Interest
I. Foundations of the Regime
II. The Regime by Single Stages and Examples
III. Conclusion
8. Inducements
I. Introduction
II. Current Legislation
III. The Dutch Inducement Ban
IV. The UK Inducement Rules
V. MiFID II
VI. Research as an Inducement
VII. Conclusion
9. Agency and Principal Dealing under MiFID I and MiFID II
I. Introduction
II. Four Transaction Types
III. Investor Protection
IV. Acting as Agent or as Principal: Not a Justified Distinction to Serve as a Basis for Determining the Degree of Investor Protection
V. Conclusion
10. MiFID II/MiFIR’s Regime for Third-Country Firms
I. General
II. Scope of MiFID II/MiFIR’s Regime for Third-Country Firms
III. Eligible Counterparties and Per Se Professional Clients
IV. Retail Clients and Opt Up Professional Clients
V. Retail Clients, Professional Clients, and Eligible Counterparties
VI. Initiative Test
VII. Conclusion
III Trading
11. Governance and Organization of Trading Venues: The Role of Financial Market Infrastructure Groups
I. Introduction
II. Trading Facilities: Concept and Regulatory Framework
III. The EU Scenario of Trading Venues
IV. FMI Groups in an MiFID II World
V. Conclusion
12. EU Financial Governance and Transparency Regulation: A Test for the Effectiveness of Post-Crisis Administrative Governance
I. Transparency Regulation and EU Financial Governance: Why It Matters
II. Transparency Regulation and EU Financial Governance
III. MiFIR, Transparency, and Supervisory Administrative Governance
IV. MiFIR, Transparency, and Administrative Regulatory Governance
V. Conclusion
13. SME Growth Markets
I. Introduction
II. Existing Alternative Markets for SMEs in Europe
III. SME Growth Markets under MiFID II
IV. Alternative Disclosure Obligations?
V. What Role for the ‘SME Growth Market’ Label? Some Reflections on Alternative Scenarios
VI. Conclusion
14. Dark Trading under MiFID II
I. Introduction
II. Dark Pools: Motivation, Classification, and Related Literature
III. Dark Pools under MiFID I
IV. Dark Pool Regulation under MiFID II/MiFIR
V. Summary
15. Derivatives: Trading, Clearing, STP, Indirect Clearing, and Portfolio Compression
I. Introduction
II. The Trading Obligation
III. The Clearing Obligation
IV. The Straight-Through-Processing Obligations
V. The Obligations in Respect of Indirect Clearing
VI. The Obligations in Respect of Portfolio Compression
VII. A Few Final Remarks
16. Commodity Derivatives
I. Introduction
II. Features of the Commodity Derivatives Markets and Its Relation to the Crisis
III. Overview of the Most Important EU Provisions on Commodity Derivatives
IV. The Regulatory Perimeter of the MiFID II/MiFIR Package
V. The New Position Limits Regime
VI. A New Set of Intervention Powers for Authorities and Trading Venues
VII. Conclusion
17. Algorithmic Trading and High-Frequency Trading (HFT)
I. Introduction
II. Algorithmic Trading
III. High-Frequency Trading
IV. Conclusion
18. MiFID II and Equity Trading: A US View
I. Basic Building Blocks
II. High-Frequency Trading
III. Dark Pools and Internalization
IV. Conclusions: Reflections on Issues Raised by MiFID II
IV Supervision and Enforcement
19. Public Enforcement of MiFID II
I. Introductory Remarks
II. Structure of the Study
III. Competent Authorities: General Aspects, Powers, and Redress Procedures
IV. Cooperation Arrangements
V. Concluding Remarks and Assessment
20. The Private Law Effect of MiFID I and MiFID II: The Genil Case and Beyond
I. General
II. May Civil Courts Be Less Strict Than MiFID I and MiFID II?
III. May Civil Courts Be Stricter Than MiFID I and MiFID II?
IV. May Contracting Parties Be Less Strict Than MiFID I and MiFID II?
V. May Contracting Parties Be Stricter Than MiFID I and MiFID II?
VI. Influence of MIFID I and MiFID II on the Principle of Relativity
VII. Influence of MiFID I and MiFID II on Proof of Causation
VIII. Influence of MiFID I and MiFID II on a Contractual Limitation or Exclusion of Liability
IX. MiFID I and MiFID II Assessments by the Courts of Their Own Motion in Relation to Private Investors?
X. Conclusion
V The Broader View and the Future of MiFID
21. MiFID II in Relation to Other Investor Protection Regulation: Picking Up the Crumbs of a Piecemeal Approach
I. Introduction
II. MiFID II versus IDD
III. MiFID II and IDD versus UCITS KII and PRIIPs KID
IV. Cross-Selling Practices
V. Conclusion
22. Shadow Banking and the Functioning of Financial Markets
I. Introduction
II. Shadow Banking: Concept and Terminology
III. Non-Regulated Activity According to the FSB Approach
IV. What Financial Activity is not Subject to Banking Supervision?
V. The European Response to Shadow Banking
VI. Conclusion
23. Investment-Based Crowdfunding: Is MiFID II Enough?
I. Introduction
II. Benefits, Risks, and Challenges
III. MiFID I
IV. National Approaches to Crowdfunding
V. MiFID II
VI. Concluding Remarks
Index

Citation preview

REGULATION OF THE EU FINANCIAL MARKETS MiFID II and MiFIR

REGULATION OF THE EU FINANCIAL MARKETS MiFID II and MiFIR

Edited by

DANNY BUSCH GUIDO FERRARINI

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © The editors and several contributors 2016 The moral rights of the author[s] have been asserted First Edition published in 2016 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Crown copyright material is reproduced under Class Licence Number C01P0000148 with the permission of OPSI and the Queen’s Printer for Scotland Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2016960865 ISBN 978–0–19–876767–1 eISBN 978–0–19–108030–2 Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

PREFACE

Investment firms and trading venues have been closely regulated by the EU Markets in Financial Instruments Directive (MiFID I), the MiFID I Implementing Directive, and the MiFID I Implementing Regulation since 1 November 2007. MiFID I aims to provide a high level of harmonized investor protection, financial market transparency, and greater competition between trading venues. On 3 January 2018—some ten years later—the MiFID I regime will be replaced by MiFID II, which comprises, among other things, a directive (MIFID II), the Markets in Financial Instruments Regulation (MiFIR), and a truly impressive number of implementing measures, commonly referred to as Level 2 legislation. MiFID I may have the reputation of being strict, but MiFID II/MiFIR tightens the reins even more. It is not hard to guess the reason: the financial crisis has also revealed gaps in the MiFID I legislation, notably in investor protection, as well as shortcomings in the functioning and transparency of financial markets. The MiFID II/MiFIR regime will have a major impact on the financial sector in Europe. This volume aims to analyse and discuss the main changes and new provisions introduced by MiFID II/MiFIR. Its chapters are grouped in a thematic way, covering the following areas: (i) general aspects, (ii) investment firms and investment services, (iii) trading, (iv) supervision and enforcement, and (v) the broader view and the future of MiFID. Part I offers an overview of the developments leading to MiFID II and clarifies the central term ‘investment firm’. It also provides an overview of the volume’s contents.

Part II considers the main changes of MiFID II’s scope and exemptions in comparison with MiFID I, followed by a treatment of the governance rules for investment firms. It also analyses the duty to act in the client’s best interest, as well as the new product governance and product intervention rules. Moreover, the novel distinction between independent and nonindependent advice is scrutinized, followed by an assessment of the conflicts of interest provisions and the inducement rules. It further discusses the regulatory relevance of the distinction between agency and principal dealing. This part concludes with an outline of the new rules for thirdcountry investment firms. Part III discusses the governance and organization of trading venues and outlines the main features of the extensive new transparency regime which will apply to trading in a wide range of asset classes under MiFIR. It also contains a treatment of the new rules for SME growth markets, as well as an analysis of the concept of dark trading. Moreover, it outlines the new mandatory trading obligation for derivatives and the new commodity derivatives provisions. This part concludes with a treatment of the new rules for algorithmic and high-frequency trading, also from a US perspective. Part IV analyses the MiFID II provisions on supervision, enforcement, and cooperation by competent authorities. It also examines to what extent the civil courts are bound by MiFID I and II under EU law. Part V clarifies the relationship between MiFID II and several other closely related directives, such as the Insurance Distribution Directive, the PRIIPs Directive, and the UCITS Directive. It also analyses and discusses shadow banking and the functioning of financial markets. Finally, this part explores the policy and regulatory issues generated by investment-based crowdfunding in Europe. The volume was preceded by a meeting on 28 and 29 January 2016 of the International Working Group on MiFID II, established as a joint initiative between the Institute for Financial Law within the Business & Law Research Centre of Radboud University, Nijmegen, the Netherlands, and the Genoa Centre for Law and Finance, University of Genoa, Italy. We thank the Business & Law Research Centre of Radboud University,

Nijmegen for its sponsorship. We also thank NautaDutilh Amsterdam for hosting the meeting. We are grateful to the distinguished members of the Working Group for their dedication to the project and, in particular, for their contributions to this book as authors. We also thank the invitees to the meeting for providing the members of the Working Group with invaluable comments on their draft chapters. Last, but not least, we acknowledge our gratitude to the editorial team at Oxford University Press, who successfully brought a lengthy and complex project to completion. The law is stated as of 1 November 2016. Danny Busch Nijmegen, the Netherlands Guido Ferrarini Genoa, Italy

CONTENTS

Table of Cases Table of Legislation List of Abbreviations Author Biographies I GENERAL ASPECTS 1. Who’s Afraid of MiFID II? An Introduction Danny Busch and Guido Ferrarini I. II. III. IV. V. VI.

Introduction Investment Firms and Investment Services Trading Supervision and Enforcement The Broader View and the Future of MiFID Final Remarks

II INVESTMENT FIRMS AND INVESTMENT SERVICES 2. The Scope of MiFID II Kitty Lieverse I. II. III. IV. V. VI. VII.

Introduction Investment Firms under MiFID II Credit Institutions Investment Firms under the CRR Structured Deposits General Exemptions to MiFID II Applicability Trading on Own Account

VIII. IX. X. XI. XII. XIII. XIV.

Dealings in Emission Allowances Financial Instruments Insurance Companies and Insurance Intermediaries Managers of Collective Investment Undertakings Data-Reporting Service Providers Market Operators Conclusion

3. Governance of Investment Firms under MiFID II Jens-Hinrich Binder I. Introduction II. Governance-related Regulation of Investment Firms between Prudential and Conduct-of-Business Regulation III. The Technical Framework under MiFID II IV. Conclusions 4. The Overarching Duty to Act in the Best Interest of the Client in MiFID II Luca Enriques and Matteo Gargantini I. Introduction II. Antecedents of the Investment Firm’s Duty to Act in the Best Interest of the Client III. The Duty to Act in the Best Interest of the Client in MiFID II IV. The Duty’s Functions and Contents V. The Articulation of the Duty with Respect to Individual Services and Activities VI. A Look Across the Atlantic VII. Assessment 5. Product Governance and Product Intervention under MiFID II/MiFIR Danny Busch I. Introduction II. Product Governance III. Product Intervention

IV. Conclusion 6. Independent Financial Advice Paolo Giudici I. II. III. IV. V.

Introduction Economic Background MiFID I MiFID II Conclusions

7. Conflicts of Interest Stefan Grundmann and Philipp Hacker I. Foundations of the Regime II. The Regime by Single Stages and Examples III. Conclusion 8. Inducements Larissa Silverentand, Jasha Sprecher, and Lisette Simons I. II. III. IV. V. VI. VII.

Introduction Current Legislation The Dutch Inducement Ban The UK Inducement Rules MiFID II Research as an Inducement Conclusion

9. Agency and Principal Dealing under MiFID I and MiFID II Danny Busch I. II. III. IV.

Introduction Four Transaction Types Investor Protection Acting as Agent or as Principal: Not a Justified Distinction to Serve as a Basis for Determining the Degree of Investor Protection V. Conclusion

10.

MiFID II/MiFIR’s Regime for Third-Country Firms Danny Busch and Marije Louisse I. II. III. IV. V. VI. VII.

General Scope of MiFID II/MiFIR’s Regime for Third-Country Firms Eligible Counterparties and Per Se Professional Clients Retail Clients and Opt Up Professional Clients Retail Clients, Professional Clients, and Eligible Counterparties Initiative Test Conclusion III TRADING

11. Governance and Organization of Trading Venues: The Role of Financial Market Infrastructure Groups Guido Ferrarini and Paolo Saguato I. II. III. IV. V. 12.

Introduction Trading Facilities: Concept and Regulatory Framework The EU Scenario of Trading Venues FMI Groups in an MiFID II World Conclusion

EU Financial Governance and Transparency Regulation: A Test for the Effectiveness of Post-Crisis Administrative Governance Niamh Moloney I. Transparency Regulation and EU Financial Governance: Why It Matters II. Transparency Regulation and EU Financial Governance III. MiFIR, Transparency, and Supervisory Administrative Governance IV. MiFIR, Transparency, and Administrative Regulatory Governance V. Conclusion

13.

SME Growth Markets Rüdiger Veil and Carmine Di Noia I. Introduction

II. III. IV. V.

Existing Alternative Markets for SMEs in Europe SME Growth Markets under MiFID II Alternative Disclosure Obligations? What Role for the ‘SME Growth Market’ Label? Some Reflections on Alternative Scenarios VI. Conclusion 14.

Dark Trading under MiFID II Peter Gomber and Ilya Gvozdevskiy I. II. III. IV. V.

15.

Introduction Dark Pools: Motivation, Classification, and Related Literature Dark Pools under MiFID I Dark Pool Regulation under MiFID II/MiFIR Summary

Derivatives: Trading, Clearing, STP, Indirect Clearing, and Portfolio Compression Rezah Stegeman and Aron Berket I. II. III. IV. V. VI. VII.

16.

Introduction The Trading Obligation The Clearing Obligation The Straight-Through-Processing Obligations The Obligations in Respect of Indirect Clearing The Obligations in Respect of Portfolio Compression A Few Final Remarks

Commodity Derivatives Antonella Sciarrone Alibrandi and Edoardo Grossule I. Introduction II. Features of the Commodity Derivatives Markets and Its Relation to the Crisis III. Overview of the Most Important EU Provisions on Commodity Derivatives IV. The Regulatory Perimeter of the MiFID II/MiFIR Package V. The New Position Limits Regime

VI. A New Set of Intervention Powers for Authorities and Trading Venues VII. Conclusion 17.

Algorithmic Trading and High-Frequency Trading (HFT) Pierre-Henri Conac I. II. III. IV.

18.

Introduction Algorithmic Trading High-Frequency Trading Conclusion

MiFID II and Equity Trading: A US View Merritt B. Fox I. II. III. IV.

Basic Building Blocks High-Frequency Trading Dark Pools and Internalization Conclusions: Reflections on Issues Raised by MiFID II IV SUPERVISION AND ENFORCEMENT

19.

Public Enforcement of MiFID II Christos V. Gortsos I. Introductory Remarks II. Structure of the Study III. Competent Authorities: General Aspects, Powers, and Redress Procedures IV. Cooperation Arrangements V. Concluding Remarks and Assessment

20.

The Private Law Effect of MiFID I and MiFID II: The Genil Case and Beyond Danny Busch I. General II. May Civil Courts Be Less Strict Than MiFID I and MiFID II?

III. May Civil Courts Be Stricter Than MiFID I and MiFID II? IV. May Contracting Parties Be Less Strict Than MiFID I and MiFID II? V. May Contracting Parties Be Stricter Than MiFID I and MiFID II? VI. Influence of MIFID I and MiFID II on the Principle of Relativity VII. Influence of MiFID I and MiFID II on Proof of Causation VIII. Influence of MiFID I and MiFID II on a Contractual Limitation or Exclusion of Liability IX. MiFID I and MiFID II Assessments by the Courts of Their Own Motion in Relation to Private Investors? X. Conclusion V THE BROADER VIEW AND THE FUTURE OF MIFID 21.

MiFID II in Relation to Other Investor Protection Regulation: Picking Up the Crumbs of a Piecemeal Approach Veerle Colaert I. II. III. IV. V.

22.

Shadow Banking and the Functioning of Financial Markets Eddy Wymeersch I. II. III. IV. V. VI.

23.

Introduction MiFID II versus IDD MiFID II and IDD versus UCITS KII and PRIIPs KID Cross-Selling Practices Conclusion

Introduction Shadow Banking: Concept and Terminology Non-Regulated Activity According to the FSB Approach What Financial Activity is not Subject to Banking Supervision? The European Response to Shadow Banking Conclusion

Investment-Based Crowdfunding: Is MiFID II Enough? Guido Ferrarini and Eugenia Macchiavello I. Introduction

II. III. IV. V. VI. Index

Benefits, Risks, and Challenges MiFID I National Approaches to Crowdfunding MiFID II Concluding Remarks

TABLE OF CASES

BELGIUM

Belgian Constitutional Court, Judgment no. 89/2016 (9 June 2016) 21.24 EUROPEAN UNION

Alpine Investments v Minister Van Financier [1995] ECR I-114 23.47 Asbeek Brusse and De Man Garabito, EU CoJ 30 May 2013, NJ 2013/487, with note by Mok 20.39 Asturcom, EC CoJ 6 October 2009, no. C-40/08, NJ 2010/11 20.39 CaixaBank France v Ministère de l’Économie, des Finances et de l’Industrie (Case C-442/02) 1 CMLR 39 23.47 Cassis de Dijon—Rewe Zentral AG v Bundesmonopolverwaltung Fuer Branntwein (Case 120/78) [1979] ECR 649 23.47 Commission of the European Communities v Federal Republic of Germany (Case 205/84) [1986] ECR 3775 23.47 Commission of the European Union v Italy [1996] ECR I-2691 23.47 Genil 48 SL and Other v Bankinter SA and Others EU CoJ 30 May 2013, no. C-604/11, AA (2013) 663 1.47, 4.02, 4.37, 9.43, 9.45, 20.03, 20.05, 20.19, 20.21, 20.23, 20.29, 20.33, 20.36, 20.37, 20.40 Littlewoods Retail and others, no. C-591/10¸19 July 2012 9.43, 20.03 Meroni v High Authority (Case 9/56) [1957–1958] ECR 133 12.40, 12.55, 22.75 Mostaza Claro, EC CoJ 26 October 2006, no. C-168/05, NJ 2007/201, with note by Mok 20.39 MyTravel (C-291/03) [2005] ECR I-8477, paragraph 17 9.43, 20.03 Nationale-Nederlanden Levensverzekering Mij NV/Hubertus Wilhelminus van Leeuwen EU CoJ 29 April 2015, no. C-51/13, AA (2015) 696 1.47, 20.05–20.17, 20.18, 20.20, 20.24, 20.40

Pennon, EC CoJ 4 June 2009, no. C-243/08, NJ 2009/395, with note by Mok 20.39 Reinhard Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano [1995] ECR 1465 23.47 San Giorgio Case No. 199/82 [1983] ECR 3595 20.03 UK v Parliament and Council (C-270/12) 22 January 2014, not yet reported 12.40, 22.75 Weber’s Wine World Handels-GmbH and others v Abgabenberufungskommission Wien (Case C-147/01) [2003] All ER (D) 45 (Oct) 20.03 C-51/13, Opinion of Advocate-General Sharpston, 12 June 2014 ECLI:EU2014:1921, para 15 20.07 ITALY

Decision no. 18615/2013 (2013) Consob, Bollettino n. 9.1 (Cassa di Risparmio di Savigliano) 6.17 Decision no. 19283/2015 (2015) Bollettino n. 9.1 (Poste Italiane) 6.17 Decision no. 19368/2015 (2016) Bollettino n. 1.2 (Banca Popolare) 6.17 Decision no. 19497/2016 (2016) Bollettino n. 1.2 (Cassa di Risparmio di Bolzano) 6.17 Tribunal of Prato, 13 June 2015 6.28 Tribunal of Turin, 20 November 2012 6.28 NETHERLANDS

ABN AMRO v Van Welzen, HR 23 March 2007, NJ 2007/333 9.41 Bugro v Rabobank, Dordrecht District Court, 29 February 2012, FR 2012/48, ground 4.17 9.37 Coöperatieve Rabobank Vaart en Vecht UA v X, HR 3 February 2012, NJ 2012/95, AA (2012) 752; JOR 2012/116 9.41, 20.20 Fortis Bank v Bourgonje, HR 24 December 2010, NJ 2011/251 9.41 Holding Westkant B.V., in liquidation v ABN AMRO Bank N.V., Den Bosch Court of Appeal 15 April 2014, JOR 2014/168, with note by Van der Wiel and Wijnberg; Ondernemingsrecht 2014/92, with note by Arons 20.34 Nationale-Nederlanden v Van Leeuwen Rotterdam District Court 28 November 2012, ECLI:NL:RBROT:2012:BY5159, para 2.9 20.13 Treek v Dexia Bank Nederland, HR 5 June 2009, JOR 2009/199 9.41, 9.51

Vereniging van Effectenbezitters and Others v World Online International NV, HR 27 November 2009, NJ 2014/201, with notes by Du Perron, AA (2010) 336, Raaijmakers, JOR 2010/43, and Frielink 20.21, 20.33, 20.34, 20.35 UNITED KINGDOM

Gorham and others v British Telecommunications plc and others [2000] 1 WLR 2129 (CA) 9.40 Grant Estates Ltd v Royal Bank of Scotland [2012] CSOH 133 9.30–9.36, 9.45 Kelly v Cooper [1993] AC 205 4.90 Seymour v Ockwell [2005] EWHC 1137 (QB) 9.40 UNITED STATES

Basic, Inc. v Levinson, 485 US 224 (1988) 6.03 Lehl v SEC, 90 F.3d 1483 4.46 SEC v Capital Gains Research Bureau, Inc., 375 US 180 (1963) 4.76 SECv Goldman, Sachs & Co. and Fabrice Tourre (ABACUS case), 10 Civ. 3229 (BJ) (S.D.N.Y. filed April 16, 2010) 4.73

TABLE OF LEGISLATION

INTERNATIONAL INSTRUMENTS

Basel II Capital Accord 2004 3.01, 3.17, 3.19, 3.23 Charter of Fundamental Rights of the European Union 19.32, 20.26 Art 16 20.26 OECD Model Tax Convention on Income and on Capital Art 26 10.47 Treaty on European Union (TEU) Art 5 21.01 Treaty on the Functioning of the European Union (TFEU) Art 26 19.03 Art 53(1) 7.10, 19.03 Art 102(d) 7.75 Art 114 12.40, 19.03, 22.77 Art 114(1) 22.07 Art 127(2) 19.106 Art 127(2), last indent 19.106 Art 289(1) 19.01 Art 290 1.02 Art 290-1 19.06 Art 291 1.02 EUROPEAN UNION Directives

73/239/EEC First Council Directive 73/239/EEC on the coordination of laws, regulations and administrative provisions relating to the taking-up and pursuit of the business of direct insurance other than life assurance, [1973] OJ L 228/13 16.67

77/780/EEC First Council Directive of 12 December 1977 on the coordination of the laws, regulations and administrative provisions relating to the taking up and pursuit of the business of credit institutions, [1977] OJ L 322/30 3.10 Art 3(2) 3.10 89/646/EEC Second Council Directive of 15 December 1989 on the coordination of laws, regulations and administrative provisions relating to the taking up and pursuit of the business of credit institutions and amending Directive 77/780/EEC, [1989] OJ L 386/1 (Second Banking Directive) 3.10, 3.12 Art 5 3.10 Art 11 3.10 92/96/EEC Council Directive 92/96/EEC of 10 November 1992 on the coordination of laws, regulations and administrative provisions relating to direct life assurance and amending Directives 79/267/EEC and 90/619/EEC, [1992] OJ L 360/1 (Third Life Assurance Directive) 20.05, 20.14 recital 23 20.16 Art 31 20.10 Art 31(3) 20.11, 20.14, 20.16 Annex II 20.10, 20.11, 20.16, 20.18 Annex II(A) 20.10 Annex II(B) 20.10 93/13/EEC Directive of 5 April 1993 on unfair terms in consumer contracts, [1993] OJ L 95/29 20.39 93/22/EEC Council Directive of 10 May 1993 on investment services in the securities field, [1993] OJ L 141/7 (Investment Services Directive (ISD 1993)) 1.04–1.06, 3.01, 3.02, 3.05, 3.07, 3.08, 3.09, 3.11, 3.12, 3.14, 3.15, 3.16, 3.23, 3.50, 3.59, 11.34, 16.38, 16.44 Preamble 3.08 recital 2 3.08 recital 5 3.08 recital 6 3.09 recital 27 1.05 recital 41 3.08 Art 3(3) 3.41 Art 3(3)(1) 3.10

Art 3(3)(1), 2nd indent 3.01 Art 3(3)(2) 3.01, 3.10 Art 3(4) 3.01, 3.11 Art 3(7)(e) 3.11 Art 4 3.01, 3.10 Art 8(1) 3.12 Art 9 3.01, 3.10 Art 10 3.01, 3.12, 3.14 Art 10, sentence 1 3.11 Art 10, sentence 2, 1st indent 3.11 Art 10, sentence 2, 2nd –5th indent 3.12 Art 11 3.12 Art 11(1) 4.06 Art 14(3) 1.06 Art 17(4) 1.05 Art 18(2) 1.05 95/46/EC Directive of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data, [1995] OJ L 281/31 19.39 Arts 25–26 19.124 97/9/EC Directive of the European Parliament and of the Council of 3 March 1997, [1997] OJ 84/22 (ICS Directive) 10.47 98/26/EC Directive on settlement finality in payment and securities systems, [1998] OJ L 166/45 (SFD) 15.83, 15.98, 15.99 Art 2(f) 15.99 Art 8 15.98 2001/24/EC Directive on the recognition and winding up of credit institutions, [2001] OJ L 125/15 (Banks RWD) 15.98 2002/83/EC Directive of the European Parliament and of the Council of 5 November 2002 concerning life assurance, [2002] OJ L 345/1 16.67 2002/87/EC Directive of the European Parliament and of the Council of 16 December 2002 on the supplementary supervision of credit institutions, insurance undertakings and investment firms in a financial conglomerate and amending Council Directives 73/239/EEC, 79/267/EEC, 92/49/EEC, 92/96/EEC, 93/6/EEC, and 93/22/EEC, and Directives 98/78/EC and

2000/12/EC of the European Parliament and of the Council, [2003] OJ L 35/1 (Financial Conglomerates Directive) 11.82, 11.84, 11.86 recital 2 11.84 Chap II 11.84 Art 6 11.84 Art 7 11.84 Art 8 11.84 Art 9 11.84 Art 9(b) 11.84 Art 10 11.84 2002/92/EC Directive of the European Parliament and of the Council of 9 December 2002 on insurance mediation, [2003] OJ L 9/3 (Insurance Mediation Directive (IMD)) 2.22, 2.52, 2.53, 21.06 Arts 12–13 21.06 2003/6/EC Directive of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation, [2003] OJ L 96/16 (Market Abuse Directive (MAD)) 13.01, 13.24, 13.33, 15.01, 17.52, 20.34 Art 6(6) 17.53 Art 6(9) 17.53, 17.55 2003/41/EC Directive of the European Parliament and of the Council of 3 June 2003 on the activities and supervision of institutions for occupational retirement provision, [2003] OJ L 235/10 Art 6(a) 16.67 2003/71/EC Directive of the European Parliament and of the Council of 4 November 2003 on the prospectus to be published when securities are offered to the public or admitted to trading and amending Directive 2001/34/EC, [2003] OJ L 345/64 (Prospectus Directive) 5.28, 10.35, 13.09, 13.21, 20.31, 20.34, 22.44 Art 3(1) 13.09 Art 6(2) 20.33 Art 6(2), 1st para 20.31 2003/87/EC Directive of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC, [2003] OJ L 275/32 2.42, 2.45, 19.97

2004/39/EC Directive of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC, [2004] OJ L 145/1 (MiFID I) 1.01, 1.02, 1.04–1.06, 1.07, 1.09, 1.11, 1.13, 1.14, 1.20, 1.21, 1.29, 1.30, 1.32, 1.34, 1.40, 1.44, 1.46, 1.47, 1.50, 1.51, 2.01, 2.16, 2.25, 2.26, 2.59, 2.61, 3.01, 3.04, 3.05, 3.07, 3.14, 3.15, 3.16, 3.17, 3.23, 3.50, 3.59, 4.02, 4.14, 4.48, 4.52, 4.53, 4.56, 4.58, 4.60, 4.61, 4.71, 4.91, 6.02, 6.15–16, 6.20, 6.23, 6.36, 6.37, 6.45, 7.02, 7.03, 7.06, 7.08, 7.09, 7.10, 7.11, 7.34, 7.36, 7.43, 7.45, 7.46, 7.48, 7.54, 7.55, 7.57, 7.62, 7.88, 8.01–8.05, 8.18–8.19, 8.24, 8.25, 8.48, 8.54, 9.01–9.06, 9.08–9.10, 9.15, 9.19–9.21, 9.23, 9.27, 9.28, 9.31, 9.33–9.34, 9.36, 9.44, 9.45, 9.47–9.48, 9.54, 11.01, 11.15, 11.16, 11.18, 11.31, 11.34, 12.02, 12.04, 12.06, 12.14, 12.16, 12.18, 12.19, 12.22, 12.27, 12.28, 12.36, 13.01, 13.04, 13.53, 14.01, 14.03, 14.07–14.09, 14.36, 14.39– 14.41, 14.44, 14.45, 14.47, 14.50, 14.52, 14.59, 14.64, 14.66, 14.67, 14.71, 14.75, 14.92, 14.101, 16.45, 16.46, 16.52, 16.56, 16.57, 16.67, 17.07, 17.23, 17.34, 17.39, 17.52, 18.97, 18.105, 18.110, 18.118, 18.125, 19.01, 19.15, 19.18, 19.23, 19.26, 19.27, 19.40, 19.78, 19.82, 19.83, 19.99, 19.100, 19.107, 19.110, 19.114, 19.117, 19.123, 19.126, 19.129, 20.01, 20.02, 20.04–20.05, 20.19, 20.20, 20.22–20.27, 20.30–20.40, 21.04, 21.05, 21.06, 21.11, 21.57, 23.12–23.17, 23.19–23.22 recital 2 7.10, 9.02, 20.25, 20.33, 20.39 recital 17 3.14 recital 18–19 3.14 recital 22 3.14 recital 24 3.14 recital 25 3.14 recital 29 7.11 recital 31 7.10 recital 33 4.71, 7.10 recital 40 4.61 recital 44 7.10 recital 53 14.44 Title II 1.09, 11.59 Title III 1.09 Art 1(2) 1.13

Art 2(1)(d) 4.52, 9.20 Art 2(1)(d)(iv) 4.52, 4.61 Art 2(1)(k) 16.54 Art 2(1)(2) 4.61 Art 3 23.15, 23.16, 23.17, 23.22, 23.37 Art 4(1) 1.09 Art 4(1)(2) 14.41, 14.42, 14.43 Art 4(1)(4) 23.16 Art 4(1)(5) 9.06, 23.15 Art 4(1)(6) 9.08 Art 4(1)(7) 9.10, 14.43 Art 4(1)(10) 9.21 Art 4(1)(14) 9.05, 14.41 Art 4(1)(15) 1.11, 9.05, 14.42 Art 4(1)(18) 21.04 Art 4(1)(19) 21.04 Art 4(2) 1.09 Art 5(1) 1.12 Art 5(5) 19.12 Art 6(3) 1.12 Art 6(31) 1.12 Art 6(32) 1.12 Arts 9–13 1.12 Art 9 3.14, 3.56 Art 10 3.14, 3.56 Art 13 3.14 Art 13(2) 3.15, 7.33, 17.23, 17.24, 17.34, 17.39, 17.55 Art 13(2)–(8) 3.15 Art 13(3) 7.09, 7.26, 7.28, 7.34, 7.36, 7.45 Art 13(4) 3.14, 7.33, 17.23, 17.24, 17.34 Art 13(5) 3.14, 3.15, 7.33, 17.23, 17.24, 17.34, 17.39, 17.55 Art 13(6) 7.26, 7.33, 17.23, 17.24, 17.34, 17.39, 17.55 Art 13(7) 3.15, 7.33 Art 13(8) 3.15, 7.33 Art 14(1) 17.34, 17.39 Art 14(4) 17.39 Art 15 10.20

Art 16 1.12 Art 16(3) 19.12 Art 17(2) 19.12 Art 18 et seq 7.09 Art 18 7.09, 7.11, 7.26 Art 18(1) 7.07 Art 18(2) 7.28, 7.48 Art 18(3) 7.09 Art 18ff 1.12 Art 19 4.61 Art 19(1) 4.06, 4.92, 7.57, 8.01, 20.20 Art 19(3), last sentence 20.20 Art 21 3.54, 4.71, 7.09, 7.63 Art 21(1) 7.67 Art 22(3) 7.38 Art 23(1) 23.42, 23.43 Art 24 4.14 Art 25 7.38 Art 25(1) 9.20 Art 26 17.39 Art 27 7.09 Art 27(1) 14.50 Art 27(1), first para 9.10 Art 28(1) 9.10 Art 29 14.52 Art 29(1) 14.51 Art 29(2) 14.51, 23.43 Art 31(1) 8.24 Art 34 11.59 Art 39(b) 17.34, 17.53 Art 39(c) 17.34 Art 39(d) 17.53 Art 42(3) 17.39 Art 43 17.53 Art 44 14.52 Art 48 19.15 Art 48(2) 19.12

Art 49 19.19 Art 50 19.27, 19.30 Art 51 19.40 Art 52 19.79 Art 53 19.82 Art 54 19.23 Art 55 19.26 Art 56 19.83 Art 56(2) 19.92 Art 56(5) 19.95 Art 57(1) 19.99 Art 57(2) 19.100 Art 58 19.107 Art 58a–59 19.110 Art 60 19.114 Art 61 19.118 Art 62 19.123 Art 62a 19.84 Art 63 19.126 Annex I, Section A 1.09, 21.04 Annex I, Section A, point (1) 3.25 Annex I, Section A, point (2) 3.25 Annex I, Section A, point (4) 3.25 Annex I, Section A, point (5) 3.25 Annex I, Section B 21.04 Annex I, Section B, point (1) 3.25 Annex I, Section C 1.09, 1.11, 21.04, 23.13 Annex I, Section C(5) 16.38, 16.45 Annex I, Section C(6) 16.38, 16.45 Annex I, Section C(7) 16.38, 16.45 Annex I, Section C(9) 16.38 Annex I, Section C(10) 16.38 2004/72/EC Directive of 29 April 2004 implementing Directive 2003/6/EC of the European Parliament and of the Council as regards accepted market practices, the definition of inside information in relation to derivatives on commodities, the drawing up of lists of insiders, the notification of managers’ transactions and the notification of suspicious

transactions, [2004] OJ L 162/70 (MAD Implementing Directive for regulated markets) Arts 7–10 17.53, 17.55 2004/109/EC Directive of the European Parliament and of the Council of 15 December 2004 on the harmonization of transparency requirements in relation to information about issuers whose securities are admitted to trading on a regulated market and amending Directive 2001/34/EC, [2004] OJ L 390/38 (Transparency Directive) 5.28, 13.01, 13.13, 13.33, 20.34 Art 3(1a) 13.33 Art 4 13.23 Art 5 13.23 Art 9 3.56 Art 10 3.56 Art 12(4) 3.56 Art 12(5) 3.56 2005/1/EC Directive of the European Parliament and of the Council of 9 March 2005 amending Directive 2002/87/EC, [2005] OJ L 79/9 11.84 2005/29/EC Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices in the internal market and amending Council Directive 84/450/EEC, Directives 97/7/EC, 98/27/EC and 2002/65/EC of the European Parliament and of the Council and Regulation (EC) No 2006/2004 of the European Parliament and of the Council, [2005] OJ L 149/22 (Unfair Commercial Practices Directive) 21.08, 21.16, 21.54, 21.60 Annex II 20.34 2005/68/EC Directive of the European Parliament and of the Council of 16 November 2005 on reinsurance and amending Council Directives 73/239/EEC, 92/49/EEC as well as Directives 98/78/EC and 2002/83/EC, [2005] OJ L 323/1 16.67 2006/43/EC Directive of the European Parliament and of the Council of 17 May 2006 on statutory audits of annual accounts and consolidated accounts, amending Council Directives 78/660/EEC and 83/349/EEC and repealing Council Directive 84/253/EEC, [2006] OJ L 157/87 3.35, 19.24 Arts 3–14 19.24 Art 41 3.35

2006/48/EC Directive of the European Parliament and of the Council of 14 June 2006 relating to the taking up and pursuit of the business of credit institutions, [2006] OJ L 177/1 (recast Banking Directive) 3.17, 16.67, 22.41 Art 22 3.47, 3.48 Art 22(1) 3.17 Annex V 3.17 2006/49/EC Directive of the European Parliament and of the Council of 14 June 2006 on the capital adequacy of investment firms and credit institutions (recast), [2006] OJ L 177/201 (recast Capital Adequacy Directive) 3.17, 3.19, 3.20 Art 34 3.18 2006/73/EC Commission Directive of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive, [2006] OJ L 241/26 (MiFID I Implementing Directive) 1.01, 3.15, 3.16, 3.48, 3.50, 6.15, 7.34, 7.39, 8.05, 8.05–8.07, 8.08–8.09, 8.25, 8.26, 8.30, 9.01, 11.01, 11.15, 20.01 recital 4 3.16 recital 5 3.16 recital 7 3.16 recital 27 4.91 recital 69 4.71, 9.09 recitals 81–83 4.39 recital 81 4.24 recital 82 4.38 Chap 2 3.15 Chap 3 3.15 Art 4 8.25, 8.27, 8.28, 8.33, 8.59 Arts 5–9 17.24, 17.34 Art 5 3.15, 17.23, 17.55 Art 6 3.15, 17.23, 17.55 Art 7 3.15, 17.23 Art 7(2) 3.48 Art 8 3.15 Art 9 3.15, 17.23, 17.55

Art 13 17.23, 17.24, 17.34, 17.39 Art 14 17.23, 17.24, 17.34, 17.39 Art 14(1) 20.37 Art 16 3.15 Arts 21–26 7.36, 7.38 Art 21 7.36, 7.38 Art 21(a) 7.36, 7.38 Art 21(b) 7.36, 7.38, 7.75 Art 21(c) 7.38 Art 21(d) 7.38 Art 21(e) 7.38, 8.09 Art 22(3) 7.34, 7.36, 7.38 Art 22(3)(2)(a) 7.38 Art 22(3)(2)(b) 7.38 Art 22(3)(2)(c) 7.38 Art 22(3)(2)(d) 7.38 Art 22(3)(2)(e) 7.38 Art 22(4) 7.48 Art 26 8.06, 8.09, 8.17, 8.25, 8.31 Art 39 6.15 Art 44(1) in fne 9.05 Art 44(3) 7.67 Art 51 17.23, 17.24, 17.34, 17.39 Art 52 4.24, 4.37 2008/25/EC Directive 2008/25/EC of the European Parliament and of the Council of 11 March 2008 amending Directive 2002/87/EC on the supplementary supervision of credit institutions, insurance undertakings and investment firms in a financial conglomerate, as regards the implementing powers conferred on the Commission, [2008] OJ L 81/40 11.84 2009/65/EC Directive of the European Parliament and of the Council of 13 July 2009 on the coordination of laws, regulations, and administrative provisions relating to undertakings for collective investment in transferable securities, [2009] OJ L 302/32 (UCITS IV) 2.32, 2.55, 5.02, 16.67, 19.17, 21.03, 21.27, 21.28, 21.30, 21.33, 21.35, 21.50, 21.61, 22.79 Art 1(2) 19.17

Art 5 19.22 Art 6 2.56 Art 6(3) 5.02 Art 50(1) 16.85 Art 50(2) 16.85 Art 52 22.79 Art 73 19.24 Art 78 21.28 Art 80 21.28 2009/111/EC Directive of the European Parliament and of the Council of 16 September 2009 amending Directives 2006/48/EC, 2006/49/EC and 2007/64/EC as regards banks affiliated to central institutions, certain own funds items, large exposures, supervisory arrangements and crisis management, [2009] OJ L 302/97 (CRD II) 3.14, 22.41 3.14, 22.41 Art 122(a) 22.55 2009/138/EU Directive of the European Parliament and of the Council of 25 November 2009 on the taking-up and pursuit of the business of Insurance and Reinsurance, [2009] OJ L 335/1 (Solvency II) 2.32, 22.44 Art 15 2.51 Art 16 2.51 2010/73/EU Directive of 24 November 2010 amending Directives 2003/71/EC on the prospectus to be published when securities are offered to the public admitted to trading and 2004/109/EC on the harmonization of transparency requirements in relation to information about issuers whose securities are admitted to trading on a regulated market, [2010] OJ L 327/1 22.110 2010/76/EU Directive of the European Parliament and of the Council of 24 November 2010 amending Directives 2006/48/EC and 2006/49/EC as regards capital requirements for the trading book and for resecuritizations, and the supervisory review of remuneration policies, [2010] OJ L 329/3 (CRD III) 3.18, 3.19, 22.41 Preamble 3.18 recitals 7–10 3.18 recital 17 3.18 recitals 17–22 3.18 Art 1(3) 3.18

2010/78/EU Directive of the European Parliament and of the Council of 24 November 2010 amending Directives 98/26/EC, 2002/87/EC, 2003/6/EC, 2003/41/EC, 2003/71/EC, 2004/39/EC, 2004/109/EC, 2005/60/EC, 2006/48/EC, 2006/49/EC and 2009/65/EC in respect of the powers of the European Supervisory Authority (European Banking Authority), the European Supervisory Authority (European Insurance and Occupational Pensions Authority) and the European Supervisory Authority (European Securities and Markets Authority), [2010] OJ L 331/120 11.84 2011/61/EU Directive of the European Parliament and of the Council of 8 June 2011 on Alternative Investment Fund Managers and amending Directives 2003/41/EC and 2009/65/EC and Regulations (EC) No 1060/2009 and (EU) No 1095/2010, [2011] OJ L 174/1 (AIFMD) 1.24, 2.32, 2.55, 2.56, 5.02, 10.23, 10.26, 16.39, 16.41, 16.67, 19.06, 22.44, 22.82 Art 4(1)(a) 19.22 Art 6 2.56 Art 6(4) 2.56, 5.02 Art 6(6) 2.56 Art 21(5) 10.26 Art 21(6) 10.26 Art 21(6), last para 10.26 Art 25(7) 22.39 Art 37 10.26 Art 37(2) 10.26 Art 37(23)(b) 10.26 Art 53 22.39 Art 67 10.26 2011/89/EU Directive of the European Parliament and of the Council of 16 November 2011 amending Directives 98/78/EC, 2002/87/EC, 2006/48/EC and 2009/138/EC as regards the supplementary supervision of financial entities in a financial conglomerate, [2011] OJ L 326/113 11.84 2013/34/EU Directive of the European Parliament and of the Council of 26 June 2013 on the annual financial statements, consolidated financial statements and related reports of certain types of undertakings, amending Directive 2006/43/EC of the European Parliament and of the Council and

repealing Council Directives 78/660/EEC and 83/349/EEC, [2013] OJ L 182/19 19.52 Art 2(11) 19.113 Art 22(1)–(2) 19.112 Art 34 19.24 2013/36/EU Directive of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC, [2013] OJ L 176/338, (Capital Requirement Directive (CRD IV)) 1.13, 1.15, 1.26, 2.15, 2.16, 2.17, 2.22–2.25, 2.28, 2.39, 3.03, 3.06, 3.18, 3.21, 3.23–3.28, 3.34–3.36, 3.44–3.50, 3.52, 3.55, 3.58–3.63, 3.65, 4.67, 10.18, 10.27, 10.34, 10.48, 10.62, 11.25, 11.84, 12.06, 12.41, 15.01, 15.45, 15.79, 15.110, 16.36, 16.53, 16.60, 17.45, 19.05, 19.50, 19.105, 22.06 Preamble 3.59 recital 55 3.26 recital 60 3.43 Art 2(1) 3.25 Art 3(1)(7) 3.26, 3.28 Art 3(1)(33) 3.56 Art 3(2) 3.26 Art 9(1) 3.25 Art 13(1) 3.29 Art 14 3.56 Art 14(1) 3.56 Art 14(2) 3.57 Art 18 19.50 Arts 22–24 3.57 Arts 22–27 3.56 Art 23(1) 3.57 Art 28(2) 2.22 Art 29(1) 2.22 Art 29(2) 2.22 Art 30 2.22 Art 31(1) 2.22 Art 31(2) 2.22

Art 67(1) 19.50 Art 67(2)(f)–(g) 3.33 Arts 74–87 3.47 Art 74 7.34 Art 74(1) 3.47 Art 74(2) 3.48 Art 76(1) 3.48 Art 76(2) 3.48 Art 76(3) 3.27, 3.35, 3.48 Art 76(4) 3.48 Art 76(5) 3.48 Arts 77–84 3.48 Art 88 3.21, 3.25, 3.26, 3.27, 3.33, 7.34, 10.47, 11.18 Art 88(1) 3.29, 3.48, 11.25 Art 88(1)(e) 3.26, 3.63 Art 88(1)(2) 3.30 Art 88(2) 3.27, 3.35, 11.25 Art 88(2)(a) 3.39 Art 88(2)(2)(a)–(d) 3.35 Art 91 3.21, 3.25, 3.33, 7.34, 10.47, 11.18 Art 91(1) 3.37, 11.25 Art 91(2) 3.37, 11.25 Art 91(7) 3.39 Art 91(8) 3.33, 3.38 Art 91(9) 3.39 Art 91(10) 3.39 Art 91(12) 3.40 Art 92(1) 3.45, 3.46 Art 94 3.45 Art 95(1) 3.27 Art 95(2) 3.35 2014/17/EU of the European Parliament and of the Council of 4 February 2014 on credit agreements for consumers relating to residential immovable property and amending Directives 2008/48/EC and 2013/36/EU and Regulation (EU) No 1093/2010, [2014] OJ L 60/34 (Mortgage Credit Directive (MCD)) 21.53, 21.54 Art 4(26) 21.54

Art 4(27) 21.54 Art 12 21.53, 21.54 2014/49/EU Directive of the European Parliament and of the Council of 16 April 2014 on deposit guarantee schemes, [2014] OJ L 173/149 (Deposit Guarantee Schemes Directive) 22.27 Art 2(1)(c) 2.27, 5.05, 19.05 Art 2(1)(3) 21.06 2014/57/EU Directive of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse, [2014] OJ L 173/179 (Market Abuse Directive (MADII)) 10.15, 10.35, 15.01, 16.22, 18.10, 18.118, 19.33 recital 1 6.03 2014/59/EU Directive of the European Parliament and of the Council of 15 May 2014 establishing a framework for the recovery and resolution of credit institutions and investment firms and amending Council Directive 82/891/EEC, and Directives 2001/24/EC, 2002/47/EC, 2004/25/EC, 2005/56/EC, 2007/36/EC, 2011/35/EU, 2012/30/EU and 2013/36/EU, and Regulations (EU) No 1093/2010 and (EU) No 648/2012, [2014] OJ L 173/190 (Bank Recovery and Resolution Directive (BRRD)) 4.67, 10.55 Art 1(1)(e) 10.55 Art 2(1)(89) 10.55 Art 43ff 4.34 Art 95 10.56 Art 96 10.55 Art 96(1), 2nd para 10.55 Art 96(3) 10.56 Art 103(1) 10.56 2014/65/EU Directive of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU (recast), [2014] OJ L 173/349 (MiFID II) 1.01, 1.02, 1.03, 1.07, 1.09, 1.11, 1.13–1.30, 1.33, 1.34, 1.40, 1.42–1.48, 1.50–1.52, 1.56, 1.57, 2.01–2.03, 2.05–2.06, 2.09, 2.11, 2.13– 2.16, 2.19–2.35, 2.37–2.39, 2.41–2.45, 2.49, 2.51–2.53, 2.55–2.61, 3.01– 3.04, 3.06, 3.18, 3.20–3.25, 3.28, 3.34–3.36, 3.44, 3.45, 3.47, 3.50–3.57, 3.59–3.63, 3.65, 4.01, 4.03, 4.07, 4.14–4.16, 4.18, 4.19, 4.24, 4.31, 4.32, 4.47–4.51, 4.54, 4.58–4.62, 4.68, 4.71, 4.72, 4.75, 4.82, 4.83, 4.88–4.92,

5.01–5.04, 5.06–5.07, 5.09, 5.11, 5.12, 5.27, 5.28, 5.40, 5.41, 5.61, 5.62, 5.74, 6.01, 6.02, 6.13, 6.19, 6.20, 6.22–6.24, 6.27, 6.28, 6.31–6.33, 6.37, 6.38, 6.44, 6.45, 6.47, 7.01, 7.02, 7.06–7.09, 7.11, 7.12, 7.18, 7.20, 7.25, 7.36, 7.37, 7.40, 7.42, 7.43, 7.46, 7.48–7.50, 7.52, 7.53, 7.55, 7.57, 7.58, 7.60, 7.63–7.65, 7.72, 7.74, 7.88, 8.03, 8.18, 8.26, 8.28, 8.37, 8.47, 8.60, 8.61, 8.64, 8.66, 9.01–9.06, 9.08, 9.10, 9.16, 9.19–9.21, 9.23, 9.44, 9.45– 9.47, 9.50, 9.54, 9.55, 10.01–10.07, 10.09–10.13, 10.14, 10.15, 10.16, 10.18, 10.20–10.23, 10.27, 10.31, 10.34, 10.41, 10.45–10.50, 10.52– 10.54, 10.62, 10.67, 10.68, 10.70, 10.72, 10.76–10.78, 11.01–11.02, 11.04, 11.15–11.17, 11.20–11.22, 11.27, 11.31, 11.34, 11.40, 11.64, 11.65, 11.73, 11.75, 11.76, 11.80, 11.81, 11.86, 12.01, 12.02, 12.04, 12.05, 12.15, 12.23, 12.31, 12.40, 12.50, 12.51, 13.02, 13.03, 13.15, 13.17–13.18, 13.27, 13.32, 13.48, 13.53, 13.54, 14.07–14.09, 14.59, 14.60, 14.62, 14.64–14.67, 14.70, 14.75, 14.91, 14.92, 14.94, 14.100, 14.101, 14.102, 14.104–14.106, 15.01, 15.04, 15.36, 15.109, 15.112, 15.126, 16.02–16.03, 16.18, 16.21, 16.29, 16.31–16.34, 16.42–16.44, 16.46–16.48, 16.50–16.53, 16.55, 16.58, 16.60, 16.65, 16.67, 16.71– 16.74, 16.82, 16.88–16.89, 17.02, 17.20, 17.23–17.24, 17.28, 17.31, 17.33, 17.34, 17.39, 17.41–17.44, 17.51, 17.56, 17.61, 17.62, 18.03, 18.40, 18.94, 18.105, 18.106, 18.111, 18.113, 18.115, 18.118, 18.127, 19.01, 19.04, 19.06, 19.08, 19.12, 19.27, 19.33, 19.34, 19.37, 19.40, 19.41, 19.43, 19.50, 19.58, 19.62, 19.75, 19.78, 19.79, 19.80, 19.83, 19.85, 19.98, 19.119, 19.121, 19.129, 19.130, 19.131, 20.01, 20.02, 20.04–20.05, 20.19, 20.20, 20.21–20.27, 20.20–20.40, 21.03, 21.04, 21.06–21.10, 21.13–21.20, 21.22–21.30, 21.32, 21.33, 21.34, 21.35, 21.38, 21.43, 21.44, 21.50, 21.52–21.55, 21.57, 21.59, 21.61, 21.62, 22.06, 22.62, 22.87, 23.12, 23.19–23.22, 23.34–23.36, 23.38–23.40, 23.43–23.44, 23.51–23.52, 23.57, 23.59 Preamble 3.20, 3.59 recital 4 3.04 recital 5 3.04, 3.23 recital 7 7.10 recital 9 2.49 recital 10 2.49 recital 11 2.45 recital 12 2.12 recital 18 2.38

recital 19 2.38 recital 20 2.38, 16.59 recital 20(2) 2.39 recital 21 16.72 recital 22 2.40 recital 24 2.08, 9.48, 9.49 recital 26 2.11 recital 27 2.32 recital 28 2.34, 10.01 recital 32 2.35 recital 33 2.35 recital 34 2.32 recital 38 2.15 recital 39 2.26 recital 40 2.28, 11.31 recital 41 2.33 recital 42 7.10 recital 45 2.06, 9.47 recital 51 7.10 recital 53 et seq 7.10 recital 54 5.30 recital 55 3.26 recital 56 7.11 recital 64 14.62 recital 69 7.11 recital 70 9.02, 20.25, 20.33, 20.39 recital 71 5.15, 5.16, 5.27, 5.32, 5.40, 6.41, 7.37, 7.40 recital 71 et seq 7.10 recital 71, 1st para 4.10 recital 71, 2nd para 4.10 recital 74 8.40, 8.41 recital 75 8.42 recital 77 4.20, 7.10 recital 78 21.30, 21.31 recital 80 7.10, 7.74 recital 81 7.10, 7.75 recital 85 4.66, 10.73

recital 87 2.52, 7.10, 21.06 recital 88 7.11 recital 89 23.34 recital 91 4.71, 7.10 recitals 91–98 7.62 recital 92 7.64 recital 93 4.72 recital 95 7.68 recital 97 7.64 recital 109 4.61, 10.45 recital 111 10.69, 10.73 recital 115 2.59 recital 116 2.59 recital 123 7.11 recital 132 13.15 recital 134 13.15, 13.49 recital 136 19.103 recital 137 19.27, 19.130 recital 138 19.37, 19.38 recital 139 19.132 recital 140 19.92 recital 141, 1st sentence 19.42, 19.43 recital 141, 2nd sentence 19.48 recital 142 19.52 recital 143 19.32 recital 144 19.33 recital 145 19.55 recital 146, 1st–5th sentences 19.56 recital 146, 6th sentence 19.69 recital 146, 7th sentence 19.62 recital 147 19.76, 19.78 recital 148 19.40 recital 149 19.42 recital 150 19.44 recital 151 19.81 recital 154 19.93

recital 155 7.10, 7.11 recital 164 4.14, 7.10 recitals 10.72 Title II 1.09, 2.04, 9.16, 11.22, 11.23 Title III 1.09, 11.17, 11.65 Title IV 2.47 Title V 10.13 Title VI 19.07 Art 1(1) 2.01, 10.04, 10.11, 19.04 Art 1(2) 19.04 Art 1(2)(b) 10.09 Art 1(23) 19.05 Art 1(3) 1.13, 2.15, 2.33, 19.05 Art 1(3)(a) 5.02 Art 1(3)(a)–(d) 2.15 Art 1(3)(b) 5.02 Art 1(4) 1.13, 2.26, 2.28, 5.05, 5.12, 5.15, 5.16, 5.18, 5.19, 5.24, 5.25, 5.27, 5.29, 5.30, 5.31, 5.32, 5.40, 10.09, 10.10, 19.05, 21.06 Art 1(4)(a) 5.05, 5.18, 5.19, 5.24, 5.25, 5.27, 5.29, 5.30, 5.32, 5.40 Art 1(4)(b) 5.05, 5.15, 5.16, 5.31 Art 1(4)(c) 5.41, 5.61, 5.62 Art 1(6) 15.106 Art 2 2.31 Art 2(1)(a) 2.31 Art 2(1)(b) 2.31, 2.34 Art 2(1)(c) 2.13, 2.31, 2.35 Art 2(1)(d) 2.19, 2.31, 2.37, 2.38, 2.39, 2.40, 9.20 Art 2(1)(d)(iii) 17.31, 17.44 Art 2(1)(e) 2.31, 2.42 Art 2(1)(f) 2.31, 2.34 Art 2(1)(g) 2.31, 2.34 Art 2(1)(h) 2.31, 2.35 Art 2(1)(i) 2.31, 2.37, 2.55, 16.53, 16.56 Art 2(1)(j) 2.19, 2.31, 2.37, 2.39, 2.40, 9.20, 16.53, 16.55, 16.58, 16.61 Art 2(1)(j), 1st sentence 2.39 Art 2(1)(k) 2.13, 2.31, 2.35, 2.37 Art 2(1)(l) 2.31

Art 2(1)(m) 2.31 Art 2(1)(n) 2.31 Art 2(1)(o) 2.31, 2.33 Art 2(2) 2.31, 2.56 Art 2(4) 2.39, 16.43 Art 2(9)–(10) 19.52 Art 3 2.30 Art 3(1)–(2) 2.30 Art 3(1) 23.37, 23.51 Art 3(1)(a)–(c) 2.30 Art 3(1)(c) 23.37 Art 3(1)(d) 2.30 Art 3(1)(e) 2.30 Art 3(2) 23.37, 23.51 Art 3(2), 2nd indent 23.37 Art 4(1) 1.09, 2.11, 2.28, 2.57 Art 4(1)(a)–(d) 2.11 Art 4(1)(1) 2.02, 2.10, 2.11, 2.16, 3.25, 19.04 Art 4(1)(2) 2.03, 2.10, 2.14, 19.04, 23.39 Art 4(1)(2)(a)–(c) 3.25 Art 4(1)(3) 2.10, 19.05 Art 4(1)(4) 2.04, 2.14, 23.40 Art 4(1)(4)–(7) 2.10 Art 4(1)(5) 2.04, 2.06, 2.14, 4.54, 4.58, 9.06, 9.47 Art 4(1)(6) 2.04, 2.14, 9.08 Art 4(1)(7) 2.14, 2.38 Art 4(1)(8) 2.04, 2.10 Art 4(1)(9) 2.10, 2.14, 9.21 Art 4(1)(11) 21.24 Art 4(1)(13) 13.17 Art 4(1)(14) 12.31 Art 4(1)(15) 1.11, 2.03, 19.34 Art 4(1)(16) 2.49 Art 4(1)(17) 2.46 Art 4(1)(18) 19.04 Art 4(1)(20) 9.10, 11.06, 12.16, 15.31 Art 4(1)(21) 9.05, 10.14, 11.17, 12.15, 19.04

Art 4(1)(22) 9.05, 10.14, 11.23, 12.15, 13.01, 19.05 Art 4(1)(23) 2.04, 2.60, 10.14, 11.22, 12.15, 23.44 Art 4(1)(24) 2.60, 10.14, 11.05, 19.05 Art 4(1)(27) 19.05 Art 4(1)(29) 23.43 Art 4(1)(30) 19.04 Art 4(1)(31) 3.56 Art 4(1)(32) 2.34, 19.52, 19.113 Art 4(1)(33) 2.34, 19.52 Art 4(1)(35(b)) 19.112 Art 4(1)(36) 3.26, 3.28, 19.42 Art 4(1)(38) 2.08, 9.48, 11.29 Art 4(1)(39) 17.03 Art 4(1)(40) 17.05 Art 4(1)(41) 17.29 Art 4(1)(42) 7.75, 21.52 Art 4(1)(43) 2.27, 5.05, 19.05, 21.06 Art 4(1)(44) 2.46, 23.39 Art 4(1)(44)(c) 15.03, 16.47 Art 4(1)(50) 19.31 Art 4(1)(51) 19.22 Art 4(1)(52)–(54) 19.04 Art 4(1)(52) 2.57, 15.105 Art 4(1)(53) 2.57 Art 4(1)(54) 2.57 Art 4(1)(55) 10.11, 19.91 Art 4(1)(56) 19.91 Art 4(1)(57) 10.04, 19.04 Art 4(1)(63) 19.04 Art 4(2) 1.09, 4.37, 23.41 Art 4(4) 6.15 Arts 5–43 19.05 Art 5 13.48, 19.47 Art 5(1) 1.12, 11.23 Art 5(2) 11.23 Art 5(4)(b) 2.11 Art 6 19.78

Art 6(2) 19.47 Art 6(3) 1.12 Art 6(34) 1.12 Art 6(35) 1.12 Art 7(2)–(3) 19.78 Art 7(3) 10.31 Art 8 10.50 Art 8(d) 3.33 Arts 9–16 1.12 Art 9 3.33, 4.37, 11.25 Art 9(1) 3.21, 3.25, 3.27, 3.29, 3.35 Art 9(3) 2.28, 3.29, 3.31, 3.34, 7.10 Art 9(3)(a) 5.30 Art 9(3)(b) 5.02, 5.12, 5.24 Art 9(3)(c) 3.34, 3.45 Art 9(3)(3) 3.32 Art 9(3)(4) 3.53 Art 9(3)–(6) 3.21 Art 9(4) 3.37 Art 9(6) 2.11, 3.29 Arts 10–13 3.21 Art 10 3.57, 11.25 Art 10(1) 3.56, 11.77, 11.78 Art 10(1)(2) 3.57 Art 12 2.56, 3.57 Art 13 2.56, 3.57 Art 13(1) 3.57 Art 16 3.50, 11.26 Art 16(2) 3.53, 7.33 Art 16(3) 2.28, 4.49, 6.40, 7.09, 7.26, 7.28, 7.34, 7.37, 7.45, 19.35, 21.08, 21.13, 21.16, 23.35 Art 16(3) 2nd para 5.18, 23.35 Art 16(3) 3rd para 5.19, 5.29, 23.35 Art 16(3) 4th para 5.29, 5.32, 23.35 Art 16(3) 5th para 5.29 Art 16(3) 6th para 5.27

Art 16(3) 7th para 5.40 Art 16(3)(1) 3.54, 7.36 Art 16(3)(2) 3.54, 7.40 Art 16(3)(2–6) 7.40 Art 16(3)(2–7) 5.02 Art 16(3)(3) 3.54 Art 16(3)(4) 3.54, 7.40 Art 16(3)(6) 7.40 Art 16(3)(7) 3.54 Art 16(4) 3.51, 7.33 Art 16(5) 7.33 Art 16(5)(2) 3.52 Art 16(5)(3) 3.52 Art 16(6) 3.54, 7.26, 7.33 Art 16(7) 3.54, 7.33 Art 16(8) 2.23, 7.33 Art 16(8)–(10) 3.54 Art 16(9) 2.23, 7.33 Art 16(12) 3.50 Art 17 3.50, 7.33, 17.20, 17.21 Art 17(1) 17.24, 17.56, 18.118 17.26, 17.46 Art 17(3) 17.27 Art 17(4) 17.27 Art 17(5) 17.31, 17.56 Art 17(6) 17.32 Art 18 11.76, 22.89 Art 18(1)–(4) 11.27 Art 18(5) 11.27 Art 18(11) 11.77 Art 19 2.56 Art 19(1) 11.26, 11.28 Art 19(2) 11.28 Art 19(3) 11.28 Art 19(4) 4.15 Art 19(5) 11.28 Art 20 2.60, 12.15 Art 20(1) 11.29

Art 20(2) 11.24, 11.29 Art 20(3) 11.29 Art 20(4) 11.29 Art 20(6) 4.15, 11.29 Art 20(6), 2nd para 9.16 Art 20(6)(8) 11.11 Art 20(7) 11.24, 11.66 Art 21 1.12 Art 22 19.52 Art 23 et seq 7.09 Art 23 3.54, 4.01, 4.25, 7.09, 7.11, 7.26, 21.13 Art 23(1) 7.07 Art 23(2) 7.28, 7.45, 7.48 Art 23(3)(a) 7.45 Art 23(3)(b) 7.48, 7.49, 7.52 Art 23(4) 7.09 Art 23ff 1.12 Arts 24–25 19.117 Art 24 2.28, 4.08, 4.11, 4.15, 4.61, 6.38, 8.40, 23.47 Art 24(1) 1.18, 4.01, 4.06, 4.07, 4.09, 4.10, 4.11, 4.12, 4.13, 4.15, 4.16, 4.24, 4.31, 4.33, 4.37, 4.60, 4.82, 4.92, 9.20, 20.20, 21.08, 21.16 Art 24(2) 4.10, 4.49, 4.60, 5.02, 7.40, 21.08, 21.38 Art 24(2) 1st para 5.15 Art 24(2) 2nd para 4.08, 5.16, 5.25, 5.31 Art 24(2)(2) 7.40 Art 24(3)–(5) 4.42 Art 24(3) 21.16 Art 24(4) 4.14, 4.60, 21.10, 21.13, 23.34 Art 24(4), last para 21.29 Art 24(4)–(9) 7.09 Art 24(4)(a) 6.20, 6.31, 6.34, 7.51 Art 24(4)(a)(i) 7.56 Art 24(4)(a)(ii) 6.25 Art 24(4)(b) 7.40, 7.51 Art 24(4)(c) 6.29, 7.51 Art 24(5) 4.14, 4.60, 23.34 Art 24(5), last sentence 20.20

Art 24(6) 4.60, 7.73, 21.13 Arts 24(7)–(9) 21.18 Art 24(7) 4.44, 6.21, 6.13, 7.58, 21.19 Art 24(7)(a) 4.35, 6.21 Art 24(7)(b) 4.08, 6.22 Art 24(8) 4.08 Art 24(9) 4.01, 4.25, 4.60, 7.59, 8.47, 21.18 Art 24(9), 1st para, sub (b) 4.09 Art 24(9), 3rd para 4.09 Art 24(10) 4.01, 4.08, 4.20, 4.60 Art 24(11) 4.60, 6.38, 21.09, 21.52, 21.55 Art 24(11)(1) 7.76 Art 24(11), 3rd subpara 4.18 Art 24(11)(2) 7.76 Art 24(12) 8.28, 8.58, 8.59, 9.54, 19.117, 21.10, 23.47, 23.48 Art 24(13) 8.38, 21.25, 21.29 Art 24(13)(d) 4.09, 4.18 Art 24(14) 21.25, 21.29 Art 24(14)(c) 7.49 Art 25 2.28, 6.38, 12.04, 23.47 Art 25(1), 4.12, 4.24, 4.60, 6.38 Art 25(2) 2.28, 4.01, 4.12, 4.60, 21.43 Art 25(2)(2) 7.76 Art 25(2)(4) 21.13 Art 25(3) 2.28, 4.01, 4.28, 4.60, 7.76 Art 25(4) 2.28, 4.13, 4.60, 4.62, 4.64, 4.72, 7.74, 23.36 Art 25(4)(a)(v) 2.28 Art 25(4)(b) 4.35 Art 25(4)(d) 4.68, 4.91 Art 25(5) 21.13 Art 25(6) 4.14, 6.43, 21.13 Art 25(7) 4.60 Art 25(9)(c) 7.49 Art 25(10)(b) 2.28 Arts 27–28 19.117 Art 27 4.01, 4.71, 4.91, 15.106, 23.47 Art 27(1)(2) 7.67

Art 27(2) 7.70 Art 27(3) 7.69 Art 27(4) 7.65 Art 27(5)(2) 7.66 Art 27(6) 7.69 Art 27(9)(a) 7.49 Art 28 23.47 Art 28(2) 12.31 Art 28(3) 21.24 Art 29 6.35 Art 29(1) 23.43 Art 29(3) 4.60 Art 29(4) 19.12 Art 30 4.14, 4.61, 4.62 Art 30(1) 4.14, 4.68 Art 30(1), 2nd indent 23.34 Art 30(2) 4.14 Art 30(6) 21.24 Art 31 et seq 7.10 Art 33 13.48, 22.116, 23.44 Art 33(1) 13.16 Art 33(2) 13.16 Art 33(3) 13.16, 13.18 Art 33(3)(a) 13.17, 13.51 Art 33(3)(e) 13.32 Art 33(4) 13.32 Art 33(7) 13.47 Art 33(8) 13.19 Arts 34–35 19.47 Art 34 10.58, 10.78, 19.118, 23.47 Art 35 10.78 Art 35(8) 23.47 Arts 35–36 11.79 Arts 36–38 11.79 Art 37(1) 22.90 Art 37(2) 22.88 Art 38 22.90

Arts 39–43 1.27, 10.04, 10.45 Art 39 10.13, 10.33, 10.42, 10.48, 10.51, 10.55, 10.64, 10.68, 10.69, 10.70, 10.78, 19.47, 23.35 Art 39(1) 10.05, 10.09, 10.10, 10.45, 10.78 Art 39(2) 10.78 Art 39(2), opening words 10.46, 10.47 Art 39(2)(a) 10.47 Art 39(2)(b) 10.47 Art 39(2)(c) 10.47 Art 39(2)(d) 10.47 Art 39(2)(e) 10.47 Art 39(2)(f) 10.47 Art 39(3) 10.46 Art 40 10.46, 16.32, 16.84, 23.35 Art 41(1), opening words 10.47 Art 41(1)(b) 10.47 Art 41(2) 10.10, 10.52, 10.62 Art 41(2), 1st para 10.48 Art 41(2), 2nd para, first part of the sentence 10.47 Art 41(2), 2nd para, second part of the sentence 10.47 Art 42 10.08, 10.10, 10.68, 10.69, 10.71, 10.75, 10.78, 16.32, 23.35 Art 42, last sentence 10.69 Art 42(2)(f) 16.32 Art 43 19.50 Art 43(a) 10.49 Art 43(b) 10.49 Art 43(c) 10.49 Art 43(d) 10.49 Art 43(e) 10.49 Art 44 10.14, 19.47 Arts 44–56 19.05 Art 44(1) 11.17 Art 44(1)(1) 11.17 Art 44(2) 11.17 Art 44(5) 19.50 Art 45(1) 11.18 Art 45(2)(a) 11.18

Art 45(2)(b) 11.18 Art 45(2)(c) 11.18 Art 45(3) 11.18 Art 45(4) 11.18 Art 45(6) 11.18 Art 46 11.20 Art 46(1) 11.78 Art 46(2) 11.77 Art 47 11.76 Art 47(1) 11.21 Art 48 11.21, 11.27, 12.05, 17.20, 17.21 Art 48(1) 11.21, 17.34, 17.36 Art 48(2) 17.37 Art 48(5) 17.37 Art 48(6) 17.40 Art 48(8) 17.40 Art 48(9) 17.40, 17.56 Art 48(10) 17.50 Art 49 11.27, 17.20, 17.21, 17.48, 17.49, 17.50 Art 50 17.20, 17.21 Art 51 11.21 Art 51(2) 22.63 Art 52 11.21 Art 54 11.21 Art 53(2) 22.88 Art 53(3) 11.28 Art 53(4) 4.15 Art 55 22.90 Art 57 2.47, 16.21, 16.63, 16.87, 19.35 Art 57(1) 16.71 Art 57(3) 16.64 Art 57(4) 16.86 Art 57(8)–(10) 16.89 Art 57(12) 16.64, 16.71 Art 57(12)(f) 16.73 Art 57(13) 16.87 Art 58 2.47, 16.29

Art 59 10.11, 19.47 Art 59(1) 2.58 Art 59(2) 10.13 Art 60(2) 10.13 Art 62 19.50 Art 63 2.58 Art 64 2.58 Art 65 2.58, 19.50 Art 66 2.58 Arts 67–88 1.45, 19.07, 19.129 Art 67(1) 19.11, 19.50, 19.103, 19.104 Art 67(2) 19.19, 19.38, 19.54 Art 67(2), 1st subpara 19.12 Art 67(2), 2nd subpara 19.12 Art 67(2), 3rd subpara 19.13 Art 67(3) 19.14 Art 68 19.17, 19.132 Art 69 19.30, 19.47 Art 69(1) 19.28 Art 69(2) 4.02 Art 69(2), 1st para 19.30 Art 69(2), 2nd para 19.29 Art 69(2), 3rd para 19.29 Art 69(2), last para 20.04, 20.21, 20.29, 20.33, 20.36, 20.37 Art 69(2), in fine 9.45 Art 69(2)(o) 19.89 Art 69(2)(p) 16.82, 16.87, 16.88, 19.89 Art 69(2)(s) 5.41, 5.62 Art 69(2)(t) 5.41, 5.61 Arts 70–71 19.40 Art 70 19.46, 19.50, 19.85 Art 70(1), 1st subpara, 1st sentence 19.41 Art 70(1), 1st subpara, 2nd sentence 19.42 Art 70(1), 2nd subpara 19.43 Art 70(1), 3rd subpara 19.46 Art 70(2) 19.42

Art 70(3) 19.47 Art 70(3)(a)(x) 4.02 Art 70(3)–(5) 19.42, 19.47 Art 70(6) 19.48, 19.50 Art 70(6)(f) 19.52 Art 70(7) 19.53 Art 71 19.49, 19.105 Art 71(1), 1st subpara 19.58 Art 71(1), 2nd subpara 19.59 Art 71(1), 3rd subpara 19.59 Art 71(2) 19.60 Art 71(3) 19.69 Art 71(3), 1st subpara 19.61 Art 71(3), 2nd subpara, 1st sentence 19.69 Art 71(3), 2nd subpara, 2nd –4th sentences 19.71 Art 71(3), 2nd subpara, 3rd sentence 19.70 Art 71(3), 2nd subpara, 4th sentence 19.70 Art 71(3)–(6) 19.63 Art 71(4), 1st subpara, 1st sentence 19.66 Art 71(4), 1st subpara, 2nd sentence 19.66 Art 71(4), 2nd subpara, 1st sentence 19.72 Art 71(4), 2nd subpara, 2nd sentence 19.73 Art 71(5) 19.67, 19.73 Art 71(6) 19.68, 19.74 Art 71(7) 19.64 Art 71(7), 3rd subpara 19.65 Art 72 19.40 Art 72(1) 19.38, 19.54 Art 72(2) 19.55 Art 73(1) 19.76 Art 73(2) 19.77 Arts 74–75 19.105 Art 74 19.105 Art 74(1) 19.78 Art 74(2) 19.79

Art 75 19.105 Art 75(1)–2) 19.81 Art 75(3) 19.82 Art 76 19.105, 19.106, 19.124 Art 76(1) 19.19 Art 76(2) 19.20 Art 76(3) 19.20 Art 76(4)–(5) 19.22 Art 77 19.104 Art 77(1) 19.24 Art 77(2) 19.25 Art 78 19.39 Arts 79–88 19.132 Art 79 11.80, 16.33, 19.83 Art 79(1) 19.103 Art 79(1), 1st subpara 19.83 Art 79(1), 2nd subpara 19.85 Art 79(1), 3rd –5th subpara 19.86 Art 79(2) 11.80, 19.91 Art 79(3) 19.87 Art 79(4) 19.88 Art 79(5) 19.90 Art 79(6)–(7) 19.98 Art 79(8) 19.94 Art 79(9) 19.56 Art 79(9), 3rd subpara 19.96 Art 80(1) 19.99 Art 80(2) 19.100 Art 80(3) 19.100 Art 80(4), 3rd subpara 19.100 Art 81(1) 19.103 Art 81(2) 19.104 Art 81(3) 19.105 Art 81(4) 19.108 Art 81(4), 3rd subpara 19.108 Art 81(5) 19.106, 19.107

Art 82(1) 19.109 Art 82(2) 19.110 Art 83 19.111 Art 84(1)–(2) 19.112 Art 84(3) 19.113 Art 84(4) 19.115 Art 84(4), 3rd subpara 19.115 Art 85 19.116 Art 86(1) 19.118 Art 86(1)–(3) 19.122 Art 86(2) 19.120 Art 86(3) 19.121 Art 86(4) 19.122 Art 87 19.84 Art 88 19.104, 19.106 Art 88(1) 19.124 Art 88(2) 19.125 Art 91 21.06 Art 93(1), 1st subpara, 1st sentence 19.02 Art 93(1), 2nd subpara 19.02 Art 94 2.16 Art 95 2.49, 16.50 Art 96 19.02 Art 97 19.02 Annex I, Section A 1.09, 2.03, 2.04, 4.55, 11.23, 19.04, 23.39, 23.40 Annex I, Section A(1) 2.20 Annex I, Section A(2) 2.20 Annex I, Section A(4) 2.20 Annex I, Section A(5) 2.20 Annex I, Section A(9) 1.09 Annex I, Section B 8.62, 19.05 Annex I, Section B(2) 7.73 Annex I, Section B(5) 4.24, 4.37 Annex I, Section C 2.03, 2.43, 2.44, 2.46, 19.04, 19.34, 23.39 Annex I, Section C(1) 2.46, 23.39 Annex I, Section C(2) 2.46 Annex I, Section C(3) 2.46

Annex I, Section C(4) 2.46, 2.47 Annex I, Section C(4)–(10) 15.03 Annex I, Section C(5) 2.46, 2.48, 16.47 Annex I, Section C(6) 2.46, 2.49, 16.47, 16.49, 16.50 Annex I, Section C(7) 2.46, 2.50, 16.47 Annex I, Section C(8) 2.46 Annex I, Section C(9) 2.46 Annex I, Section C(10) 2.46, 2.47, 16.47, 16.60 Annex I, Section D 2.57 Annex I, Section II 10.06 2014/91/EU Directive of the European Parliament and of the Council of 23 July 2014 on the coordination of laws, regulations, and administrative provisions relating to undertakings for collective investment in transferable securities (UCITS) as regards depositary functions, remuneration policies, and sanctions, [2014] OJ L 257/186 (UCITS V) 1.48, 5.02, 16.39, 16.40, 16.85 Art 2(1), 2nd para 5.02 Art 49 16.85 2014/92/EU Directive of the European Parliament and of the Council of 23 July 2014 on the comparability of fees related to payment accounts, payment account switching and access to payment accounts with basic features, [2014] OJ L 257/214 (Payment Accounts Directive (PAD)) 21.53, 21.54 Art 4(3) 21.53 Art 5(2) 21.53 Art 8 21.53 2016/97/EU Directive of the European Parliament and of the Council of 20 January 2016 on insurance distribution (recast), [2016] OJ L 26/19 (Insurance Distribution Directive (IDD)) 1.48, 2.53, 2.54, 21.03, 21.04, 21.06, 21.07, 21.08, 21.09, 21.10, 21.11–21.13, 21.15, 21.16, 21.18– 21.26, 21.27, 21.29, 21.20, 21.32–21.35, 21.38, 21.50, 21.53, 21.54, 21.55, 21.59 recital 56 21.06 Art 17(1) 21.08, 21.16 Art 17(2) 21.08, 21.16 Art 18 21.10 Art 19 21.06, 21.10

Arts 20–22 21.10 Art 20 21.11 Art 20(1) 21.11 Art 21 21.53 Art 22(2) 21.10 Art 23 21.08 Art 24 21.09 Art 25 21.06, 21.08 Art 25(2) 21.16 Arts 27–28 21.13 Art 29(1) 21.13 Art 29(2) 21.14, 21.18 Art 29(3) 21.14, 21.18 Art 29(3), 4th para 21.19 Art 29(3), last para 21.21 Art 29(4) 21.19 Art 29(4), 2nd para 21.29 Art 30 21.13 Art 30(1) 21.11 Art 30(4) 21.13 Art 30(5) 21.13 2016/1034/EU Directive of the European Parliament and of the Council of 23 June 2016 amending Directive 2014/65/EU on markets in financial instruments, [2016] OJ L 175/8 1.02 Draft Commission Delegated Directive of 7 April 2016, supplementing Directive 2014/65/EU of the European Parliament and of the Council with regard to safeguarding of financial instruments and funds belonging to clients, product governance obligations and the rules applicable to the provision or reception of fees, commissions or any monetary or nonmonetary benefits, C(2016) 2031 final 5.07, 5.08, 5.32, 8.39, 8.46, 8.48, 8.52, 8.53, 8.63, 8.67 recital 15 5.07 recital 15 in fine 5.09 recital 17 5.06 recital 17, 2nd sentence 5.29 recital 18 5.09 recital 19 5.22

recital 20 5.36 recital 23 8.51 recital 24 8.41 recital 28 8.63 recital 29 8.67 Art 1(2) 5.05, 5.07, 5.10, 5.12, 5.13, 5.17, 5.19, 5.20, 5.22, 5.24, 5.26, 5.27, 5.28, 5.29, 5.30, 5.31, 5.32, 5.33, 5.34, 5.35, 5.37, 5.38, 5.39, 5.40 Art 9(1), 1st para 5.07 Art 9(1), 2nd para 5.07 Art 9(3) 5.24 Art 9(4) 5.24 Art 9(5) 5.30 Art 9(6) 5.12 Art 9(6) 2nd sentence 5.13 Art 9(6) 3rd sentence 5.13 Art 9(7) 5.17 Art 9(8) 5.08 Art 9(9) 1st sentence 5.23 Art 9(9) 2nd sentence 5.23 Art 9(9) 1st para, 2nd sentence 5.19 Art 9(9) 1st para, 3rd sentence 5.19 Art 9(9) 2nd para 5.20 Art 9(10) 5.24 Art 9(11) 5.24 Art 9(12) 5.24 Art 9(14) 5.32, 5.33 Art 9(15) 5.32 Art 9(15) 1st sentence 5.34 Art 9(15) 2nd sentence 5.34 Art 9(15) 3rd sentence 5.35 Art 9(15) 4th sentence 5.35 Art 10(1), 1st para 5.09 Art 10(1), 2nd para 5.28 Art 10(1), 3rd para 5.28

Art 10(2), 1st para 5.26 Art 10(2), 2nd para 5.27 Art 10(2), 3rd para 5.28, Art 10(2), 3rd para 1st sentence 5.29 Art 10(2), 3rd para 2nd sentence 5.29 Art 10(3) 5.40 Art 10(4) 5.32, 5.38 Art 10(5) 5.32, 5.39 Art 10(6) 5.17 Art 10(7) 5.31 Art 10(8) 5.12 Art 10(8) 2nd sentence 5.13 Art 10(8) 3rd sentence 5.13 Art 10(9) 5.37 Art 10(10), 1st sentence 5.10 Art 10(10)(a) 5.10 Art 10(10)(b) 5.10 Art 10(10)(c) 5.10 Art 10(13) 5.29 Art 11 8.43 Art 11(2) 8.49 Art 11(5) 8.55 Art 12 8.41 Art 12(1) 8.41 Art 12(3) 8.44 Art 13 8.65 Proposal for a Directive of the European Parliament and of the Council on insurance mediation (recast), COM(2012) 360 21.06 Proposal for a Directive of the European Parliament and of the Council on markets in financial instruments repealing Directive 2004/39/EC of the European Parliament and of the Council (Recast), COM/2011/0656 final 10.52, 11.15 recital 73 10.52 Art 4 10.52 Regulations

45/2001/EC Regulation of the European Parliament and of the Council of 18 December 2000 on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data, [2001] OJ L 8/1 19.39, 19.126 Art 9 19.124 1606/2002/EC Regulation of the European Parliament and of the Council of 19 July 2002 on the application of international accounting standards, [2002] OJ L 243/1 Art 3 16.71 1287/2006/EC Commission Regulation implementing Directive 2004/39/EC as regards record-keeping obligations for investment firms, transaction reporting, market transparency, admission of financial instruments to trading, and defined terms for the purposes of that Directive, [2006] OJ L 241/1 (MiFID I Implementing Regulation) 1.01, 9.01, 11.01, 11.15, 14.57, 14.68, 14.75, 20.01 Art 2(1) 16.45 Art 16 19.95 Art 18(1)(a) 14.53 Art 18(1)(b) 14.55 Art 18(2) 14.56 Art 20 14.57 Art 21 9.10 Art 22 14.50 Annex I 19.69, 19.72 Annex II 19.66 Annex II, Table 1 14.45 Annex III 19.72 1234/2007/EC Council Regulation of 22 October 2007 establishing a common organization of agricultural markets and on specific provisions for certain agricultural products, [2007] OJ L 299/1 (Single CMO Regulation) 5.56, 5.71, 16.32, 16.84 713/2009/EC Regulation of the European Parliament and of the Council of 13 July 2009 establishing an Agency for the Cooperation of Energy Regulators, [2009] OJ L 211/1 19.90 1060/2009/EC Regulation of the European Parliament and of the Council of 16 September 2009 on credit rating agencies, [2009] OJ L 302/1 (CRA Regulation) 1.24, 10.23, 10.25, 10.27

Art 5(6) 10.25 583/2010/EU Commission Regulation of 1 July 2010 implementing Directive 2009/65/EC of the European Parliament and of the Council as regards key investor information and conditions to be met when providing key investor information or the prospectus in a durable medium other than paper or by means of a website, [2010] OJ L 176/1 (UCITS Implementing Regulation) 21.28 Art 4(8) 21.36 Art 4(9) 21.29 Art 6 21.28 Art 8 21.36 Art 8(1) 21.36 Arts 10–13 21.29 Annex I 21.36 1092/2010/EU Regulation of the European Parliament and of the Council of 24 November 2010 on European Union macro-prudential oversight of the financial system, [2010] OJ L 331/1 19.106 1093/2010/EU Regulation of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Banking Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/78/EC, [2010] OJ L 331/12 (EBA Regulation) 5.64, 5.73 Art 9 5.64 Art 9(2) 5.64 Art 9(5) 5.65, 5.66, 5.73 Art 9(5), 3rd para 5.66 Art 18 5.73 1095/2010/EU Regulation of the European Parliament and of the Council establishing a European Supervisory Authority, [2010] OJ L 331/84 (ESMA Regulation) 5.64, 5.73, 12.36, 12.47, 19.83 Art 1(2) 19.110 Art 9 5.64 Art 9(2) 5.64 Art 9(5) 4.84, 5.65, 5.66, 5.73 Art 9(5), 3rd para 5.66 Arts 10–14 19.100 Art 10 1.02

Art 10(1) 5th subpara 19.100 Art 15 1.02, 17.24, 19.65, 19.96, 19.100, 19.108, 19.115 Art 15(1) 19.65, 19.100 Art 17 19.110 Art 18 5.73 Art 19 12.36, 19.109, 19.110, 19.111, 19.122 Art 21 19.100 Art 31(c) 19.109 Art 33 19.124 Arts 35–36 19.84 1227/2011/EU Regulation of the European Parliament and the Council of 25 October 2011 on wholesale energy market integrity and transparency, [2011] OJ L 326/1 (REMIT) 2.49, 16.05, 16.49 Art 2(4) 16.49 236/2012/EU Regulation of the European Parliament and of the Council of 14 March 2012 on short selling and certain aspects of credit default swaps, [2012] OJ L 86/1 (Short Selling Regulation) 12.43, 16.92, 22.75 Art 14(2) 22.76 648/2012/EU Regulation of the European Parliament and of the Council of 4 July 2012, on OTC derivatives, central counterparties and trade repositories, [2012] OJ L 201/1 (European Market Infrastructure Regulation (EMIR)) 1.24, 1.29, 1.39, 2.49, 10.16, 10.24, 10.25, 10.27, 11.01, 11.65, 11.73, 11.80, 11.82, 11.83, 11.85, 11.86, 12.06, 12.49, 15.01, 15.04, 15.13, 15.15, 15.16, 15.17, 15.18, 15.19, 15.23, 15.24, 15.25, 15.26, 15.31, 15.33, 15.40, 15.45, 15.46, 15.50, 15.51, 15.57, 15.74, 15.77, 15.78, 15.79, 15.82, 15.83, 15.103, 15.104, 15.106, 15.110, 16.03, 16.14, 16.25, 16.28, 16.35, 16.46, 16.47, 16.50, 16.67, 16.69, 16.71, 16.72, 16.73, 22.62, 22.63, 22.64, 22.65, 22.68, 22.69, 22.72, 22.73, 22.104, 22.105, 22.139 recital 23 22.68 recital 26 15.16 recital 34 11.31 recital 38 15.15 recital 52 11.83 recital 64 22.74 Title IV 10.24, 10.27 Title V 11.31, 11.59

Art 1(1) 16.67 Art 2(1) 19.22, 22.66 Art 2(3) 15.36 Art 2(6) 15.24 Art 2(7) 15.04 Art 2(8) 15.13, 16.67, 22.67 Art 2(9) 15.13 Art 2(12) 11.31 Art 2(14) 15.49 Art 3 15.15 Art 4(4) 15.14 Art 5(2) 15.33 Art 5(4) 15.18 Arts 7–8 11.79 Art 7 11.59 Art 10 22.64 Art 10(1)(b) 15.13 Art 10(3) 15.13, 16.25, 22.67 Art 16 22.66 Art 17(4) 15.83 Art 18 11.83 Art 25 15.104, 22.73 Art 25(1) 10.24 Art 25(6) 10.24 Art 34 22.63 Art 39 15.45, 15.76, 15.104, 22.74 Art 39(7) 22.74 Art 39(9) 15.46 Art 41 22.66 Art 42 22.66 Art 43 22.66 Art 45 22.66 Art 48 15.76, 15.104 Art 48(5) 15.47 Art 48(6) 15.47 Art 48(7) 15.47 Arts 51–54 11.79

Art 89 15.16 918/2012/EU Commission Delegated Regulation of 5 July 2012 supplementing Regulation (EU) No 236/2012 of the European Parliament and of the Council on short selling and certain aspects of credit default swaps with regard to definitions , the calculation of net short positions, covered sovereign credit default swaps, notification thresholds, liquidity thresholds for suspending restrictions, significant falls in the value of financial instruments and adverse events, [2012] OJ L 274/1 22.75 149/2013/EU Commission Delegated Regulation of 19 December 2012 supplementing Regulation (EU) No 648/2012 of the European Parliament and of the Council with regard to regulatory technical standards on indirect clearing arrangements, the clearing obligation, the public register, access to a trading venue, non-financial counterparties, and risk mitigation techniques for OTC derivatives contracts not cleared by a CCP, [2013] OJ L 52/11 (EMIR Regime (EMIR ICA RTS)) 15.79, 15.99, 15.103, 15.110 Chap II 15.82 Art 1(a) 15.80 Art 4(4) 15.79 Art 7 15.18 Art 10 15.13, 16.25, 16.68.16.71 Art 11 15.13, 16.25, 16.68 575/2013/EU Regulation of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012, [2013] OJ L 176/1 (Capital Requirements Regulation (CRR)) 1.26, 2.15, 2.16, 2.17, 2.19, 2.20–2.25, 3.58, 4.67, 10.18, 10.27, 10.34, 10.62, 12.06, 15.01, 15.110, 16.36, 16.37, 22.06, 22.39, 22.44, 22.50 Art 4(1) 2.16 Art 4(1)(b) 3.25 Art 4(1)(c) 3.25 Art 4(1)(1) 19.05, 22.27 Art 4(1)(2) 2.16, 2.20 Art 4(1)(2c) 2.22 Art 4(1)(3) 2.20, 3.25 Art 4(1)(4) 2.18 Art 4(1)(26) 19.17

Art 4(1)(36) 3.56 Art 4(2) 2.16, 3.25 Art 52 4.51 Art 92 2.22 Art 95 2.22 Art 96 2.22 Art 395(2) 22.39 Art 435 3.58 Art 450 3.58 Art 493 2.25 Art 493(1) 16.38 Art 498 2.25 Art 498(1) 16.38 1308/2013/EU Regulation of the European Parliament and of the Council of 17 December 2013 establishing a common organization of the markets in agricultural products and repealing Council Regulations (EEC) No 922/72, (EEC) No 234/79, (EC) No 1037/2001 and (EC) No 1234/2007, [2013] OJ L 347/671 19.98 596/2014/EU Regulation of the European Parliament and of the Council of 16 April 2014 on market abuse repealing Directive 2003/6/EC of the European Parliament and of the Council and Commission Directives 2003/124/EC, 2003/125/EC and 2004/72/EC, [2014] OJ L 173/1 (Market Abuse Regulation (MAR)) 1.33, 10.15, 10.35, 13.02, 13.24, 13.25, 13.27, 13.32, 13.33, 13.41, 13.45, 13.54, 15.01, 16.23, 16.31, 17.56, 19.33 recital 5 13.32, 13.33 Art 1 13.33 Art 2(1) lit. (b) 13.24 Art 7 13.31 Art 7(2) 13.28 Art 17 13.29 Art 17(1) 13.27, 13.31 Art 17(4) 13.28 Art 18(6) 13.25 Art 19(1) 13.27 Art 19(4) 13.27 Art 30(2)(j)(ii) 13.29

600/2014/EU Regulation of the European Parliament and of the Council of 15 May 2014 on Markets in Financial Instruments and amending Regulation (EU) No 648/2012, [2014] OJ L 173/84 (MiFIR) 1.02, 1.03, 1.07, 1.19, 1.24, 1.25, 1.26, 1.28, 1.30, 1.31, 1.32, 1.38, 1.40, 1.50, 1.56, 2.45, 2.49, 3.54, 4.51, 4.83, 5.01, 5.40, 5.42, 5.58, 5.65, 5.66, 5.68, 5.73, 7.08, 9.01, 9.45, 10.01–10.07, 10.09, 10.10, 10.14, 10.15, 10.19, 10.20– 10.23, 10.27, 10.32, 10.34, 10.41, 10.45, 10.62, 10.67, 10.70, 10.72, 10.74, 10.76–10.78, 11.01, 11.31, 11.73, 11.75, 11.81, 12.01–12.09, 12.12, 12.13–12.17, 12.19, 12.23, 12.29–12.31, 12.34, 12.36, 12.40, 12.42, 12.44, 12.45–12.51, 12.53, 12.55–12.57, 14.07–14.09, 14.60, 14.64, 14.69, 14.71, 14.77, 14.102, 14.106, 15.01, 15.03, 15.04, 15.10, 15.14, 15.15, 15.16, 15.26, 15.27, 15.33, 15.35, 15.36, 15.40, 15.50– 15.52, 15.57, 15.75, 15.77, 15.94, 15.104, 15.106, 15.111, 15.114, 15.117, 15.131, 15.132, 16.02–16.03, 16.14, 16.18, 16.25, 16.26, 16.31, 16.32, 16.42–16.43, 16.47, 16.71, 16.72, 16.84, 16.88, 16.89, 19.01, 19.03, 19.33, 19.34, 19.58, 19.75, 19.78, 19.79, 20.01, 20.04, 22.06, 22.63, 22.76, 22.139, 23.12, 23.34, 23.35 recital 3 11.16 recital 6 14.92 recital 7 11.22 recital 8 14.60, 15.06, 15.106, 22.65 recital 9 7.07, 9.16 recital 26 7.07, 15.06 recital 27 15.06 recital 28 14.61 recital 34 22.76 recital 37 15.33 recital 41 10.01, 10.02, 10.19, 10.22, 10.28 recital 43 10.67, 10.74 recitals 10.72 Title III 15.31 Title V 1.35, 11.30, 11.65, 15.01, 15.02, 15.06, 15.32, 15.49, 15.76, 15.105 Title VI 11.30, 11.65 Title VII, Chap 1 5.48 Art 1(1)(f) 10.04, 10.09, 22.63 Art 1(30) 2.47

Art 1(32) 15.04 Art 2(1)(28) 22.62 Art 2(1)(29) 15.03 Art 2(1)(30) 16.47, 19.31 Art 2(1)(34) 15.105 Art 2(1)(42) 10.04 Art 2(1)(47) 15.109, 15.111 Arts 3–22 12.02 Art 3(1) 12.18, 14.64 Art 3(2) 12.18 Art 4 12.19, 12.36, 12.37, 14.67 Art 4(1)(a) 14.67, 14.77 Art 4(1)(b) 14.69 Art 4(1)(b)(i) 14.77, 14.81 Art 4(1)(b)(ii) 14.73 Art 4(1)(c) 14.75 Art 4(1)(d) 14.71 Art 4(2)(a) 14.68 Art 4(2)(b) 14.68 Art 4(3)(a) 14.69 Art 4(3)(b) 14.69 Art 4(4) 12.37 Art 4(7) 12.37, 14.106 Art 5 12.20, 12.48, 14.78, 18.106, 18.110 Art 5(2) 14.82 Art 5(3) 14.82 Art 5(7) 14.84 Art 5(8) 14.85 Art 6 2.57, 12.21 Art 7 2.57, 12.21, 12.38 Art 7(1), 3rd sentence 19.47 Art 8 12.22, 12.23, 15.106 Art 8(1) 14.64 Art 9 12.23, 12.39 Art 9(4) 12.24 Art 9(5) 12.32 Art 10 2.57, 12.25, 12.39, 15.105, 15.106, 15.125

Art 11 12.25 Art 11(1) 19.47 Art 12 2.57 Art 13 2.57 Arts 14–15 12.27 Arts 14–26 19.117 Art 14 22.63, 23.47 Art 14(1) 9.10, 12.28 Art 14(2) 12.28 Art 14(3) 12.28 Art 14(4) 12.28 Art 14(5) 12.28 Art 15(1) 12.28 Art 15(2) 12.29 Art 15(3) 12.29 Art 17 12.30 Art 18 12.27, 12.33, 15.106 Art 18(1) 9.10, 12.32 Art 18(3) 12.32 Art 18(5) 12.32 Art 18(6) 12.33 Art 18(7) 12.33 Art 18(9) 12.32 Art 20 2.57, 12.26, 22.62 Art 20(1) 9.10 Art 21 2.57, 12.26, 15.106 Art 21(1) 9.10 Art 23 14.93 Art 23(1) 14.93 Art 23(2) 14.93 Art 25 15.126 Art 26 22.62 Arts 28–34 11.30, 12.15 Art 28 15.06, 15.106, 16.26 Art 28(1) 10.16, 15.15, 15.16 Art 28(2) 15.14, 15.28 Art 28(4) 10.15, 10.16

Art 29(1) 15.32 Art 29(2) 15.49, 15.57 Art 30 15.76, 15.103 Art 30(1) 15.104 Art 31 15.105, 22.65 Art 31(2) 15.117, 15.124 Art 31(3) 15.126 Art 31(4) 15.117 Art 31(4)(a) 15.117 Art 31(4)(b) 15.117 Art 32 10.16, 15.18, 16.26, 22.63 Art 32(1) 15.17, 15.27 Art 32(1)(b) 15.26 Art 32(2) 15.18, 15.21, 15.25 Art 32(2)(b) 15.20 Art 32(3) 15.20 Art 32(4) 15.25, 15.31, 22.63 Art 32(5) 15.27 Art 33 15.02 Art 34 10.16, 15.29, 16.26 Arts 35–36 10.15 Art 35 15.34 Art 35(1) 11.30 Art 36 15.34 Art 36(1) 11.30 Art 37 10.15 Art 37(1) 10.22 Art 38 10.15 Art 38(1) 10.15 Art 38(2) 10.15 Art 38(3) 10.15 Arts 39–43 5.41 Arts 40–42 19.35 Art 40 4.84, 5.62, 5.65, 16.32 Art 40(1) 5.66 Art 40(2), 1st para 5.67 Art 40(2), 2nd para 5.66

Art 40(2)(c) 4.85 Art 40(3), 1st para 5.71 Art 40(3), 2nd para 5.72 Art 40(4) 5.66 Art 40(5) 5.66 Art 40(6) 5.66 Art 40(7) 5.70 Art 40(8) 5.68 Art 40(8)(b) 5.69 Art 41 5.62, 5.65 Art 41(1) 5.66 Art 41(2), 1st para 5.67 Art 41(2), 2nd para 5.66 Art 41(3), 1st para 5.71 Art 41(3), 2nd para 5.72 Art 41(4) 5.66 Art 41(5) 5.66 Art 41(6) 5.66 Art 41(7) 5.70 Art 41(8) 5.68 Art 41(8)(b) 5.69 Art 42 4.84, 5.62, 16.84, 23.48 Art 42(1) 5.42 Art 42(2) 2nd para 5.42 Art 42(2) 3rd para 5.42 Art 42(2)(a) 5.45 Art 42(2)(b) 4.85, 5.52 Art 42(2)(c) 5.53, 16.84 Art 42(2)(d) 5.54, 16.84 Art 42(2)(e) 5.55 Art 42(2)(f) 5.56 Art 42(3)(a) 5.57 Art 42(3)(b) 5.57 Art 42(3)(c) 5.57 Art 42(4) 5.58 Art 42(5) 5.43

Art 42(6) 5.44 Art 42(7) 5.46 Art 42(7)(d) 5.69 Art 43(1) 5.59 Art 43(2) 5.60, 5.72 Art 43(3) 5.60 Art 45(1) 16.88 Arts 46–49 1.26, 10.04, 10.17 Art 46 10.66, 10.67 Art 46(1) 10.05, 10.09, 10.10, 10.29, 10.33, 10.40, 10.60, 10.78 Art 46(2) 10.29 Art 46(2)(b) 10.29 Art 46(2)(c) 10.29 Art 46(3) 10.34, 10.62 Art 46(4) 10.17, 10.78 Art 46(4), 1st para 10.29 Art 46(4), 3rd para 10.29 Art 46(4), 5th para 10.67 Art 46(4), last para 10.40, 10.41 Art 46(5), 1st para 10.34 Art 46(5), 2nd para 10.34 Art 46(5), 3rd para 10.08, 10.10, 10.65, 10.66, 10.67, 10.69, 10.75, 10.78 Art 46(6) 10.34, 10.44 Art 46(7) 10.29 Art 47 10.20, 10.23, 10.27, 10.29 Art 47(1) 10.18, 10.22 Art 47(2) 10.29 Art 47(3) 10.33 10.57, 10.58, 10.60, 10.62, 10.63, 10.64, 10.78 Art 47(3), 1st para 10.58 Art 47(3), 1st para, last sentence 10.58 Art 47(3), 2nd para 10.59 Art 47(4) 10.40 Art 49(1)(a) 10.37 Art 49(1)(b) 10.37 Art 49(1)(c) 10.37 Art 49(1)(d) 10.37

Art 49(2) 10.38 Art 49(3) 10.39, 10.41, 10.63 Art 52 12.05 Art 54(1) 10.36 Art 55 15.50 Art 55, first subpara 19.02 Art 55, last subpara 19.02 806/2014/EU Regulation of the European Parliament and of the Council of 15 July 2014 establishing uniform rules and a uniform procedure for the resolution of credit institutions and certain investment firms in the framework of a Single Resolution Mechanism and a Single Resolution Fund and amending Regulation (EU) No 1093/2010, [2014] OJ L 225/1 Art 18(1)(c) 22.27 Art 27 4.34 909/2014/EU Regulation of 23 July 2014 on improving securities settlement in the EU and on central securities depositories and amending Directives 98/26/EC and 2014/65 and Regulation (EU) No 236/2012, [2014] OJ L 257/1 (CSDR) 22.91, 22.92, 22.93, 22.96, 22.97, 22.139 recital 5 22.92 Art 2(1)(1) 19.22 Art 12(1) 22.94 Art 25 22.98 Art 49 22.95 Art 54 22.98 Arts 59–60 22.98 1187/2014/EU Commission Delegated Regulation of 2 October 2014 supplementing Regulation (EU) No 575/2013 of the European Parliament and the Council as regards regulatory technical standards for determining the overall exposure to the client or a group of the connected clients in respect of transactions with underlying assets, [2014] OJ L 324/1 22.38 1286/2014/EU Regulation of the European Parliament and of the Council of 26 November 2014 on key information documents for packaged retail and insurance-based investment products, [2014] OJ L 352/1 (PRIIPs Regulation) 1.48, 7.80, 10.09, 21.03, 21.24, 21.27, 21.28, 21.30, 21.33, 21.35, 21.37, 21.49, 21.50, 21.61, 22.114 recital 35 21.49 Art 2 10.09

Art 6(4) 21.24, 21.28 Art 8(3)(d) 21.36 Art 8(3)(f) 21.29 Art 13 21.28 Art 32(1) 21.49 Art 33(1) 21.49 2205/2015/EU Commission Delegated Regulation of 6 August 2015 supplementing Regulation (EU) No 648/2012 of the European Parliament and of the Council with regard to regulatory technical standards on the clearing obligation, [2015] OJ L 314/13 15.33 1033/2016/EU Regulation of the European Parliament and of the Council of 23 June 2016 amending Regulation (EU) No 600/2014 on markets in financial instruments, Regulation (EU) No 596/2014 on market abuse and Regulation (EU) No 909/2014 on improving securities settlement in the European Union and on central securities depositories, [2016] OJ L 175/ 1 1.02 Draft Commission Delegated Regulation (EU) of 25 April 2016 supplementing Directive 2014/65/EU of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive, C(2016) 2398 final, 25 April 2016 3.03, 3.28, 3.50, 3.52, 3.53, 3.54, 17.21, 21.08, 21.25, 21.29, 21.31, 21.38 Arts 12–17 15.31 Art 20 17.29 Art 31(1), 1st sentence 20.37 Arts 46–50 21.29 Art 46 21.08 Arts 47–50 21.08 Art 48(2)(c) 21.43 Art 50(1) 21.32 Art 50(4) 21.31 Art 50(5) 21.30, 21.32 Art 50(6) 21.30 Draft Commission Delegated Regulation (EU) of 18 May 2016 supplementing Regulation (EU) No 600/2014 of the European Parliament and of the Council with regard to definitions, transparency, portfolio compression and supervisory measures on product intervention and

positions, C(2016) 2860 final, 18 May 2016 (Draft Commission Delegated Regulation) 2.28, 3.50, 4.37, 5.41, 5.47, 5.48, 5.49, 5.50, 5.51, 9.10 recital 15 4.24, 4.38 recital 16 4.38, 15.117 recital 17 4.38 recital 18 5.51 recital 19 5.50 recital 48 4.91 recital 103 4.71, 9.09 recital 114 13.23 Art 9 4.24 Art 12 9.10 Art 13 9.10 Art 14 9.10 Art 15 9.10 Art 16 9.10 Art 17(2) 15.123 Art 17(3) 15.123 Art 17(4) 15.119 Art 17(5) 15.121 Art 17(6) 15.122 Art 18(1) 15.124 Art 18(2) 15.125 Arts 19–21 5.41 Art 19(1) 2nd para 5.50 Art 19(2) 1st para in fine 5.59, 5.68 Art 20(1) 2nd para 5.50 Art 20(2) 1st para in fine 5.59, 5.68 Art 21 3.50 Art 21(1) 1st para 5.47 Art 21(1) 1st para in fine 5.48 Art 21(1) 2nd para 5.50 Art 21(2)(a)–(v) 5.47, 5.50 Art 22 3.50, 3.53 Art 22(2) 3.53

Art 22(3) 3.53 Art 23 3.50, 3.52 Art 24 3.50, 3.52 Art 25 3.50 Art 27 3.45 Arts 33–35 3.54 Art 41(2) 4.68 Art 41(4) 4.67 Art 64(1), in fine 9.05, 9.09 Art 77(1) 13.17 Art 77(2) 13.01 Art 78(2)(c) 13.21 Art 78(2)(d) 13.22 Art 78(2)(e) 13.22 Art 78(2)(g) 13.23 Art 78(2)(h) 13.41 Annex 15.117 Draft Commission Delegated Regulation (EU) of 26 May 2016 supplementing Regulation (EU) No 600/2014 of the European Parliament and of the Council on markets in financial instruments with regard to regulatory technical standards on criteria for determining whether derivatives subject to the clearing obligation should be subject to the trading obligation (Draft Commission Delegated Regulation on the Trading Obligation) 15.21 recital 3 15.22 recital 10 15.30 Draft Commission Delegated Regulation (EU) of 13 June 2016 supplementing Regulation (EU) No 600/2014 of the European Parliament and of the Council on markets in financial instruments with regard to regulatory technical standards on the direct, substantial and foreseeable effect of derivative contracts within the Union and the prevention of the evasion rules and obligations recital 6 15.14 Art 2 5.14 Draft Commission Delegated Regulation (EU) of 19 June 2016 supplementing Regulation (EU) No 600/2014 of the European Parliament and of the Council with regard to regulatory technical standards

specifying the obligation to clear derivatives traded on regulated markets and timing of acceptance for clearing (Draft Commission Delegated Regulation on STP) recital 1 15.50 recital 9 15.68 recital 10 15.71 recital 11 15.72 recital 12 15.73 Art 1 15.58 Art 2(1) 15.63 Art 2(2) 15.59 Art 2(3)(a) 15.59 Art 2(3)(b) 15.59 Art 2(4)(a) 15.59 Art 2(4)(b) 15.59 Art 3(2) 15.60 Art 3(3) 15.60 Art 3(4) 15.61, 15.62 Art 4(1) 15.64 Art 4(2) 15.66 Art 4(3) 15.68 Art 4(4) 15.70 Art 4(5) 15.68 Art 5(1) 15.71 Art 5(2) 15.72 Art 5(3) 15.73 Draft Commission Delegated Regulation on Algorithmic Trading 15.70 Draft MIFIR ICA RTS 15.79, 15.81, 15.84, 15.97, 15.99, 15.101, 15.103, 15.104 recital 1 15.77 recital 3 15.83 recital 4 15.87 recital 6 15.87 recital 17 15.93 Art 1(1) 15.80 Art 2(1) 15.83 Art 3(2) 15.86, 15.92

Art 3(3) 15.83 Art 4(2) 15.85 Art 4(3) 15.92 Art 4(4) 15.86 Art 4(5) 15.89 Art 4(6) 15.89 Art 4(7) 15.79, 15.90 Art 4(8) 15.83, 15.93 Art 5(1) 15.85 Art 5(2) 15.88 Art 5(4) 15.92 Art 5(5) 15.86 Art 6(1) 15.95 Art 6(1)(c) 15.95 Art 6(2) 15.95 Art 6(2)(c) 15.95 Art 6(3) 15.94, 15.95 Art 6(3)(c) 15.95 Art 6(3)(d) 15.95 Art 6(4) 15.96 Art 6(4)(b) 15.82 Art 6(5) 15.96 Art 6(5)(b) 15.82 Art 6(6) 15.96 Art 6(6)(b) 15.82 Proposal for a Regulation laying down common rules on securitization and creating a European framework for simple, transparent and standardized securitization and amending Directives 2009/65/EC, 2009/138/EC, 2011/61/EU and Regulations (EC) No 1060/2009 and (EU) 648/2012, COM (2015) 472 final 22.49, 22.50 recital 11 22.56 Art 2(12) 22.55 Art 3(1) 22.55 Art 3(3) 22.50, 22.55 Art 4(2) 22.52 Art 4(4) 22.55 Art 4(5) 22.51

Art 5 22.53 Arts 8–10 22.52 Art 10(2) 22.57 Art 14(4) 22.58 Proposal for a Regulation of the European Parliament and of the Council on markets in financial instruments, SEC(2011) 1227 final 11.15 Proposal for a Regulation of the European Parliament and of the Council on the prospectus to be published when securities are offered to the public or admitted to trading, COM (2015) 583 final recital 25 21.33, 22.111 Art 7 21.33 Proposal for a Regulation on Money Markets Funds, COM (2013) 615 final 22.83 Proposal for a Regulation on reporting and transparency of securities financing transactions, COM (2014) 040 final 22.101, 22.103, 22.104, 22.108 Art 2(1) 22.104 Proposal for a Regulation on structural measures improving the resilience of EU credit institutions, COM (2014) 43 final 22.101 NATIONAL LEGISLATION France

Code monétaire et financier Art 421-16-2 16.82 Art 547-1 et seq 23.22 Art L547-3 23.23 Art L547IV 23.23 Art L547-9 23.26 Decree no. 2014-1053 of 16 September 2014 implementing Ordinance no. 2014-559 23.22 Ordinance no. 2014-559 of 30 May 2014 23.22 Separation and Regulation of Banking Activities Act L.2013-672 of July 2013 16.82 Art 580-1 16.82 Art 580-2 16.82 Germany

Banking Act (Kreditwesengesetz–KWG) 23.24 § 1(1a), sentence 2 no 2 23.24 § 2.6 23.24, 23.25 § 32.1 23.24 BGB (Bürgerliches Gesetzbuch) §181 9.11, 9.17 Commercial Code (Wetboek van Koophandel) Arts 76–85a 9.07 HGB (Handelsgesetzbuch) §§ 383–406 9.07 § 400 9.13 Small Investor Protection Act (Kleinanlegerschutzgesetz) of 23 April 2015 23.24 Stock Corporation Act (Aktiengesetz) s 76 3.33 s 77 3.33 s 91 3.33 s 91(1) 3.33 s 91(2) 3.33, 3.64 s 93 3.33 Trade, Commerce and Industry Regulation Act (Gewerbeordnung–GewO) s 34c 23.13, 23.25 s 34f 23.24 Italy

Delibera Consob no. 18592/2013 (Consob Crowdfunding Regulation) 23.22 Art 13(3) 23.27 Art 13(5) 23.27 Art 14 23.27 Art 15 23.27 Art 15(1)(g) 23.27 Art 15(2) 23.32 Art 17(3) 23.22 Arts 24–25 23.27 Art 24(1) 23.27 Art 24(1)(a) 23.33 Art 24(1)(b) 23.27

Art 24(2) 23.22 Annex 3 23.27 Delibera Consob no. 19520/2016 (Consob Regulation) 23.22 Art 13(5-bis) 23.22 Art 13(5-ter) 23.22 Art 17(2)–(3) 23.22 Art 17(4) 23.22 Art 24(2-bis) 23.22 TUF (Testo Unico della Finanza) Art 50-quinquies(3) 23.22 Luxembourg

Civil Code Art 6 20.22 Netherlands

Bulletin of Acts and Decrees 8.21, 8.23, 8.26 Decree implementing the Pensions Act and the Occupational Pension Scheme (Obligatory Membership) Act Chap 4 20.27 Art 13(2)(e) 2.27 Dutch Civil Code (DCC) Art 6:2 20.15, 20.16, 20.17, 20.18 Art 6:2(1) 20.15 Art 6:2(2) 20.15 Art 6:233(a) 9.37 Art 6:235(1) 9.37 Art 6:248 20.15, 20.16, 20.17 Art 6:248(2) 9.37 Art 7:401 9.37 Art 7:416 9.14 Art 7:416(1) 9.14 Art 7:416(2) 9.14 Art 7:416(3) 9.14 Art 7:417 9.18 Art 7:417(1) 9.18 Art 7:417(2) 9.18

Dutch Draft Legislative Proposal for Implementation of MiFID, p. 11 10.42, 10.48 s 1:19c 10.67 Dutch Market Conduct Decree Art 168a 8.19 Exemption Regulation FSA (Vrijstellingsregeling Wft) Art 10 10.42, 10.48 Art 18 10.42 Art 35 10.42 Financial Supervision Act (FSA) (Wet op het financieel toezicht, Wft) 8.24 Art 2:96 10.42 Art 3:5 10.42 Art 3:6 10.42 Art 3:7 10.42 Art 4:20(6) 20.20 Occupational Pension Scheme (Obligatory Membership) Act (Wet verplichte beroepspensioenregeling) 20.27 s 43(1) 20.27 Pensions Act (Pensioenwet) 20.27 s 34(1) 20.27 Spain

Ley de fomento de la financiación empresarial (28 April 2015 No 5) 23.21 Art 46 23.21, 23.49 Art 47 23.49 Art 48 23.21, 23.49 Art 51 23.21 Art 51(1) 23.21 Art 52 23.21 Arts 55–56 23.21 Art 60 23.28 Art 61 23.28 Art 62 23.28 Art 66 23.21 Art 68 23.30 Art 71 23.28 Art 81 23.30

Arts 89–93 23.21 United Kingdom

Financial Services Act 1986 4.05 Financial Services and Markets Act 2000 (FSMA) 13.04 Pt X, Chap I 9.33 s 138D 9.33 s 150 9.33, 9.34 Financial Services and Markets Act 2000 (Regulated Activities) Order 2001 Art 19 16.56 Art 53 4.39 United States

Commodity Exchange Act (CEA) 1936 16.77 s 5(d), core principle 5 16.77 Dodd-Frank Wall Street Reform and Consumer Financial Protection Act 1.36, 4.75, 11.67, 16.78, 16.91 Title VII 16.07, 16.78 Title VII, Pt II 15.18 s 723(8) 15.18 s 733 11.67 s 737 16.78 s 913 4.75 s 913(b) 4.80 s 913(f) 4.80 s 913(g) 4.79, 4.80 Investment Advisers Act 1940 4.74 s 206 4.76 s 206(3) 4.79 s 211(g) 4.79 Securities and Exchange Act Section 6, 15 USC § 78f (1934) 18.55 § 80b-2(a)(11)(C) 4.74 Securities Exchange Act s 10(b) 18.10 r 10b-5 18.10 s 15(k) 4.79, 4.80

LIST OF ABBREVIATIONS

ABCP ABS ABSPP ACER ACPR ADT AFM AFME AGR AIF AIFM AIFMD AIM AMF APA APM ARM AT ATS

Asset-Backed Commercial Paper Asset-Backed Securities Asset-Backed Securities Purchase Programme Agency for the Cooperation of Energy Regulators Autorité de contrôle prudentiel et de résolution Average Daily Turnover Authority for the Financial Markets Association for Financial Markets in Europe Average Growth Rate Alternative Investment Fund Alternative Investment Fund Manager Alternative Investment Fund Managers Directive Alternative Investment Market Autorité des Marchés Financiers Approved Publication Arrangement Adverse Price Movement Approved Reporting Mechanism Algorithmic Trading Alternative Trading System

Banks RWD BCBS BDCN BGB BIS BRRD BTSs

Directive 2001/24/EC on the Reorganisation and Winding Up of Credit Institutions Basel Committee on Banking Supervision Broker/Dealer Crossing Network Bürgerliches Gesetzbuch—German Civil Code Bank for International Settlements Banking Recovery and Resolution Directive Binding Technical Standards

CB CCP CDS CEA CEMA CEO CESR CFTC CIP CLOB CMU CN CNAV CNVM COBS COFIA CP CRA CRD I CRD IV CRR CSD CSDR CSMAD CTP

Clearing Broker Central Counterparties Credit Default Swaps Commodity Exchange Act Committee for Economic and Markets Analysis Chief Executive Officer Committee of European Securities Regulators Commodity Futures Trading Commission Conseiller en Investissement Participatif—French crowdfunding investment advisor Central Limit Order Book Capital Market Union Crossing Network Constant Net Asset Value Spanish Securities Commission Conduct of Business Sourcebook Classes of Financial Instruments Approach Consultation Paper Credit Rating Agency Capital Requirements Directive I Capital Requirements Directive IV Capital Requirements Regulation Central Securities Depository Central Securities Depositories Regulation Criminal Sanctions for Market Abuse Directive Consolidated Tape Provider

DCB DCC DEA DFSA DGS DJIA DM DMA DP DVC

Dutch Central Bank Dutch Civil Code (Burgerlijk Wetboek) Direct Electronic Access Dutch Financial Supervision Act Deposit Guarantee Scheme Dow Jones Industrial Average Dealer Market Direct Market Access Discussion Paper Double Volume Cap

DVP

Delivery versus Payment

EB EBA EBBO ECB ECGI ECJ EEA EFAMA EIOPA ELA EMIR EP ESA ESCB ESMA ESRB ETD ETF ETP EU

Executing Broker European Banking Authority European Best Bid or Offer European Central Bank European Corporate Governance Institute European Court of Justice European Economic Area European Fund and Asset Management Association European Insurance and Occupational Pensions Authority Emergency Liquidity Assistance European Market Infrastructure Regulation European Parliament European Supervisory Authority European System of Central Banks European Securities and Markets Authority European Systemic Risk Board Exchange-Traded Derivative Exchange Trade Fund Exchange Traded Product European Union

FATF FC FCA FINRA FMI FR FSA FSB FSMA FSMA 2000 FVC

Financial Action Task Force Financial Counterparty Financial Conduct Authority Financial Industry Regulatory Authority Financial Markets Infrastructure Final Report Financial Services Authority Financial Stability Board Financial Services and Markets Authority Financial Services and Markets Act 2000

GDP

Gross Domestic Product

Financial Vehicle Corporation

GEL GewO GOICA

Grant Estates Limited Gewerbeordnung—German Trade, Commerce, and Industry Regulation Act Gross Omnibus Indirect Client Account

HFT HGB

High-Frequency Trading/Trader Handelsgesetzbuch—German Commercial Code

IDD IEX IF IFP

Insurance Distribution Directive Investors Exchange Investment Firm intermédiaire en financement participative— French crowdlending investment advisor

IFRS IM IMD IMF IOSCO IPO IRSA ISA ISD 1993 ISDA ISICA ITS

International Financial Reporting Standards Initial Margin Insurance Mediation Directive International Monetary Fund International Organization of Securities Commissions Initial Public Offering Interest Rate Swap Agreement Individually Segregated Account Investment Services Directive 1993

KID KII KOM KWG KYC

Key Investor Document Key Investor Information Key Operating Milestone Kreditwesengesetz—German Banking Act Know Your Customer

LIS LSE LP

Large-in-Scale London Stock Exchange Liquidity Premium

International Swap and Derivatives Association Individually Segregated Indirect Client Account Implementing Technical Standards

MAC MAD MAR MCD MI MiFID MiFIR MMF MRM MTF

Mercato Alternativo del Capitale Market Abuse Directive Market Abuse Regulation Mortgage Credit Directive Market Impact Markets in Financial Instruments Directive Markets in Financial Instruments Regulation Money Market Funds Market Risk Measure Multilateral Trading Facility

NASD NBB NBO NCA NFC NMS NMS NT

National Association of Securities Dealers National Best Bid National Best Offer National Competent Authority Non-Financial Counterparty Normal Market Size National Market System Negotiated Trade / Negotiated Transaction

OAM OICA OFIs OMF OSA OTC OTF OTR

Officially Appointed Mechanism Omnibus Indirect Client Account Other Financial Institutions Order-Management Facility Omnibus Segregated Account Over the Counter Organized Trading Facility Order-to-Transaction Ratio

P2B P2P PAD PBBO PD PRA PRIIP PSI

Person-to-Business Peer-to-Peer Payment Accounts Directive Primary Best Bid and Offer Prospectus Directive Prudential Regulation Authority Packaged Retail and Insurance-based Investment Product Prestataire en Services d’investissement— French investment

firms RBS RDR REMIT RFQ RIE RMs RP RTS SA SDP SFTs SEC SEF SFD SFT SI SIB SIM

Royal Bank of Scotland Retail Distribution Review Regulation on Wholesale Energy Markets Integrity and Transparency Request for Quote Recognized Investment Exchange Regulated Markets Reference Price Regulatory Technical Standard

SMEs SMSG SPV SRM SRRI SSM SSS STP STS

Sponsored Access Single-Dealer Platform Securities Financing Transactions Securities and Exchange Commission Swap Execution Facility Settlement Finality Directive Securities Financing Transactions Systematic Internalizer Securities and Investments Board Società di investimento mobiliare—Italian banks and investment firms Small and Medium-Sized Enterprises Securities and Markets Stakeholder Group Special Purpose Vehicle Single Resolution Mechanism Synthetic Risk and Reward Indicator Single Supervisory Mechanism Systems of Settlement of Securities Straight-Through-Processing Simple, Transparent and Standardized Securitization

T2S TD TEU TFEU

Target 2 Securities Transparency Directive Treaty on European Union Treaty on the Functioning of the European Union

TR

Trade Repository

UCITS US

Units for Collective Investment in Transferable Securities United States

VC VDO VM VNAV

Venture Capitalist Volume Discovery Order Variation Margin Variable Net Asset Value

Wft

Wet op het financieel toezicht (Dutch Act on Financial Supervision) Wertpapierhandelsgesetz (German Securities Trading Act) Written Statement on Suitability

WpHG WSS

AUTHOR BIOGRAPHIES

Editors Danny Busch is Professor of Financial Law and Director of the Institute for Financial Law, Radboud University, Nijmegen, the Netherlands. He is Visiting Professor at Università Cattolica del Sacro Cuore di Milano, Visiting Professor at Università degli Studi di Genova and a Member of the Dutch Banking Disciplinary Committee (Tuchtcommissie Banken). He is extensively engaged in the provision of training to attorneys-at-law, financial regulators and financial professionals. He is author of many articles in the field of financial and commercial law, and editor of several books, including A Bank’s Duty of Care (with C.C. van Dam), Hart, 2017; Agency Law in Commercial Practice (with L. J. Macgregor and P. Watts), OUP, 2016; European Banking Union (with G. Ferrarini), OUP, 2015; Alternative Investment Funds in Europe (with L.D. van Setten), OUP, 2014; and Liability of Asset Managers (with D.A. DeMott), OUP 2012. After having graduated with highest honours in Dutch law from Utrecht University in 1997, he was awarded the degree of Magister Juris in European and Comparative Law by the University of Oxford (St. John’s College) in 1998. From 1998 until 2001 he held the position of lecturer and researcher at the Molengraaff Institute of Private Law in Utrecht. In 2002 he defended his PhD in Utrecht (Indirect Representation in European Contract Law, KLI, 2005). From 2002 until 2010 he was an attorney-at-law (advocaat) with the leading Dutch international law firm De Brauw Blackstone Westbroek in Amsterdam where he practised banking and securities law (both the private law and regulatory aspects). Guido Ferrarini is Professor of Business Law, University of Genoa, Italy, and Professor of Governance of Financial Institutions, University of

Nijmegen, the Netherlands. He holds a J. D. (University of Genoa, 1972), an LL.M. (Yale Law School, 1978) and a Dr. jur. (h.c., Ghent University, 2009). He is a founder and fellow of the European Corporate Governance Institute (ECGI), Brussels. He was a member of the Board of Trustees, International Accounting Standards Committee (IASC), London, an independent director at several Italian blue-chip companies, and Chairman of EuroTLX (an Italian MTF). He was an adviser to the Draghi Commission on Financial Markets Law Reform, to Consob (the Italian Securities Commission), and to the Corporate Governance Committee of the Italian Stock Exchange. He has held Visiting Professor positions at several universities in Europe (Bonn, Católica Lisbon, Frankfurt, Ghent, Hamburg, LSE, UCL, Tilburg, and Duisenberg) and the US (Columbia, NYU, and Stanford), teaching courses on comparative corporate governance and financial regulation. He is author of many articles in the fields of financial law, corporate law, and business law, and editor of several books, including Financial Regulation and Supervision: A Post-Crisis Analysis (with E. Wymeersch and K. Hopt) OUP, 2012, Boards and Shareholders in European Listed Companies (with M. Belcredi) CUP, 2013, and European Banking Union (with D. Busch) OUP, 2015. He is presently board chair of a securities firm, board member of a private bank, and adviser on corporate law and financial regulation.

Contributors Antonella Sciarrone Alibrandi is Professor of Law at the Università Cattolica del Sacro Cuore di Milano. Aron Berket is an attorney-at-law at Simmons & Simmons, Amsterdam. Jens-Hinrich Binder is Professor of Law and holds a Chair in Private, Commercial, and Corporate Law at Eberhard-Karls-University, Tuebingen, Germany. Veerle Colaert is Professor of Financial Law at KU Leuven University, a member of the advisory Securities and Markets Stakeholder Group of

ESMA and a member of the Sanctions Commission of the Belgian Financial Services and Markets Authority (FSMA).. Pierre-Henri Conac is Professor of Commercial Law at the University of Luxembourg and ECGI Research Associate. Carmine di Noia is a Commissioner at CONSOB (Italian Securities and Exchange Commission). The chapter was written when the author was deputy director General at ASSONIME and member of the Securities and Markets Stakeholder Group at ESMA. Luca Enriques is Allen & Overy Professor of Corporate Law at the University of Oxford and ECGI Research Fellow. Merritt B. Fox is Michael E. Patterson Professor of Law and NASDAQ Professor of the Law and Economics of Capital Markets at Columbia Law School. Matteo Gargantini is a Senior Research Fellow at the Max Planck Institute, Luxembourg. Paolo Giudici is Professor of Business Law at the Free University of Bozen-Bolzano and ECGI Research Associate. Peter Gomber is Professor of e-Finance and Co-Chair of the E-Finance Lab, Research Center SAFE, at Goethe University, Frankfurt. Christos V. Gortsos is Professor of Public Economic Law at the Law School of the National and Kapodistrian University of Athens and visiting Professor at the European Institute of the University of Saarland.. Edoardo Grossule is a Researcher at Università Cattolica del Sacro Cuore di Milano. Stefan Grundmann holds the Chair of Transnational Private Law at the European University Institute, and is on leave from Humboldt University where he is a Professor of Private and Business Law.

Ilya Gvozdevskiy is Research Assistant to the Chair of e-Finance, EFinance Lab, at Goethe University, Frankfurt. Philipp Hacker is a postdoctoral fellow and research assistant at Humboldt University of Berlin; in autumn 2016, he joined the European University Institute as a Max Weber Fellow. Kitty Lieverse is Professor of Financial Regulatory Law at the Radboud University, Nijmegen, and an attorney-at-law and partner at Loyens & Loeff, Amsterdam. Marije Louisse is an attorney-at-law at Loyens & Loeff, Amsterdam. Eugenia Macchiavello is a Lecturer in Banking Law at the University of Genoa and Senior Research Fellow at the Genoa Centre for Law and Finance. Niamh Moloney is Professor of Financial Markets Law at the London School of Economics and Political Science. Paolo Saguato is a Senior Fellow at the Institute of International Economic Law at Georgetown Law Centre and a Senior Fellow at the Genoa Centre for Law and Finance. Larissa Silverentand is an attorney at law and partner at NautaDutilh, Amsterdam. Lisette Simons is an attorney at law at NautaDutilh, Amsterdam. Jasha Sprecher is an attorney at law at NautaDutilh, Amsterdam. Rezah Stegeman is an attorney-at-law and partner at Simmons & Simmons, Amsterdam. Rüdiger Veil holds the Alfried Krupp Chair for Civil Law, German and International Business, and Corporate Law and is managing director of the Institute for Corporate and Capital Markets Law at Bucerius Law School.

Eddy Wymeersch is Professor Emeritus at the University of Gent. He has been Chairman of the Committee of European Securities Regulators (CESR) and of the European Regional Committee of IOSCO. He was Chairman of the CBFA (Chief Executive 2001–7 and Chairman of the Supervisory Board 2007–10).

PART I

GENERAL ASPECTS

1 WHO’S AFRAID OF MIFID II? An Introduction Danny Busch and Guido Ferrarini

I. Introduction 1. General 2. From ISD to MiFID I 3. From MiFID I to MiFID II 4. The Term ‘Investment Firm’ II. 1. 2. 3. 4. 5. 6. 7. 8. 9. III. 1. 2. 3. 4. 5. 6.

Investment Firms and Investment Services Scope and Exemptions Governance of Investment Firms The Duty to Act in the Client’s Best Interest Product Governance and Product Intervention Independent and Non-Independent Advice Conflicts of Interest Inducements Agency and Principal Dealing Third-Country Investment Firms Trading Governance and Organization of Trading Venues The New Transparency Regime for Trading SME Growth Markets Dark Pools Derivatives Trading and the New Mandatory Trading Obligation Commodity Derivatives

7. Algorithmic Trading and High-Frequency Trading 8. MiFID II and Equity Trading Regulation: a US Perspective IV. Supervision and Enforcement 1. Public Enforcement of MiFID II 2. The Private Law Effect of MiFID I and MiFID II V. The Broader View and the Future of MiFID 1. MiFID II and Investor Protection: Picking Up the Crumbs of a Piecemeal Approach 2. Shadow Banking and the Functioning of Financial Markets 3. Investment-Based Crowdfunding: Is MiFID II Enough? VI. Final Remarks

I. Introduction 1. General [1.01] Investment firms and regulated markets have been closely regulated by the EU Markets in Financial Instruments Directive (MiFID I), the MiFID I Implementing Directive, and the MiFID I Implementing Regulation since 1 November 2007.1 MiFID I and the MiFID I Implementing Directive have been transposed into national legislation in the various Member States of the European Union (EU) and the European Economic Area (EEA). Naturally, the MiFID I Implementing Regulation has not been transposed into national law. The regulation has direct effect and, under European law, may not therefore be transposed into national law. MiFID I aims to provide a high level of harmonized investor protection, financial market transparency, and greater competition between trading venues. [1.02] On 3 January 20182—some ten years later—the MiFID I regime will be replaced by MiFID II, which comprises, among other things, a Directive (MIFID II), the Markets in Financial Instruments Regulation (MiFIR), and a truly impressive number of implementing measures,

commonly referred to as Level 2 legislation.3 MiFID I may have the reputation of being strict, but MiFID II/MiFIR tightens the reins even more. It is not hard to guess the reason: the financial crisis has also revealed gaps in the MiFID I legislation, notably in investor protection, as well as shortcomings in the functioning and transparency of financial markets. The MiFID II/MiFIR regime will have a major impact on the financial sector in Europe. [1.03] This book aims to analyse and discuss the main changes and new provisions introduced by MiFID II/MiFIR. The book chapters are grouped in a thematic way, covering the following areas: (i) investment firms and investment services, (ii) trading, (iii) supervision and enforcement, and (iv) the broader view and the future of MiFID II/MiFIR. This chapter provides a summary and overview of the chapters of this book. But before doing so, we will first briefly outline the history of the European regulation of investment firms and regulated markets, followed by a treatment of the central term ‘investment firm’.

2. From ISD to MiFID I [1.04] The European regulation of investment firms and regulated markets started not with the MiFID I regime but even earlier. 1993 saw the adoption of the Investment Services Directive (ISD).4 The ISD introduced a rather elementary scheme, which was fairly soon found to have deficiencies. The ISD provided for insufficient harmonization, weak investor protection, and limited competition between trading venues. In the view of lawmakers, the MiFID I regime was intended to rectify these shortcomings.5 [1.05] It is hardly surprising that the ISD provided insufficient harmonization since it was only intended to provide minimum harmonization. This meant that the Member States were able to introduce stricter rules, thereby detracting from the concept of a level playing field.6 In the interests of the general good, the Member States could also create obstacles for investment firms from other Member States.7 And this actually happened in many Member States, especially in relation to conduct of business rules (i.e. the rules on how investment firms should treat their

clients). Investor protection was also inadequate. This was mainly because the ISD did not cover all services and financial instruments. Indeed, the ISD did not even cover such an essential service in practice as investment advice, and commodity derivatives did not come under the ISD because they were not treated as financial instruments. This posed risks for investors and jeopardized the efficient and stable operation of financial markets. MiFID I has largely addressed these points. Unlike the ISD, MiFID I for the most part provides for maximum harmonization so that the Member States may not introduce supervision rules that are more strict or less strict than the MiFID I regime. Moreover, MiFID I covers investment advice, and the definition of ‘financial instrument’ has been expanded in such a way that it now also includes commodity derivatives.8 [1.06] Finally, as various rules had gradually become outdated, the ISD no longer provided an answer to changed market structures such as the advent of alternative trading venues. Alternative trading venues fulfil the same economic functions as regulated markets and came across to Europe in the 1990s from the United States, where they had been around for longer. In the 1990s alternative trading venues tended to be based in London. Regulated markets are, in common parlance, stock markets. The purpose of both established stock markets and alternative trading venues is to marry up supply and demand for financial instruments in a multilateral setting. It should be noted that financial instruments listed on a stock market can be traded not only on the stock market concerned but also on other stock markets or through an alternative trading venue. During the negotiations on the ISD, free competition between the established (at that time still mainly national) stock markets and other stock markets or alternative trading venues was already a reality.9 The northern and southern Member States took diametrically opposed positions on this subject during the negotiations. The United Kingdom, Germany, Luxembourg, the Netherlands, Ireland, and Denmark were in favour of free competition between trading venues because competition reduces trading costs. As they saw it, investors only stood to benefit from this. But France, Italy, Spain, Portugal, Greece, and Belgium saw things differently and pointed to the risk of market fragmentation. If an order in a financial instrument could be executed on a multitude of trading venues, this would mean that there would be insufficient liquidity at each trading venue to establish efficient price

formation. And this would actually be to the disadvantage of investors. Naturally, national interests also played a role here. The southern Member States were undoubtedly fearful that the position of their own national stock markets would be weakened. They preferred not to see trade ebbing away to alternative trading venues in London. Conversely, the United Kingdom had for this very reason much less difficulty in accepting the concept of free competition between trading venues. Whatever the case, the two sides failed to reach agreement on this during the negotiations on the ISD. As a compromise, the so-called optional concentration rule was included in the ISD.10 This rule meant that it was up to the Member States themselves to decide whether retail orders in financial instruments should necessarily be executed by sending them to the stock markets, which were still mainly national. Hardly surprisingly, the northern Member States did not make use of this option, unlike the southern Member States.11 However, the issue was unexpectedly resolved during the negotiations on MiFID I. The optional concentration rule was abolished, and since 1 November 2007 there has been free competition between stock markets and alternative trading venues throughout the EU/EEA. Judging by the figures, this competition has become fairly well established in recent years.12

3. From MiFID I to MiFID II [1.07] MiFID I is a major advance on the ISD. Indeed, MiFID I has justifiably been called the ‘core pillar of EU financial market integration’.13 Nonetheless, as noted at the start of this chapter, the financial crisis has been one factor that has exposed major gaps in investor protection and shortcomings in the operation and transparency of financial markets. As a result of these gaps and shortcomings, strenuous efforts have been made since 2010 to carry out a thorough review of the MiFID I regime.14

4. The Term ‘Investment Firm’ [1.08] We have already used the term ‘investment firm’ on a number of occasions. Unlike terms such as ‘insurer’ or ‘bank’, ‘investment firm’ is

little used in common parlance, and even in the financial press it is rarely encountered. Moreover, it is often confused with the term ‘investment fund’.15 So there is more than sufficient reason to devote a section to clarify this central term. [1.09] Under MiFID I and MiFID II, an investment firm is defined as an entity that provides investment services and/or performs investment activities. Both the services and the activities relate by definition to financial instruments.16 The investment services and activities are (1) reception and transmission of orders in financial instruments; (2) execution of orders in financial instruments on behalf of clients;17 (3) portfolio management; (4) investment advice; (5) underwriting or placing of financial instruments with and without a firm commitment basis; (6) dealing on own account; and (7) operation of a multilateral trading facility (MTF).18 MIFID II adds an additional activity to the list, namely the operation of an organized trading facility (OTF).19 For the sake of clarity, it should be noted that although the operator of a regulated market is not an investment firm, it is subject to MiFID I and MiFID II, albeit to a different set of rules than those applicable to investment firms.20 [1.10] It should be realized in this connection that the categories listed above can overlap. When providing a service, an individual portfolio manager or an investment adviser will often execute orders in financial instruments on behalf of the client (although there are also portfolio managers and investment advisers who leave the execution of orders on behalf of clients to other parties). In short, individual portfolio management and investment advice may include the power to execute orders on behalf of clients. [1.11] If a given financial product is not a financial instrument, it cannot give rise to an investment service or activity and the relevant entity is not an investment firm in relation to that product and is not subject to the MiFID I and MiFID II regime for investment firms. It should be noted, incidentally, that the term financial instrument is a fairly broad concept. It includes not only equities and bonds but also interest rate swaps and many other derivative products. Under MiFID II the definition of financial instrument has been broadened still further.21

[1.12] In any event, the basic rule is that an investment firm needs authorization granted by the home Member State’s competent authority in order to act as such.22 Once an investment firm has this authorization, it can provide investment services and perform investment activities throughout the entire EU/EEA, insofar as these services and activities are covered by the authorization (the so-called European passport).23 To be eligible for an authorization, an investment firm must fulfil extensive requirements.24 These requirements also apply after the investment firm has obtained authorization. It follows that investment firms must comply with the conditions for initial authorization at all times.25 Moreover, they must also comply with any additional requirements that may be imposed.26 [1.13] Credit institutions authorized under Directive 2013/36/EU can also provide investment services and/or perform investment activities, in which case they are largely subject to the same MiFID I and MiFID II rules applicable to investment firms.27

II. Investment Firms and Investment Services 1. Scope and Exemptions [1.14] In Chapter 2, Lieverse discusses the main changes of the scope of MiFID II, in comparison with MiFID I. MiFID II introduces several changes to the scope of the supervision of investment firms and some other business related to the investment services industry, including data reporting service providers. The list of investment services has been extended by adding the operation of an OTF. Also, the revision and—in respect of some items—extension of the definition of ‘financial instrument’ has an impact on the scope of MiFID II supervision, including emission allowances. In addition, advisory and distribution services in respect of structured deposits by investment firms and credit institutions have been brought within the scope of MiFID II. Furthermore, supervision comparable to the MiFID II system is being introduced for insurers and intermediaries in respect of investment-based insurance products. At the same time, the

scope of some of the exemptions to MiFID II is revised, with an impact in particular for trading on own account. Trading on own account through high-frequency algorithmic trading techniques is brought within the scope of MiFID II, and the broad exemption for traders in commodities derivatives has also been limited. As a result, the scope of MiFID II is considerably broader than MiFID I. This has an effect beyond the applicability of MiFID II regulation, as currently the prudential requirements are as a starting point also linked to the MiFID II qualification, though with specific exemptions. In addition, the concept of linking prudential supervision to the MiFID II qualification, rather than to the prudential risk profile of the underlying services and activities, is subject to review and reconsideration.

2. Governance of Investment Firms [1.15] In Chapter 3, Binder discusses the governance of investment firms. In response to widespread concerns about the quality of governance arrangements in financial intermediaries as a source of systemic risk, the revised framework for the regulation of investment services in the EU has not just reinforced the governance requirements that have been in place for a long time, but has significantly increased their complexity and intensity. Investment firms, which are subject to both the Capital Requirements Directive (CRD) IV and MiFID II, now have to comply with a wide range of regulatory requirements, covering board structure and composition, as well as individual duties of board members and their remuneration, general organizational and risk management arrangements, and the supervisory scrutiny of shareholders and owners with qualifying holdings. The author concludes that the underlying policy (to address systemic implications of financial intermediation irrespective of the business model) is certainly consistent, but that the equal treatment of banks and investment firms under broadly identical regulatory frameworks (albeit with additional requirements imposed by MiFID II) is nonetheless questionable in view of the diversity of existing business models and activities. Starting with an analysis of the historical emanation of the new approach and the underlying policy, the author seeks to identify potential weaknesses and implications for its implementation.

3. The Duty to Act in the Client’s Best Interest [1.16] In Chapter 4, Enriques and Gargantini analyse the scope, contents, and implications of MiFID II’s new framework regarding the duty to act in the client’s best interest. The authors analyse the duty as an autonomous source of obligations for investment firms and as a guidance principle for both EU bodies in charge of implementing MiFID II and judges and supervisory authorities interpreting more specific duties. The authors also discuss the implications of extending the duty to intrinsically at-arm’slength activities such as dealing on own account and self-placement. [1.17] MiFID II confirms, in line with its predecessors, that investment firms must act in accordance with the best interests of their clients. Besides being an overarching principle for firms’ professional conduct, the duty to act in the client’s best interest is further specified by more detailed provisions within the body of MiFID II, and therefore plays a role in their interpretation. Rules concerning the management of conflicts of interest, including at the level of staff remuneration practices, limitations to inducements, and best execution are just a few examples of obligations aimed at ensuring that clients’ interests are pursued even when firms’ incentives may be tainted. The ‘best interest’ principle may have direct application as well, thus grounding firms’ liability even in cases where no specific rule of conduct is violated. [1.18] Although the general principle set forth by Article 24(1) is not new, MiFID II further specifies its implications and, overall, enhances its effects: it does so by introducing new specific rules of conduct implementing the ‘client’s best interest’ standard and by broadening its scope of application. For instance, new provisions ensure that firms pursue clients’ interests both during product distribution and before customers are approached: MiFID II mandates a thorough understanding of the marketed financial instruments and of clients’ needs when financial instruments are offered or recommended, and requires firms to design financial instruments in the light of the characteristics of the target groups at the early stage of product manufacturing. At the same time, the expansion of the regulated activities brings new business areas within the purview of the ‘best interest’ standard: this is particularly the case with placement of firms’ own products on the

primary market (self-placement), now explicitly included in the list of investment services and activities.

4. Product Governance and Product Intervention [1.19] In Chapter 5, Busch analyses and discusses the product governance and product intervention rules introduced by MiFID II/MiFIR. According to the author, the combination of these two approaches designed to exclude harmful products from the market is a major step forward in investor protection. On the other hand, complying with the product governance rules will entail costs for the firms concerned. They will have to put in place the requisite internal procedures and there will be a statutory duty for the firm developing the product and the firm distributing it to exchange a considerable volume of information. All in all, the author believes that these extra costs (which will undoubtedly be discounted in the cost price of the product) are acceptable. In any event, they are dwarfed by the social costs caused by the marketing of harmful financial products (e.g. interest rate swaps sold to small and medium-sized enterprises in many European countries). The author argues that MiFID II’s introduction of both product governance rules and product intervention rules is no more than common sense. It would be naive to think that product governance rules could in practice guarantee that harmful products are no longer marketed.

5. Independent and Non-Independent Advice [1.20] In Chapter 6, another noteworthy innovation is discussed by Giudici: MiFID II (unlike MiFID I) draws a distinction between independent and non-independent advice. The author submits that the quickest policy indication for increasing households’ trust in financial markets, to the benefit of the economic system, seems to be the offer of professional financial advice on affordable terms. The problem is how to convince investors to pay for advice, and how to protect investors who do not want to pay for conflicted advice and for hard sell under the guise of personal recommendation—an area where MiFID I has not performed well. MiFID II’s answer is to pose a new set of information duties on financial

advisors with the clear intention of nudging investors towards independent, fee-only advice. The intention is good. However, the author argues that the new regime raises many important issues, among which are: the ambiguity of the ‘independent’ suit, which can be taken or demised by investment firms from time to time and from client to client; the interaction between the product governance regime and the suitability assessment; the regulatory inconsistency that is emerging between investment advice and portfolio management; and the potential costs of the written statement of suitability.

6. Conflicts of Interest [1.21] In Chapter 7, Grundmann and Hacker analyse and discuss the issue of conflicts of interest in the ambit of services regulated by MiFID. The authors do so mainly with respect to the MiFID II regime, but take the MiFID I regime as a background because it has been more abundantly discussed and because the largest part of the basic structure and also many single solutions have remained unchanged. The chapter combines (several) legal and social science theories with an analysis of the substantive law solutions, and presents them in an evolutionary perspective, but also with a focus on a more detailed analysis of the status now reached. The authors argue that the legislature did not take sufficient care in specifying the different approaches which (may have) influenced the substantive law solutions found, and did not even specify clearly the interests involved. The authors further argue that it is best to expose the regime by segregating the single main stages of an investment services relationship and approaches which have been used (and solutions which have been found). Such segregation is all the more important as solutions and approaches used at the different stages are indeed highly heterogeneous in kind. The authors do not criticize, but clarify this—it may well be that the factual background of the different stages and situations calls for such heterogeneity in regulatory approach (which will be discussed for the single situations). For presenting this programme, the authors first expose the foundations—(i) the most important theoretical approaches to a regulation of conflicts of interest, and (ii) the legislative bases (and history)—and then proceed to analyse the five different stages and situations (and, in addition, linked contracts) step by

step. The authors always start by characterizing the type of approach taken in the legislative regime, then move on to the main features of innovation in the transition from MiFID I to MiFID II, before finally discussing in greater detail the substantive solution found in MiFID II.

7. Inducements [1.22] In Chapter 8, Silverentand, Sprecher, and Simons analyse and discuss the MiFID II inducement rules. During the negotiations on MiFID II, it became clear that inducements were a topic on which there was no easy agreement between the Member States. Where certain Member States pressed for a total ban on inducements, other Member States were unwilling to impose such strict rules. As is not uncommon in these situations, the political compromise that was reached was to allow for deviating rules by those Member States that wished to see stricter rules. Allowing for this may be regarded as counter to the general trend set by the European legislator to create less room for Member State options by creating ‘single rulebooks’ and an increase in the use of regulations. The authors conclude that it is disappointing to see that on such an important topic, the European market will continue to have rules that deviate per Member State. The Dutch legislator has already indicated that it will make use of the (continued) room for ‘gold plating’, and it can according to the authors be expected that other Member States where stricter rules apply will do the same, leaving an un-level playing field for investment firms regarding the use of inducements.

8. Agency and Principal Dealing [1.23] In Chapter 9, Busch focuses on the question of whether allowing the extent of the protection afforded to an investor under MiFID to be largely dependent on the distinction between dealing on own account on one hand and trading on behalf of the client (and other forms of investment service) on the other is justified. The author submits that this question must be answered in the negative. An investor may reasonably expect the investment firm used by him to look after his interests adequately and thus

to observe certain duties of care towards him. The investment firm is, after all, ideally placed to use its expertise. Its fund of knowledge is bound to be superior to that of an investor, particularly a retail investor. Nor is this any different where the investment firm acts purely as the investor’s contractual counterparty. In such cases, the investor is reasonably entitled to expect the investment firm to observe the same duty of care that would apply if it were providing an execution-only service. Moreover, the distinction between dealing on own account (principal dealing) on one hand and trading on behalf of the client (and other forms of investment service) on the other is tenuous, arbitrary, and easy to manipulate. This is all the more so where a contractual clause providing that an investment firm acts solely as contractual counterparty is claimed to apply even where an employee of the investment firm advises the investor, contrary to the terms of the agreement. According to the author, MiFID II provides no practicable criterion either. To achieve an adequate level of investor protection MiFID II resorts to the artifice of reclassifying certain types of dealing on own account as acting on behalf of the client. Finally, both the UK Government (in response to the Kay Review) and the Dutch Supreme Court take the view that duties of care must also apply where an investment firm acts solely as an investor’s contractual counterparty. Under a future MiFID III, an investment firm which acts solely as contractual counterparty should be required to observe the same duty of care as applies in the case of the investment service of execution of orders on behalf of the client.28

9. Third-Country Investment Firms [1.24] In Chapter 10, Busch and Louisse analyse and discuss the MiFID II/MiFIR rules for third-country investment firms (i.e. investment firms established outside the EU/EEA). Under MiFID II the position of thirdcountry firms will no longer be a purely national matter. MiFID II now itself sets the parameters. But, as the authors explain, the legislation is unfortunately complex and does not provide for full harmonization. The Commission and the Member States found it hard to reach agreement on this point. The phenomenon of differing national regimes is therefore not precluded even under MiFID II. The fact that MiFID II regulates the position of third-country firms reflects a European trend. The European

Market Infrastructure Regulation (EMIR), the CRA Regulation, and the AIFMD all deal with the position of parties established in a third country. These may be CCPs, CRAs, AIFMs, or depositaries. [1.25] MiFID II/MiFIR introduces a common regulatory framework that should harmonize the existing fragmented framework for the provision of services by third-country firms, ensure certainty and uniform treatment of third-country firms accessing the EU/EEA, ensure an assessment of effective equivalence by the Commission, and provide for a comparable level of protection to clients in the EU/EEA receiving services from thirdcountry firms. [1.26] Articles 46–49 MiFIR (the MiFIR third-country regime) regulate the provision of investment services to eligible counterparties and per se professional clients and performance of investment activities by thirdcountry firms following an equivalence decision by the Commission. Upon adoption of such equivalence decision, a third-country firm can register itself with ESMA. Once a third-country firm has been registered by ESMA, it may provide investment services or perform investment activities to eligible counterparties and per se professional investors throughout the EU/EEA. Member States may not impose any additional requirements on such third-country firms in respect of matters covered by MiFID II/MiFIR and CRD IV/CRR. In the absence of an equivalence decision in relation to a third country, the national regimes of the Member States continue to apply. [1.27] The provision of investment services to retail and opt up professional clients and performance of investment activities through the establishment of a branch is regulated in Articles 39–43 MiFID II (the MiFID II third-country regime). Under the MiFID II third-country regime, Member States have the option to require the opening of a branch by a third-country firm wishing to provide investment services and/or perform investment activities (with or without ancillary services) in a Member State to retail clients or to opt up professional clients (i.e. clients who have obtained professional client status by opting up). If a Member State requires the opening of such a branch harmonized requirements will apply. Otherwise, the national regimes of the Member States continue to apply.

[1.28] Although the basics of the third-country regime under MiFID II/MiFIR seem to be set, the authors show that the devil is in the detail. There is still a considerable lack of clarity in relation to the scope of the MiFID II/MiFIR’s third-country regime, as in where it concerns investment activities, and activities other than investment services and investment activities (such as selling and advising in relation to structured deposits), or the initiative test. In addition, the concurrence between MiFID II and MiFIR, if a third-country firm provides investment services eligible counterparties and retail clients, raises interesting questions that are not easily answered.

III. Trading 1. Governance and Organization of Trading Venues [1.29] In Chapter 11, Ferrarini and Saguato analyse the governance and organization of trading venues. The authors show that MiFID II brings modest changes to the EU landscape of trading venues. The newly introduced Organized Trading Facilities (OTFs) are going to be the reference venues for a significant portion of the derivatives trading in years to come. The regulated markets (RMs) and Multilateral Trading Facilities (MTFs) regimes have been aligned, and specific provisions strengthen the governance of the venues and their operators. However, market dynamics are already challenging the MiFID II regulatory framework for the governance and organization of trading venues. Trading venues have developed over the last twenty years into Financial Markets Infrastructures (FMI) groups that provide both trading and post-trading services. These new conglomerates test the capacity of the current regulatory and supervisory regime of financial markets—and MiFID II itself—to oversee their activities and to guarantee competition and stability in the trading and post-trading industry. MiFID II does not explicitly take FMI groups into account—trading venues are regulated as individual entities, which may or may not be operated within a group. In MiFID II, only three sets of rules address, to varying degrees, some of the potential risks underlying FMI

groups: (i) conflicts of interest between firms operating and RMs, MTFs, and OTFs; (ii) transparency of ownership of trading venues and suitability of trading venues’ shareholders; and (iii) non-discriminatory access to trading and post-trading services. However, the prudential regulation and supervision of the FMI group has not been included in the MiFID I review process. The authors conclude that this regulatory gap might be a threat to the stability of financial markets, and regulators should consider a regulatory intervention to fill it. The experience of the regulatory and supervisory colleges of CCPs under EMIR and the regulatory framework of the financial conglomerates Directive could be two possible ways to strengthen the oversight of FMI groups.

2. The New Transparency Regime for Trading [1.30] In Chapter 12, Moloney outlines the main features of the extensive new transparency regime which will apply to trading in a wide range of asset classes under MiFIR. By contrast with MiFID I, which limited transparency requirements to the equity markets and which contained extensive exemptions and waivers, MiFIR adopts a maximalist approach to transparency. The most extensive transparency requirements apply to the three forms of ‘trading venue’ for multilateral trading which are established under the MiFID II/MiFIR venue classification system (RM, MTF, and OTF). Bilateral/OTC trading between counterparties is subject only to posttrade transparency requirements. Overall, MiFIR’s regulatory design has been shaped by a driving concern to protect liquidity, particularly in nonequity asset classes. Indeed, exemptions, waivers, suspensions, and calibrations, designed to protect liquidity, are a recurring feature of the new transparency regime. The pre-trade equity/equity-like transparency regime is subject, for example, to a series of waivers designed to allow venues to support ‘dark’ trading, given the efficiencies which dark trading can bring, notably for institutional investors and for market liquidity generally. [1.31] Once the Level 2 process which governs the adoption of administrative rules is completed, the MiFIR transparency rulebook will form a regulatory regime of vast breadth and depth. But operational supervisory decision-making by national supervisors will, in addition,

become subject to EU-level controls and processes. The waivers which are available from the transparency requirements for equity and non-equity trading can be regarded as the key battleground on which national supervisors (and the Member States) will seek to protect national territory, in that the waivers afford national supervisors the possibility to apply tailored treatment to distinct national trading systems and practices. [1.32] In one of the key changes from MiFID I, ESMA is now empowered to oversee national supervisory decision-making with respect to equity and non-equity pre-trade transparency waivers. Under MiFID I, waiver decisions in relation to the equity transparency rules are currently at the discretion of national supervisors, but under voluntary arrangements proposed supervisory waiver decisions are notified to ESMA, which then adopts a common position across its member supervisors on the waiver. This informal process has now been formalized. The author submits that while the new regime will have significant market-shaping effects, it will also prove revealing as to the effectiveness of the administrative governance apparatus which currently supports EU financial governance. In particular, the new transparency regime is exposing weaknesses in regulatory administrative governance, notably with respect to how the EU deals with uncertainty. Specifically, the absence of a power to suspend administrative rules where they prove to have destabilizing effects may prove a significant challenge to the effectiveness of EU financial governance, given the uncertain impact of MiFIR on financial markets.

3. SME Growth Markets [1.33] In Chapter 13, Veil and Di Noia analyse and discuss SME Growth Markets, introduced under MiFID II as an important strategy to improve access to finance for SMEs in Europe. SME Growth Markets should be more flexible than regulated markets. However, the authors submit that this will not be the case. The regime for SME Growth Markets will consist of strict rules about insider trading and market manipulation which are subject to supervision by NCAs. These parts of the regime are important to ensure investor confidence. It is also convincing to protect investors by a disclosure regime ensuring the publication of price-relevant information on

SME Growth Markets. However, the authors argue that it is neither necessary nor recommendable to apply the respective disclosure obligations (as well as the similar regime for insiders’ lists and the identical one for managers’ transactions) under the MAR. Instead, a system based on current event reports is sufficient in order to tackle information asymmetries on SME Growth Markets. The authors conclude that the disclosure regime for SME Growth Markets in Europe should be re-assessed with the aim of allowing market operators to experiment with alternative disclosure obligations on SME Growth Markets.

4. Dark Pools [1.34] In Chapter 14, Gomber and Gvozdevskiy focus on the concept of dark trading in the context of MiFID II. This is analysed against the background of the MiFID I regulation and its economic consequences for European equity markets. Dark trading implies orders with partial or missing pre-trade transparency, meaning that these orders are hidden from the rest of the market. Dark pools of regulated venues and in (un-regulated) OTC markets allow institutional investors to execute large blocks of shares in one lot, reducing information leakage, diminishing market impact, and providing price improvements. MiFID II aims to increase market transparency and to bring trading of financial instruments into regulated platforms. Extending the waivers introduced by MiFID I, the new Directive announces the double volume cap regime. The first cap of the mechanism restricts the usage of waivers by individual trading venues if the proportion of total value traded under negotiated transactions and reference price waivers together exceeds 4 per cent of total trading volume. The second cap suspends all trading venues across the Union from trading under negotiated transactions and reference price waivers, if the cumulated proportion of total value traded under these waivers across the Union exceeds 8 per cent of total trading volume. And additional trading obligation of shares will reduce the extent of OTC trading in Europe. Some market participants and trading venues recently introduced MiFID II-ready solutions preventing dark executions from being subject to the double volume cap regime either by classifying the orders as large in scale or by introducing trading systems

based on auction market models. These models and functionalities, which already anticipate the future MIFID II regime, are also discussed.

5. Derivatives Trading and the New Mandatory Trading Obligation [1.35] In Chapter 15, Stegeman and Berket analyse and discuss the new mandatory trading obligation for derivatives provided in Title V of MiFIR. The authors conclude that time will tell whether the new rules discussed in this chapter will impact on the derivatives markets as they stand today, and, if so, to what extent and in what ways. [1.36] The opinions on whether the trading obligation will have any effect are very fragmented. This makes it hard to predict the exact consequences it may have. However, looking at the US, activity and liquidity in the USD and EUR plain vanilla IRS markets have increased following the Dodd– Frank trading mandate, according to a Bank of England Staff Working Paper.29 Furthermore, it is concluded that the associated execution costs in these markets are significantly reduced, the causes of this reduction likely to be the shortening in dealer intermediation chains and more competitive pricing by dealers. Both are potentially attributed to mandatory exchange trading, as the results of the research suggest that inter-dealer market activity has reduced following the trading mandate entering into force. However, the researchers could not scientifically prove a causal link between the trading mandate and the reduction in inter-dealer trading. As such, the full impact of the trading mandate remains uncertain. [1.37] With respect to the clearing obligation and the rules on the timing of acceptance for clearing (STP obligation) the authors expect the impact to be relatively limited as they seem to be in accordance with current market practices, or at least not too far off. However, it remains to be seen whether the mandatory reduced timeframes increase the risk of errors. [1.38] With the portfolio compression rules under MiFIR the European legislator continues stimulating portfolio compression by regulation. The light-touch regime seems to strike the right balance between, where

possible and appropriate, encouraging parties to compress their portfolios on one hand and increasing transparency for the entire market on the other hand. According to the authors it is therefore conceivable that portfolio compression will remain a much used exercise in the foreseeable future. [1.39] When comparing the new obligations, according to the authors the rules relating to indirect clearing have the most velar impact on the current market practice. Since, different from the indirect clearing rules for OTC derivatives under EMIR, indirect clearing is commonly seen in the ETD space, these rules will definitely impact these existing client clearing processes.

6. Commodity Derivatives [1.40] In Chapter 16, Sciarrone Alibrandi and Grossule analyse and discuss the MiFID II/MiFIR regulation of the commodity derivatives sector. The authors submit that this sector is one of the areas most affected by the financial markets reform process, particularly in the MiFID I review. Under the G20 mandate, EU policymakers issued specific provisions to increase transparency, tackle the turbulences affecting commodity markets, and curb the negative effects of speculation. [1.41] This chapter analyses the position limits regime establishing limits on ex ante positions related to all commodity derivatives and prohibiting participants from holding contracts beyond a certain limit. Furthermore, this chapter deals with the new rules amplifying the regulatory and supervisory powers of ESMA, of national competent authorities (NCAs), and trading venues, introducing a range of interventionist tools that can affect operators’ investment business. The chapter also stresses the need to introduce a specific regulation depending on different commodity derivatives (e.g. energy and food). The main provisions and the specific technical standards are analysed, paying particular attention to controversial measures such as the definition of the ancillary activities, the methodology to calculate the position limits, and the authorities’ new powers (especially in the field of product intervention).

7. Algorithmic Trading and High-Frequency Trading [1.42] In Chapter 17, Conac analyses the MiFID II rules on algorithmic trading (AT), including high-frequency trading (HFT). The author argues that AT raises serious issues of volatility and systemic risk while HFT raises issues of systematic front-running of investors. However, there are serious differences of opinion within the EU as to the benefits and risks of those two trading techniques, and especially of HFT. Therefore, MiFID II takes a technical approach mostly focused on prevention of a repeat of the 2010 Flash Crash and some provisions on market abuse. [1.43] Since the MiFID II will not be implemented before 2018, the ESMA 2012 Guidelines will remain the most effective regulation to frame the development of HFT. The author submits that this reveals the usefulness of having ESMA adopting Guidelines in order to tackle market developments with relative speed. However, since the EU legislator decided not to tackle HFT strongly, and since implementation of the Directive is still far away, it is probable that some legislators and supervisors will do it themselves. The successful case of enforcement in France is a sign of this possible trend. Therefore, regulation by prosecution of market abuse by HFT traders could lead to a de facto ban of HFT in some Member States. The author argues that this would be a huge cost for supervisors as they would need to allocate scarce resources to this topic, which means that only the most motivated supervisors will do it.

8. MiFID II and Equity Trading Regulation: a US Perspective [1.44] In Chapter 18, Fox provides a US perspective on the MiFID II equity trading regulation. The author concludes that a comparison of the EU and US market structure rules, and the concerns that generated them, suggests three key differences. Relative to the United States, the EU shows (i) more concern with having an effective price-formation process, (ii) more concern with the possibility that HFTs contribute to price instability and engage in market abuse, and (iii) less concern with promoting competition among trading venues. These differences have characterized the MiFID I

era and are reflected in MiFID II as well, although MiFID II does evince somewhat greater concern about competition among trading venues than was true before.

IV. Supervision and Enforcement 1. Public Enforcement of MiFID II [1.45] In Chapter 19, Gortsos provides a systematic presentation, analysis, and assessment of the MiFID II provisions (Articles 67–88) on supervision, enforcement, and cooperation by competent authorities. The author addresses the role of Member States’ competent authorities within the system of the MiFID II, with particular emphasis on the competent authorities’ supervisory powers, their power to impose administrative sanctions and measures, as well as criminal sanctions and redress procedures. Cooperation arrangements between Member States’ competent authorities, the obligation to cooperate with the ESMA, and cooperation with third countries are addressed as well. Finally, the rules are briefly assessed on the basis of three (out of the five) elements pertaining to financial supervision, which, in the author’s view, are essential for the preservation of financial stability and the attainment of the other goals underlying (public) capital markets law and which are addressed by the provisions of the MiFID II: (i) micro-prudential supervisory effectiveness, (ii) the efficient and unobstructed exercise of competent authorities’ sanctioning powers, and (iii) the effectiveness of supervisory cooperation arrangements.

2. The Private Law Effect of MiFID I and MiFID II [1.46] In Chapter 20, Busch examines to what extent the civil courts are bound by MiFID I/MiFID II under European law. In addressing this subject, the author discusses the following topics: (i) whether the civil courts may be less strict or stricter than MiFID I/MiFID II, (ii) whether the contracting

parties may be less strict or stricter than MiFID I/MiFID II, (iii) whether there is any influence of MiFID I/MiFID II on the principle of ‘relativity’ or ‘proximity’ in the Member States where this is a requirement for liability in tort, (iv) whether there is any influence of MiFID I/MiFID II on proof of causation, (v) whether there is any influence on a contractual limitation or exclusion of liability, and (vi) whether the civil courts are obliged to determine of their own motion whether there has been an infringement of MiFID I/MiFID II (conduct of business) rules in disputes between investment firms and private investors. [1.47] The author concludes that the last word has not been spoken about the effect of MiFID I/MiFID II on private law. MiFID II is as unclear about this as MiFID I. Although the possible contours are somewhat clearer as a result of the judgments in the Genil30 and Nationale-Nederlanden31 cases, the EU Court of Justice has not yet explicitly answered the main questions. For more definitive answers it will be necessary to await the further judgments of the Court of Justice.

V. The Broader View and the Future of MiFID 1. MiFID II and Investor Protection: Picking Up the Crumbs of a Piecemeal Approach [1.48] In Chapter 21, Colaert clarifies the relationship between the MiFID II and several other closely related Directives, such as the Insurance Distribution Directive, the PRIIPs Directive, and the UCITS Directive. The interaction between different pieces of EU financial legislation becomes ever more important—and difficult. [1.49] The analysis reveals numerous inconsistencies and interpretation difficulties. The author argues that while some of the shortcomings are inherent to the EU piecemeal approach, other problems should and could have been avoided, by adopting a holistic approach of financial regulation and supervision.

2. Shadow Banking and the Functioning of Financial Markets [1.50] In Chapter 22, Wymeersch analyses and discusses shadow banking and the functioning of financial markets; that is, activities that are outside the regulatory perimeter of MiFID I and MiFID II/MiFIR. The author shows that financial activity that is taking place outside the traditional and often regulated financial sphere of the securities markets has become quite substantial and especially diverse. This segment is often designated as ‘shadow banking’—a misnomer. The activities to be classified under this characterization have been mapped by the Financial Stability Board, and include a wide variety of entities which either specialize in specific financial activities, or offer financial services as part of their overall product offer. The risks they create are essentially of a macro or ‘systemic’ nature, leading to major financial disruption and contagion. In the securities markets, this activity is often indirectly regulated as part of the wider market regulation. For example, this is the case for the market in derivatives, or for the use of securitized instruments; for derivatives, certain trading venues have been mandated, not always with comprehensive results. The author submits that the post-trade sector may be a source of considerable risk: therefore OTC derivatives are subject to mandatory clearing in the CCP, while title security for most traditional securities will be organized within the Central Securities Depositories. The recent Securities Financing Regulation, dealing with repos and similar instruments, clarifies the position of investors that see their securities ‘reused’ in a second collateral transaction, for which in the future their express consent will be necessary. The author concludes that, although thought to be on the border of the traditional securities systems, these matters and the related regulation have a considerable impact on the framing of certain classes of securities and may determine the safety of the overall system, including that of the final investor.

3. Investment-Based Crowdfunding: Is MiFID II Enough?

[1.51] In Chapter 23, Ferrarini and Macchiavello explore the policy and regulatory issues generated by investment-based crowdfunding in Europe. First, the authors argue that crowdfunding raises serious investor protection concerns, particularly when directed to retail investors. As governments try to stimulate innovation and the formation of new enterprises, a trade-off is created between investor protection and economic growth. EU law and the laws of Member States try to solve this trade-off in different ways, as the authors show with reference to MiFID I and the laws of the UK, France, Italy, Spain, and Germany. Second, the authors focus on MiFID II and show that the new Directive, while enhancing investor protection in general and furthering harmonization, does not create all the conditions needed for the formation of a pan-European crowdfunding market. At the same time, MiFID II narrows the potential for exemptions under which some Member States have adopted special regimes for crowdfunding, therefore restricting the scope for an enabling approach to investment-based crowdfunding at the national level.

VI. Final Remarks [1.52] On the whole, this volume’s chapters show that MiFID II will have a major impact on the financial services sector. MiFID II introduces a veritable flood of new rules (not only at Level 1 but certainly also at Level 2) and tightens up certain aspects of existing rules. [1.53] Naturally, only time will tell whether the package of measures under MiFID II will have the desired effect and always allow the right balance to be struck between costs and benefits. And the question arises of whether compliance with the flood of new regulatory provisions is even possible.32 [1.54] In any event, the Commission has recently indicated that it will pay more attention to the regulatory burden. Under the direction of Frans Timmermans, the Commission has developed an improved regulation programme: fewer new rules and more evaluation of existing rules.33 This is yet another real challenge.

[1.55] It is also worth noting here that the Commission published a call for evidence on 30 September 2015. In the space of just six years the Commission has published no fewer than forty Directives and regulations relating to the financial markets with the aim of preventing another crisis. The Commission now wishes to examine whether the structure as a whole is consistent and whether any of the rules overlap or give rise to undesirable economic consequences. On 17 May 2016, the Commission published a summary of the 288 responses it had received. Overall, stakeholders did not dispute the reforms of recent years and many expressed support, highlighting the benefits of the new rules. But the Call for Evidence was also welcomed as giving all interested parties the opportunity to assess the potential interactions, overlaps, and inconsistencies between different pieces of legislation.34 [1.56] Finally, there is Brexit. A referendum was held on Thursday 23 June 2016, to decide whether the UK should leave or remain in the EU. Leave won by 52 per cent to 48 per cent.35 Only time will tell what this means for the financial sector in Europe and the UK financial sector in particular. However, it seems safe to assume that the UK will not formally leave the EU before MiFID II/MiFIR becomes binding on the financial sector in Europe. So Brexit should not delay a UK MiFID II implementation beyond the EU’s 3 January 2018 effective date.36 [1.57] Whichever way you look at it, MiFID II will be a fact of life for the financial sector in Europe for the years to come.

1

Directive 2004/39/EC, OJ L 145, 30 April 2004, pp. 1–47 (MiFID I); Directive 2006/73/EC, OJ L 241, 2 September 2006, pp. 26–58 (MiFID I Implementing Directive); Regulation (EC) No 1287/2006, OJ L 241, 2 September 2006, pp. 1–25 (MiFID I Implementing Regulation). 2 Initially, Directive 2014/65/EU, OJ L 173, 15 May 2014, pp. 349–496 (MiFID II) and Regulation (EU) No. 600/2014, OJ L 173, 15 May 2014, pp. 84–148 (MiFIR) stipulated that the bulk of the new legislation would become binding on the financial sector as per 3 January 2017, but this has been extended to 3 January 2018. See Directive 2016/1034/EU OJ L 175, 23 June 2016, pp. 8–11; (2) Regulation (EU) No. 2016/1033 OJ L 175, 23 June 2016, pp. 1–7. The reason for the extension lies in the complex technical infrastructure that

needs to be set up for the MiFID II package to work effectively. The European Securities and Markets Authority (ESMA) has to collect data from about 300 trading venues on about 15 million financial instruments. To achieve this result, ESMA must work closely with national competent authorities and the trading venues themselves. However, the European Commission was informed by ESMA that neither competent authorities nor market participants would have the necessary systems ready by 3 January 2017. In light of these exceptional circumstances and in order to avoid legal uncertainty and potential market disruption, an extension was deemed necessary. See . 3 MiFID I and MiFID II are both based on the ‘Lamfalussy process’ (a four-level regulatory approach recommended by the Committee of Wise Men on the Regulation of European Securities Markets, chaired by Baron Alexandre Lamfalussy and adopted by the Stockholm European Council in March 2001 aiming at more effective securities markets regulation) as developed further by Regulation (EU) No 1095/2010 of the European Parliament and of the Council, establishing a European Supervisory Authority (European Securities and Markets Authority or ESMA): at Level 1, the European Parliament and the Council adopt a Directive in co-decision which contains framework principles and which empowers the Commission acting at Level 2 to adopt delegated acts (Article 290 The Treaty on the Functioning of the European Union C 115/47) or implementing acts (Article 291 The Treaty on the Functioning of the European Union C 115/47). In the preparation of the delegated acts the Commission will consult with experts appointed by Member States within the European Securities Committee. At the request of the Commission, ESMA can advise the Commission on the technical details to be included in Level 2 legislation. In addition, Level 1 legislation may empower ESMA to develop draft regulatory or implementing technical standards according to Articles 10 and 15 of the ESMA Regulation, which may be adopted by the Commission (subject to a right of objection by Council and Parliament in case of regulatory technical standards). At Level 3, ESMA also works on recommendations and guidelines and compares regulatory practice by way of peer review to ensure consistent implementation and application of the rules adopted at Levels 1 and 2. Finally, the Commission checks Member States’ compliance with EU legislation and may take legal action against non-compliant Member States. See e.g. European Commission, MiFID Impact Assessment (COM(2011) 656 final), p. 356, footnote 1. 4 Directive 93/22/EEC, OJ L 141, 11 June 1993, pp. 27–46. For a commentary, see Guido Ferrarini (ed.), European Securities Markets: The Investment Services Directive and Beyond (Kluwer Law International 1998). 5 Cf. Ryan Davies, Alfonso Dufour, and Brian Scott-Quinn, ‘The MiFID: Competition in a New European Equity Market Regulatory Structure’ in Guido Ferrarini and Eddy Wymeersch (eds) Investor Protection in Europe: Corporate Law Making, the MiFID and Beyond (OUP 2006), p. 163; Barbara Alemanni, Giuseppe Lusignani, and Marco Onado, ‘The European Securities Industry. Further Evidence on the Roadmap to Integration’, ibid, p. 199; Guido Ferrarini and Fabio Recine, ‘The MiFID and Internalisation’, ibid, p. 235;

Johannes Köndgen and Erik Theissen, ‘Internalisation under MiFID: Regulatory Overreaching or Landmark in Investor Protection’, ibid, p. 271; Peter Mülbert, ‘The Eclipse of Contract Law in the Investment Firm–Client Relationship: The Impact of the MiFID on the Law of Contract from a German Perspective’, ibid, p. 297; Luca Enriques, ‘Conflicts of Interest in Investment Services: The Price and Uncertain Impact of MiFID’s Regulatory Framework’, ibid, p. 321. 6 See Recital 27 ISD. 7 Article 17(4) and Article 18(2) ISD. See David O’Keeffe and Nelius Carey, ‘Financial Services and the Internal Market’ in Guido Ferrarini, Klaus Hopt, and Eddy Wymeersch (eds) Capital Markets in the Age of the Euro: Cross-border Transactions, Listed Companies and Regulation (Kluwer Law International 2002), p. 3; Niamh Moloney, ‘Investor Protection and the Treaty: an Uneasy Relationship’, ibid, p. 17; Michel Tison, ‘Conduct of Business Rules and Their Implementation in the EU Member States’, ibid, p. 65. 8 Cf. Niamh Moloney, ‘Effective Policy Design for the Retail Investment Services Market: Challenges and Choices Post FSAP’ in Ferrarini and Wymeersch (eds) (n. 5), 381; Michel Tison, ‘Financial Market Integration in the Post FSAP Era: In Search of Overall Conceptual Consistency in the Regulatory Framework’, ibid, p. 443. 9 Cf. Marco Pagano, ‘The Changing Microstructure of European Equity Markets’ in Ferrarini (ed.) (n. 4), 177. 10 Article 14(3) ISD. Cf. Guido Ferrarini, ‘Exchange Governance and Regulation: An Overview’ in Ferrarini (ed.) (n. 4), 245. 11 Davies, Dufour, and Scott-Quinn, (n. 5), 179–87; Ferrarini and Recine, (n. 5), 235; Köndgen and Theissen, (n. 5), 271 (specifically about Germany). 12 See European Commission, MiFID Impact Assessment (COM(2011) 656 final), p. 88 ff. 13 As regards the use of the description ‘core pillar of financial market integration’, see European Commission, 20 October 2011, COM(2011) 656 final, Proposal for a Directive of the European Parliament and of the Council on markets in financial instruments repealing Directive 2004/39/EC of the European Parliament and of the Council (Recast) (MiFID II), p. 1. 14 The Commission formally kick-started the process through a public hearing over two days on 20–21 September 2010, followed by a public consultation dated 8 December 2010. Cf. also the Commission’s impact assessment dated 20 October 2011 (SEC(2011) 1226 final), p. 7, § 2.1., accompanying the Commission’s initial proposal for MiFID II and MiFIR, p. 7, § 2.1. 15 Investing by participating in an investment fund is a form of collective investment. The invested capital is then raised by a number of parties, whose capital is then collectively invested by the investment manager. In other words, profits and losses are apportioned among the participants in proportion to the amount of their participation. On this subject, see e.g. Danny Busch and Lodewijk van Setten, ‘The Alternative Investment Fund

Managers Directive’ in Lodewijk van Setten and Danny Busch (eds) Alternative Investment Funds in Europe: Law and Practice (Oxford: OUP, 2014), pp. 1–122. 16 See Article 4(1) and (2) MiFID I in conjunction with Annex I, Sections A and C; Article 4(1) and (2) MiFID II in conjunction with Annex I, Sections A and C. 17 In practice this type of investment service is often termed ‘execution only’. This is however a confusing term, as in practice ‘execution only’ is also used to refer to the investment activity of dealing on own account (see main text). 18 See Annex I, Section A of both MiFID I and MiFID II. Most investment services and activities are defined in more detail. 19 See Annex I, Section A(9) MiFID II. 20 Title II MiFID I and MiFID II sets out the rules for investment firms, and Title III MiFID I and MiFID II the rules for regulated markets. 21 See Article 4(1)(15) in conjunction with Annex I, Section C MiFID I; Article 4(1)(15) in conjunction with Annex I, Section C MiFID II. See on the broadening of the scope of ‘financial instrument’ under MiFID II: § II sub (1), below. 22 See Article 5(1) MiFID I and Article 5(1) MiFID II. 23 See Article 6(3), (31), and (32) MiFID I and Article 6(3), (34), and (35) MiFID II. 24 Articles 9–13 MiFID I and Articles 9–16 MiFID II. 25 Article 16 MiFID I and Article 21 MiFID II. 26 Article 18 ff MiFID I and Article 23 ff MiFID II. 27 See Article 1(2) MiFID I and Article 1(3) and (4) MiFID II. 28 See also n. 17 above. 29 E. Benos, R. Payne, and M. Vasios, ‘Centralized trading, transparency and interest rate swap market liquidity: evidence from the implementation of the Dodd–Frank Act’, Bank of England Staff Working Paper, No. 580, January 2016. 30 EU CoJ 30 May 2013, no. C-604/11, AA (2013) 663, with note by Busch; JOR 2013/274, with note by Busch (Genil 48 SL and Other v Bankinter SA and Others). 31 EU CoJ 29 April 2015, no. C-51/13, AA (2015) 696, with note by Busch and Arons (Nationale-Nederlanden Levensverzekering Mij NV/Hubertus Wilhelminus van Leeuwen). 32 See on the (im)possibility of compliance with the flood of new regulatory requirements recently: V. Colaert, Normvlucht en systeemdwang in de financiële sector— Wetsnaleving in tijden van normatieve expansie (Acta Falconis VI) (inaugural lecture K.U. Leuven), Intersentia, Antwerpen/Cambridge 2015. 33 On the subject of better regulation, see for example the European Parliament, Briefing of September 2014, Hearings of European Commissioner-designate—Frans Timmermans, p. 2: ‘The aim of Better Regulation is to promote the simplification of EU law, reducing the administrative burden, especially for business, and to ensure respect for the principles of subsidiarity and proportionality.’ (Available at: .)

34

The Call for Evidence, the 288 responses from stakeholders, and the Commission’s summary of the responses are available at: . 35 See . 36 See .

PART II

INVESTMENT FIRMS AND INVESTMENT SERVICES

2 THE SCOPE OF MIFID II Kitty Lieverse

I. Introduction II. Investment Firms under MiFID II III. Credit Institutions IV. Investment Firms under the CRR V. Structured Deposits VI. General Exemptions to MiFID II Applicability VII. Trading on Own Account VIII. Dealings in Emission Allowances IX. Financial Instruments X. Insurance Companies and Insurance Intermediaries XI. Managers of Collective Investment Undertakings XII. Data-Reporting Service Providers XIII. Market Operators XIV. Conclusion

I. Introduction

[2.01] The basic provision on the scope of MiFID1 II is Article 1(1), stating that MiFID II applies to investment firms, market operators, data reporting services providers, and third-country investment firms providing investment services or performing investment activities through the establishment of a branch in the European Union. In this chapter, the focus will be on investment firms located in the EU.2

II. Investment Firms under MiFID II [2.02] MiFID II defines an investment firm as any legal person whose regular occupation or business is the provision of one or more investment services to third parties and/or the performance of one or more investment activities on a professional basis.3 This definition contains four elements that are relevant to determining the scope of MIFID II. [2.03] The first and most important element is the reference to (i) the provision of investment service to third parties and (ii) the performance of investment activities on a professional basis. These investment services and investment activities are defined by reference to Section A of Annex I to MiFID II, relating to any of the instruments listed in Section C of Annex I.4 As a result, the list of investment services and investment activities relating to any of the financial instruments listed in Section C are the primary tool for determining the scope of MiFID II. The list of financial instruments is further discussed below. [2.04] The investment services and activities listed in Section A of Annex I are: (1) Reception and transmission of orders in relation to one or more financial instruments. (2) Execution of orders on behalf of clients. This means: acting to conclude agreements to buy or sell one or more financial instruments on behalf of clients and includes the conclusion of agreements to sell financial instruments issued by an investment firm or credit institution at the moment of issuance.5

(3) Dealing on own account. This means: trading against proprietary capital resulting in the conclusion of transactions in one or more financial instruments.6 (4) Portfolio management. This means: managing portfolios in accordance with mandates given by clients on a discretionary clientby-client basis where such portfolios include one or more financial instruments.7 (5) Investment advice. This means: the provision of personal recommendations to a client, either upon its request or at the initiative of the investment firm, in respect of one or more transactions relating to financial instruments.8 (6) Underwriting of financial instruments and/or placing financial instruments on a firm committed basis. (7) Placing of financial instruments without a firm committed basis. (8) Operation of an MTF.9 (9) Operation of an OTF.10 [2.05] Compared to MiFID, there are two changes in this list of investment services and activities that are worth noting. The first one is that the operation of an OTF has been added as the ninth category on the list of investment services and activities. As a result, under MiFID II the operation of an OTF is subject to an authorization requirement and other supervisory requirements.11 [2.06] The second change relates to the definition of ‘execution of orders on behalf of clients’, which has now been extended to include ‘the conclusion of agreements to sell financial instruments issued by an investment firm or credit institution at the moment of issuance’.12 The background of this addition has been clarified in Recital 45, which states that MiFID II applies if an investment firm (including a credit institution) provides investment advice to its clients on financial instruments that it distributes and issues itself. If an investment firm distributes financial instruments issued by itself without providing any advice, such service should also be covered by MiFID II, according to Recital 45. This is effectuated by extending the definition of the service ‘execution of orders on behalf of clients’ to this effect. This leads to the question whether every

issuance of financial instruments constitutes an investment service by the issuer. As a starting point, it would seem unlikely that an issuer which is an ordinary commercial company would qualify as an investment firm solely because of the qualification of conclusion of agreements to sell financial instruments in the primary market as ‘execution of orders on behalf of a client’. Under Recital 45 it would seem that the European legislator rather aims to include ‘execution only’ distribution activities to clients, regardless of whether the financial instruments are issued by the investment firm/credit institution itself or by another issuer. As a result, the test would be to determine whether the distribution activities towards the clients would trigger the qualification ‘execution of orders on behalf of a client’ if the distribution is done by a firm that is not the issuer. If this is the case, the same qualification applies if the distribution is done by the issuer itself. In other words, the fact that the issuer of the financial instruments does not engage a third-party investment firm or credit institution for the distribution to its clients but performs such distribution itself would no longer mean that the distribution activities are not subject to MiFID regulation. The qualification of such distribution activity towards clients as ‘execution of orders on behalf of clients’ would seem more likely if the financial instruments are investment type products than in the case where the financial instruments are shares in the capital of the investment firm or credit institution, or corporate bonds. The investors that would purchase such shares or bonds could typically not be qualified as clients of the issuer. [2.07] As a result, the extension of the definition of ‘execution of orders on behalf of clients’ in my view does not mean that an investment firm (or credit institution) acting as issuer in the primary market would at all times simultaneously act as an investment firm, providing the service of order execution to a client, if a client subscribes for shares or other financial instruments issued by the investment firm. This is primarily based on the fact that this category of service implies the existence of a client relationship which follows from the definition itself: ‘execution of orders on behalf of clients’. As a result, if an investment firm/credit institution is solely acting in a capacity as issuer in the primary market and not simultaneously servicing clients in this regard, this extended definition would not be relevant. This requires that a determination must be made whether the relation between the investor and the investment firm/issuer is

to be qualified as one of servicing a client, that is, whether the investment firm/issuer is acting in the capacity of an investment firm towards clients when issuing and placing financial instruments. [2.08] Reference is also made to the definition of matched principal trading, as set out in Article 4(1), para. 38 MiFID II. ‘Matched principal trading’ means a transaction where the facilitator interposes itself between the buyer and the seller in such a way that it is never exposed to market risk throughout the execution of the transaction, with both sides executed simultaneously, and where the transaction is concluded at a price where the facilitator makes no profit or loss, other than a previously disclosed commission, fee, or charge for the transaction. Based on Recital 24, this system of matched principal trading should be viewed as ‘dealing on own account when executing client orders’ if in this way the investment firm executes orders from different clients. This manner of order matching is considered to be, under Recital 24, both dealing on own account and execution of orders on behalf of clients. [2.09] Based on these items, MiFID II has effectively extended the scope of the service ‘execution of orders on behalf of clients’ and has limited the scope of dealing on own account, in the sense that in the instances mentioned above an investment firm would both be dealing on own account and executing client orders. Based on the revisions in MiFID II just discussed, it is now clear that if the investment firm acts as issuer (of a financial instrument) and also as investment firm in a client relationship, the rules for order execution apply. In addition, if the investment firm acts as facilitator to match opposing client orders, this role as facilitator includes that the investment firm provides order execution services to each client, even though technically the investment firm in its role as facilitator would also be dealing on own account vis-à-vis each individual counterparty. [2.10] The existence of a client relationship is a defining element of any investment service.13 Based on the definitions above and the type of service these typically entail, items (1), (2), and (4)–(7) qualify as investment services, while items (3), (8), and (9) qualify as investment activities.

[2.11] The second element of the definition of investment firm is the reference to ‘legal person’. Although Recital 26 to MIFID II makes clear that any reference to persons should be understood to include both natural and legal persons, Article 4(1) para. 1 MiFID II makes clear that natural persons can only be included in the scope of the definition of ‘investment firm’ if: (i) their legal status ensures a level of protection for third parties’ interests equivalent to that afforded by legal persons, and (ii) they are subject to equivalent prudential supervision appropriate to their legal form.14 The possibility that natural persons qualify as investment firms and are authorized as such under MiFID II will not be further discussed in this chapter.15 [2.12] The third element entailed is that the conduct of the investment services or activities must constitute the regular business or profession of the firm. This is confirmed by Recital 12 which reads as follows: The purpose of this Directive is to cover undertakings the regular occupation or business of which is to provide investment services and/or perform investment activities on a professional basis. Its scope should therefore not cover any person with a different professional activity.

[2.13] This does not mean, however, that any incidental performance of an investment service or activity by a person with a different professional capacity would be per se outside the scope of MiFID II. This can be derived from the exemptions in Article 2(1)(c) and (k) MiFID II. These exemptions entail that persons who provide investment services in the course of another professional activity are excluded from the scope of MiFID II only if certain conditions are met. The first available exemption is for persons who provide investment services in an incidental manner in the course of a professional capacity if such capacity is regulated by legal or regulatory provisions or a code of ethics which do not exclude the provision of that service,16 provided that such activity is regulated and the relevant rules do not prohibit the provision, on an incidental basis, of investment services. The second exemption is for persons providing investment advice in the course of another profession (not covered by MiFID II) provided that the provision of investment advice is not specifically remunerated.17

[2.14] The fourth element is the addition of ‘third party’ as the addressee of the investment services. In my view, this reflects that the provision of investment services is by definition directed to a third party; a client. Simply said, the absence of a client relationship means that the relevant service does not qualify as an investment service. In Article 4(1), para. 9 MiFID II, ‘client’ is defined as any legal or natural person to whom an investment firm provides investment or ancillary services. As set out above, in the discussion of the scope of the various investment services, the element that a third party/client is involved is also part of the description or definition of the various individual services.18 Despite the fact that the absence of a client relationship would mean that the relevant service should not qualify as an investment service under MiFID II, there are a couple of exemptions to the scope of MiFID II that apply this principle, as will be further set out in what follows in the discussion of the general exemptions.

III. Credit Institutions [2.15] Credit institutions that provide investment services and/or investment activities technically (reading the definition set out above) qualify as investment firms but are subject to a revised scope of MiFID II, as set out in Article 1(3)(a–d) MiFID II. This is determined by the fact that credit institutions are primarily regulated19 by the Capital Requirements Directive, CRD IV, and the Capital Requirements Regulation, CRR.20 The MiFID II provisions provide a ‘top-up’ regulation for credit institutions to the extent that they provide investment services and activities. An additional authorization is however not required for credit institutions that wish to conduct investment services and/or investment activities.21

IV. Investment Firms under the CRR [2.16] As set out above, the MiFID II definition of an investment firm basically covers all persons who conduct investment services and activities which by definition relate to financial instruments as defined under MiFID

II. As will be further set out in Sections VI, VII, and VIII, there are a number of exemptions from the scope of MiFID II, for example for persons who are solely dealing on their own account, provided certain conditions are met. As a starting point, the scope of MiFID II also determines the applicability of prudential requirements on the basis of CRD IV22 and CRR.23 This is based on Article 4(1) under (2) of the CRR. For the definition of investment firm under the CRR reference is made to the definition in MiFID I (which should be read as a reference to MiFID II24). More specifically, in Article 4(1) under (2) of CRR, reference is made to a person as defined in point (1) of Article 4(1) of MiFID II, ‘which is subject to the requirements of that Directive’, while this provision subsequently excludes three categories of investment firms, as will be further set out in the next paragraphs. The reference to persons being subject to the requirements of MiFID II implies that persons that technically qualify as investment firms but are exempted from the scope of MiFID II are also as a starting point excluded from the prudential requirements under CRR. The additional three exclusions from the definition of an investment firm under CRR are: (a) credit institutions, (b) local firms, and (c) certain other investment firms depending on the scope of the services that they perform. [2.17] The exclusion of credit institutions logically follows from the fact that even though they technically qualify as investment firms if they conduct investment services and/or activities, they are already captured by CRD IV/CRR as their primary source of regulation, as set out in the previous paragraph. [2.18] The second exclusion relates to so-called local firms. Under Article 4(1), para. 4 CRR, a local firm means a firm dealing for its own account on markets in financial futures or options or other derivatives and on cash markets for the sole purpose of hedging positions on derivatives markets, or dealing for the accounts of other members of those markets and being guaranteed by clearing members of the same markets, where responsibility for ensuring the performance of contracts entered into by such a firm is assumed by clearing members of the same markets.

[2.19] The reference to ‘local’ should historically be read rather literally,25 but this additive has basically lost its meaning. The defining elements of a

local firm include that it is a firm dealing for its own account, but it may have external clients as the definition includes that it may also be ‘dealing for the accounts of other members of those markets’. The markets on which a local firm may deal are restricted (‘financial futures or options or other derivatives and on cash markets for the sole purpose of hedging positions on derivatives markets’). This specific definition of the concept of a local firm is for the CRR-exemption purposes only. It does not coincide with a MiFID II definition or exemption. As a result, a local firm within the meaning of CRR typically qualifies as an investment firm under MiFID II (for dealing on own account on a professional basis), and may or may not be exempt from requirements under MiFID II depending on the exact scope of its dealing on own account activities.26 [2.20] The third CRR exclusion from the MiFID II concept of an investment firm basically comprises investment firms that are not permitted to hold client money or securities (and for that reason may not at any time place themselves in debt to those clients) and provide certain MiFID II investment services only. These are the services (1), (2), (4), and (5) of Section A of Annex I. These services basically entail: receipt and transmission of orders, execution of orders, portfolio management, and investment advice. As a final condition, the investment firm, in order to be CRR exempt under this category, may not be authorized to provide the ancillary service safekeeping.27 An investment firm that fulfils these three cumulative requirements is an investment firm within the meaning of MiFID II, but not an investment firm (and thus it also does not qualify as an ‘institution’28) within the meaning of the CRR. [2.21] As will be further discussed in Section VII, compared to MiFID, the scope of certain exemptions under MiFID II, including the exemption for dealing on own account, is more restrictive. Based on the approach as set out above, the result of this is an extension of the prudential requirements under CRR to investment firms that were previously exempted under MiFID, but no longer have an exempted status under MiFID II. [2.22] In addition to the CRR-specific definition of an investment firm, as set out above (which is based on the MiFID II scope and provides for some

specific exemptions), there is also an additional relevance to the MiFID II list of investment services and activities in determining the initial capital and solvency requirements contained in CRD IV. As a starting point, each CRR investment firm must have initial capital of €730,000.29 A reduction to €125,000 of initial capital, however, applies to such a CRR investment firm that does not deal for its own account or underwrite issues of financial instruments on a firm commitment basis, but does hold client money and securities and performs one or more of the services as listed (receipt and transmission of orders, execution of orders, and/or portfolio management).30 Furthermore, local firms are made subject to a specific initial capital requirement of €50,000 if they are authorized under MiFID II and use a MiFID II passport.31 Finally, the investment firms that are exempt under CRR in the third category as reflected above32 are subject to an initial capital of €50,000 or €25,000.33 A somewhat comparable system applies to solvency requirements, in the sense that there are solvency requirements for CRR investment firms that apply by default,34 while specific requirements apply to other categories of CRR investment firms and also to some of the investment firms that are exempted from the CRR definition.35 [2.23] Notably, and as specified above, the list of investment services and activities under MiFID II co-determine the applicable initial capital and solvency requirements, in addition to the element whether or not the investment firm is permitted to hold client money or securities. On this latter element, it is noted that this element is added by CRD IV and CRR, without a clear link to the meaning of this element, in view also of the fact that under MiFID II asset segregation rules apply.36 The national implementation of these rules would typically entail that investment firms are in fact not allowed to hold client money and securities, which makes it somewhat unclear what this element in fact means.37 The link to the MiFID II scope means that the interpretation of the MiFID II services and activities not only determines the scope of the MiFID II authorization requirement and additional rules, but is equally important to determine the applicability of the prudential requirements under CRD IV and CRR. [2.24] In this regard, it is noted that in its Report on Investment Firms,38 the EBA has provided an analysis for its view that this link between the

MiFID II investment services and investment activities on one hand and the determination of the applicability of CRD IV/CRR, including the scope thereof, on the other, should be reconsidered as the MiFID II description of services and activities does not necessarily properly reflect the risk profile of the relevant investment firm and the appropriateness of prudential requirements. Ultimately, this position may lead to a situation where the CRD IV/CRR requirements will be reserved for ‘bank-like’ investment firms, while a tailored prudential system for other types on investment firms will be introduced.39 [2.25] As will be further discussed in Section VII, certain exemptions, including the exemption for dealers in commodity derivatives, are more limited under MiFID II than under MiFID I. Based on the system described above, this would bring such dealers within the scope of CRD IV/CRR. In this regard, it is worth noting that the EBA Report on Investment Firms recommends that the current waiver for certain prudential requirements for dealers in commodity derivatives40 remains in place until 31 December 2020.41 From the EBA Report it can be deduced that this is also the date that is linked to the views of the EBA on the proposed timing of a new prudential regime for investment firms as referred to earlier.

V. Structured Deposits [2.26] The fact that investment firms and credit institutions are regulated by MiFID II for their investment services and investment activities is no different from MiFID I. However, MiFID II has added so-called structured deposits to the scope of supervision for investment firms and credit institutions. Under MiFID II, structured deposits do not qualify as financial instruments, but nevertheless are subject to MiFID II regulation. The background of this change is clarified in the Recitals to MIFID II, where it is stated that structured deposits have emerged as a form of investment product but are not covered under any legislative act for the protection of investors at EU level, while other structured investments are covered by such legislative acts. In the interest of a level playing field and to enhance an adequate level of investor protection across the EU, it has been

considered appropriate, so the Recital states, to include structured deposits in the scope of MiFID II.42 This inclusion, however, only concerns investment firms and credit institutions ‘when selling or advising’ clients in relation to structured products. This means that selling or advising in respect to structured products by other parties is not subject to MiFID II. Also, other investment services relating to structured deposits (i.e. portfolio management) are not covered by this extension.43 [2.27] Structured deposits are defined in MiFID II44 as deposits45 which are fully repayable at maturity and under the terms of which interest or premium will be paid or is at risk, on the basis of a formula containing factors such as an index, a financial instrument, a commodity, and/or an exchange rate, or combination thereof. Structured deposits do not include deposits linked solely to interest rates, such as Euribor or Libor, regardless of whether or not the interest rates are predetermined, or whether they are fixed or variable.46 [2.28] The inclusion of structured products under MiFID II means that when investment firms and credit institutions are selling or advising clients in relation to structured deposits, this should be understood as their acting as intermediaries for those products issued by credit institutions that can take deposits in accordance with Directive 2013/36/EU.47 Article 1(4) MiFID II lists the MiFID II provisions that should be observed if distribution or advisory services in respect of such structured deposit are provided to a client. These provisions include the suitability test of Article 25(2) MiFID II or the appropriateness test of Article 25(3) MiFID II, depending on the service that is provided. 48 In the event the services are limited to execution of orders or the receipt and passing on of orders, the appropriateness test does not have to be performed.49 This exception is however not applicable if the structured products qualify as complex, as set out in Article 25(4) under (a)(v) MiFID II. These are products that incorporate a structure which makes it difficult for a client to understand the risk of return of the costs of exiting the product before the term has lapsed. Under Article 25(10)(b) MiFID II, it is the task of ESMA to develop and keep updated guidelines to determine which products fall under this description (and thus also require an appropriateness test if the services are

limited to order execution or the receipt and transmission of orders).50 Also, the provisions on product governance as included in Article 24 apply to structured deposits. In short, these provisions entail that investment firms and credit institutions that manufacture or distribute structured deposits are subject to the product-governance requirements, which include that a product approval process must be in place.51

VI. General Exemptions to MiFID II Applicability [2.29] As stated, the provision of investment services to clients and/or the conduct of investment activities on a professional basis, in each case to the extent relating to financial instruments, triggers the qualification as an investment firm and thus MiFID II applicability. MiFID II maintains the general exemptions that were provided by MiFID, however, narrowing the scope of the exemptions for some of them, as further set out in the next section. [2.30] Next to these general exemptions, Article 3 of MiFID II offers an optional exemption to Member States. Member States may choose not to apply MiFID II to persons for which they are the home Member State, provided that national authorization and regulation apply. This optional exemption may only be applied to persons that meet certain conditions52 and are made subject to certain local requirements.53 [2.31] The general exemptions to MiFID II include exemptions for the following categories:54 (i) Insurance companies;55 (ii) Persons providing investment service ‘intra-group’;56 (iii) Persons providing investment services incidentally or investment advice in the course of another professional activity, subject to some further conditions;57 (iv) Persons dealing on own account;58 (v) Operators dealing in emission allowances;59

(vi) Administrators of employee participation schemes, whether or not in combination with providing intra-group services;60 (vii) Public bodies;61 (viii) Pension funds and collective investment undertakings, including their depositaries and managers;62 (ix) Persons that can benefit from some Member States’ (Denmark, Finland, and Italy) specific exemptions;63 (x) Transmission system operators;64 (xi) Central Securities Depositories (CSDs).65 [2.32] Some of these exemptions are triggered by the fact that the relevant entity would conduct the investment services or activities in the context of business that is already regulated at the EU level through a different set of rules. This is true for the insurance companies’ exemption66 and also for the pension funds and collective investment undertakings, including their managers.67 [2.33] As noted, for credit institutions a system applies whereby the authorization and operation are governed by the EU law specific to credit institutions, while for any investment services and activities the MiFID II rules apply.68 A similar system is considered to apply to CSDs. As is reflected in the Recital (41) of MiFID II, CSDs are to be specifically regulated under EU law and are to be subject, in particular, to authorization and certain operating conditions. In the Recital it is noted that CSDs might, in addition to the core services referred to in other EU law, provide investment services and activities which are regulated under MiFID II. In order to ensure that any entities providing investment services and activities are subject to the same regulatory framework, it is appropriate, so it is stated in the Recital, to ensure that such CSDs are not subject to the requirements of this MiFID II relating to authorization and certain operating conditions but that EU law regulating CSDs as such should ensure that they are subject to the provisions of MiFID II when they provide investment services or perform investment activities in addition to the services specified in that EU law. Based on this reasoning, CSDs are exempt from MiFID II to the extent they are regulated under EU law on the basis of Article 2(1)(o) MiFID II. Reference is also made to the specific exemption

for transmission system operators: when carrying out tasks under specific EU law and in accordance with this specific regulation, they are exempt from MiFID II.69 [2.34] Other exemptions seem to be triggered by the fact that the services are not provided to ‘clients’, but rather to a pre-determined group of related persons. This is true for the exemption of persons who provide investment services consisting exclusively in the administration of employeeparticipation schemes and who therefore do not provide investment services to third parties.70 The limited scope of the clients (namely only the parent company, the other subsidiaries, or the own subsidiaries71 of such parent) is also a trigger for the so-called intra-group exemption.72 This exemption is in fact an application of the criterion that MiFID II aims to regulate the provision of services to third parties, while the group entities as described in this exemption (parent, other subsidiaries, own subsidiaries) could be considered not to be such third parties. This is also reflected in Recital 28: Persons who do not provide services for third parties but whose business consists in providing investment services solely for their parent undertakings, for their subsidiaries, or for other subsidiaries of their parent undertakings should not be covered by this Directive.

[2.35] Also, exemptions are available for persons with a different main professional activity to providing investment services or investment activities, subject to certain conditions.73 Exemption (h) is triggered by the public task of certain entities which may trigger the conduct of investment services or activities.74 [2.36] The other exemptions will be discussed in more detail in Sections VII, VIII, and IX.

VII. Trading on Own Account [2.37] Under MiFID II, the exemptions for trading on own account have been revised and reshaped. In general, the scope of the exemptions for trading on own account has been narrowed by MiFID II.75 As can be

derived from the Recitals, the following reasoning has been applied in redefining the exemptions for trading on own account. [2.38] As a starting point, it is expressed in the Recitals76 that persons administering their own assets and undertakings, who do not provide investment services or perform investment activities other than dealing on own account in financial instruments, should not be covered by MiFID II. There are, however, two important limitations to this general exception for dealing on own account. The first one is that a special regime applies for dealing on own account in the following financial instruments: (i) commodity derivatives and (ii) emission allowances or derivatives thereof, as set out in the next paragraph. The second limitation is that persons who deal on their own account are nevertheless covered by MiFID II, if they are market makers,77 members or participants of a regulated market or an MTF or have direct electronic access to a trading venue, apply a high-frequency algorithmic trading technique, or deal on own account when executing client orders. Based on this rationale, the exemption of Article 2(1)(d) MIFID II has been formulated. [2.39] Persons not covered by this exemption may be able to benefit from the exemption reflected in Article 2(1)(j) MiFID II. This exemption is available for (i) persons who deal on own account (including market makers) in commodity derivatives, emission allowances or derivatives thereof, but excluding persons who deal on own account when executing client orders, or (ii) persons who provide investment services (so no dealing on own account) in commodity derivatives or emission allowances or derivatives thereof to the customers or suppliers of their main business. These persons are not covered by the scope of MiFID II, provided that (a) that activity is an ancillary activity to their main business on a group basis, and that main business is neither the provision of investment services within the meaning of MiFID II nor of banking activities within the meaning of CRD IV, nor market making in commodity derivatives, and (b) those persons do not apply a high-frequency algorithmic trading technique. Notably, having direct electronic access to a trading venue is not an exclusion to this exemption.78 Technical criteria for when an activity is ancillary to such a main business should be clarified in regulatory technical standards, taking into account the criteria specified in MiFID II.79 As a

condition to this exemption, it is required to report the contemplated use of the exemption to the relevant authorities.80 The criteria set out above are aimed to ensure that non-financial firms dealing in financial instruments in a disproportionate manner compared with the level of investment in the main business are covered by the scope of MiFID II.81 [2.40] Exemptions (d) and (j) can be combined. This means that persons that deal in commodity derivatives, emission allowances, and derivatives thereof may also deal in other financial instruments as part of their commercial treasury risk-management activities to protect themselves against risks, such as exchange rate risks (and rely on exemption (d) for this activity).82 [2.41] Based on these exemptions, market makers in financial instruments (other than market makers in commodity derivatives, emission allowances, or derivatives thereof provided that their market-making activity is ancillary to their main business considered on a group basis and provided that they do not apply a high-frequency algorithmic trading technique) are always covered by the scope of MiFID II and cannot benefit from any exemption. The same applies to persons: (a) dealing on own account when executing client orders or (b) applying a high-frequency algorithmic trading technique. These persons are covered by the scope of MiFID II and cannot benefit from any exemption.

VIII. Dealings in Emission Allowances [2.42] A new exemption has been added to the list.83 This new exemption is for operators dealing in emission allowances. This exemption is triggered by the fact that under MiFID II (as further set out in paragraph 2.47) emission allowances have been added to the list of financial instruments. This brings all investment services and activities with regard to emission allowances into the scope of MiFID II as a starting point. However, under certain conditions an exemption is available. These conditions comprise: (i) the operator dealing in emission allowances is acting on the basis of compliance obligations under Directive 2003/87/EC, (ii) the transactions

constitute solely dealing on own account, and (iii) there is no execution of client orders and no application of HFT.

IX. Financial Instruments [2.43] As set out above, one of the most defining instruments to determine the scope of MiFID II is that MiFID II regulates investment services and activities to the extent only that these relate to financial instruments, which are defined by means of reference to Section C of Annex A to MiFID II. An exemption to this has been discussed above: even though structured deposits do not qualify as financial instruments, advising on and selling of these products by investment firms and credit institutions have been brought under the scope of MiFID II, at least to a certain extent. [2.44] On the demarcation of MiFID II by means of the reference to financial instruments as listed on Section C of Annex I, the following comments can be made. [2.45] First of all, an addition has been made to the list of financial instruments by including emission allowances. The background for this is clarified in Recital 11 to MiFID II, which states that a range of fraudulent practices have occurred in spot secondary markets in emission allowances which could undermine trust in the emissions trading scheme.84 One of the measures to address this is bringing emission allowances fully into the scope of MiFID II and MiFIR, by classifying them as financial instruments. As discussed above, a limited exemption has been made available for operators that deal on own account in emission allowances, subject to some further conditions. [2.46] Other parts of Section C of Annex A have remained unaltered. This is true for items (1), (2), (3), (8), and (9): (1) (2) (3) (8)

Transferable securities;85 Money-market instruments;86 Units in collective investment undertakings; Derivative instruments for the transfer of credit risks; and

(9) Financial contracts for differences. The other items ((4), (5), (6), (7), and (10)) have been reshaped. This can be clarified as follows. [2.47] Item (4) now contains a reference to emission allowances, which was previously included under item (10). As a result of this replacement, derivative contracts in respect of emission allowances do not qualify as ‘commodity derivatives’ as defined in Article 1(30) of MiFIR.87 Derivatives of emission allowances qualify as financial instruments (by means of their inclusion in item (4)), but do not qualify as commodity derivatives which provides a exclusion from the applicability of the new provisions on the determination of position limits and position management controls in respect of commodity derivatives.88 [2.48] In item (5) the reference to interest rate forward contracts has been replaced by the more general reference to forward contracts. [2.49] The type of commodity derivatives referred to in item (6) (physically settled commodity derivatives) is extended to those traded on an OTF. As a result, commodity derivatives traded on an OTF which can be physically settled qualify as financial instrument. However, outside the scope of the definition are wholesale energy products traded on an OTF that must be physically settled. The element ‘must be physically settled’ will be specified in a delegated act which, according to Recital 10, will take into account at least the creation of an enforceable and binding obligation to physically deliver, which cannot be unwound and with no right to cash settle or offset transactions except in the case of force majeure, default, or other bona fide inability to perform. The background to the exclusion of wholesale energy contracts is that these products are regulated by REMIT. This is confirmed in Recital 9 which states that the scope of financial instruments will include physically settled energy contracts traded on an OTF, except for those already regulated under REMIT.89 A special regime applies to a subcategory of instruments covered by item (6) of Section C, which are the so-called C6 energy derivative contracts.90 These are commodity derivatives relating to coal or oil that are traded on an OTF and must be physically settled. The C6 energy derivatives contracts are not

covered by REMIT and qualify as financial instruments. As a result, in respect of these financial instruments the provisions on position limits and transaction reporting apply upon the effectiveness of MiFID II and MiFIR. In respect of certain other obligations that are triggered by the qualification as financial instrument, Article 95 MiFID II provides for a transitional regime.91 [2.50] Item (7) of Annex C contains the remainder of commodity derivatives that can be physically settled, do not fall under item (6), are not for commercial purposes, and which have the characteristics of other derivative financial instruments.

X. Insurance Companies and Insurance Intermediaries [2.51] As set out above, the general exemption available to insurance companies remains unaltered under MiFID II. This exemption entails that insurance companies that provide investment services or conduct investment activities are not covered by MiFID II. Obviously, the potential for insurance companies to conduct any investment services or activities is limited by the fact that insurance companies may not engage in any other business than insurance business.92 This allows insurance companies to conduct all investment services and activities that may be deemed to be within the scope of the business of an insurance company (such as the investment of premium income in financial instruments) without applicability of MiFID II. [2.52] MiFID II, however, adds a new angle for so-called investmentbased insurance products. The background for this is clarified in Recital 87. In this Recital reference is made to the fact that investments can be made by clients through insurance contracts, and such a route to make investments would often been seen as an alternative or substitute for financial instruments regulated under MiFID II: To deliver consistent protection for retail clients and ensure a level playing field between similar products, it is important that insurance-based investment products are subject to appropriate requirements. Whereas the

investor protection requirements in this Directive should therefore be applied equally to those investments packaged under insurance contracts, their different market structures and product characteristics make it more appropriate that detailed requirements are set out in the ongoing review of Directive 2002/92/EC rather than setting them in this Directive. Future Union law regulating the activities of insurance intermediaries and insurance undertakings should thus appropriately ensure a consistent regulatory approach concerning the distribution of different financial products which satisfy similar investor needs and therefore raise comparable investor protection challenges. [2.53] Insurance-based investment products therefore still do not qualify as financial instruments. The revisions to the Insurance Mediation Directive93 prescribed by MiFID II, however, arrange for the applicability of certain MiFID II rules to such insurance contracts, based on the rationale that these insurance-based investment products are comparable, from the client’s perspective, to MiFID II-regulated investment products and thus the client should be offered a similar level of investor protection. For this purpose, MiFID II includes revisions to the IDD, incorporating a definition of insurance-based investment product. Based on this new definition, such product is basically an insurance product which offers a maturity of surrender value and where such value is wholly or partially exposed to market fluctuations. Non-life products and life products that provide for payable benefits only in the event of death or in respect of incapacity due to injury, sickness, or infirmity as well as certain pension products are excluded from the definition. [2.54] The additional rules included in the IDD for the insurance-based investment products include: (i) rules on conflict of interest, (ii) general principles relating to acting in the best interest of clients, and (iii) rules on information to clients. As an option, Member States may prohibit or limit the acceptance or receipt of fees paid or provided to insurance intermediaries or insurers by any third party (other than the client itself).

XI. Managers of Collective Investment Undertakings

[2.55] On the basis of Article 2(1)(i) MiFID II, collective investment undertakings as well as their managers and the depositaries are exempted from MiFID II. These collective investment undertakings cover the alternative investment funds as regulated under the AIFMD94 as well as the undertakings for collective investments in securities.95 Based on Recital 34, this exemption is based on the fact that the collective investment undertakings and their managers are subject to specific rules directly adapted to their activities. From this reasoning, it can be deduced that the exemption from MiFID II applies to any investment services and activities typical to the business of the manager of a collective investment undertaking, such as the execution of orders in units in the collective investment undertaking. [2.56] Notably, the specific Directives applicable to collective investment undertakings and their managers allow the managers to be authorized for the provision of certain investment services. On the basis of Article 6 AIFMD,96 Member States may authorize an external manager of a collective investment undertaking under the AIFMD to provide, inter alia, management of portfolios of investments, investment advice, and reception and transmission of orders in relation to financial instruments. In that case, certain MiFID II provisions should be made applicable.97

XII. Data-Reporting Service Providers [2.57] MiFID II introduces supervision on a new group of servicers, namely providers of data-reporting services. Based on Section D of Annex I to MiFID II, these services include three different categories, namely: (i) operating an APA, (ii) operating a CTP, and (iii) operating an ARM. These services are defined in Article 4, para. 1 MiFID II. APA or ‘Approved Publication Arrangement’ means a person authorized under MiFID II to provide the service of publishing trade reports on behalf of investment firms pursuant to Articles 20 and 21 of MiFIR.98 CTP or ‘Consolidated Tape Provider’ means a person authorized under MiFID II to provide the service of collecting trade reports for financial instruments listed in Articles 6, 7, 10, 12, and 13 of MiFIR from market operators and APAs and consolidating

them into a continuous electronic live data stream providing price and volume data per financial instrument.99 ARM or ‘Approved Reporting Mechanism’ means a person authorized under MiFID II to provide the service of reporting details of transactions to competent authorities or to the ESMA on behalf of investment firms.100 [2.58] The supervision that is introduced by MIFID II on these providers includes that they are subject to prior authorization by their home Member State.101 In addition, requirements apply to the management body of a data reporting-service provider.102 Organizational provisions apply which are specific to the type of data-reporting service provider.103 [2.59] The background to this new supervision introduced by MiFID II on data service providers is the renewed focus on the importance of transparency, both pre and post trade. MiFID II extends the transparency requirements,104 and the simultaneous introduction of supervision on data service providers underlines the importance that accurate and comprehensive information on market data is to be available to the users of such data, which include the supervisors.105

XIII. Market Operators [2.60] The scope of MiFID II in respect of market operators has basically remained unchanged, with the exception of adding the operation of an OTF to the list of investment services and activities that trigger an authorization requirement and applicability of MiFID II requirements.106

XIV. Conclusion [2.61] MiFID II introduces several changes to the scope of the supervision of investment firms and some other business related to the investment services industry, including data-reporting service providers. The list of investment services is extended by adding the operation of an OTF. Also,

the revision and—in respect of some items—extension of the definition of ‘financial instrument’ has an impact on the scope of MiFID II supervision. This includes emission allowances. In addition, advisory and distribution services in respect of structured deposits by investment firms and credit institutions are brought within the scope of MiFID II. Furthermore, supervision comparable to the MiFID II system is being introduced for insurers and intermediaries in respect of investment-based insurance products. At the same time, the scope of some of the exemptions to MiFID II is revised, with an impact in particular for dealing on own account. Dealing on own account through a high-frequency algorithmic trading technique is brought within the scope of MiFID II, and the broad exemption for traders in commodities derivatives has also been limited. As a result, the scope of MiFID II is considerably broader than MiFID I. This has an effect beyond the applicability of MiFID II regulation, as currently the prudential requirements are also linked as a starting point to the MiFID II qualification —however, with specific exemptions. In addition, the concept of linking prudential supervision to the MiFID II qualification, rather than to the prudential risk profile of the underlying services and activities, is subject to review and reconsideration.

1

MiFID I stands for the Markets in Financial Instruments Directive (Directive 2004/39/EG). MiFID II stands for the Markets in Financial Instruments Directive II (Directive 2014/65/EG). 2 For this chapter, use has been made of preparations for a publication on the scope of MiFID II with a focus on the Dutch law implementation by Kitty Lieverse and Marije Louisse in ‘De reikwijdte van MiFID II’, Tijdschrift voor Financieel recht, December 2015, pp. 437–46. 3 Article 4, para. 1(1) MiFID II. 4 Article 4, para. 1(2) MiFID II. Financial instruments are defined in Article 4, para. 1(15) MiFID II. 5 Article 4, para. 1(5) MiFID II. 6 Article 4, para. 1(6) MiFID II. 7 Article 4, para. 1(8) MiFID II. 8 Article 4(1), para. 4 MiFID II. 9 MTF stands for Multilateral Trading Facility. 10 OTF stands for an Organised Trading Facility, defined in Article 4(23) MiFID II as a multilateral system which is not a regulated market or an MTF and in which multiple third-

party buying and selling interests in bonds, emission allowances, structured financial products, or derivatives are able to interact in the system in a way that results in a contract in accordance with Title II of MiFID II. 11 This change to the scope of MiFID II is not further discussed in this chapter. 12 Article 4(1) para. 5 MiFID II. 13 Based on the definition in Article 4(1), para. 1 MiFID II. 14 If a natural person is an investment firm and provides the service of holding thirdparty funds or transferable securities, Article 4(1) MiFID II requires that additional items listed as (a–d) are complied with. 15 Additional requirements for natural persons acting as investment firms can be found in Article 5(4)(b) MiFID II and Article 9(6) MiFID II. 16 Article 2(1)(c) MiFID II. 17 Article 2(1)(k) MiFID II. 18 The element of a ‘client’ is included in the description of service under (2) and in the definition of service under (4) and (5). Services (6) and (7) in my view imply that these underwriting and placing services are typically provided for the relevant issuer as a client. 19 Reference is made to Recital 38 to MiFID II. The credit institution-specific regulation is included in CRD IV: Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC, EU L 176/338 (Activity of Credit Institutions Directive); and Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012, EU L 176/1. 20 Regulation (EU) 573/2013, EU L 176/1. 21 Article 1(3) MiFID II does not refer to the authorization requirements under MiFID II. 22 Directive 2013/36/EU of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms (CRD IV), EU L 176/338. 23 In fact, on this basis the scope of MiFID II is also a relevant factor for determining the scope of other pieces of legislation, including the Banking Resolution and Recovery Directive. 24 Article 94 MiFID II. 25 This is based on the EBA Report on Investment Firms of December 2015: Response to the Commission’s Call for Advice of December 2014, EBA/Op/2015/20. In this report interesting background on the concept of a local firm is provided: Historically, when there was still a physical presence or ‘open market outcry’ on an exchange, futures (derivatives) trading floors included a type of individual trader, known as the ‘local’, who made their living by buying and selling and seeking a profit from the spread. In effect, they were market makers, providing intra-day liquidity to

the market. With trading taking place physically on the floor of the exchange, ‘locals’ sought to place themselves in close proximity to the traders working for the major proprietary houses, so that they could benefit through such relationships and from being part of larger orders. They were, therefore, genuinely local in terms of their presence and their business with the relevant exchange. On further developments in this regard, the EBA reports continues: The assumption has been that local firms are small enterprises with a limited risk profile. The definition of ‘local firm’ in the CRR has remained virtually unchanged from the definition in the original version of the CAD in 1993. Local firms were, indeed, originally small firms (usually self-employed traders), only active on a single exchange. However, over time these small trading firms have grown, resulting in firms that are, to all intents and purposes, no longer local despite still being able to qualify under the ‘local firm’ definition. Some local firms are active in multiple jurisdictions and across all continents, employing hundreds of employees and having average trading positions of several billion. The assumption that a local firm is small and relatively low risk is, therefore, becoming more and more obsolete. 26

As further discussed below, see: Article 2(1)(d) and (j) MiFID II. Article (4)(1), para. 2 under c, CRR. 28 Article (4)(1), para. 3 CRR. 29 Article 28(2) CRD IV. 30 Article 29(1) CRD IV. Article 29(2) CRD IV provides for a Member State option to reduce this amount to €50,000 where a firm is not authorized to hold client money or securities, to deal for its own account, or to otherwise underwrite issues on a firm commitment basis. 31 Article 30 CRD IV. 32 Article 4(1), para. 2c, CRR. 33 Article 31(1) CRD IV and Article 31(2) CRD IV for those investment firms in this category that are also subject to registration under the Insurance Mediation Directive (Directive 2002/92/EC of the European Parliament and of the Council of 9 December 2002 on insurance mediation, OJ L 9/3), in both cases with the option to wholly or partially seek replacement by means of professional indemnity insurance as set out under items (b) and (c) of this provision. 34 Article 92 CRR. 35 Articles 95 and 96 CRR. A useful overview is provided in the EBA Report on Investment Firms (2015), on p. 15 and also on p. 17. 36 Article 16(8) and (9) MiFID II. 37 As is also noted in the EBA Report on Investment Firms (2015), p. 14. 38 EBA/Op/2015/20. 27

39

Reference is made to para. 2.5.3 of the EBA Report on Investment Firms and the recommendations, the analytical review of investment firms’ risk in Chapter 3, and the recommendation resulting therefrom in Chapter 4. On this topic the two main recommendations are as follows. Recommendation 1: ‘Recommendation for a new categorisation of investment firms distinguishing between systemic and “bank-like” investment firms to which the full CRD/CRR requirements should be applied; other investment firms (“non-systemic”) with a more limited set of prudential requirements; and very small firms with “non-interconnected” services.’ Recommendation 2: ‘Recommendation for the development of a prudential regime for “non-systemic” investment firms.’ 40 Articles 493 and 498 CRR, relating to the large exposures and capital adequacy provisions, currently set to expire by 31 December 2017. 41 Reference is made to recommendation 3 on p. 87 of the EBA Report on Investment Firms. 42 Reference is made to Recital 39. 43 This is based on the wording of Article 1(4) MiFID II, with the proviso that the terminology ‘selling’ is somewhat vague in the MiFID II context. This is further discussed below in respect of the scope of the suitability and appropriateness test. 44 Article 4(1), para. 43, MiFID II. 45 As defined in point (c) of Article 2(1) of Directive 2014/49/EU of the European Parliament and of the Council of 16 April 2014 on deposit guarantee schemes, OJ L 173, 12.6.2014, 149–78. 46 Article 4(1), para. 43 under a, MiFID II. 47 Reference is made to Recital 40 to MiFID II. 48 The terminology of Article 25 MiFID II is not tailored to the services in respect of structured deposits as mentioned in Article 4(1) of MiFID II (selling or advising) that have been made subject to MiFID II. I would understand the reference to either the suitability test or appropriateness test to be determined by whether or not the involvement of the investment firm or credit institution when distributing structured deposits includes advice. If so, the suitability test applies. Otherwise, when ‘selling’ without advice the appropriateness test applies. 49 Article 25(4), under (a)(v) MiFID II. 50 Reference is made to the Commission Delegated Regulation which is to be determined on the basis of the ESMA RTS on Complex debt instruments and structured deposits, on the basis of Article 25(10)(b) MiFID II. 51 Article 1(4), Article 9(3), Article 16(3), and Article 24 MiFID II. 52 These are the cumulative conditions listed under Article 3(1)(a–c), or (d) or (e) MiFID II. 53 Article 3(1)(2) MiFID II. 54 Article 2 MiFID II. 55 Article 2(1)(a) MiFID II.

56

Article 2(1)(b) MiFID II. Article 2(1) under (c) and (k) MiFID II. 58 ibid under (d) and (j) MiFID II. 59 ibid under (e) MiFID II. 60 ibid under (f) and (g) MiFID II. 61 ibid under (h), Article 2, para. 2 MiFID II. 62 ibid under (i) MiFID II. 63 ibid under (l) and (m) MiFID II. 64 ibid under (n) MiFID II. 65 ibid under (o) MiFID II. 66 Insurance companies are regulated at the EU level through Directive 2009/138/EU of the European Parliament and of the Council of 25 November 2009 on the taking-up and pursuit of the business of Insurance and Reinsurance (Solvency II); reference is made to Recital 27 to MiFID II. 67 Reference is made to Recital 34 to MiFID II. Collective investment undertakings, or at least their managers, are regulated by means of Directive 2011/61/EU of the European Parliament and of the Council of 8 June 2011 on Alternative Investment Fund Managers and amending Directives 2003/41/EC and 2009/65/EC and Regulations (EC) No 1060/2009 and (EU) No 1095/2010 (AIFMD); or Directive 2009/65/EU of the European Parliament and of the Council of 13 July 2009 on the coordination of laws, regulations and administrative provisions relating to undertakings for collective investment in transferable securities, OJ L 302/32 (UCITS Directive). 68 Article 1(3) MiFID II. 69 Article 2(1)(o), as explained in Recital 41. 70 This applies to the exemption of Article 2(1) under (f) and (g) MiFID II. 71 It is noted that the concepts ‘parent’ and ‘subsidiaries’ are defined in Article 4(1), paras. 32 and 33 MiFID II. 72 Article 2(1) under (b) MiFID II. 73 Reference is made to Article 2(1) under (c) and (k) and the comments made above under ‘Investment firm under MiFID II’. 74 Reference is made to Recital 32: 57

It is necessary to exclude from the scope of this Directive central banks and other bodies performing similar functions as well as public bodies charged with or intervening in the management of public debt, which concept covers the investment thereof, with the exception of bodies that are partly or wholly state-owned the role of which is commercial or linked to the acquisition of holdings. See also Recital 33 for the limitation of the exclusion of the ESCB, etc. from the scope of MiFID II.

75

Article 2(1)(k) of MiFID, the commodity dealer exemption, has been deleted and Article 2(1)(i) MiFID, the ancillary exemption, has been narrowed and replaced to Article 2(1)(j) MiFID II, while the dealing on own account exemption of Article 2(1)(d) MiFID has also been amended. 76 Reference is made to Recitals 18, 19, and 20 to MiFID II. 77 ‘Market maker’ is defined in Article 4(1)(7) MiFID II as a person who holds himself out on a continuous basis as being willing to deal on own account by buying and selling financial instruments against that person’s proprietary capital at prices determined by that person. 78 As set out above, this is different for the exemption mentioned in Article 2(1)(d) MiFID II, as set out under item (iii). 79 Reference is made to Article 2(4) MiFID II. On the basis of this provision ESMA has published a RTS: Criteria for establishing when an activity is to be considered to be ancillary to the main business. 80 Article 2(1)(j), final sentence, MiFID II. On the contents of these RTSs to determine whether an activity is ancillary at group level, it is stated in Recital 20, para. 2 to MiFID II: In doing so, those criteria should take at least into consideration the need for ancillary activities to constitute a minority of activities at group level and the size of their trading activity compared to the overall market trading activity in that asset class. It is appropriate that where the obligation to provide liquidity on a venue is required by regulatory authorities in accordance with Union or national laws, regulations, and administrative provisions or by trading venues, the transactions entered into to meet such an obligation should be excluded in the assessment of whether the activity is ancillary. 81

Reference is made to Recital 20, second paragraph, to MiFID II. Recital 22 to MiFID II. 83 Article 2(1)(e) MiFID II. 84 Set up by Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC, OJ L 275/32. 85 As defined in Article 4(1)(44) MiFID II. 86 As defined in Article 4(1)(17) MiFID II. 87 This definition refers to an underlying value referred to in item (10) of Section C of Annex I which now no longer includes emission allowances. 88 Title IV MiFID II, Articles 57 and 58. 89 Regulation (EU) No 1227/2011 of the European Parliament and the Council of 25 October 2011 on wholesale energy market integrity and transparency, OJ L 326/1. 90 Reference is made to Article 4(1)(16) MiFID II. 82

91

This applies to the clearing obligation and margining requirements under Regulation 648/2012. 92 Reference is made to Articles 15 and 16 of Solvency II. 93 Replaced by the Insurance Distribution Directive (IDD) of 20 January 2016, Directive 2016/97/EU of the European Parliament and of the Council of 20 January 2016 on insurance distribution (recast), OJ L 26/19. 94 AIFMD (2011/61/EU). 95 UCITS (2009/65/EU). 96 A comparable provision is included in Article 6 of Directive 2009/65/EU (UCITS Directive). 97 Reference is made to Article 6(6) AIFMD which states that Article 2(2) and Articles 12, 13, and 19 of MiFID II shall apply to the provision of the services referred to in para. 4 of this Article 6 by managers. 98 Article 4(1)(52) MiFID II. 99 Article 4(1)(53) MiFID II. 100 Article 4(1)(54) MIFID II. 101 Article 59(1) MiFID II. 102 Article 63 MiFID II. 103 Article 64 for APAs, Article 65 for CTPs, and Article 66 for ARMs. 104 For example, while MiFID I limited the pre- and post-trade transparency to equity instruments, MIFID II also extends as a starting point to non-equity instruments such as bonds (but some waivers are allowed). Reference is made to II and III of MiFIR. 105 Reference is made to Recitals 115 and 116 to MiFID II. 106 Article 4(1)(23) MiFID II, which adds the definition of an OTF, which qualifies as a trading venue, in Article 4(1)(24). OTFs are subject to specific requirements; reference is made for example to Article 20 MiFID II.

3 GOVERNANCE OF INVESTMENT FIRMS UNDER MIFID II Jens-Hinrich Binder*

I. Introduction II. Governance-related Regulation of Investment Firms between Prudential and Conduct-of-Business Regulation 1. The Case for Governance-Related Regulation According to MiFID II—Some Preliminary Observations 2. Governance-Related Regulation since the ISD 1993 3. Some Preliminary Conclusions III. 1. 2. 3. 4. 5. 6.

The Technical Framework under MiFID II General Principles and Problems The Board Organization and Risk Management Shareholders and Owners with Qualifying Holdings Governance Reporting Assessment

IV. Conclusions

I. Introduction

[3.01] The regulation of the governance of investment firms—for present purposes to be understood as referring to the internal corporate governance, that is, the rules and arrangements relating to the board, the organization, and internal procedures1—is by no means a post-financial-crisis phenomenon, and MiFID II is not the first European law instrument to prescribe relevant requirements in this regard. In fact, their history reaches back beyond MiFID I2 and, to a considerable extent, has been part of the regulatory framework for the prudential supervision of banks and investment firms generally. Long before the financial crisis reinforced concerns about the management of financial intermediaries, the first components of the regulatory concept were already laid down in the Investment Services Directive (ISD) of 19933—then, as continues to be the case to date, in the form of fundamental prudential requirements that must be met both as a precondition for authorization as a licensed investment firm and on a day-to-day basis in the course of the firm’s operations.4 In addition, non-bank investment firms have been subject to the governancerelated provisions applicable to banks in particular after the implementation of the Basel II capital accord in Europe from 2006.5 In short, governancerelated regulations have been part and parcel of the European framework regulation of investment intermediaries from the start, as a precondition for authorization and a complement to conduct-of-business regulation. [3.02] Against this backdrop, a superficial observer could be forgiven for concluding that governance-related regulatory requirements appear to have been both an uncontroversial and a rather immutable component of the relevant EU framework for the setting-up and operation of financial intermediaries engaged in the business of investment services to date. On closer inspection, however, neither holds true. As will be discussed in detail in subsection II below, both the rationale of governance-related regulatory requirements and their technical content have changed significantly over time. It is also as a result of this rather complex history that some of the technical regulatory concepts, reinforced and expanded under MiFID II, continue to pose significant challenges to the diverse range of financial intermediaries that are subject to these requirements across the European Union. Again, as will be discussed below, these challenges are reinforced by the absence of clear-cut policy foundations, which has been a

characteristic of the regulatory framework for the pursuit of investment services ever since the adoption of the ISD 1993. [3.03] The present chapter seeks to explore both the relevant policy background and the technical content of governance-related regulation of investment firms under MiFID II and, in addition, secondary legislation complementing the Level 1 requirements,6 and the CRD IV package, which extends not just to credit institutions but also to investment firms. Section II below will first analyse the rationale for governance-related regulation against the backdrop of the historic emanation of governance-related requirements for investment firms in European financial regulation. Section III then examines the technical features of the new regime, and identifies some challenges for transposition and enforcement. Section IV concludes.

II. Governance-related Regulation of Investment Firms between Prudential and Conduct-of-Business Regulation 1. The Case for Governance-Related Regulation According to MiFID II—Some Preliminary Observations [3.04] Judging from the preamble to MiFID II, the case for governancerelated regulation would appear to be rather straightforward. In addition to the general policy objective to combat ‘weaknesses in the functioning and in the transparency of financial markets’,7 the Directive expressly identifies ‘weaknesses in corporate governance in a number of financial institutions, including the absence of effective checks and balances within them’, as a ‘contributory factor to the financial crisis’, and then continues to set out the rationale for regulatory interference by arguing that [e]xcessive and imprudent risk taking may lead to the failure of individual financial institutions and systemic problems in Member States and globally. Incorrect conduct of firms providing services to clients may lead to investor detriment and loss of

investor confidence. In order to address the potentially detrimental effect of those weaknesses in corporate governance arrangements, Directive 2004/39/EC should be supplemented by more detailed principles and minimum standards.8

[3.05] Starting from this wording, governance-related regulation of investment firms can thus be said to serve a dual objective, namely (i) the prevention of failures of financial institutions with a view to protecting systemic stability, and (ii) the prevention of losses to investors and the protection of investor confidence. On the surface, this is hardly surprising, as it merely reflects the traditional set of—interrelated—objectives of financial regulation generally: the preservation of the economic functions of financial markets (in particular, the effective allocation of capital within a market economy), and the protection of customers (whose confidence is essential for the preservation of financial stability).9 Irrespective of this general background, however, the relevance of the above statement of policy objectives goes far beyond stating the obvious. It is worth noting that, although the relevant regulations under MiFID I are expressly referred to as a starting point for the reform, systemic stability considerations, for the first time in the history of European securities regulation, are expressly mentioned as a key driver for the further refinement of the relevant requirements. This takes up corresponding considerations first laid down in the Commission’s 2010 Green Paper on the governance of financial institutions.10 By contrast, as will be examined in further detail below, neither the ISD 1993 nor MiFID I made any reference to such considerations in the context of governance-related regulation. To be sure, neither of these earlier instruments were very outspoken with regard to the motives for governance-related regulation. This was consistent with the original objective to realign the prudential requirements pertaining to nonbank investment firms with those applicable to credit institutions in the interest of equal market access,11 but indicates that the rationale for regulatory intervention was rather obscure under the predecessors of the present regime. [3.06] In view of lessons learnt from the global financial crisis, the reinforcement and expansion of governance-related requirements by MiFID II therefore reflect not just a desire for further refinement of existing regulatory requirements and strategies but also, importantly, a shift in the

underlying policy objectives: In addition to the protection of investors and investor confidence, systemic considerations have entered the scene—and seem to play a more important role than the traditional focus on investor protection as such. To be sure, this development reacts to widespread concerns about the quality of governance arrangements in financial intermediaries generally. These have arisen as a result of (well-founded or perceived) lessons learnt from the global financial crisis and have inspired not just the substantive overhaul of the governance-related provisions in the MiFID regime but also, and to a larger extent, the corresponding reforms of governance-related requirements for banks under the CRD IV package of 2013.12 As illustrated, in particular, by the Lehman Brothers failure in September 2007, systemic implications triggered by the insolvency of financial intermediaries are not necessarily confined to licensed banks. They arise out of the relevant firm’s size, its connectedness to other market participants and to providers of market infrastructure, and the complexity of its operations and legal structure, as well as the market perception of the relevance of the firm—rather than out of the nature of its business (deposittaking, provision of investment services) as such.13 On closer examination, however, the available evidence is mixed. While it is generally agreed that risk management systems failed to mitigate excessive risk-taking in financial intermediaries in the run-up to the crisis,14 empirical findings with regard to other aspects of corporate governance are, at best, inconclusive.15 In particular, it is far from settled that those corporate governance standards that had been established already prior to the financial crisis can be said to have contributed to safer organizational arrangements.16 Specifically to what extent prudential concerns about firm soundness and, ultimately, financial stability implications are warranted in the case of (different types of) investment firms is still rather uncharted territory. As international regulatory standards, in response to the global financial crisis, have understandably focused on systemically important financial institutions,17 it is, in principle, hardly surprising that the relevant provisions under MiFID II and secondary legislation, as a rule, are intended to apply indiscriminately across the full range of investment firms operating within the European Union. Given the high level of diversity of market participants in terms of size, business models and interconnectedness with counterparties, the underlying rationale is nonetheless somewhat

questionable.18 Against this backdrop—and somewhat in contrast to the exposition of the policy rationale expressed in the preamble to MiFID II— the case for governance-related regulation of financial intermediaries in general thus seems to be less clear than might be expected in view of the wave of failures of investment banks experienced globally in recent years. The same applies, and to a greater extent, to the technical design of the reformed regulatory framework under MiFID II.

2. Governance-Related Regulation since the ISD 1993 [3.07] To understand the full dimension of the considerations presented above, it is appropriate to explore the historical evolution of governancerelated regulations for investment firms in some more detail. In this context, the political objective to realign the regulatory frameworks for banks and non-bank financial intermediaries with the ISD 1993 deserves particular attention (infra II.2.A). Arguably, neither this Directive nor the subsequent amendments to existing requirements under MiFID I (infra II.2.B) fully acknowledged the functional parallels and differences between banking regulation, from which the relevant concepts were adapted, on the one hand, and investment services on the other hand.

A. Governance-Related Regulation under the ISD 1993 [3.08] In the context of governance-related requirements, specific references to the objective of investor protection in the ISD 1993 are rare. To be sure, the Preamble did acknowledge the need ‘to protect investors and the stability of the financial system’19 as key policy objectives generally. In addition, it argued that […] it is necessary, for the protection of investors, to guarantee the internal supervision of every [investment] firm, either by means of two-man management or, where that is not required by this Directive, by other mechanisms that ensure an equivalent result.20

[3.09] Beyond these few—and erratic—remarks, however, the rationale for governance-related regulations received no further explanation. Instead,

the focus clearly was on the equal treatment of banks and investment firms under the new framework.21 As such, the rationale for governance-related prudential requirements in the industry—as distinct from corresponding regulations for credit institutions in the technical sense—was mentioned neither in the Directive nor in contemporary literature. Consequently, governance-related regulation under the ISD 1993 remained somewhat haphazard, and limited to very few specific requirements. [3.10] Among these, the requirement that ‘the direction of a firm’s business must be decided by at least two persons’ of ‘sufficiently good repute and […] sufficiently experienced’22 was an early predecessor of today’s fit-and-proper regime,23 modelled after the corresponding provisions in the Second Banking Directive of 1989.24 In addition, the Directive prescribed a regime for supervisory scrutiny of owners of qualified holdings in investment firms,25 which again took up earlier precedents established in the area of banking regulation.26 [3.11] By comparison with subsequent developments, the ISD 1993 did not prescribe any substantive requirements with regard to business models or organizational issues. As part of the licensing process, however, enterprises seeking to be recognized as investment firms had to submit a ‘programme of operations setting out inter alia the types of business envisaged and the organizational structure of the investment firm’.27 While this ensured access of supervisory authorities to the relevant information, the definition of more specific requirements was left to the Member States, which were to ‘draw up prudential rules which investment firms shall observe at all times’,28 a serious and ‘systematic’ infringement of which could result in the revocation of the respective firm’s licence.29 Specifically, investment firms, in this context, were to be required to have sound administrative and accounting procedures, control and safeguard arrangements for electronic data processing, and adequate internal control mechanisms including, in particular, rules for personal transactions by […] employees.30

[3.12] In addition, Member States had to draw up requirements with regard to the protection of client assets and funds as well as the

documentation of transactions, and had to prescribe institutional arrangements designed to minimize the impact of conflicts of interest on clients.31 It is in this regard that the ISD 1993 went beyond the level of governance-related requirements already present in contemporary European banking regulation, reflecting earlier US American precedents.32 And it is here that the development of a specific organizational regime for investment firms in European financial regulation, distinct from governance-related requirements for banks, can be said to have its roots—a regime that goes beyond a mere replication of the organizational requirements for banks and specifically addresses the agency problems characteristic for investment firms which, because of the wide range of different types of investment services, are more complex than within the relationship between a bank and depositors.33 In these provisions, the functional link between prudential regulation and conduct-of-business regulation,34 discussed above, is particularly obvious. However, it is also in this regard that the lack of a consistent rationale for governance-related regulation of investment firms in early European regulation is evident. While investor protection is clearly the key motive for most of the organizational requirements stipulated by Article 10, the case for the remaining provisions examined above—and the rationale for the substantive synchronization of governance requirements for investment firms with those for banks generally—remains somewhat obscure. At least for the early stages of European regulation of investment firms and, specifically, the ISD 1993 it can only be explained on the grounds of the general motive to create a separate but equal single licence for the non-bank securities firms while ensuring that they do not have a competitive advantage in terms of funding requirements vis-à-vis credit institutions when such institutions are providing investment firms.35

[3.13] In other words, in addition to the general objective of investor protection, concerns about the conditions for fair competition among different types of providers of investment services—more specifically, universal banks and non-bank investment firms—and considerations of equal market access across the Common Market rather than concerns for financial stability, have to be considered as a key rationale for the early regulatory framework for the governance of investment firms.

B. Governance-Related Regulation under MiFID I and the CRD II package [3.14] In principle, the substitution of the ISD 1993 by MiFID I as such hardly changed this picture. To be sure, the new Directive, just as its predecessor, quoted both the protection of investors and the ‘stability of the financial system’ as key policy objectives.36 Whether and to what extent financial stability concerns actually warranted regulation of the corporate governance of investment firms was not discussed, however. Just as under the ISD 1993, governance-related requirements under MiFID I, in principle, mainly reflect investor protection concerns and, to the extent that they replicate corresponding requirements in banking regulation, merely continue to synchronize both regimes with a view to ensuring a level playing field between universal banks and specialized investment firms.37 In principle, the Directive carried forth the main elements of governancerelated regulations already established by the ISD 1993. Directors and shareholders as well as members with qualifying holdings continued to be subject to suitability tests in order to prevent undue influence on the firms’ management.38 In terms of organizational requirements, Article 13 also took up the requirements established by the ISD 1993, but with some alterations to the catalogue stipulated by Article 10 ISD 1993 that reflected more specific concerns about operational risks.39 [3.15] Notwithstanding the continuity of substantive regulations visible in the wording of ISD 1993 and MiFID I, the latter clearly triggered a conceptual departure from the earlier instrument in that it formed the basis for extensive Level 2 legislation on organizational requirements and operating conditions.40 In addition to these rather general provisions, the governance and organization of investment firms now were subjected to a complex, harmonized regime of institutional and procedural substantive requirements, compliance with which would be monitored by the licensing authorities. In addition to general requirements relating to the internal organization,41 Directive 2006/73/EC on organizational requirements and operating conditions, insofar as it is of interest for the purposes of this chapter, stipulated detailed provisions and definitions on compliance,42 risk management,43 internal audits,44 the corresponding responsibility of senior

management,45 and on specific organizational requirements designed so as to ensure the safeguarding of client funds.46 [3.16] Irrespective of the substantial increase in detail brought about by the Organisational Requirements Directive 2006/73/EC, however, the fundamental policy basis for governance-related requirements did not deviate conceptually from earlier regulation. Just as in MiFID I, considerations of equal market access on one hand and investor protection on the other hand can be identified as the key motives inspiring the technical specifications stipulated by the Organisational Requirements Directive. If the new regime differed at all from its predecessor in this respect, ensuring a level playing field for all providers of investment services throughout the (then) European Community—rather than systemic stability concerns—appears to have been an even stronger determinant of governance-related requirements than under the ISD 1993. This motive was expressly mentioned as the objective for technical harmonization of organizational requirements and conditions for authorization, as well as for the principle of maximum harmonization imposed by the Implementing Directive,47 whereas investor protection was identified as a key rationale for requirements for operating conditions48 (i.e. conduct-of-business rules) —and systemic stability concerns were not quoted at all in either MiFID I or the Organisational Requirements Directive 2006/73/EC. [3.17] To stop the analysis at this point would be misleading, however, as it was outside the MiFID I regime that governance-related requirements for investment firms first received a much more comprehensive treatment in EU financial services regulation. With the transposition of the Basel II capital accord of 200449 into European law by the CRD I package in 2006,50 not just credit institutions but also investment firms became subject to a reformed set of comprehensive governance-related requirements, which reflected the Basel Committee’s emphasis on the need for sound and robust governance arrangements in addition to reliable capital cushions in financial intermediaries.51 In particular, Article 22(1) of the recast Banking Directive, which was supplemented with complex technical requirements in Annex V to that Directive, required banks to establish and maintain

robust governance arrangements, which include a clear organisational structure with well defined, transparent and consistent lines of responsibility, effective processes to identify, manage, monitor and report the risks it is or might be exposed to, and adequate internal control mechanisms, including sound administrative and accounting procedures.

[3.18] The scope of these provisions was extended to investment firms pursuant to Article 34 of the recast Capital Adequacy Directive. In 2010, the new regime was supplemented by new rules on remuneration policies as part of the CRD III reform in November 2010.52 To be sure, particularly this latter reform can be characterized as a first step towards the subsequent comprehensive overhaul of the regulatory framework for banks and investment firms by the CRD IV package and MiFID II, in that the Preamble discusses at length weaknesses in remuneration policies and risk governance as a determinant for failures in the financial crisis and also reflects the Commission’s plans for extensive enhancements of governancerelated provisions developed in its 2010 Green Paper.53 [3.19] In short, the modern trend for comprehensive regulation of investment firms’ governance arrangements clearly had its roots already in the CRD I package (as recast in 2006), which implemented the Basel II Accord within the European Union. As such, the new regime clearly was motivated by systemic stability considerations and did not reflect specific investor protection concerns. By and large, the extension of the organizational requirements not just to credit institutions but also to nonbank investment firms continued to reflect the traditional rationale of ensuring equal market access and a level playing field for both types of financial intermediaries. Until the 2010 reform of the CRD III package, which reflected already the more comprehensive insights into the determinants of financial crisis gained during 2007–9, the inclusion of investment firms in the general regulatory framework for credit institutions cannot be said to have been based on an analysis of the functional parallels and differences in the respective business models, and on the resulting challenges to financial stability associated with them.

3. Some Preliminary Conclusions

[3.20] In retrospect, the policy rationale for the treatment of investment firms in European financial regulation can be said to have oscillated between three rather different aspects: investor protection, the protection of financial stability, and the creation of a ‘level playing field’, that is, the harmonization of requirements for market access for investment firms in the Internal Market. As a closer analysis of the emanation of governancerelated regulations in European law reveals, their conceptual basis has been rather weak from the early stages of EC Securities Law to the present day. The recent shift towards systemic risk considerations as the key rationale formulated in MiFID II is certainly consistent with a broader trend in postcrisis financial regulation, which started to influence the regulation of governance arrangements in financial intermediaries already under the CRD I package in the form of the 2006 reforms. Specifically with regard to the governance of non-bank investment firms, however, the relevant policy foundations continue to be open to doubts, mainly in two respects: First, as discussed above, their historical emanation since 1993 has by no means been organic. In contrast to the wording of the relevant Recitals in the Preamble to MiFID II, the key drivers for the EU-wide harmonization of relevant standards until MiFID II have been considerations of equal market access within the Common Market, and also, to a somewhat lesser extent, investor protection. Systemic stability concerns, by contrast, were cited only superficially in the earlier documents, with no significant role in the design of the relevant provisions. In this respect, the new emphasis on systemic stability concerns in MiFID II, now expressly identified as key determinants for the governance-related regulation of investment firms, appears to mark a significant shift in the underlying policy foundations, whereas the relevant technical instruments, to a large extent, are mere adaptations of the earlier regime. It remains to be seen whether and to what extent the rearranged and expanded set of regulatory requirements can be reconciled with the specific challenges posed by non-bank investment firms. Second, empirical evidence as to concrete deficiencies in existing arrangements is rather limited, which further weakens the case for specific regulatory interventions in the organizational choices of regulated firms. Both aspects will have to be considered in the course of the analysis of specific technical requirements discussed in Section III below.

III. The Technical Framework under MiFID II 1. General Principles and Problems [3.21] In principle, the governance-related requirements stipulated by MiFID II—and, in addition, the new Organisational Requirements Regulation54—continue to combine general prudential requirements established in the field of banking regulation with specific additions designed to address the specific characteristics of investment services (as distinct from banking activities). This is particularly visible with regard to the Directive’s provisions on the management body, which expressly adapt the regime set out in the CRD IV of 201355 and add further requirements in the interest of investor protection56 (see infra III.2). Just as in EU banking regulation, these duties are inextricably intertwined with organizational and procedural requirements, which are directed both to the firm and to the management body and, again, address both general prudential concerns and issues specific to the provision of investment services (as distinct from core banking services) (infra III.3). With regard to regulatory requirements for shareholders and owners with qualifying holdings, the new regime builds on, and develops further, the corresponding provisions under MiFID II,57 showing significant similarities with, but no direct cross-references to, the corresponding provisions in EU banking regulation (infra III.4). Finally, investment firms are subject to certain governance reporting requirements (infra III.5). [3.22] In terms of technical content, the departure from earlier policy concepts, at least at first sight, thus seems to be far less dramatic than the changes in the underlying policy would suggest. This finding, in turn, triggers a number of interrelated questions, to be addressed in the overal assessment (infra III.6): If it is true that the reformed set of governancerelated requirements under the MiFID II regime (unlike the corresponding requirements under the earlier instruments) focuses on the preservation of systemic stability rather than the protection of individual investors, to what extent is that change in policy reflected in the substantive content of the

relevant provisions, and what are the practical implications in terms of construction, implementation, and enforcement of the new regime? [3.23] To be sure, the policy objectives formulated in MiFID II58 (in conjunction with those set forth in the CRD IV) still do not quite reflect a clear-cut understanding of the relationship between systemic concerns and investor protection, or at least fail to present a consistent case for regulation in this regard. However, they certainly do highlight the functional link between prudential regulation, that is, institutions-oriented regulation aiming at enhancing the resilience and soundness of intermediaries, and conduct-of-business regulation, that is, transactions-oriented standards for the provision of financial services. ‘Detriment’ to investors, in the words of the Directive, can result both from the failure of individual firms and ‘incorrect conduct’; and both types of risks, in the view of the authors of MiFID II, can be traced back to ‘weaknesses in corporate governance’.59 While these considerations are consistent with the determinants of the reformed governance-related requirements under the CRD IV,60 they deserve some attention especially in the present context, as it is for the first time in the history of EU securities regulation that the functional link between the two regimes has been spelled out with such clarity. From this perspective, governance-related regulation of investment firms almost by definition serves a dual objective—the protection both of the interests of systemic stability and of individual investors. This clearly goes back to the considerations formulated above—and it suggests that, indeed, the crisis has helped to refine the understanding of relevant risks as well as to recalibrate the regulatory responses to such weaknesses. In this respect, securities regulation appears finally to have caught up with the area of banking regulation, where the dual objective of (prudential) regulation has always been defined as encompassing both the preservation of financial stability and the protection of depositors. In sum, based on the construction of the formulation of relevant policy objectives quoted above, the concept of governance-related regulations under MiFID II, in comparison with its predecessors in the ISD 1993 and MiFID I, can be said to have converged with the rationale of the corresponding requirements in the area of banking regulation: In European (and international) banking regulation, requirements for the internal governance of regulated institutions have been an integral part of the prudential toolbox since the implementation of the

Basel II Capital Accord—and, as such, have been aiming at the protection of both systemic stability and depositors.61 This, again, evokes some key questions for further exploration in the present chapter: In view of the functional differences between the economic nature of banking activities and those associated with the provision of investment services, should that conceptual convergence be welcomed? Similarly, and even more farreaching: To what extent is the underlying rationale corroborated by the available evidence on governance failures? As discussed before,62 the empirical case for enhanced governance-oriented regulation after the global financial crisis is, at best, mixed, making a refined understanding of the specific governance standards for non-bank investment services even more desirable. [3.24] Finally, in a more technical respect, the following analysis will have to address the question of whether the new regime, despite the increase in technical detail in comparison with its predecessors, continues to be sufficiently flexible with regard to the variety of regulated investment firms across the EU. Just as is the case in the area of the reformed regime for the prudential regulation of banks under CRD IV,63 this is not just a question of proportionality in view of differences in terms of size and complexity of the regulatees, but also of flexibility with regard to differences in terms of their legal nature and respective governance structures. Under the MiFID II regime, these concerns are particularly pressing because implementing Level 2 instruments no longer take the form of Directives but now come in the form of regulations.

2. The Board A. Organizational Structure [3.25] Just like banks, investment firms under the new regime are subject to a rather complex set of fundamental requirements as to the organizational structure of the board. In this respect, MiFID II does not establish independent criteria, but merely incorporates the corresponding

requirements prescribed by CRD IV (Article 9(1) in conjunction with Article 88 CRD IV). In this regard, it should be noted that the reference to the CRD IV in Article 9(1) MiFID II is of a declaratory nature for the majority of investment firms within the EU, which are covered by the definition of ‘investment firms’ set out in Article 4(2) of the Capital Requirements Regulation (the ‘CRR’)64 and therefore qualify as ‘institutions’ as defined in Article 4(1)(3) of the CRR which in turn includes them directly in the scope of the CRD IV requirements anyhow.65 Consequently, Article 9(1) MiFID II merely extends the CRD IV regime to (non-CRR) investment firms not covered by the CRD IV package.66 [3.26] Within this framework, Article 88 CRD IV lays down specific requirements both for the division of responsibilities and for the organization of the board. In principle, although both the CRD IV and MiFID II seek to establish a neutral system for adaption across different systems of corporate governance,67 this framework is tailored primarily to one-tier board models, where board management and supervisory functions are exercised from within a single board of directors (as distinct from— German-style—two-tier board systems, which separate management and supervisory boards). Both Directives leave it to the Member States to make the necessary adjustments for two-tier boards.68 Nonetheless, the prescription whereby the chairman of the management body in its supervisory function must not exercise simultaneously the functions of a chief executive officer within the same institution, unless justified by the institution and authorised by competent authorities,69 clearly addresses two-tier board systems70 and imposes a separation of functions which, while commonly practised at least within the United Kingdom for some time, has never been established as preferable over a combination of the two functions in one person.71

[3.27] Within the board, firms must establish a risk committee, a nomination committee, and a remuneration committee, if they are ‘significant in terms of their size, internal organization and complexity of their activities’. This, again, follows directly from provisions in the CRD IV,72 of which only Article 88 is expressly referred to in Article 9(1) of the MiFID II.

B. Duties i. The Board as a Whole [3.28] The duties of the management body in general and its committees—in a rather complex and partly duplicative way —are set forth in both the CRD IV and MiFID II, as well as in the provisions of Chapter II of the Commission Delegated Regulation of 25 April 2016.73In line with the concept of functional neutrality with regard to different legal structures in the national laws on corporations and partnerships,74 neither the CRD IV nor MiFID II lays down a positive definition of the board’s duties, which are determined exclusively by autonomous legislation in the Member States. This is clearly reflected in the identical definition of the term ‘management board’ in both Directives, which expressly refers to those strategic, oversight, and management duties that are determined by the applicable national laws.75 While accepting the general duties of management (and supervisory) boards as prescribed by national legislation in principle, both the CRD IV and the MiFID II lay down additional specifications with regard to, in particular, organizational duties, however. For regulated entities, the organizational duties of management (and supervisory) boards form a two-tier system, with the first level consisting of general principles under the applicable national law of business associations (companies or partnerships, as the case may be), which are then complemented and, in part, superseded by specific organizational duties prescribed by European financial regulation. It is worth noting that the requirements addressed directly to the management body in both CRD IV and MiFID II are, to a large extent, of a procedural rather than a substantive nature; they specify the responsibility of the board for the definition of strategies and procedures and for the implementation of relevant standards, rather than defining the relevant standards themselves (which are specified in other provisions and in additional Level 2 legislation76). [3.29] Generally, pursuant to the identical formulation in both Article 88(1) CRD IV (referred to in Article 9(1) MiFID II) and Article 9(3) MiFID II, the management body77 is required to define[ ], oversee[ ] and [be] accountable for the implementation of the governance arrangements that ensure effective and prudent management of the [firm], including

the segregation of duties in the [firm] and the prevention of conflicts of interest, to which Article 9(3) MiFID II adds the further specification that the management body must exercise these functions ‘in a manner that promotes the integrity of the market and the interest of clients’.

[3.30] Article 88(1)(2) CRD IV then specifies these functions further and requires the board, in particular, to (a) approve and oversee the firm’s ‘strategic objectives, risk strategy and internal governance’, (b) ensure ‘the integrity of the accounting and financial reporting systems’, (c) oversee ‘the process of disclosure and communications’, and (d) effectively oversee senior management. [3.31] These requirements are then complemented by additional, albeit to some extent duplicative, specifications in Article 9(3) MiFID II, pursuant to which the management board shall define, approve and oversee

(a) the organisation of the firm for the provision of investment services and activities and ancillary services, including the skills, knowledge and expertise required by personnel, the resources, the procedures and the arrangements for the provision of services and activities, taking into account the nature, scale and complexity of its business and all the requirements the firm has to comply with; (b) a policy as to services, activities, products and operations offered or provided, in accordance with the risk tolerance of the firm and the characteristics and needs of the clients of the firm to whom they will be offered or provided, including carrying out appropriate stress testing, where appropriate; (c) a remuneration policy of persons involved in the provision of services to clients aiming to encourage responsible business conduct, fair treatment of clients as well as avoiding conflict of interest in the relationships with clients. [3.32] Moreover, the management board is required

to monitor and periodically assess the adequacy and the implementation of the firm’s strategic objectives in the provision of investment services and activities and ancillary services, the effectiveness of the investment firm’s governance arrangements and the adequacy of the policies relating to the provision of services to clients and take appropriate steps to address any deficiencies.78

[3.33] Both Article 88 of the CRD IV and, to an even larger extent, Article 9 MiFID II thus formulate a rather comprehensive catalogue of standards for the board’s responsibilities with regard to organizational and compliance aspects. In principle, the set of functions thus defined does not significantly deviate from the allocation of powers and responsibilities under general company or partnership law (whereby the relevant functions will usually be allocated to the board, even if, in many cases, the applicable law will be much less detailed than its regulatory counterpart79). However, the far more explicit definitions of responsibilities in regulatory law, a breach of which may ultimately trigger the revocation of the licence,80 evidently not only seek to enhance legal certainty, but also to facilitate effective supervisory control of the different aspects. This is reflected in the specific sanction provided for breaches of the requirements set forth in Article 91 CRD IV, which may trigger administrative penalties of up to €5 million or double the loss that has been incurred as a consequence of the breach, with the sanction, as a rule, to be made public.81 When assessing compliance, authorities will apply a specific standard of care that has been defined in Article 91(8) CRD IV, within the context of requirements on the suitability of board members.82 [3.34] All in all, the reformed set of specific board duties primarily reflects prudential concerns about the soundness of risk profiles and operational arrangements of institutions (i.e., credit institutions and investment firms) and, as such, systemic stability considerations. This is illustrated, first and foremost, by the fact that the most important part of the relevant duties is set forth in the CRD IV rather than MiFID II—without any differentiation whatsoever between the business models characteristic for credit institutions and those of non-bank investment firms. Even the additional requirement to a similar effect imposed by Article 9(3) MiFID II evidently focuses on the definition of standards aiming at the soundness of operational arrangements and risk sensitivity rather than on such aspects

that may directly benefit the position of investors, with the limited exception of Article 9(3)(c) MiFID II (quoted above, paragraph 3.31). As indicated above, this is not inconsistent with lessons learnt from the global financial crisis, which clearly reinforced the notion that systemic risk is not confined to commercial banking but attributable to other factors (size, interconnectedness, complexity of business organizations).83 Against this backdrop, it may be considered helpful that the new regulatory framework, by specifying the board’s responsibility for organizational and strategic decisions and risk management in most comprehensive terms, clarifies the delineation of powers and duties in regulated firms, which could potentially lead to increased awareness among board members as to the specific challenges and needs associated with their respective supervisory and management functions. ii. Board Committees [3.35] Just as with credit institutions, investment firms, pursuant to Article 88(2) CRD IV (in conjunction with Article 9(1) MiFID II, where applicable), are subject to detailed requirements with regard to the formation and tasks of the nomination committee. The committee is not only responsible for the selection of candidates for management body positions and the evaluation of the knowledge, skills, and experience, as well as periodical assessments and reviews of the policy for the selection and appointment of senior management, but also for the promotion of gender equality.84 Remuneration committees are responsible for ‘the preparation of decisions regarding remuneration, including those which have implications for the risk profile and risk management of the institution concerned’.85 Risk committees are to advise the management body with regard to the ‘risk appetite and strategy’, and to assist with the implementation of the strategy. The risk assessment also has to review the impact of prices of liabilities and assets on the firm’s business model and risk strategy, and to propose remedies if deficiencies are detected.86 By contrast, no specific requirements are set forth in either the CRD IV or MiFID II with regard to the duties of audit committees, which credit institutions and investment firms are required to establish under Article 41 of Directive 2006/43/EC.87

C. Personal Requirements [3.36] Reflecting widespread concerns about the quality of management and supervision in the boards of financial institutions before and during the global financial crisis,88 enhanced requirements for the eligibility for board membership and for board diversity have become a central aspect of the reformed regulatory framework for both credit institutions and securities. Just as in the area of board duties, discussed above, the relevant requirements are laid down partly in the CRD IV (for credit institutions and investment firms alike), with—again, to some extent duplicative—additions stipulated by MiFID II. [3.37] Generally, all board members, executive and non-executive, are subjected to a rigorous fit-and-proper test. Pursuant to Article 91(1) and, in identical wording, to Article 9(4), they are required at all times ‘[to] be of sufficiently good repute and possess sufficient knowledge, skills and experience to perform their duties’ and to commit sufficient time to perform their functions.89 [3.38] Moreover, pursuant to a requirement that links personal qualifications with a standard of care for the execution of the board duties, each member of the management body is required to act with honesty, integrity and independence of mind to effectively assess and challenge the decisions of the senior management where necessary and to effectively oversee and monitor decision-making.90

[3.39] In addition to the individual requirements, the board as a whole is required to have ‘adequate collective knowledge, skills and experience to be able to understand the [firm’s] activities, including the main risks’.91 In order to maintain the standard, regulated firms are required to ‘devote adequate human and financial resources to the induction and training of members of the management body’.92 Particular attention has been given to ensuring board diversity, which is reflected not only in the requirement to engage a broad set of qualities and competences when recruiting members to the management body and for that purpose to put in place a policy promoting diversity on the management body,93

but also in the definition of tasks of the nomination committee, which must not simply strive for ‘diversity’ in its proposals for vacant management body positions, but is also required to decide on a target for the representation of the underrepresented gender in the management body and prepare a policy on how to increase the number of the underrepresented gender in the management body.94

[3.40] Pursuant to Article 91(12) CRD IV, the relevant criteria are to be complemented by EBA guidelines in due course.95 [3.41] With these requirements, the new regime clearly goes far beyond the design and content of traditional ‘fit-and-proper’ tests in the area of financial regulation, which usually set a rather broadly defined set of benchmarks against which supervisory authorities would have assessed the qualifications and character of persons nominated for board positions.96 To be sure, these earlier requirements already facilitated a significant degree of supervisory control over board decisions, if and to the extent that the relevant authority, under the applicable national law transposing the European requirements, was willing to use board members’ readiness to comply even with informal supervisory guidance as indicative for the purposes of the ‘fit and proper’ test.97 [3.42] Under the reformed regime, by contrast, the traditional, flexible approach, which left further specification to the discretion of national legislators and authorities, has been complemented with a considerable range of new procedural and substantive elements. Among these, requirements that the personal qualification of board members be commensurate with the cognitive and strategic problems to be expected in the performance of their respective functions can surely be said to be rather uncontroversial, although it remains to be seen whether the more prescriptive approach really yields a more effective enforcement of qualitative standards by relevant authorities. In particular, it remains open to doubt whether the (still rather abstract) standard of ‘honesty, integrity and independence of mind’ can be operationalized in a way that substantively adds to the effectiveness of authorities’ scrutiny of board composition.98

[3.43] More controversial, however, has been the policy decision to promote gender diversity as a means to foster good corporate governance in financial institutions. The underlying rationale is to prevent ‘groupthink’, which, according to a widely held view of the global financial crisis, can supposedly be traced back not least to a greater risk appetite of male management than could be expected to be found among a gender-diverse board.99 As empirical analysis of what has become known as the ‘Lehman Sisters Hypothesis’ (insinuating that, with a female board of directors, excessive risk-taking and the ensuing crisis would not have happened) appears to be inconclusive,100 the effectiveness of mandatory board diversity as a precaution against excessive risk-taking practices can, at best, be described as doubtful. It is to be expected that regulated firms, in redefining their diversity policies, will attend to the relevant supervisory authority’s expectations rather than to what they themselves would perceive to be an optimal balance.101 [3.44] At any rate, just as has been diagnosed for the highly prescriptive set of board duties above,102 the focus of the reformed personal requirements applicable to board members under the new CRD IV/MiFID II regime clearly reflects systemic stability rather than investor protection considerations. Also in this context, credit institutions and non-bank investment firms are indiscriminately subjected to a regime which seeks to improve the quality of corporate governance in the interest of the individual firm’s soundness of operations and, ultimately, the prevention of contagious effects that would arise from their financial failure. To be sure, clients that have an ongoing relationship with the firm (e.g. in the context of assetmanagement contracts) could benefit as well, but rather indirectly.

D. Remuneration of Board Members and Senior Management [3.45] Among the different policy initiatives to foster effective corporate governance in financial institutions in recent years, prescriptive requirements for remuneration (or ‘compensation’) arrangements have clearly played a major, albeit somewhat independent, role. This is

particularly visible in the EU Commission’s 2010 Green Paper, which discusses remuneration policies as a distinct part of the regulatory agenda on issues related to the corporate governance of financial institutions.103 At the international level, sound remuneration standards, designed to remove incentives to engage in excessive risk-taking, have been promoted as a priority within the post-crisis reform agenda of the Financial Stability Board,104 whose recommendations have been influential not just for the design of relevant provisions at the European level, but also—prior to EUwide harmonization—within individual Member States.105 Under the European framework harmonized by the CRD IV, investment firms are subject to the regime applicable also to credit institutions, to which MiFID II adds the requirement that the management board has to develop a specific ‘remuneration policy’ for persons involved in the provision of services to clients,106 while more detailed requirements for the development of remuneration policies and practices, including on the protection of client interests in this context, are prescribed in Article 27 of the Delegated Council Regulation of 25 April 2016.107 Leaving this aside, investment firms have to comply with the general qualitative principles governing the remuneration policies for senior management, so-called risk-takers, staff in control functions, and any employee whose remuneration is equivalent to that of senior management and risk-takers and persons whose decisions have a material impact on the firm’s risk profile stipulated in Article 92(1) CRD IV. Specifically, Article 94 delineates the extent to which variable elements may be included. As mentioned before, remuneration committees must be established within ‘significant’ firms108 as an institutional means to ensure sound remuneration policies. [3.46] The substance of these requirements, as well as the underlying rationale, has been discussed extensively elsewhere.109 This and the fact that investment firms, to the extent that they are included in the scope of application of the CRD IV,110 are subjected to the same regime as credit institutions hardly necessitates in-depth treatment of the remuneration regime within the present chapter. It should be noted, however, that the extension of the relevant requirements to investment firms does not reflect a careful analysis of the similarities and the differences existing between remuneration practices in credit institutions and investment firms,

respectively—and of their implications for risk-taking and, ultimately, systemic risk. In this regard, the same applies as to other parts of the regulatory framework for investment firms, which also extend the regulatory regime for banks without a sound empirical basis. While it may be possible to address such differences under the proportionality test prescribed by Article 92(1) CRD IV, the policy foundations for the equal treatment still appear to be rather weak.

3. Organization and Risk Management A. General Requirements Laid Down in the CRD IV Regime [3.47] To be sure, the governance-related provisions for investment firms, as laid down in MiFID II, follow a more autonomous approach with regard to organizational duties and risk management than with regard to the regulation of the board and its duties (infra III.3.B). Nonetheless, the CRD IV regime also lays down a general basis applicable without discrimination to banks and non-bank investment firms in this respect. The substantive requirements on risk management in Articles 74–87 CRD IV clearly form the core of the Directive’s provisions on internal governance arrangements and clearly aim at enhancing the soundness of its business as a whole.111 This becomes obvious from the very first provision on governance arrangements, which builds on the wording of Article 22 of the recast Banking Directive of 2006 and requires that [i]nstitutions shall have robust governance arrangements, which include a clear organisational structure with well-defined, transparent and consistent lines of responsibility, effective processes to identify, manage, monitor and report the risks they are or might be exposed to, adequate internal control mechanisms, including sound administration and accounting procedures, and remuneration policies and practices that are consistent with and promote sound and effective risk management.112

[3.48] While the relevant arrangements, processes, and mechanisms are to be specified further by EBA Guidelines,113 the following provisions of the CRD IV themselves continue to lay down a rather tight framework of

procedural and substantive requirements for the different aspects of risk governance. As mentioned before, this includes the need to establish a risk committee with both advisory and assistant functions in the formulation and implementation of risk strategies.114 In addition, ‘significant’ credit institutions and investment firms covered by the CRD IV are required to establish a ‘risk management function’ with independent resources and access to the management body.115 But these requirements are ‘without prejudice to the application of Directive 2006/73/EC to investment firms’, which in Article 7(2) already prescribes the creation of an independent risk management function and thus prevails over the CRD IV. The management body itself is expressly required to approve[ ] and periodically review[ ] the strategies and policies for taking up, managing, monitoring and mitigating the risks the institution is or might be exposed to, including those posed by the macroeconomic environment in which it operates in relation to the status of the business cycle,116

and has to ‘devote[ ] sufficient time to consideration of risk issues’.117 These core provisions are then complemented with detailed technical requirements in Articles 77–84. [3.49] In general, the approach to risk management taken by the CRD IV can best be described as a multi-tier concept, which seeks to incentivize the firms (or, more specifically, management boards) to actively define, control, and improve the culture of their respective firms, the substantive standard for which is defined in rather abstract terms (‘robust’ arrangements; ‘clear’ structure; ‘well-defined, transparent, and consistent’ lines of responsibility, etc.). While this is clearly reflective of a principles-based approach to governance regulation, the Directive is much more specific with regard to organizational features designed to bolster the firms’ own efforts. In particular, the specialist risk-management function that has to be established under the new regime has rightly been explained by the desire to create an institutionalized ‘antidote to overcome certain risk management flaws observed in the banks and financial institutions that suffered in the global financial crisis’,118 and other features designed to increase the board’s sensitivity vis-à-vis the risk culture fit into the picture too.119 In particular, the new approach evidently seeks to allocate specific responsibilities and to

formulate qualitative standards for the risk-management organization and the resources (financial and human) dedicated to it, in order to activate the boards that are perceived to have been too passive and complacent before the crisis.120 Whether or not this concept is enough to overcome residual deficiencies remains to be seen, however. It could well be the case that firms’ future behaviour may be driven more by the need to demonstrate compliance with the regulator’s expectations rather than a genuine effort to improve the quality of existing arrangements. In this context, it is worth noting that proceduralist, principles-based approaches of corporate and financial behaviour have repeatedly been criticized on account of their structural tendency to foster ‘compliant non-compliance’121—to provoke formalistic efforts to ‘tick the boxes’ in generic to-do lists rather than encourage the regulatees’ creativity with a view to accomplishing the regulatory objectives through instruments that match the needs of the individual case.122 At any rate, specific models for the design of riskmanagement arrangements, which could be referred to as ‘best practice’ by both authorities and regulated institutions, have evolved rather slowly, and widely advocated standards are still far from universally accepted.123

B. Specific Organizational Requirements under MiFID II [3.50] Although the CRD IV regime on risk management is highly prescriptive and rather detailed, the corresponding requirements stipulated by MiFID II go still further. The relevant provisions are to be found in Articles 16 and 17 MiFID II, of which the former lays down general requirements applicable for all investment firms, while the latter (not to be examined in detail hereafter) specifically addresses risk-management issues in firms engaging in algorithmic trading. As noted before, this regime is supplemented by implementing Level 2 legislation promulgated by the Commission pursuant to Article 16(12) MiFID II, which has adopted a rather abstract approach and does not provide very detailed specification, however.124 Thus, the regulatory framework for organizational duties for investment firms—which takes up the coexistence of general requirements applicable to both credit institutions and non-bank investment firms under

the predecessors of the present regime125—will continue to reflect a complex interplay of different sources of law. Conceptually, the new regime will become even more complex to implement than its predecessors because of the fact that, unlike under MiFID I, implementing legislation now takes the form of a Council Regulation rather than a Directive. Investment firms regulated under both CRD IV and MiFID II will thus simultaneously be subject to risk-management requirements under the national laws transposing the CRD IV and MiFID II (and Directive 2006/73/EC) and the Council Regulation Implementing MiFID II. This requires, at the very least, a careful design of national regimes with a view to minimizing the potential for interferences, but will still prove problematic in cases where the provisions of CRD IV and MiFID II overlap. [3.51] In substantive terms, part of the relevant provisions reflect, again, prudential concerns aiming at enhancing the stability and soundness of the individual firms, while others focus on improving the legal and commercial position of investors and other clients. With respect to the former, MiFID II requires investment firms to take reasonable steps to ensure continuity and regularity in the performance of investment services and activities. To that end the investment firm shall employ appropriate and proportionate systems, resources and procedures.126

[3.52] Moreover, in what is in part a (redundant) repetition of the general CRD IV requirements examined above, investment firms must have sound administrative and accounting procedures, internal control mechanisms, effective procedures for risk assessment, and effective control and safeguard arrangements for information processing systems

as well as […] sound security mechanisms in place to guarantee the security and authentication of the means of transfer of information, minimise the risk of data corruption and unauthorised access and to prevent information leakage maintaining the confidentiality of the data at all times.127

[3.53] To be sure, the latter requirement aims not exclusively at the soundness of the relevant firm as such, but also at the protection of the commercial relationships with investors and other clients. The same applies to the duty to establish a compliance organization within the firm, which must include adequate policies and procedures sufficient to ensure compliance of the firm including its managers, employees and tied agents with its obligations under this Directive as well as appropriate rules governing personal transactions by such persons.128

In this context, Article 22 of the Commission Delegated Regulation of 25 April 1016129 provides a considerable degree of further specification, including on the responsibilities and duties of the compliance function (Article 22(2)) and the requirements that must be satisfied in terms of organizational design (Article 22(3)). Generally, firms must ensure that the management board has access to all relevant information.130 [3.54] Exclusively in the interest of investors, the Directive prescribes a detailed set of procedural requirements which build on Article 21 MiFID I but are considerably more detailed and prescriptive. These include institutional arrangements designed to prevent conflicts of interest,131 product approval processes for the ‘manufacturing’ (!) of financial products for sale to clients,132 and obligations to review the financial instruments offered or marketed by the firm.133 They supplement, but are without prejudice to, the general requirements of MiFIR relating to disclosure, suitability or appropriateness, identification, and management of conflicts of interests and inducements.134 Additional requirements are made with regard to record keeping and documentation135 and to the preservation of ownership and other rights of clients.136 [3.55] While the underlying policy rationale of the organizational provisions stipulated by MiFID II thus differs in part from those set out in the CRD IV, with investor protection coming into play as a second determinant in addition to systemic stability concerns, the regulatory strategy employed is essentially the same in both instruments.137 Just like in the CRD IV, we find a combination of substantive standards in the form of a multitude of rather vague criteria (‘reasonable’ steps, ‘sound’ mechanisms

and processes, ‘effective’ procedures, ‘adequate’ policies, etc.), which leave a considerable discretion to supervisory authorities, with procedural requirements that expressly allocate responsibilities and require the implementation of specific organizational arrangements, such as the creation of a compliance function. Just like the CRD IV in this respect, the MiFID II organizational framework can thus be said to reflect the expectation that the management board’s sense of responsibility can be activated in a way that ultimately may yield improvements in the corporate culture—the attitude of directors and senior and junior management with regard to risk-taking and qualitative standards in the provision of investment services to clients. Just as with the CRD IV regime, however, doubts as to the efficacy of proceduralist approaches may also be warranted in this context.

4. Shareholders and Owners with Qualifying Holdings [3.56] With regard to the supervisory scrutiny of shareholders and owners of qualifying holdings, the new regulatory framework takes up the previous regime established by Articles 9 and 10 MiFID I. The new framework adopts a far more complex approach, however, underlining the relevance of shareholder control as a means to promote sound business practices and good corporate governance. Significantly, the new provisions are broadly in line with the corresponding requirements under Articles 14 and 22–27 of the CRD IV (on the disclosure and supervisory assessment of shareholders and owners prior to authorization and in connection with a proposed acquisition at a later stage, respectively). Both regimes cover ‘the shareholders or members, whether direct or indirect, natural or legal persons, that have qualifying holdings’,138 defined as a direct or indirect holding in an investment firm which represents 10% or more of the capital or of the voting rights, as set out in Articles 9 and 10 of Directive 2004/109/EC […], taking into account the conditions regarding aggregation thereof laid down in Article 12(4) and (5) of that Directive, or which makes it possible to exercise a significant influence over the management of the investment firm in which that holding subsists.139

[3.57] As a condition for authorization, the competent authorities must be ‘satisfied as to the suitability’ of the holders of such positions, ‘taking into account the need to ensure the sound and prudent management of an investment firm’.140 With regard to the acquisition of positions in an already authorized firm, these requirements are supplemented with specific qualitative criteria including, inter alia, the reputation of the proposed acquirer, the reputation and experience of persons who will direct the business as a result of the acquisition, the financial soundness of the acquirer, and the likely impact of the acquisition on future compliance with regulatory requirements.141 Articles 12 and 13 MiFID II lay down the procedural framework for the assessment process in this context, which is clearly modelled on the corresponding provisions of Articles 22–24 CRD IV.

5. Governance Reporting [3.58] Again, just as with credit institutions, investment firms covered by the CRD IV/CRR framework are subject to the governance reporting requirements stipulated by Articles 435 and 450 CRR, which provide for the disclosure on the risk-management function and certain aspects of board organization, as well as remuneration policies, respectively.142 While details remain outside the scope of this chapter, it should be noted that the general approach to using external disclosure as a means to foster compliance with prudential standards is thereby also extended to the governance-related regulation of investment firms.

6. Assessment A. Systemic Stability Concerns versus Investor Protection: A Reappraisal of Policy Rationales [3.59] As has been discussed above, the policy rationale for governancerelated provisions in the CRD IV/MiFID II regime remains somewhat obscure, in particular if measured by the formulations of the respective Preambles. For the first time, specific systemic stability considerations

explicitly have made their way into the policy foundations of the organizational framework set out in MiFID II, reflecting a departure from the earlier approach which was informed mainly by the motive to realign the regulatory frameworks for non-bank investment firms with those applicable to universal banks, and, in addition, to introduce some specific requirements addressing the principal–agent conflicts characteristic for the provision of investment services.143 Although the new regime still falls short of providing a fully consistent and convincing regulatory answer to the functional parallels and differences of universal banks and non-bank investment firms regarding the two fundamental objectives of securities regulation—market stability and investor confidence—it should be noted that the functional relationship between systemic stability considerations on one hand and investor protection on the other has become more balanced, and the policy mix pursued with regard to the regulation of investment firms is more convincing than under the predecessors. Both Directives jointly seek to address a list of shortcomings in the governance arrangements of financial intermediaries that have come to be perceived as drivers for the excessive risk-taking and insufficient control of risks that ultimately triggered the global financial crisis. Rightly, unlike under the ISD 1993 and MiFID I, the inclusion of investment firms into the governance-related regulatory framework for credit institutions is no longer justified merely on the grounds of equal market access (a concept that had been rather unconvincing from the start), but reflects genuine considerations of the risks pertaining to the legal and economic nature of the commercial relationships between (non-bank) investment firms and their clients and other counterparties. In many ways, the shift in policy rationales is also reflected in the substantive and procedural technical framework established by both Directives, as both the technical requirements aiming at enhancing the soundness of strategies and operations and those aiming at improving investor protection have become more refined, compared to the earlier concept. [3.60] In sum, the CRD IV/MiFID II framework, at a very abstract level, should therefore be regarded as an improvement of the status quo ante. It should be noted, however, that this positive assessment is confined to the policy foundations as such; that is, to the concept of addressing systemic risk wherever it occurs and, therefore, irrespective of the commercial nature

of the relevant entity (‘credit institution’ or ‘investment firm’). To be sure, as noted before, this approach reflects a broader, international approach to dealing with systemic relevance of financial institutions generally, which effectively removes the traditional sector-specific boundaries in financial regulation in response to lessons learnt during the global financial crisis.144 Whether the rather indiscriminate treatment of both types of institutions within the scope of the CRD IV is actually justified is a different story, however. In this regard, it should be noted that the empirical basis for the assessment of systemic risk associated with the specific activities undertaken by investment firms continues to be extremely weak. On closer inspection, the decision to include a wide range of diverse firms, which engage in an equally diverse range of activities, within the scope of a legal framework whose main focus still remains on the activities of commercial banks may well turn out to be too simplistic to actually satisfy the test of proportionality. In this context, it should be noted that the categorization of investment firms laid down in MiFID II and the CRD IV is far from uncontroversial even within the European regulatory landscape. It is particularly telling that the European Banking Authority, in a recent report, strongly promotes the introduction of a new approach, which would be more sensitive to the implications of the firms’ activities not just in terms of their individual soundness, but also in terms of systemic stability.145 [3.61] Irrespective of these general considerations, it is far from clear whether the technical design draws the right conclusions; that is, whether the new framework addresses real deficiencies with adequate, effective remedies. While the need for enhanced systemic stability and investor protection can hardly be disputed as such, the first doubts arise just below this rather high-level set of objectives. As noted before, it is not just the empirical evidence of weaknesses in governance arrangements prior to the global crisis that is rather thin, but also the available body of ‘best practice’ that could be referred to by both supervisors and regulated firms in the course of implementing the new regime. In this respect, while it is true that the CRD IV/MiFID II reflects a sound definition of fundamental policy objectives, the question of whether the same could also be said of the corresponding mix of policies and strategies remains open.

B. Principles-Based Regulation versus Ever Tighter Rules: Some Observations on Regulatory Strategy [3.62] As noted previously in respect of organizational requirements,146 the preference for a combination of a large amount of vaguely defined principles (compliance with which will be monitored by authorities enjoying a rather high level of discretion) with highly prescriptive procedural requirements has become characteristic for both the CRD IV and the MiFID II. This could well turn out to be a major problem for the effective implementation in future practice. If anything, this approach is likely to reduce the flexibility associated with a genuine principles-based strategy, where the supervisor is free to act on the basis of a few, broad principles rather than the present array of detailed qualitative standards. Effective implementation will succeed only if the regulated firms and their supervisors manage to interpret, and apply, the new framework in ways flexible enough to cater for the needs arising out of individual business models and the legal and organizational structures of firms. This in turn will happen only if supervisors abstain from one-size-fits-all approaches that would streamline existing arrangements without any positive yield in terms of either systemic resilience or improved investor protection.147 In this context, it should not go unnoticed that the complex set of vague definitions is likely to give rise to substantial legal uncertainty on the part of regulated firms. This in turn could incentivize board members to strive for uncontroversial, if formalistic, compliance with the supervisor’s expectations rather than creative, albeit perhaps controversial, independent solutions which might be preferable not just for the firm itself but also for systemic stability.148

C. Technical Inconsistencies with Company and Partnership Law [3.63] These concerns are aggravated by the technical inconsistencies between the governance framework laid down by both the CRD IV and MiFID II on one hand and the real-life variety of different legal and organizational structures in investment firms on the other hand, which in turn reflect differences in the underlying national company and partnership

laws. Such inconsistencies are resolvable, although perhaps not in the most efficient way; the Directives clearly restrict the scope for organizational choices of firms. For example, as investment firms must introduce a separation of the functions of chairman of the board and chief executive officer,149 it is clear that a combination of functions is not permissible from the regulatory perspective, even where general company or partnership law would allow it and although the rationale may well be debatable.150 Technical inconsistencies between the regulatory framework and the applicable national regimes for the formation of business enterprises are more difficult to resolve in cases where the Directives’ approach is too inflexible to take account of the mandatory restrictions of national law. As pointed out by Peter O. Mülbert and Alexander Wilhelm in a careful analysis of the applicable CRD IV provisions, this is particularly the case with regard to the Directive’s organizational duties for the management board, which cannot easily be adapted to the legal environment of two-tier boards or corporate forms without any mandatory supervisory functions.151

D. Impediments to Effective Enforcement: The Public– Private Dichotomy [3.64] From a company law perspective, the new regulatory framework for the governance of investment firms can be characterized as a complex supplement to the existing company (or partnership) law provisions relating to the formation of business enterprises, board duties, and, in particular, organizational requirements that have hitherto been defined exclusively by the applicable laws of the country of registration. As discussed above, these provisions do not necessarily conflict with the regulated firms’ best interests. Viable risk-management systems, for example, are surely desirable not just for financial intermediaries and, indeed, will be required by some national company laws.152 In certain cases, regulatory requirements may nonetheless conflict with the commercial interests of a regulated firm.153 At any rate, the management board may, rightly or wrongly, find a particular organizational measure required by the supervisory authorities to be undesirable commercially, for example because the costs of implementation are high while the possible benefits

appear to be rather abstract. In any such cases, the management board will find themselves in a conflict of interest between the need to comply with supervisory requirements on one hand and the duty to best serve the company’s commercial interests on the other. If that is the case, the effective enforcement of regulatory requirements may be compromised by the interference with private-law duties of directors under the applicable company law. At the very least, private enforcement of regulatory standards is unlikely to happen in such circumstances, which could well weaken the de facto influence of regulatory standards.154

IV. Conclusions [3.65] Compared with credit institutions, investment firms under the reformed framework laid down in the CRD IV and MiFID II have been subjected to an even more complex set of corporate governance requirements, which extend not just to the composition and organization of the board of directors and risk management, but also to remuneration issues and the supervisory scrutiny of qualified holdings. The underlying rationale to address systemic risk wherever it occurs—irrespective of the character of an intermediary’s business—is certainly sound as such, and MiFID II clearly constitutes a step forward in terms of consistency of the underlying policy objectives. However, as argued in the present chapter, the equal treatment of banks and investment firms in the area of corporate governance requirements should clearly be justifiable on the grounds of systemic risk considerations and, in this regard, the empirical evidence supporting the convergence of regulatory approaches in the fields of banking and securities regulation remains weak. The notion that all investment firms covered by MiFID II, irrespective of size and business models, should be required to comply with rules designed to address governance failures in large, systemically important firms is open to doubt. At the very least, this calls for a careful handling of the proportionality tests required to be applied under both the CRD IV and MiFID II, while calls for a recategorization of the scope of application, so as to enhance sensitivity to potential systemic implications of firm failures, should be taken up in the medium and long run. Moreover, this chapter has identified a number of general concerns as

to the effectiveness of the new framework, which are not only confined to the regulation of investment firms. Whether or not the new set of requirements will succeed in promoting sound corporate governance of financial intermediaries remains to be seen. This chapter argues that there are reasons to remain sceptical.

* The author would like to thank Guido Ferrarini and Victor de Seriere, as well as other participants in the International MiFID II Working Group conference in Amsterdam, for insightful comments. The usual disclaimer applies. 1 See, for a similar approach, e.g., Klaus J. Hopt, ‘Corporate Governance of Banks and Other Financial Institutions After the Financial Crisis’, (2013) Journal of Corporate Law Studies 13, 219, 222; Klaus J. Hopt, ‘Corporate Governance of Banks after the Financial Crisis’ in Eddy Wymeersch, Klaus J. Hopt, and Guido Ferrarini (eds) Financial Regulation and Supervision—A Post-Crisis Analysis (Oxford: OUP, 2012), para. 11.01. 2 Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC, OJ L 145 p. 1 (hereafter: ‘MiFID I’). 3 Council Directive 93/22/EEC of 10 May 1993 on investment services in the securities field, OJ L 141 p. 7 (hereafter: ‘ISD 1993’). 4 See, in particular, ISD 1993, Article 3(3)(1), second indent (directors of investment firms to be ‘of sufficiently good repute and … sufficiently experienced’) and 3(3)(2) (at least two responsible directors required), Article 3(4) (organizational structure to be submitted to supervisory scrutiny before authorization), Article 4 (identity of owners and qualifying holdings to be disclosed and their ‘suitability’ to be assessed prior to authorization), Article 9 (notification procedure for the acquisition of qualifying holdings), and Article 10 (specific governance requirements, including ‘sound administrative and accounting procedures, control and safeguard arrangements for electronic data processing, and adequate internal control mechanisms’). 5 See infra II.2.B. 6 Commission Delegated Regulation (EU) no. …of 25 April 2016 on Directive 2014/65/EU of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive. As of end October 2016, the document has not yet been published in the Official and has not entered into force. The version adopted by the Commission is document no. C (2016)2398 final. 8 MiFID II, Preamble, Recital 5. 7 MiFID II, Preamble, Recital 4.

9

See generally, e.g., Rüdiger Veil, ‘Concept and Aims of Capital Markets Regulation’ in Rüdiger Veil (ed.) European Capital Markets Law (Oxford: Hart Publishing, 2013), pp. 18–19. 10 Commission, Green Paper on Corporate Governance in Financial Institutions, COM(2010) 284 final. See also Niamh Moloney, EU Securities and Financial Markets Regulation, 3rd edn (New York: OUP, 2014), 357–8. 11 See further infra II.2. 12 Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC, OJ L 176, 27.6.2013, p. 338. For an in-depth analysis of that regime, see Peter O. Mülbert and Alexander Wilhelm, ‘CRD IV Framework for Banks’ Corporate Governance’ in Danny Busch and Guido Ferrarini (eds) European Banking Regulation (Oxford: OUP, 2015), Chapter 6. 13 Cf., for useful analyses of the Lehman collapse and its systemic implications, e.g., Michael Fleming and A. Sarkar, ‘The Failure Resolution of Lehman Brothers’ (2014) Special Issue: Large and Complex Banks, FRBNY Econ. Pol’y Rev. 20, 175; Stephen J. Lubben and Sarah Pei Woo, ‘Reconceptualizing Lehman’ (2014) Texas International Law Journal 49, 297; Rainer Kulms, ‘Lehman’s Spill-over Effects’ (2012) Peking University Journal of Legal Studies 3, 3. And see, from a policy perspective, Basel Committee on Banking Regulation, ‘Report and Recommendations of the Cross-border Bank Resolution Group’ (March 2010), available at http://www.bis.org/publ/bcbs169.pdf, paras 49 and 50; Centre for Economic Policy Research, A Safer World Financial System: Improving the Resolution of Systemic Institutions, Geneva Reports on the World Economy No. 12 (London: CEPR, 2012) (available online at ), pp. 42–6. 14 E.g., Marco Becht, Patrick Bolton, and Ailsa Röell, ‘Why Bank Governance is Different’ (2011) Oxford Review of Economic Policy 3, 437, 455.; Klaus J. Hopt, ‘Corporate Governance of Banks after the Financial Crisis’ in Wymeersch, Hopt, and Ferrarini (eds) (n. 2), paras 11.16 et seq., 11.45 et seq.; see also René M. Stulz, ‘Governance, Risk Management, and Risk-Taking in Banks’ (2014), available at http://ssrn.com/abstract=2457947. 15 Cf., e.g., Becht, Bolton, and Röell (n. 15), 437 (discussing both board incompetence and deficient risk management); on the relationship between shareholder-friendly governance structures and risk appetite of intermediaries, see (with mixed results) Deniz Anginer et al., ‘Corporate Governance and Bank Insolvency Risk. International Evidence’ (2014), available at http://ssrn.com/abstract=2491490; Andrea Beltratti and René M. Stulz, ‘Why Did Some Banks Perform Better During the Credit Crisis? A Cross-Country Study of the Impact of Governance and Regulation’ (2012) Journal of Financial Economics 105, 1; David H. Erkens, Mingyi Hung, and Pedro P. Matos, ‘Corporate Governance in the 2007– 2008 Financial Crisis: Evidence from Financial Institutions Worldwide’ (2012) Journal of

Corporate Finance 18, 389; Luc Laeven and Ross Levine, ‘Bank Governance, Regulation and Risk Taking’ (2009) Journal of Financial Economics 93, 259. For comprehensive reviews of the available literature, see, e.g., James R. Barth, Chen Lin, and Clas Wihlborg (eds) Research Handbook on International Banking and Governance (Oxford: OUP, 2012); Jakob de Haan and Razvan Vlahu, Corporate Governance of Banks: A Survey (Oxford: OUP, 2013), available at . See also, discussing possible implications from a legal perspective, Andreas Kokkinis, ‘A Primer on Corporate Governance in Banks and Financial Institutions: Are Banks Special?’ in Iris Chiu (ed.) The Law on Corporate Governance of Banks (Cheltenham: Edward Elgar, 2015), paras 1.37 et seq.; Klaus J. Hopt, (2013) JCLS 13, 219, 237 ff.; Klaus J. Hopt, in Wymeersch, Hopt, and Ferrarini, (n. 2), paras 11.16–11.22; Peter O. Mülbert, ‘Corporate Governance of Banks’ (2009) European Business Organization Law Review 10, 411, 433– 4; Peter O. Mülbert and Ryan Citlau, ‘The Uncertain Role of Banks’ Corporate Governance in Systemic Risk Regulation’ in Hanne S. Birkmose, Mette Neville, and Karsten E. Sørensen (eds) The European Financial Market in Transition (Aalphen aan den Rijn: Kluwer, 2012), p. 275; Christoph Van der Elst, Corporate Governance and Banks: How Justified is the Match? (Oxford: OUP, 2015), available at . 16 Contrast, e.g., Aslɩ Demirgüç-Kunt and Enrica Detragiache, ‘Basel Core Principles and Bank Soundness’, World Bank Policy Research Working Paper No. 5129 (2009), available at (reaching a rather negative conclusion); with the more positive assessment by Richard Podpiera, ‘Does Compliance with Basel Core Principles Bring Any Measurable Benefits?’, IMF Working Paper WP/04/204 (2004), available at . 17 Starting with the 2009 London summit, G20 Finance Ministers and Governors have called for guidance on the assessment of the systemic importance of financial institutions and special regulatory measures to address relevant risks; see International Monetary Fund, Bank for International Settlements and Financial Stability Board, ‘Report to G20 Finance Ministers and Governors: Guidance to Assess the Systemic Importance of Financial Institutions, Markets and Instruments: Initial Considerations’, October 2009, available at ; and see further Financial Stability Board, ‘Reducing the moral hazard posed by systemically important financial institutions’, 20 October 2010, available at , and see Financial Stability Board, Press Release: ‘G20 Leaders endorse Financial Stability Board policy framework for addressing systemically important financial institutions’, 12 November 2010, available at For an industry perspective, cf. also Institute of International Finance, ‘Systemic Risk and Systemically Important Firms: An Integrated Approach’ (May 2010), available at , in particular pp. 17–40. 18 See further infra III.6.A.

20

ibid., Preamble, Recital 5. ISD 1993, Preamble, Recitals 2 and 41. 21 ibid., Preamble, Recital 6. 22 ibid., Article 3(3) sentences 1 and 2. 23 As to which, see infra III.2. 24 First Council Directive 77/780/EEC of 12 December 1977 on the coordination of the laws, regulations, and administrative provisions relating to the taking up and pursuit of the business of credit institutions, OJ L 322, 17.12.1977, p. 30, Article 3(2). The origins of this requirement can be traced to earlier precedents in national legislation, e.g., in Germany, see Jens-Hinrich Binder, ‘Organisationspflichten und das FinanzdienstleistungsUnternehmensrecht: Bestandsaufnahme, Probleme, Konsequenzen’ (2015) ZGR— Zeitschrift für Unternehmens- und Gesellschaftsrecht, 667, 677. 25 ISD 1993, Articles 4 and 9. 26 See Second Council Directive 89/646/EEC of 15 December 1989 on the coordination of laws, regulations, and administrative provisions relating to the taking up and pursuit of the business of credit institutions and amending Directive 77/780/EEC, OJ L 386, 30.12.1989, p. 1, Articles 5 and 11. 30 ibid., Article 10 sentence 2, first indent. 27 ISD 1993, Article 3(4). 28 ibid., Article 10, sentence 1. 29 ibid., Article 3(7)(e). 31 ibid., Article 10 sentence 2, indents 2–5, respectively. 32 Binder (n. 25), 688. The organizational requirements for banks under the Second Banking Directive of 1989 (Second Council Directive 89/646/EEC of 15 December 1989 on the coordination of laws, regulations and administrative provisions relating to the taking up and pursuit of the business of credit institutions and amending Directive 77/780/EEC, OJ L 386, 30.12.1989, p. 1) were confined to the very vaguely defined rule ‘that every credit institution have sound administrative and accounting procedures and adequate internal control mechanisms’ (Article 13(2)). Taking an even narrower perspective, Article 8(1) of the ISD 1993 required that ‘institutions’ internal control mechanisms and administrative and accounting procedures permit the verification of their compliance with [the Directive’s requirements] at all times.’ 33 See, for an excellent overview of the relevant issues, Moloney (n. 11), 320–3. And see further infra III.6.A. 34 As prescribed by Article 11 of the Directive. 35 In the words of a contemporary analysis, see Marc Dassesse, Stuart Isaacs, and Graham Penn, EC Banking Law, 2nd edn (London: Lloyd’s of London Press, 1994), para. 7.2. 36 MiFID I, Preamble, Recital 17, see also Recital 25 (scope of prudential regulation to be confined to ‘those entities which, by virtue of running a trading book on a professional basis, represent a source of counterparty risk to other market participants’). 19

37

For an exposition of the underlying policy considerations, cf. MiFID I, Preamble, Recitals 18–19, 22, and 24. 38 MiFID I, Articles 9 and 10, respectively. 39 Significantly, Article 13(4) MiFID I now required specific precautions in the interest of business ‘continuity and regularity in the performance of investment services and activities’, while Article 13(5) stipulated more detailed provisions for the outsourcing of operational functions. 40 See Commission Directive 2006/73/EC of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive, OJ L 241, 2.9.2006, p. 26 (hereafter: ‘Implementing Directive 2006/73/EC’), Chapters 2 and 3. 41 ibid., Article 5 (specifying the requirements under Article 13(2) to (8) MiFID I). 42 ibid., Article 6 (specifying the requirements under Article 13(2) MiFID I). 43 ibid., Article 7 (specifying the second subparagraph of Article 13(5) MiFID I). 44 ibid., Article 8 (specifying the second subparagraph of Article 13(5) MiFID I). 45 ibid., Article 9 (specifying Article 13(2) MiFID I). 46 ibid., Article 16 (specifying Article 13(7) and (8) MiFID I). 47 ibid., Preamble, Recitals 4 and 7. 48 ibid., Recital 5. 49 Basel Committee on Banking Supervision, ‘International Convergence of Capital Management and Capital Standards—A Revised Framework’ (June 2004). 50 Directive 2006/48/EC of the European Parliament and of the Council of 14 June 2006 relating to the taking up and pursuit of the business of credit institutions, OJ L 177, 30.6.2006, p. 1 (the ‘recast Banking Directive’); Directive 2006/49/EC of the European Parliament and of the Council of 14 June 2006 on the capital adequacy of investment firms and credit institutions (recast), OJ L 177, 30.6.2006, p. 201 (the ‘recast Capital Adequacy Directive’). 51 See, e.g., Mülbert and Wilhelm (n. 13), para. 6.05. 52 Directive 2010/76/EU of the European Parliament and of the Council of 24 November 2010 amending Directives 2006/48/EC and 2006/49/EC as regards capital requirements for the trading book and for re-securitizations, and the supervisory review of remuneration policies, OJ L 329, 14.12.2010, p. 3, Article 1(3). 53 ibid., Preamble, Recitals 7–10 and 17–22. As for the Green Paper, see ibid., Recital 17 and supra text and n. 11. 54 Supra n. 7. 55 See MiFID II, Article 9(1), referring to CRD IV, Articles 88 and 91. 56 MiFID II, Article 9(3)–(6). 57 MiFID II, Articles 10–13. 58 See, again, supra text accompanying n. 9.

59

See, again, MiFID II, Preamble, Recital 5 (quoted supra text accompanying n. 9). See, for further discussion, Mülbert and Wilhelm (n. 13), paras 6.75 and 6.76. 61 See, for further discussion, infra III.6.A. 62 Supra II.A. 63 For a good introduction to the problem and analysis of the CRD IV framework in this respect, see Mülbert and Wilhelm (n. 13), paras 6.38–6.47. 64 Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012, OJ L 176, 27.6.2013, p. 1. Article 4(2) CRR extends to persons subject to the 60

requirements imposed by [MiFID I], excluding (a) credit institutions; (b) local firms; (c) firms which are not authorized to provide the ancillary service referred to in point (1) of Section B of Annex I to Directive 2004/39/EC, which provide only one or more of the investment services and activities listed in points 1, 2, 4 and 5 of Section A of Annex I to that Directive, and which are not permitted to hold money or securities belonging to their clients and which for that reason may not at any time place themselves in debt with those clients. Obviously, this definition must now be read as referring to the definition of investment firms as stipulated by Article 4(1)(1) MiFID II, with the exception of firms specified in Article 4(1)(2)(a)–(c) MiFID II; see Mülbert and Wilhelm (n. 13), para. 6.11 notes 23 and 24. 65 Cf. Articles 88, 91 in conjunction with Article 2(1) CRD IV. 66 That is, those firms mentioned in CRR, Article 4(1)(b) and (c): ‘local firms’ and ‘firms which are not authorized to provide the ancillary service referred to in point (1) of Section B of Annex I to Directive 2004/39/EC, which provide only one or more of the investment services and activities listed in points 1, 2, 4 and 5 of Section A of Annex I to that Directive, and which are not permitted to hold money or securities belonging to their clients and which for that reason may not at any time place themselves in debt with those clients’. 69 CRD IV, Article 88(1)(e). 67 Cf. CRD IV, Preamble, Recital 55; replicated in MiFID II, Preamble, Recital 55. Consequently, the definition of a ‘management body’ set out in Article 3(1)(7) CRD IV and, in almost identical wording, Article 4(1)(36) MiFID II follows a functional approach, referring to the ‘body or bodies […] which are appointed in accordance with national law, which are empowered to set the institution’s/entities strategy, objectives and overall direction, and which oversee and monitor management decision-making, and include the persons who effectively direct the business of the institution/entity’. 68 Cf. Article 3(2) CRD IV and Article 4(1)(36) MiFID II, pursuant to which, ‘Where this Directive refers to the management body and, pursuant to national law, the managerial and supervisory functions of the management body are assigned to different bodies or

different members within one body, the Member State shall identify the bodies or members of the management body responsible in accordance with its national law, unless otherwise specified by this Directive’. See, for an in-depth discussion and critique, Mülbert and Wilhelm (n. 13), paras 6.38–6.47. 70 Mülbert and Wilhelm (n. 13), paras 6.38. 71 See, critiquing the underlying policy decision against the available empirical evidence, Luca Enriques and Dirk Zetzsche, ‘Quack Corporate Governance, Round III? Bank Board Regulation Under the New European Capital Requirement Directive’ (2015) Theoretical Enquiries in Law 16, 211, 233–5; similarly Mülbert and Wilhelm (n. 13), para. 6.77 (expressing their preference for a more flexible comply-or-explain approach). 72 CRD IV, Articles 76(3), 88(2), and 95(1). 73 Supra n. 7. 74 See supra n. 69. 75 See, again, CRD IV, Article 3(1)(7) and MiFID II, Article 4(1)(36), quoted supra n. 68. 76 To be discussed in detail infra III.3. 77 At least two persons, cf. CRD IV, Article 13(1) and MiFID II, Article 9(6)—note the exception in the second subparagraph of the latter provision. 78 MiFID II, Article 9(3)(3). 79 Contrast, e.g., the requirements pursuant to Article 88 CRD IV and Article 9 MiFID II with the general exposition of tasks of the management body in Sections 76, 77, 91, and 93 of the German Aktiengesetz (Stock Corporation Act), which are formulated in much more general terms and, with the exception of Section 91(1) (on bookkeeping) and Section 91(2) (on risk-control arrangements) hardly mention any specific responsibilities at all. 80 Cf. MiFID II, Article 8(d). 81 CRD IV, Article 67(2)(f)–(g). 82 As to which, see infra text accompanying n. 90. 83 See supra text accompanying n. 18. 84 See, for details, CRD IV, Article 88(2)(2)(a)–(d). For a discussion of the diversity requirement as such, see infra 3. 85 CRD IV, Article 95(2). 86 CRD IV, Article 76(3). 87 Directive 2006/43/EC of the European Parliament and of the Council of 17 May 2006 on statutory audits of annual accounts and consolidated accounts, amending Council Directives 78/660/EEC and 83/349/EEC and repealing Council Directive 84/253/EEC, OJ L 157, 9.6.2006, p. 87. But cf. CRD IV, Article 76(3), pursuant to which non-significant institutions may combine their risk committee and their audit committee in one committee. For further discussion, see Mülbert and Wilhelm (n. 13), para. 6.13. 88 Cf., e.g., Hopt, in Wymeersch, Hopt, and Ferrarini (n. 2), para. 11.18; for empirical assessments, see again supra n. 16. 89 CRD IV, Article 91(2) and MiFID II, Article 9(4).

90

CRD IV, Article 91(8). CRD IV, Article 91(10). 94 CRD IV, Article 88(2)(a). 91 CRD IV, Article 91(7). 92 CRD IV, Article 91(9). 95 Promulgation of which was required by 31 December 2015. As of October 2016, however, the guidelines have yet to be published. 96 Cf., e.g., the much less complex requirements under the ISD 1993, Article 3(3) (quoted supra text accompanying n. 23). 97 Cf., discussing the German practice in banking supervision in this regard, JensHinrich Binder, ‘Vorstandshandeln zwischen öffentlichem und Verbandsinteresse— Pflichten- und Kompetenzkollisionen im Spannungsfeld von Bankaufsichts- und Gesellschaftsrecht’ (2013) ZGR Zeitschrift für Unternehmens- und Gesellschaftsrecht, 760, 774–5. 98 See, expressing similar doubts, also Mülbert and Wilhelm (n. 13), para. 6.67; see also Jaap Winter, ‘The Financial Crisis: Does Good Corporate Governance Matter and How to Achieve It?’ in Wymeersch, Hopt, and Ferrarini (eds) (n. 2), para. 12.28; Enriques and Zetzsche (n. 72), 226. 99 Cf. CRD IV, Preamble, Recital 60, according to which board composition ought to be ‘sufficiently diverse as regards age, gender, geographical provenance and educational and professional background to present a variety of views and experiences’. 100 Contrast, e.g., Irene van Staveren, ‘The Lehman Sisters hypothesis’ (2014) Cambridge Journal of Economics 38, 995 (finding that women in boards tend to perform better than men when deciding under uncertainty), with Renee B. Adams and Vanitha Ragunathan, ‘Lehman Sisters’ (2015), FIRN Research Paper, available at SSRN: (finding that this is not necessarily the case). For an extensive review of the available economic literature without a special focus on the financial sector, see Enriques and Zetzsche (n. 72), 219–25. For a general analysis of ‘groupthink’ as a corporate governance problem, cf. Andrew Howard, ‘Groupthink and Corporate Governance Reform: Changing the Formal and Informal Decision-making Processes of Corporate Boards’ (2011) S. Cal. Interdisc. L. J. 20, 425. 101 Enriques and Zetzsche, ibid., pp. 224–5. 102 Supra paragraph 3.34. 103 Green Paper (n. 11), 17–18. 104 Cf. Financial Stability Board (then Financial Stability Forum), Principles for Sound Compensation Practices (2 April 2009), available at , and FSB, Principles for Sound Compensation Practices— Implementation Standards (25 September 2009), available at . 105 Cf., e.g., discussing the introduction of relevant regulations in Germany, Jens-Hinrich Binder, ‘Steuerung und Kontrolle von Vergütungssystemen durch die BaFin’ in Volker 93

Rieble, Abbo Junker, and Reinhard Giesen (eds) Finanzkriseninduzierte Vergütungsregulierung und arbeitsrechtliche Entgeltsysteme (Munich: ZAAR-Verlag, 2011), p. 63. 106 Cf. MiFID II, Article 9(3)(c), pursuant to which the policy has to aim ‘to encourage responsible business conduct, fair treatment of clients as well as avoiding conflict of interest in the relationships with clients’. 107 Supra, n. 7. 108 Supra n. 73 and accompanying text. 109 See, for introductions to the regime, e.g., Moloney (n. 11), 388–90; Mülbert and Wilhelm (n. 13), at paras 6.16–6.20. And see generally Tom Dijkhuizen, ‘The EU’s Regulatory Approach to Banks’ Executive Pay: From Pay Governance to Pay Design’ (2014) European Company Law 11, 30; Eilís Ferran, ‘New Regulation of Remuneration in the Financial Sector in the EU’ (2012) ECFR 9, 1; Guido Ferrarini and Maria Cristina Ungureanu, ‘An Overview of the Executive Remuneration Issue Across the Crisis’ in Birkmose, Neville, and Sørensen (eds) (n. 16), 349; Guido Ferrarini and Maria Cristina Ungureanu, ‘Lost in Implementation: The Rise and Value of the FSB Principles for Sound Compensation Practices at Financial Institutions’ (May 2011) Revue Trimestrielle de Droit Financier, 1–2, 60; Andrew Johnston, ‘Preventing the Next Financial Crisis? Regulating Bankers’ Pay in Europe’ (2014) Journal of Law and Society 41, 6; for a US perspective, cf. Lucian Bebchuk and Holger Spamann, ‘Regulating Bankers’ Pay’ (2010) Geo. L. J. 98, 247. 110 See, again, supra n. 65 and accompanying text. 112 CRD IV, Article 74(1). 111 See, for a brief comparison with the CRD I–III regimes, Mülbert and Wilhelm (n. 13), para. 6.21. 116 CRD IV, Article 76(1). Note the similarities with the formulation of board duties in Article 88(1), quoted supra text accompanying n. 77. 113 CRD IV, Article 74(2), The present version dates back to the corresponding provision of Article 22 in the recast Banking Directive of 2006 (supra text and n. 51), see EBA, Guidelines on Internal Governance (GL 44) (27 September 2011). 114 CRD IV, Article 76(3) and (4), see supra text accompanying n. 86. 115 See, for details, ibid., Article 76(5); Mülbert and Wilhelm (n. 13), para. 6.23. 117 ibid., Article 76(2). 118 Iris H.-Y. Chiu, Regulating (from) the Inside—The Legal Framework for Internal Control in Banks and Financial Institutions (Oxford: Bloomsbury, 2015), p. 87. 119 See, again, supra III.2.A. 120 For an excellent, in-depth analysis see Chiu (n. 119), 77–118. 121 In the words of John Braithwaite, ‘Rules and Principles: A Theory of Legal Certainty’ (2002) Australian Journal of Legal Philosophy 27, 47, 55–6. 122 See generally, e.g., Jens-Hinrich Binder, Regulierungsinstrumente und Regulierungsstrategien im Kapitalgesellschaftsrecht (Tübingen: Mohr Siebeck, 2012), p.

181; Jens-Hinrich Binder, ‘Prozeduralisierung und Corporate Governance— Innerbetriebliche Entscheidungsvorbereitung und Prozessüberwachung als Gegenstände gesellschaftsrechtlicher Regulierung’, (2007) ZGR Zeitschrift für Unternehmens- und Gesellschaftsrecht 745, 783–7. And see, stressing similar potential weaknesses in the context of governance-related regulation of financial institutions, also Chiu (n. 119), 30–3. For a more optimistic call for a regulatory focus on board processes, cf. also Nicola Faith Sharpe, ‘Process Over Structure: An Organizational Behavior Approach to Improving Corporate Boards’, (2011) S. Cal. L. Rev. 85, 261. 123 Binder (n. 25), 703. 124 See Commission Delegated Regulation of 25 April 2016 (n. 7), in particular Articles 21 (general requirements), 22 (on compliance requirements), 23 (on risk management), 24 (on internal audit), and 25 (responsibilities of senior management). 125 See, again, the summary of organizational requirements under the ISD 1993 and MiFID I supra II.B. 126 MiFID II, Article 16(4). 127 ibid., Article 16(5), subparas 2 and 3, respectively. These requirements are complemented—again, in rather abstract and vague terms—by corresponding implementing provisions in Articles 23 and 24 of Commission Delegated Regulation of 25 April 2016 (n. 7). 128 ibid., Article 16(2). 129 Supra, n. 7. 130 ibid., Article 9(3)(4). 131 ibid., Article 16(3), first subparagraph. This requirement, in conjunction with Article 23 of the Directive, is specified further in Articles 33–35 of the Commission Delegated Regulation of 25 April 2016 (n. 7). 132 MiFID II, subparas 2 and 3. 133 ibid., subpara. 4. 134 ibid., subpara. 7. 135 ibid., Article 16(6) and (7). 136 ibid., Article 16(8)–(10). 137 It is perhaps telling, in this context, that the relevant international standards on riskmanagement approaches for securities firms have been developed in line with prudential standards developed for credit institutions—compare, e.g., International Organization of Securities Commissioners, ‘Risk Management and Control Guidance for Securities Firms and their Supervisors’ (May 1998), available at , with Basel Committee on Banking Supervisors, ‘Risk Management Practices and Regulatory Capital’ (November 2001), available at . 139 MiFID II, Article 4(1)(31), see also CRD IV, Article 3(1)(33), referring to CRR, Article 4(1)(36). 138 Article 10(1) MiFID II and CRD IV, Article 14(1).

140

MiFID II, Article 10(1)(2), cf. CRD IV, Article 14(2). MiFID II, Article 13(1), cf. CRD IV, Article 23(1). Note that—convincingly, but in contrast to the MiFID II, where Article 10 makes no reference to Article 13—these requirements are also to be applied in conjunction with the supervisory scrutiny of qualifying holdings in the initial authorization process (CRD IV, Article 14(2)). In practice, however, authorities are also likely to apply the same criteria under the initial authorization process in a ‘pure’ MiFID II scenario. 142 See also Moloney (n. 11), 391. 143 See supra II. 144 See, again, supra n. 18 and accompanying text. 145 EBA, ‘Report on Investment Firms—Response to the Commission’s Call for Advice of December 2014’, EBA/Op/2015/20 (2015), available at . See also EBA, ‘Opinion of the European Banking Authority on the First Part of the Call for Advice on Investment Firms’, EBA-Op-2016-16 (2016), available at . 146 See, again, supra III.3. 147 See, again, Binder, ZGR Zeitschrift für Unternehmens- und Gesellschaftsrecht (2015), pp. 667, 702–3; cf. also, discussing the prospect of governance-oriented regulation of banks participating in the Banking Union, Binder, ‘The Banking Union and the Governance of Credit Institutions: A Legal Perspective’ (2015) European Business Organization Law Review 16, 467, 478–87. 148 See, again, Binder, ZGR Zeitschrift für Unternehmens- und Gesellschaftsrecht (2015), pp. 667, 702–3, 707; Chiu (n. 119), 30–3; Enriques and Zetzsche (n. 72), 227–9. 149 See, again, CRD IV, Article 88(1)(e), on which, see supra III.2.A. 150 See, again, Enriques and Zetzsche (n. 72), 232; Mülbert and Wilhelm (n. 13), para. 6.77. 151 See, again, Mülbert and Wilhelm (n. 13), paras 6.38–6.47. 152 Cf., e.g., Section 91(2) of the German Stock Corporation Acts. 153 Cf. Binder, ZGR Zeitschrift für Unternehmens- und Gesellschaftsrecht (2015), pp. 667, 704 (discussing group-wide organizational requirements under EU and national banking regulation, which may conflict with limits to organizational choices within groups under German group law). 154 ibid., pp. 705–6. 141

4 THE OVERARCHING DUTY TO ACT IN THE BEST INTEREST OF THE CLIENT IN MIFID II Luca Enriques and Matteo Gargantini

I. Introduction II. Antecedents of the Investment Firm’s Duty to Act in the Best Interest of the Client III. The Duty to Act in the Best Interest of the Client in MiFID II IV. The Duty’s Functions and Contents V. The Articulation of the Duty with Respect to Individual Services and Activities 1. The Duty and Its ‘Natural Bedfellows’: The Boundaries of Investment Advice 2. The Duty and Its ‘Odd Bedfellows’: At-Arm’s-Length Services 3. Manufacturing 4. Self-Placement 5. Dealing on Own Account VI. A Look Across the Atlantic VII. Assessment

I. Introduction [4.01] MiFID II confirms, in line with its predecessors, that firms providing investment services (hereafter, ‘investment firms’) shall act in accordance with the best interests of their clients (Article 24(1)). Besides being an overarching principle for firms’ professional conduct, the duty to act in accordance with the client’s best interest (hereafter, ‘the best interest duty’ or simply ‘the duty’) is further specified by more detailed provisions within the body of MiFID II and its implementing measures, playing a role itself in their interpretation. Rules concerning the management of conflicts of interest, including at the level of staff remuneration practices (Articles 23 and 24(10)), limitations to inducements (Article 24(9)), and best execution (Article 27) are just a few examples of obligations aimed at ensuring that clients’ interests are pursued even when firms’ incentives may be tainted. Another typical specification of the duty is the duty to apply a suitability or, depending on the service provided, an appropriateness assessment before entering a transaction with a client (Article 25(2) and (3)). [4.02] The duty may have direct application as well, thus grounding firms’ liability even in cases where no specific rule of conduct is violated. Not only does breach of the generic best interest duty trigger administrative sanctions (Article 70(3)(a)(x) MiFID II), but it may also entail private enforcement by dissatisfied investors (Article 69(2) MiFID II).1 [4.03] Although the duty is not new, MiFID II further specifies its implications and, overall, enhances its effects: it does so, as we shall see, by introducing new specific rules of conduct implementing the ‘client’s best interest’ standard and by broadening its scope of application. After sketching out the origins of this duty (Section II), the chapter maps its use in MiFID II (Section III) and analyses it in its various functions: first, as a self-standing source of obligations for investment firms; second, as guidance for the interpretation of the more specific duties MiFID II imposes on investment firms; third, as guidance for EU policymakers in their drafting of implementing rules (Section IV). Section V focuses on how the duty fits well with some of the traditional investment services (portfolio management, advice, and order execution), while it is hard to reconcile with

the features of typically arm’s-length activities such as dealing on own account and self-placement. Section VI takes a look at the ongoing US debate on whether fiduciary duties – to which the best interest duty is normally ascribed – should apply to brokers, reporting the problems with their extension as highlighted by US legal scholars. Finally, Section VII provides an overall assessment of the MiFID II framework concerning the best interest duty and of its efficiency.

II. Antecedents of the Investment Firm’s Duty to Act in the Best Interest of the Client [4.04] The best interest duty has a long history in common law countries, where it is often associated with fiduciary relationships. In the common law tradition, the duty was first developed in equity as a flexible tool capable of ensuring protection in circumstances where this would be impossible under predetermined rules.2 The principle offered a highly protective standard implying the duty to pursue the clients’ interests above the fiduciaries’,3 if not even to act in the clients’ sole interests.4 Such an obligation is regarded as a core element of fiduciary relationships—so much so that it is sometimes regarded as the archetypical component thereof5—and has a pervasive role in ensuring investor protection in trust-based relationships.6 [4.05] Financial market law in the UK has initially been reluctant to make explicit reference to this duty as a general principle. For instance, in setting out the rules applicable to the advice and distribution industry, the Core Conduct of Business Rules adopted by the Securities and Investments Board (SIB) upon delegation by the Financial Services Act 1986 made independent financial advisers and tied agents subject to a ‘best advice rule’ when they made personal recommendations on packaged investment products (including units in regulated collective investment schemes). The complexity and the opacity of such products justified strengthened rules of conduct that prevented both independent advisers and tied agents from recommending a packaged product if they were ‘aware of a […] packaged product which would better meet [the client’s] needs’.7 However, while

identification of the best product was to be made out of the products generally available on the market for independent financial advisers, tied agents could also refer to products of their firm or marketing group alone.8 For this reason, the ‘best interest’ requirement was considered a distinguishing feature of independent advisors, in line with their nature as investors’ agents, with tied agents acting as salespersons on behalf of product manufacturers.9 In this context, scholars considered the best advice rule as an application of the best execution principle,10 where the duty to act in the best interest of the client was explicitly stated,11 in line with its origin as a typical fiduciary duty.12 The 1986 ‘polarized’ system was subsequently abandoned for a more homogeneous regulatory framework for investment advice,13 and in 2004 the UK FSA (now FCA) Principles for Businesses adopted the current milder14 formulation that refers to the duty to pay ‘due regard’ to the interests of customers, and to treat them fairly.15 [4.06] International recommendations and codifications of best practices have been less reluctant to openly expand the obligation to pursue the customers’ best interest into an across-the-board principle applicable to any business activity, with no explicit differentiation between arm’s-length (e.g. dealing) and fiduciary-based activities (e.g. advice): the 1990 IOSCO International Conduct of Business Principles—which were overall strongly influenced by the UK conduct of business rules16—recommended that firms should act ‘with due skill, care and diligence, in the best interests of [their] customers and the integrity of the market’;17 a wording subsequently reflected in Article 11(1) Directive 93/22/EEC and, later on, in Article 19(1) MiFID I and Article 24(1) MiFID II.

III. The Duty to Act in the Best Interest of the Client in MiFID II [4.07] According to Article 24(1), ‘member states shall require that, when providing investment services or, where appropriate, ancillary services to clients, an investment firm act […] in accordance with the best interests of its clients’ (emphasis added). Such a wording may seem to imply something

less than a ‘duty to act in the best interest of the client’, but the two can be regarded as synonymous, given their interchangeable use throughout MiFID II. [4.08] In fact, the expression ‘[duty to act] in accordance with the best interests of its clients’, the seemingly broader ‘[duty to act] in the best interests of its clients’, or even the expression (emphasis added in the following instances) ‘[duty to act] in the interest of the client’ can be found elsewhere in Article 24 to refer to the same duty. First, Article 24(2), second para., requires investment firms ‘to ensure that financial instruments are offered or recommended only when this is in the interest of the client’. Second, Article 24(7)(b) clarifies that advice can be qualified as independent also when the investment firm receives ‘[m]inor nonmonetary benefits that are capable of enhancing the quality of service provided to a client and are of a scale and nature such that they could not be judged to impair compliance with the investment firm’s duty to act in the best interest of the client’. The exact same wording can also be found in Article 24(8), which allows investment firms providing portfolio management services to accept inducements with the same characteristics. In addition, according to Article 24(10), an investment firm ‘shall ensure that it does not remunerate or assess the performance of its staff in a way that conflicts with its duty to act in the best interests of its clients’. [4.09] Other provisions in the same Article rather repeat the wording of Article 24(1) (‘duties [or obligation] to act … in accordance with the best interests of its clients’), namely Article 24(9), first para., sub (b), and third para., and Article 24 (13)(d), addressing the conditions for receiving inducements. [4.10] As anticipated, there is no reason whatsoever to consider the duty to act in the best interest of the client as a different—and possibly stricter— standard than the one in Article 24(1). Recital 71 first para., confirms that the two are one and the same: referring to Article 24(2), it states that investment firms need to understand the features of the financial instruments they offer or recommend as this enables them to comply with the duty to ‘act in accordance with the best interest of their clients’, while Article 24(2), as reported above, considers the same duty as a means ‘to

ensure that financial instruments are offered or recommended only when this is in the interest of the client’.18 [4.11] References to Article 24(1) in other MiFID II provisions provide a better understanding of the duty’s scope. In some cases, such provisions clarify that Article 24 applies to a specific service, in others that it does not. For the sake of exposition, we can identify four different settings, depending on how the best interest duty combines with other general standards and with more detailed rules of conduct. [4.12] In the first setting, the best interest duty applies along with the companion Article 24(1) duty ‘to act honestly, fairly and professionally’ (hereinafter, the companion standards) and with more detailed conduct of business rules. For instance, Article 25(1) and (2) specify how investment firms shall assess the suitability/appropriateness of a service or a product, a conduct of business rule which is normally regarded as a specification of the best interest duty.19 Investment firms performing sales activities must make sure that investors have a proper understanding of the investments they are entering into (appropriateness test), while for the higher added value services of financial advice and individual portfolio management the assessment must also include customers’ ability to bear the risk of the losses financial instruments may cause as well as the ability of those instruments to match customers’ investment objectives (suitability test). This regulatory technique raises the questions of whether and, if so, to what extent the best interest duty coexists with requirements established by detailed rules of conduct and therefore potentially adds on further duties to those stemming from such rules. [4.13] In a second setting, the best interest duty, together with other general standards of conduct, applies, but one or more specific rules of conduct are explicitly waived. For example, the duty to assess the appropriateness of investment services and financial instruments is waived for the execution or reception and for transmission of trading orders concerning non-complex financial instruments, to the extent that such activities are performed at the initiative of the clients (‘execution-only’: Article 25(4) MiFID II). This means that investment firms do not have to obtain information on the (potential) clients’ knowledge and experience,

nor must they assess whether the envisaged investment services or products are appropriate for customers in the light of their knowledge and expertise. However, Article 25(4) does not exempt investment firms from the Article 24(1) duties, including that of best interest. Here, the question is to what extent some of the waived requirements may still be deemed applicable by way of interpretation of the best interest duty and its companion standards, irrespective of the lighter protective regime set forth by detailed rules of conduct. [4.14] In a third setting, investment firms are exempted from the best interest duty, but not from the companion standards. This happens when investment firms perform straightforward execution or transmission of trading orders of—as well as when they deal on own account with —‘eligible counterparties’ (a subset of professional clients: Article 30). In this case, MiFID II requires pre-contractual information and reports on the service provided and its costs (Article 24(4) and (5) and Article 25(6)), with the exclusion of any appropriateness assessment.20 Eligible counterparties receiving those services are also protected by investment firms’ companion standards and by the duty to provide clear and not misleading information, but they cannot rely on the best interest duty (Article 30(1)). This is a remarkable innovation in the MiFID II framework, as MiFID I exempted pure sales transactions with eligible counterparties from both the duty and the companion standards (Article 24 MiFID I). The inapplicability of the best interest duty reflects the fact that parties to an arm’s-length transaction on the wholesale market do not normally protect each other’s interests, unless otherwise agreed. However, MiFID II acknowledges that eligible counterparties may still expect investment firms to comply with companion standards, even if they do not need special protection when making their investment choices. The expansion of fairness principles to such transactions reflects the policy consideration that conduct of business rules should serve not only investor protection, but also market integrity purposes (MiFID II, Recital 164). Hence, the requirement that investment firms behave according to professional standards in all their activities allows for public enforcement intervention beyond the interests and the incentives of clients, who can only be expected to enforce their own rights when their private interest so commands.

[4.15] Fourth, and finally, in some cases MiFID II excludes application of Article 24(1) in its entirety. For instance, non-discretionary crossing of buying and selling interests represents a key protective tool for multilateral trading facilities (MTFs) and regulated markets’ members and participants when they route their orders to such trading venues: no space is left for assessing counterparties’ interests with regard to transactions concluded on MTFs and regulated markets and therefore Article 24 does not apply (Articles 19(4) and 53(4)).21

IV. The Duty’s Functions and Contents [4.16] As a preliminary matter, let us first clarify that it would make no sense to ask whether the best interest duty stems from the fact that investment services give rise, as such, to fiduciary relationships. Fiduciary relationships are a product of national private (common) law traditions, and nothing authorizes the consideration of MiFID II as a vehicle for the legal transplanting of fiduciary obligations from those jurisdictions to others. Continental European jurisdictions have developed alternative, albeit equivalent, tools in order to satisfy the need to integrate clear-cut but rigid rules with more flexible standards of behaviour, the most cited example being the duty of good faith and fairness.22 MiFID II also refers to this fairness principle (Article 24(1)), along with the best interest duty, thus juxtaposing legal principles with different origins23 in a somewhat uneasy combination. In addition, MiFID II is an autonomous instrument and contains a set of overlapping and complementary duties which would make an inquiry into the fiduciary nature of the best interest duty quite purposeless. [4.17] We are rather interested in understanding, first, what the functions and the common EU core content of the best interest duty are and, second, how the duty applies to the provision of the various investment services. The present section deals with the former question, while Section V focuses on the latter. Both try to answer these questions from a positive, rather than normative, perspective.

[4.18] The best interest duty plays a key role in MiFID II’s framework. First, it acts as a self-standing duty filling the gaps in investor protection left by more specific conduct of business rules. Here, the duty complements such rules of conduct when these do not apply because of their limited scope or when they fall short of delivering a sufficient level of investor protection, however subjective this assessment can be. Second, the duty provides justification for pro-client interpretations, both by supervisors and courts, of the more specific MiFID II conduct of business rules, once again when narrow interpretations would not yield satisfactory results. Third, as an overarching principle in MiFID II, the duty provides guidance to EU policymakers in charge of drafting implementing regulations. Implementing rules or guidelines will sometimes have to be enacted to clarify whether some practices are consistent with the duty: this is namely the case with cross-selling and inducements (Article 24(11), third subpara. and 24(13) (d)). [4.19] As a gap filler, the best interest duty complements the specific conduct of business rules, such as the obligations related to conflicts of interest management, the duty to run appropriateness or suitability tests, the duty of best execution, and so on. First of all, the duty reaches beyond the scope of these specific rules when these do not apply. By the same token, the duty ensures that compliance with the specific rules effectively results in the degree of investor protection that in the given circumstances MiFID II requires. [4.20] Note that, as a self-standing duty, the best interest principle is not limited to areas left uncovered by the specific conduct of business rules. In one case, this is even explicit: Article 24(10) prevents firms from remunerating or evaluating their staff in a way that conflicts with the duty to act in the best interest of their clients. Here, the duty explicitly shapes the contours of a specific governance rule imposed on the investment firm. As the provision itself specifies [i]n particular, [an investment firm] shall not make any arrangement by way of remuneration, sales targets or otherwise that could provide an incentive to its staff to recommend a particular financial instrument to a retail client when the investment firm could offer a different financial instrument which would better meet that client’s needs.

Recital 77 makes clear that, despite the reference to recommendations, such a provision applies to both advisory services and selling activities. [4.21] More generally, and even in the silence of the Directive, the duty also applies to behaviours that specific conduct of business rules do cover: in circumstances where substantial compliance with such rules is insufficient to ensure that the investor’s interest is duly protected, the duty itself kicks in.24 [4.22] In other words, full compliance with the specific conduct of business rules provides no safe harbour for the investment firm: if the treatment the client receives as an outcome of such compliance is not in line with her best interest, then the best interest duty will have been breached. [4.23] But what exactly are the duty’s contents? And how far does it reach? The regulatory framework for investment advice offers a good starting point to illustrate the duty’s implications. [4.24] Investment advice falls within the scope of investment services insofar as it consists in personal recommendations on specific financial instruments (Article 9 EU Commission Delegated Regulation supplementing MiFID II as regards organisational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive—hereinafter ‘Delegated Regulation’; previously Article 52 Directive 2006/73/EC). This would in principle exclude the provision of general advice on a generic kind of financial instrument from its scope, even when that kind of financial instrument is presented as suitable. However, according to Recital 15 Delegated Regulation (previously Recital 81 Directive 2006/73/EC), providing generic advice of this sort – i.e., accompanied by a misleading or false statement of suitability or of previous consideration of the circumstances of a client – would likely result in the violation of the duty: direct application of the duty fills the gaps left open by the inapplicability of the suitability test, because the provision of general recommendations would merely qualify as an ancillary service, subject to less stringent conduct of business rules (Articles 24(1) and 25(1) and Annex I(B)(5) MiFID II).

[4.25] The duty can also complement specific rules of conduct. Rules on conflicts of interests and inducements set obligations aimed at ensuring unbiased recommendation of financial products and investment services (Articles 23 and 24(9)), while the suitability test prevents investment firms from recommending inadequate transactions. When more than one financial instrument or investment service matches the suitability test for a specific client, investment firms will comply with the suitability rule by recommending any of them. And, so long as they comply with processbased rules on conflicts of interest, they are not prevented from recommending products in respect of which they are conflicted.25 [4.26] Such conduct of business rules are also compatible with a model of investment advice where investment firms simply assess whether investment products are suitable and, hence, recommendable, but clients retain full responsibility for selecting among them.26 A more protective form of advice may however add on top of this an investment firm’s duty to identify the best option, in terms of risk-adjusted performance and fees, within the set of suitable products.27 Whether investment firms take on this responsibility may depend on the specific contractual terms, also in the light of the specific circumstances,28 but the best interest duty may facilitate such a result and has occasionally been indicated as functionally equivalent to a duty to select the most suitable product, together with other conduct of business rules.29 [4.27] As hinted already, the best interest duty also acts as guidance to interpretation of specific conduct of business rules. Attempts to enhance investor protection in non-advised services, where specific conduct of business rules aim to ensure appropriateness rather than suitability of the products concerned, provide good examples of how our duty may perform this function. [4.28] In general, the appropriateness test aims to verify whether clients have the expertise and skills required to understand the features of the specific type of product or service they are going to receive. Investment firms have to warn their clients in case they are not satisfied that the product or the service is appropriate, but this does not prevent investment firms and their clients from performing the devised transactions. Therefore,

the appropriateness test helps clients evaluate their ability to take adequate choices, but once the test is performed—with either a positive or a negative result—investors retain full responsibility for their choices and caveat emptor would seem to apply (Article 25(3) MiFID II). The appropriateness test does not ensure the same level of investor protection as the suitability test: it leaves the final say to clients and is not designed to ensure that transactions in financial instruments meet clients’ needs. [4.29] Noteworthy attempts to enhance investor protection in services requiring the mere assessment of appropriateness have been carried out in the wake of the mis-selling scandals that have affected investors in a number of EU Member States.30 Recently, diffuse losses among retail investors were reported in the UK, Germany, Spain, Italy, the Netherlands, and elsewhere.31 Investment decisions taken in connection with such misselling practices were flawed in many respects. Sometimes, investors could not seemingly grasp the actual risks hiding behind the complexity of prepackaged structured products,32 and therefore received a payoff which underperformed that of non-structured products with an equivalent risk.33 In its opinion on MiFID I practices for firms selling complex products, ESMA acknowledged that investment firms’ compliance with rules on (suitability and) appropriateness in the placement of complex financial instruments had been insufficient.34 In what appears to be an implicit use of the general standard to act in accordance with the client’s best interest, the opinion invites national competent authorities to make sure that firms do not rely excessively on standardized processes to assess appropriateness, as they should rather focus on a substantial assessment of clients’ ability to understand complex products.35 At the national level, this guideline has been further specified by some national competent authorities in the sense that relevant information concerning clients’ features should be gathered in ‘indirect’ ways, so as to avoid any self-assessment by customers.36 [4.30] Observing that the best interest duty can fill regulatory gaps or complement detailed conduct of business rules, including by way of interpretation, begs of course the question of what kind of behaviour the duty requires. Of course, no precise a priori answer to this question can be given: as a general standard of behaviour, only ex post specification by the

relevant enforcement actors is possible. Hence, what can be said is that the duty is open-ended, lending itself to imposing client-regarding behaviour and therefore protection of the client’s interests in the absence of specific conduct of business rules or whenever existing conduct of business rules do not cover a given situation, but at the same time the investor-protection rationale that justifies the application of specific rules in similar circumstances applies to that very situation. Another way of putting this is that the best interest duty moves the guiding principles underlying client– firm relationships in investment services provisions away from caveat emptor, but to what extent that results in requiring investment firms to neglect their own interests to the benefit of their counterparties depends on the specific circumstances and, crucially, given decentralized enforcement, on prevalent interpretations at national level. [4.31] This is the result of MiFID II’s approach of shaping the contractual relationships among private parties—a technique sometimes referred to as of ‘regulatory contract law’37—by mandating the application of general principles whose contents cannot easily be defined in a uniform manner across the EU. The best interest duty and the companion standards are largely inspired by national legal systems and, absent a sufficiently broad EU case law and ESMA guidance, national courts and regulators will inevitably interpret them according to their own legal traditions. In this respect, it is of little consequence whether, in principle, such duties should be defined by way of autonomous interpretation or whether Article 24(1) should instead be understood as an implicit reference to national legal systems. Irrespective of the abstract answer to such questions, the second prevails in practice, so that those general duties are sometimes considered as a way to ensure that investment firms comply with the national private law duties underlying the core aspects of the investment service provided in the light of the contractual agreement between the parties.38 Far from being a mere repetition of already applicable rules, this regulatory technique would strengthen investor protection by enabling public enforcement of such duties.

V. The Articulation of the Duty with Respect to Individual Services and Activities [4.32] Having articulated the core functions of the best interest duty in the MiFID II framework, we are ready to move on to delineate how the duty operates with respect to individual investment services. To do so, we will distinguish between services based on a trust-like relationship and services (otherwise) entailing arm’s-length transactions (hereinafter also arm’slength services). The analysis will confirm the intuition that the less an investment service or activity displays the features of a fiduciary relationship, the harder it is to gauge how the duty applies to it and, correspondingly, the more awkward the position of the investment firm is if, in these cases as well, the duty is to be taken seriously. [4.33] In fact, the operation of the investment firm’s best interest duty is unproblematic when it comes to the provision of investment advice and individual portfolio management, which are normally classified as fiduciary relationships.39 The problems come when applying the same duty to (otherwise) arm’s-length (or transactional) relationships such as selfplacement and dealing on one’s own account. Here, the question is to what extent Article 24(1) reflects an actual reliance on investment firms’ commitment to act to the benefit of their counterparties in the same fashion as in a relationship of pure trust and confidence.40 Introducing quasifiduciary legal obligations in transactional investment services may not always be efficient, as it may increase uncertainty as to investment firms’ expected behaviours even when the parties prefer to negotiate clear-cut rules in advance rather than relying on flexible, but fatally vague, general standards.41

1. The Duty and Its ‘Natural Bedfellows’: The Boundaries of Investment Advice [4.34] Investment advice and discretionary portfolio management are, so to speak, the natural bedfellows of the best interest duty. These financial services are inevitably prone to contractual incompleteness, because the

parties cannot foresee all future states of the world and, hence, set in advance how they should behave with regard to any possible circumstances. By the same token, investments normally need quite a long time before their results can be evaluated, meaning that investors are only able to assess the quality of the provider’s performance when it might be too late to react; in addition, they may well never be able to tell whether a lower than expected return is the outcome of disloyalty, incompetence, or just bad luck.42 It is therefore inherent to both investment advice and discretionary portfolio management services that the investment firm serves its clients’ interests in making delegated investment and divestment choices and in providing recommendations on which investment choices to make. In this context, the best interest duty may still play a decisive and sometimes problematic role in the determination of the boundaries of investment advice, in particular when this combines with other services and activities, such as placing of financial instruments and self-placement. Some of the mis-selling scandals that have occurred in recent years reveal that household investors’ portfolios were excessively concentrated. Recent bank resolutions through bail-in43 in markets characterized by widespread direct access by retail investors to bonds44 have provided anecdotal evidence that in a number of cases small investors’ portfolios’ exposure to subordinated bonds would not have stood a suitability scrutiny. Whether this excessive concentration depends on diffuse violation of rules implementing MiFID I is hard to tell, as its determinants may be manifold. [4.35] ESMA has recently qualified subordinated bonds as complex financial instruments,45 but this was not necessarily the interpretation at national level beforehand.46 Thus, some purchases might have taken place according to the execution-only regime, which may be opted into if the service is provided at the initiative of the client (Article 25(4)(b)). Although a limited offer of products for sale can easily skew the composition of investors’ portfolios in peripheral market contexts where competition is reduced, high concentration of investments can reasonably be taken as indirect evidence that the relevant financial instruments had been recommended to clients. The critical element is whether such messages were presented as suitable or based on a consideration of the clients’ characteristics, bearing in mind, though, that no duty exists for providers of

non-independent advice to consider a broad range of financial instruments (Article 24(7)(a), a contrario). [4.36] Whether or not MiFID I conduct of business rules were violated, it still comes as no surprise that disgruntled investors or national competent authorities may invoke an extensive interpretation of rules mandating a suitability test to hold investment firms accountable in these scenarios, and an even more protective effect may be achieved by stressing the flexibility of general standards such as the duty to act in the client’s best interest. [4.37] The definition itself of investment advice, as previously complemented by Level 2 Directive 2006/73/EC (Article 52), was somewhat ambiguous in providing that such an investment service was only provided when recommendations were presented as suitable for the client or were based on a consideration of the circumstances of that client. The Delegated Regulation adopted pursuant to Article 4(2) MiFID II substantially replicates the previous rule (Article 9), so that the question on non-tailored recommendations remains unanswered at the legislative level.47 As a by-product of an apparently misguided attempt to regulate through definitions rather than by way of straightforward rules of conduct, recommendations that are neither grounded on the client’s characteristics nor presented as such fall outside the scope of investment advice and merely qualify as ancillary services (Annex I(B)(5) MiFID II). Requested to clarify the point, the Commission has acknowledged that Article 52 requirements on personalization of recommendations belong to the definition of investment advice, while stating at the same time that a recommendation which is given to a client with no consideration of its suitability would be forbidden ‘in the light of the fiduciary obligations firms are subject to’.48 The Commission’s position on recommendations that are not presented as suitable highlights the role of the quasi-fiduciary duty to act in the client’s best interest as a potential gap-filler, because Article 24(1) prevents circumvention of the suitability test. As a result, investment firms cannot take advantage of the narrow definition of investment advice with a view to recommending unsuitable financial products when they perform pure brokerage or dealing activities.49

[4.38] In principle, brokers’ required or voluntary provision of information to clients entails no presumption that investment firms assess the suitability of the displayed products. This distinguishes selling services coupled with generic advice from investment advice tout court, although the previous analysis shows this distinction may be blurred in practice. As mentioned, Recital 15 Delegated Regulation specifies that providing generic advice on a type of financial instrument which is presented as suitable would not qualify as investment advice—as this requires a recommendation be given on a specific financial instrument—but would contravene the best interest duty and its companion standards if that advice were not suitable. By the same token, Recitals 16 and 17 expand conduct of business rules from the core investment service provided (e.g. order execution) to preliminary steps such as the provision of generic advice or of general recommendations through distribution channels. This broadened scope of investor protection beyond the limits of clear-cut rules of conduct in some critical circumstances such as placing of financial instrument heavily relies on the gap-filling function of the best interest duty. In particular, it prevents investment firms from easily circumventing those rules by steering their clients’ investment decisions while providing nonadvised services. [4.39] This is a very sensitive area where policymakers are still struggling to find a proper regulatory balance, 50 and simply relying on a broad interpretation of the best interest duty may not be sufficient.51 Empirical evidence suggest that extending the scope of investment advice, or of the duties attached to its provision, outside a well-defined area of personalized recommendations concerning specific products may induce investment firms to refrain from providing any kind of information to their clients, as this would entail the risk of being bound by higher standards of behaviour and of being exposed to the ensuing costs and liability risk.52 Such unintended consequences may occasionally reduce investor protection for those clients who cannot afford the costs of fully fledged investment advice.53 [4.40] This makes remuneration of advice a particularly sensitive issue. Fee-free advice is widespread across many EU jurisdictions because of households’ notorious reluctance to pay to obtain personalized

recommendations.54 Conflicts of interests associated with commissionbased remuneration are however particularly problematic, and facilitated some past mis-selling scandals: in the UK, investor protection concerns led to national super-equivalent provisions aimed at curbing commission-based advice.55 However, non-advised sales accompanied by provision of complete information on a wide range of financial instruments entail significant costs: such a business model may not be economically viable without at least implicit personal recommendation and the subsequent higher take-up of products.56 This brings us back to the original problem: is commission-based advice preferable to a system where investment firms have no incentives to provide investors with sufficient information? Or is no information still better than biased information, considering investment firms’ incentives? [4.41] When investment services and activities other than investment advice and discretionary portfolio management are involved, things become even less straightforward. In the absence of a statutory duty to act in the client’s best interest, parties may or may not explicitly contract for such a duty. In their silence, however, an external observer could hardly tell whether a transaction is accompanied by an investment firm’s implicit commitment to act in the client’s best interest, or by the corresponding client’s expectation, and ex post unilateral statements on the matter could hardly be relied upon.57 We deal with these issues in the following subsections.

2. The Duty and Its ‘Odd Bedfellows’: At-Arm’sLength Services [4.42] When providing pure execution and ‘sale’ services, such as brokerage, placement, self-placement, and dealing on own account, investment firms, which we refer to as broker-dealers in the following, do not take final decisions on client’s investments, nor are they per se required to provide assistance in the assessment of the financial instrument that is being traded or sold, let alone of alternative investment choices. Their contribution to the decision-making is in principle null when investors are

addressing them in order to trade products they have already identified as the target of their purchase, although information concerning the services provided and the financial instruments forming the object of those services is to some extent required (Article 24(3)–(5) MiFID II) or may be provided voluntarily. [4.43] Broker-dealers may also take a more active stance and make potential customers aware that they have financial instruments for sale on their ‘shelves’ and inform them about such products’ features, thus taking the initiative for the execution of trades with or on behalf of clients. Still, this entails no presumption that marketed financial instruments and investment services are regarded as suitable to the potential customers addressed, nor that they are recommended as such, but the risk is of course high that promotional communications slip into concealed recommendations and, thus, investment advice. Concerns that investment firms may circumvent fiduciary-style obligations seem to underlie some national competent authorities’ opinion that investment firms should design standardized internal models for business-to-client relationships and enforce their application by firms’ staff, as any free interaction in such critical circumstances may easily result in investment advice.58 [4.44] It is certain that no presumption exists that the products sold by broker-dealers comprise a wide enough selection of what is on the market or, in other words, that investment firms will make any attempt to display all the available products that would satisfy investors’ needs. This is not even required of investment advisers, unless they present themselves on the market as independent advisers (Article 24(7) MiFID II). [4.45] While the best interest duty entails no obligation to offer a minimum range of financial instruments, it still has some implications that are hard to reconcile with the very nature of sales activities. This is the case, for instance, with the no-profit rule, one of the implications of the duty in its purest form,59 which prevents the agent from taking advantage of its position by extracting wealth from the activities it performs for the beneficiary without authorization (as is the case with incentive-based compensation arrangements60). This restriction is shaped on agency relationships where no room is left for the fiduciary to pursue her own

interest, even when doing so does not harm the beneficiary: intuitively, it sits uneasily with arm’s-length contractual relationships. [4.46] Because the typical remuneration system for investment firms operating as principals, rather than as their customers’ agents, consists of mark-ups and mark-downs61—as opposed to commissions and fees— dealers are inevitably earning from spreads on market prices. This may produce an inherent conflict of interests only to the extent that the dealer is subject to duties that go beyond what is expected from a mere buyer or seller; in any contract, one of the parties has an incentive to appropriate the maximum possible part of the counterparty’s reservation price, but qualifying this as a form of misappropriation would simply place dealers out of business.62 Indeed, the no-profit rule and the remedies typically associated to it, such as disgorgement of profit,63 can hardly fit the normal activity of dealers, because the investment firm has its own (legitimate) interest in the transaction and no one would expect dealers to transfer earnings made out of such a transaction to their clients. At most, the duty of fair dealing may become relevant in this context, if an ‘excessive’ mark-up or mark-down can justify a finding that the contract lacks equilibrium, to the extent that the applicable private law provides for specific remedies (e.g. for unconscionability) or that the competent supervisor can set the appropriate level of spreads. In the US, this is for instance the case with FINRA Rule 2121, which specifies the duty of fair dealing64 by preventing dealers from excessively marking up the price of securities.65 [4.47] Occasionally, a stringent interpretation of the best interest duty may lead to more pervasive effects in the MiFID I and II framework, too. Investment firms offering derivative contracts and acting at the same time as their clients’ counterparties inevitably make (or lose) money when their clients lose (or make) money. An ESMA Q&A document has recently stressed this divergence between the investment firms’ and the clients’ interests in the context of contracts for difference (CFDs) and other complex speculative products such as binary options and rolling spot forex. In the Authority’s view, those contracts entail a conflict of interest that cannot be adequately managed as it is inherent to the investment firms’ proprietary position, unless investment firms hedge, in whole or in part, their client orders. As a consequence, ESMA believes that the incentives to

profit from investors’ losses can by no means comply with the best interest duty, and that the offer of CFDs and other speculative products to retail clients should therefore be avoided altogether.66 [4.48] The Q&A document therefore adopts a very broad notion of conflict of interests. This latter is normally not understood by scholars as to encompass situations where contractual counterparties, which are not linked by any principal–agent relationship, simply aim to maximize their own welfare on a market-based transaction.67 However, the flexibility of the best interest duty seems to offer supervisors a ready-to-use tool to address such situations.

3. Manufacturing [4.49] Not only does MiFID II strengthen rules concerning firm–client relationships, but it also sets specific obligations to consider clients’ interests in manufacturing68 and distribution activities (Article 24(2)).69 These obligations also apply when no suitability assessment is required,70 as they also include mere securities offers besides recommendations, so that the workability of the new provisions for activities falling within the scope of the lighter appropriateness regime remains uncertain. [4.50] By addressing the organizational side of potential misbehaviours, MiFID II leverages on firms’ incentives, as a complement to directly regulating their conduct. This focus on upstream elements de facto broadens the scope of the (quasi-fiduciary) suitability test to the detriment of lighter appropriateness scrutiny regardless of the service actually provided downstream. [4.51] Rules setting out obligations to consider clients’ interests in manufacturing and distribution activities may also result in a selling ban when the client’s best interest principle would be impossible to meet (or to ascertain) for any possible target. In ESMA’s view, when this is the case, national competent authorities ‘should monitor that firms do not offer advice on that envisaged product, or sell it at all’:71 the reference to mere selling activities shows how remarkable the effects of fiduciary-like duties

can be when they apply to arm’s-length transactions. While some competent authorities had already restricted the marketing of certain complex financial instruments at national level long before ESMA’s opinion was released,72 this opinion may facilitate the adoption of similar prohibitions by other authorities, as demonstrated by recent national recommendations that products falling within certain categories of ‘complex products’ not be sold to retail customers, unless ESMA’s standard is met.73 As the provision of arm’s-length investment services and activities creates frictions with the best interest duty, one can easily see how a strict interpretation of this general standard of behaviour may drift into a creeping form of product intervention, and how this may take place outside the procedural requirements for imposing restrictions on the distribution of certain financial instruments or to the performance of certain financial activities under the MiFID II/MiFIR system (Articles 40 ff MiFIR).74

4. Self-Placement [4.52] Under the MiFID I regime, the treatment of direct placement of proprietary financial instruments (self-placement) by investment firms was uncertain. At least in some jurisdictions, the traditional view was that direct placement by issuers—whether corporate or financial institutions—did not fall within the scope of MiFID-regulated activities, because the ability to raise capital that is inherent to the very essence of every company inevitably requires direct contacts with multiple potential investors, and no indication exists that such day-to-day contacts qualify as reserved activities.75 In other countries, self-placement by investment firms was regarded as an investment activity, rather than as an investment service, and hence excluded from the application of conduct of business rules on the basis that dealing solely as a contractual counterparty does not entail any activity performed on behalf of investors.76 Pursuant to either interpretation, the MiFID I investor-protection regime did not extend to self-placement,77 so that in order to make MiFID I conduct of business rules applicable to it, some countries enacted specific provisions.78

[4.53] The Commission’s Q&A Database on MiFID I addressed the point in an inconclusive way. While making clear that financial institutions were subject to ‘know your client’ requirements when their sales persons, on their own initiative, addressed potential investors and recommended financial instruments issued by their own financial institutions, no clear indication was given for cases where no such active solicitation was performed.79 In the 2010 MiFID I review exercise, CESR called for greater clarity on self-placement, no matter whether in connection of investment advice or on a stand-alone basis.80 [4.54] In MiFID II, self-placement explicitly qualifies as an investment service: order execution now includes ‘the conclusion of agreements to sell financial instruments issued by an investment firm or a credit institution at the moment of their issuance’ (Article 4(1)(5)). This broadened definition brings new activities into the scope of EU rules, thereby extending the scope of the best interest duty to an area where firms’ duties as investment firms may combine with firms’ interests as issuers, or as contractual counterparties in a derivative contract, in problematic fashions. In their capacity as issuers or parties to a derivative contract, banks and investment firms have an incentive to reduce the cost of their (debt and equity) capital and to maximize their returns. This can easily jeopardize investor protection because investment firms will benefit from increasing the price of the financial instruments they place, or from placing financial instruments even in cases where these would not stand a suitability or appropriateness scrutiny. [4.55] Self-placement displays peculiar features that might have in principle suited different investment services. Indeed, it shares similarities with at least three investment services listed in MiFID II Annex I (Section A), namely placing of financial instruments, dealing on own account, and order execution. Far from being confined to taxonomy, such qualification has remarkable regulatory implications on the applicability of the best interest duty and its specifications, especially when it comes to the exemptions regime, as we will show. Similarities between placing of financial instruments and self-placement flow from the common element that (debt or equity) capital is raised

through newly issued financial instruments marketed (with or without a public offering81) to a large number of investors. At the same time, it is widely understood that placing requires by its own nature a separation between a service provider and an issuer (or an owner, on secondary markets) because the service provider facilitates access to investors by making its sales network available and therefore operates on the issuer’s (or the owner’s) behalf.82 This explains why no placement service can be generally carried out absent such separation between issuer and investment firm83 and why—as stated by the EU Commission84—no placement occurs when financial instruments are issued on a continuous basis upon investors’ demand rather than being actively marketed on behalf of issuers, as is typically the case with securities offerings during IPOs. [4.56] This does not, of course, mean that that placement (or selfplacement, as the case may be) of financial instruments cannot combine with other investment services, such as investment advice or execution of orders: in this case, investment firms owe their obligations also (or exclusively, in the case of self-placement) to clients buying the relevant financial instruments.85 As CESR noticed under the MiFID I regime, absent such combination, investment firms may be considered as acting on behalf of issuers or offerors alone or, in the case of self-placement, in their own interest, with the risk of leaving investors unprotected.86 [4.57] Self-placement is even closer to selling financial instruments on own account.87 In both circumstances, investment firms operate as direct counterparties to their clients and, at least at the level of the investment firm (as opposed to that of its salespersons or traders), have their incentives driven by price considerations, rather than by commissions or fees. The distinction here concerns not so much the activity provided (as is the case with placing of financial instruments), but rather the context where principal trading is performed, namely the primary versus the secondary market. [4.58] Perhaps in light of the differences between self-placement on one hand, and placing and dealing on own account on the other, the new regime opts instead for the third alternative and assimilates self-placement to execution of orders on behalf of clients (Article 4(1)(5) MiFID II).88 MiFID

II’s focus is therefore on the investor’s side of the transaction, while the role of investment firms in marketing their own financial instruments and/or acting as clients’ direct contractual counterparties plays a minor role.89 [4.59] As a consequence, the sale of financial instruments in the context of self-placement is now subject to MiFID II conduct of business rules. How much this will strengthen investor protection and make the supervisory and judicial scrutiny on investment firms’ behaviour more intense remains to be seen, as legal uncertainty remains high for many reasons. At least four sources of uncertainty can be identified. First, the remaining uncertainties on the reach of the new definition of execution of orders on behalf of clients; second, the MiFID II generous exemptions regime; third, the interactions between self-placement and investment advice; fourth, the implications of the conflict-of-interest regime. [4.60] Let us start with the uncertain reach of the broadened definition of order execution. Some commentators have submitted that the new definition does not include primary market transactions where issuers are not acting in their capacity as investment firms, but just as mere contractual counterparties in arm’s-length transactions.90 This opinion is based on the distinction between the provision of investment services (in favour of a client) and the performance of an investment activity (with a counterparty not receiving any service),91 a distinction which the wording of some of the provisions outlining conduct of business rules would seem to justify: some key MiFID II provisions setting forth rules of conduct explicitly refer, just like MiFID I, to the provision of investment (and in some cases also ancillary) services, but not to the performance of investment activities.92 [4.61] Yet, the Commission has adopted a different interpretation with regard to MiFID I, on the basis that the Directive provided no clear criterion for distinguishing ‘between cases where a service is being provided to a client and […] those where an activity is simply being carried on with a person who is not a client.’93 Therefore, every person entering into a transaction with an investment firm should (shall) be regarded as either an eligible counterparty—and hence as a client, as per Recital 40 MiFID I, now Recital 109 MiFID II94—or as a professional or retail client, and the transaction would (will) necessarily be subject to MiFID conduct of

business rules (Article 19 MiFID I, now Article 24 MiFID II), with the limited exception of best execution in transactions that are not carried out on behalf a client. Nor does MiFID II seem able to materially affect this conclusion, absent any further clarification on the point. For instance, the exemption for persons dealing on own account applies to the extent that such persons do not provide any ‘other’ investment service or perform any ‘other’ investment activity, which suggests that dealing on own account may constitute both a service and an activity (Article 2(1)(2)). In a similar vein, the same exemption does not include dealing on own account when executing client orders (Article 2(1)(d)(iv)), but this does not clarify whether pure dealing on own account would qualify as a service or an activity in the taxonomy of MiFID II.95 All in all, these remaining uncertainties may pave the way to divergent national practices on a crucial matter for investment firms’ proprietary trading, in spite of MiFID II’s harmonization efforts. [4.62] Doubts on the boundaries of order execution reverberate on the exemptions regime, which is the second element of uncertainty. The classification of self-placement as order execution, rather than as placing or dealing on own account, ensures at first sight the widest possible exemptions regime. In fact, order execution qualifies for exemption from MiFID II conduct of business rules both in the case of transactions with eligible counterparties (Article 30, which also entails inapplicability of the best interest duty) and in the case of transactions carried out as executiononly services (Article 25(4)). To the contrary, dealing on own account falls within the special rules for eligible counterparties only, while placement does not qualify for any of the two exemptions. [4.63] The rationale of such policy choice and the related implications are not entirely straightforward. One may wonder why placement of the investment firms’ own products on the primary market can take advantage of the execution-only regime, while this would not seem to be possible for secondary market transactions carried out on a principal basis as long as these qualify as dealing on own account. The opposite might rather be expected, as primary market transactions are normally riskier, for retail clients, than secondary market transactions (if only for lack of a market price). This contradiction would be partially solved if one adopted the

opinion mentioned above that dealing on own account is not subject to (most) conduct of business rules because it is an investment activity rather than an investment service, and is therefore not performed vis-à-vis any ‘client’. In this manner, dealing on own account would simply need no exemptions because it would fall outside the scope of application of the rules addressed by the exemption for execution-only services. However, the EU Commission’s reading of the MiFID I regime on the impossibility of distinguishing between investment activities performed with counterparties and investment services provided to clients makes this interpretation uncertain at best.96 [4.64] Other commentators submit instead that, despite the wording of Article 25(4), the execution-only regime includes cases where an investment firms executes client orders by dealing on its own account.97 As dealing on own account is not mentioned in the rule, this interpretation applies more easily to cases where dealing on own account and execution of client orders—which is explicitly exempted—are jointly performed. But it is hard to see how the execution-only regime could be extended to cases where dealing on own account is carried out at the initiative of clients without entailing an execution of order on their behalf: this is the case when clients request to buy or sell financial instruments ‘from the investment firm’.98 Yet, clients would seem to be more in need of protection when investment firms are executing orders on their behalf, because in that case investment firms’ discretion is broader. [4.65] Furthermore, it is not clear why placing of own products—that qualifies as order execution and is therefore eligible for the execution-only regime—is regarded as less risky than placing of other issuers’ products— which is not eligible instead. Again, one might expect the opposite, due to obvious inherent conflicts of interests. More generally, the scope of the exemption may be significant if one considers that a service is provided at the initiative of the client even if the client demands it on the basis of a promotion or an offer ‘that by its very nature is general and addressed to the public’ (Recital 85). In the light of this, a selling strategy where firms deliver informative brochures concerning their own products to clients after being asked whether they have some interesting securities on their shelves would seem to be compatible with the execution-only regime.

[4.66] In addition, and moving to the the third source of regulatory uncertainty, how much the new definition of order execution and the exemption regime will shape investment firms’ behaviour in the context of self-placement will also depend, in practice, on whether self-placement will be accompanied by personal recommendations. When this is the case, the relationship between investment firms and their clients switches from order execution to investment advice, and no exemption for execution-only services applies. But while differentiating investor-specific recommendations and pure sales is easy on the books, we have already seen how old and recent mis-selling scandals show that primary markets are often characterized by strong selling pressure concerning firms’ own products, so that investors cannot always tell advice from sales.99 Fee-free advice, which often accompanies self-placement, remains a key feature of many Member States’ financial markets.100 This persistent dominance may easily drift into mis-selling practices when generic recommendations not qualifying as investment advice are dropped to clients, thus increasing selling pressure outside the context of suitability-based protective rules. [4.67] The complex interaction between self-placement and the best interest duty justified an ad hoc joint statement by the Joint Committee of the three ESAs in July 2014, which echoes concerns that financial instruments qualifying as regulatory capital for prudential regulation under CRD/CRR IV and BRRD might be abusively sold to unaware investors even if inappropriate or unsuitable.101 The Joint Committee statement provides some further guidance on a sensitive area, for instance by restating that the suitability test should not focus on financial instruments in isolation but rather consider the importance of a balanced portfolio composition, as risk diversification represents a crucial issue in investor protection.102 Unfortunately, the statement falls short of delivering clear indications as to when self-placement encroaches upon investment advice, thus triggering the suitability test, and leaves questions on the extent of the best interest duty in this setting, as well as the ensuing litigation risk, unresolved. Against this backdrop, Article 41(4) Delegated Regulation relies on enhanced disclosure to protect investors, as it requires banks and investment firms offering own financial instruments that are included in the calculation of prudential requirements to clarify the differences between those instruments and bank deposits.

[4.68] Finally, how protective the new regime for self-placement will be also depends on the conflict-of-interests regime, which remains applicable both to execution-only services and to transactions with eligible counterparties and cannot therefore be escaped in the context of selfplacement (Articles 25(4)(d) and 30(1) MiFID II).103 In its consultation paper on the technical advice to be delivered to the Commission in view of the adoption of MiFID II Level 2 measures, ESMA stressed that selfplacement inevitably entails conflicts of interests for banks, and suggested that procedures adopted for their identification and management ‘may include consideration of refraining from engaging in the activity, where conflicts of interest cannot be appropriately managed so as to prevent any adverse effects on clients’.104 Upon the suggestion of the Securities and Markets Stakeholders Group, the requirement was strengthened in ESMA’s final technical advice, subsequently reflected by Article 41(2) Delegated Regulation, according to which investment firms’ procedures on conflicts of interest ‘must’ include the option of refraining from engaging in selfplacement.105 [4.69] How should these rules apply within the context of the executiononly regime, where firms operate as mere order-takers? A possible answer is that, if the issuing investment firm is simply informing its potential clients that its products are available for sale, rules on conflicts of interests still apply, but the firm should not be expected to refrain from selling the product as long as the client’s decision to purchase is taken autonomously (and is therefore not influenced by the firm’s conflicting interest).106

5. Dealing on Own Account [4.70] Extending fiduciary-style duties to dealing on own account is obviously an uncomfortable task. When firms are operating as principals, incentives to sell financial instruments they own at the best possible price are inherent to the very essence of the activity they perform, and charging them with protection of client’s best interest is problematic.107 In this case, the fiduciary duty to put the ‘best interest’ of clients first can hardly be

reconciled with the fiduciary duty that trading desk employees owe to their principals.108 [4.71] These contradictions have initially induced some national competent authorities to interpret MiFID I in the sense that its conduct of business rules do not apply when investment firms simply provide quotes and then deal with investors on the basis of such information: this would qualify as investment activity and not as investments service because no client (or no client order) exists.109 However, the same Commission opinion110 that ruled out the possibility to confine conduct of business rules to the provision of investment services to clients, with the exclusion of investment activities with mere contractual counterparties, militates against disapplication of MiFID II rules to provision of quotes (either continuously or in response to a request). In a similar vein, an ESMA Q&A document on CFDs111 stresses that conflicts of interest relevant for the MiFID I framework may also arise when investment firms execute orders on such financial instruments on the basis of quotes they provide. As a consequence, dealing on own account is always subject to conduct of business rules, including the client’s best interest rule.112 Some qualification should however be made for investment firms’ obligation to execute orders on terms most favourable to the client (best execution: Article 21 MiFID I and Article 27 MiFID II).113 Recital 103 Delegated Regulation (previously Recital 69 Directive 2006/73/EC) makes clear that best execution also applies when client orders are executed against proprietary capital,114 but it also points out that if a client simply accepts a quote provided by an investment firm that complies with the best execution duty, then the firm does not have to take further steps when executing the quote, unless this has become manifestly out of date under the new market conditions. Furthermore, MiFID II Recital 91 (former MiFID I Recital 33) specifies that best execution applies ‘where a firm owes contractual or agency obligations to the client’. The Commission has interpreted this statement in the sense that, when orders are executed against proprietary capital, investment firms are subject to the obligation to execute orders on the terms most favourable to the client only if they are operating ‘on behalf of the client’, as is the case when the decision to deal

on own account is taken by investment firms themselves, other alternatives being available.115 This narrow scope is accompanied by a presumption that best execution applies when the counterparties are retail clients, as these are more likely to rely on the firm to act in their best interests.116 However, this may be a weak safeguard when the relevant financial instruments are illiquid and the firm is the only available counterparty.117 To avoid regulatory loopholes, some scholars have submitted that in such cases the rule should be interpreted as mandating the adoption of a ‘fair price’ rule.118 The effect of this interpretation echoes the US FINRA Rule 2121 summarized above.119 [4.72] MiFID II appears also to be aware that some limitations to best execution are necessary in the context of proprietary trading. MiFID II Recital 93 specifies that the obligation to execute orders on terms most favourable to the client involve no duty to compare own execution policies and commissions or fees with those of other firms. As a consequence, investor protection in this scenario mainly relies on disclosure of execution policies under Article 25(4) MiFID II.

VI. A Look Across the Atlantic [4.73] Concerns on the extension of fiduciary-style duties to transactional relationships are not confined to the EU. In the US, a wide echo was triggered by the ABACUS case, where enforcement actions were brought for the violation of (trust-based) anti-fraud provisions in the context of a securitization transaction involving professional counterparties.120 [4.74] Conduct of business standards applicable to broker-dealers have also been troubling the US policy and scholarly debate in recent years. Here, the absence of clear-cut guiding criteria on the boundaries of incidental advice has paved the way to a broad interpretation of fiduciarystyle duties.121 In the US, broker-dealers may give advice without falling into the (more stringent) fiduciary standards set forth by the Investment Advisers Act (1940) if such advice is ‘solely incidental to the conduct of [their] business as a broker or a dealer’ and is not remunerated with any

‘special compensation’ (15 USC 80b-2(a)(11)(C)).122 However, an express or implied representation of fair dealing by brokers triggers application of Rule 10b-5, so that misrepresentation of the risk involved is regarded as a violation of the duty of fair dealing grounded on the ban on securities fraud (the so-called shingle theory).123 Although the theory is mostly invoked to enforce fair commercial practices in purely brokerage or dealing services,124 it also applies to investment recommendations made by brokerdealers on the basis that clients may legitimately rely on the broker-dealer’s implied representation that it had good reasons to believe the recommended investment was suitable.125 [4.75] The troublesome task of identifying the boundaries between pure salesmen’s activities and advice under EU law echoes the attempts to clarify similar questions in the US based on the shingle theory. In a similar vein, uncertainties in the interpretation of the MiFID II duty to act in accordance with the client’s best interest mirror the US debate on the applicability of fiduciary (or fiduciary-like) obligations to firms providing brokerage or dealing activities. The Dodd–Frank Act granted the SEC rulemaking powers on the broker-dealers’ duty of care, to be exercised after conducting an ad hoc study (Section 913). At the time of writing, the SEC has published a proposal (‘the 2011 SEC proposal’) to extend advisory-style fiduciary duties to broker-dealers providing advice,126 but no such regulation has been adopted to date. [4.76] Whether a broadened scope of fiduciary standards would actually strengthen protection of US investors vis-à-vis broker-dealers is debated, the doubts partially depending on uncertainties concerning the current scope of the duty of fair dealing for these firms.127 For instance, the duty of loyalty is normally more intense for advisers as per Section 206 of the Investment Advisers Act, as interpreted by the US Supreme Court,128 which entails a duty to eliminate or, if this is not possible, to declare any conflict of interest that can taint the provision of advice. The duty of care is better specified for broker-dealers when they provide implicit129 or explicit recommendations on a transaction or an investment strategy, as in these circumstances a suitability rule applies, as well as ‘know your customer’ and ‘know your security’ rules (FINRA Rule 2111).130

[4.77] Some of the arguments raised by US scholars and practitioners in favour of or against the 2011 SEC proposal are worth recalling because they may shed light on the complex role of fiduciary-style duties when these mandatorily apply to at arm’s-length services. For instance, an agency relationship only exists when the broker is representing its client while purchasing securities, rather than matching the order on a principal basis, hence performing an activity that goes beyond mere selling and involves professional skills in finding the desired investment; to the contrary, no such relationship can be construed when the firm is operating as a counterparty.131 In the latter scenario, stronger investor protection requires, normally, a finding of fraud. That finding, in turn, can be based either on the breach of the duties stemming from an implicit fiduciary relationship or on the shingle theory.132 However, both solutions are prone to high legal uncertainty, as no consolidated criteria can be identified in the relevant case law.133 [4.78] Against this backdrop, and in line with the nature of these activities, the SEC 2011 proposal does not purport to extend fiduciary duties to brokerage and dealing services which are provided without (explicit or implied) recommendations.134 By the same token, US courts may qualify brokers as fiduciaries on appropriate facts, such as when they act as agents while pursuing their customer’s mandate according to their discretion, but refrain from doing so when brokers are acting as mere ordertakers.135 [4.79] For similar reasons, no duty exists upon brokers to offer the best available securities, that is, the ‘most suitable ones’,136 even if they know that other cheaper products are available for sale on the market.137 Section 15(k) Securities Exchange Act—introduced by the Dodd–Frank Act Section 913(g)—enables the SEC to mandate that brokers and dealers, as well as investment advisers, act in the client’s best interest without regard to their financial or other interests, but only insofar as such broker-dealers are providing personalized investment advice about securities (either to a retail customer or to any other customers as the Commission may provide by rule).138 To the contrary, the same provision excludes that brokers and dealers have a continuing duty of care or loyalty to the customer after

providing personalized investment advice about securities. Most importantly, mere trading of proprietary products does not, as per an explicit provision, violate fiduciary standards the SEC may impose, if customers are duly informed of such limitation.139 [4.80] Although the provisions above restrict SEC’s rulemaking powers, the scope of the study commissioned by Section 913(b) of the Dodd–Frank Act is broader, as the Commission is enabled to formulate any other consideration it may deem necessary and appropriate.140 Fears were therefore voiced that the study could be a prelude to possible further additional legislation by Congress extending fiduciary duties to pure brokerage and dealing services.141 Among the drawbacks of this possible regime, commentators often mention the risk of market distortions. Increased compliance costs—so the opinion goes—may oust retail customers from the market for financial services,142 so that the reform may harm the group of investors it is supposed to protect.143 We do not purport to assess whether these fears are grounded, especially because they are partially based on the expectation that commission-based remuneration (i.e., inducements) may be restricted in the future,144 a measure not recommended by the SEC 2011 proposal. In the EU, indeed, commissionbased advice is also allowed.145 Although associated with reduced quality of recommendations,146 commission-based remuneration is also considered a way to ensure investors have cheap access to investment advice.147 [4.81] However, concerns for increased costs for retail investors as a consequence of tightened suitability standards have also been voiced in more general terms,148 and particular attention is devoted in the US debate to the duty to recommend the best suitable product in all circumstances.149

VII. Assessment [4.82] MiFID II confirms, in line with its predecessors, that investment firms shall act in accordance with their clients’ best interest. This overarching fiduciary-style duty is directly applicable to investment firms,

which may be held liable if they do not comply with this principle in their professional conduct. In the preceding sections, we have highlighted how the best interest rule set forth by the MiFID II regime applies across the board, encompassing situations where such a duty does not easily fit with the nature of the activities performed.150 Fiduciary-style duties are a wellgrounded regulatory tool to reduce the risk of opportunistic behaviour in long-term relationships, such as portfolio management and investment advice, characterized by information asymmetries, as they curb agents’ incentives to take advantage of their principals’ reduced ability to monitor agents’ performance.151 To the contrary, the fiduciary paradigm is hard to reconcile with the very nature of arm’s-length services. The ensuing divergence in the interpretation of Article 24(1) duties results in reduced legal certainty for investment firms and investors alike. [4.83] As we have seen, the best interest duty fits some investment firms’ activities so little that, under the current regime as interpreted by ESMA, some of these activities cannot be performed at all.152 In other words, the best interest rule acts as an indirect ban on some conduct. Whether such conduct is only high-risk from the clients’ perspective or is inherently abusive is key in understanding the impact of the overarching best interest duty on the market for investment services. The MiFID II/MiFIR system provides for a special product intervention regime with a view to tackling significant investor-protection concerns, but the best interest duty may achieve the same purpose by virtue of its broad scope and its inherent flexibility. [4.84] Yet, product intervention is subject to specific procedural requirements that ESMA and national competent authorities have to comply with in order to restrict or prohibit the sale of certain financial instruments or the performance of certain financial activities. For instance, national competent authorities have to notify all other fellow authorities—after consulting them if these are likely to be affected by the measure—one month (or twenty-four hours in the case of urgency) in advance of the planned restriction or prohibition. ESMA must also be notified, so that it can release a non-binding opinion on whether the measure is proportionate and non-discriminatory. Furthermore, evidence shall be provided that the conditions for the restriction or prohibition measure are met, including that

existing regulatory requirement under Union law would not suffice to address the relevant regulatory concerns (Article 42 MiFIR). For measures taken by ESMA, Article 9(5) Regulation (EU) No 1095/2010 applies, so that restrictions and prohibitions must be reviewed at least every three months and expire otherwise (Article 40 MiFIR). [4.85] In light of these burdensome procedural rules, supervisors may be inclined to address investor-protection concerns by exploiting the catch-all provision on clients’ best interests. Stretching the best interest duty can indeed curb excessive risk-taking by retail clients even in the absence of a clear abuse by investment firms acting as mere contractual counterparties. Whether this is the best tool to implement a paternalistic supervisory policy agenda—which we do not question here—is, however, debatable. Some overlap between strict supervision and the straight ban of practices that jeopardize investor protection is probably unavoidable, as demonstrated by the requirement that product intervention measures may only be taken, at national and EU level, insofar as the enforcement of existing requirements would not adequately address regulatory concerns such as a threat to investor protection (Articles 40(2)(c) and 42(2)(b) MiFIR). [4.86] Take, for example, the ESMA Q&A document on CFDs.153 ESMA holds that unhedged execution of retail client orders concerning certain complex financial instruments would inevitably breach the best interest duty and should therefore be avoided. The conclusion is based on the economic incentives in arm’s-length transactions, as any increase in the investment firms’ returns is mirrored by a corresponding disadvantage for their clients. However, the complexity of the product—combined with asymmetric information vis-à-vis retail clients—seems to be the actual determinant of the threat to investor protection in this case. The Q&A document therefore relies on the uneasy relationship between the best interest duty and purely transactional relationships to protect retail investors from the risk of being mis-sold high-risk products. A regulatory strategy such as that makes the boundaries of dealers’ legitimate market practices highly uncertain. Contractual counterparties’ interests naturally diverge in any transaction carried out on a principal basis, and situations where investment firms’ and clients’ profits are inversely correlated are not exclusive to complex derivative instruments or to retail clients. Hence,

market participants may not be in a condition to assess when arm’s-length transactions violate the best interest duty and when they do not. [4.87] A sceptical observer may stress that the expansion of quasifiduciary duties to cover arm’s-length services tilts towards indulging investors—strikingly, no matter whether retail or professional, the only exception being eligible counterparties—at the risk of going along with misguided or biased perceptions of an investment firm’s role. Paving the way to redress for breach of fiduciary-style duties when no portfolio or investment advice relationship is established entails the risk of facilitating investors’ actions to obtain restoration for wrong investments even in the absence of any investment firm’s wrongdoing.154 Such a strategy may lure policymakers because investors and the public at large have full perception of individual investors’ losses triggered by major defaults or resolutions, while the increased costs for all investors under the form of higher commissions stemming from undeserved redress normally go unnoticed. [4.88] At the same time, broad principles setting fiduciary-style duties are a key tool for effective enforcement in critical situations where detailed rules of conduct leave loopholes that investment firms can exploit to the detriment of their clients. MiFID II rules of conduct are not immune to this risk, as the previous analysis demonstrates. Broad principles such as the best interest duty and its companion standards play an important role in ensuring a reasonable regulatory equilibrium. To avoid the risk that invoking general standards may lead to over-enforcement, fiduciary-style obligations need to be interpreted in a flexible manner,155 and in the light of the nature of the service provided.156 While increased protection of clients’ best interest makes perfect sense when a fiduciary relationship between clients and investment firms combines with personalized recommendations of firms’ own products, purely transactional relationships taking place in the context of own-account dealing do not warrant such intense protections. In this second scenario, the best interest duty is sometimes implemented, at national level, as a requirement that investment firms pay due regard to their clients’ interests, rather than focusing exclusively on them.157 The MiFID II principle that firms shall act ‘in accordance with’ the client’s best interest would seem flexible enough to be compatible with such

interpretation, when the very nature of the relationship between firms and their clients has no fiduciary component. [4.89] Besides being directly applicable, the best interest duty is further specified by more detailed provisions within the body of MiFID II, including rules concerning the management of conflicts of interest and inducements, best execution, and suitability or appropriateness of financial instruments and investment services. The combination of detailed provisions setting out conduct of business rules and of more general standards of behaviour has many advantages, as it can ensure that firms are subject to a regime that is clear and, at the same time, flexible enough to address ex post shortcomings that the lawmaker cannot possibly predict because of limited rationality in forecasting future states of the world.158 However, this regulatory technique requires careful coordination between rules and standards at the enforcement stage, in order to avoid conflicts and suboptimal outcomes.159 [4.90] A crucial factor to make sure that the duet between rules and standards is harmonious is regulatory consistency. MiFID II, just like its predecessor,160 falls short of clarifying how the two regulatory layers interact. This lack of clarity seems to be the result of an unresolved tension between two different approaches. The first approach reflects the idea, which has gained traction in recent decades,161 that parties should be able to deviate from general standards of conduct, as private-ordering solutions are able to reduce or to better address contractual incompleteness.162 The second approach relies instead more heavily on mandatory rules, on the assumption that deviations from the law may deliver undesirable results both for individual investors and in terms of social welfare. [4.91] Inherent to the first approach is the qualification of conduct of business rules as default provisions whose compliance costs may not always be justified. When compliance costs exceed the corresponding benefits, opting out of those rules may occasionally produce a net benefit that may accrue to one party only or be allocated through bargaining. Clearly representative of this philosophy are MiFID II rules on investment suitability and appropriateness, as they modulate the intensity of investor protection on the basis of the service provided. 163 Relevant elements of

mandatory law—attributable to the second regulatory approach—remain to maintain higher protection in favour of a core subset of ‘weaker’ retail investors that cannot be classified as professional even upon request:164 for them, the risk that investment firms will take advantage of investors’ overconfidence (false negatives) is considered higher than the risk of imposing an excessive regulatory burden in some cases (false positives).165 However, MiFID II allows166 contracting parties, including weaker retail clients, to deviate from some core specific conduct of business rules when investment firms are operating within the execution-only regime, that is, as mere order-takers to the benefit of customers that are only looking for execution services at a low cost. Here, the first regulatory approach prevails and enables significant deviations from core specific conduct of business rules: among these, only conflict-of-interest provisions remain in place for execution-only services (Article 25(4)(d) MiFID II). More precisely, disclosure is the principal—though not exclusive—safeguard addressing possible organizational and administrative ineffectiveness in managing alien interests.167 In such circumstances, the duty to take all reasonable steps to obtain the best possible result in the execution of clients’ orders (Article 27 MiFID II) buttresses investment firms’ duty of care in the provision of the trading service, while no explicit obligation exists to ensure that the traded financial instruments are appropriate for the ordering customer. [4.92] Therefore, with the qualifications outlined above, specific conduct of business rules are optional, the parties being allowed to tailor the level of protection for the client. To the contrary, the second regulatory approach seems to prevail for more general standards. As a consequence, the overarching duty to pursue the client’s best interest, on one hand, and specific conduct of business rules, on the other, are the expression of divergent and potentially conflicting approaches.168 The general standards set forth in Article 24(1) MiFID II are indeed applicable even in cases where some of their main specifications, such as the suitability/appropriateness test, are waived.169 This second regulatory approach reflects the understanding that mandatory standards of behaviour aimed at protecting investors are needed to curb the risk that individual

choices impair market integrity (one of MiFID II’s underlying regulatory objectives).170 [4.93] Yet, once again, uncertainties remain. In particular, it is unclear how firms can comply with the general principle to act in the client’s best interest when they are explicitly exempted from applying rules that are best understood as a specification thereof.171 Investment firms performing execution-only sales services, and thus enabled not to collect information on customers’ features, cannot of course pursue their clients’ best interests in the same way as when such information is available to them.172

1

See also ECJ, C-604/11, Genil, 30 May 2013, § 57 (principles of equivalence and effectiveness bind Member States in the determination of private-law consequences of MiFID I violations). 2 J. Benjamin, Financial Law (Oxford: OUP, 2007), 556–7. 3 ibid 558. 4 The ‘sole interest’ standard prevents transactions that may be in the clients’ interests, as well, as long as the fiduciary also takes a benefit out of them and irrespective of any detriment to the beneficiary; this is not the case with the ‘best interest’ standard, where clients may not reject transactions that satisfied their needs in the best possible manner; see T. Frankel, Fiduciary Law (Oxford: OUP, 2011) 149–52. 5 L. D. Smith, ‘Can We Be Obliged to Be Selfless?’ in A. S. Gold and P. B. Miller (eds) Philosophical Foundations of Fiduciary Law (Oxford: OUP, 2014), pp. 143–58 (fiduciary duty is a requirement to exercise judgement in what the fiduciary perceives to be the best interests of the beneficiaries). 6 A. Hudson, The Law of Finance (2nd edn, London: Sweet & Maxwell, 2013), pp. 299– 300. 7 J. Black, Rules and Regulators (Oxford: OUP, 1997), pp. 140–3, 146, 149, 152. 8 SIB, Core Conduct of Business Rules, Core Rule 17. 9 N. Moloney, How to Protect Investors. Lessons from the EC and the UK (Cambridge: CUP, 2010), p. 268. 10 Black (n. 7), 176. 11 SIB, Core Conduct of Business Rules, Core Rule 21. 12 J.-P. Casey and K. Lannoo, The MiFID Revolution (Cambridge: CUP, 2009), pp. 58– 60. 13 For a description see Moloney (n. 9), 268; G. McMeel, ‘Agency and the Retail Distribution of Financial Products’ in D. Busch et al. (eds) Agency Law in Commercial

Practice (Oxford: OUP, 2016), pp. 182–3. 14 See n. 157 and accompanying text. 15 UK FCA Principles for Business, Principle 6. 16 G. Ferrarini, ‘Towards a European Law of Investment Services and Institutions’ (1994) Common Market Law Review 31, 1283, 1304; M. Tison, ‘Conduct of Business Rules and their Implementation in the EU Member States’ in G. Ferrarini et al. (eds) Capital Markets in the Age of the Euro (Oxford: OUP, 2002), p. 68. 17 IOSCO, International Conduct of Business Principles (1990) Principle 2. The 2003 IOSCO Objectives and Principles of Securities Regulation (at 36) specified that firms should ‘observe high standards of integrity and fair dealing and should act with due care and diligence in the best interests of [their] customers’ in the comment to Principle 12, which required market intermediaries to comply with standards for operational conduct that aim to protect the interests of clients; see now Principle 31 IOSCO Objectives and Principles of Securities Regulation (2010). 18 Furthermore, the same Recital 71, second para., refers to the duty to act ‘in the interest of the client’ as a perfect substitute of the duty to act ‘in accordance with’ the same interest referred to in the first para. 19 Casey and Lannoo (n. 12), 46. 20 Eligible counterparties may ask to be treated according to the ordinary investor protection regime (Article 30(2)). 21 Of course the general Article 24(1) duties and their corollaries will continue to apply between members and participants of trading venues on one hand, and their respective clients whose orders are transmitted to the trading venue on the other (Articles 19(4) and 53(4) MiFID II). Note also that Article 24(1) applies, instead, to organized trading facilities (OTFs), which match orders on a discretionary basis (Article 20(6)). 22 German law has also traditionally mandated agents in a contract concluded on a commission basis (Kommissionvertrag) to protect the principal’s interest and to deliver him any asset obtained in the execution of the business commissioned (P. Mülbert, ‘The Eclipse of Contract Law in the Investment Firm–Client Relationship: The Impact of the MiFID on the Law of Contract from a German Perspective’ in G. Ferrarini and E. Wymeersch (eds) Investor Protection in Europe (Oxford: OUP, 2006), p. 300). 23 Benjamin (n. 2), 568, 573 (fairness to consumers does not originate in English case law and is inherently unclear and historically alien to English law). 24 With specific reference to conflict of interests rules, see L. Enriques, ‘Conflicts of Interests in Investment Services: The Price and Uncertain Impact of MiFID’s Regulatory Framework’ in G. Ferrarini and E. Wymeersch (eds) (n. 22), 326 (complying with the specific conflict-of-interest rules will never be enough for firms, as these will always have to ensure that their clients are treated in accordance with their best interests). 25 Moloney (n. 9), 218. 26 See e.g., for Italy, F. Parrella, ‘Consulenza in materia di investimenti’ in R. D’Apice (ed.) L’attuazione della MiFID in Italia (Bologna: Il Mulino, 2010), p. 191 (no specific

rule requires investment firms to recommend the most suitable product). 27 See G. Rubin, ‘Advisers and the Fiduciary Duty Debate’ (2015) Business and Society Review 120, 519–20. 28 For instance, Consob requires investment firms to compare different financial instruments and to recommend the best available option when advice is spontaneously given with a view to dissuading investors from placing an order (see Consob Comm. No 9019104 (2 March 2009)). 29 See FSA, Reforming Conduct of Business Regulation (Consultation Paper 06/19) (2006), 77 (an explicit obligation for advisers to recommend the ‘most suitable’ packaged product from the prescribed range on which they advise would go beyond MiFID I, but the combination of other MiFID I rules, including the obligation to act in the best interests of the client, and the suitability test, reaches the same outcome); this view was subsequently confirmed in FSA, Reforming Conduct of Business Regulation (Policy Statement 07/06) (2007), 50. 30 Benjamin (n. 2), 579 (arm’s-length approach ensures effectiveness of risk-transfer function of financial markets, but it might not be viable when retail investors face hardship in bear markets). 31 Moloney (n. 9), 199, 256–63. 32 Pre-packaged financial products are widespread among investors in the UK (ibid, 210). 33 ESMA’s Committee for Economic and Markets Analysis (CEMA), Economic Report: Retailisation in the EU (ESMA/2013/326) (2013) (performance of structured products with 100% capital protection is relatively low when compared to risk-free investments). 34 ESMA, MiFID practices for firms selling complex products (ESMA/2014/146) (2014) 2. 35 ibid 5. 36 Consob, Comm. No 97966 (22 December 2014). 37 The expression ‘regulatory contract law’ describes the increasingly pervasive role of lawmakers and authorities in defining mandatory contractual provisions and in ensuring their enforcement when contractual remedies prove insufficient (J. Köndgen, ‘Policy Responses to Credit Crisis: Does the Law of Contract Provide an Answer?’ in S. Grundmann and Y. Atamer (eds) Financial Services, Financial Crisis and General European Contract Law. Failure and Challenges of Contracting (Alphen aan den Rijn: Kluwer Law International, 2011), pp. 39–40). 38 M. Kruithof, ‘A Differentiated Approach to Client Protection: The Example of MiFID’ in Grundmann and Atamer (eds) (n. 37), 147 f. 39 Smith (n. 5), 148 (advisory services, just like portfolio management, belong to fiduciary relationships). 40 Hudson (n. 6), 299. 41 S. M. Davidoff Solomon et al., ‘The SEC v Goldman Sachs: Reputation, Trust, and Fiduciary Duties in Investment Banking’ (2012) Journal of Corporation Law 37, 529,

550–1 (explaining the reasons why mandating fiduciary duties in the context of arm’slength transactions may be inefficient and submitting that this may hold true, to some extent, even when retail investors are involved). 42 Black (n. 7), 141; A. M. Pacces, ‘Financial Intermediation in the Securities Markets: Law and Economics of Conduct of Business Regulation’ (2000) International Review of Law & Economics 20, 478, 484. 43 Articles 43 ff. Directive 2014/59/UE (BRRD); Article 27 Reg. (EU) No 806/2014 (SRM). 44 This is typically the case with Italy, see e.g. ESME, Non-Equity Market Transparency (2007), 10 (bonds comprise 22.4% of total financial assets for Italian retail investors— compared with 1.5% in the UK and 6.9% in the US—and are often held directly). Other countries with significant direct participation in bond markets are Germany and Belgium, while figures are lower for France: CEPR, European Corporate Bond Markets: transparency, liquidity, efficiency (2006), 32. 45 ESMA (n. 34), 3. More explicitly ESMA, Guidelines on complex debt instruments and structured deposits (ESMA/2015/1783) (2015), 21, 32. 46 Consob, Comm. No 97996/14 (22 December 2014) (subordinated debt not listed among complex financial products). 47 See ECJ Genil (n. 1), §§ 51–5. 48 See EU Commission, ‘Investment advice’—meaning of ‘personal recommendation’, Your Questions on Legislation, Question 158 (p. 335). 49 The question remains whether firms may be required, under the general Article 24(1) duties, to warn investors if non-personalized advice is given (Moloney (n. 9), 204). 50 Recently ESMA, MiFID Suitability Requirements. Peer Review Report (ESMA/2016/584) (2016) 7–8 (national competent authorities are often trying to clarify the boundary between information and advice); see also McMeel (n. 13), 184 (reporting UK case law were banks’ liability has been excluded on the basis that they did not cross the dividing line between giving information on, and selling, a product, on one hand, and the activity of giving advice on the other). 51 For instance, in the UK investment advice not involving personal recommendations is still subject to the best interest duty in COBS Rule 2.1.1R (FCA, Retail Investment Advice. Finalized Guidance (FG15/1) (2015) 11), but this apparently reduces legal certainty for investment firms. 52 See FCA, Financial Advice Market Review. Final report (2016), 28–32 (considering a MiFID-like definition of advice as preferable to that provided by Article 53 Financial Services and Markets Act 2000 (Regulated Activities) Order 2001, which includes generic advice, and suggesting adopting MiFID I definition to reduce legal risk. Remarkably, the analysis does not take into account the effects of Recital 81–83 Directive 2006/73/EC on the scope of application of investment advice). 53 ibid.

54

FCA (n. 52), 7; see also Chapter 6 in this volume [p. 3 working paper] (households normally expect advice to be provided for free). 55 FCA COBS 6.1A.4. See EU Commission, MiFID Transposition state of play, Notification 6, at (reporting on the reasons for ‘banning retail investment product providers from offering predetermined amounts of commission to UK adviser firms, in relation to the recommendation of their products’); see also Moloney (n. 9), 210–11. 56 FSA, Distribution of Retail Investments: Delivering the RDR (CP 09/18) (2009), 19. 57 Davidoff Solomon et al. (n. 41), 542. 58 Consob, Prime linee di indirizzo in tema di consulenza in materia di investimenti [Preliminary guidelines on financial advice] (2007) 4; Consob, Comm. No 9019104 (n. 28), 8–9. 59 Smith (n. 5), 150. 60 R. H. Sitkoff, ‘The Economic Structure of Fiduciary Law’ (2011) Boston University Law Review 91, 1039, 1041 (incentive-based compensation arrangements normally used to reduce—although not eliminate—agency problems). 61 SEC, Study on Investment Advisers and Broker-Dealers (2011), 10–11. 62 In fiduciary relationships, misappropriation infringes the duty of loyalty (Sitkoff (n. 60), 1043). 63 Sitkoff (n. 60), 1048; J. Getzler, ‘Financial Crisis and the Decline of Fiduciary Law’ in N. Morris and D. Vines (eds) Capital Failure (Oxford: OUP, 2014), p. 199. 64 J. Campbell et al., ‘The Regulation of Consumer Financial Products: An Introductory Essay with Four Case Studies’ HKS Faculty Research Working Paper Series RWP10-040 (2010) 16; T. L. Hazen, ‘Are Existing Stock Broker Standards Sufficient?: Principles, Rules and Fiduciary Duties’ (2010) Columbia Business Law Review 709, 731. Lack of disclosure of excessive mark-ups is also considered a violation of antifraud provisions: L. Loss et al, Fundamentals of Securities Regulation (Alphen aan den Rijn: Wolters Kluwer, 2011), pp. 1420–31. See also A. F. Tuch, ‘Conduct of Business regulation’ in N. Moloney, E. Ferran, and J. Payne (eds) Handbook of Financial Regulation (Oxford: OUP, 2015), p. 551 (FINRA ‘catch-all’ Rule 2010 allows policing maintenance of ‘high standards of commercial honor’ and ‘just and equitable principles of trade’ by broker-dealer); J. J. Park, ‘The Competing Paradigms of Securities Regulation’ (2007) Duke Law Journal 57, 625, 646, 649 (FINRA Rule 2010—former NASD Rule 2110—provided the legal basis for principle-based enforcement against kickbacks in the IPO allocation process). 65 See FINRA Rule 2121. Fair Prices and Commissions (dealers shall buy or sell at a price which is fair, taking into consideration all relevant circumstances; a similar provision applies to commissions charged by brokers. Commissions exceeding 5% of market prices are in principle regarded as excessive, but in the case of low-priced securities a slightly higher percentage may be justified). See also Lehl v SEC, 90 F.3d 1483, 1488 (5% benchmark calculated by NASD, now FINRA, on the basis of practices adopted by the majority of broker-dealers). A similar provision also existed in the UK in the past but was

subsequently repealed. See G. McMeel, ‘Standards of Conduct for Investment Advice, Stockbroking, and Portfolio Management’ in G. McMeel and J. Virgo (eds) McMeel and Virgo on Financial Advice and Financial Products (Oxford: OUP, 2014) 333 (reporting that the FSA, which had the power to prevent investment firms from excessively charging their clients under the old COB Rule 5.6, had never used such power). 66 See ESMA, Questions and Answers relating to the provision of CFDs and other speculative products to retail investors under MiFID (ESMA/2016/590) (2016) 18, 20. 67 C. Kumpan and P. Leyens, ‘Conflicts of Interest of Financial Intermediaries’ (2008) European Company and Financial Law Review 5, 72, 79. See also Chapter 7 in this volume (Section§ A.I.1) (only situations where a position of influence is entrusted to the counterparty and accompanied by a consideration are relevant for MiFID I and II regime on conflicts of interest). 68 For further specifications see ESMA, Structured Retail Products: Good practices for product governance arrangements (ESMA/2014/332) (2014). 69 Article 16(3) requires identification of a target market of end clients and consideration of all risks relevant to such a target when manufacturing financial instruments; the distribution strategy shall be coherent with the target, too. Article 24(2) reinforces this duty by ensuring that manufacturing and distribution of financial instruments meet the target market’s needs. 70 Article 16(3) para. 7 MiFID II clarifies that rules on product manufacturing and distribution apply without prejudice to requirements concerning appropriateness. See also N. Moloney, EU Securities and Financial Markets Regulation (3rd edn, Oxford: OUP, 2014), p. 800. 71 ESMA (n. 34), 3. 72 For a summary of national measures see Moloney (n. 70), 825–6. 73 Consob, Comm. 97966, 22 December 2014, 5 (also establishing some specific procedural guarantees if such complex products are sold notwithstanding Consob’s recommendation. Among the complex products to which the recommendation applies are asset-backed securities (ABS), CoCo bonds, financial instruments qualifying as additional tier 1 capital as per Article 52 Reg. (EU) No. 575/2013 (CRR), credit linked notes, and derivatives not negotiated in a trading venue and not held for hedging purposes). 74 See Section VII. 75 For Italy see Consob, Comm. DAL/RM/96011036, 11 December 1996; Consob, Del. No DIN/58349, 29 July 2000; see also Consob, Del. No 18696, 12 November 2013 (Del. No DIN/58349 still applicable). In the UK, an appropriateness test is required to nonMiFID firms that arrange or deal in relation to certain financial instruments (such as nonlisted securities or derivatives) with retail clients, to the extent that the client order is in response to a promotion (FCA Conduct of Business Sourcebook (COBS) R.10.1.2; FSA (n. 29), 85). 76 D. Busch, ‘Agency and Principal Dealing under the Markets in Financial Instruments Directive’ in D. Busch et al. (eds) (n. 13), 142, 146, 151.

77

For other entities, an exemption could be found in Article 2(1)(d) MiFID I, which excluded from the Directive’s scope of application persons exclusively acting as dealers on own account. See now Article 2(1)(d)(iv) (restricting the same exemption to cases where dealing on own account is not performed when executing client orders; see n. 93 and accompanying text). 78 In Italy, banks and insurance companies are subject to MiFID I conduct of business rules when placing own financial products (Article 25-II Consolidated Law on Finance). 79 EU Commission, Credit institutions. Application of MiFID, Your Questions on Legislation, Question 230 (p. 420). 80 CESR, Responses to EU Commission Questions on MiFID Review (CESR/10-860) (2010) 3. 81 ibid 2–3. 82 ibid 2. 83 C. Comporti, ‘La sollecitazione all’investimento’ in A. Patroni Griffi et al. (eds) Intermediari finanziari, mercati e società quotate (Turin: Giappichelli, 1999), p. 551; R. F. Schiavelli, ‘Il contratto di collocamento’ in E. Gabrielli and R. Lener (eds) I contratti del mercato finanziario (Turin: UTET, 2004), pp. 1003–4. 84 EU Commission, Fund management. UCITS—characterisation of investment service, Your Questions on Legislation, Question 92 (p. 259); EU Commission, Fund management. Investment services—application to distribution of units in collective investment undertaking, Question 96 (p. 263). 85 ESMA, Consultation Paper. MiFID II/MiFIR (ESMA/2014/549) (2014) 77 (on ensuing conflicts of interests). For Italy see Consob, Comm. No 9019104 (n. 28), 3 (placing of financial products shall also be performed in the client’s interests). 86 CESR (n. 80), 2. 87 See Busch (n. 76), 172. 88 See Moloney (n. 70), 344, 799. 89 In the MiFID I framework, the focus by some rules on client orders was sometimes understood in the sense that such rules did not apply when client orders were matched by dealers’ own positions (CESR (n. 80), 3). 90 See Chapter 2 in this volume [p. 2 of the working paper]. 91 See Chapter 2 ibid [p. 3 of the working paper]. See also I. MacNeil, An Introduction to the Law on Financial Investment (2nd edn, Oxford: Hart Publishing, 2012), p. 463 (appropriateness test applies to arm’s-length transactions as well, although legal presumptions reduce the regulatory burden when investment firms are dealing with professional clients). 92 See MiFID II, Article 24, Sections (1), (2), (4), (5), (6), (9), (10), and (11) on general principles; Article 25, Sections (1), (2), (3), (4), and (7) on the appropriateness test; Article 29(3) on tied agents. 93 EU Commission, Answers to CESR scope issues under MiFID and the implementing directive (Working Document ESC-07-2007) (2007). See also P. Nelson, Capital Markets

Law and Compliance, The Implications of MiFID (Cambridge: CUP, 2008), p. 228–9 (FSA initially excluded application of MiFID I to ‘non-clients’, which was clearly not the intention of MiFID I, and later on tried to stress that in some circumstances client orders are missing, but the EU Commission ‘killed the argument’). See also R. Kent et al. ‘Conduct of Business’ in R. Fox and B. Kingsley (eds) A Practitioner’s Guide to the UK Financial Services Rulebook (London: Sweet & Maxwell, 2013), pp. 218–19 (EU Commission’s restrictive position on the possibility to exclude investor protection on the basis that no client relationship exists suggests erring on the side of caution when excluding that a counterparty qualifies as a ‘client’). 94 See also Kruithof (n. 38), 142–4 (in MiFID I regime, arm’s-length transactions between eligible counterparties are regarded as services even if this would not be the case in day-to-day language, so that such transactions are subject to MiFID I conduct of business rules unless these are explicitly made inapplicable). 95 At the same time, exemptions from application of conduct of business rules for dealing on own account with eligible counterparties can only make sense if such rules would otherwise be applicable (Article 30). See however Busch (n. 76), 151, 153 f., 159 (dealing on own account not performed vis-à-vis clients, hence not subject to client classification duties and to conduct of business rules—with the possible exception of fair dealing—including on conflict of interests). 96 See n. 93 and accompanying text. 97 A. Srivastava, ‘Conduct of Business Standards: Fair Dealing with Clients’ in J. Herbst (ed.) A Practitioner’s Guide to MiFID II (2nd edn, London: Sweet & Maxwell, 2015), p. 158. 98 EU Commission (n. 93), § 7. 99 See also Chapter 6 [p. 3 and 8 working paper] (distinguishing between actual advice and ‘cheap talk’ where advice is simply mimicked, possibly to promote own products, is difficult). 100 Moloney (n. 9), 259–61. 101 ESAs Joint Committee, Placement of financial instruments with depositors, retail investors and policy holders (‘Self placement’) (JC 2014 62) (2014). 102 ibid 8–9. 103 See Moloney (n. 70), 353. 104 ESMA (n. 85), 82–3 and 86. 105 ESMA (n. 50), 86 (‘may’ replaced with ‘must’). 106 This approach would be similar to that adopted by the EU Commission on best execution (n. 115 and accompanying text). 107 SEC (n. 61), (‘engaging in principal trading with customers or clients represents a clear conflict for any fiduciary’); A. B. Laby, ‘Reforming the Regulation of Broker-Dealers and Investment Advisers’ (2010) Business Lawyer 65, 395, 425 (profit-driven motive behind principal trading runs counter to traditional fiduciary notions); Busch (n. 76), 142.

108

See S. W. Simon, reported by Rubin (n. 27), 536: ‘That’s an inherent conflict of interests that no serious-minded person well-versed in these issues can ‘harmonize’ in any way’. See also SEC (n. 61) (acknowledging that principal trading brings to inextricable contradictions with the best interest duty, as ‘engaging in principal trading with customers or clients represents a clear conflict for any fiduciary’). 109 UK FSA, Implementing MiFID’s Client Classification Requirements (2006), 20–3. The UK FSA original position on best execution reflected its original interpretation on client classification: UK FSA, Reforming Conduct of Business Regulation (2006), 94. For further analysis see G. Ferrarini, ‘Best Execution and Competition Between Trading Venues: MiFID’s Likely Impact’ (2007) Capital Markets Law Journal 2, 404, 411–13. 110 See n. 93 and accompanying text. 111 ESMA (n. 66), 20. 112 EU Commission (n. 93), Issue 1, § 3. See also EU Commission, Review of the Markets in Financial Instruments Directive (MiFID), 8 December 2010, 64 f. 113 Namely, investment firms shall endeavour to obtain the best possible result for their clients taking into account, inter alia, price, costs, speed, likelihood of execution and settlement, size, and nature of the order. For retail customers, investment firms shall take into account the total consideration required for executing the order. 114 According to the Recital, ‘dealing on own account with clients by an investment firm should be considered as the execution of client orders, and therefore subject to […] obligations in relation to best execution’. 115 EU Commission (n. 93), Issue 1, § 6. For a more detailed analysis see Nelson (n. 93), 393 ff. 116 EU Commission (n. 93), Issue 1, § 9. 117 Consob Comm. No 9019104 (n. 28), 2–3. 118 Casey and Lannoo (n. 12), 72. 119 See n. 65 and accompanying text. 120 See Davidoff Solomon et al. (n. 41), 535–42. 121 D. C. Langevoort, ‘Brokers as Fiduciaries’ (2010) University of Pittsburgh Law Review 71, 439, 444. 122 Tuch (n. 64), 548. 123 A. Marcacci, ‘The US Suitability Rule and Fiduciary Duty and Their EU “StepSister”’ (2014–15) Consumer Finance Law Quarterly Report 68, 234, 236. 124 Loss et al. (n. 64), 1422–3. 125 R. Karmel, ‘Is the Shingle Theory Dead?’ (1995) Washington and Lee Law Review 52, 1271, at 1276. 126 SEC (n. 61), (proposing a ‘uniform fiduciary standard’ for advisers and brokerdealers providing personalized investment advice). 127 For a general description see Tuch (n. 64), 549–52. 128 SEC v Capital Gains Research Bureau, Inc., 375 US 180 (1963).

129

Implicit recommendations to hold securities are not covered by the definition of advice (FINRA Rule 2111 (Suitability) FAQ). 130 Relaxation is allowed when recommendations are addressed to institutional investors (FINRA Rule 2111(2)). 131 A. D. Madison, ‘Derivatives Regulation in the Context of the Shingle Theory’ (1999) Columbia Business Law Review 271, at 279: Langevoort (n. 121), 444 (pure sales of securities by brokerage firms might involve a high degree of trust, but this does not necessarily go along with a legal ‘fiduciarization’ of the relationship). 132 J. D. Cox et al., Securities Regulation. Cases and Materials (6th edn, New York: Wolters Kluwer, 2009) 1027–30. 133 ibid. See also Tuch (n. 64), 549 (‘no clear consensus exists as to when broker-dealers owe fiduciary duties’). 134 Langevoort (n. 121), 444 n. 21. 135 Hazen (n. 64), ,735–7, 741–3, 746–7 (with respect to non-discretionary accounts, the brokers’ duty is limited to proper execution of transactions. Upon completion of the trade, brokers acing upon a customer’s order do not have further duties to call upon their professional skills concerning the wisdom of customers’ trades). 136 See Rubin (n. 27), 536 (‘broker-dealers, unlike [investment advisers], have an obligation to serve their firm’s interest ahead of those of the customers’). 137 Langevoort (n. 121), 445. 138 Section 211(g) Investment Advisers Act (1940) sets a similar restriction. 139 See also Section 206(3) Investment Advisers Act (1940) (acting as principals for their own account is unlawful for investment advisers if they do not disclose in writing such capacity and obtain the client’s consent. This prohibition does not apply to brokerdealers that are not acting as an investment adviser in relation to the relevant transaction). 140 Furthermore, Section 913(f) of the Dodd–Frank Act, which also contains rulemaking provisions in favour of the SEC, is slightly broader in its scope (A. B. Laby, ‘Implementing Regulatory Harmonization at the SEC’ (2010–11) Rev. Banking & Fin. L. 30, 189, 193). The SEC is more likely to rely on Section 913(g) as this provides a more solid background to its rulemaking authority (ibid, 194). 141 Laby (n. 140), 190–1. 142 The matter is debated: compare e.g. M. S. Finke and T. S. Langdon, ‘The Impact of the Broker-Dealer Fiduciary Standard on Financial Advice’ (2012) Journal of Financial Planning 25, 28 (finding no evidence that heightened fiduciary duties would harm small investors) with Langevoort (n. 121), 445 (except in the true fee-based account, commissions and mark-ups pay for the advice customers receive when remuneration is not fee-based; threatening such revenue stream may freeze customers out of any but the lowest-cost brokerage services). 143 Confusion on the different rules applicable to broker-dealers and to advisers is one of the driving factors behind the SEC reform proposal (SEC (n. 61), 94).

144

Rubin (n. 27), 531 (reporting industry’s concerns that the SEC study may be the first step of a longer series of regulatory acts tightening rules applicable to broker-dealers, including those on remuneration). Commission-based remuneration is explicitly allowed by Section 15(k) Securities Exchange Act (commission-based remuneration shall not, in and of itself, represent a violation of fiduciary standards that might be made applicable by the SEC if broker-dealers provide such recommendations). 145 Yet restrictions on inducements may narrow down its scope in the future: see the concerns expressed by ESMA Securities and Markets Stakeholder Group, Investor Protection Aspects of the Consultation Paper on MiFID II and MiFIR (ESMA/2014/SMSG/035) (2014), 15–18. 146 Moloney (n. 9), 257–62; Tuch (n. 64), 553. 147 At the same time, some scholars notice that fee-based advice is no panacea: see Langevoort (n. 121), 449 (‘sufficiently gullible investors would end up paying too much in advisory fees, unless those, too, were regulated’). The fact remains, of course, that high fees are more observable than inducements and other indirect fees, no matter how transparent these are thanks to regulation. 148 This risk was also pointed out in a dissenting opinion by two SEC Commissioners: see K. L. Casey and T. A. Paredes, Statement Regarding Study on Investment Advisers and Broker-Dealers, 21 January 2011. 149 Rubin (n. 27). 150 For a different opinion see, however, Busch (n. 76) (considering purely transactional activities outside the scope of MiFID II investor-protection regime, but suggesting a more homogeneous regulation). 151 Sitkoff (n. 60), 1042–5. 152 This may be the case with complex products that cannot meet the best interests of their clients or the risk of which cannot be assessed for lack of sufficient information (n. 71 and accompanying text); similarly, investment firms selling CFDs or other complex derivative financial instruments would breach their (retail) clients’ best interest unless they hedge, at least in part, their exposure (n. 66 and accompanying text). 153 ESMA (n. 66). 154 Park (n. 64), 667–72; Benjamin (n. 2), 581–2 (widespread hardship for retail investors may trigger political desire to find financial institutions responsible on the basis of ex post losses rather than ex ante risks); Moloney (n. 9), 218 (risks of fairness standards are ‘considerable if the fairness obligation becomes a supervisory or judicial occasion for reflecting, ex post, wider societal discontent with the financial markets, particularly in times of market turbulence’). 155 Moloney (n. 9), 218 (reasonableness criteria needed to countervail excessive discretion in ex post review). 156 Cf. Sitkoff (n. 60), 1043–5 (principles should not be considered in isolation. Duty of care establishes a reasonableness standard that ‘is informed by industry norms and practices’).

157

Benjamin (n. 2), 558, 572, 579 (under UK FSA’s Principles for Business—Principle 6—firms ‘must pay due regard to the interests of customers, and treat them fairly, i.e. balance the interests of the clients with their own, rather than always and loyally promote the clients’ interests above their own’ (emphasis in original)). 158 Park (n. 64), 640 (rules are promulgated ex ante, while principles are defined ex post). Similarly Hazen (n. 64), 718 ff. 159 Benjamin (n. 2), 556–61, 580–3 (highlighting risk of uncertainty inherent to widespread reliance on equity and standards as opposed to rules and the common law). 160 Moloney (n. 9), 212. (wondering whether MiFID I regime maintains an appropriate balance between rules and principles). 161 Benjamin (n. 2), 559 (also referring to Kelly v Cooper [1993] AC 205, per Lord Browne-Wilkinson). See also H. N. Butler and L. E. Ribstein, ‘Opting Out of Fiduciary Duties: A Response to the Anti-Contractarians’ (1990) Washington Law Review 65, 1; R. Cooter and B. J. Freedman, ‘The Fiduciary Relationship: Its Economic Character and Legal Consequences’ (1991) New York University Law Review 66, 1045; F. H. Easterbrook and D. R. Fischel, ‘Contract and Fiduciary Duty’ (1993) Journal of Law & Economics 36, 425. 162 Underpinning this scholarly view is also the consideration that fiduciary duties are often regarded as a recessive feature in investment services and are increasingly deviated from by contractual parties: Getzler (n. 63), 201; Davidoff Solomon et al. (n. 41), 543–6. 163 G. Ferrarini, ‘Contract Standards and the Markets in Financial Instruments Directive: An Assessment of the Lamfalussy Regulatory Architecture’ (2005) European Review of Contract Law 1, 19, 26 (at least some conduct of business rules are inherently contractual and, hence, optional). 164 Moloney (n. 9), 209 (MiFID I conduct of business regime allows opt-out only for retail investors that can qualify as professionals, while vulnerable trusting investors remain protected). 165 Sitkoff (n. 60), 1047 (mandatory provisions addressing fiduciary duties have a protective and cautionary function). 166 A different problem is whether national law can allow contractual opt-out even in other circumstances. The question is normally answered in the negative, although with some uncertainties in some jurisdictions: see D. Busch, ‘Why MiFID Matters to Private Law: The Example of MiFID’s Impact on Asset Managers’ Civil Liability’ (2012) Capital Markets Law Journal 7, 386, 402. 167 Enriques (n. 24), 330–1. To be sure, under MiFID I it was clarified that disclosure of conflicts of interest provided no safe harbour with regard to the requirement that conflicts of interest be managed. See Recital 48 Delegated Regulation (previously Recital 27 Directive 2006/73/EC). 168 This troublesome interaction between contractarian and non-contractarian views on fiduciary duties also afflicts the US debate: Langevoort (n. 121), 443.

169

CESR (n. 80), 3 (‘even under the execution-only regime, firms must comply with all the applicable conduct of business rules (such as, e.g., the obligation under Art. 19(1) [now Art. 24(1) MiFID II] to act honestly, fairly, and professionally in accordance with the best interest of its [sic] clients)’). 170 M. Tison, ‘The Civil Law Effects of MiFID in a Comparative Law Perspective’ in S. Grundmann et al. (eds) Festschrift für Klaus J. Hopt (Berlin: De Gruyter, 2010), pp. 2612, 2633–4. 171 I. MacNeil, ‘Rethinking Conduct Regulation’ (2015) Butterworths Journal of International Banking and Financial Law 30, 413, 415–16 (MiFID I rules often result in ‘a regulatory scheme that implements a diluted form of fiduciary duties’). 172 An interesting combination of principle-based and rule-based regulation is offered by Australia, where a best interest standard is coupled with more detailed conduct of business rules that have the function of a safe harbour: while sticking to the safe harbour is sufficient for compliance purposes, firms are free to find other ways to ensure compliance with the general standards (Tuch (n. 64), 557–8).

5 PRODUCT GOVERNANCE AND PRODUCT INTERVENTION UNDER MIFID II/MIFIR Danny Busch

I. Introduction II. 1. 2. 3. III. 1. 2. 3.

Product Governance Scope Manufacturers and Distributors Three Perspectives Product Intervention Introduction and Scope Product Intervention by NCAs Temporary Intervention Powers of ESMA and EBA

IV. Conclusion

I. Introduction [5.01] Some of the financial products sold in recent years have not been in the interests of the client, such as interest rate swaps sold to small- and medium-sized enterprises in many European countries. This is why consideration has been given to ways of nipping this problem in the bud; in other words by preventing harmful products from even reaching the market.

Under MiFID II this has taken the form of a mandatory product-approval process. But, as usual, firms will look for ways around these requirements. It would be naive to think that product-approval schemes could in practice guarantee that harmful products are no longer marketed. This is why there must be a safety net. This safety net takes the form of a power for the national competent authorities (NCAs) and also for the ESMA and EBA to remove harmful products from the market—a system known as product intervention. In this chapter, the new MiFID II/MiFIR rules on product governance and product intervention are analysed and discussed.

II. Product Governance 1. Scope [5.02] The product-governance rules apply first and foremost to investment firms and banks that may provide investment services.1 However, they also apply to other entities which can provide investment services, in particular managers of undertakings for collective investment in transferable securities (UCITS) and alternative investment funds (AIFs) when such entities are authorised to perform MiFID investment services (pursuant to Article 6(3) of UCITS2 and Article 6(4) of [the Alternative Investment Fund Managers Directive, or] AIFMD3 respectively) and only in connection to the performance of such services. Such UCITS management companies or alternative investment fund managers that distribute or manufacture UCITS or AIFs to investors will only be directly subject to the requirements applicable to the investment services they provide.4

[5.03] In short, the product-governance rules apply to these managers only in connection with the provision of investment services. This means that UCITS management companies and AIF managers which offer investors units in UCITS or AIFs managed by them (without also providing them with investment services) are not subject to the product-governance rules of MiFID II as a result of making such an offer. It follows that UCITS management companies and AIF managers which provide no investment

services at all will therefore not be subject in any way to the productgovernance rules of MiFID II. ESMA rightly states that it would be preferable for the UCITS and AIFM Directives to be amended in such a way that these managers too are subject to the product-governance rules under MiFID II.5 [5.04] Nonetheless, many managers will also be bound de facto by the product-governance rules under MiFID II as the unit rights are often marketed through intermediaries—investment firms. These investment firms will qualify as distributors.6 To be able to comply with their productgovernance obligations, the distributors are reliant on cooperation with the manufacturer of the product, in this case the UCITS management company or AIF manager. Although managers are admittedly not bound by the product-governance rules, it seems to me that they have every interest in ensuring that their distributors comply with the rules. It follows that they must determine the target market properly so that distributors can offer the units to the correct investors.7 In the remainder of this chapter I will disregard the situation in which a UCITS management company or AIF manager provides investment services. [5.05] Finally, it should be noted that the product-governance rules apply not only to financial instruments but also to selling and advising on structured deposits.8 A party that offers or advises on structured deposits may be a bank which does not provide investment services. The productgovernance rules therefore do not apply solely to investment firms, banks, UCITS management companies, and AIF managers that may provide investment services, but also to ‘ordinary’ banks. In this chapter I will often use the more neutral terms ‘firm’ and ‘product’.

2. Manufacturers and Distributors A. General [5.06] MiFID II distinguishes between firms that manufacture the product (manufacturer) and those that distribute the product (distributor). Each of these categories has its own responsibility in respect of product governance.

In practice, a firm can naturally be both a manufacturer and a distributor. In such a case, it must comply with both sets of product-governance rules.9

B. Manufacturers [5.07] Following ESMA’s technical advice, the Draft Commission Delegated Directive published on 7 April 2016 makes it clear that the term ‘manufacturer’ must be interpreted broadly. Manufacturing of products ‘encompasses the creation, development, issuance and/or design of financial instruments’.10 This includes firms ‘advising corporate issuers on the launch of new financial instruments’.11 In short, if an investment firm advises a company on the issue of shares or bonds, it must complete the product-approval process.12 It has been argued in the literature that is going rather far,13 but we must bear in mind that the product-governance requirements should be complied with in an appropriate and proportionate manner, ‘taking into account the nature of the financial instrument, the investment service and the target market for the product’.14 ESMA also recommends that the product-governance obligations should apply to all financial instruments, for example ordinary shares and bonds, even if the term ‘manufacture’ perhaps does not seem (linguistically) applicable to them.15 Nonetheless, ESMA’s approach seems to me to be correct in view of the object of investor protection and the text of MiFID II and the Draft Commission Delegated Directive, which both provide, after all, that the product-governance rules apply to all financial instruments (and structured deposits). [5.08] In practice, it is naturally quite conceivable that two or more manufacturers collaborate to create a new product. In such a case, following ESMA’s technical advice, the Draft Commission Delegated Directive stipulates that they outline their mutual responsibilities in a written agreement.16

C. Distributors

[5.09] The Commission also chooses a broad approach in the case of distributors. In response to the previous consultation, various market participants had requested that the obligations for distributors should not be extended to investment firms which distribute financial instruments on an execution-only service basis. However, ESMA saw no cause for such a restriction,17 and this approach, which seems to me to be correct, is followed by the Commission. See Recital (18) of the Draft Commission Delegated Directive: In light of the requirements set out in [MiFID II] and in the interest of investor protection, product governance rules should apply to all products sold on primary and secondary markets, irrespective of the type of product or service provided and of the requirements applicable at point of sale. However, those rules may be applied in a proportionate manner, depending on the complexity of the product and the degree to which publicly available information can be obtained, taking into account the nature of the instrument, the investment service and the target market. Proportionality means that these rules could be relatively simple for certain simple, products distributed on an execution-only basis where such products would be compatible with the needs and characteristics of the mass retail market.18

[5.10] In practice, there is quite often likely to be a chain of distributors. According to the Commission (following ESMA’s technical advice), the final distributor in the chain (i.e. the party with the direct client relationship) has ultimate responsibility for meeting the product-governance obligations that apply to distributors.19 But the intermediate distributors too have responsibilities. They must ensure that relevant product information is passed from the manufacturer to the final distributor in the chain.20 Similarly, if the product manufacturer requires information on product sales in order to comply with its own product-governance obligations, the intermediate firm must enable him to obtain it.21 And, finally, an intermediate distributor must apply any relevant product-governance obligations for manufacturers in relation to the service they provide.22 This provision is as opaque as it is remarkable. After all, an intermediate distributor is only a manufacturer if it has ‘manufactured’ the product or, for example, has itself made modifications to the product, whether or not in cooperation with one or more other manufacturers (for this purpose, the term ‘manufacturer’ should be broadly interpreted23). If that is not the case,

I can see no good reason why an intermediate distributor should be bound by the product-governance rules for manufacturers.

3. Three Perspectives [5.11] MiFID II approaches the subject of product governance from three perspectives: (1) corporate governance, (2) investor protection, and (3) organizational requirements.

A. Corporate Governance [5.12] From the corporate governance perspective, it is apparent that the management body has final responsibility for product governance. MiFID II provides that the management body defines a policy as to products (and other matters) which is in accordance not only with the risk tolerance of the firm but also with the characteristics and needs of the clients of the firm to whom they will be offered or provided, including carrying out appropriate stress testing. The management body is also responsible for approving this policy and overseeing its implementation.24 In other words, firms must ensure that the management body has effective control over the firm’s product-governance process. The provision applies both to firms that manufacture products (manufacturers) and to firms that offer or recommend those products (distributors).25 [5.13] Firms must ensure that the compliance reports to the management body systematically include information about (1) the products they manufacture, including information on the distribution strategy, if they are manufacturers, and (2) the products they offer or recommend and the services provided, if they are (also) distributors. Firms must make the reports available to their NCA on request.26

B. Investor Protection

[5.14] From the investor protection perspective the ultimate goal of product governance is apparent, namely protection of both professional and non-professional clients. The relevant provisions distinguish between the duties of care of firms which manufacture products (manufacturers) and parties which offer or recommend those products (distributors). [5.15] Manufacturers must ensure that (1) products are manufactured which meet the needs of an identified target market of end clients within the relevant category of clients, (2) the strategy for distribution of the products is compatible with the identified target market, and (3) they take reasonable steps to ensure that the product is distributed to the identified target market.27 [5.16] Distributors must (1) understand the products they offer or recommend, (2) assess whether the products are compatible with the needs of the clients to whom they provide investment services, and (3) ensure that the products are offered or recommended only when this is in the interests of the client.28 This requirement applies both to firms which sell products they have manufactured themselves (distributors-cum-manufacturers) and to firms which sell or recommend products manufactured by others (pure distributors).

C. Organizational Requirements i. General [5.17] Finally, from the perspective of the organizational requirements it can be seen how product governance must be firmly embedded within the firm’s organization. The compliance departments of manufacturers and distributors play an important role in this connection. It is their responsibility to ensure that the product-governance obligations are implemented and complied with within the organization.29 ii. Product-Approval Process for Manufacturers [5.18] Each manufacturer must have put in place a product-approval process. Through this process each newly manufactured product (and each significant adaptation of an existing product) must be approved before it is marketed or distributed to clients.30

[5.19] A product is always assessed in relation to the intended target market, for example retail or wholesale (or naturally a more specific target market). The approval process must ensure that all risks to such identified target market are assessed. The process must also ensure that the intended distribution strategy is consistent with the identified target market.31 As part of this process, the manufacturer must also identify any groups of investors for whose needs, characteristics, and objectives the product is not compatible.32 Where firms collaborate to manufacture a product, only one target market needs to be identified.33 [5.20] Manufacturers distributing their products through other parties (i.e. distributors) must determine the needs and characteristics of clients for whom the product is compatible based on their theoretical knowledge of and past experience with the product or similar products, the financial markets, and the needs, characteristics, and objectives of potential end clients.34 [5.21] Various market participants indicated in the previous consultations that they would appreciate guidance from ESMA on how to determine the target market for a new product. However, according to ESMA this was not necessary: ESMA considers that it would be inappropriate to specify in too much detail the level of granularity that is required, since this will vary according to the specific circumstances. For simpler, more mainstream investments, such as ordinary shares, it is likely that the target market will be identified with less detail. In many cases, it is understood that such products can be considered compatible with the mass retail market. For more complicated, less mainstream products, such as contingent convertible securities or structured products with complicated return profiles, the target market should be identified with more detail.35

[5.22] This approach is followed by the Commission, see Recital (19) of the Draft Commission Delegated Directive: The level of granularity of the target market and the criteria used to define the target market and determine the appropriate distribution strategy should be relevant for the product and should make it possible to assess which clients fall within the target market, for example to assist the ongoing reviews after the financial instrument is launched. For simpler, more common products, the target market could be identified

with less detail while for more complicated products such as bail-inable instruments or less common products, the target market should be identified with more detail.36

[5.23] See also Article 9(9), first sentence, Draft Commission Delegated Directive, stipulating that Member States shall require investment firms to identify at sufficiently granular level the potential market for each financial instrument and specify the type(s) of client for whose needs, characteristics and objectives the financial instrument is compatible.37

[5.24] What other aspects should the manufacturer take into account in the product-approval process? 1. Firms manufacturing products must ensure that the design of the product, including its features, (a) does not adversely affect end clients or (b) does not lead to problems with market integrity by enabling the firm to mitigate and/or dispose of its own risks or exposure to the underlying assets of the product, where the investment firm already holds the underlying assets on own account.38 2. The manufacturers must prepare a conflict-of-interest analysis each time a product is manufactured. In particular, firms must assess whether the product creates a situation where end clients may be adversely affected if they take (a) an exposure opposite to the one previously held by the firm itself, or (b) an exposure opposite to the one that the firm wants to hold after the sale of the product.39 3. Firms must consider whether the product may represent a threat to the orderly functioning or to the stability of financial markets before deciding to proceed with the launch of the product.40 4. The manufacturer must also prepare a scenario analysis. This involves studying the risk of poor end client outcomes posed by the product and in which circumstances these outcomes may occur. The firm must assess the financial instrument under negative conditions (stress testing). The following circumstances can be taken into account here: (a) the market environment deteriorates, (b) the manufacturer or a third party involved in the manufacturing and/or functioning of the product experiences financial difficulty or other counterparty risk materializes, (c) the product fails to become commercially viable, or (d) demand for

the product is much higher than anticipated, putting a strain on the firm’s resources and/or on the market of the underlying product.41 5. The manufacturer must consider whether the product meets the identified needs, characteristics, and objectives of the target market. It is necessary to check, for example, whether (1) the product’s risk/reward profile is consistent with the target market, and (2) the product design is driven by features that benefit the client and not by a business model that relies on poor client outcomes to be profitable.42 6. Manufacturers need to consider the charging structure proposed for the product and check, for example, that (a) the product’s costs and charges are compatible with the needs, objectives, and characteristics of the target market; (b) charges do not undermine the return expectations of the product; for example, where the costs or charges are equal, exceed or remove almost all the expected tax advantages linked to a product, and (c) the charging structure of the product is appropriately transparent for the target market (e.g. it must not be too complex to understand and must not disguise charges).43 iii. Product-Governance Procedures for Distributors [5.25]  Distributors do not need an approval process, but, as noted above,44 they must, among other things, (1) assess whether the products meet the needs of the clients to whom they offer investment services, taking account of the identified target market of end clients, and (2) ensure that products are offered or recommended only when this is in the interests of the client.45 [5.26] This objective must be implemented within the firm’s organization as follows. If a distributor decides what products and services it wishes to provide or recommend, it must have implemented adequate productgovernance procedures which ensure that (1) the products and services it intends to offer are compatible with the needs, characteristics, and objectives of the identified target market, and (2) the intended distribution strategy is consistent with the identified target market. Distributors must therefore determine the needs of the clients that they intend to focus on, so as to ensure that clients’ interests are not compromised as a result of commercial or funding pressures. As part of this process, distributors must identify any groups of investors for whose needs, characteristics, and objectives the product or service is not compatible.46

[5.27] A distributor must obtain from manufactures that are subject to MiFID II information to gain the necessary understanding and knowledge of the products they intend to recommend or sell, in order to ensure that these products will be distributed in accordance with the needs, characteristics, and objectives of the identified target market.47 [5.28] A distributor must take all reasonable steps to ensure it also obtains adequate and reliable information from manufacturers not subject to MiFID II to ensure that products will be distributed in accordance with the characteristics, objectives, and needs of the target market. Where relevant information is not publicly available, the distributor must take all reasonable steps to obtain such relevant information from the manufacturer or its agent. Acceptable publicly available information is information which is clear, reliable, and produced to meet regulatory requirements, such as disclosure requirements under the Prospectus Directive48 or the Transparency Directive.49 This obligation is relevant for products sold on primary and secondary markets and applies in a proportionate manner, depending on the degree to which publicly available information is obtainable and the complexity of the product.50 [5.29] A distributor which has not itself manufactured a product is dependent on the manufacturer for the provision of information on the product. This is why the manufacturer is obliged to provide adequate information about the product and the product-approval process to all distributors, including the product’s identified target market.51 This includes information about the appropriate channels for product distribution, the product-approval process, and the target market assessment. The information should be of an adequate standard to enable distributors to understand and recommend or sell the product properly.52 A distributor, in its turn, must use the information obtained from manufacturers and information on its own clients to identify the target market and distribution strategy.53 Where a firm is both manufacturer and distributor, it is naturally necessary to determine the target market only once.54 iv. Product-Governance Requirements in Respect of Staff [5.30]  Manufacturers must ensure that relevant staff involved in the manufacturing

of products possess the necessary expertise to understand the characteristics and risks of the products they intend to manufacture.55 [5.31] Distributors have a comparable obligation. However, they must ensure not only that relevant staff understand the characteristics and risk of the products they are distributing, but also the needs, characteristics, and objectives of the identified target market.56 v. Monitoring Obligation General [5.32] Product-governance obligations do not only apply at the start of the process. A firm must regularly review products which it ‘offers or markets’, taking into account any event that could materially affect the potential risk to the identified target market, in order to assess at least (1) whether the product remains consistent with the needs of the identified target market and (2) whether the intended distribution strategy remains appropriate (together known as the monitoring obligation).57 Although the words ‘offers or markets’ are not crystal clear, this must be read as meaning that both the manufacturer and the distributor have a monitoring obligation. In any event, following ESMA’s technical advice, the Draft Commission Delegated Directive subjects both manufacturers and distributors to a monitoring obligation.58 Monitoring Obligation—Manufacturer [5.33] A manufacturer must review the products it manufactures on a regular basis, taking into account any event that could materially affect the potential risk to the identified target market. A manufacturer must consider if the product remains consistent with the needs, characteristics, and objectives of the target market and if it is being distributed to the target market, or is reaching clients with whose needs, characteristics, and objectives the product is not compatible.59 [5.34] A manufacturer must review (i) products prior to any further issue or relaunch, if it is aware of any event that could materially affect the potential risk to investors and (ii) at regular intervals to assess whether the products function as intended. A manufacturer must determine how

regularly to review its products based on relevant factors, including factors linked to the complexity or the innovative nature of the investment strategies pursued.60 [5.35] Manufacturers must also identify crucial events that would affect the potential risk or return expectations of the product, such as (a) the crossing of a threshold that will affect the return profile of the product, or (b) the solvency of certain issuers whose securities or guarantees may impact the performance of the product.61 What happens if such an event occurs? The manufacturer must take ‘appropriate action’. It is up to the manufacturer to determine what action it will take, provided that this is appropriate. However, following the lead of market participants, the following suggestions are made (the list is not exhaustive): (1) the provision of any relevant information on the event and its impact on the product to (a) clients, or (b) the distributors of the product if the manufacturer itself does not offer the product directly to the clients; (2) changing the productapproval process; (3) stopping further issuance of the product; (4) changing the product to avoid unfair contract terms; (5) considering whether the sales channels through which the products are sold are appropriate if it transpires that the product is not being sold as envisaged (e.g. if a product was designed for a niche market of sophisticated investors, but is being sold to a much larger group of clients); (6) contacting the distributor to discuss a modification of the distribution process; (7) terminating the relationship with the distributor; or (8) informing the relevant NCA.62 [5.36] To be able to comply with its monitoring obligation, however, a manufacturer needs information about its distributors. ESMA proposes that distributors must inform the manufacturer periodically about their experience with the product. This proposal has encountered resistance from the market. The criticism is that a reporting duty of this kind would entail extra costs and be so disproportionate to the expected benefits that small distributors would limit the number of manufacturers with which they work in order to minimize their reporting obligations. Ultimately, this would also not be in the interests of investors since they would have less choice. However, most consumer organizations consider that a reporting obligation would be useful, particularly if feedback from clients were to be included in the information supplied to the manufacturer. After considering all these

comments, ESMA decided to maintain its proposal, on the basis that such reporting can be beneficial for the functioning of product-governance obligations.63 The Draft Commission Delegated Directive closely follows ESMA’s technical advice. See Recital (20): For the efficient functioning of product governance obligations, distributors should periodically inform the manufacturers about their experience with the products. While distributors should not be required to report every sale to manufacturers, they should provide the data that is necessary for the manufacturer to review the product and check that it remains consistent with the needs, characteristics and objectives of the target market defined by the manufacturer. Relevant information could include data about the amount of sales outside the manufacturer’s target market, summary information of the types of clients, a summary of complaints received or by posing questions suggested by the manufacturer to a sample of clients for feedback.

[5.37] See finally Article 10(9) Draft Commission Delegated Directive, stipulating that ‘[…] distributors provide manufacturers with information on sales and, where appropriate, information on […] reviews to support product reviews carried out by manufacturers’.64 Monitoring Obligation—Distributor [5.38] But as the distributor too is subject to a monitoring obligation, it also has a responsibility. Distributors must periodically review and update their product-governance arrangements in order to ensure that they remain robust and fit for their purpose, and take appropriate actions where necessary.65 [5.39] Distributors must also review the products they offer or recommend and the services they provide on a regular basis, taking into account any event that could materially affect the potential risk to the identified target market. Distributors must assess at least whether the product or service remains consistent with the needs, characteristics, and objectives of the identified target market and whether the intended distribution remains appropriate. Distributors must reconsider and/or update the product-governance arrangements if they become aware that (1) they have wrongly identified the target market for a specific product or service, or (2) the product or service no longer meets the circumstances of the identified target market (e.g. where the product becomes illiquid or very volatile due to market changes).66

vi. Relationship with KYC Rules and Other MiFID II/MiFIR Provisions [5.40] The measures, processes, and arrangements described above are without prejudice to all other requirements of MiFID II and MiFIR, including those relating to disclosure, suitability or appropriateness, identification, and management of conflicts of interest and inducements.67 This means, among other things, that the firm must still comply with the applicable KYC rules even in relation to a client who comes within the product’s target group. In ESMA’s words: [t]he analysis of the target market for the purposes of product governance arrangements is distinct from and does not replace the suitability/appropriateness assessments which are conduct of business rules that take place for each specific transaction concluded by a given investor in relation to a given product.68

III. Product Intervention 1. Introduction and Scope [5.41] The previous section dealt with the MiFID II rules on product governance. As noted in the introduction to this chapter, it would be naive to think that these rules are watertight and guarantee that harmful products will never again be marketed. It is therefore no more than realistic to introduce (and enforce) ex post product-intervention provisions in addition to the ex ante product-governance rules. A distinction is made in this connection between (1) product intervention by NCAs and (2) the intervention powers of ESMA and EBA. Like the product-governance rules, the product-intervention rules apply to both financial instruments and structured deposits (referred to collectively below as ‘products’).69

2. Product Intervention by NCAs A. Powers of NCAs under MiFIR

i. General [5.42] The NCAs are the first in line to act. They may prohibit or restrict the following: 1. the marketing, distribution, or sale of (a) certain products or (b) products with certain specified features; 2. a type of financial activity or practice in or from the Member States concerned.70 [5.43] The NCA may impose a prohibition or restriction on a precautionary basis before a product has been marketed, distributed, or sold to clients. A prohibition or restriction may apply in circumstances, or be subject to exceptions, specified by the NCA.71 The NCA must publish on its website notice of any decision to impose a prohibition or restriction.72 [5.44] An NCA may impose a prohibition or restriction only if it is satisfied on reasonable grounds that the below cumulative requirements have been complied with. Once the below conditions are no longer all complied with (insofar as they are applicable), the NCA must revoke the prohibition or restriction.73 ii. ‘Significant Concern’ or ‘Threat’ [5.45] (1) (a) A product, financial activity, or practice gives rise to ‘significant investor protection concerns’, poses ‘a threat to the orderly functioning and integrity of financial markets or commodity markets or to the stability of whole or part of the financial system within at least one Member State’; or (b) a derivative has a detrimental effect on the price-formation mechanism in the underlying market.74 [5.46] But what exactly constitutes ‘a significant investor protection concern’? And when is there ‘a threat to the orderly functioning and integrity of financial markets or commodity markets’ or a threat to ‘the stability of whole or part of the financial system within at least one Member State’? The Commission adopts delegated acts specifying criteria and factors to be taken into account by NCAs in order to determine this. The criteria and factors comprise:

(a) the degree of complexity of a product and the relation to the type of client to whom it is marketed, distributed, and sold; (b) the degree of innovation of a product, activity, or practice; (c) the leverage a product or practice provides; (d) in relation to the orderly functioning and integrity of financial markets or commodity markets, the size or the notional value of an issuance of products.75 [5.47] Article 21(2)(a)–(v) Draft Commission Delegated Regulation published on 18 May 2016 specifies the criteria and factors to be taken into account in a detailed manner.76 The NCAs must assess the relevance of all factors and criteria listed in the Draft Commission Delegated Regulation, and take into consideration all relevant factors and criteria in determining when (1) the marketing, distribution, or sale of (a) certain products or (b) products with certain specified features or (2) a type of financial activity or practice, creates ‘a significant investor protection concern’, ‘a threat to the orderly functioning and integrity of financial markets or commodity markets or to the stability of the whole or part of the financial system within at least one Member State’.77 [5.48] In its technical advice, ESMA stated that for NCAs the listed factors and criteria ‘are not intended to represent an exhaustive list’.78 It is not entirely clear whether this advice has been followed by the Commission. First of all, the Draft Commission Delegated Regulation provides that [t]he factors and criteria to be assessed by competent authorities to determine whether there is a significant investor protection concern or a threat to the orderly functioning and integrity of financial markets or commodity markets or to the stability of the whole or part of the financial system within at least one Member State shall include [emphasis added] the following […].79

[5.49] In respect of ESMA and EBA, the Draft Commission Delegated Regulation uses the words ‘shall be the following’, instead of ‘shall include’.80 This provides strong evidence that for NCAs the listed criteria and factors are not intended to represent an exhaustive list, whereas for ESMA and EBA they are intended to represent such an exhaustive list.

[5.50] But there are some doubts. The Draft Commission Delegated Regulation also provides that an NCA may determine the existence of ‘a significant investor protection concern’, ‘a threat to the orderly functioning and integrity of financial markets or commodity markets or to the stability of the whole or part of the financial system within at least one Member State’ based on ‘one or more’ of the factors and criteria specified in Article 21(2)(a)–(v) Draft Commission Delegated Regulation, suggesting that for NCAs the listed criteria and factors are intended to represent an exhaustive list.81 The exact same formulation is used for ESMA and EBA.82 See also Recital (19) Draft Commission Delegated Regulation: The need to assess all criteria and factors that could be present in a specific factual situation should not prevent however the temporary intervention power from being used by competent authorities, ESMA and EBA where only one of the factors or criteria leads to such a concern or threat.

[5.51] In relation to investor protection, reference is made to a ‘significant concern’, whereas in relation to the orderly functioning and integrity of financial and commodity markets, reference is made to a ‘threat’. Recital (18) Draft Commission Delegated Regulation explains the difference: The existence of a ‘threat’, one of the prerequisites of the intervention in the perspective of the orderly functioning and integrity of financial or commodity markets or stability of the financial system, would require the existence of a greater concern than a ‘significant concern’, which is the prerequisite of the intervention for investor protection.83

iii. Other Requirements [5.52] (2) The existing regulatory requirements under Union law applicable to the relevant product or financial activity or practice do not sufficiently address the risks, and the issue would not be better addressed by improved supervision or enforcement of existing requirements.84 [5.53] (3) The action must be proportionate taking into account (a) the nature of the risks identified, (b) the level of sophistication of investors or market participants concerned, and (c) the likely effect of the action on investors and market participants who may hold, use, or benefit from the product, activity, or practice.85

[5.54] (4) The NCA has properly consulted NCAs in other Member States that may be significantly affected by the action.86 [5.55] (5) The action does not have a discriminatory effect on services or activities provided from another Member State.87 [5.56] (6) The NCA has properly consulted public bodies competent for the oversight, administration, and regulation of physical agricultural markets under Regulation (EC) No 1234/2007, where a product or activity or practice poses a serious threat to the orderly functioning and integrity of the physical agricultural market.88

B. Consultation with Other NCAs and ESMA/EBA [5.57] Before an NCA may impose a prohibition or restriction it must consult with the other NCAs and ESMA/EBA. It is not entirely clear whether there must be consultation with all other NCAs or only with NCAs in other Member States that may be significantly affected by the action. In the latter case, the provision would to this extent seem to repeat cumulative condition (4) under paragraph 5.54 above. Whatever the case, not less than one month before the measure is intended to take effect, it must notify all other NCAs and ESMA in writing (or through another medium agreed between the authorities) the details of: (a) the product89 or activity or practice to which the proposed action relates; (b) the precise nature of the proposed prohibition or restriction and when it is intended to take effect; and90 (c) the evidence upon which it has based its decision and upon which it is satisfied that each of the conditions at paragraphs 5.43–5.56 above, are met.91 [5.58] The period of a month is naturally very long in cases where speed is of the essence. MiFIR therefore provides for an exception in such cases: if the NCA deems it necessary to take urgent action in order to prevent detriment arising from products, practices, or activities, it may take action

on a provisional basis with no less than twenty-four hours’ written notice, before the measure is intended to take effect, to all other NCAs and ESMA or, for structured deposits, EBA, provided that all the criteria are met and that, in addition, it is clearly established that a one-month notification period would not adequately address the specific concern or threat. However, the NCA may not take action on a provisional basis for a period exceeding three months.92

C. Coordinating Role of ESMA and EBA [5.59] ESMA (for financial instruments) or EBA (for structured deposits) perform a facilitation and coordination role in relation to action taken by NCAs. In particular they ensure that action taken by an NCA is justified and proportionate and that where appropriate a consistent approach is taken by NCAs.93 [5.60] After receiving notification of any action that is to be imposed, ESMA or EBA will adopt an opinion on whether the prohibition or restriction is justified and proportionate. If ESMA or EBA considers that the taking of a measure by other NCAs is necessary to address the risk, it will state this in its opinion. The opinion must be published on ESMA’s website (in the case of financial instruments) or EBA’s website (in the case of structured deposits).94 Where an NCA proposes to take, or takes, action contrary to an opinion adopted by ESMA or EBA or declines to take action contrary to such an opinion, it must immediately publish on its website a notice fully explaining its reasons for so doing (comply or explain).95

D. Powers of NCAs under MiFID II [5.61] NCAs also have product-intervention powers under MiFID II. First of all, the NCA may suspend the marketing or sale of products where the firm (which may be a manufacturer or a distributor) has not developed or applied an effective product-approval process or has otherwise failed to comply with the product-governance rules.96 I assume that this provision relates to the situation where the product-governance rules have not been

complied with in relation to a given product and there is therefore no guarantee that the product is ‘safe’. To ensure this safety after all, the NCA can then suspend the marketing and sale of the product. Another interpretation is also possible, but seems less likely. If the productgovernance rules have not been observed in relation to a given product, the NCA may also suspend trading in other products which are marketed or sold by the firm in question, even if the product-governance rules have been adequately complied with in relation to such products. In such a case the power of suspension would be more in the nature of a sanction. [5.62] Second, the NCA may suspend the marketing or sale of products where the conditions of Article 40, 41, or 42 MiFIR are not met.97 Articles 40 and 41 MiFIR relate to the temporary intervention measures of ESMA and EBA.98 So should we interpret MiFID II as meaning that the NCAs may suspend the marketing or sale of products in all cases in which ESMA and EBA are unable to do so because the conditions of Articles 40 (ESMA) and 41 (EBA) MiFIR have not been met? This would then constitute a very broad intervention power. And what about the reference to Article 42 MiFIR? This provision relates to the intervention measures which the NCAs can themselves take under MiFIR (see above99). On a literal interpretation of Article 69(2)(s) MiFID II, an NCA may, in all cases where it cannot prohibit or restrict the marketing or sale of a product under Article 42 MiFIR because the conditions of application are not met, still suspend the marketing or sale of products under Article 69(2)(s) MiFID II. Although suspension is only a temporary measure and an outright prohibition or restriction under Article 42 MiFIR seems much more final, I cannot help but wonder whether Article 69(2)(s) MiFID II is really intended to confer an unconditional power of suspension. Perhaps the provision should be read as meaning that if a legal prohibition or restriction under Article 40, 41, or 42 MiFIR is not complied with by the firm concerned, the NCA may halt the actual marketing and sale.

3. Temporary Intervention Powers of ESMA and EBA A. General

[5.63] Besides the NCAs, ESMA and EBA may themselves also intervene. However, these powers are secondary to those of the NCAs. [5.64] But let us first retrace a few steps. Under the ESMA and EBA Regulations, ESMA and EBA have obtained tasks relating to consumer protection and financial activities.100 ESMA and EBA supervise new and existing financial activities and can adopt guidelines and recommendations with a view to promoting the safety and soundness of markets and convergence of regulatory practice.101

B. Powers under MiFIR [5.65] ESMA and EBA may temporarily prohibit or restrict certain financial activities that threaten (1) the orderly functioning and integrity of financial markets or (2) the stability of the whole or part of the financial system of the EU/EEA in the cases specified and under the conditions laid down in further legislative acts.102 MiFIR is a further legislative act of this kind. MiFIR sets out what activities may be prohibited and on what conditions.103 The powers of ESMA and EBA are always identical, although ESMA is in each case the competent authority in relation to financial instruments and EBA in relation to structured deposits. ESMA/EBA may prohibit or restrict the following activities in the EU/EEA: (1) the marketing, distribution, and sale of (a) certain products or (b) products with certain specified features; (2) a type of financial activity or practice. [5.66] A prohibition or restriction may apply in circumstances, or be subject to exceptions, specified by ESMA/EBA.104 ESMA/EBA may impose the prohibition or restriction on a precautionary basis before a product has been marketed, distributed, or sold to clients.105 ESMA/EBA publishes on its website notice of any decision to take any action.106 In short, ESMA/EBA has powers and duties of publication comparable to those of the NCAs (see above107), although the powers relate to the entire EU/EEA and are not limited to a Member State, and any action taken by

ESMA/EBA will always be temporary. Before ESMA/EBA decides to take action, it must notify the NCAs of the action it is proposing.108 ESMA/EBA must review a prohibition or restriction at appropriate intervals and at least every three months. If the prohibition or restriction is not renewed after that three-month period, it will expire.109 As the action under MiFIR must be taken ‘in accordance with Article 9(5) [ESMA/EBA Regulations]’,110 a Member State may request ESMA/EBA to reconsider its decision under MiFIR. In such a case ESMA/EBA will decide whether to maintain its decision.111 [5.67] ESMA/EBA may act only if the following cumulative conditions have been fulfilled: 1. The proposed action addresses a ‘significant investor protection concern’ or a ‘threat to the orderly functioning and integrity of financial markets or commodity markets or to the stability of the whole or part of the financial system’ in the EU/EEA. 2. Regulatory requirements under Union law that are applicable to the relevant product or activity do not address the threat. 3. An NCA or NCAs have not taken action to address the threat, or the actions that have been taken do not adequately address the threat.112 [5.68] The first two of these conditions are virtually identical to those imposed in respect of product intervention under MiFIR by NCAs.113 In the case of the first requirement, the Commission (like in relation to the NCAs) must adopt delegated acts specifying criteria and factors to be taken into account in determining when there is a ‘significant investor protection concern’, a ‘threat to the orderly functioning and integrity of financial markets or commodity markets or to the stability of the whole or part of the financial system’ of the EU/EEA. This covers the same criteria and factors as those which are relevant to NCAs.114 For ESMA and EBA there are no doubts that the list of criteria and factors is intended to represent an exhaustive list.115 [5.69] The size or the notional value of an issuance of products is mentioned separately in the case of ESMA/EBA, whereas in the case of the

NCAs this factor is explicitly related to the orderly functioning and integrity of financial markets or commodity markets.116 [5.70] It is apparent from the third condition that the powers of ESMA and EBA are secondary to those of the NCAs. But if ESMA or EBA takes action, its measures prevail over those of an NCA.117 I understand this provision to mean that supervised investment firms/banks which are affected by the product intervention must be guided by the measures taken by ESMA/EBA and can ignore any previous measures taken by NCAs. This would in any event seem to be the most logical explanation, because in view of condition (3) above ESMA/EBA can act only once the measures taken by NCAs have failed to deal with the threat adequately. If the national measures are not adequate, it makes little sense for them to be complied with by the sector unless it is assumed that cases can arise where the NCA’s action is not sufficient on its own but is sufficient in combination with a measure taken by ESMA/EBA. [5.71] When taking action ESMA/EBA must ensure that the action: (a) does not have a detrimental effect on the efficiency of financial markets or on investors that is disproportionate to the benefits of the action; (b) does not create a risk of regulatory arbitrage; and (c) has been taken after consulting the public bodies competent for the oversight, administration, and regulation of physical agricultural markets under Regulation (EC) No 1234/2007, where the measure relates to agricultural commodities derivatives.118 [5.72] As we saw above,119 after receiving notification of any action that is to be imposed, ESMA/EBA must adopt an opinion on whether the prohibition or restriction is justified and proportionate. If ESMA or EBA considers that the taking of a measure by other competent authorities is necessary to address the risk, it must state this in its opinion.120 However, if an NCA has already taken a measure without awaiting the opinion, ESMA or EBA may itself take a corrective or other measure without adopting the opinion.121

C. Power under ESMA/EBA Regulation: Emergency Situations [5.73] Besides the intervention powers which ESMA/EBA has under MiFIR (see above) and independently of any elaboration in further legislative acts, ESMA/EBA may also temporarily prohibit or restrict certain financial activities in the case of an emergency situation in accordance with and under the conditions laid down in Article 18 of the ESMA/EBA Regulations.122 The requirements for intervention on this basis are strict. As the conditions for exercise of the powers which ESMA/EBA has under MiFIR are more flexible, the power to intervene in an emergency is expected to have little added value in practice.

IV. Conclusion [5.74] Both the ex ante product-governance rules and the ex post productintervention rules are far-reaching. The combination of these two approaches to preventing harmful products from reaching the market is an important step forward in protecting investors. On the other hand, compliance with the product-governance rules is bound to entail costs for the firms concerned. As seen above, the internal procedures must be configured and the arrangement requires a considerable exchange of information between the firm that manufactures a product and the firm that distributes it. All in all, I believe that these extra costs (which will undoubtedly be discounted in the cost price) are acceptable. The social costs caused by marketing harmful products are many times greater. In my view, the fact that MiFID II introduces not only product-governance rules but also far-reaching product-intervention rules shows a sense of realism. After all, as I noted in the introduction, it would be naive to think that the product-governance rules could in practice guarantee that harmful products are no longer marketed.

1

See Article 9(3)(b); Article 16(3), second to seventh subparagraphs; Article 24(2) (non-bank investment firms) in conjunction with Article 1(3)(a) and (b) MiFID II (banks

which also act as investment firms). 2 Directive 2009/65/EC of the European Parliament and of the Council of 13 July 2009 on the coordination of laws, regulations, and administrative provisions relating to undertakings for collective investment in transferable securities, OJ L 302/32 (UCITS IV), as amended by Directive 2014/91/EU of the European Parliament and of the Council of 23 July 2014 on the coordination of laws, regulations, and administrative provisions relating to undertakings for collective investment in transferable securities (UCITS) as regards depositary functions, remuneration policies, and sanctions, OJ L 257/186 (UCITS V). UCITS V should have been implemented in the national laws of the Member States of the EU/EER from 18 March 2016 (Article 2(1), second paragraph, UCITS V). 3 Directive 2011/61/EU of the European Parliament and of the Council of 8 June 2011 on Alternative Investment Fund Managers and amending Directives 2003/41/EC and 2009/65/EC and Regulations (EC) No 1060/2009 and (EU) No 1095/2010 (AIFM Directive or AIFMD), OJ L 174/1. 4 ESMA/2014/1569, Final Report—ESMA’s Technical Advice to the Commission on MiFID II and MiFIR (19 December 2014), p. 52 (no. 9). 5 ibid. 6 See paragraphs 5.09–5.10, below. 7 See paragraphs 5.25–5.29. 8 It is evident from Article 1(4), opening words and (a) and (b), MiFID II, as well as from Article 1(2), Draft Commission Delegated Directive, C(2016) 2031 final, 7 April 2016, that the product-governance rules also apply to structured deposits. A structured deposit is a deposit on which the interest to be paid is not determined by reference to an interest rate (such as Euribor), but is instead dependent, for example, on the position of the AEX index. See the definition of structured deposit in Article 4, para. 1(43) MiFID II in conjunction with Article 2(1)(c) of Directive 2014/49/EU of the European Parliament and of the Council of 16 April 2014 on deposit guarantee schemes, OJ L 173/149 (Deposit Guarantee Schemes Directive). On the subject of structured deposits, see ESMA/2015/610, Consultation Paper, Draft guidelines on complex debt instruments and structure deposits (24 March 2015). 9 See Recital (17), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 51. 10 See Article 9(1), first paragraph, Draft Commission Delegated Directive (n. 8); ESMA (n. 4), 51, 55 (no. 2). And the same obviously applies to structured deposits, see Article 1(2), Draft Commission Delegated Directive (n. 8). 11 See Recital (15), Draft Commission Delegated Directive (n. 8); ESMA (n. 4), 51. And the same obviously applies to structured deposits, see Article 1(2), Draft Commission Delegated Directive (n. 8). 12 See below at paragraphs 5.18–5.24. 13 L. J. Silverentand, ‘MiFID II—Product Governance’, Tijdschrift voor Financieel Recht (2015), p. 63.

14

See Article 9(1), second paragraph, Draft Commission Delegated Directive (n. 8). And the same obviously applies to structured deposits, see Article 1(2), Draft Commission Delegated Directive (n. 8). 15 ESMA (n. 4), 53–4 (no 12). Cf. L. J. Silverentand (n. 13). 16 See Article 9(8), Draft Commission Delegated Directive (n. 8); ESMA (n. 4), 56 (no. 9). 18 See Recital (18), Draft Commission Delegated Directive (n. 8). See also Recital (15), in fine, Draft Commission Delegated Directive (n. 8), stipulating that firms ‘that offer or sell financial instruments and services to clients should be considered distributors’. That product-governance obligations for distributors must be applied in an appropriate and proportionate manner is also apparent from Article 10(1), first paragraph, Commission Delegated Directive (n. 8). See also ESMA (n. 4), 55 (no. 1). 17 ESMA (n. 4), 56 (no. 9). 19 See Article 10(10), first sentence (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 61 (no. 31, first sentence). 20 See Article 10(10)(a) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive, (n. 8). See also ESMA (n. 4), 61 (no. 31 sub (i)). 21 See Article 10(10)(b) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 61 (no. 31 sub (ii)). 22 See Article 10(10)(c) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 61 (no. 31 sub (iii)). 23 See above, paragraph 5.07. 24 Article 9(3)(b) (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II. 25 Article 9(6) (manufacturers) and Article 10(8) (distributors) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 7) (manufacturers), 60 (no. 28) (distributors). 26 Article 9(6), second and third sentence (manufacturers) and Article 10(8), second and third sentence (distributors) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 7, second sentence) (manufacturers), 60 (no. 28, second sentence) (distributors). See also on the role of the compliance function paragraph 5.17. 27 Article 24(2), first paragraph (financial instruments) in conjunction with Article 1(4), opening words and (b) (structured deposits) MiFID II. See also Recital (71) MiFID II. 28 Article 24(2), second paragraph (financial instruments) in conjunction with Article 1(4), opening words and (b) (structured deposits) MiFID II. See also Recital (71) MiFID II.

29

See Article 9(7) (manufacturers) and Article 10(6) (distributors) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 8) (manufacturers), 60 (no. 26) (distributors). See on the compliance function also paragraph 5.13. 30 Article 16(3), second paragraph (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II. 31 Article 16(3), third paragraph (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II. 32 Article 9(9), first paragraph, second sentence (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 10). 33 Article 9(9), first paragraph, third sentence (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 10). 34 Article 9(9), second paragraph (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56–7 (no. 11). 35 ESMA (n. 4), 51 (no. 5), 56–7 (nos. 10 and 11). 36 See Recital (19), Draft Commission Delegated Directive (n. 8). And the same obviously applies to structured deposits, see Article 1(2), Draft Commission Delegated Directive (n. 8). 37 And the same obviously applies to structured deposits, see Article 1(2), Draft Commission Delegated Directive (n. 8). 38 Article 9(2), second sentence (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 3). 39 Article 9(3) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 4). 40 Article 9(4) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 56 (no. 5). 41 Article 9(10) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 57 (no. 12). As regards stress tests, see also Article 9(3)(b) (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MIFID II, referred to in paragraph 5.12 above. 42 Article 9(11) (financial instruments), in conjunction with Article 1(2) (structured deposits) Commission Delegated Directive (n. 8). See also ESMA (n. 4), 57 (no. 13). 43 Article 9(12) (financial instruments), in conjunction with Article 1(2) (structured deposits), Commission Delegated Directive (n. 8). See also ESMA (n. 4), 57 (no. 14). 44 At paragraph 5.16.

45

Article 24(2), second paragraph (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II. 46 Article 10(2), first paragraph (financial instruments), in conjunction with Article 1(2) (structured deposits), Commission Delegated Directive (n. 8). See also ESMA (n. 4), 59 (no. 21, first three sentences). 47 Article 10(2), second paragraph (financial instruments), in conjunction with Article 1(2) (structured deposits), Commission Delegated Directive (n. 8). See also ESMA (n. 4), 60 (no. 29). See Article 16(3), sixth paragraph (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II and Recital 71 MiFID II. 48 Directive 2003/71/EC of the European Parliament and of the Council of 4 November 2003 on the prospectus to be published when securities are offered to the public or admitted to trading and amending Directive 2001/34/EC, OJ L 345/64. 49 Directive 2004/109/EC of the European Parliament and of the Council of 15 December 2004 on the harmonization of transparency requirements in relation to information about issuers whose securities are admitted to trading on a regulated market and amending Directive 2001/34/EC, OJ L 390/38. 50 Article 10(2), third paragraph (financial instruments), in conjunction with Article 1(2) (structured deposits), Commission Delegated Directive (n. 8). See also ESMA (n. 4), 60–1 (no. 30). See also Article 10(1), second and third paragraph, Commission Delegated Directive (n. 8). See also ESMA (n. 4), 59 (no. 20). 51 Article 16(3), fifth paragraph (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II. 52 Article 10(13) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 57–8 (no. 15). 53 Article 10(2), third paragraph, first sentence, (financial instruments), in conjunction with Article 1(2) (structured deposits) Commission Delegated Directive (n. 8). See also ESMA (n. 4), 59 (no. 21, sentences four and five), where in sentence five reference is made to Article 16(3), fourth paragraph, MiFID II. This is evidently an error, since ESMA intends to refer to Article 16(3), third paragraph, MiFID II. 54 Article 10(2), third paragraph, second sentence, and Recital (17), second sentence, (financial instruments), in conjunction with Article 1(2) (structured deposits), Commission Delegated Directive (n. 8). See also ESMA (n. 4), 59 (no. 21, sentence 6). 55 Article 9(5) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive. See also ESMA (n. 4), 56 (no. 6). Cf. also the corporate governance requirement that the management body should approve the internal organization of the firm, including criteria for the selection, training, knowledge, skills, and experience of the staff. See Article 9(3)(a) (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II; see also Recital (54) MiFID II.

56

Article 10(7), (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 60 (no. 27). See also Article 24(2), second paragraph, (financial instruments) in conjunction with Article 1(4), opening words and (b) (structured deposits) MiFID II, which provides that ‘an investment firm shall understand the financial instruments they offer or recommend’. See also paragraph 5.16 above. 57 Article 16(3), fourth paragraph (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II. 58 Article 9(14) and (15) (manufacturers) and Article 10(4) and (5) (distributors), (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4) 58 (nos. 16–19) (manufacturers), 60 (nos. 23 and 24) (distributors). As to manufacturers see also Recital (71) MiFID II, making it clear that ‘[i]nvestment firms that manufacture financial instruments […] periodically review the identification of the target market of and the performance of the products they offer’ (emphasis added). 59 Article 9(14) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). 60 Article 9(15), first and second sentence (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). 61 Article 9(15), third sentence (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). 62 Article 9(15), fourth sentence, in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive(n. 8). See also ESMA (n. 4), 58–9 (no. 19). See also pp. 54–5 (nos. 15–17). 63 ESMA (n. 4), 54 (no. 14) and 60 (no. 25). 64 See also ESMA (n. 4), 60 (no. 25). As for structured deposits, Article 10(9) must of course be read in conjunction with Article 1(2), Draft Commission Delegated Directive (n. 8). 65 Article 10(4) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 60 (no. 23). 66 Article 10(5) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8). See also ESMA (n. 4), 60 (no. 24). 67 Article 16(3), seventh paragraph (financial instruments) in conjunction with Article 1(4), opening words and (a) (structured deposits) MiFID II, Recital 71 MiFID II. See in a similar vein Article 10(3) (financial instruments), in conjunction with Article 1(2) (structured deposits), Draft Commission Delegated Directive (n. 8), with the addition that ‘[…] particular care shall be taken when distributors intend to offer or recommend new products or there are variations to the services they provide’. See also ESMA (n. 4), 59 (no. 21, in fine). 68 ESMA (n. 4), 51 and 59–60 (no. 22).

69

See the text of the Articles 39–43, MiFIR and Articles 19–21, Draft Commission Delegated Regulation , C(2016) 2860 final, 18 May 2016. See also Article 69(2)(s) and (t) (in conjunction with Article 1(4)(c)) MiFID II. 70 Article 42(1) MiFIR. 71 Article 42(2), second and third paragraphs, MiFIR. 72 Article 42(5) MiFIR. The notice must specify details of the prohibition or restriction, a time after the publication of the notice from which the measures will take effect, and the evidence upon which it is satisfied each of the conditions are met. A prohibition or restriction applies only in relation to actions taken after the publication of the notice. 73 Article 42(6) MiFIR. 74 Article 42(2)(a) MiFIR. 75 Article 42(7) MiFIR. 76 Draft Commission Delegated Regulation (n. 69). See also ESMA (n. 4), 191–6 (no. 3). 77 Article 21(1), first paragraph, Draft Commission Delegated Regulation (n. 69). 79 Article 21(2), first paragraph, in fine, Commission Delegated Regulation (n. 69). 78 ESMA (n. 4), 191 (no. 3, second sentence). EBA even seems to suggest that in respect of structured deposits the list of criteria is non-exhaustive not only for the NCAs but also for EBA. See EBA/Op/2014/13, Technical Advice on possible delegated acts on criteria and factors for intervention powers concerning structured deposits under Articles 41 and 42 of Regulation (EU) No 600/2014 (MiFIR) (11 December 2014), p. 7 (no. 3): ‘In accordance with the overall conditions for intervention specified under Title VII, Chapter 1 of MiFIR, EBA and CAs should also be able to intervene in new instruments, services or activities that may not meet these factors or criteria or, conversely, not necessarily intervene if given criteria are met but overall detriment is not foreseen or detected, or the relevant proportionality test is not satisfied.’ 80 Article 19(2), first paragraph, in fine (ESMA), Article 20(2), first paragraph, in fine (EBA), Draft Commission Delegated Regulation (n. 69). 81 Article 21(1), second paragraph, Draft Commission Delegated Regulation (n. 69). 82 Article 19(1), second paragraph (ESMA), Article 20(1), second paragraph (EBA), Draft Commission Delegated Regulation (n. 69). 83 Draft Commission Delegated Regulation (n. 69). See also ESMA (n. 4), 190–1 (no. 2). 84 Article 42(2)(b) MiFIR. 85 Article 42(2)(c) MiFIR. 86 Article 42(2)(d) MiFIR. 87 Article 42(2)(e) MiFIR. 88 Article 42(2)(f) MiFIR. 89 Article 42(3)(a) MiFIR refers only to ‘financial instrument’ and not also to ‘structured deposit’. This seems to me to be an error. In the body of the text, I have therefore assumed that Article 42(3)(a) MiFIR also relates to structured deposits.

90

Article 42(3)(b) MiFIR. Article 42(3)(c) MiFIR. 92 Article 42(4) MiFIR. 93 Article 43(1) MiFIR. 94 Article 43(2) MiFIR. 95 Article 43(3) MiFIR. 96 Article 69(2)(t) (financial instruments) in conjunction with Article 1(4)(c) (structured deposits) MiFID II. 97 Article 69(2)(s) (financial instruments) in conjunction with Article 1(4)(c) (structured deposits) MiFID II. 98 See below, paragraphs 5.63–5.72. 99 At paragraphs 5.42–5.56. 100 Article 9, Regulation (EU) No 1095/2010, OJ L 331, 15 December 2010, pp. 84–119 (ESMA Regulation); Article 9 Regulation (EU) No 1093/2010, OJ L 331, 15 December 2010, pp. 12–47 (EBA Regulation). 101 Article 9(2) ESMA Regulation; Article 9(2) EBA Regulation. 102 Article 9(5) ESMA Regulation; Article 9(5) EBA Regulation. 103 Article 40 (ESMA’s temporary intervention powers) and 41 (EBA’s temporary intervention powers) MiFIR. 104 Article 40(1) (ESMA); Article 41(1) (EBA) MiFIR. 105 Article 40(2), second paragraph (ESMA); Article 41(2), second paragraph (EBA) MiFIR. 106 Article 40(5) (ESMA); Article 41(5) (EBA) MiFIR. The notice must specify details of the prohibition or restriction and a time after the publication of the notice from which the measures will take effect. A prohibition or restriction will only apply to action taken after the measures take effect. Article 40(5) (ESMA); Article 41(5) (EBA) MiFIR. 107 At paragraphs 5.43–5.56. 108 Article 40(4) (ESMA); Article 41(4) (EBA) MiFIR. 109 Article 40(6) (ESMA); Article 41(6) (EBA) MiFIR. See also Article 9(5), third paragraph, ESMA/EBA Regulations. 110 See Articles 40(1) (ESMA) and 41(1) (EBA) MiFIR. 111 Article 9(5), third paragraph, ESMA/EBA Regulations. 112 Article 40(2), first paragraph (ESMA); Article 41(2), first paragraph (EBA) MiFIR. 113 See above at paragraphs 5.42–5.56. 114 Article 40(8) (ESMA); Article 41(8) (EBA) MiFIR. See also paragraphs 5.42–5.56 above. 115 See above, paragraphs 5.45–5.51, and see Article 19(2), first paragraph, in fine (ESMA), Article 20(2), first paragraph, in fine (EBA), Draft Commission Delegated Regulation (n. 69). 91

116

Compare Article 40(8)(b) (ESMA) and Article 41(8)(b) (EBA) MiFIR with Article 42(7)(d) MiFIR (NCAs). 117 Article 40(7) (ESMA); Article 41(7) (EBA) MiFIR. 118 Article 40(3), first paragraph (ESMA); Article 41(3), first paragraph (EBA). 119 At paragraphs 5.59–5.60. 120 Article 43(2) MiFIR. 121 In any event, this is how I understand Article 40(3), second paragraph (ESMA); Article 41(3), second paragraph (EBA) MiFIR. 122 Article 9(5) ESMA Regulation; Article 9(5) EBA Regulation. See R. H. Maatman and M. H. J. M. ter Braak, ‘Financiële crisis en noodsituatie’ (2012) Ondernemingsrecht, pp. 291–3.

6 INDEPENDENT FINANCIAL ADVICE Paolo Giudici

I. Introduction II. 1. 2. 3. 4.

Economic Background Households’ Participation in Financial Markets Households’ Bias Advice Missing Research

III. MiFID I 1. MiFID I on Advice 2. The Problem with the MiFID I Regime IV. 1. 2. 3. 4.

MiFID II MiFID II on Information Independent Advice and Other Investment Services Contractual Obligations Investment Advice and Portfolio Management

V. Conclusions

I. Introduction [6.01] In this chapter I analyse the new MiFID II regime concerning investment advice, and in particular independent investment advice. There are important new provisions concerning investment advice and it is clear

that the directive places a lot of emphasis on the role of independent investment advice as an instrument to restore trust and help investors, especially retail ones, in taking financially sound decisions concerning their savings. [6.02] The chapter is organized into four further sections. Section II provides background for the growing importance attributed to financial advice as an instrument to improve the overall efficiency of financial markets. It reviews the main economic literature on households’ biases and the role that advice plays in their investment decisions. Section III briefly analyses the MiFID I regime and highlights its main issue, namely the ambiguous boundary between investment advice and the related suitability assessment, and guidance or promotion, with the much more forgiving appropriateness test or execution-only regime. Section IV focuses on MiFID II provisions concerning investment advice, exploring some of the main problems emerging from the new framework, which must be understood as an attempt to nudge European investors towards independent investment advice. Section V concludes.

II. Economic Background 1. Households’ Participation in Financial Markets [6.03] Trust is a fundamental value in financial markets. Any investment decision requires reliance on the information available, and therefore trust in the market’s integrity. There is a burgeoning literature on the role of trust in financial markets,1 and legal rules2 and doctrines3 abound with references to trust in the market. A significant amount of studies report that over the course of the financial crisis trust in financial institutions has decreased. Given the importance of financial markets as intertemporal and interpersonal resource-transfer mechanisms, along with their monetary role, this lack of trust might have a significant impact on economic development. Traditionally, households have not been as active on financial markets as might have been expected.4 Levels of participation, especially in equity

markets, vary with wealth, age, race, education, cognitive capacities, risk appetite, and social trust.5 In any event, and generally speaking, household participation appears to be suboptimal.6

2. Households’ Bias [6.04] Households have major problems with financial investments. Indeed, when households participate in financial markets, their investment choices are often ripe with mistakes. Households are generally financially illiterate.7 They tend to be overconfident8—male investors more so than women.9 They trade too much.10 Indeed, self-ignorance is one of the most important biases, and wealthy and self-confident financially illiterates represent a big group. [6.05] Financial ignorance and overconfidence are amongst the leading factors in investors’ lack of diversification.11 Households are prone to a home-bias that leads them to hold too many local financial instruments.12 Households tend to maintain fixed beliefs about the fundamental value of shares in the face of market price movements, even though they do not do so with regard to mutual funds.13 They are therefore extremely vulnerable to the disposition effect and thus to ‘momentum behaviour’, which lead them to ride losses and quickly realize gains.14 [6.06] The question as to how well retail investors fare by trading in financial markets should be rephrased: ‘how much do retail investors lose?’. The answer is straightforward: they are likely to be losers, and to lose significant amounts of money in comparison with professional traders.15 Financial literacy and access to financial advice seems to improve retail investors’ portfolios. Recent research convincingly shows that big losers in financial markets are retail investors who neither seek financial advice nor possess a good understanding of basic financial–numerical operations.16 This research confirms experimental evidence concerning retail investors’ inability to take optimal investment choices, even in an artificial and highly simplified investment environment.17 The policy implications are clear,

even though it is not so clear whether financial literacy and advice are complements or substitutes and, therefore, which one should have a preferred track.18 However, since it is not yet evident that policy measures aiming at improving financial literacy are really working, and any benefits from them will only come through in the long term, the quickest policy indication for increasing households’ trust in financial markets and improving their portfolios to the benefit of the economic system seems to be the offering of professional financial advice on affordable terms.19 Of course, this policy indication will not be particularly effective if advice and financial literacy are complements rather than subsitutes.

3. Advice A. Supply Side [6.07] Whilst overconfidence is a trait of many retail traders on stock markets, a large number of households struggle to understand today’s complex financial markets, and expect to be helped by banks and other financial institutions. A large cross-country survey in Europe has shown that almost 90 per cent of investors expect financial intermediaries to give advice through their personnel.20 However, retail investors are not used to paying for advice21 and expect it to be provided for free, which it often is, as part of the interpersonal relationships established with investment firms’ personnel. This advice can be part of a dedicated service offered by the financial institution and paid for indirectly by the client through distribution fees or commissions. Quite often, however, it is what can be termed ‘false advice’—perhaps just a chat, often with a non-specialized employee or agent,22 or a promotional effort by the financial intermediary’s salespeople where a hard sell comes under the guise of advice. [6.08] In a large online survey among purchasers of investment products in the EU, the great majority of the investors reported that their choice was directly influenced by an ‘adviser’.23 The economic research, however, is not able to distinguish the nature and quality of the advice through the data, precisely because it is very difficult if not impossible for many households

to distinguish between true professional advice and the cheap talk referred to above.

B. Is Conflicted Advice Better Than No Advice? [6.09] Investors do not like, and, in significant numbers, do not accept, paying for advice, which they somehow consider an accessory of the provision of financial services. Accordingly, advice is either paid for by product providers through ‘commissions and kickbacks’24 and is therefore open to potential conflicts of interests, or it is cheap talk offered by a salesperson or, even worse, a promotional effort which mimics qualified professional advice. [6.10] Nevertheless, there are models of ‘cheap talk’25 concerning the financial industry, where reputation protection and the fear of legal sanctions tame conflict of interest. According to these models, particularly illiterate households may benefit from cheap talk, because making judgements based on cheap talk is better than having no idea whatsoever about how to employ savings. In an influential paper Inderst and Ottaviani have modelled the different situations that can occur when two firms compete through commissions paid to an adviser. In their model, the adviser controls producers’ access to investors, and monitors the suitability of the financial products that she suggests to clients. The analysis shows commission and kickbacks are not always necessarily bad, and that mandatory disclosure can backfire and lead to suboptimal situations. In the authors’ words, ‘policies that chill commissions through mandatory disclosure or bans may have unintended consequences for efficiency because they inefficiently reduce the responsiveness of advice to supplyside differences’.26

C. Demand Side [6.11] The policy problems are increased by a demand-side puzzle. Customers expect to receive advice, but there is no real understanding of when and how retail investors rely on it. According to recent studies, more

financially educated retail investors seek more professional advice but are, at the same time, less dependent on it.27 Less sophisticated investors do not often look for advice; but if they do, they rely more on it.28 A much-cited paper that deals with this demand-side puzzle reports a situation where an investment bank proactively offered to provide advice that, according to the researchers involved, was unbiased and theoretically sound. The results are striking. A small percentage of the customers (5%) accepted the offer of advice. They were the more financially sophisticated clients. However, and rather surprisingly, they generally did not follow the advice and, accordingly, did not improve their portfolio performance, just as with all the other customers that simply refused the offer.29 It is not clear what the determinants are that make private investors ask and then rely on advice. [6.12] The demand-side puzzle apparently does not worry ‘robo-advisers’, which are joining the market and hope to attract the attention of investors whose portfolio is too small for the costs of independent, personal investment advice. Of course, the advent of robo-advisers might represent an unexpected technological shock in this area of finance, and might change the market environment and investors’ mindset.30

4. Missing Research [6.13] There are many missing parts in the economic research on investment advice. One missing part concerns the financial literacy of employees and agents of European financial institutions. There is some anecdotal evidence that employees who are more prone to recommending unsuitable products are those that do not themselves distinguish a suitable product from an unsuitable one. MiFID II addresses the problem of the competence of personnel,31 but empirical research is missing in this area. It would be worth knowing whether low-competence personnel can really be more prone to misselling and, more generally, how personnels’ financial literacy compares to those of their institutions’ customers. [6.14] Another black box concerns the contractual mechanisms that govern the relationship between financial intermediaries and the natural

persons who work on advisory services.32 Commissions and kickbacks apparently delink customers’ interest from the financial intermediary providing advisory services. However, contractual mechanisms between the financial institution and its advisory workforce might realign interest, though even a simple mechanism that connects the adviser’s income to the client’s portfolio story would add a layer of complexity to the analysis of conflict of interest conducted so far by economists. Research in this area would be very welcome.

III. MiFID I 1. MiFID I on Advice [6.15] Under the MiFID I regime, investment advice was included as a stand-alone, core investment service, subject to the suitability requirement. Investment advice was restricted to advice on particular financial instruments, presented as suitable for a person or based on a consideration of the circumstances of that person,33 and it did not require a written basic agreement.34 Accordingly, the suitability requirement could operate in any situation where an investment firm was communicating to the client that an investment in a specific financial product was ‘good for him’. This regime was not related to any specific contractual setting. Indeed, the suitability test was a legal obligation that did not arise out of a specific contract, but was the consequence of specific communications between the investment firm and the client. Indeed, investment advice means the provision of personal recommendations to a client in respect of one or more transactions relating to financial instruments.35

2. The Problem with the MiFID I Regime [6.16] The MiFID I regime was heralded as a great improvement in customer protection, even though in some Member States it actually limited the range of the suitability test, thereby reducing the protection of retail

investors.36 The problem is that when the suitability requirement does not attach to placing and other forms of distribution, ‘investment advice’ can still be mimicked by a seller who wants to promote a financial product as particularly well suited for the client. Notwithstanding the absence of an investment advice contract, firms can still place financial products with intense ad hoc personal recommendations attached. These personal recommendations can take the form of an aside where the seller rather casually ‘recommends’ the transaction to the customer, or be part of a systematic attempt to influence clients’ decisions without facing the burdens of the investment advice regime. [6.17] In civil litigation, the real nature of these personal recommendations as an expression of a ‘non-contractualized’ investment advice service, being part of ‘one-to-one’ communication between the investment firm’s employee or agent and the client is very difficult to prove. Securities watchdogs, instead, can employ a range of investigative instruments to spot recurrent red flags that signal such practices, amongst which the most important is probably an unreasonable concentration of clients’ investments in the firm’s products. Other red flags that have been mentioned in some decisions by the Italian authority are the massive reprofilings of clients, when they take place just before the issuance or distribution of a specific financial product to be sold to those clients, or the financing of the purchase through special loans.37 In any event, it is clear that in placing services the distinction between promotion, guidance, and advice creates many ambiguities, and that a private litigant can face great problems of burden of proof when he seeks to offer evidence that the investment firm was not only distributing but was also advising, and was doing so outside any contractually regulated framework. [6.18] Some prominent European cases have shown situations where, under the guise of distribution, retail investors as well as small and medium-sized businesses were allegedly subject to misselling that was largely grounded on biased and hidden advice. The Spanish case of ‘partecipaciones preferentes’ (‘preferred shares’) is probably the most famous one concerning retail investors,38 but many others followed. At the end of 2015, an alleged case of misselling hit the Italian political arena. Four small insolvent regional banks—Banca Etruria, Banca Marche,

CariFerrara and CariChieti—were rescued by the Italian government. The rescue deal forced shareholders and junior bondholders to incur losses, and it was thus discovered that a disproportionate amount of junior bonds were held by retail investors who, allegedly, were personally recommended to buy them and were holding them in very undiversified portfolios, where the concentration of risk on a single issuer was beyond any reasonable level.39 Similar cases followed in the same period with regard to clients of Banca Popolare di Vicenza and Veneto Banca.40 [6.19] Thus, the issue is whether MiFID II will substantially improve this grey area of European regulation.

IV. MiFID II 1. MiFID II on Information [6.20] The MiFID II regime poses a new set of information duties on financial advisers. When investment advice is provided, the investment firm must, in good time before providing investment advice, inform the client about three specific issues. These are: whether the advice is independent or not; whether it is based on broad or restricted analysis of financial instruments; and whether the firm will periodically assess the suitability of the financial instruments recommended, and inform the client of its findings.41

A. Independent Basis of Advice [6.21] Is the advice is provided on an independent basis or not? If the advice is labelled independent, pursuant to Article 24(7) the investment firm must ‘assess a sufficient range of financial instruments available on the market which must be sufficiently diverse with regard to their type and issuers or product providers to ensure that the client’s investment objectives can be suitably met’.42 More specifically, the range of financial instrument

must not be limited to financial instruments issued or provided by (i) the investment firm itself or by entities having close links with the investment firm; or (ii) other entities with which the investment firm has such close legal or economic relationships, such as contractual relationships, as to pose a risk of impairing the independent basis of the advice provided. In short, the financial instruments must be of a sufficiently various type and origin; and as to origin, the financial instruments must not exclusively come from related issuers or providers. ‘Sufficient’ is the key word here. [6.22] Where the advice is independent, the investment firm shall not accept and retain fees, commissions, or any monetary or non-monetary benefits paid or provided by any third party or a person acting on behalf of a third party in relation to the provision of the service to clients.43 MiFID II exclusively allows non-monetary benefits that are capable of enhancing the quality of service provided to a client and are of a scale and nature such that they could not be judged to impair compliance with the investment firm’s duty to act in the best interest of the client. These minor non-monetary benefits must be clearly disclosed to the client. [6.23] Accordingly, independence is not a prerequisite of a financial advisory service under MiFID II, as it was not under MiFID I. Independence is merely a MiFID II label. MiFID I did not mention independence. On the market, however, firms were actively promoting themselves as independent advisers and this self-promotion could only be challenged on the grounds of misleading marketing. Now MiFID II sets clear rules for promoting advisory services as ‘independent’. [6.24] Financial advice can be non-independent, with the adviser paid by third parties and assessing a limited range of financial instruments, which can be issued or provided for by related parties. The European legislator has therefore decided that a straightforward ban on commission from third parties might make financial advice unaffordable for investors who need it most and that it is better to have potentially conflicted advice than no advice at all, as the economic literature has suggested. Thus, the UK experience of the Retail Distribution Review (RDR)44 was not followed. Whereas the RDR bans all advisers offering either ‘independent’ or ‘restricted’ advice

from receiving commissions or other benefits from third parties, MiFID II applies the ban exclusively to advisers that pretend to be ‘independent.’

B. Broad or Restricted Analysis [6.25] Once it has specified whether its advice is independent or not, the investment firm must inform the client on ‘whether the advice is based on a broad or on a more restricted analysis of different types of financial instruments and, in particular, whether the range is limited to financial instruments issued or provided by entities having close links with the investment firm or any other legal or economic relationships, such as contractual relationships, so close as to pose a risk of impairing the independent basis of the advice provided’—Article 24(4)(a)(ii). [6.26] As mentioned, the investment firm which claims to be providing independent advice must assess a sufficient range of sufficiently diverse financial instruments. Here, instead, the information concerns the broadness of the analysis. Accordingly, an independent adviser might choose to inform the client that its analysis is based on a restricted but sufficiently wide and diverse base of financial instruments, which of course is not limited to those offered or issued by related parties. [6.27] Also in this respect, MiFID II does not follow the UK RDR distinction between independent and restricted advice. Under that distinction, independent advisers are able to consider and recommend all types of retail investment products, whereas restricted advisers work with a limited range of products or focus on a particular market, and do not cover all the types of retail investment able to meet the customer’s need and objectives. Under MiFID II, instead, an independent adviser can consider and recommend a sufficient range of investment products, whereas a nonindependent adviser might potentially consider and recommend a broad (i.e. wider) base of investment products and yet be non-independent because it receives commissions from some of the financial product providers or issuers.

C. Periodic Assessment of Suitability [6.28] Finally, the investment firm must inform the client as to whether it will provide the client with a periodic assessment of the suitability of the financial instruments recommended to that client. In various European jurisdictions investors have brought claims against investment firms and banks alleging that the financial intermediary had an ongoing duty to inform the client about the fall in value of the financial instruments that the institution had previously suggested to the client as a suitable investment. Financial intermediaries objected that they had no duty of this sort, owing to the absence of a specific contractual provision on the issue.45 Accordingly, MiFID II tackles the issue from an information perspective. The underlying idea is probably that this information could help reduce customer reluctance about paying for advice. Indeed this particular issue shows the limits of the market praxis where customers do not pay for advice. If advice is paid for by financial product providers and issuers through commissions and kickbacks, or if advice is paid for through brokerage fees or other cross-subsidies, the incentives for the adviser to continuously track the client’s portfolio and suggest modifications might be not aligned with the client’s interest.46

D. Other Information [6.29] Pursuant to Article 24(4)(c), the investment firm must also provide information on all costs and associated charges. In the case of investment advice, this information concerns the cost of advice and how the client may pay for it, and encompasses any third-party payments. This means that there must be full disclosure on how advice services that are not directly paid for by the client are financed.

2. Independent Advice and Other Investment Services A. Offering Independent and Non-Independent Advice at the Same Time

[6.30] There are some issues concerning the relation between independent advice and other services. The first is whether the financial intermediary can label its advice as independent with some clients and non-independent with others. In other words, is independent advice exclusive or not? The answer depends on whether the term ‘independent’ is attached to the service or the firm. [6.31] The wording of MiFID II is not straightforward. However, Article 24(4)(a) seems to attach the independent requirement to the service and not to the firm (‘when investment advice is provided, the investment firm must …’). The same seems to be true under Article 24(7) (‘Where an investment firm informs the client that investment advice is provided on an independent basis … that investment firm shall not accept and retain fees, commissions or any monetary or non-monetary benefits paid or provided by any third party or a person acting on behalf of a third party in relation to the provision of the service to clients’). [6.32] If this construction of MiFID II rules is correct, an investment firm could, for instance, opt to offer independent advice to clients that are willing to pay for the service, while offering non-independent advice to clients that are not prepared to face its costs or that cannot sustain those costs because of the amount of money involved. Accordingly, there is no ‘independent adviser’ in MiFID II. Instead, there is ‘independent advisory service’. The ‘independent’ suit can be taken or left from time to time, and from client to client.

B. Coupling Independent Advice and Distribution [6.33] A strictly related issue is whether ‘independent’ advice can be coupled with distribution. Can a firm be a distributor of proprietary or thirdparty products and, at the same time, be an independent adviser? Again, since MiFID II’s ‘independent’ regime appears to attach to the service and not to the investment service provider, it seems that the firm can offer independent advisory services to selected clients even though it offers distribution services or sells its own products to another class of clients. If this is true, the firm can promote itself on the market as an independent

consultant while in reality primarily being a product distributor. Since research shows that retail investors do not read the very long contracts that they sign, and that the nature of the service they receive is, in their perception, characterized by what the person they interact with tells them, as well as by advertisement, the ‘independent adviser’ label might be used to attract clients and then offer them distributed products, and in particular proprietary products. [6.34] If this is true, what has really changed? In such an ‘ambiguity scenario’ certain retail clients are not offered independent advice contracts, and therefore the information mentioned by Article 24(4)(a) is not even referred to. Yet, the investment firm can advertise itself as an ‘independent adviser’ because it actually offers independent advice to other clients, and this can be sufficient to make the former think that the promotional efforts of the salespeople or the guidance they give in relation to the investment products they are placing is advice and, even worse, is ‘independent advice’. In the end, the problem has not changed much. The core problem is how to help clients distinguish ‘advice’ from ‘guidance’ or from promotion.47

C. Tied Agents and Advice [6.35] An area of further potential confusion is advice by tied agents. Under Article 29 MiFID II, Member States shall allow an investment firm to appoint tied agents ‘for the purposes of promoting the services of the investment firm, soliciting business or receiving orders from clients or potential clients and transmitting them, placing financial instruments and providing advice in respect of such financial instruments and services offered by that investment firm’. Thus, tied agents provide advice concerning financial instruments. [6.36] Is this advice different from investment advice? Investment firms have tried to argue that it is, under MiFID I, thereby seeking to distribute products with advice attached without facing the burden of the suitability test. The attempt was generally rebutted, but confusion still reigns, as a recent amendment to the Italian law shows, which now refers to Italian tied

agents as ‘financial advisers allowed to make off-premises offers’ (consulenti finanziari abilitati all’offerta fuori sede). The change reflects the new reality of the financial market, where tied agents are no longer door-to-door salesmen, but are moving towards financial planning and true financial advisory work. However, there are still many financial networks where tied agents operate as salesmen of products distributed by their investment firms or, even worse, salesmen of proprietary products. Needless to say, to term these people ‘advisers’ increases confusion.

3. Contractual Obligations A. Suitability Assessment and Investment Firms’ Personnel [6.37] As was the case under MiFID I, when providing investment advice the investment firm has to obtain the necessary information regarding the client’s or potential client’s knowledge and experience in the investment field relevant to the specific type of product or service, that person’s financial situation including his ability to bear losses, and his investment objectives including his risk tolerance. This ‘passive’ information (from the client to the financial intermediary) must enable the investment firm to recommend to the client or potential client the investment services and financial instruments that are suitable for him and, in particular, are in accordance with his risk tolerance and ability to bear losses.48 [6.38] There are glaring misunderstandings about what a suitable financial instrument is,49 and financial intermediaries continue to have problems in coping with suitability assessments, as a recent FCA review of suitability of retail investment portfolios provided by wealth management and private banking shows.50 Amongst the reasons, there is certainly the fact that many employees and agents that work for financial intermediaries and give financial advice are not financially literate.51 On this point, MiFID II rightly takes a stand. It establishes that Member States shall require investment firms to ensure and demonstrate to competent authorities on request that natural persons giving investment advice or information about financial instruments, investment services, or ancillary services to clients on behalf

of the investment firm possess the necessary knowledge and competence to fulfil their Articles 24 and 25 MiFID II obligations. Pursuant to Article 25(1), Member States shall publish the criteria to be used for assessing such knowledge and competence.

B. Suitability Assessment and Product Governance [6.39] There is considerable confusion about the interaction between product governance and the suitability assessment. According to the product governance provisions, the manufacturer has to ensure that the financial instruments are designed to meet the needs of an identified target market of end clients within the relevant category of clients. Moreover, the manufacturer shall ensure that the strategy for distribution of the financial instruments is compatible with the identified target market, and that the investment firm takes reasonable steps to ensure that the financial instrument is distributed to the identified target market. [6.40] With regard to firms that offer or recommend financial instruments, they shall understand the financial instruments and assess the compatibility of the financial instruments with the needs of the clients to whom they provide investment services. In doing so, the investment firms shall also take into account the identified target market of the clients, and ensure that financial instruments are offered or recommended only when this is in the interest of the clients.52 [6.41] Accordingly, there is a double-check system. The manufacturer monitors the distribution system, and the offering or recommending firm takes into account the target market that has been identified by the manufacturer. For this purpose, the latter must have appropriate arrangements in place to obtain and understand the relevant information concerning the product-approval process, including the identified target market and the characteristics of the product they recommend.53 How does this system actually work in the area of advice? What happens if the adviser believes that a certain financial product fits a client’s investment portfolio, and recommends the purchase of that product even though the

product, in isolation, does not match the client’s target market? It should be clear that the suitability assessment concerns a financial product within an investment portfolio, and that the single financial product should not be considered in isolation under a suitability assessment. Diversification is the only free lunch in finance, and a large part of the suitability assessment is about portfolio composition and correct asset allocation and diversification.54 Accordingly, the suitability assessment does not overlap with product governance. They are two distinct steps in the protection of clients’ interests, and the suitability assessment should not be absorbed by the product governance process. The logical consequence is that an adviser can recommend a financial product that is not targeted to the client’s class identified by the manufacturer, when that product fits well with the clients’ portfolio. [6.42] Nevertheless, the interaction between product governance and suitability assessment will create problems. It is clear that the courts will have problems in understanding how an investment adviser might have recommended a product to a client that was not part of the product’s target market. Therefore, product governance risks creating a warped idea about what the suitability assessment really is. Of course, a related issue will be whether the manufacturer should consider the adviser as a distributor and intervene when it appears that the product is destined for the portfolios of investors who are outside the target market.

C. The Written Statement of Suitability [6.43] Pursuant to Article 25(6), the investment firm must specify in a written statement on suitability (WSS) how the advice given meets the preferences, needs and other characteristics of the retail client. Accordingly, Member States are free not to impose the WSS to non retail clients. The WSS, of course, increases the cost of advice. The amount of the increase depends on how detailed the WSS has to be. Drafted in a very short form, the cost can be limited but the role of the document as an instrument to understand ex post the basis of the recommendation becomes negligible. By contrast, if the WSS must be a clear document that offers a

thorough view of how the financial adviser formed his judgement, then the cost of drafting it can become significant. The question concerning retail clients’ openness to pay for this extra cost resurfaces as one of the major problems of the MiFID II regime.

4. Investment Advice and Portfolio Management [6.44] The last point is about the relationship between investment advice and portfolio management. Under both services, a suitability assessment is required. Portfolio management is something more than investment advice, because the investment firm, instead of recommending an investment product, directly manages an investment portfolio owned by the client. However, and rather surprisingly, under MiFID II advice is subject to stricter requirements in terms of information to the client. The portfolio manager is not required to inform the client on whether the management is provided on an independent basis, nor on whether the advice is based on a broad or on a more restricted analysis of different types of financial instruments and whether the range is limited to financial instruments issued or provided by entities having links so close as to pose a risk of impairing the independent basis of the service offered. If the portfolio manager asserts to be offering independent portfolio management, it is not obliged to respect the stringent rules attached to independent advice. Moreover, the portfolio manager has no duty to write any statement of suitability. The peculiar inconsistency of this regulatory divergence is striking.

V. Conclusions [6.45] The main problem of the current MiFID I regime concerning advice is ambiguity. Investment firms can too easily add personal recommendations to their distribution efforts, and then defend themselves against accusations of a breach of the suitability assessment by asserting that they were placing financial instruments and not advising on them. The shift from the appropriateness test of distributary services to the suitability assessment attached to personal recommendations is extremely difficult to

prove for an investor, in the absence of written evidence or an explicit investment advice contract. This issue is not addressed by MiFID II, which mainly addresses information issues, above all the ones concerning independent advice. Ample room for ambiguities continues to exist, because investment firms will be able to wear too many hats, as will their tied agents. This is the natural implication of not going the UK or Dutch way, imposing fee-only advice. MiFID II policy effort is to nudge investors, especially retail investors, towards fee-only advisory services. [6.46] The written statement of suitability is a good idea. However, if rigorously construed it will also be an expensive one, because it takes time to write a statement that one day might be used against you. This is one of the areas where the regulation of advice and portfolio management diverge, with the latter curiously treated more gently than the former. This might create room for some forms of regulatory arbitrage between the two services. [6.47] The interaction between product governance and the suitability assessment looks likely to generate confusion. If MiFID II delivers the message that the suitability assessment is a sub-exercise that takes place within the boundaries of product governance (i.e. a derivate of the product governance process), there will be less reason to look for advice, and MiFID II as a law signal might have an undesirable impact on financial education—at least on that of laywers and courts.

1

Luigi Guiso, Paola Sapienza, and Luigi Zingales, ‘Trusting the Stock Market’ (2008) The Journal of Finance 63, 2557. 2 Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse (Market Abuse Directive), OJ L 173/179, Recital no. 1: ‘An integrated and efficient financial market and stronger investor confidence requires market integrity. The smooth functioning of securities markets and public confidence in markets are prerequisites for economic growth and wealth.’ 3 Basic, Inc. v. Levinson, 485 US 224 (1988), ‘reliance on the integrity of the price set by the market’. 4 Luigi Guiso, Michael Haliassos, and Tullio Jappelli, Household Portfolios (Cambridge, MA: MIT Press, 2002).

5

John Y. Campbell, ‘Household Finance’ (2006) The Journal of Finance 61, 1553; Dimitris Christelis, Tullio Jappelli, and Mario Padula, ‘Cognitive Abilities and Portfolio Choice’ (2010) European Economic Review 54, 18. 6 Campbell (n. 5). 7 Maarten van Rooij, Annamaria Lusardi, and Rob Alessie, ‘Financial Literacy and Stock Market Participation’ (2011) Journal of Financial Economics 101, 449. 8 Hans-Martin Von Gaudecker, ‘How Does Household Portfolio Diversification Vary with Financial Literacy and Financial Advice?’ (2015) The Journal of Finance 70, 489. 9 W. Montford and R. E. Goldsmith, ‘How Gender and Financial Self-Efficacy Influence Investment Risk Taking’ (2016) International Journal of Consumer Studies 40, 101; B. M. Barber and T. Odean, ‘Boys Will Be Boys: Gender, Overconfidence, and Common Stock Investment’ (2001) Quarterly Journal of Economics 116, 261; cf. also M. Gentile, N. Linciano, and P. Soccorso, ‘Financial Advice Seeking, Financial Knowledge and Overconfidence’ (2016) Consob, Quaderni di finanza 83, 10–11. 10 Brad M. Barber and Terrance Odean, ‘Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors’ (2000) Journal of Finance 55, 773. 11 William N. Goetzmann and Alok Kumar, ‘Equity Portfolio Diversification’ (2008) Review of Finance 12, 433. 12 Laurent E. Calvet, John Y. Campbell, and Paolo Sodini, ‘Down or Out: Assessing the Welfare Costs of Household Investment Mistakes’ (2007) Journal of Political Economy 115, 707. 13 Laurent E. Calvet, John Y. Campbell, and Paolo Sodini, ‘Fight or Flight? Portfolio Rebalancing by Individual Investors’ (2009) Quarterly Journal of Economics 124, 301; Tom Y. Chang, David H. Solomon, and Mark M. Westerfield, ‘Looking for Someone to Blame: Delegation, Cognitive Dissonance, and the Disposition Effect’ (2016) Journal of Finance 71, 267. 14 Andrea Frazzini, ‘The Disposition Effect and Underreaction to News’ (2006) Journal of Finance 61, 2017; Camelia M. Kuhnen, ‘Asymmetric Learning from Financial Information’ (2015) Journal of Finance 70, 2029. 15 Brad M. Barber et al., ‘Just How Much Do Individual Investors Lose by Trading?’ (2009) Review of Financial Studies 22, 609. 16 H.-M. von Gaudecker, ‘How Does Household Portfolio Diversification Vary with Financial Literacy and Financial Advice?’ (2015) The Journal of Finance 70, 489. 17 Nick Chater, Steffen Huck, and Roman Inderst, ‘Consumer Decision-Making in Retail Investment Services: A Behavioural Economics Perspective’ (2010) Final Report 22, para. 590. 18 Cf. Riccardo Calcagno and Chiara Monticone, ‘Financial Literacy and the Demand for Financial Advice’ (2015) Journal of Banking & Finance 50, 363 (complements); M. Gentile, N. Linciano, and P. Soccorso (n. 9), 30 ff (complements); H.-M. von Gaudecker (n. 16), 500 (neither complements nor substitutes).

19

Commission Staff Working Document, Economic Analysis Accompanying the document Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions Action Plan on Building a Capital Markets Union (2015) 64. 20 ICPSR Eurobarometer 60.2, 21. 21 In an experimental setting, Chater, Huck, and Inderst (n. 17) found that a significant minority of retail investors are disproportionately averse to paying an up-front fee for advice: see in particular at para. 603. 22 See paragraph 6.37. 23 Chater, Huck, and Inderst (n. 17) 131 ff. 24 Roman Inderst and Marco Ottaviani, ‘Competition through Commissions and Kickbacks’ (2012) American Economic Review 102, 780. 25 Patrick Bolton, Xavier Freixas, and Joel Shapiro, ‘Conflicts of Interest, Information Provision, and Competition in the Financial Services Industry’ (2007) Journal of Financial Economics 85, 297; Roman Inderst and Marco Ottaviani, ‘Misselling through Agents’ (2009) American Economic Review 99, 883. 26 Inderst and Ottaviani (n. 24), 782. Cf. also Andreas Hackethal, Michael Haliassos, and Tullio Jappelli, ‘Financial Advisors: A Case of Babysitters?’ (2012) Journal of Banking & Finance 36, 509. 27 Calcagno and Monticone (n. 18). 28 Dimitris Georgarakos and Roman Inderst, ‘Financial Advice and Stock Market Participation’ (2014). Available at SSRN 1641302. 29 Utpal Bhattacharya et al., ‘Is Unbiased Financial Advice to Retail Investors Sufficient? Answers from a Large Field Study’ (2012) Review of Financial Studies 25, 975. 30 Financial Times, ‘UK banks set to launch “robo-advisers”’, 22 January 2016; FINRA, ‘Report on Digital Investment Advice’ 15 March 2016, available at . 31 See paragraph 6.38. 32 But see Inderst and Ottaviani (n. 25). 33 Commission Directive 2006/73/EC of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive, OJ L 241/26 (MiFID I Implementing Directive). 34 Article 39, MiFID I Implementing Directive (2006/73/EC). 35 This is the definition offered by Article 4(4) MiFID II, which contains minor amendments to the one contained in MiFID I. 36 Italy was a case in point. Up to MiFID I, Italian courts considered the suitability test to be applicable to any investment service and, in particular, to distributional services such as placing. 37 Consob, decision no. 18615/2013, (2013) Bollettino n. 9.1 (Cassa di Risparmio di Savigliano); decision no. 19283/2015, (2015) Bollettino n. 9.1 (Poste Italiane); decision no.

19368/2015 (2016) Bollettino n. 1.2 (Banca Popolare); decision no. 19497/2016 (2016) Bollettino n. 1.2 (Cassa di Risparmio di Bolzano). 38 Fernando Zunzunegui, ‘Mis-selling of Preferred Shares to Spanish Retail Clients’ (2014) Journal of International Banking Law and Regulation 174; Securities and Markets Stakeholder Group, ‘Advice to ESMA: Investor Protection Aspects of the Consultation Paper on MiFID II and MiFIR’, 8 August 2014, available at . 39 J. Brunsden and P. Jenkins, ‘Bank Rescue: Putting Bondholders on the Hook’, Financial Times, 4 January 2016. 40 With regard to small and medium enterprises, the mis-selling cases have mainly involved derivatives, and more specifically swaps. However, they are mainly pre-MiFID I cases. 41 See MiFID II, Article 24(4)(a). 42 Article 24(7)(a). 43 Article 24(7)(b). 44 This is a set of new rules enforced since 2013 in the UK, which is currently under review precisely because it apparently enlarged the ‘advice gap’: see HM Treasury—FCA, ‘Financial Advice Market Review’, March 2016, available at . 45 In Italy courts have not offered a uniform answer to the problem: cf. Tribunal of Prato, 13 June 2015 (the investment firm had a duty to inform the client about the collapse in value of Greek bonds); Tribunal of Turin, 20 November 2012 (no ongoing duty on the investment firm after the brokerage service has been rendered). 46 As far as I know, economic research has not analysed this issue. 47 Niamh Moloney, UE Securities and Financial Markets Regulation (Oxford: Oxford University Press, 2014), p. 805 (‘the reform … sits uneasily in an EU market dominated by sales of proprietary products’). 48 Whereas the concept of suitability of a financial instrument or a financial portfolio is understood, at least in its general terms, it is more difficult to understand what it is referred to with regard to a suitable investment service. Here MiFID II, as well as MiFID I, mentions something that it is outside its own boundaries: whereas ‘investment advice’ is defined as the provision of personal recommendations to a client in respect of one or more transactions relating to financial instruments, both the old and the new Directives cover recommendations of ‘services’, which differ conceptually from recommendations of ‘financial instruments’. 49 It is even more so, of course, with regard to bundled products, to which the suitability test applies as well—assessing them as an overall bundled package: Article 24(11) MiFID II. 50 FCA, Wealth management firms and private banks. Suitability of investment portfolios. TR15/12 (2015). 51 Curiously, this issue is not covered by the blossoming economic literature on financial literacy, where the true financial literacy of banks’ and investment firms’ employees is

never investigated. 52 Article 16(3). 53 Recital 71. 54 See ESMA, ‘Guidelines on certain aspects of the MiFID suitability requirements’ (2012), para. 60: ‘A firm should establish policies and procedures which enable it to ensure inter alia that: (a) the advice and portfolio management services provided to the client take account of an appropriate degree of risk diversification.’

7 CONFLICTS OF INTEREST Stefan Grundmann and Philipp Hacker*

I. Foundations of the Regime 1. The Ambit of the Topic 2. Theoretical Approaches for Regulating and Limiting Conflicts of Interests 3. The Range of Regulatory Approaches II. The Regime by Single Stages and Examples 1. Organizational Requirements Minimizing Conflicts of Interest 2. Conflicts of Interest in the Formation of Investment Services Contracts 3. In Particular: Conflicts of Interest and the Regime on Fees 4. Conflicts of Interest in the Execution of Investment Services Contracts (‘Best Execution’) 5. Conflicts of Interest in the Case of Linked Contracts 6. New Strategies: the Future of the Regulation of Conflicts of Interest III. Conclusion

I. Foundations of the Regime [7.01] We shall first provide an overview of the foundations of the regulation of conflicts of interest in investment services both with respect to

theory and to positive law.1 First, however, we must specify which conflicts of interest we are interested in.

1. The Ambit of the Topic [7.02] Conflicts of interest span a wide area in the law. We shall therefore first demonstrate what specific conflicts of interest this chapter will focus on and then argue that the specific types of problems of these conflicts of interest do not sit comfortably within the general framework of the objectives of the MiFID I and II regulations.

A. Which Conflicts of Interests? [7.03] Conflicts of interest abound in many settings in private law. Real estate brokers recommend flats to incoming residents, directors manage companies for the shareholders, rating agencies are paid by those whose papers they are evaluating, and car dealers resell used cars. In all of these cases, the agents are subject to conflicts of interest that decisively impact on their strategies and actions. In a first, very broad approach, conflicts of interest typically arise under two conditions. First, an agent is entrusted to look after the assets, health, or other interests of another person.2 Second, her utility-maximizing strategy does not maximize the benefit of the other party, that is, the agent has an incentive to act opportunistically or to otherwise not fully maximize the benefit of the other party.3 While some conflicts of interest have been regulated for a long time (corporate fiduciary duties) or are coming under increasing regulatory scrutiny (rating agencies), many conflicts of interest in general contract law remain fairly untouched by mandatory or even default law. They form part and parcel of the bargaining process which is often entrusted to work as a balancing device between the parties: as a matter of positive law, both parties are free, within certain limits, to pursue their own interests without regard to the interest of the other party or even disclosing conflicts of interest. Therefore, the core question is: What distinguishes a conflict of interest worth regulating from one whose solution is left to bargaining? This difference can also be coined as one of terminology, for instance contrasting opposed interests from a

conflict of interest in the strict sense.4 Terminology, however, exists for the sake of clarity mainly; it provides a term for a distinction for which there are plausible or even mandatory substantive reasons. The substantive reasons are what really matters. In a contract relationship, these have to be found primarily in the implied terms agreed upon or in the expectations and interests of the parties (and the client–intermediary relationship is contractual in essence even if it is also subject to public regulation and supervision). In the light of this starting point, the main difference would seem to be whether the situation is one where the parties involved know that they are bargaining about a certain solution or whether the decision on the solution is left by one party to the other—with the expectation that that party will then use this power for the benefit of the first.5 From a normative perspective, this explanation/expectation becomes even more convincing when this power to influence the assets/rights/interests of the other is handed over by the counterparty to the decision-maker without the counterparty receiving any compensation in return and without any investment made by the decision-maker.6 In this case, the decision-maker can be seen to have received a position not only with an implied term to see solely after the interests of the other party, but also without her own efforts legitimizing her interest to be taken into account. Thus, in such a case, the rule that the decision-maker should take solely the interests of the other party as the guideline to her action can be justified both by (implied) agreement and by objective reasons. The additional criterion—entrusting a position of influence plus doing so without compensation—also helps to distinguish borderline cases. If an enterprise drafts standard contract terms, it certainly has a position of power to influence the other parties’ interests; but it does not receive it for free, as the overall package of conditions (with the price) cannot be split into parts without distorting the balance of price and duties. Imposing a duty to take only the other party’s interest as a guideline would force the enterprise to forego the possibility to calibrate its own duties when making an offer in return for a certain price. The absurd result would be that whoever drafts standard contract terms always has to draft them taking the clients’ interests as sole guideline for such drafting. Since the regulation in standard contract terms relates to duties costly to comply with, drafting in one way or another is not without costs and hence such an absurd conclusion is not warranted.

[7.04] One can frame our focus—on conflicts of interest in the narrow, regulated sense—not only by looking to the case patterns, but also by defining the legal consequences attached. We focus on those situations in which the party subject to a fiduciary duty does indeed have to take the interests of the other party as the sole guideline (strict fiduciary duty) and— moreover—where this duty is even accompanied by some mechanism which is aimed at rendering compliance with this substantive guideline even more efficient. [7.05] While this situation and its legal consequence are, of course, not limited to financial services, that is, while they do exist also in other areas of the law—clearly in company law, but also in distribution chains, for instance with respect to sales’ agents7—both the situation and legal consequence are particularly important for the area of financial services; moreover, the area is one by which this situation and its legal consequence have been shaped in a particularly meaningful way. They are so important here that one could even say that they are all-pervasive and that financial services are unique in the mode and consistency with which conflicts of interest have been regulated. In this sense, the regulatory and theoretical landscape looks quite different in the financial services field. [7.06] In this domain, two, perhaps even three aspects are striking. First, as has already been stated, the interests of one party strictly prevail in the regulatory setting—those of the client—and this is or can be justified by the fact that an (entrusted) position of influence has been conveyed (without compensation). Second, this general rule is reinforced by specific, in large part mandatory provisions about on one hand organizational features intended to minimize conflicts of interest and on the other hand the disclosure of conflicts of interest. This immediately raises the question of why conflicts of interest are treated differently or more intensively in the area of financial services than in other parts of contract law or private law more generally, or at least with more emphasis on a strict rule to have clients’ interests prevail. A third element may be seen in the fact that the duty to take the client’s interest as a guideline seems to be strong even if it is difficult to see conveyance of an entrusted position. In other words, there seems to be an overspill from the many situations in which there is indeed conveyance of an entrusted position to other situations in the bank–client

relationship. Why, for instance, is there a right to banking secrecy in a very general sense, and also in the credit business (where solely the client or her interests decide whether the bank may pass on the details of the relationship), and why does there also seem to be a duty to disclose conflicts of interests in situations where a loan agreement is bargained for? However, as the cases regulated by MiFID I and II fall into the core area of entrusted positions being conveyed, the questions of overspill are not paramount in the following, for our purpose. Section I.2 on theoretical approaches will provide an account of the characteristics that distinguish these cases of uncompensated conveyance of entrusted positions from others and legitimize regulation. [7.07] There is a second and often neglected difference within conflicts of interest, but this time not concerning the field in which they arise but rather the parties whose interests are affected. On one hand, we can distinguish between the interests of the client and the interest of the financial services provider which are in conflict. On the other hand, however, the provider may also be conflicted between the interests of different clients that he simultaneously advises.8 Both types are mentioned in Article 18(1) MiFID I and Article 23(1) MiFID II. The latter case, however, has not received much attention in the literature9 and the regulatory provisions; we shall thus focus on the difference between the provider’s and the client’s interest, and only briefly address the other situation: it is important mainly when clients really ‘compete’, namely in questions of ‘execution’, for instance to know which order should take preference, and what ‘best execution’ is more generally. In other areas of the law, such as directors’ duty to respect the share of minority shareholders according to strict equal treatment standards, this problem is much more important.10

B. No Consistent System of Goals in the Exposition of MiFID I and II [7.08] If one asks about the importance of conflicts of interest and their regulation in the ambit of application of MiFID I and II, the overall image seems ambivalent.

[7.09] On one hand one does not only find both specific provisions for the organization of the provider of investment services aimed at minimizing conflicts of interests (Article 13(3) MiFID I and Article 16(3) MiFID II) and contract rules (namely disclosure rules, Article 18 et seq. MiFID I and Article 23 et seq. MiFID II), but there is also one specific rule on the overall problem of conflicts of interest (Article 18 MiFID I and Article 23 MiFID II). Moreover, two core problems of conflicts of interest, the regime on ‘best execution’ (Article 21 MiFID I and Article 27 MiFID I) and the one on inducements (Article 24(4) through 24(9) MiFID II), have aroused enormous interest, both in theory and in the courts, and/or were at the core of legislative development in the reform (with, again, one, now two, specific norms only on this). Finally, there is also a special provision allowing for particular Level 2 and Level 3 legislation, especially with respect to the key aspects of conflicts of interest (definition, main organizational means, main kinds of disclosure) (Article 18(3) MiFID I and Article 23(4) MiFID II). All this would seem to speak for an absolutely core role of the regulation of conflicts of interest in the overall regime of MiFID I and II. [7.10] On the other hand, however, despite their importance, the aims of the regime on conflicts of interest are not systematically dealt with or related to the overarching regulatory objectives in MiFID I and II. This is partly fuelled by the confusion about the aims of MiFID I and II themselves. The most often and most notably stated purpose of MiFID I and II is to protect investors;11 in some parts, this is complemented by the— necessarily interlinked—aim of safeguarding the integrity of capital markets (explicitly in this sense Article 9(3) MiFID II as the core rule defining the board’s duties, see also Article 31 et seq. MiFID II).12 Since the regulation was also passed in an effort to harmonize the conditions for offering investment services in the European single market, the two aforementioned goals are joined by a third: the creation of one (harmonized) European Financial Market, which some even see as the main goal, really,13 and which could probably be supported by Recital 7 MiFID II,14 but is contradicted by what is arguably the most meaningful Recital, as it puts all goals into one context: Recital 164 states that the objective of this Directive [is to create] … an integrated financial market in which investors are effectively protected and the efficiency and

integrity of the overall market are safeguarded, [and this] requires the establishment of common regulatory requirements … so as to prevent opacity or disruption on one market from undermining the efficient operation of the Union financial system as a whole which cannot be sufficiently achieved by the Member States but can rather [by the] … European Union. This Recital can probably be seen as a general one, which also states the set of goals with respect to conflicts of interest regulation—in fact as the most general one—and it clearly establishes the goal of harmonization (only) as instrumental to achieving (i) ‘investor protection’ and (ii) ‘efficiency and integrity of the overall market’. [7.11] Conflicts of interests are mentioned a few times in the MiFID I and II regime, but the relationship to the three aims is not clarified, particularly not with respect to investor protection. In MiFID I, only Recital 29 speaks about conflicts of interest and explains the need of regulating them by the proliferation of products. In MiFID II, this is taken up in Recital 56 where now—at last and at least—the link to investor protection is also specified.15 Therefore, it seems clear that mitigation of conflicts of interests is central to investor protection (a point apparently confirmed in Article 9(3) MiFID II already mentioned, which also mentions market integrity); this goal is chiefly pursued through business conduct rules. However, the main provisions regulating conflicts of interest in MiFID I and II, Articles 18 and 23 respectively, do not form part of the section on investor protection which comprises the business conduct rules. [7.12] The prevailing view for capital market law more generally would seem to be that good investor protection is also functional for furthering market efficiency, even being one of the main tools for doing so. Therefore, it seems fair to start from the assumption that conflicts of interest are combated with the aims of investor protection and market integrity and also market efficiency—at least in the core cases of conflicts of interest. Here, the argument is accepted more than in any other area of regulation that sound protection of individual investors’ interests also helps the market to function.16 This, however, is only a rough description. First, the link between conflicts of interest and market integrity and efficiency is not

explicitly made, even in MiFID II. Second, there is no specification of how investor protection or market integrity should be conceived in more detail, for instance whether investor protection should be directed towards rationally acting investors or those with bounded rationality, or both (in which combination?), and for instance whether the goal of market integrity should be understood as requiring ‘fair’ solutions in each single case or as only defying more general distrust in markets. [7.13] As the goals are not described in sufficient detail (namely in the Recitals), the discussion has to be based on a survey of the main discussed and possible approaches, refining the goals of investor protection and market integrity/efficiency (see Section I.2)—followed by a short survey of the main rules and their main thrust (see Section I.3)—before we enter into a stage-by-stage analysis (Section II below).

2. Theoretical Approaches for Regulating and Limiting Conflicts of Interests [7.14] Against this partly dissatisfying state of positive law, it is all the more important to consider the theoretical foundations for the regulation of conflicts of interest. Three different, yet interlinking, approaches can be uncovered: an appeal to fairness, the pursuit of efficiency, and the acknowledgment of cognitive errors and limitations.

A. Fairness: Entrusting Positions of Influence (without Compensation) [7.15] Fairness is a key principle both in economics17 and in the law.18 Ever since Aristotle, the regulation of exchange through contract has been closely linked to notions of justice and fairness.19 In modern times this was highlighted, in different ways and with different assumptions, by the leading political philosophers of the twentieth century, John Rawls20 and Jürgen Habermas.21

[7.16] In contract law, fairness standards have mainly been derived (i) from what has been called ‘justice of consensus’ and/or (ii) on the basis of objective considerations. With ‘justice of consensus’, all those considerations are addressed which argue that agreement, at least if duly formed (information, no overreaching, etc.), can be seen as sufficient source of legitimacy—certainly when the parties’ will and interests are seen as the ultimate yardstick, but also when contract law is seen to be inspired first and above all by the interests of society at large.22 Among the explanations already given when defining which conflicts of interest are discussed in this chapter, the first criterion of differentiation would seem to go particularly into this direction. If regulation of conflicts of interest targets situations where a position of influence has been entrusted from one party to another and if in this situation a rule is meant to apply which requires the latter to act solely in the interest of the former, the main argument would seem to be this: The position which allows one party to influence the affairs of the other has been entrusted to the former on the condition—at least on the implied condition—that the former has to exercise such power solely to the benefit of the latter.23 Thus, the rule envisaged is justified as being agreed upon and hence ‘just’, at least in the consensual notion. The second explanation given above for such a rule (when defining which conflicts of interest are discussed in this chapter) points to the fact that the situation can also be defined as one where the party to whom such power has been entrusted has not given anything in return for it and does not make sacrifices (‘uncompensated’ entrustment). Indeed, the argument is that this criterion has to be satisfied in addition to the one of entrusting powers/influence if one wants to justify the rule that solely the interest of the client has to be used as a guideline of action. It can neither be assumed nor would it be obviously ‘fair’ that a party who receives a power to influence needs to act solely in the interest of the other party if this implies a loss, that is, if the party entrusted would have to give in assets or funds which she owned before acting in this way.24 [7.17] It should be noted that the fairness concerns apply equally to all cases of conflicts of interest. This is not the case or is at least less so with the two remaining theoretical approaches, efficiency and cognitive limitations (which, however, are perhaps more telling on the question of

whether financial services providers are different from other segments in this respect).

B. Efficiency: Avoiding Hidden Gains (Fundamental Lack of Transparency) [7.18] The second line of argument is about efficiency and the effect of hidden gains. In the transactions governed by MiFID II, there are many where the outcome of the action—for instance a piece of advice given— could be compared with so many other outcomes possible that this renders any comparison with outcomes which would have been achieved without any conflict of interest influencing the outcome practically impossible. One such example is the advice given on the—theoretically unlimited range of —investment instruments. In this case, hidden gains derived from the use of the position of power lead to a fundamental lack of transparency in the offers made for the client. This is so with respect to the cost of the investment advice (as the adviser earns the fees plus the hidden gains) and with respect to the value of the investment advice to the client, that is, how high the chances are that the client would have been offered a better investment opportunity without such conflicts of interest existing.25 In this case, obliging the fiduciary to act solely in the interest of the client is aimed at avoiding such fundamental opacity—and the well-known adverse selection problems associated with it.26 As we will see, even such an obligation as this is not seen as being sufficient; rather a procedural rule preventing conflicts of interest is added to the legal framework. [7.19] There are also, however, situations where hidden gains may occur, but end prices can be compared—be they the result of a better price which the investment services provider could get, but to which he added a hidden gain, or of a less favourable price to which then, however, no additional hidden gains would have been added. Such a situation is possible once the investment instrument is chosen and it comes to ‘best’ execution. In this situation, the exercise of discretion by the investment services provider may well be much better observable for the client, and the end price paid may not differ for him. In these situations, the conflict of interests may have repercussions at other levels—for instance with respect to the choice of

trading venue, but need not lead to an adverse selection problem for the client.27 Thus, the argument from a hidden gains and adverse selection (efficiency) perspective might imply that divergent strategies of dealing with conflicts of interest at different stages of the investment relationship are indeed at least plausible. [7.20] This leads to yet another, more segment-specific consideration of conflicts of interest with respect to efficiency concerns: the efficiency argument seems particularly strong in the case of financial services. While in other areas of the law (renting a house, buying a car) clients can at least theoretically be entrusted to take care of their own interests, this seems close to impossible in investment contracts due to the complexity of the investment products and the resulting size of the information asymmetry between the service provider and the client. This speaks in favour of segment-specific concerns. Thus, Enriques for instance suggests that financial services are special because of their vague standards and the often small amounts of damages awarded in cases of conflicts of interest, leading to under-enforcement. Moreover, he sees reputational negative externalities for the entire industry that arise from ‘scandals’.28 In our view, some doubt is cast upon at least the second point by the fact that poor—or biased— investment advice often affects good parts of a person’s wealth or savings (for old age, for estate planning, etc.) and that, at least in Germany, the large majority of investment services cases centre on allegedly undisclosed conflicts of interest and the large majority of all cases based on violation of conflicts of interest rules resides in the area of financial services. We even think that MiFID II transactions are particularly prone to leading to existential risk and to potentially ruinous transactions in a relatively high ratio of cases. The argument that a particular design of conflicts of interest rules can be traced back to or even justified by scandals only seems convincing from a post-crisis perspective. However, most of the MiFID II reform had already been discussed and even designed before 2007/08. Therefore, the vagueness argument would seem to be the most credible. As has been shown above, this affects only certain cases of conflicts of interest in the financial services area, not really all of these cases. Moreover, this is a phenomenon which is indeed particularly important in financial services, but is certainly not limited to it. In terms of information theory, the problem of vagueness is that of quality of the advice (service) being a credence

rather than an experience or even an inspection good.29 This would seem to imply that since quality is so difficult to observe for clients, a procedural rule—to avoid conflicts of interest as much as possible and to disclose them when they do occur—seems needed more here than in other segments.

C. Cognitive Limitations: Mitigating Processes of Suboptimal Use of Information [7.21] A third aspect has received increasing attention in the wake of the financial crisis: the cognitive limitations of investors. This argument draws heavily on research in cognitive psychology and behavioural economics. In recent decades, these areas have engendered a shift of paradigms in economics: homo economicus has increasingly come under attack and is perhaps even in part being replaced by a more realistic, more empirical figure.30 The behavioural approach to economics and the law has uncovered two sets of specific limitations that are of interest here: cognitive capacity limits and bounded rationality.31 The first notion refers to the limited number of pieces of information that the working memory of the human brain can simultaneously process. The concept is simple: Once all channels of information processing are fully used, any additional information provokes information overload.32 It leads not only to stress and scrambling but to a deterioration in the quality of decisions.33 In a classic article, George Miller suggested back in 1956 that the maximum number of pieces that can be processed simultaneously (so-called chunks) is roughly seven.34 It turns out that this number varies greatly between agents and tasks. However, contemporary research shows that the limit will often already be reached at four chunks.35 This has deep implications for traditional regulatory strategies such as disclosure. If too many items are disclosed at a time, they might not reach the working memory. Moreover, limited attention may result in clients ignoring vital information in the first place.36 [7.22] Furthermore, even if the amount of information does remain within cognitive capacity limits, the correct processing of the information may be subject to cognitive biases. In recent decades behavioural economics research has identified a whole range of different biases that testify to what

has come to be called ‘bounded rationality’, borrowing a term from Herbert Simon.37 Bounded rationality may interact with the disclosure of conflicts of interest in several ways.38 First, optimism bias may lead a client to believe that the negative impact of conflicts of interest will not materialize in her case.39 Second, a confirmation bias40 may operate if the client is inclined to have a positive view of the service provider: she will filter information given by the provider in a way that reflects her benevolent attitude toward the provider, thus downplaying her perception of the potential impact of conflicts of interest. This is particularly pertinent in the case of investment advice since trust in advisers is generally quite high: More than 80 per cent of investors in the EU used advisers in the retail investment market,41 and naïve trust is generic,42 with more than 80 per cent of the advisees generally trusting their advisers.43 Consequently, confirmation bias can be considered to be widespread among advisees. Perseverance—a similar phenomenon denoting the continued holding of beliefs even after the rational basis for these beliefs has been removed— may have a similarly frustrating effect on the effectiveness of disclosure of conflicts of interest.44 [7.23] However, as in all discussions of bounded rationality, a note of caution is in order.45 On one hand, other biases may cut against the ones just mentioned. For example, if the client believes the chance of being affected by unwanted behaviour by the service provider is quite small, probability weighting may lead to an overestimation of this probability: one of the central tenets of Kahneman and Tversky’s Prospect Theory is that very small probabilities tend to be overweighted.46 Furthermore, the availability heuristic may work in either direction depending on whether stories of benevolent or malevolent providers are mentally available to the client.47 On the other hand, it is crucial to note that not all clients will be acting in a boundedly rational way. Rather, they will act along a continuum ranging from full to bounded rationality. Regulatory theory must thus take uncertainty inherent in empirical research into account.48 [7.24] Despite all uncertainty, however, it remains clear that cognitive limitations must be dealt with thoroughly in financial markets. This is not only a consequence of the financial crisis of 2008 but also reflects the fairly

advanced state of knowledge on empirical behaviour of investors. In fact it is quite ironic to see that behavioural economics has found most traction within the economic discipline in the area which was formerly a stronghold of rationality: finance.49 Richard Thaler traces the rise of behavioural finance to the exceptional amount of data available in financial markets and hence the possibility to run empirical studies easily.50 The truth of this claim notwithstanding, the prevalence of behavioural finance—coupled with the often existential impact of unplanned consequences of financial contracts on clients—perhaps provides an explanation of why conflicts of interest, which may form an unholy alliance with biased perceptions, are regulated so tightly within the finance industry. It is this range of regulatory approaches that we turn to now.

3. The Range of Regulatory Approaches [7.25] To address these issues, that is, to safeguard fairness, to enhance efficiency, and to mitigate cognitive limitations, the law has adopted a range of regulatory approaches. The two most important ones are organizational rules on one hand and contract or disclosure rules on the other (see below Section I.3.A).51 However, if analysed more carefully and within this twofold main thrust, MiFID II seems to follow a considerable range of quite heterogeneous regulatory strategies for different questions and different stages (see below Section I.3.B.). In all the explanations in the following, while primarily commenting the MiFID II regulatory body, we also investigate both the theoretical underpinning and—at least where appropriate—the alternative strategies possible.

A. Organizational Rules and Contract/Disclosure Rules [7.26] The division between organizational rules and contract or disclosure rules is reflected in the provisions of MiFID I and II. Article 13(3) MiFID I holds that ‘[a]n investment firm shall maintain and operate effective organizational and administrative arrangements with a view to taking all reasonable steps designed to prevent conflicts of interest […] from adversely affecting the interests of its clients.’ This is reiterated in

Article 16(3) MiFID II. To ensure effective monitoring of the competent authorities, Article 13(6) MiFID I and Article 16(6) MiFID II impose a duty of record keeping. Disclosure rules can be found in Article 18 MiFID I and Article 23 MiFID II. The first paragraph of each provision enjoins the provider to identify and manage all possible conflicts of interest between himself and the client or between different clients. In fact, this two-step approach still largely relates to organizational measures and not yet to contractual or disclosure rules proper. (Only) the second paragraph, adding a third step, truly embarks on the route of the contractual relationship and contains a duty to disclose conflicts of interest whenever organizational matters are insufficient to protect the client. [7.27] Both types of provisions, organizational and disclosure rules, are designed to prevent conflicts of interest in the first place (organizational) and, where not possible, to at least prevent damage to the investor by enabling an informed decision (disclosure). Both features are also an expression of the general duty of the provider to safeguard the interests of the client (see Chapter 4). However, this rule was deemed too vague and had to be qualified by provisions on the conflicts of interest mandating specific prospective measures in the organizational and contractual domain. Thus, the regulation is not only based on the assumption that the services provider has to be bound to strictly follow the client’s interests as the sole guideline, but also requires that a rule of precaution be added: All three steps are meant to make the provider (i) aware of a conflict of interest and document it; then (ii) minimize the conflict, that is, avoid as far as possible situations where it is more likely that she might violate her obligation to strictly observe the client’s interests; and (iii) at least call on the client to be alert that this risk is particularly high in a conflict of interests situation. Thus, the regime on conflicts of interest constitutes a preventive measure which is aimed at a better compliance with the obligation to strictly observe the client’s interests. [7.28] The first question that arises from this dichotomy is the relationship of the two types of prospective measures to one another. The answer of positive law is that organizational measures need to be pushed as far as possible in order to ensure the maximum mitigation of conflicts of interest; it is not sufficient simply to disclose and to abstain from the best

organizational prevention structures. This becomes apparent in the formulation of Article 18(2) MiFID I and Article 23(2) MiFID II where disclosure is envisaged as a subsidiary tool to which the company may only have recourse if organizational measures prove insufficient. The same can be deduced from the unconditional nature of the organizational duties in Article 13(3) MiFID I and Article 16(3) MiFID II. While being unconditional, the organizational duties only go as far as is ‘feasible’. The exact amount of effort required will be investigated further below.52 It may be noted that the institutionalization of the duty to safeguard the clients’ interests by organizational means is a typical instrument in banking and financial services supervision whenever the mere duty to act in the interest of the client is deemed insufficient. [7.29] Second, if the disclosure duty is triggered, the question of the reach of disclosure arises. More particularly, it is unclear from the provisions in MiFID I and II whether the provider only has to warn that a conflict of interest exists or whether she also has to explain the specific conflict she is subject to. This too will be investigated in detail below.53

B. Heterogeneous Approaches in the Different Stages of the Client Relationship [7.30] Disclosure rules are most often contract rules. However, the new MiFID II regime in particular operates with a whole range of different regulatory approaches that go well beyond such a contract standards approach to deal with conflicts of interest. More specifically, three further types of approaches can be discerned: a market structure approach, a strict investors’ interest approach, and an unfair trade practices/competition law approach. These different methods correspond to the regulation of different stages of the transaction. The formation of the contract is subject to the contract standards and the market structure approaches, the execution to a strict investors’ interest approach, and linked contracts are governed by an unfair trade practices/competition law approach. In the remainder of this chapter, we will describe the details of the different approaches and relate them to the corresponding stages of the transaction.

II. The Regime by Single Stages and Examples [7.31] The different stages and corresponding duties contemplated in the next sections are as follows: organizational duties; formation of the investment services contract and disclosure; remuneration regimes in particular; execution and multiplicity of contracts, and the duty of best execution; and linked contracts. Thus, while the organizational duties are meant to minimize and even exclude conflicts of interest and therefore take precedence, the stages concerning the contractual relationship are much more manifold and rich in regulatory ideas. The section concludes by outlining new strategies with regard to the future of the regulation of conflicts of interest.

1. Organizational Requirements Minimizing Conflicts of Interest [7.32] One of the specific components of the regulatory regime on conflicts of interest in the field of financial services is the precautionary rules that mandate specific organizational features within the company of the provider.

A. A Company Structure Solution [7.33] The regulation of the organization of investment firms is not restricted to the mitigation of conflict of interests. Rather, the MiFID regime mandates effective organizational structures for general compliance (Article 13(2) MiFID I, Article 16(2) MiFID II), continuity of service (Article 13(4) MiFID I, Article 16(4) MiFID II), the avoidance of undue operational risk in the outsourcing of functions (Article 13(5) MiFID I, Article 16(5) MiFID II), record keeping (Article 13(6) MiFID I, Article 16(6), (7) MiFID II), protection of the client’s ownership rights (Article 13(7), (8) MiFID I, Article 16(8), (9) MiFID II), and algorithmic trading (Article 17 MiFID II). The depth of regulation shows that organizational requirements are considered a key component of risk regulation in financial services by the

European legislator. While this regime can be seen to supplement company law in that it regulates the organizational structure of companies engaging in investment banking, it is striking for at least one novelty as compared to typical company law:54 it does not only specify the work of the organs of the company and impose duties on members of such organs, but also independently specifies structural duties well below the level of organs. Under general company law, these aspects would rather be subject of duties of proper organization imposed on the board, but whose arrangement as such and certainly whose details would be left to the board’s discretion.55 [7.34] Thus, it does not come as a surprise to see that conflicts of interest are also subject to organizational precautionary provisions in Article 13(3) MiFID I and Article 16(3) MiFID II. The Level 2 Directive implementing MiFID I56 provided for further detail in organizational regulation, mandating inter alia the establishment of Chinese walls (Article 22(3) MiFID I ID). The core idea behind the regulation of the organization of the provider is to set up a company structure designed to minimize conflicts of interest in the first place. Since conflicts of interest raise the mentioned problems of fairness, efficiency, and cognitive limitations on one hand, and since the mere duty to act in the client’s best interest does not seem sufficient to exhort providers to do so in fact, financial services regulation, also in other fields such as general banking supervision law,57 makes specific provisions concerning the governance and organizational structure of the company. These provisions erect the first line of defence against the negative impact of conflicts of interest: while they will often not be sufficient to fully eradicate them, the hope is to contain them to a minimum. Only where these organizational measures are deemed insufficient is disclosure of the remaining conflicts required as a second line of defence (see below, Section II.2.C). [7.35] The guiding principle in the organizational requirements is that while they are necessary, they should not be too rigid. Rather, the regime of MiFID I and II spells out general principles and guidelines and leaves the discovery of the best organizational structure for a particular firm to the firm itself. The above-mentioned exception of Chinese walls shows their importance. The open-endedness of the regime is aimed at providing

incentives for innovation and at tapping the efficiency of decentralized decision-making.58

B. From MiFID I to MiFID II [7.36] The paragraph governing organizational requirements has been copied and pasted word for word from Article 13(3) MiFID I to Article 16(3)(1) MiFID II. It mandates the maintenance and operation of effective organizational and administrative arrangements to prevent conflicts of interest. This general provision was spelled out in greater detail in Articles 21–26 MiFID I ID. Article 21, for example, provides that firms must identify situations which are particularly likely to provoke conflict of interest, such as when the firm is likely to make a financial gain at the expense of the client (Article 21(a)) or when the firm has an interest in a specific outcome that differs from the best outcome for the client (Article 21(b)). In its para. 3, Article 22 notably compels the firm to establish Chinese walls, that is, provisions that interrupt the flow of information between different sections of the firm in order to minimize conflicts of interest. [7.37] The key difference between the old and the new regimes is the insertion of a provision on product governance in Article 16(3) MiFID II.59 The fact that the provision was added specifically to the section governing organizational requirements concerning conflicts of interest suggests that the European legislator intended to introduce the rules on product governance as part of the regulatory strategy to minimize conflicts of interest. This is also highlighted by Recital 71 of MiFID II which connects product governance with the duty to act in the best interest of the client.

C. Core Parameters of the MiFID II Regime [7.38] The key components of the organizational measures undertaken to prevent conflicts of interest are spelled out in Articles 21–26 MiFID I ID. According to Article 21, the firm must identify all situations from which conflicts of interest may arise. These comprise situations of direct financial

gain of the firm (Article 21(a)), an interest of the firm in an outcome different from the best one for the client (Article 21(b)), incentives to favour one client over another (Article 21(c)), direct competition between the firm and the client (Article 21 (d)), and finally inducements other than the standard commission fees (Article 21(e)). To counter conflicts engendered by such tempting circumstances, the key organizational feature contemplated by Article 22(3) MiFID I is the establishment of Chinese walls between persons and departments within a firm whenever the flow of information between these would generate or aggravate conflicts of interest. Article 25 specifies that this implies in particular the separation of financial analysts from other persons within the firm who are in client relationships, creating ‘areas of confidentiality’. These structures are intended to promote a sufficient degree of independence of the actors in the client relationship from the general interests of the firm or specific interests of other departments and persons within the firm. They include the prevention of the exchange of information (Article 22(3)(2)(a)), separate supervision (Article 22(3)(2)(b)), the independence of remuneration of one person from the actions of another person within the firm (Article 22(3)(2)(c)), prevention of inappropriate influence (Article 22(3)(2)(d)), and the avoidance of involvement of a single person in different business activities (Article 22(3) (2)(e)) wherever non-compliance with any of these would give rise to conflicts of interest. [7.39] The core features of the ‘areas of confidentiality’ are the following: non-public information may not travel from the one department to the other and both departments have to take their decisions independently from each other. So-called Chinese walls have to be erected:60 via mutual exclusion from access, if possible also physically; via instruction of employees about the function and the functioning of these walls, namely through guidelines; via observation of the information flows and the ways by which information proliferates. The core decision to be taken is which zone should be separated from which: typically at least the area for the loan business and for issuing of securities from the area of investment advice and the latter again from fund management and portfolio management.61 The fine tuning is done via ad hoc decisions in individual cases for which the compliance management is responsible: either by introducing additional bans on transfer of information, if it is particularly sensible (‘secondary Chinese

wall’), or in favour of exceptionally disregarding the wall (‘wall crossing’). The core criterion is how vital the knowledge also is in the area with which information could potentially be shared, particularly whether it is necessary to properly carry on the business (‘need-to-know’). The ESMA Final Report on MiFID II does indeed go even further in suggesting that ‘[t]he MiFID Implementing Directive should be amended by inserting new provision to require a physical separation between the financial analysts involved in the production of the investment research and other relevant persons whose responsibilities or business interests may conflict with the interests of the persons to whom the investment research is disseminated’.62 This would, for example, tackle cases in which the provider simultaneously provides investment banking services to an issuer,63 a duality of roles from which conflicts of interest between different clients may arise. [7.40] The central novelty of the MiFID II regime, however, beyond the establishment of Chinese walls, is the product governance regime laid out in Articles 16(3)(2–6) and 24(2) MiFID II. It complements the assessment of appropriateness and suitability of financial products for clients’ needs by establishing—early, in the manufacturing stage of the product—a match between product characteristics and a specific target market,64 even though product governance does not prejudice the assessment of appropriateness or suitability (Recital 71). The basic idea is that manufacturers of financial products must now identify a suitable target market of end clients for these products and distribute them within this target market (see Article 16(3)(2) MiFID II). The target market is defined primarily by the risk structure of the product, but it is also defined by the time horizon of the investor.65 Furthermore, the manufacturers must monitor the performance of the product post sale and make adaptations if it fails to conform with the needs of the target market (see Article 16(3)(4) MiFID II). Thus, the design of the product must primarily be aimed at coinciding with the core interest of the clients in the targeted market. Moreover, investment firms offering these products must ensure, by organizational means, that they understand the product-approval process and the characteristics of the different target markets and of the products they offer in these (see Article 16(3)(6) MiFID II). This includes the appropriate training of staff and control over the governance process by the management body.66 In doing so, a greater alignment of particularly the risk preferences of clients and the products

they are being offered is meant to be achieved. The main connection to the sphere of conflict of interest is the following: once the client has been allocated to a specific target market, the firm must offer products in the client’s best interest, taking account of the target market (Article 24(2)(2) MiFID II). In our view, this prevents the firm from simply offering products outside the target market;67 at the very least, such an extra-target offer must be accompanied with a clear warning to the client.68 Generally, the firm must take the target market into account when disclosing the risks associated with a recommended product (see Article 24(4)(b) MiFID II). [7.41] The establishment of target markets testifies to the fact that even within the group of retail or professional investors there are various subgroups with different risk preferences. The subgroups can now be identified with a certain target market and products that are tailored to their needs, both in the manufacturing and in the recommendation process. Since the target market helps to define the preferences of the client, it reduces discretion of the firm to offer products and thus limits the scope for biased advice based on conflicts of interest. A weakness of the proposal lies in the fact that firms will be in a position to determine what clients form part of what target market,69 permitting discretion that can be affected by conflicts of interest. However, if the target market is neatly defined in terms of risk preferences and time horizon, and if these are elicited by the firm from the client, there is a good chance that the allocation of clients to target markets will be fairly obvious and thus less open to manipulation. [7.42] With respect to all of these organizational duties, the question remains of how much effort a firm must exert in order to minimize conflicts of interest, and from what point on the disclosure regime concerning the remainder of conflicts of interest takes over. Here again, the ESMA Final Report offers a convincing reading of the MiFID II provisions: it describes the disclosure of information as a measure of last resort.70 The point where organizational measures end and disclosure begins is determined by the companies themselves: they must be reasonably confident that the range of organizational measures will suffice to avoid conflicts of interest if they intend to avoid disclosure.71 In this, they need not take every possible prudential measure; rather it is deemed sufficient by the majority of literature to limit the efforts to those strategies that a firm can reasonably be

expected to be undertaken given its size and structure.72 Eventually, the provider has to choose the most effective strategy amongst those that it is required to consider under the proportionality principle.

2. Conflicts of Interest in the Formation of Investment Services Contracts [7.43] We shall first discuss the provisions pertaining to the general regime on formation of the investment services contract (then to be followed by explanations on fees, below Section II.3). After a brief theoretical introduction, the differences between MiFID I and II will be highlighted before turning to the core parameters of the MiFID II regime.

A. A Contract Standards Solution [7.44] As was noted earlier, the regime of MiFID I and II comprises the general duty to safeguard the interests of the client and combines this with a more concrete prospective rule: the disclosure of conflicts of interest which endanger the due discharge of the duty to safeguard the clients’ interest. These duties are boundary conditions that influence the content of the investment service contract. The MiFID regime therefore tries to mitigate conflicts of interest in the formation of investment services contracts by giving the client the information necessary to make an informed choice on the kind of contract she wants to conclude. Thus, the organizational regime of precaution is supplemented by a second, contract-based regime still of precaution—all to safeguard as far as possible compliance with the actual substantive law standard: to take as sole guideline for action the interest of the client. There does indeed seem to be a need for (precautionary) regulation. In a study for the European Commission, Nick Chater, Steffen Huck, and Roman Inderst showed that more than half of the consumers were unaware of conflicts of interest of financial advisers and wrongly believed that advice was provided on a fully independent basis.73 The beneficial effect of disclosure would not only be ideally to enable clients to make an informed decision but also to discipline providers so that commissions do not lead to biased advice but rather incentivize information

seeking and selection of efficient investment instruments.74 Thus, the scope is threefold, really: first, to make (a good part of the) clients aware of the fact that there are considerable conflicts of interests potentially affecting investment advice; second, to give them the information needed to properly assess the number and the mechanisms of the different sources of conflict of interests possible or existing in this case, which should empower clients to ask more specific questions and take more specific precautions, for instance to choose between providers subject to more or to less conflicts of interests; and third, to serve as a disincentive for providers to cede to their conflicts of interests. In Section II.6, focusing on cognitive biases, we ask the additional question of whether this regime should be—or is already—supplemented by debiasing features for those investors who do not draw the appropriate conclusions from such information and whether the information required is already calibrated in a cognitively optimal way (see below Section II.2.C, but mainly Section II.6).

B. From MiFID I to MiFID II [7.45] Except for the introduction of product governance, the organizational duties of the provider have remained exactly the same in the transition from Article 13(3) MiFID I to Article 16(3) MiFID II. The disclosure regime, however, did see some changes. First, disclosure now needs to be made in a durable medium (Article 23(3)(a) MiFID II); such a provision was lacking in MiFID I.75 Second, the client must not only be informed about the conflicts of interest per se, as in MiFID I, but also about the organizational measures taken to mitigate them (Article 23(2) MiFID II). Third, and perhaps most importantly, the disclosure must be sufficiently detailed, but also take into account the nature of the client (Article 23(3)(a) MiFID II). The implications of this last provision will be discussed in Section II.2.C. [7.46] Generally, however, MiFID II deals with conflicts of interest at the pre-contractual stage in a fashion very similar to the one inaugurated under MiFID I: the organizational duties have remained exactly the same. The general structure of tackling precontractual conflicts of interest with organizational measures and disclosure was kept. The disclosure regime

was slightly, but not significantly, updated—unless one puts emphasis on the disclosure rule potentially requiring much more meaningful information and even de-biasing. What is undoubtedly a huge novelty at this stage, the introduction of a market structure approach, will be discussed under Section II.3, when it comes to fees.

C. Core Parameters of the MiFID II Regime [7.47] Under the regime of MiFID I and II, disclosure steps in where organizational measures fail. It is meant to warn the client of the conflict of interest so that she can reassess her stance on the quality of advice accordingly. In order for the disclosure to be effective, two key questions must be addressed: first, how much detail must be disclosed about the concrete conflict in question; and second, whether repeated disclosure is necessary. [7.48] The question alluded to above about the reach of the disclosure duty thus arises with renewed force under MiFID II. Does the provider only have to mention that she is conflicted or also explain the concrete facts that lead to the specific conflict of interest? The provisions are ambivalent about this point. Article 18(2) MiFID I held that ‘the general nature and/or sources of conflicts of interest’ must be disclosed. Article 23(2) MiFID II reiterates that phrase. The question therefore comes down to whether the general nature of the conflict of interest is to be portrayed in more abstract or concrete terms. Here, Article 23(3)(b) MiFID II adds that the disclosure shall ‘include sufficient detail, taking into account the nature of the client’.76 Thus, two questions arise. First, what is sufficient? Second, what characteristics define the nature of the client? In fact, the second question has to be answered first because the degree of detail is meant to depend on the nature of the client. [7.49] A first interpretation of the phrase ‘nature of the client’ can point to other instances in MiFID II in which it is used. For example, Article 24(14) (c) holds that delegated legal acts protecting the investor shall take into account the ‘retail or professional nature of the client’. This is reiterated, for

delegated acts in similar contexts, in Article 25(9)(c) and Article 27(9)(a). On one hand, one could conclude that the nature of the client in Article 23(3)(b) also refers to its professional or retail status. This would mean that greater detail can be given to professional actors, which seems reasonable. On the other hand, however, one might also argue that the fact that the retail/professional investor dichotomy was not mentioned in the specific provision points to the possibility of taking other features into account as well. Against the backdrop of the theoretical approaches to conflicts of interest mentioned above, one possible and sensible distinction might be the one between fully and boundedly rational actors. Indeed, it seems convincing to mandate disclosure in greater detail vis-à-vis rational actors who can be expected to adequately process the information while cutting back on the amount of information vis-à-vis boundedly rational ones who might be overwhelmed and overloaded. The problem with this interpretation is that in many cases it will be difficult for the provider to assess whether the client is either more fully or more boundedly rational, especially given that this distinction does not necessarily coincide with the retail or professional status of the investor. Therefore, the most sensible construction of the phrase ‘nature of the client’ would probably attach first to the retail or professional status of the investor, which can be easily assessed. Only in a second step, if possible, can the degree of rationality be taken into account if and only if it is determinable by the provider. What remains clear, however, is that this new addition in MiFID II calls for greater personalization of disclosure. ESMA seems to support this, suggesting that ‘disclosure of risks arising as a result of the conflict is essential for clients to understand the nature of the conflict of interest itself’.77 [7.50] This has a direct influence on the reach of the disclosure duty, the other question mentioned above. If the client is a professional investor and the provider does not have any reason to believe that she is boundedly rational, disclosure must be of greater detail. If, however, the client is a retail investor, disclosure should be of less detail unless the provider has reason to believe that the investor is fairly rational. What remains to be analysed is: How far should the degree of detail comprise the exact concrete type of conflict of interest? On one hand, a more specific exposition of the conflict of interest may put the client in a better position to judge which

recommendations may be affected and which may not. Strictly speaking, rational clients will therefore benefit from greater detail of disclosure. On the other hand, boundedly rational clients may be overwhelmed or bored by too much detail, leading in both cases to a neglect rather than a deeper appreciation of the information. This remains a problem even under a more personalized regime because the provider often will not be able to adequately ascertain whether she is dealing with a boundedly or fully rational type. In this context, it is to be kept in mind that under MiFID II even the organizational measures undertaken to mitigate conflict of interest need to be disclosed, adding to the amount of information delivered (with questionable benefit to the client). [7.51] The answer to the question of the reach of disclosure must take the whole situation in which disclosure takes place into account. Notably, conflicts of interest are certainly not the only items that need to be disclosed —they are followed by information on the specific nature of the service (Article 24(4)(a)) and of the investment products (Article 24(4)(b) and (c)), particularly its risk components. Therefore, to prevent information overload, it seems wise to avoid driving the details of disclosure to a maximum.78 However, if the conflict is of a specific nature so that it affects only a part of the obligations of the provider and if it can be expected that rational parties will be able to digest and act on the information, then more specific disclosure is warranted. In this case, rational parties stand to gain much from more detailed disclosure, while the disadvantage to boundedly rational actors may be limited. At least a part of those would not have taken the disclosure into account even if it were shorter; the rest may benefit from a disciplining effect that the detailed disclosure unfolds if rational agents act upon it and providers are unable to distinguish prima facie between rational and boundedly rational agents. To make things more complex, however, these disciplining effects of the disclosure are called into question by the results of an experiment by Ismayilov and Potters, who noted that disclosure did not affect either the providers’ nor the clients’ actions.79 All in all, disclosure should thus be limited in detail unless specific reasons (high rationality of the client; particular nature of the conflict) speak in favour of divulging the details of the conflict in question. A short description of the most relevant features of the mechanism generating the

conflict of interest would seem to be what is required if such rationality cannot be assumed. [7.52] This position arguably finds itself in accord with the recommendations by ESMA. Speaking on the reach of disclosure, the authority finds that it must be made ‘in sufficient detail to enable that client to make an informed investment decision’.80 However, clients will only be able to make an informed decision when they are not overwhelmed by the disclosed data, that is, if the amount of disclosure remains within their cognitive capacity limits. Therefore, making the reach of disclosure depend on the status as well as, at a secondary level, the rationality of the investor sits comfortably within both the wording and the purpose of Article 23(3) (b) MiFID II. [7.53] Going beyond interpretation, the following proposal for a reform of the disclosure strategies might fit both fully and boundedly rational types as well as retail and professional investors. Providers have to disclose to all investors how the rational professional investors typically react to the most important sources of conflicts of interests inherent in the particular case. The less experienced could thus opt for ‘following’ the more experienced ones. This solution would link investor protection closely to overall concerns of market integrity. It thus leverages the reactions of rational professional investors for the sake of other types of investors and simultaneously enhances the incentive structure for providers. Empirical tests should be conducted to analyse whether such an enhanced disclosure would make any difference for investor choice. Then, even Level 2 legislation may suffice, as the main thrust is laid down in MiFID II already. [7.54] Finally, we can now turn to the issue of repeated disclosure. Here, the literature and practice are still divided about the best approach.81 One may argue at the outset that providing the same disclosure before every transaction if the conflict of interest has not changed is cumbersome to both the provider and the client: it causes compliance costs on the side of the provider and cognitive costs of processing the information on the side of the client. That said, people tend to forget—more boundedly rational investors tend to forget more. Therefore, repeating disclosure may be necessary in some cases to raise the salience of conflicts of interest and to ensure that

their effects are really taken into account by the investor in every decision. CESR’s Technical Advice for the implementation of MiFID I noted that the frequency of disclosure should take into account the nature of the client.82 Luca Enriques thus suggests that generally retail clients should be informed more frequently.83 This makes sense since retail clients, other than more sophisticated institutional investors, will be less generally aware of conflicts of interest; in fact, 80 per cent of advisees in the retail sector fully trusted their advisers and did not seem to be aware of conflicts of interest.84 Enriques also notes that ‘standard’ conflicts of interest should only be disclosed only once. 85 However, the survey just mentioned which uncovered blithe and undeserved trust in advisers arguably speaks for a continual updating of the relevance of conflicts of interest in the minds of retail investors. Therefore, it is suggested here that the disclosure of generic conflicts of interest should be made at least periodically, perhaps roughly at every fifth encounter with the adviser. This disclosure of specific conflicts of interest remains necessary according to the guidelines visited in paragraph 7.50.

3. In Particular: Conflicts of Interest and the Regime on Fees [7.55] The disclosure strategy discussed in the previous section is complemented by the regulation of the fees of the service provider. These provisions have seen a particular and noteworthy change in the transition from MiFID I to MiFID II; since these démarches are the subject of separate chapters in this volume, we shall restrict ourselves to a few observations.86

A. A Market Structure Approach [7.56] In tackling conflicts of interest that arise by virtue of the fee structure, the new MiFID II provisions now take a market structure approach. They create two different markets for the offering of investment services: dependent services where the remuneration mostly comes from

inducements and where it should therefore be visible to all clients that conflicts of interests may considerably influence the advice given (where, however, the conflict of interest still has to be disclosed); and independent services where the client directly pays for the advice, and inducements (and also some other specific sources of conflicts) are ruled out. Thus, the rule is organizational—preceding disclosure and enhanced by it (Article 24(4)(a) (i))—but different from the prime organizational requirements affecting the organization of the (one) provider of investment services in that it now concerns the market structure and thus (consciously structured) competition between different kinds of providers of investment services.

B. From MiFID I to MiFID II [7.57] Indeed, this market structure approach is perhaps the greatest novelty in the MiFID II regime concerning conflicts of interest. It was absent under MiFID I, which only contained, under Article 19(1), the duty to act in the best interest of the clients, but did not say anything about how that duty should affect the fee structure of the company. Via gold-plating, important Member States had, however, introduced sophisticated regimes on inducements, statutory or in case law. The cases of the UK (Retail Distribution Review banning commissions) and of Germany (§ 31d Wertpapierhandelsgesetz (WpHG)) are noteworthy.87 The former in particular strongly influenced the MiFID II market structure regime, to which we now briefly turn.

C. Core Parameters of the MiFID II Regime [7.58] The two markets created by dependent and independent services treat conflicts of interest quite differently. The market for independent services attempts to reduce conflicts of interest to a minimum by removing their core element: first, one of the, perhaps even the, key source of conflict of advice,88 namely fees from issuers or other entities with an interest in selling their investment products; second, limited product range (Article 24(7)).

[7.59] In the market for dependent services, however, the source of conflict remains and has to be disclosed, and further requirements have to be met (Article 24(9)). Thus, in the dependent regime, the market is structured so as to maximize the probability that inducements will still benefit, or at least not damage, the client. And these market structure requirements are coupled with disclosure rules. Whoever wishes to contract for the high-quality, independent market has to pay for it directly out of her own pocket, while in the other market, the client pays indirectly through worse quality or higher prices (part of which are passed on to the provider).89 [7.60] It remains to be seen in how far these schemes will reshape the advice landscape in Europe. In the UK, following the Retail Distribution Review mentioned before which foreshadowed the MiFID II regime,90 empirical evidence suggests that product bias has indeed decreased.91 However, as a reaction to the ban on commissions Lloyds banking group has begun to restrict investment advice (of an independent nature) to retail customers investing more than £100,000. The decision was taken in response to an in-house study conducted by Lloyds purportedly showing that willingness to pay directly for investment advice decreases with the sum to be invested.92 If more companies follow this lead, and refrain from even offering independent investment advice to small investors, the wellintentioned rules could end up hurting small retail investors by making advice less accessible to them—broadening what has been termed the ‘advice gap’ in investment services, especially for less affluent customers.93 This is particularly worrying given that, as noted earlier, 80 per cent of retail investors in the EU use advice to inform their purchase decisions.94 Ex post evaluation of the MiFID II regime should therefore inquire into these potentially regressive effects.

4. Conflicts of Interest in the Execution of Investment Services Contracts (‘Best Execution’) [7.61] Conflicts of interest do not only arise in the formation of the contract but also in its execution. The regime of MiFID I and II deals with

this in a range of specific provisions.

A. A Strict Investors’ Interest Approach [7.62] The central idea behind the protection of the interest of the investor in the execution phase of the contract is to specifically oblige the provider to act in the best interest of the client and to specify precisely the benchmark against which the actions taken are to be assessed. This is necessary because the protection of the client interest faces two difficulties in this domain. First, the execution is out of reach for the client; she is not in a position to accept or decline certain kinds of execution after the formation of the contract. Rather, the provider executes the trade on her own, which is one of the reasons she was retained in the first place. Second, the provider must elaborate an execution policy against which her performance is measured ex post. However, it seems utopian to hope for market solutions based solely on the best terms of execution. The information asymmetry between the provider and the client in this technical domain is so great that it seems difficult to overcome by means of disclosure alone, which would be a prerequisite, however, for a functioning market on the terms of the execution. Therefore, the MiFID regime specifies what constitutes ‘best execution’; that definition, in turn, varies substantially between the two Directives—so fundamentally that, in fact, the two regimes can be seen as following different philosophies and that the philosophy based almost exclusively on disclosure has been largely given up or at least modified (see now Recitals 91–8 MiFID II, instead of only three Recitals in MiFID I).

B. From MiFID I to MiFID II [7.63] Article 21 MiFID I set out the ancient regime of best execution. This required the provider to design a best execution policy that would work in the best interest of the client. This policy had to be disclosed to and consented by the clients before the execution of an order. The success of the policy needed to be monitored, and assessed comparatively with alternative policies, such as those of competitors, the policy being adapted if necessary.

Material changes again had to be disclosed to clients. The freedom of the provider to define the execution policy was therefore indirectly limited by the obligation to report on the relative success of the chosen strategy. [7.64] MiFID II keeps this regime in general but updates it in two dimensions in view of the greater variety of trading venues available within the European Union (Recital 92). First, disclosures of execution policy were considered too generic to make a meaningful difference (Recital 97). Second, the new provisions seek to protect retail clients in a specific manner (Recital 92).

C. Core Parameters of the MiFID II Regime [7.65] According to MiFID II, providers still have to elaborate an execution policy that is to further the clients’ interests in the best way possible (Article 27(4)), regularly monitor and update it, as well as disclose it to and obtain consent from the client. However, as noted above, some key changes were undertaken. [7.66] First, the European legislator intended to improve the quality of the disclosure in order to spur, where possible, market selection processes based on the terms of the execution policy. In what seems to be an attempt to square the circle, Article 27(5)(2) notes that the information must be given ‘clearly, in sufficient detail and in a way that can be easily understood by clients’. What seems problematic is that more detail will often entail less understanding. It remains to be seen therefore whether the ‘improved’ disclosure will have any beneficial effects on the content of the execution policies or on execution itself. [7.67] Second, the best interest of the client is defined differently for professional and retail investors (Article 27(1)(2)). For professional investors, the provider has to take into account, as under Article 21(1) MiFID I, the ‘price, costs, speed, likelihood of execution and settlement, size, nature or any other consideration relevant to the execution of the order’. For retail investors, however, only the best total price is installed as a proxy for best interest. Under the old regime, this narrow definition of the

best execution was mentioned only on a Level 2 implementation;95 it has now moved up the ladder to Level 1. For retail clients, this change takes into consideration the tension between their information needs and willingness to retrieve information (which is typically lower) and the interest in the parameters of execution other than price (which typically is lower as well). The total price (total consideration) comprises the actual price as well as any costs incurred directly by the execution such as venue trading fees, commissions, etc. The advantage of looking exclusively at the total consideration is clear: while the speed of the transaction is often a lesser concern for retail investors, the price is not only a key dimension retail investors care about but is also relatively easy to ascertain and monitor.96 This clear provision therefore helps to safeguard clients’ interests even if the market does not function adequately on the terms of the execution policies. However, it should be noted that the rule may present an advantage to established, and thus more liquid, trading venues, thus mitigating competition in the transaction market, at least in the retail investment services.97 There is reason to believe, though, that the trade of large blocks of products in the professional investor market, for which price is not the determining characteristic, will be sufficient to spur competition between established and new trading venues. Since retail clients certainly have more limited power to (algorithmically or cognitively) process information than professional ones, it seems sensible to restrict ‘best execution’ in their case to a single, easy-to-determine parameter. [7.68] Third, the focus on price as the key proxy for best execution in the retail financial services market is complemented by an anti-discrimination rule, see Recital 95: All costs incurred in the transaction market, that is, in the execution of the order, must be passed on to the client as they are. Particularly, this means that a sudden drop in prices on the transaction market must be reflected in the total consideration charged to the client; the provider may not use the transaction market to charge hidden fees to the client by topping up on the transaction fees. This seems sensible since such hidden fees cannot be easily discovered by the client (screening problem) and providers should not be allowed to use this information asymmetry which would also be difficult to overcome. The ultimate desideratum is that the beneficial effects of competition of venues are not reduced, due to the

fact that price reductions by one venue or another are not passed on to the investors. [7.69] Fourth, still more reporting duties are installed to raise the accountability of the provider in the execution. Providers themselves have to publish an annual report in which they name the top five execution venues in terms of trading volume of the last year (Article 27(6)). And trading and execution venues have to make available to the public, without any charges, data relating to the quality of execution of transactions in that venue on at least an annual basis (see Article 27(3)). This comprises details about price, costs, speed, and likelihood of execution for individual financial instruments. The public disclosure of these parameters will enable checking the validity of the provider’s reports on the relative success of their execution strategy and thus contributes to the disciplining function of public scrutiny in indirectly delimiting the freedom-to-choose execution policies. [7.70] Fifth, and finally, it is restated that providers may not receive any remuneration from trading execution venues that would contravene the interest in taking away adverse incentives from investment service providers on key competition parameters—just like in the rules on inducement described in the previous section, Article 27(2).

5. Conflicts of Interest in the Case of Linked Contracts [7.71] Finally, conflicts of interest can also arise when different contracts are linked in the hands of the service provider. On one hand, she may opt to not only to provide the investment service but also credit services, such as a loan agreement, designed to finance the contemplated investments. On the other hand, different investment services can be linked by the provider by means of cross-selling, that is, by bundling or tying together essentially different services in a single product.

A. A Balancing and Unfair Trade Practices/Competition Law Approach [7.72] The linking of different contracts provides two different challenges to regulation: first, the possible exploitation of investor errors in the financing of investment services; and second, in addition to concerns of investor protection, harm to competition in the wake of cross-selling techniques. The MiFID II regime addresses these concerns by employing a balancing approach which includes an unfair trade practices/competition law approach for cross-selling: the freedom to conduct a business is thus traded off against investor protection on one hand and against the safeguarding of competition on the other. In the case of financing of investment through loans, the practice is generally permitted because the duties imposed on investment advice are high enough to also include the risks stemming from an investment financed by a loan. Therefore, financing is forbidden where this protection is largely missing, namely in cases in which the degree of investor protection is greatly reduced, namely the mere execution of client orders. Conversely, cross-selling is seen as a risk to both competition and investors and is thus subjected to heightened scrutiny.

B. Investment Contracts Financed by Loan Contracts [7.73] The first regulatory challenge arises when investment contracts are simultaneously financed by credit contracts. The granting of credits or loans to an investor to allow him to carry out a transaction in a financial instrument is defined as an ancillary service in Annex I Section B (2) MiFID II. What are the legal implications of the coupling of credit and investment services? In the realm of disclosure, MiFID cedes to the specific provisions on the regulation of consumer credit, that is, it is sufficient to fulfil the disclosure obligations contained in the consumer credit regulation (see Article 24(6) MiFID II). [7.74] Most notably, however, according to the MiFID II regime, loan contracts may not be coupled with the performance of investment services

whenever the latter consist only in the execution of client orders (see Article 25(4)). Indeed, since the degree of investor protection through exploration and disclosure is reduced to a minimum in the case of execution-only services, it seems sensible to rule out the rendering of such services on credit basis (cf. Recital 80). Otherwise, the provider might be tempted to persuade clients wishing to speculate on their own (through execution only) to take out loans in order to take greater chances. In this case, the conflict of interest would with good likelihood come together with some of the biases discussed in Section I.2.C. Execution-only clients, by ordering these services, consider themselves not to be in need of the protection of exploration and disclosure provided by the usual investment services regime. While this may be true with regard to many of them, at least some will come to this conclusion through a bias towards optimism and overconfidence in their own ability to make assessments concerning the future development of the products they invest in and their own capacity to bear risk. It would indeed be a mischievous coincidence if such clients, having erroneously opted out of most investor-protection schemes, were now faced with aggressive credit strategies by the service providers. While some rational clients might be able to benefit from these, the boundedly rational, overoptimistic, and overconfident ones would most likely also underestimate the chance of defaulting on that credit. Therefore, severe financial harm could ensue for these investors. Article 25 (4) thus wisely excludes the simultaneous provision of credit and investment services in the guise of execution-only agreements.

C. Cross-selling of Different Investment Services [7.75] The second case of linked contracts is contemplated in Recital 81 MiFID II: cross-selling. In this situation, the provider ‘offers an investment service together with another service or product as part of a package or as a condition for the same agreement or package’ Article 4(1)(42). ESMA helpfully distinguishes between bundled and tied packages:98 in a bundled package, each of the components can be purchased separately by the client; in a tied package, at least one of the components can only be bought through the package. For example, a tied package may force the investor, when purchasing stock, to also open a current account with the same

provider. Not surprisingly, the Recital highlights the potential anticompetitive effects of such practices. After all, cross-selling is one of the classic examples of the abuse of a dominant position, explicitly cited in Article 102(d) TFEU. The very provision of tied packages in most cases reflects a conflict of interest since the tying benefits the provider while reducing investor choice and the possibility to combine services from different providers—one of the situations contemplated in Article 21(b) MiFID I ID as an epitome of conflict of interest. Recital 80 therefore notes that such practices should be carefully assessed to safeguard competition and consumer choice. Indeed, as antitrust scholarship has shown99 and as the Recital suggests, tying may distort competition, reduce mobility, and lead to worse outcomes for investors. [7.76] Under the MiFID II regime, Article 24(11)(1) holds that providers must inform clients which of the different components can be purchased separately and provide evidence of the costs for each of these separately. If the risks of the package are different from the risks of the individual components, the modification of risk by the bundling or tying strategy must be explained to the investor (see Article 24(11)(2)).100 Here, it may be doubted whether it will always be possible to do so in a manner understandable to most investors; therefore, the focus should be put on simple categories (such as ‘greater risk’ or ‘lower risk’, potentially coupled with colour coding or even a ‘traffic light’ system). Furthermore, according to Article 25(2)(2) and Article 25(3) the firm must ensure that the whole package is suitable or appropriate to the investor. [7.77] In December 2015, ESMA published guidelines on cross-selling practices.101 It stressed that price transparency is key to investor protection in this area, and that prices need to be disclosed in good time before the investment decision and in an understandable manner.102 The same holds true for risk disclosure; risk components must furthermore be disclosed with equal salience as price components.103 Importantly, in the case of bundled packages, the default must not be the purchase of the bundled package but rather of the separate components (no pre-ticked boxes).104 This strategy makes use of default effects such as the status quo bias.105 Finally, ESMA notes that staff must be trained and incentive and

remuneration practices designed to minimize conflicts of interest in crossselling practices.106

6. New Strategies: the Future of the Regulation of Conflicts of Interest [7.78] Particularly in view of bounded rationality, the literature has suggested a host of further regulatory tools that might mitigate the problem of conflicts of interest. In this section, we shall therefore briefly review cognitively optimized disclosure, debiasing, and caps on commissions.

A. Cognitively Optimized Disclosure [7.79] The beneficial effects of disclosure depend on its effectiveness in educating clients. As mentioned, clients may suffer from limited attention, weak motivation, and cognitive capacity limits which prevent the proper cognitive reception of the disclosure of conflicts of interest. Against the backdrop of the possibility of bounded rationality and information overload, disclosure thus seems at least problematic. As Omri Ben-Shahar and Carl E. Schneider have powerfully argued, disclosure cannot be trusted to automatically work its magic.107 Therefore, it should be empirically tested in every specific instance whether the provision of disclosure really enhances decision-making.108 In a recent experiment, Huseyn Ismayilov and Jan Potters demonstrated that the rate by which clients follow providers’ advice is not affected by the disclosure of conflicts of interest: they conclude that it is neither beneficial for nor detrimental to the decisionmaking of the client.109 [7.80] Does this mean that disclosure should be discarded altogether? This chapter claims that such a reaction would ‘throw the baby out with the bathwater’. Rather, disclosure should be cognitively optimized. Behavioural economics has recently focused much regulatory attention on the concrete format and design of disclosures, going beyond the mere provision of ever more detailed forms of information.110 For example, the ESMA guidelines

on cross-selling stress the importance of providing price information in simple language.111 This exhortation joins a full array of provisions designed to simplify disclosure.112 However, in a recent paper, Ben-Shahar and Chilton have presented results of a series of experiments that point to the limits of cognitively optimizing disclosure.113 In the context of privacy disclosures, even the most advanced formats fared no better than traditional boilerplate disclosures—indeed, both were equally ignored. As with every empirical study, however, close attention must be paid to the limits of external validity. In the context of conflicts of interest, Inderst and colleagues were actually able to demonstrate that a strong health warning (‘[n]ote that this means that the advisor did not necessarily have your own investment earnings in mind when he gave his advice’) did make a difference: people spent more time on the disclosure when the warning was present.114 While one must be careful not to attach warning signs to all disclosures—this would raise the danger of wear out—the central status of conflicts of interests might justify enhancing the salience of disclosure with warning signs. In another recent study,115 Geneviève Helleringer demonstrated that ‘explanatory disclosures’, which not only disclosed a conflict of interest but also explained that this implied that the interests of provider and client were not aligned,116 did significantly affect clients’ trust in the advice and, under additional conditions,117 their willingness to invest. Such enhanced disclosure formats that increase the salience of conflicts of interest present a potential way forward to make naïve clients more sophisticated, which in turn promises to set market forces in motion that enhance efficiency and client welfare.118

B. Debiasing [7.81] As was further mentioned above, clients may not only suffer from information overload but also from different biases which prevent the proper processing of information. In particular, the significance and degree of conflicts of interest may be misunderstood due to optimism bias, confirmation bias, perseverance, and other phenomena. The behavioural literature has long suggested debiasing through law as one of the strategies to cope with such cognitive errors.119 In this context, it is very important to

reiterate that not all clients suffer from the same biases, or from biases at all. For debasing strategies to be justified, two conditions therefore must be met: debiasing must be expected to be effective among the boundedly rational clients, and it must not unduly affect rational types.120 [7.82] The effectiveness of debiasing continues to be an issue of scientific debate.121 Much depends on the concrete bias and the concrete practice of debiasing. One strategy would be to simply disclose the relevant bias to clients, for example in the form of a warning. In some settings, such strategies have proven successful to mitigate biases; in others not.122 [7.83] As to the second condition, the effect of the debiasing on rational actors obviously also strongly depends on the type of intervention. While disclosure of the bias is not particularly intrusive, it may have two negative effects on rational actors. First, it increases the amount of information to be disclosed and thus makes overload more likely. Second, it may induce bias in formerly rational actors if these correct for biases they are not actually subject to. [7.84] Therefore, more empirical research is needed before debiasing strategies can be fruitfully installed in this context. Moreover, information overload seems to be a greater problem than biases in regulating conflicts of interest. However, if simple and unintrusive debiasing strategies can be operationalized in this setting, they would present a welcome addition to the regulatory toolbox.

C. Caps [7.85] Price caps on commissions are a final regulatory instrument. This can be considered a measure of last resort in case the previously mentioned strategies fail. If the cap is set at a low level, this will reduce the incentive to the provider to give biased advice. If most of the clients are naïve about the dependent nature of the services and the commission and fee structures that induce these biases (as the study by Chater, Huck, and Inderst suggests123), the cap can therefore serve to realign the interests of provider and client. However, as Inderst and Ottaviani have rightly pointed out,

commissions can also have beneficial effects that are endangered by outright caps.124 For example, they can generate incentives for otherwise sluggish and disinterested providers to search for information about new financial products that could be marketed with a commission fee, enlarging the pool of products that providers can recommend. Furthermore, it may be expected that at least to a certain extent the most efficient products will come with the highest commissions so that a free commission scheme channels clients to better products. These functions of information search and efficient allocation are hindered by caps. [7.86] In a recent paper, Oren Bar-Gill125 has analysed the effects of price caps on consumer goods and pointed out a final complexity: often, products will comprise multiple price components. If only one of the components is regulated by a cap, this can have the unintended consequence of sellers reacting with an increase in other price dimensions which renders the cap futile but still produces significant administration costs. In the case of financial services, it might for example be expected that providers would potentially raise or introduce different kinds of ‘service fees’ to make up for a loss of commission revenue. However, while it is true that this would eventually maintain a similarly high price structure, it might have the beneficial effect of reducing incentives to recommend specific products only because of the high commissions attached to them. While the price effect of a cap may therefore be disputable, it could have a bias-reducing effect. Nevertheless, if a cap is to be considered, a clear focus must be put on ruling out the substitution of commissions by other kinds of monetary or non-monetary rewards offered by third parties for marketing and selling of their specific products.

III. Conclusion [7.87] This chapter has inquired into both the theory of the regulation of conflicts of interest in general and the practice of such regulation in the case of financial services regulation under the MiFID regime. We first reviewed theoretical approaches potentially legitimizing the regulation of conflicts of interests and then applied these insights in a step-by-step survey of the

different stages over which the client–service provider relationship evolves, from the first point of contact to the execution of the investment services contract in specific trading venues. Finally, going beyond positive law, we discussed the conclusions that the theoretical overview warrants for the future of the regulation of conflicts of interests. [7.88] In doing so, we have argued for the following seven distinct propositions. First, not all conflicts of interest are worthy of regulation; rather such a necessity arises only when one party is entrusted with a position of influence while simultaneously not offering any compensation for this discretion to the counterparty. Second, the domain of financial services stands out in several respects vis-à-vis general contract law with respect to such conflicts of interest: on one hand, the interest of the client is often the sole guideline under MiFID; on the other, we suspect that this result may be fuelled by the nature of investment advice as a credence good and by the prevalence of behavioural finance as a discipline highlighting cognitive biases. Third, given the centrality of conflicts of interest for financial services regulation, it is all the more deplorable that the current regime of European positive law, both under MiFID I and under MiFID II, does not provide a convincing account of the goals of the regulation of conflicts of interest. Therefore, fourth, an inquiry into the theoretical and interdisciplinary foundations of such regulation is warranted. We claim that the current regime can be explained by having recourse to duties of fairness (resulting from the conveyance of a position of influence without compensation); to efficiency (aiming at the reduction of the possibility of hidden gains and thus generating trust in investment advice); and to cognitive limitations (highlighting the effects of biases on investment choice and evaluation of investment advice). Fifth, these three main legitimizing forces for the regulation of conflicts of interests can be uncovered in the range of regulatory approaches under the MiFID regime. These generally fall into either the category of organizational or that of contract/disclosure rules. However, a mix of heterogeneous strategies is employed at different stages of the client–service provider relationship: the organizational requirements follow a company structure solution; the regime on the formation of investment services contracts a contract standards solution; the regime on fees and inducements a market structure approach; the execution stage a strict investors’ interest approach; and the

regulation of linked contracts a balancing and unfair trade practices/competition law approach. Sixth, while much continuity exists, the most innovative parts of the MiFID II regime can probably be found in the product governance rules and in the market structure approach governing the regime of fees, with its characteristic division of services into dependent and independent ones. Finally, we argue that some of the theoretical concerns, such as cognitive limitations, have been insufficiently addressed by the MiFID II regime and that future strategies should include cognitively optimized disclosure, and in the appropriate cases also debiasing, and perhaps even mandatory price caps on commissions.

* The authors would like to thank Mikaella Yiatrou and George Papaconstantinou for excellent research assistance. 1 For an early assessment of the positive law of conflicts of interest, see Guido Ferrarini, ‘Contract Standards and the Markets in Financial Instruments Directive (MiFID): An Assessment of the Lamfalussy Regulatory Architecture’, (2005) ERCL 1, 19, 33–5, 37–9; cf. for a more policy-oriented approach Luca Enriques, ‘Conflicts of Interest in Investment Services: The Price and Uncertain Impact of MiFID’s Regulatory Framework’ in Guido Ferrarini and Eddy Wymeersch (eds) Investor Protection in Europe. Corporate Law Making, the MiFID and beyond (Oxford: OUP, 2006), p. 321; and (with respect to the transition to MiFID II) namely Christoph Kumpan and Patrick C. Leyens, ‘Conflicts of Interest of Financial Intermediaries: Towards a Global Common Core in Conflicts of Interest Regulation’ (2008) ECFR 72. 2 Cf. Kumpan and Leyens (n. 1), 79–80: ‘discretionary decision-making power to affect the wealth of another’, and at 84 for a proposal of a definition of conflicts of interest; more extensively, Christoph Kumpan, Der Interessenkonflikt im deutschen Privatrecht (Tübingen: Mohr-Siebeck, 2014), p. 14–29. 3 Cf. also CESR’s Technical Advice on Possible Implementing Measures of the Directive 2004/39/EC on Markets in Financial Instruments, 1st Set of Mandates, January 2005, at 41: ‘It is not enough that the firm stands to make money or that the client stands to lose money, this must then involve a conflict with a duty the firm owes to the client.’ 4 See, for instance, Kumpan and Leyens (n. 1), 79 who distinguish between conflicts of interest and opposition of interests. 5 First discussed clearly in this sense, to our knowledge, in Wolfgang Zöllner, Die Schranken mitgliedschaftlicher Stimmrechtsmacht bei den privatrechtlichen Personenverbänden (Munich: Beck, 1963), pp. 341–56 et passim; more ample development of this theory (and literature, also in English), below Section I.2.

6

First discussed clearly for this additional criterion, to our knowledge, in Stefan Grundmann, Der Treuhandvertrag: die werbende Treuhand (Munich: Beck, 1997), Chapters 4 and 5; and summary in English in Stefan Grundmann, ‘Trust and Treuhand at the End of the 20th Century: Key Problems and Shift of Interests’ (1999) The American Journal of Comparative Law 47, 401–28; more ample development of this theory, below Section I.2. 7 For a broad survey of cases from different areas, see both Stefan Grundmann (above n. 7), Chapters 9–12 and Christoph Kumpan (above n. 3), 105–17, 201–20, and 508–55. 8 Kumpan and Leyens (n. 1), 80; Marc Kruithof, ‘Conflicts of Interest: Is the EU Approach Adequate?’, December 2005, Ghent University Financial Law Institute Working Paper No. 2005-07, available at , at 12–18. 9 For a treatment in the literature, see Kumpan and Leyens (n. 1), 97–8; Kruithof (n. 9), at 13–14. 10 While in the scope of application of MiFID II such clashes between interests of different clients do not seem paramount, in the scope of application of MiFIR this is different. Conflicts of interest are regulated with a view to mitigating clashes of interest between different (groups of) clients, via rules regulating the organization of particular kinds of markets, see Recitals 9 and 26 of Regulation (EU) Nr. 600/2014 of the European Parliament and of the Council of 15 May 2014 on Markets in Financial Instruments and amending Regulation (EU) Nr. 648/2012, EU OJ 2014 L 173/84. For more detail on these rules on market organization, see Chapter 5. 11 See Recitals 42, 51, 53 et seq. (together with market integrity); 71 et seq., 77 (for inducements and ‘best execution’); 80, 81, 87, 91 (again for ‘best execution’); and 155 and 164 of MiFID II (the latter two for all three goals). See also Recitals 2 (for the regulation as such, plus facilitating EU-wide offers), 31 (investor protection requiring differentiation between kinds of clients), 33 (for ‘best execution’ in particular), and 44 of MiFID I (for aftersales transparency). 12 See Recitals 53 et seq. (together with investor protection) and 155 and 164 of MiFID II (the latter two for all three goals), and Recital number 44 of MiFID I (for aftersales transparency). 13 Enriques remarks that market integration may even be the central aim of securities regulation in the EU and thus also the overriding aim of the MiFID (I) regime: Enriques (n. 1), 338. 14 This Recital reads as follows: ‘Since the main objective and subject-matter of this Directive is to harmonise national provisions …’. One realizes, however, that this phrase is mainly meant to differentiate between several legal bases which seemed possible (and is not so much about the validity of different goals of the Directive) when one continues to read. The Recital continues as follows: ‘… concerning the areas referred to, it should be based on Article 53(1) of the Treaty on the Functioning of the European Union (TFEU)’. 15 Recital 69 where conflicts of interest are mentioned a second time relates only to trading venues and is therefore of limited interest for this chapter, where the client

relationship is core. Recitals 88, 123, and 155 relate to procedural rules and do not contain additional content for our purposes. 16 For protection of functioning markets and of individual investors (as mutually reinforcing each other), see the groundbreaking (and not only for German literature) Klaus Hopt, Kapitalanlegerschutz im Recht der Banken (Munich: Beck, 1975), p. 51 et seq., 334– 7; today, see for instance, Stefan Grundmann, European Company Law: Organization, Finance and Capital Markets (2nd edn, Antwerp/Oxford: Intersentia, 2012), § 19 paras 16–18; also Niamh Moloney, EC Securities Regulation (3rd edn, Oxford: Oxford University Press, 2014), pp. 564–71; apparently, however, of little importance in UK practice, see: Alistair Alcock, The Financial Services and Markets Act 2000: A Guide to the New Law (Bristol: Jordans, 2000), pp. 178–80 (‘In the UK, such private resort to the courts has been much rarer.’). 17 See, e.g., Werner Güth, Rolf Schmittberger, and Bernd Schwarze, ‘An Experimental Analysis of Ultimatum Bargaining’ (1982) J. Econ. Behav. & Organization 3, 367; for an overview, see Ernst Fehr and Simon Gächter, ‘Fairness and Retaliation: The Economics of Reciprocity’ (2000) J. Econ. Persp. 14, 159; see further Wolfgang Fikentscher, Philipp Hacker, and Rupprecht Podszun, FairEconomy. Crises, Culture, Competition and the Role of Law (Heidelberg: Springer, 2013). 18 Grundmann (n. 7), Chapters 4 and 5; and summary in English in Stefan Grundmann, ‘Trust and Treuhand at the End of the 20th Century: Key Problems and Shift of Interests’ (1999) American Journal of Comparative Law 47, 401–28; Philipp Hacker, Verhaltensökonomik und Normativität (Tübingen: Mohr-Siebeck, forthcoming), Part 2. 19 Aristotle, Nicomachean Ethics, Book V. 20 John Rawls, A Theory of Justice (Cambridge, MA: Belknap Press of Harvard University Press, rev. edn, 1999), e.g. Chapter 1: ‘Justice as Fairness’; for context, see also Stefan Grundmann, ‘Gesellschaftsordnung und Privatrecht’ in Stefan Grundmann, Hans Micklitz, and Moritz Renner (eds) Privatrechtstheorie (Tübingen: Mohr-Siebeck, 2015), Vol. 1, pp. 405–43 (forthcoming in English). 21 Jürgen Habermas, ‘Discourse Ethics: Notes on a Program of Philosophical Justification’ in Moral Consciousness and Communicative Action 43 (Christian Lenhardt and Shierry Weber Nicholsen trans., 1991); Jürgen Habermas, Between Facts and Norms (Cambridge, MA: The MIT Press, William Rehg. trans., 1996), e.g. Chapter 3: ‘A Reconstructive Approach to Law I: The System of Rights’. 22 This cannot be developed in detail here. Highly influential in this, at least in Germany, but paradigmatic well beyond German private law: Walter Schmidt-Rimpler, ‘Grundfragen einer Erneuerung des Vertragsrechts’ (1941) Archiv für die civilistische Praxis (AcP) 147, 130–97 (the ‘guarantee of justness’, today, under the influence of Habermas, rather coined as the ‘chance of justness’); and Ludwig Raiser, ‘Vertragsfunktion und Vertragsfreiheit’ Festschrift zum hundertjährigen Bestehen des Deutschen Juristentages 1860–1960, pp. 101–34; short survey on the main explanations given (in different jurisdictions, but also beyond strict legal theory, namely in economics) in Stefan Grundmann, ‘Privatautonomie,

Vertragsfunktion und “Richtigkeitschance”?’ in Grundmann, Micklitz, and Renner (n. 21), Vol. 1, pp. 875–902. 23 Most explicitly and with lengthy explanation (and potentially also the first to address this): Zöllner (n. 6), 341–56 et passim; the argument can, however, also be found in the common law literature and case law on fiduciary duties, see from the early development of this idea J. C. Shepherd, The Law of Fiduciaries (Toronto: Carswell, 1981), pp. 35–42, 93– 123 (‘Encumbered Powers’); critical in this respect, however: Deborah DeMott, ‘Beyond Metaphor: an Analysis of Fiduciary Obligation’ (1988) Duke L. J. 879–924, 912 seq.; see also, with ample reference to case law, Paul Finn, Fiduciary Obligations (Sydney: Law Book Company, 1977); Ernest Vinter, A Treatise on the History and Law of Fiduciary Relationship and Resulting Trusts (3rd edn, Cambridge: Heffer, 1955); Tamar Frankel, ‘Fiduciary Law’ (1983) Cal. L. Rev. 71, 795–836. 24 See nn. 6 and 7. 25 For the efficiency implication of hidden gains with respect to the cost of investment advice, see more in detail Chris Bryant and Graham Taylor, ‘Fund Management Charges, Investment Costs and Performance’ (May 2012) Investment Management Association, Statistics Series Paper 3, Chart 1; Council of Economic Advisers of the President of the United States, ‘The effects of conflicted investment advice on retirement savings’ (February 2015) Accessed on 30/03/2016 from ; David Blake, ‘On the Disclosure of the Costs of Investment Management’ (May 2014) Pensions Institute Discussion Paper PI-1407; and with respect to the value of the investment advice to the consumer see: Michael Finke, ‘Financial Advice: Does it Make a Difference?’ in Olivia S. Mitchell and Kent Smetters (eds) The Market for Retirement Financial Advice (Oxford: OUP, 2013), p. 229; in general see: Jason G Cummins and Ingmar Nyman, ‘The Dark Side of Competitive Pressure’ (2002) Divisions of Research & Statistics and Monetary Affairs, Federal Reserve Board. 26 For more detail on the adverse selection implications of fundamental lack of transparency in market offers, see Daniel Schwarcz, ‘Beyond Disclosure: The Case for Banning Contingent Commissions’ (2007) Yale Law & Policy Review 25, 289–336; Stewart C Myers and Nicholas S Majluf, ‘Corporate Financing and Investment Decisions When Firms Have Information that Investors Do not Have’ (1984) Journal of Financial Economics 13, 187–221, 196. 27 On conflicts of interest implications that lead to issues other than adverse selection see more in detail Jeremy Burke, Angela A Hung, Jack W Clift, Steven Garber, and Joanne K Yoong, ‘Impacts of Conflicts of Interest in the Financial Services Industry’ (2015) RAND Working Paper WR-1076, 1–65. 28 Enriques (n. 1), 324; see also Paul A. Gompers and Josh Lerner, ‘Conflict of Interest and Reputation in the Issuance of Public Securities: Evidence from Venture Capital’ (April 1999) Journal of Law & Economics 42(1), 1–28. 29 On this (rough) categorization of situations into three paradigmatic cases—in which quality can either be assessed before the transaction, with hindsight, or not at all—see more

in detail Anthony Creane and Thomas D. Jeitschko, ‘Endogenous Entry in Markets with Unobserved Quality’ (2012) Economic Analysis Group Discussion Paper 12-6, 1–37; as for search goods and experience and credence goods, see the pioneering works of Phillip Nelson, ‘Information and Consumer Behaviour’ (1970) Journal of Political Economy 78, 311–29 and Michael Darby and Edi Karni, ‘Free Competition and the Optimal Amount of Fraud’ (1973) Journal of Law and Economics 16, 67–88; and today Wesley A. Magat and W. Kip Viscusi, Informational Approaches to Regulation (Cambridge, MA: MIT Press, 1992); Wesley A. Magat, ‘Information Regulation’ in Paul Newman (ed.) The New Palgrave Dictionary of Economics (London: MacMillan, 1998); Eirik G. Furubotn and Rudolf Richter, Institutions and Economic Theory: The Contribution of the New Institutional Economics (2nd edn, Ann Arbor: University of Michigan Press, 2005), pp. 318–26; and Hans-Bernd Schäfer and Claus Ott, Lehrbuch der ökonomischen Analyse des Zivilrechts (5th edn, Berlin: Springer, 2012), p. 465 et seq. 30 See, e.g., Colin Camerer, Behavioral Game Theory: Experiments in Strategic Interaction (Princeton, NJ: Princeton University Press, 2003); Colin Camerer, George Loewenstein, and Matthew Rabin (eds), Advances in Behavioral Economics (New York: Russell Sage Foundation/Princeton and Oxford: Princeton University Press, 2004); and more recently, George A. Akerlof and Robert J. Shiller, Phishing for Phools. The Economics of Manipulation and Deception (Princeton and Oxford: Princeton University Press, 2015); Richard Thaler, Misbehaving. The Making of Behavioral Economics (New York and London: W. W. Norton & Company, 2015). 31 For an overview of bounded rationality, and the classical distinction between bounded rationality, bounded willpower, and bounded self-interest, see, e.g., Richard Thaler, ‘Doing Economics Without Homo Economicus’ in Steven G. Medema and Warren J. Samuels (eds) Foundations of Research in Economics: How Do Economists Do Economics? (1996), pp. 227 et seq.; for cognitive capacity limits, see, e.g., Nelson Cowan, ‘The Magical Number 4 in Short-Term Memory: A Reconsideration of Mental Storage Capacity’ (2000) Behav. & Brain Sci. 24, 87; Baddeley, ‘Working Memory: Theories, Models, and Controversies’ (2012) Ann. Rev. Psychol. 63, 1. 32 See, e.g., Martin J. Eppler and Jeanne Mengis, ‘The Concept of Information Overload: A Review of Literature from Organization Science, Accounting, Marketing, MIS, and Related Disciplines’ (2004) Info. Soc’y. 20, 325, 326. 33 Angela Edmunds and Anne Morris, ‘The Problem of Information Overload in Business Organizations: A Review of Literature’ (2000) Int’l J. Info. Mgmt. 20, 17, 19. 34 George A. Miller, ‘The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity For Processing Information’ (1956) Psychol. Rev. 63, 81; see further Robert S. Owen, ‘Clarifying the Simple Assumption of the Information Load Paradigm’ (1992) Advances Consumer Res. 19, 770, 773 (noting that evidence supports Miller’s magical number ‘as a rough benchmark’); see also Cowan (n. 32); Baddeley (n. 32), 15. 35 Cowan (n. 32); Alan Baddeley (n. 32); on the upper end, in some consumer choice situations 10 product alternatives or information on 15 attributes of a product were reported

as a limit: Naresh K. Malhotra, ‘Information Load and Consumer Decision Making’ (1982) J. Consumer Res. 8, 419, 427. 36 On limited attention, see, e.g., Christopher Chabris and Daniel Simons, the invisible gorilla [sic] (New York: Crown, 2010). 37 Herbert Simon, ‘A Behavioral Model of Rational Choice’ (1955) Quarterly Journal of Economics 69, 99, 115 (‘approximate rationality’); Simon, ‘Theories of Bounded Rationality’ in C. B. McGuire and Roy Radner (eds) Decision and Organization (Amsterdam and London: North Holland, 1972), pp. 161, 163. 38 Cf. also Kumpan and Leyens (n. 1), 89 who name overconfidence as a further possible source of concern. 39 On optimism bias, see Neil D. Weinstein, ‘Unrealistic Optimism About Future Life Events’ (1980) Journal of Personality and Social Psychology 39, 806; Neil D. Weinstein and William M. Klein, ‘Unrealistic Optimism: Present and Future’ (1996) Journal of Social and Clinical Psychology 15, 1; for a cautionary note from a mathematical perspective, however, see Adam G. L. Harris and Ulrike Hahn, ‘Unrealistic Optimism about Future Life Events: A Cautionary Note’ (2011) Psychological Review 118, 135. 40 See Charles Lord, Mark R. Lepper, and Elizabeth Preston, ‘Considering the Opposite: A Corrective Strategy for Social Judgment’ (1984) J. Personality & Soc. Psychol. 47, 1231 (calling the effect ‘biased assimilation of new evidence’); Raymond S. Nickerson, ‘Confirmation Bias: A Ubiquitous Phenomenon in Many Guises’ (1998) Rev. Gen. Psychol. 2, 175. 41 Nick Chater, Steffen Huck, and Roman Inderst, Consumer Decision-Making in Retail Investment Services: A Behavioural Economics Perspective. Final Report, November 2010, available at , at 385. 42 ibid., at 343; trust is lower when advisers are biased, however: ibid., at 371. 43 ibid., at 385. 44 On perseverance, see, e.g., Cathy McFarland, Adeline Cheam, and Roger Buehler, ‘The Perseverance Effect in the Debriefing Paradigm: Replication and Extension’ (2007) Journal of Experimental Social Psychology 43, 233. 45 See Hacker (n. 19); Hacker, Overcoming the Knowledge Problem in Behavioral Law and Economics: Uncertainty, Decision Theory, and Autonomy (17 July 2015), available at . 46 This is a characteristic of the function they introduce to formalize probability weighting, see Daniel Kahneman and Amos Tversky, ‘Prospect Theory: An Analysis of Decision under Risk’ (1979) Econometrica 47, 263, 281; for a more nuanced assessment of the treatment of small probabilities, however, see Colin Camerer and Howard Kunreuther, ‘Decision Processes for Low Probability Events: Policy Implications’ (1989) Journal of Policy Analysis and Management 8, 565. 47 See, e.g., Amos Tversky and Daniel Kahneman, ‘Availability: A Heuristic for Judging Frequency and Probability’ (1973) Cognitive Psychol. 5, 207; Baruch Fischhoff, Paul Slovic, and Sarah Lichtenstein, ‘Fault Trees: Sensitivity of Estimated Failure Probabilities

to Problem Representation’ (1978) J. Experimental Psychol.: Hum. Perception & Performance 4, 330; Amos Tversky and Daniel Kahneman, ‘Extensional versus Intuitive Reasoning: The Conjunction Fallacy in Probability Judgment’ (1983) Psychol. Rev. 90, 293. 48 The authors have made two proposals that are differently nuanced with respect to the treatment of uncertainty concerning the existence, direction, and intensity of biases, see on one hand Stefan Grundmann, ‘Targeted Consumer Protection’ in Dorota Leczykiewicz and Stephen Weatherill (eds) The Images of the Consumer in EU Law (Oxford: Hart, 2016), pp. 223 et seq (holding that a fully rational model should be chosen unless danger for existential goods, such as a threat to health or large financial losses, looms), and on the other hand, Hacker (n. 46) (arguing that regulatory theory should by default adopt a pluralistic model of rationality that posits the simultaneous presence of both rational and boundedly rational actors and comparatively weighs the impact on the different groups). 49 See, e.g., David Hirshleifer, Behavioral Finance, MPRA Paper No. 59028, available at ; Nicholas C. Barberis and Richard H. Thaler, ‘A Survey of Behavioral Finance’ in Richard H. Thaler (ed.) Advances in Behavioral Finance, Vol. II (New York, Princeton, and Oxford: Princeton University Press, 2005), 1 et seq. 50 Thaler (n. 31), Conclusion. 51 See Kumpan and Leyens (n. 1), 85. 52 See nn. 71 et seq. and accompanying text. 53 See Section II.2.C. 54 Yet another feature where banking company law reaches considerably beyond general contract law, is, of course, the imposition on boards to take financial stability as the guideline of action which takes precedence over all others: see, among others, Jens-Hinrich Binder, ‘Vorstandshandeln zwischen öffentlichem und Verbandsinteresse – Pflichten- und Kompetenzkollisionen im Spannungsfeld von Bankaufsichts- und Gesellschaftsrecht’ (2013) Zeitschrift für Gesellschaftsrecht (ZGR) 760–801; Stefan Grundmann, ‘The Banking Union Translated into (Private Law) Duties: Infrastructure and Rulebook’ (2015) EBOR 16, 357–82, 369–79. 55 In relation to the exploration of the legal control on managerial discretion, see J. E. Parkinson, Corporate Power and Responsibility: Issues in the Theory of Company Law (Oxford: Oxford University Press, 1995), pp. 73–96; for an economic analysis of managerial discretion, see Oliver Williamson, ‘Managerial Discretion and Business Behaviour’ (1963) The American Economic Review 53(5), 1032–57; for an up-to-date empirically oriented paper pertaining to the role of the board’s discretion in a company’s internal organization, see D. B. Wangrow, D. J. Schepker, and V. L. Barker, ‘Managerial Discretion: An Empirical Review and Focus on Future Research Directions’ (2015) Journal of Management 45(1), 99–135. 56 Commission Directive 2006/73/EC of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards organisational

requirements and operating conditions for investment firms and defined terms for the purposes of that Directive, OJ 2006 L 241/26 (hereinafter MiFID I ID). 57 See, for instance, Articles 74, 88, and 91 of the Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms (CRD IV), OJ 2013 L 176/338. 58 Cf. CESR’s Technical Advice (n. 4), 41; see also Stefan Grundmann, ‘Das Wertpapier- und Effektengeschäft’ in Karlheinz Ebenroth, Carsten Thomas Boujong, Detlev Joost, and Lutz Strohn (eds) Handelsgesetzbuch (Munich: C. H. Beck/Franz Vahlen, 3rd edn, 2015), Vol. 2, n. VI 294. 59 For more on product of governance under MiFID II, see, e.g., Hacker (n. 19), Part 3, under Product Governance; Hartmut Renz, Ingrid Kalisch, Sandra Pfister, Stuart Axford, and George M. Williams, Jr., ‘ESMA Opinion on Structured Retail Products: Good Practices for Product Governance Arrangements’ (2014) Journal of Investment Compliance 15, 19. 60 Harry McVea, Financial Conglomerates and the Chinese Wall: Regulating Conflict of Interest (Oxford: Clarendon Press, 1993), pp. 122–234; Thomas Möllers, ‘Kölner Kommentar zum Wertpapierhandelsgesetz (WpHG)’ (2nd edn, Cologne: Heymanns, 2014), § 33, para. 69 et seq.; in more detail Thomas Lösler, Compliance im Wertpapierdienstleistungskonzern (Berlin: De Gruyter, 2003), pp. 73–92. 61 Lösler (n. 61), 79–84; Dieter Eisele and Alexander Faust, Bankrechtshandbuch (4th edn, Munich: Beck, 2011), § 109, para. 141 et seq. 62 ESMA, Final Report: ESMA’s Technical Advice to the Commission on MiFID II and MiFIR, ESMA/2014/1569, December 2014, at 83 (emphasis added). 63 ibid., at 82. 64 Cf. European Commission, Public Consultation: Review of the Markets in Financial Instruments Directive (MiFID), December 2010, at 68–9; ESMA, Consultation Paper MiFID II/MiFIR, ESMA/2014/549, May 2014, at 41. 65 See, e.g., Martin Brenncke, ‘Der Zielmarkt eines Finanzinstruments nach MiFID II’ (2015) Wertpapiermitteilungen (WM) 1173, 1174–5. 66 ESMA (n. 63) Final Report ESMA’s Technical Advice to the Commission on MiFID II and MiFIR, ESMA/2014/1569, December 2014, at 60 nn. 27–8. 67 See for an assessment of the general legality of offering products outside the target market ESMA (n. 63) 52, 59. 68 This is the view of Brenncke (n. 66), 1173, 1179. 69 See ESMA (n. 63), 52 n. 6. 70 ESMA (n. 63), 82. 71 ESMA (n. 63), 82. 72 See, e.g., Grundmann (n. 59), n. VI 315; contra Christoph Kumpan and Alexander Hellgardt, ‘Haftung der Wertpapierdienstleistungsunternehmen nach Umsetzung der EU-

Richtlinie über Märkte für Finanzinstrumente (MiFID)’ (2006) Der Betrieb (DB) 1714, 1715–16. 73 Chater, Huck, and Inderst (n. 42), 7. 74 Roman Inderst and Marco Ottaviani, ‘Regulating Financial Advice’(2012) European Business Organization Law Review 13, 237, 241. 75 It did appear in the CESR’s Technical Advice (n. 4), 53. 76 Again, the same provision can be found in the MiFID I ID, Article 22(4), which was moved up to Level 1 in MiFID II. Therefore, under MiFID I, a minority view already held that disclosure has to be such that the mechanism for how the conflict of interest works and why lack of neutrality has to be feared has to be disclosed: see namely, even before introduction of Level 2 legislation: Grundmann (n. 59), n. VI 227 (also in the previous editions and with further references). 77 ESMA (n. 63), 80. 78 Consumer Financial Protection Bureau/Kleimann Communication Group, Inc., Know Before You Owe. Evolution of the Integrated TILA-RESPA Disclosures, July 2012, available at , at 7 and 175; see also James M. Lacko and Janis K. Pappalardo, Improving Consumer Mortgage Disclosures: An Empirical Assessment of Current and Prototype Disclosure Forms, Federal Trade Commission Bureau of Economics Staff Report, June 2007, available at . 79 Huseyn Ismayilov and Jan Potters, ‘Disclosing Advisor’s Interests neither Hurts nor Helps’(2013) Journal of Economic Behavior & Organization 93, 314, 317. 80 ESMA (n. 63), 83. 81 See, e.g., Enriques (n. 1), 331. 82 CESR’s Technical Advice (n. 4), 44. 83 Enriques (n. 1), 331. 84 Chater, Huck, and Inderst (n. 42), at 385. 85 Enriques (n. 1), 331. 86 For a more detailed treatment of inducements, see Chapter 8; for the division between dependent and independent services, Chapter 6; see also brief discussion of the need for the market structure (‘governance’) approach on remuneration in Rik Mellenbergh, ‘MiFID II: New Governance Rules in Relation to Investment Firms’ (2014) European Company Law 11, 172. 87 See George A. Papaconstantinou, ‘Investment Bankers in Conflict: The Regime of Inducements in MiFID II and the Member States’ Struggle for Fairness’ (2016) European Review of Contract Law (forthcoming); for the UK, see also the Retail Distribution Review commenced in 2006 and implemented in 2012 with the aim of raising minimum qualifications for advisers, improving transparent disclosure of charges, and removing commission payments to advisers: ; for Germany, see short survey on the large number of decisions rendered by the Supreme Court on Inducements: Grundmann (n. 59), n. VI 287–90. 88 Cf. Inderst and Ottaviani (n. 75), 237, 239. 89 See Inderst and Ottaviani (n. 75), 237, 240. 90 See n. 88 and accompanying text. 91 Europe Economics, Retail Distribution Review Post Implementation Review, 16 December 2014, available at , p. 2. 92 Michelle Abrego, ‘Lloyds U-Turns and Scraps Plans for Mass Market Advice’, New Model Adviser (27 September 2012), available at . 93 See, e.g., Towers Watson, Advice Gap Analysis: Report to FCA, 5 December 2014, available at , particularly pp. 8–9. 94 Chater, Huck, and Inderst (n. 42), 385. 95 Article 44(3) Commission Directive 2006/73/EC of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards organisational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive, OJ 2006 L 241/26. 96 On the difficulty to assess the quality of an execution in a multi-dimensional scheme taking into account factors other than price, see Guido Ferrarini, ‘Contract Standards and the Markets in Financial Instruments Directive (MiFID): An Assessment of the Lamfalussy Regulatory Architecture’ (2005) ERCL 1, 19, 38; Ferrarini, ‘Best Execution and Competition between Trading Venues: MiFID’s Likely Impact’ (2007) Capital Markets Law Journal 2, 404, 407–9. 97 Guido Ferrarini, ‘Best Execution and Competition between Trading Venues—MiFID’s Likely Impact’ (2007) Capital Markets Law Journal 2, 404, 407. 98 ESMA, Final Report: Guidelines on cross-selling practices, ESMA/2015/1861 (December 2015), at 36–7. 99 Loci classici include Ward S. Bowman, Jr., ‘Tying Arrangements and the Leverage Problem’ (1957) Yale Law Journal 67, 19; Donald F. Turner, ‘The Validity of Tying Arrangements under the Antitrust Laws’ (1958) Harvard Law Review 72, 50; for a more recent treatment, see Jean Tirole, ‘The Analysis of Tying Cases: A Primer’ (2005) Competition Policy International 1, 1; W. David Slawson, ‘Excluding Competition without Monopoly Power: The Use of Tying Arrangements to Exploit Market Failure’ (1991) Antitrust Bulletin 36, 457. 100 See also ESMA (n. 99), at 39. 101 ESMA (n. 99). 102 ESMA (n. 99), 38. 103 ESMA (n. 99), 40. 104 ESMA (n. 99), 41.

105

See William Samuelson and Richard Zeckhauser, ‘Status Quo Bias in Decision Making’ (1988) Journal of Risk and Uncertainty 1, 7. 106 ESMA (n. 99), 41–2. 107 Omri Ben-Shahar and Carl E. Schneider, More Than you Wanted to Know (Princeton and Oxford: Princeton University Press, 2014); see also Hacker (n. 19), Part 3. 108 See, e.g., the exemplary study of the US Consumer Financial Protection Bureau on the effectiveness of different mortgage disclosure formats (n. 79). 109 Ismayilov and Potters (n. 80) 314, 318. 110 See Hacker (n. 19), Part 3. 111 ESMA (n. 99), 38. 112 See, e.g., Regulation (EU) No 1286/2014 of the European Parliament and of the Council of 26 November 2014 on key information documents for packaged retail and insurance-based investment products (PRIIPs), OJ 2014 L 352/1. 113 Omri Ben-Shahar and Adam S. Chilton, Simplification of Privacy Disclosures: An Experimental Test, Working Paper (2016), available at . 114 Chater, Huck, and Inderst (n. 42), at 336–7. 115 Geneviève Helleringer, Trust Me, I Have a Conflict of Interest! Testing the Efficacy of Disclosure in Retail Investment Advice, Working Paper, 2016, available at ; the results do not show how far actual investment decisions would follow the stated changes in attitudes analysed by the study; however, the study makes such behavioural change at least a plausible result of enhanced disclosure. 116 The explanation reads: ‘For your information, your financial advisor will be paid a percentage of your total investment. Please note that this means that the advisor did not necessarily have your own investment earnings in mind when giving advice.’ Helleringer (n. 116), 41. 117 These include the possibility to compare offers, advice, and disclosures among different statements advertising advised investment opportunities, see Helleringer (n. 116), 21. 118 Inderst and Ottaviani (n. 75), 237, 241–2. 119 Christine Jolls and Cass R. Sunstein, ‘Debiasing through Law’ (2006) Journal of Legal Studies 35, 199. 120 See for a theoretical account of such asymmetrical effects of behavioral interventions Camerer, Issacharoff, Loewenstein, O’Donoghue, and Matthew Rabin, ‘Regulation for Conservatives: Behavioral Economics and the Case for “Asymmetric Paternalism”’ (2003) University of Pennsylvania Law Review 151, 1211. 121 See Hacker (n. 19), Part 3. 122 See, for successful debiasing strategies through warnings, e.g., Marc Alpert and Howard Raiffa, ‘A Progress Report on the Training of Probability Assessors’ in Daniel Kahnemann, Paul Slovic, and Amos Tversky (eds), Judgment under Uncertainty: Heuristics and Biases (Cambridge, UK: Cambridge University Press, 1982), p. 294; for

unsuccessful strategies, e.g., Neil D. Weinstein and William M. Klein, ‘Resistance of Personal Risk Perceptions to Debiasing Interventions’ in Thomas Gilovich, Dale Griffin, and Daniel Kahneman (eds) Heuristics and Biases: The Psychology of Intuitive Judgment (Cambridge, UK: Cambridge University Press, 2002), p. 313; more generally Hacker (n. 19), Part 3; Baruch Fischhoff, ‘Debiasing’ in Kahnemann, Slovic, and Tversky (eds) (ibid.), p. 422; Richard P. Larrick, ‘Debiasing’ in Derek J. Koehler and Nigel Harvey (eds) Blackwell Handbook on Judgment and Decision Making (Malden, MA.: Blackwell Publishing, 2004), p. 316. 123 Chater, Huck, and Inderst (n. 42). 124 Inderst and Ottaviani (n. 75), 237, 241–2. 125 Oren Bar-Gill, ‘Price Caps in Multi-Price Markets’ (2015) The Journal of Legal Studies 44, 4 53.

8 Inducements Larissa Silverentand, Jasha Sprecher, and Lisette Simons

I. Introduction II. 1. 2. 3.

Current Legislation MiFID I MiFID I Implementing Directive Guidance by ESMA

III. 1. 2. 3.

The Dutch Inducement Ban Introduction The Dutch Inducement Ban Dutch Inducement Rules for Investment Services Outside the Scope of the Inducement Ban

IV. 1. 2. 3.

The UK Inducement Rules Introduction MiFID I Implementation Retail Distribution Review

V. MiFID II 1. Introduction 2. Investment Advice on an Independent Basis and Portfolio Management 3. Other Investment Services and Ancillary Services 4. Deviating Requirements VI. Research as an Inducement 1. Introduction

2. Definition of ‘Research’ 3. Research as an Inducement VII. Conclusion

I. Introduction [8.01] Pursuant to the Markets in Financial Instruments Directive1 (MiFID I), an investment firm must act honestly, fairly, and professionally in accordance with the best interests of its clients.2 This general duty forms the basis for all conduct rules applicable to the provision of investment services. [8.02] With the entry into force of MiFID I, rules regarding inducements were introduced as part of a set of more detailed rules expanding on this general duty of care. Inducements are payments or other benefits received by or paid for by an investment firm in relation to investment services. When an investment firm receives payments or benefits from other persons or entities than its clients, there can be a conflict between the interests of the investment firm and the interests of the client. In such case, there is a concern whether the investment firm will be led by its obligation to act in accordance with the best interests of its clients, or whether the investment firm will be led by its own interests in receiving payments or other benefits from other persons. As follows from the reference to ‘benefits’, it is not only payments made to an investment firm that may bring about such a conflict of interests; non-monetary benefits, such as free trips or gifts, may also lead to reward-driven behaviour by the investment firm. [8.03] Ever since the introduction of the inducement rules in MiFID I, the inducement rules have led to a lot of debate and discussion.3 This chapter provides an overview of the current rules on inducements, as well as the changes under the Markets in Financial Instruments Directive II4 (MiFID II).

II. Current Legislation 1. MiFID I [8.04] As set out above in paragraph 8.01, the inducement ban finds its origin in MiFID I. MiFID I was adopted in April 2004 and the national legislation implementing MiFID I was due to apply as of November 2007. MiFID I prescribes amongst others inducement rules. These rules are intended to better protect customers against reward-driven advice behaviour, in view of the fact that at that time many investment firms were being partly paid by issuers or providers of financial instruments. [8.05] As set out above, the legal basis for the inducement ban is the loyalty obligation laid down in MiFID I. This obligation stipulates that investment firms, when providing investment services and/or ancillary services, must act honestly, fairly, and professionally, in accordance with the best interests of the client.5 This general duty is set out in more detail in both MiFID I itself as well the MiFID I Implementing Directive.6 It includes, in addition to the inducement rules, more detailed information requirements and the know-your-customer obligation.7 In the following section, we will describe the rules set out in the MiFID I Implementing Directive in more detail.

2. MiFID I Implementing Directive [8.06] In August 2006, the European Commission adopted Level 2 implementing measures, including a regulation and a Directive, to further specify the requirements and obligations laid down in MiFID I. The Level 2 Directive (the ‘MiFID I Implementing Directive’) introduced the inducement ban. Pursuant to this Directive, investment firms are not regarded as acting honestly, fairly, and professionally in accordance with the best interests of a client if they pay or are paid any fee or commission, or provide or are provided with any non-monetary benefit. As a result, the

Directive contains, as a starting point, a general prohibition on inducements. There are, however, three exceptions to this prohibition:8 1. when the inducement is paid or provided to or by the client (or a person on behalf of the client); 2. when the inducement is paid to or by a third party, provided that each of the following conditions are met: i. prior to the provision of the investment service, the existence, nature, and amount of the inducement must be disclosed to the client in a comprehensive and understandable manner. An investment firm may disclose the required information in summary form, provided that it provides more detailed information at the client’s request;9 ii. the payment must be designed to enhance the quality of the relevant service to the client; and iii. the inducement does not impair the investment firm’s duty to act in the best interest of the client. 3. when the inducements are proper and necessary for the provision of investment services, such as custody costs, settlement and exchange fees, regulatory levies, or legal fees, and which, by their nature, cannot give rise to conflicts with the investment firm’s duties to act honestly, fairly, and professionally in accordance with the best interests of its clients. [8.07] The conditions included in this Article are fairly open-ended. As a result, the introduction of the inducement ban caused quite a bit of confusion and uncertainty in practice. Specifically the question under what circumstances an inducement enhances the quality of the provision of the service and does not impair the firm’s duty to act in the best interest of the client was discussed extensively.

3. Guidance by ESMA [8.08] The Committee of European Securities Regulators (currently and hereafter referred to as the European Securities and Markets Authority, ‘ESMA’) adopted recommendations10 (the ‘ESMA Recommendations’) on

inducements in May 2007. The main objective of these recommendations was to provide guidance on inducements and, consequently, ensure a consistent implementation and application of the MiFID Implementing Directive.11 In what follows, we will elaborate upon the conditions set out in the provision on inducements on the basis of the ESMA Recommendations and its ‘Good and Bad Practices’, of April 2010 (the ‘ESMA Practices’).12 First, we will discuss the scope of the inducement ban, and second, we will analyse the application of the exceptions thereto.

A. Scope of the Inducement Ban i. Standard Fees [8.09] Standard fees are payments that are considered standard in the specific market. In consultations preceding the ESMA Recommendations, market participants had argued that such standard fees should be placed outside the scope of the inducement ban.13 It was argued, for example, that normal business arrangements between two investment firms, whereby one firm pays the other, should fall outside the scope of the inducement ban, as such payments should be considered part of normal business, rather than inducements. It was argued that the provision in the MiFID I Implementing Directive with respect to the conflict of interest, which stated that ‘other than the standard commission or fee for that service’, implied that standard fees should be allowed.14 ESMA, however, stipulated in its recommendations that Article 26 of the MiFID I Implementing Directive covers (emphasis added): ‘any fee or commission or non-monetary benefit that an investment firm may receive or pay in connection with the provision of investment and ancillary services to clients including commissions or fees that […] are standard in the market’.15 It is clear from this statement that the inducement ban has a broad scope of application. ii. Intra-group Payments [8.10] In addition to that, ESMA stated in its ESMA Recommendations that the inducement rules also apply to a payment provided to or made by a legal entity within the same group as the investment firm providing the investment service.16 This means that intragroup payments for investment services are, in principle, not excluded from

the scope of the inducement ban. In the ESMA Practices, however, ESMA took a different position on payments with regard to tied agents, which can be intra-group payments. A tied agent has a special status under MiFID, as it acts under the full and unconditional responsibility of the investment firm. It does not have its own licence, but it acts under the licence of the investment firm. ESMA stipulates in the ESMA Practices that, under certain circumstances, payments made from the investment firm to the tied agent may be seen as an internal payment. Consequently, the inducement ban does not apply to such payments. However, if such payment is related to a payment from a third party, such as a product provider, to the investment firm, the whole of the payment from the product provider has to be considered as a third-party payment, which falls within the scope of the inducement ban.17

B. Exceptions i. Direct Fees [8.11] The first exception to the inducement ban are fees that are paid by clients (we refer to this category as ‘direct fees’). With regard to direct fees, ESMA clarified that a fee is (also) deemed to be provided by the client when a third party pays the fee on behalf of the client. An example of such a situation is a lawyer or accountant paying an invoice for a client relating to the investment service. Relevant in this respect is whether there is a specific instruction by the client to the third party to make or receive a payment. In addition, ESMA stipulates that the fact that the client bears the economic cost of the fee is not by itself sufficient to be regarded as provided by the client, and, consequently, to qualify as a direct fee.18 There should be other factors indicating that the payment is made directly by or to the client. ii. Fees Enhancing the Quality of the Service [8.12] As set out above, inducements are also allowed if two conditions are met. First, the fee should enhance the quality of the service and should not impair the obligation to act in the client’s best interests. Second, prior to the provision of the investment service, the fee should be disclosed to the client in a clear manner.19

[8.13] With respect to the first condition, the following factors are, amongst others, relevant in determining whether fees/commissions can be considered to be designed to enhance the quality of the service and not to impair the firm’s duty to act in the best interest of the client: (i) the type of service and any specific duties the investment firms owes to the client, (ii) the expected benefit to the client versus the expected benefit for the investment firm, (iii) the existence of an incentive for the investment firm not to act in the client’s best interest, (iv) the relationship between the investment firm and the entity which is receiving or providing the benefit, and (v) the nature of the payment, that is, does it cause any conflicts with the conduct of business obligation to act honestly, fairly, and professionally in the best interest of the client.20 [8.14] With regard to the second condition, the requirement to disclose the fees, it is important that the investment firm discloses the details of the payment to the client. This should be done in such way as to enable the client to have clear insight into the payment structure. As set out above, an investment firm may disclose the required information in summary form, as long as it provides more detailed information at the client’s request. iii. Proper Fees [8.15] Finally, there is a third exception, which we shall refer to as ‘proper fees’.21 In order for a fee to fall under the third exception, the following two conditions must be fulfilled: (i) the payment must enable or be necessary for the provision of the service, and (ii) the fee, by its nature, cannot give rise to conflicts with the firm’s duty to act honestly, fairly, and professionally in accordance with the best interests of the client. [8.16] In order to pass the test referred to under (ii), the nature of the payment should be kept in mind rather than the question of whether the result of the payment actually gives rise to a conflict for the investment firm to act in the client’s best interest.22 [8.17] Article 26 of the MiFID I Implementing Directive refers to custody costs, settlement and exchange fees, regulatory levies, and legal fees as proper fees. According to ESMA, the list of examples stated in the article itself is not exhaustive. For other fees it should be determined whether they

meet the two requirements set out above.23 ESMA indicates that payments that are made in order to access and operate on a given execution venue should also be considered proper fees.24 However, the ESMA Practices indicate that this category of fees should not be considered as a broad category and should be applied strictly.25

III. The Dutch Inducement Ban 1. Introduction [8.18] After the financial crisis, both the Dutch legislator and the AFM intensified their scrutiny on inducements in relation to all financial products available in the Netherlands (such as financial instruments, insurance products, and credit products). This led to an inducement ban on a wide range of financial products in the Netherlands in 2013, with the exception of financial instruments. Financial instruments were not covered by the inducement ban, because MiFID I already provided for rules on inducements and the Dutch government intended for this to be part of MiFID II. In the meantime, however, the AFM did not want to wait for a Dutch or European ban on inducements for financial instruments. The AFM therefore approached the largest retail banks in the Netherlands and announced in early 2013 that it had come to a ‘voluntary’ agreement that these banks would no longer receive distribution fees with respect to investment funds from 1 January 2014. In the same period it became apparent to the Dutch legislator that the inducement ban it had envisaged had not received sufficient political support at a European level in MiFID II. In response, the legislator introduced the Dutch ban on inducements in relation to investment services from 1 January 2014, effectively rendering the AFM agreement moot. In the following paragraph, we will describe the Dutch ban on inducements in more detail.

2. The Dutch Inducement Ban

[8.19] The inducement ban for investment firms is set out in the Dutch Market Conduct Decree,26 which is the ‘Dutch Level 2’ legislation for code of conduct rules. Just as in MiFID I, the inducement ban (and other inducement rules for investment firms) is a further elaboration of the general duty of an investment firm to act honestly, fairly, and professionally in accordance with the best interests of the client. The Dutch inducement ban is only applicable in relation to investment services provided to retail investors. For professional investors the (Dutch implementation of) MiFID inducement rules are still applicable (we refer to paragraph 8.06 above). [8.20] The scope of the inducement ban is quite broad. It prohibits investment firms to directly or indirectly pay and receive any inducement in relation to the provision of investment services or ancillary services to a retail client. Inducements are defined as any remuneration or fee in any form in relation to the provision of an investment service or ancillary service. The inducement ban applies to all ancillary services and to the following investment services: (i) (ii) (iii) (iv)

reception and transmission of orders; execution of orders on behalf of clients; investment advice; and portfolio management.

[8.21] There are only a few exceptions to the inducement ban: • inducements in relation to investment services/ancillary services provided to professional investors (see a further description below); • inducements in relation to placing/underwriting services (see a further description below); • business gifts that do not exceed €100 on an annual basis; • inducements that are necessary for the service provided or inducements that enable the provision of the service. In practice, this exception does not provide much room for parties to pay or receive inducements, because this exception is interpreted in a very strict way by the regulators. Examples of inducements that are permitted in this category are custody fees, settlement fees, transaction fees of a trading venue, statutory fees, and legal fees;27

• fees paid to and by tied agents; and • inducements paid to a crowdfunding platform. This exception was introduced in 2016 to take away hurdles for equity based crowdfunding platforms that effectively provide the investment service of reception and transmission of orders. [8.22] The effect of the inducement ban is that only payments made by the retail client directly to the investment firm are a permitted way of receiving income for the investment firm. In practice the prohibition on receiving fees has proved to be especially burdensome. This part of the ban not only prohibits the retention of the fees by the investment firm, but also prohibits the firm from receiving the inducement and subsequently paying such inducements on to the client. This has resulted in a substantial effort of Dutch distributors to move all their clients into non-rebate-paying financial instruments and sometimes resulted in firms requiring their clients to sell instruments that could not be made ‘rebate free’. The choice of an inducement ban that allowed for on-payment of inducements would have been far less burdensome with the same end result for the retail client. [8.23] The investment services of underwriting/placement with or without a firm commitment basis fall outside the scope of the inducement ban. We refer to paragraph 8.06 for a description on the inducement regime applicable to these investment services. The Dutch legislator has given several arguments for excluding these services from the scope of the inducement ban.28 First, because these are services that relate to the primary markets and not the secondary markets. Second, the fees paid in the context of these investment services are generally paid by the investment firm for the preparation of the prospectus and research reports, the structuring of transactions, and assuming risks. In the eyes of the Dutch legislator, this does not run counter to the purpose of the inducement ban.

A. Territorial Scope [8.24] The Dutch inducement ban applies to investment firms as defined in the Dutch Financial Supervision Act (DFSA). This includes almost all investment firms that provide investment services in the Netherlands

(including banks, third-country investment firms, and branches of EEA investment firms operating under a European passport). Investment firms active in the Netherlands that are not subject to the Dutch inducement ban are EEA investment firms making use of their European passport to provide investment services in the Netherlands on a cross-border basis. The reason for this is that MiFID I provides that Member States may not impose any additional requirements on such investment firms where this concerns matters covered by MiFID I.29

B. Legal Basis and Necessity for ‘Gold Plating’ [8.25] MiFID I provides for maximum harmonization. This means that national rules may not impose stricter rules than those laid down in MiFID I, unless such is explicitly provided for in MiFID I. Because the Dutch inducement ban is clearly stricter than the inducement rules provided for in MiFID I, the Dutch legislator needed to find a legal basis for imposing these stricter rules. As discussed in paragraph 8.06, the MiFID I inducement rules follow from Article 26 MiFID I Implementing Directive, which does not provide for a Member State option to impose stricter rules. The Dutch legislator therefore decided to make use of the option provided by Article 4 MiFID I Implementing Directive, allowing Member States to impose additional requirements on investment firms. A Member State must be able to demonstrate that certain exceptional circumstances require the Member State to impose such additional rules. In short, the Member State may use this option: (a) in an exceptional case; (b) if such requirement is objectively justified and proportionate; (c) if such requirement addresses risks to investor protection or market integrity; (d) if such requirement is not adequately addressed by the MiFID I Implementing Directive; and (e) (i) if the risks addressed are of particular importance looking at the market structure of the Member State, or (ii) the risk has emerged/became evident after the date of application of the MiFID I Implementing Directive.

[8.26] In the view of the Dutch legislator, the measures under the MiFID I Implementing Directive were not adequate to prevent negative incentives of inducements.30 Furthermore, in its view transparency was not an adequate measure to balance the information gap between market parties and clients, especially in respect of the influence of inducements on services provided by investment firms. The Dutch legislator furthermore argued that the deficiency in the rules is evidenced by the fact that MiFID II (which was at that point still under negotiation) provides for stricter rules and a partial inducement ban and that for instance in the United Kingdom measures had been taken to ban certain inducements. Finally, the Dutch legislator argued that Netherlands market parties were relatively dependent on the revenue source of inducements and that banning inducements would be favourable for competition on the Dutch market because of greater price transparency. [8.27] In our view, it can be questioned whether the argumentation of the Dutch legislator meets the (seemingly) strict criteria of the exception provided by Article 4 of the MiFID I Implementing Directive. For instance, one could question whether the inducement ban is proportionate. In particular, the ban on paying-on fees seemed to be disproportionately burdensome on market parties without any added benefit. Having said that, the inducement ban in the Netherlands has so far not been challenged nationally or in a European context. [8.28] Because MiFID II includes a very similar possibility for stricter inducement rules in Article 24(12), it is likely that the Dutch legislator will therefore continue to rely on this exception and argumentation under MiFID II in imposing a stricter inducement ban than introduced by MiFID I (we refer to paragraph 8.05 of this chapter). Unlike the exception provided for in Article 4 of the MiFID I Implementing Directive, MiFID II requires the European Commission to provide its opinion on the proportionality of and justification for the additional requirement. Unfortunately, we will not see whether the European Commission deems such stricter rules to be proportionate and justifiable in the light of the inducement requirements that have been agreed on at a European level in MiFID II, given the fact that MiFID II provides Member States with the option to continue to rely on stricter rules that were set under Article 4 of the MiFID I Implementing Directive prior to the implementation of MiFID II.

3. Dutch Inducement Rules for Investment Services Outside the Scope of the Inducement Ban [8.29] As discussed above, the Dutch inducement ban does not apply where it concerns the investment services of placement or underwriting (with or without a firm commitment) and in respect of services provided to professional investors. The Dutch implementation of the MiFID I inducement rules applies to these categories of services. In short, this means that inducements are permitted if the client is informed of the nature and amount of the inducement and the inducement does not negatively affect the quality of the service provided or impair the investment firm’s compliance with its duty to act in the best interests of the client. For a more detailed description of these rules we refer to our discussion thereof in paragraphs 8.06 et seq.

IV. The UK Inducement Rules 1. Introduction [8.30] Similar to the Netherlands, the United Kingdom has also implemented rules providing for stricter requirements in relation to inducements than what is set out in MiFID I and the MiFID I Implementing Directive. In this section we will describe the rules applicable in the United Kingdom and the scope thereof.

2. MiFID I Implementation [8.31] The UK implemented the conduct of business obligations laid down in MiFID I in the Conduct of Business Sourcebook (COBS). Article 26 of the MFID I Implementing Directive has been implemented in COBS 2.3. This clause is effectively identical to Article 26 of the MiFID I Implementing Directive, but, in addition, also elaborates on the application of the provision to non-MiFID-business. This provision was effective as of

1 November 2007. Hereinafter, we refer to this provision as the ‘MiFID implementation provision’.

3. Retail Distribution Review [8.32] In addition to the implementation of MiFID I, the Financial Services Authority (the predecessor of the Financial Conduct Authority, ‘FCA’) started the Retail Distribution Review (RDR) in 2006. The RDR was aimed at improving the retail financial investment market and increasing consumers’ confidence and trust in the market. The main objectives of the RDR were: (i) raising the level of qualification and professionalism of advisers, (ii) improving the transparency of costs charged for advice, and (iii) changing remuneration structures between providers, platforms, and advisers.31 Specifically relevant for this chapter are (ii) and (iii).

A. Inducement Rules [8.33] The FCA decided that it was necessary to introduce more stringent rules with respect to inducements in order to realize a change in the remuneration structures between providers, platforms, and advisers. By introducing more stringent rules, the UK went beyond what was required by MiFID I. Similar to the Netherlands, as stated in paragraph 8.26, gold plating was deemed to be justified on the basis of Article 4 of the MiFID Implementing Directive. The UK implemented the following rules with respect to inducements in the COBS: • ban on payments from providers to advisers for advice on retail investment products;32 • ban on payments by providers to platforms and cash rebates by product providers to consumers using platforms;33 • requirements for advisers to establish a fee structure for advice to consumers, as well as disclosing requirements in respect of the fees

charged.34 [8.34] The rules under (i) and (iii) became effective as of 31 December 2012, one day before the inducement ban entered into force in the Netherlands. However, the FCA stated that it would refrain from enforcement measures during the first year post implementation. The rule under (ii) with regard to platforms became effective as of 6 April 2014. [8.35] The inducement rules only apply to the retail market and do not involve investment services that are provided to professional investors. The rules are applicable to advisers, platforms, and product providers. [8.36] The inducement rules in principle also apply to tied agents. The COBS stipulates that the fact that a fee is paid by a tied agent does not mean that the inducement rules do not apply.35

V. MiFID II 1. Introduction [8.37] Under MiFID II,36 the rules regarding inducements have been further restricted. Under MiFID II, it is important to determine which investment service is provided when analysing the inducement requirements. A distinction is made between (i) investment advice on an independent basis and portfolio management, and (ii) all other investment services and ancillary services. [8.38] Pursuant to Article 24(13) of MiFID II, the European Commission may adopt delegated acts, amongst others, setting out the criteria to assess compliance of firms receiving inducements with the obligation to act honestly, fairly, and professionally, in accordance with the best interest of the client. [8.39] The European Commission mandated ESMA to provide it with technical advice on possible delegated acts. On 22 May 2014, ESMA

published a consultation paper (the ‘Consultation Paper’) asking for input with respect to its technical advice.37 ESMA delivered its technical advice (the ‘Technical Advice’) on 19 December 2014.38 On 7 April 2016, the European Commission published a Draft Delegated Directive (the ‘Draft Delegated Directive’) based on the Technical Advice.39 At the time of writing this chapter, the Directive was still in draft. Below we will describe the provisions of the draft Directive. In addition, we will (where relevant) pay attention to the Technical Advice.

2. Investment Advice on an Independent Basis and Portfolio Management [8.40] Pursuant to Article 24 of MiFID II, an investment firm that provides investment advice on an independent basis, or that provides portfolio management, may not accept and retain fees, commissions, or any monetary or non-monetary benefits (hereinafter ‘inducements’) paid or provided by any third party in relation to the provision of the service to clients. This is often referred to as the ‘inducement ban’. Recital 74 of MiFID II states that this ban applies particularly to inducements from issuers or product providers. [8.41] The wording ‘accept and retain’ means that inducements may be paid by third parties to an investment firm, but that they must be paid on in full to the client as soon as possible after receipt of those payments by the firm, and that the firm is not allowed to offset any third-party payments from the fees due by the client to the firm.40 Recital 24 of the Draft Delegated Directive provides that no specific timeframe is imposed—other than ‘as soon as possible’—since third-party payments may be received by the investment firm at various points in time and for several clients at once. Firms providing independent advice or portfolio management must draw up a policy, as part of their organizational requirements, to ensure that thirdparty payments received are allocated and transferred to each individual client.41 The client must be accurately and, where relevant, periodically, informed about all inducements that the investment firm has received in

connection with the investment service provided to the client and transferred to the client.42 [8.42] The inducement ban does not apply to inducements paid or provided by a client or a person acting on behalf of a client. Inducements paid or provided by a person on behalf of the client should be allowed only as far as the client is aware that such payments have been made on that client’s behalf. In addition, the amount and frequency of any payment must be agreed between the client and the investment firm and not determined by a third party.43 According to Recital 75 of MiFID II, there are several examples which would satisfy that requirement, including where a client pays a firm’s invoice directly. Examples in respect of permitted payments by third parties on behalf of the client are those cases where the investment firm is paid by an independent third party who has no connection with the investment firm regarding the investment service provided to the client and is acting only on the instructions of the client and cases where the client negotiates a fee for a service provided by an investment firm and pays that fee. According to this same Recital, this would generally be the case for accountants or lawyers acting under a clear payment instruction from the client or where a person is acting as a mere conduit for the payment.

A. Minor Non-Monetary Benefits [8.43] The above inducement ban does not apply to minor non-monetary benefits. These must meet three criteria: they should be capable of enhancing the quality of service provided to a client; they must be of a scale and nature such that they could not be judged to impair compliance with the investment firm’s duty to act in the best interest of the client; and they must be clearly disclosed. According to Article 11 of the Draft Delegated Directive, disclosure of minor non-monetary benefits must be made prior to the provision of the relevant investment or ancillary services to clients. Such disclosures may be made in a generic way. [8.44] Article 12(3) of the Draft Delegated Directive states that benefits shall only qualify as acceptable minor non-monetary benefits if they are:

(a) information or documentation relating to a financial instrument or an investment service, that is generic in nature or personalized to reflect the circumstances of an individual client; (b) written material from a third party that is commissioned and paid for by a corporate issuer or potential issuer to promote a new issuance by the company, or where the third-party firm is contractually engaged and paid by the issuer to produce such material on an ongoing basis, provided that the relationship is clearly disclosed in the material and that the material is made available at the same time to any investment firms wishing to receive it or to the general public. Contrary to the other items in this list, this particular non-monetary benefit was not specifically included in the list as set out in ESMA’s Technical Advice; (c) participation in conferences, seminars, and other training events on the benefits and features of a specific financial instrument or an investment service; (d) hospitality of a reasonable de minimis value, such as food and drink during a business meeting or a conference, seminar, or other training event mentioned under point (c); and (e) other minor non-monetary benefits which a Member States deems capable of enhancing the quality of service provided to a client and, having regard to the total level of benefits provided by one entity or group of entities, which are of a scale and nature unlikely to impair compliance with an investment firm’s duty to act in the best interest of the client. [8.45] As per ESMA’s Technical Advice, this list appears to be intended to be exhaustive, despite the fact that in the consultation most respondents disagreed with ESMA’s proposal. One of the reasons considered by many was that an exhaustive list would be too rigid. [8.46] In addition, the Draft Delegated Directive provides that acceptable minor non-monetary benefits must be reasonable and proportionate and of such a scale that they are unlikely to influence the investment firm’s behaviour in any way that is detrimental to the interests of the relevant client.

3. Other Investment Services and Ancillary Services [8.47] The fact that MiFID II includes a specific ban on inducements for advice on an independent basis and portfolio management only does not mean that there are no inducement rules applicable to investment firms that provide other investment services or ancillary services. Article 24(9) of MiFID II provides that investment firms may not pay or be paid any inducement in connection with the provision of an investment service or an ancillary service, to or by any party other than the client, except where the inducement: (a) is designed to enhance the quality of the relevant service to the client; and (b) does not impair compliance with the investment firm’s duty to act honestly, fairly, and professionally, in accordance with the best interest of its clients. [8.48] These criteria are the same as under MiFID I. However, in the Draft Delegated Directive, the European Commission provides more detailed rules on when inducements are considered to be designed to enhance the quality of the relevant service. [8.49] Article 11(2) of the Draft Delegated Directive provides that an inducement shall be considered to be designed to enhance the quality of the relevant service to the client (i.e. the criterion set out under (a) above) if all of the following conditions are met: (a) the inducement is justified by the provision of an additional or higherlevel service to the relevant client, proportional to the level of inducements received, such as: (i) the provision of non-independent investment advice on, and access to, a wide range of suitable financial instruments including an appropriate number of instruments from third-party product providers having no close links with the investment firm; (ii) the provision of non-independent investment advice combined with either: an offer to the client, at least on an annual basis, to assess the continuing suitability of the financial instruments in which the client has invested; or with another ongoing service that

is likely to be of value to the client such as advice about the suggested optimal asset allocation of the client; or (iii) the provision of access, at a competitive price, to a wide range of financial instruments that are likely to meet the needs of the client, including an appropriate number of instruments from third-party product providers having no close links with the investment firm, together with either the provision of added-value tools, such as objective information tools helping the relevant client to take investment decisions or enabling the relevant client to monitor, model, and adjust the range of financial instruments in which they have invested, or providing periodic reports of the performance and costs and charges associated with the financial instruments; (b) the inducement does not directly benefit the recipient firm, its shareholders, or employees without tangible benefit to the relevant client; and (c) the inducement is justified by the provision of an ongoing benefit to the relevant client in relation to an ongoing inducement. [8.50] Contrary to ESMA’s Technical Advice, the above list appears to be intended to be exhaustive. [8.51] The above requirements must be fulfilled on an ongoing basis by the investment firm as long as the firm pays or receives the inducement. However, this does not imply that investment firms are required to ensure a continuously increasing quality of services over time. Recital 23 of the Draft Delegated Directive only requires that investment firms should, once they have fulfilled the quality-enhancement criterion, maintain the enhanced level of quality. [8.52] The Draft Delegated Directive goes on to say that an inducement shall not be considered acceptable if the provision of relevant services to the client is biased or distorted as a result of inducement. [8.53] The Draft Delegated Directive also provides that investment firms must hold evidence that any inducements paid or received by the firm are designed to enhance the quality of the relevant service to the client:

(a) by keeping an internal list of all inducements received by the investment firm from a third party in relation to the provision of investment or ancillary services; and (b) by recording how the inducements paid or received by the investment firm (or the inducements that the investment firm intends to use) enhance the quality of the services provided to the relevant clients, and the steps taken in order not to impair the firm’s duty to act honestly, fairly, and professionally, in accordance with the best interests of the client. [8.54] As under MiFID I, the existence, nature, and amount of the inducement, or, where the amount cannot be ascertained, the method of calculating that amount, must be clearly disclosed to the client, in a manner that is comprehensive, accurate, and understandable, prior to the provision of the relevant investment or ancillary service. New is that, where applicable, the investment firm must also inform the client on mechanisms for transferring the inducement received in relation to the provision of the investment or ancillary service to the client. [8.55] According to Article 11(5) of the Draft Delegated Directive, the following information must be disclosed to the client: (a) prior to the provision of the relevant investment or ancillary service, the investment firm must disclose to the client information on the inducement concerned in a manner that is comprehensive, accurate, and understandable. Minor non-monetary benefits may be described in a generic way. Other non-monetary benefits received or paid by the investment firm in connection with the investment service provided to a client must be priced and disclosed separately; (b) where an investment firm is unable to ascertain on an ex ante basis the amount of any inducement to be received or paid, and instead discloses to the client the method of calculating that amount, the firm shall also provide its clients with information of the exact amount of the payment or benefit received or paid on an ex post basis; and (c) at least once a year, as long as (ongoing) inducements are received by the investment firm in relation to the investment services provided to the relevant clients, the investment firm shall inform its clients on an

individual basis about the actual amount of payments or benefits received or paid. Minor non-monetary benefits may be described in a generic way. [8.56] When more firms are involved in a distribution channel, each investment firm providing an investment or ancillary service shall comply with its disclosure obligations to its clients. [8.57] The above requirements do not apply to ‘necessary fees’, that is, inducements which enable or are necessary for the provision of investment services, such as custody costs, settlement and exchange fees, regulatory levies, or legal fees, and which by their nature cannot give rise to conflicts with the investment firm’s duties to act honestly, fairly, and professionally, in accordance with the best interests of its clients.

4. Deviating Requirements [8.58] According to Article 24(12) of MiFID II, Member States may, in exceptional cases, impose additional requirements on investment firms (we also refer to paragraph 8.28). These requirements must be objectively justified and proportionate so as to address specific risks to investor protection or to market integrity which are of particular importance in the circumstances of the market structure of that Member State. If a Member State wishes to impose additional requirements, it must notify the European Commission. This notification must also include a justification for the additional requirement. The European Commission shall then provide its opinion on the proportionality of and justification for the additional requirements. In addition, the Commission shall make public on its website the additional requirements imposed by Member States. [8.59] MiFID II also provides that Member States may retain additional requirements that were notified to the Commission in accordance with Article 4 of the MiFID Implementing Directive before 2 July 2014, provided that the conditions laid down in that Article were met.

VI. Research as an Inducement 1. Introduction [8.60] As discussed in paragraphs 8.40–8.41, the inducement rules in MiFID II prohibit investment firms from accepting and retaining monetary and non-monetary benefits paid or provided by third parties in relation to the provision of independent advice or portfolio management. Only a limited category of inducements will be permitted, that is, minor nonmonetary benefits that enhance the quality of the service to the client. In its Consultation Paper and Technical Advice for the delegated acts under MiFID II, ESMA gives specific attention to research as a possible form of such non-monetary inducement. In this section we will discuss how the rules regarding the use of research in relation to the inducement ban will most likely turn out under MiFID II. As this will be laid down in Level 2 legislation, of which only a draft is available at the time of writing of this chapter, the final rules regarding research as described below may turn out differently.

2. Definition of ‘Research’ [8.61] In the Consultation Paper and the Technical Advice, ESMA uses the terms ‘research’, ‘financial research’, and ‘investment research’ without defining such terms and without much consistency. The scope of what type of research is meant here is therefore not entirely clear. In any event, it can be derived from the Consultation Paper that ESMA considers investment research as defined in Annex 1 Part B of MiFID II to fall within the category of research within the scope of these rules.44 As the term ‘research’ is used most predominantly in the Consultation Paper and the Technical Advice, we will use that term in the rest of our chapter. [8.62] A further explanation of what is to be considered research in this respect can be found in the Draft Delegated Directive. In Recital 28, research is described as follows:

Research in this context should be understood as covering research material or services concerning one or several financial instruments or other assets, or the issuers or potential issuers of financial instruments, or be closely related to a specific industry or market such that it informs views on financial instruments, assets or issuers within that sector. That type of material or services explicitly or implicitly recommends or suggests an investment strategy and provides a substantiated opinion as to the present or future value or price of such instruments or assets, or otherwise contains analysis and original insights and reach conclusions based on new or existing information that could be used to inform an investment strategy and be relevant and capable of adding value to the investment firm’s decisions on behalf of clients being charged for that research.

3. Research as an Inducement [8.63] Research as indicated above and investment research (the provision of which qualifies as an ancillary service) is provided to and used by investment firms such as portfolio managers in determining their investment advice or portfolio management. Because research has value to investment firms, it can be seen as an inducement if provided free of charge or at a discount. In the course of ESMA’s consultation and Technical Advice, market parties feared that rules would be imposed that would hinder investment firms’ access to research and would pose concerns for the viability of firms providing research. Concerns were also expressed by France, Germany, and the United Kingdom in a letter to the European Commission and ESMA, in which letter these Member States indicated that they considered the limitations on research to go beyond what was agreed in MiFID II.45 [8.64] The European Commission decided to incorporate the rules regarding research as an ‘exception’. The European Commission included Article 13 in the Draft Delegated Directive, which includes the circumstances in which research is not to be considered an inducement. This exception leaves two permitted ways of receiving research. The first option is research in return for direct payments by the investment firm out of its own resources. The second exception is an elaborate and complicated exception. It leaves room for the investment firm to use client money to fund research, if a number of strict rules are followed, including using a

separate research payment account, a specific research charge to the client, regular assessments by the investment firm of the quality of the research, making research budgets, provision of overviews to regulators, drawing up a policy, and several requirements safeguarding transparency to clients. This rather complicated and detailed requirement seems to be the opposite of what the three Member States requested in their letter regarding research as an inducement. The three Member States indicated that ‘the mechanism for the payment of research shouldn’t be overly prescriptive’. It therefore remains to be seen whether these Member States will be happy with this part of the draft delegated acts and whether this will be cause for further negotiations between the parties involved.

A. Research as a Minor Non-Monetary Benefit [8.65] Besides the exceptions under which research does not qualify as an inducement, research can also qualify as a minor non-monetary benefit. As discussed in paragraphs 8.43–8.46, certain minor non-monetary benefits are not caught by the inducement requirements and are thus permitted under MiFID II. In its Consultation Paper ESMA considered that certain ‘tailored or bespoke’ research would likely not fall within the category of nonmonetary inducements, including bespoke reports or analytical models, market data services, or investor field trips.46 In the same Consultation Paper, ESMA indicates that an example of research that may fall within the category of minor non-monetary benefits is financial analysis accessible by a large number of persons or the public at the same time.47 [8.66] In the Draft Delegated Directive, the European Commission gives a description of areas of research that are likely to qualify as minor nonmonetary benefit. Such research includes short-term market commentary, company results, and information on upcoming events.48 The Draft Delegated Directive makes clear that substantive analysis, or substantiated opinions or any other research to which valuable resources are allocated, should not be considered as minor and therefore does not fall under the exemption of such minor benefits that fall outside the scope of the inducement rules.

VII. Conclusion [8.67] During the negotiations on MiFID II, it became clear that inducements were a topic on which there was no easy agreement between the Member States. Where certain Member States pressed for a total ban on inducements, other Member States were unwilling to impose such strict rules. As is not uncommon in these situations, the political compromise that was reached was to allow for deviating rules by those Member States that wished to see stricter rules. Allowing for this may be regarded as counter to the general trend set by the European legislator to create less room for Member State options by creating ‘single rulebooks’ and an increase in the use of regulations as a legislative tool. It is therefore disappointing to see that on such an important topic in the MiFID rules the European market will continue to have deviating rules per Member State. The Dutch legislator has already indicated that it will make use of the (continued) room for ‘gold plating’, and it can be expected that other Member States having applied stricter rules will do the same, thus not creating a level playing field for investment firms regarding the use of inducements.

1

Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments amending Council Directives 85/611/EEC and 93/6/EEC, OJ L 145/1; and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC, OJ L 177/1. 2 Article 19(1) MiFID. 3 In addition, according to a report from 2011 commissioned by the European Commission (Synovate, Consumer Market Study on Advice within the Area of Retail Investment Services—Final Report, 2011), it seemed likely that a number of firms in that study failed to comply with the MiFID rules regarding inducement. 4 Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU. 5 See in this respect D. Busch and D. A. DeMott, Liability of Asset Managers (Oxford: Oxford University Press, 2012), p. 45. 6 Commission Directive 2006/73/EC of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive, OJ L 241/26.

7

See also Busch and DeMott (n. 5), 42; R. Panasar and Philip Boeckman, European Securities Law (Oxford: Oxford University Press, 2010). 8 See in this respect also M. Kenny, J. Devenny, and L. Fox O’Mahony, Unconscionability in European Private Financial Transactions (Cambridge: University Press, 2010), p. 331 and onwards. 9 Article 26 MiFID Implementing Directive. See also J. P. Casey and K. Lannoo, The MiFID revolution (Cambridge: University Press, 2009), p. 131. 10 CESR Recommendations Inducements under MiFID, CESR 07-228b, May 2007. 11 ESMA Recommendations (n. 10), p. 2. 12 Inducements: Good and poor practices, CESR/10-29619, April 2010. 13 See for instance comments by the Luxembourg Bankers’ Association on CESR Consultation paper on inducements, ABBL, ‘CESR consultation paper on Inducements under MiFID’, available at . 14 Article 21(e) of the MiFID Implementing Directive. 15 ESMA Recommendations (n.10), p. 6. 16 ibid. 17 ESMA Practices (n. 12), p. 16. 18 ESMA Recommendations (n. 10), p. 4. 19 ibid, p. 6. 20 ibid, pp. 8–9. 21 See also D. Busch and D. A. DeMott, Liability of Asset Managers (Oxford: Oxford University Press, 2012), p. 45. 22 ibid, p. 7. 23 ibid, p.7. 24 ESMA Practices (n. 12), p. 16. 25 ibid, p. 17. 26 Besluit gedragstoezicht financiële ondernemingen Wft, Article 168a. 27 Stb. 2013, 537, p. 51 (Bulletin of Acts and Decrees). 28 Stb. 2013, 537, p. 30 (Bulletin of Acts and Decrees). 29 Article 31(1) MiFID I. 30 Stb. 2013, 537, p. 28 (Bulletin of Acts and Decrees). 31 See FSA, ‘A Review of Retail Distribution’, Discussion Paper 07/1 (1 June 2007), p. 17, available at . 32 COBS 6.1.A.4. 33 COBS 6.1.B.5. 34 The rules under (i) and (iii) do not apply to fees charged or payments made before 30 December 2012. 35 COBS 2.3.7.

36

Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU (recast). 37 Consultation Paper, MiFID II/MiFIR, 22 May 2014, ESMA/2014/549. 38 Final Report, ESMA’s Technical Advice to the Commission on MiFID II and MiFIR, 19 December 2014 ESMA/2014/1569. 39 Commission Delegated Directive (EU) …/… of 7.4.2016 supplementing Directive 2014/65/EU of the European Parliament and of the Council with regard to safeguarding of financial instruments and funds belonging to clients, product governance obligations, and the rules applicable to the provision or reception of fees, commissions, or any monetary or non-monetary benefits. 40 Recital 74 of MiFID II and Article 12(1) of the Draft Delegated Directive. 41 ibid. 42 Recital 74 of MiFID II and Article 12 of the Draft Delegated Directive. 43 Recital 75 of MiFID II. 44 Consultation Paper (n. 37), p. 121. 45 Letter of 25 August 2015. 46 Consultation Paper (n. 37), p. 121. 47 ibid. 48 Recital 29 Draft Delegated Directive.

9 AGENCY AND PRINCIPAL DEALING UNDER MIFID I AND MIFID II Danny Busch

I. Introduction II. Four Transaction Types 1. General 2. Execution of Client Orders through the Regulated Market or an MTF 3. Dealing Solely on Own Account 4. Execution of Client Orders through Internalisation or Systematic Internalization 5. Execution of Client Orders by Means of Agency Crosses III. Investor Protection IV. Acting as Agent or as Principal: Not a Justified Distinction to Serve as a Basis for Determining the Degree of Investor Protection 1. General 2. Investors’ Reasonable Expectations 3. Tenuous, Arbitrary, and Easy to Manipulate 4. Grant Estates Ltd v Royal Bank of Scotland 5. No Practicable Distinction under MiFID II either 6. Dutch Supreme Court has Already Extended Civil Duty of Care to Dealing on own Account 7. The UK Government Response to the Kay Review

V. Conclusion

I. Introduction [9.01] Investment firms regularly act as agent or intermediary in the execution of transactions in financial instruments. Such transactions can take place in the context of an execution-only service, investment advice, or portfolio-management service. Investment firms also carry out transactions as principal (i.e. on own account). In such a case, the investment firm acts solely as the investor’s contractual counterparty (principal dealing). In some transactions an investment firm acts both as intermediary and on own account. Although transactions in financial instruments carried out by an investment firm are traditionally governed by the applicable private law, transactions in financial instruments are now also very tightly regulated by financial supervision law, in particular the European Markets in Financial Instruments Directive (MiFID I) and the related MiFID I Implementing Directive and MiFID I Implementing Regulation.1 On 3 January 2018 the MiFID I regime will be replaced by the MiFID II regime, which includes a directive (MiFID II) and the Markets in Financial Instruments Regulation (MiFIR).2 Below, MiFID I and MiFID II will be jointly referred to as ‘MiFID’. [9.02] One of MiFID’s key objectives is to offer investors a high level of protection.3 In the case of transactions in financial instruments the level of protection afforded to an investor under MiFID largely depends on the distinction between acting on behalf of the investor and dealing on own account. If an investment firm deals wholly or partly on behalf of the investor (as intermediary or representative), it has all kinds of duties of care to the investor. If, on the other hand, an investment firm enters into a transaction with an investor solely as a contractual counterparty, it has few if any duties of care. Once it has been established that the firm is acting on behalf of the client, the level of protection depends next on the classification of the client and the exact framework in which the transactions are carried out (i.e. whether the transactions involve executiononly, investment advice, or portfolio-management services).

[9.03] The central question addressed in this chapter is whether allowing the level of protection afforded to an investor under MiFID to be largely dependent on the distinction between dealing on own account on one hand and acting on behalf of the client on the other is justified.

II. Four Transaction Types 1. General [9.04] Investment firms can carry out transactions in financial instruments in various ways.4 It is important to distinguish between these different transaction types, as the level of protection afforded to an investor under MiFID may differ from one type to another.

2. Execution of Client Orders through the Regulated Market or an MTF [9.05] The first transaction type involves cases where an investment firm executes an order on behalf of a client and uses an external party for this purpose. This can be done in two ways. For example, suppose that the client instructs the investment firm to buy 100 Shell shares for it (for a price not exceeding x per share). The investment firm can execute the order by transmitting it to the regulated market5 on which the financial instrument is admitted to trading. A regulated market is known in common parlance as a stock exchange. Examples are NYSE Euronext Amsterdam, and the London Stock Exchange (LSE). Second, the investment firm can execute the order by transmitting it for further settlement to one of the now many competitors of the established stock exchanges—the so-called alternative trading platforms, otherwise known as multilateral trading facilities (MTFs), for example Turquoise.6 A regulated market or MTF is referred to in the terminology of MiFID as the ‘execution venue’.7

[9.06] Whether the investment firm executes the client order by transmitting it to a regulated market or to an MTF, it is treated in both cases as executing an order on behalf of the client and in both cases an external party is involved (i.e. a regulated market or an MTF).8 Through the regulated market or the MTF, which brings together multiple buying and selling interests in financial instruments, a party is found who is willing to sell 100 Shell shares.9 In MiFID terminology, the investment firm is then providing the investment service known as ‘execution of orders on behalf of clients’. This service has been defined in MiFID as ‘acting to conclude agreements to buy or sell one or more financial instruments on behalf of clients’.10 As the investment firm is acting on behalf of the client, it is providing an investment service and accordingly owes a high duty of care to the client.11 [9.07] For private law purposes, the investment firm will be treated as acting as intermediary in the transaction type described above. It may act as the client’s direct representative (direct vertegenwoordiger in the Netherlands, direkter Stellvertreter in Germany, and disclosed agent in England and Wales).12 In the case of civil law jurisdictions such as Germany and the Netherlands, there may also be forms of indirect representation (middellijke vertegenwoordiging in the Netherlands and mittelbare Stellvertretung in Germany), in particular in the case of stockbrokers (commissionair in effecten in the Netherlands and Kommissionär in Germany).13 Although the position of stockbroker is still separately regulated by law in Germany (Handelsgesetzbuch, §§ 383–406), this is no longer the case in the Netherlands.14 In England and Wales, where the legal concept of indirect representation is unknown, commission agency is a somewhat similar concept. However, this is a rather controversial concept whose exact legal consequences are not entirely clear.15 It would also be possible to think in terms of undisclosed agency in England and Wales, at least in theory.16

3. Dealing Solely on Own Account

[9.08] A second transaction type concerns cases in which the investment firm only deals on its own account and does not also act on behalf of the investor (principal dealing). In such cases, the investment firm does not in any way act on behalf the investor (either as representative or agent), but instead acts merely as the contractual counterparty of a party buying or selling financial instruments. ‘Dealing on own account’ is defined in MiFID as ‘trading against proprietary capital resulting in the conclusion of transactions in one or more financial instruments’.17 As this solely concerns dealing on own account, the investment firm has few if any duties of care to the investor.18 Naturally, where a firm solely deals on own account, there is no representation or agency under private law.

4. Execution of Client Orders through Internalisation or Systematic Internalization [9.09] A third transaction type concerns cases in which an investment firm acts in different capacities on the two sides of a single transaction, on one hand on behalf of a client when executing an order and on the other on own account. An example may help to clarify this. A client once again places an order with an investment firm for the purchase of 100 Shell shares (at a price not exceeding x per share). Unlike the situation in transaction type 1, the investment firm does not execute the order by transmitting it to the regulated market or an MTF. In this third transaction type, the investment firm executes the client order by selling its own Shell shares (i.e. Shell shares which it has ‘on its own books’). The investment firm acts in its own name (dealing on own account) on the selling side of the transaction and on behalf of the client (execution of orders on behalf of the client) on the buying side. In the supervision legislation, this third transaction type is known as internalizing the order (or systematic internalizing if it is carried out on a systematic basis).19 This is therefore a way of executing an order without the involvement of an external party. Internalization (or systematic internalization if it is carried out on a systematic basis) is not a separate investment service or activity, but a combination of the investment service known as ‘execution of orders on behalf of the client’ (see Section II.2 above) and dealing on own account

(see Section II.3 above). As the investment firm is therefore acting partly on behalf of the client, this constitutes in part an investment service and thus entails a high duty of care to the client.20 [9.10] Although systematic internalization is not a separate investment service or activity but a combination of dealing on an account and acting on behalf of the client, the fact that the internalization is described as ‘systematic’ does mean that an investment firm must fulfil certain transparency obligations prior to the dealing.21 This is why MiFID provides a detailed definition of the term ‘systematic internalizer’.22 It should also be noted that after the dealing investments firms are also subject to transparency obligations, but these apply regardless of whether or not the internalization is ‘systematic’.23 [9.11] In private law this third transaction type is sometimes said to involve self-dealing (England and Wales). In Germany this is one of the two subcategories of the Insichgeschäft (§ 181 BGB).24 This concerns the situation in which an agent or direkter Stellvertreter acts as the investor’s contractual counterparty. [9.12] As far as self-dealing is concerned, a distinction must first be made between a purchase from the principal and a sale to the principal. In the former case the agent cannot purchase [… property] for himself without full and fair disclosure of all the facts to the principal. The onus in such a case is on the agent to show that the price was adequate, that the sale was as advantageous to his principal as any other sale he could have obtained from a third party, and that he disclosed all the relevant facts to his principal before the purchase and that the principal gave his informed consent.25

In the latter case, an agent may not sell his own property to his principal without full and fair disclosure and the obtaining of his principal’s informed consent […]. He must also prove that the transaction was fair. But however fair the transaction may be, it can still be set aside by the principal if there has not been full disclosure.26

Finally, there is the special case of self-dealing where the agent has received specific instructions from the principal:

[w]here, however, the agent has specific instructions to buy or to sell at a particular price and no element of advice or discretion is involved, there may be cases where he is free to sell his own property to the principal or buy the principal’s property himself, there being no possibility of conflicting interest.27

[9.13] Owing to the risk of conflicts of interest, Insichgeschäft is permitted only with the client’s prior consent.28 Likewise, the German concept of Selbsteintritt in relation to Kommissionsgeschäft is a form of internalization, systematic or otherwise. Here an indirect representative acts as the client’s counterparty. Strictly speaking, this is impossible. A Kommissionär acts in his own name (but on behalf of the client) and has no power to bind his client directly to a contract with a third party. If the Kommissionär acts as contractual counterparty in the transaction, the contract would have only one discernible contracting party, whereas two parties are required for a valid contract. For this reason, the consequence of Selbsteintritt is nonetheless that the ultimate client (Kommittent) becomes the contractual counterparty of the Kommissionär, given the express statutory provision to this effect (§ 400 HGB). Unlike the situation with Insichgeschäft, the client’s prior consent is not required. However, Selbsteintritt is permitted only if there is a stock exchange or market price available for the financial instrument in question and the price charged for the transaction is in keeping with it. [9.14] In the Netherlands this transaction type is regulated in Article 7:416 of the Dutch Civil Code (DCC). This provision relates not only to cases in which a direct representative acts as the client’s contractual counterparty but also to cases where the contractual counterparty is an indirect representative. To ensure that in the latter case the contract is concluded between the indirect representative in his capacity as counterparty on one hand and the client on the other, it is expressly provided that a mandatory (lasthebber) who may act only in his own name may nonetheless act as the counterparty of the mandatory (lastgever) (Article 7:416(2) DCC). In both the former and the latter cases it is necessary for the content of the juristic act to be determined so precisely that the possibility of a conflict of interest is excluded (Article 7:416(1) and (2)). However, the parties may derogate from this provision and use this method even where the possibility of a conflict of interest between them is

not excluded. If the client is a consumer, such consent must always be given in writing as the juristic act is otherwise voidable (Article 7:416(3) DCC).29

5. Execution of Client Orders by Means of Agency Crosses [9.15] A fourth type of transaction involves the matching of opposite client orders (i.e. ‘sell’ orders and ‘buy’ orders) by an investment firm. Such transactions are also known as in-house matching or agency crosses. In these cases, the investment firm acts on one side of the transaction on behalf of one client (the seller) and on the other side of the transaction on behalf of the other client (the buyer). Such a situation may occur, for example, where an investment firm is instructed by one client to buy 100 Shell shares and by another client to sell 100 Shell shares.30 The firm can then buy the 100 Shell shares in the name (or, in any event, on behalf) of one client and sell the 100 Shell shares in the name (or, in any event, on behalf) of the other client. Unlike transaction type 1, no external party (a regulated market or an MTF) is called in for the execution of the order, and unlike transaction type 3 the investment firm does not execute the order by itself acting as the counterparty. In type 4, the investment firm executes an order on both sides of the transaction on behalf of a client within the meaning of MiFID I. It therefore acts on both sides of the transaction on behalf of a client. Consequently, the investment firm must observe a high duty of care in relation to both clients.31 [9.16] Under MiFID II the supervision law definition of matching opposite client orders seems rather more complicated. In certain cases, the matching of opposite client orders may amount to the operation of an organized trading facility (OTF). OTF is defined in MiFID II as follows: ‘a multilateral system which is not a regulated market or an MTF and in which multiple third-party buying and selling interests in bonds, structured finance products, emission allowances or derivatives are able to interact in the system in a way that results in a contract in accordance with Title II of this Directive’. This definition includes cases where an investment firms matches opposite client orders with one another (agency crossing

systems).32 As is apparent from the definition of OTF, trading through an OTF can relate only to non-equity instruments (such as derivatives and bonds) and not to equities (shares). Nonetheless, it can be assumed that insofar as client orders are executed on an OTF by matching opposite orders with one another, such transactions in each case involve acting on behalf of the client and thus constitute an investment service, and the investment firm must therefore observe a high duty of care to both clients.33 [9.17] In private law, the fourth transaction type described here is said to involve multiple agency (England and Wales). In Germany this is one of the two subcategories of Insichgeschäft (§ 181 BGB).34 This involves a situation in which a disclosed agent or direkter Stellvertreter acts on behalf of both the buying and the selling client. In England and Wales, one manifestation of the agent’s fiduciary duty to avoid conflicts is that he must not serve two principals whose interests may conflict. So he may not act for both parties to a transaction unless he ensures that he fully discloses all the material facts to each party and obtains their informed consent to his so acting. In this the agent does not prefer his own interest, but equally may not act entirely in the interests of either single principal. There may also be a breach of the duty of loyalty in the sense that the loyalty must be undivided. In such cases he may cause loss to one by failure to disclose information acquired in connection with the other—information, indeed, the disclosure of which would be a breach of the duty of the first.35

[9.18] The above provides a summary of the general rules. However, there may be an exception to these general rules: [p]erhaps where it is notorious that agents do act for both sides to a transaction, and where the agent explains all the circumstances fully to the principal and the principal consents to the agent receiving two commissions, he cannot subsequently call the agent to account for the commission paid by the other party, nor can he himself refuse to pay his own commission. Even if the agent is improperly acting for two opposing principals, the court will not normally make an order on the application of one of the principals which will result in the agent breaking his confidence towards his other principal, at least where the other principal has acted in good faith.36

In view of the risk of a conflict of interest, this variant of Insichgeschäft too is permitted only if the client consents in advance.37 In the Netherlands a provision of this kind is contained in Article 7:417 DCC (serving two masters). A mandatory may also act as mandatory of the counterparty only

if the content of the juristic act is determined so precisely that the possibility of a conflict of interest between the two parties is excluded (Article 7:417(1) DCC). However, the parties may derogate from this provision and use this method even where the possibility of a conflict of interest between them is not excluded. If the client is a consumer, such consent must always be given in writing as the juristic act is otherwise voidable (Article 7:417(2) DCC).38

III. Investor Protection [9.19] It has been explained above that an investment firm can execute an order for a client in a financial instrument through (i) a regulated market, (ii) an MTF, (iii) internalization, or (iv) agency crosses. Under MiFID, the firm is said in all these cases to be acting on behalf of the client and consequently to be providing, among other things, an investment service. Just as in the case of other investment services, the basic principle here is that the investment firm owes a high duty of care to the investor. Where an investment firm acts for both parties to a transaction (internalization and agency crosses), it has additional obligations to prevent conflicts of interest. [9.20] If, on the other hand, an investment firm deals solely on own account (i.e. acts solely as an investor’s contractual counterparty), the investor has little if any protection since most conduct-of-business rules are not applicable. It can at most be argued in such cases that the general obligation of fair dealing is applicable. Article 25(1) MiFID I provides, after all, in a general sense that investment firms must act honestly, fairly, and professionally. Under MiFID II this provision is included in Article 24(1) MiFIR. It can therefore be argued that this also applies where an investment firm is dealing solely on its own account in relation to an investor. Most legal practitioners will probably disagree with this analysis. They generally tend to assume that where an investment firm deals on own account it owes no MiFID duties of care whatsoever. As some situations where a firm deals on own account are, in fact, entirely exempted from MiFID, the general obligation of fair dealing does not apply at all. For an investment firm to be entirely exempted from MiFID I, it must deal solely

on own account (and not provide any investment services or activities) unless (i) it is a market-maker or (ii) it deals on own account outside a regulated market or an MTF on an organized, frequent, and systematic basis by providing a system accessible to third parties in order to engage in dealings with them.39 This exception has been tightened up still further in MiFID II in the sense that it has become more difficult to invoke it successfully.40 [9.21] It should be emphasized that the client classification obligation applies only if the investment firm provides investment services to an investor. Where an investment firm acts solely as a contractual counterparty of an investor, its actions constitute dealing on own account. As this is an investment activity and not an investment service, the client classification obligation does not apply. This is also apparent from the definition of ‘client’ in MiFID: ‘any natural or legal person to whom an investment firm provides investment or ancillary services’.41 [9.22] In brief, if an investment firm acts solely as an investor’s contractual counterparty (dealing on own account), this constitutes an investment activity and the investor is not a ‘client’. Nor, therefore, is there a client classification obligation.

IV. Acting as Agent or as Principal: Not a Justified Distinction to Serve as a Basis for Determining the Degree of Investor Protection 1. General [9.23] As we have seen, as soon as an investment firm does business on behalf of the client (acts as agent), it owes a high duty of care under MiFID. By contrast, where an investment firm acts solely as a contractual counterparty with an investor (principal dealing), it has few if any duties of care. In my view, allowing the degree of investor protection to be dependent on this distinction is unjustified. This proposition is expounded below.

2. Investors’ Reasonable Expectations [9.24] First of all, an investor is reasonably entitled to expect that an investment firm with which he deals will look after his interests adequately and thus fulfil certain duties of care towards him. The firm is, after all, the financial expert. As such, its knowledge is bound to be superior to that of the investor, certainly a retail client. This is no different where an investment firm acts solely as an investor’s contractual counterparty. In such cases the investor is reasonably entitled to expect that the investment firm will discharge the duty of care that also applies in the case of an execution-only service. [9.25] In fact, the European Commission acknowledged in its letter to the Committee of European Securities Regulators (CESR) of 19 March 2007 that the investor’s reasonable expectations play a role of some importance in answering the question of whether in a given case the investment firm transacts as agent or solely as principal. That is understandable, since whether or not the investment firm transacts as agent or solely as principal is a matter of interpretation of the legal relationship. But this approach has its limits. If it is absolutely clear on the facts that the investment firm transacted solely as principal, it is not possible to argue that the investment firm in fact transacted as agent. Preferably, therefore, the distinction between acting as agent and acting as principal should simply no longer be treated as relevant in determining the degree of investor protection.42

3. Tenuous, Arbitrary, and Easy to Manipulate [9.26] Moreover, in practice the distinction between acting on behalf of a client and dealing on own account is often tenuous, arbitrary, and easy to manipulate. For this reason it does not justify different treatment. This can be illustrated by two examples.

A. Example 1

[9.27] An investment firm gives an undertaking to a client to enter into a swap agreement on his behalf with another party. At that moment, this entails execution of an order in a financial instrument on behalf of the client (i.e. an execution-only service). Subsequently, the investment firm may often decide on the basis of the agreement with the client that it will itself act as contractual counterparty in the swap agreement (internalization/selfdealing). In that case, the investment firm combines dealing on own account with acting on behalf of the client and is therefore subject to the high duties of care referred to above as the firm is acting on the client’s behalf. Where, however, it is established from the outset that the investment firm is the client’s contractual counterparty (and the firm does not otherwise provide advice or another investment service), it is treated under MiFID I as dealing solely on own account, which means, in practice, that the firm has virtually no duty of care.

B. Example 2 [9.28] An investment firm can match opposite client orders with one another (agency crosses). In that case the firm can be said to be acting on behalf of the client on both sides of the transaction and therefore owes the high duty of care discussed above to both clients. However, this transaction can also be arranged in such a way that the investment firm takes opposite positions on own account in relation to different counterparties with the same maturity (back-to-back trading or matched principal trading). In such cases the investment firm does not in fact run any risk. The economic result is the same as if opposite client orders were matched (agency crosses). Nonetheless, there are virtually no duties of care under MiFID I in the case of matched principal trading since the investment firm is dealing solely on own account on both sides of the transaction. This means, in short, that whether there is a high duty of care or virtually no duty of care depends on how what is essentially the same economic transaction is arranged in law. [9.29] Naturally, it could be argued that it is up to the investor himself to urge the investment firm to execute the order on his behalf, but most investors will not be aware that the rather arcane legal distinction between dealing on own account and executing an order on behalf of the client can

have far-reaching consequences for the level of protection he enjoys. This is especially true in the case of retail clients. Moreover, the type of transaction is not explicitly discussed with the client. To find this out, the client must read the ‘small print’ of the contract.

4. Grant Estates Ltd v Royal Bank of Scotland [9.30] As noted above, where an investment firm arranges a transaction with an investor in such a way that it deals solely on own account, it can exclude virtually all duties of care under MiFID I. This is why interest rate swap agreements often contain a provision that the investment firm acts only as contractual counterparty and does not provide advice. But what happens if, despite the agreement, an employee of the investment firm nonetheless provides advice? Is the contractual clause limiting the role of the investment firm to that of counterparty still valid in such a case? If it is wished to answer this question in the affirmative, the distinction between dealing on own account on one hand and acting on behalf of the client and providing other investment services (such as advice) on the other would be even more likely to produce unjustified outcomes from the perspective of investor protection. [9.31] This can be illustrated by the Scottish case Grant Estates Ltd v Royal Bank of Scotland plc.43 In this case Lord Hodge44 held that a clause providing that the investment firm acted solely as contractual counterparty was valid, despite the fact that an employee had advised the investor. The facts were as follows. Grant Estates Limited (GEL) was a small property developer. On 6 and 11 July 2007 it took out two loans with the Royal Bank of Scotland (RBS) for a total amount of ₤775,000. The rate of interest on the loans was variable, namely 1.4 per cent over base rate (which was 5.75 per cent (LIBOR) on 6 July). In December 2007 RBS concluded with GEL an interest rate swap agreement (IRSA) to cover the risk of a rise in interest rates on the underlying loans. As a result of the financial crisis, however, the interest rate fell below the level of the fixed rate under the IRSA, and GEL ended up paying more than it would have done without the IRSA. In the course of 2008 GEL got into financial difficulties as a result of the economic downturn and was no longer able to meet its obligations to RBS.

On 25 February 2011 RBS applied for GEL to be placed in administration. GEL then instituted proceedings against RBS. In essence, GEL alleged that RBS had mis-sold the IRSA to GEL. GEL sought, inter alia, damages for (1) infringement of the financial supervision requirements of the Conduct of Business Sourcebook (COBS) issued by what was then the Financial Services Authority (FSA), now known as the Financial Conduct Authority (FCA) (these rules constitute the implementation of MiFID I in the United Kingdom); (2) negligent advice and negligent misrepresentation;45 and (3) breach of contract.46 [9.32] The applicable Terms of Business (ToB) provided, inter alia, that RBS ‘will not provide [GEL] with advice on the merits of a particular transaction’ and that GEL ‘should obtain […] independent financial, legal and tax advice’ (ToB 3.3). The ToB also provided that RBS ‘will provide [GEL] with general dealing services on an execution-only47 basis’ (ToB 3.2) and that [a]s we or our affiliates will act as principal in all transactions covered by these Terms, no duty of best execution arises under FSA Rules as no transactions are executed in circumstances giving rise to duties similar to those arising on an order to execute a transaction as agent. [GEL] may either accept or reject the price we offer [GEL] and we do not represent that such price is necessarily the best price (ToB 10).48

The ToB also provided that ‘[w]here these Terms of Business conflict with Applicable Regulations, the latter shall prevail. Applicable Regulations shall include the FSA Rules’ (ToB 2.2), and that ‘[n]othing in these terms shall oblige us to do anything that we believe to be contrary to Applicable Regulations. Nothing in these terms will exclude or restrict any liability that we owe you under the FSA Rules’ (ToB 19.1). Finally, schedule 1 to the ToB provided that GEL had been classified under the supervision rules as a ‘private customer’. This does not mean, by the way, that GEL had to be treated as a private investor. GEL was an SME company and was, as such, a retail, non-private investor. [9.33] GEL argued that the express provision in the ToB that no advice would be provided and that RBS would act only as contractual counterparty in the IRSA did not change the reality of the situation, namely that RBS staff had provided GEL with advice within the meaning of the FSA’s

supervision rules. This was viewed by GEL as decisive, since the ToB clearly provide that the FSA rules prevail in the event of a conflict with the ToB. GEL therefore considered that it could institute a civil claim for an infringement of the supervision rules relating to the provision of advice on the basis of the then section 150 of the Financial Services and Markets Act 2000 (FSMA), which provided that a contravention of supervision rules issued by the FSA on the basis of Part X, Chapter I FSMA ‘is actionable at the suit of a private person who suffers loss as a result of the contravention, subject to the defences and other incidents applying to actions for breach of statutory duty’.49 GEL argued that the limitation to private persons was in conflict with the EU legal principle of equal treatment, and that professional clients such as GEL could also therefore bring a civil claim for infringement of MiFID I provisions as implemented in the United Kingdom.50 [9.34] In evaluating this submission, Lord Hodge held that MiFID I does not stipulate that clients—whether professional or private—must be able to bring a civil claim against banks for a breach of MiFID I rules implemented in national law. Although MiFID I is admittedly intended to protect both retail and professional clients, the protection is solely under supervision law. MiFID I makes no provision for protection under civil law. Although section 150 FSMA provides that a contravention of supervision rules is actionable at the suit of a private person, this is a choice made by the legislator in the United Kingdom even before the implementation of MiFID I and is not in any way connected with the MiFID I provisions. Lord Hodge also stated that he was also not aware of any EU legal principle of equal treatment between corporate and private persons. As GEL was quite clearly not a private person but a corporate client, he held that GEL had in any event no right of action under section 150 FSMA.51 [9.35] Nor was there any contract with GEL to provide advice. The ToB are clear. Whether or not advice was provided by RBS staff in practice is immaterial in view of the clear provision of the ToB. Nor, according to Lord Hodge, was the situation altered by the fact that advice was possibly given in practice within the meaning of supervision law, even taking into account the provision of the ToB to the effect that ‘[w]here these Terms of Business conflict with Applicable Regulations, the latter shall prevail. Applicable

Regulations shall include the FSA Rules’. In Lord Hodge’s view, the sole purpose of the provision is to make clear that contractual provisions that differ from financial supervision law cannot be invoked by the bank against the FSA as regulator. In short, the contractual provision that no advice would be provided cannot be invoked by the bank against the FSA if the latter considers that for the purposes of supervision law advice has been provided and the applicable COBS rules have been contravened. According to Lord Hodge, the aim of this provision is not to incorporate the COBS rules into the contract, which would mean that the classification as advice under supervision law would also apply for civil law purposes, including all supervision rules applicable to the advice. An infringement of supervision rules on advice would then automatically constitute a breach of contract. However, Lord Hodge was not prepared to take this step.52 [9.36] Lord Hodge was crystal clear. A contractual provision to the effect that the bank acts solely as contractual counterparty and not as adviser is applicable even where a bank employee does actually provide the client with investment advice as referred to in MiFID I. This is so even if the IRSA contains a provision to the effect that financial supervision law prevails over the contract in cases where contractual provision is not compatible with the applicable regulatory regime. In Lord Hodge’s view, a statutory provision of this kind is intended solely to make clear that contractual provisions that differ from financial supervision law cannot be invoked by the bank against the FSA as regulator. In other words, the contractual provision that no advice is provided cannot be invoked by the bank against the FSA if the latter considers that advice was given for the purposes of supervision law and that the applicable supervision rules were infringed. According to Lord Hodge, the aim of this provision is not to incorporate the COBS rules into the contract, which would mean that the classification as advice under supervision law would also apply for civil law purposes, including all supervision rules applicable to the advice. [9.37] The civil courts in the Netherlands would probably take a different view of the matter. This is apparent, for example, from an interim judgment of Dordrecht District Court of 29 February 2012.53 This concerned the question of whether Rabobank had breached its civil duty of care in entering into an interest rate swap with Bugro BV (the owner of an inland

waterways oil tanker). In the transaction documents Rabobank had included a provision that it did not act as adviser and that the documents could not be construed as advice to enter into the specified transactions. The District Court took the view that this did not alter the fact that Rabobank had advised Bugro. In providing the advice, Rabobank was therefore subject to the obligation to meet the standard of care expected of a good services provider (Article 7:401 DCC). In other words, it had to demonstrate the competence and conduct that could reasonably be expected of a firm in this business. In short, in the light of what actually happened, the contractual provision that the bank did not provide advice had no effect since a bank employee had, in reality, provided investment advice to the corporate (nonprivate) client. This approach is in keeping with the German literature and also seems to me to be correct for Dutch law.54 Although in this case Dordrecht District Court made no mention of the principle that a rule may be derogated from on the grounds of good faith (redelijkheid en billijkheid), this could serve as a basis for this approach in the Netherlands. Another possible basis would be the argument that, in the case of SMEs, a contractual provision stipulating that the bank acts solely as contractual counterparty and does not provide advice is unreasonably onerous in view of the actual course of events.55 [9.38] If the contract explicitly provides that supervision law prevails over the contract (as in the Scottish case), a Dutch or German civil court would, in my view, have little hesitation in giving precedence to financial supervision law. A provision stipulating that the bank acts solely as contractual counterparty and does not provide advice would then be set aside on the basis of a reasonable interpretation of the contract. It seems unlikely that the Dutch and German civil courts would follow the lead of the Scottish court by interpreting the provision as merely clarifying the fact that the contractual provision (to the effect that no advice is provided) cannot be invoked by the bank against the regulator if the latter concludes that advice within the meaning of supervision law has been given and that the supervision rules have been breached. [9.39] GEL based its claim for damages not only on a direct infringement of the supervision rules but also, more indirectly, on negligent misrepresentation and negligent advice. However, Lord Hodge did not go

into this because he considered that the contract was clear. RBS never intended to enter into an investment advice relationship with GEL that would give rise to a duty of care. Insofar as GEL relied on the statements of RBS staff as investment advice, this reliance was not justified in view of the background of the clear arrangements between the parties as contained in the ToB, namely that there was no investment advice relationship. [9.40] More generally, Lord Hodge held that a common law duty of care cannot be based on supervision rules. The correctness of this assertion seems rather dubious under Scottish law. This is due in part to two English judgments, namely Gorham and others v British Telecommunications plc and others and Seymour v Ockwell. These judgments were not brought to the attention of Lord Hodge by the parties. In the latter judgment one of the judges held in relation to the supervision rules that ‘whilst the ambit of the duty of care owed by a financial adviser at common law is not necessarily co-extensive with the duties owed by that adviser under the applicable regulatory regime, the regulations afford strong evidence as to what is expected of a competent adviser in most situations’.56 Although English judgments are not binding on Scottish courts, they do constitute persuasive authority. If the parties had submitted these judgments to Lord Hodge, he might well have come to a different decision. [9.41] Whatever the case, the Dutch courts would presumably have concluded, unlike Lord Hodge, that the bank did have a civil duty of care (in functional terms somewhat comparable to common law duty of care) to GEL, and that this civil duty of care could also be defined in more concrete terms by reference to the applicable supervision rules. However, this is not entirely certain. As yet there is no case law whatever from the Dutch Supreme Court (Hoge Raad) on the civil duty of care to corporate (nonprivate) investors such as GEL. Naturally, however, there is extensive Supreme Court case law on the civil duty of care owed to retail investors, but it is debatable whether this case law can be applied one-on-one in relation to corporate (non-private) investors.57 [9.42] Finally, GEL based its claim on breach of contract. According to GEL, the contractual provision that no advice would be given amounted to an exclusion of liability for incorrect advice and was therefore an

unreasonably onerous clause to be incorporated in terms of business. The Scottish judge rejected this argument as well. This line of reasoning, based on the unreasonably onerous nature of the clause, might well stand a greater chance of success in the Netherlands. See my previous remarks about this. [9.43] Whatever the case, it is certain that a judgment exists in Scotland58 which provides that a contractual clause stating that an investment firm acts solely as contractual counterparty will be respected, even if, in reality, bank staff do provide advice. It is questionable whether this is compatible with the European principle of effectiveness. To assess this we first need to go back a little. In the judgment of the European Court of Justice in Genil 48 SL v Bankinter SA, it was held that in the absence of European legislation it was up to the Member States themselves to determine the contractual consequences of an infringement of the KYC rules, subject to observance of the principles of equivalence and effectiveness (effet utile) (para. 57).59 The Court of Justice referred here to paragraph 27 of a judgment of 19 July 2012 concerning a tax case (Littlewoods Retail and others, no. C-591/10) and the case law cited there. This reads as follows: In the absence of EU legislation, it is for the internal legal order of each Member State to lay down the conditions in which such interest must be paid, particularly the rate of that interest and its method of calculation (simple or compound interest). Those conditions must comply with the principles of equivalence and effectiveness; that is to say that they must not be less favourable than those concerning similar claims based on provisions of national law or arranged in such a way as to make the exercise of rights conferred by the EU legal order practically impossible [emphasis added]. (See, to that effect, San Giorgio, paragraph 12; Weber’s Wine World, paragraph 103; and Case C-291/03 MyTravel [2005] ECR I-8477, paragraph 17.)

[9.44] First, in the context of MiFID, the effectiveness principle would therefore seem to mean that the conditions on which an investor can hold an investment firm contractually liable may not be of such a nature as to make the successful exercise of rights impossible or extremely difficult in practice. It would thus seem to follow from this judgment that civil courts may not apply a more relaxed interpretation of MiFID rules. If, according to MiFID, the KYC rules are infringed in a specific case and the aggrieved investor sues for damages, the civil courts may not dismiss the claim on the ground that, in the particular circumstances of the case, it was not necessary to observe the KYC rules. This would, after all, be contrary to the

effectiveness principle. Moreover, it would jeopardize the high level of investor protection which MiFID aims to provide and be detrimental to the level playing field envisaged by MiFID. MiFID sets out, wherever possible, to maximize harmonization of financial supervision law for investment firms that provide investment services and carry out investment activities. It follows that the Member States no longer have the power to adopt more stringent or lenient supervision rules, unless this is apparent from MiFID itself. This approach is also conducive to legal certainty in the financial services sector and can be extended to claims for damages in respect of infringements of other MiFID provisions, such as an infringement of other conduct-of-business rules. [9.45] Second—and this brings us back to the Scottish case—it could be inferred from the Genil judgment that the civil courts are bound to hold that a contractual clause which is less stringent than MiFID cannot be applied. Naturally, this is not as strong as the argument that the civil courts may not apply a more relaxed interpretation of MiFID rules since the investor has himself agreed to the contract. Having said this, investors (even professional investors) often have little influence over contract terms. Needless to say, the fact that clauses of this kind jeopardize the high level of investor protection which MiFID aims to provide and are detrimental to the level playing field envisaged by MiFID is also a factor of importance. An example of a contract clause less stringent than MiFID is where an investment firm stipulates that it is acting solely as contractual counterparty, although in reality bank staff advise the client (as in the Scottish case). In keeping with the aim of the contract, the investment firm will then not fulfil the MiFID duties of care that apply to an investment firm. If, as a consequence of one or more of these infringements, the investor suffers loss, it could be argued that under the EU effectiveness principle he should have a real possibility of obtaining compensation for his loss. He will not do so if a contractual clause to the effect that the investment firm acts solely as contractual counterparty is held by the courts to remain fully effective. After all, a claim for compensation for infringement of MiFID rules on investment services would then be frustrated by the contractual clause.60

5. No Practicable Distinction under MiFID II either

[9.46] It is also apparent from MiFID II that the distinction between dealing on own account and executing orders on behalf of the client (and other forms of investment service) is not a practicable criterion for determining the degree of protection to which an investor is entitled. To achieve adequate investor protection, after all, MiFID II resorts to the artifice of reclassifying certain types of dealing on own account as acting on behalf of the client, specifically with a view to improving investor protection. [9.47] First, the definition of ‘execution of orders on behalf of clients’ has been modified to such an extent that some instances of dealing on own account have been reclassified and brought within its ambit, with the result that the definition of ‘dealing on own account’ is now much narrower. Likewise, under MiFID II the phrase ‘the conclusion of agreements to sell financial instruments issued by an investment firm or credit institution at the moment of their issuance’ comes within the definition of ‘execution of orders on behalf of clients’.61 What is the exact scope of this change? Some examples may help to clarify this. If an investment firm sells an investor shares in its own capital at the time of issuance and the sale does not involve the provision of any form of investment service, the investment firm acts solely as the investor’s contractual counterparty. Under MiFID I this is an instance of dealing on own account. Under MiFID II, however, it is reclassified as acting on behalf of the client and is suddenly treated as a form of investment service. Issuance is usually taken to mean the issuance of marketable shares and bonds, but in MiFID II it has a broader meaning. In the terminology of MiFID II the concept of issuance is linked to financial instruments. This means that where an investment firm acts as contractual counterparty in an interest rate swap this too is treated as the conclusion of an agreement for the sale, at the time of issuance, of a financial instrument issued by an investment firm. After all, an interest rate swap is a financial instrument, like many other derivatives. This interpretation also benefits investor protection, which is one of the key objectives of MiFID. Recital 45 in the preamble to MiFID II explicitly states that this reclassification is intended ‘to eliminate uncertainty and strengthen investor protection’. [9.48] Second, although this is apparent not from the broadening of the definition of ‘execution of orders on behalf of clients’ but from Recital 24

in the preamble to MiFID II, matched principal trading (back-to-back trading) is regarded, inter alia, as execution of orders on behalf of the client, although under MiFID I it was treated solely as dealing on own account. In Article 4(1), point (38) of MiFID II matched principal trading is defined as: a transaction where the facilitator interposes itself between the buyer and the seller to the transaction in such a way that it is never exposed to market risk throughout the execution of the transaction, with both sides executed simultaneously, and where the transaction is concluded at a price where the facilitator makes no profit or loss, other than a previously disclosed commission, fee or charge for the transaction.

In terms of economic result, matched principal trading resembles the position in which the firm acts on both sides of a transaction for the client, that is, matching opposite client orders (agency crosses). [9.49] More precisely, Recital 24 in the preamble to MiFID II provides that dealing on own account when executing client orders [i.e. (systematic) internalization] should include firms executing orders from different clients by matching them on a matched principal basis (back-to-back trading), which should be regarded as acting as principal and should be subject to the provisions of this Directive covering both the execution of orders on behalf of clients and dealing on own account.

Equating matched principal trading with (systematic) internalization is in fact based on a fallacy. In economic terms, matched principal trading much more closely resembles agency crosses, as opposite client orders are in fact matched with one another. [9.50] In any event, these two instances of reclassification enhance investor protection, but this is in my view not sufficient. If an investment firm sells a financial instrument that it has not issued itself, I cannot see any reason why the investor should not enjoy the protection of the MiFID II duties of care that apply to execution-only services.

6. Dutch Supreme Court has Already Extended Civil Duty of Care to Dealing on own Account

[9.51] Moreover, the Dutch Supreme Court has already extended the special civil duty of care to dealing on own account. In a case involving the offering of risky and complex financial products to retail investors, it held that it followed from the special civil duty of care that there was a duty to warn investors of the risks involved and a duty to comply with KYC rules, even though the bank was only acting as contractual counterparty.62

7. The UK Government Response to the Kay Review [9.52] Finally, it is worth briefly mentioning the Kay Review of UK Equity Markets and Long-term Decision Making. This report indicated that all participants in the equity investment chain should observe the same standards.63 The UK Government in its response to the Report agreed: The Government believes that it would be appropriate to instead adopt the following principle for equity markets, which reflects Kay’s Good Practice Statements for Asset Managers and Asset Holders: All participants in the equity investment chain should act: • in good faith; • in the best long-term interests of their clients or beneficiaries; • in line with generally prevailing standards of decent behaviour. This means ensuring that the direct and indirect costs of services provided are reasonable and disclosed, and that conflicts of interest are avoided wherever possible, or else disclosed or otherwise managed to the satisfaction of the client or beneficiary. These obligations should be independent of the classification of the client. They should not be contractually overridden.64

[9.53] The UK Government has asked the FSA (and its successor the FCA) ‘to consider to what extent current regulatory rules in this area align with this principle, with particular reference to the issues raised in the Kay Report around conflicts of interest requirements and contractual mechanisms to limit the obligations of intermediaries, to determine what action might be desirable’.65 It has added as follows: ‘We note that the current regulatory rules in this area are substantially influenced by harmonised EU legislation, and that changes to regulatory requirements at EU level may therefore be desirable.’

[9.54] This latter statement is significant, given this chapter’s argument that even the new MiFID II regime falls short. Given that MiFID II, like MiFID I, provides for the most part for maximum harmonization, the FCA would not seem to have the authority to introduce rules extending the high duties of care to all intermediaries, including those who act purely as contractual counterparties. The FCA could conceivably invoke Article 24(12) MiFID II, which provides that in exceptional circumstances Member States may impose additional requirements on investment firms. However, such requirements must be objectively justified and proportionate so as to address specific risks to investor protection or to market integrity which are of particular importance in the circumstances of the market structure of the Member State concerned. It seems unlikely that this test could be met.

V. Conclusion [9.55] This chapter has focused on the question of whether allowing the extent of the protection afforded to an investor under MiFID to be largely dependent on the distinction between dealing on own account on one hand and trading on behalf of the client (and other forms of investment service) on the other is justified. This question must be answered in the negative. An investor may reasonably expect the investment firm used by him to look after his interests adequately and thus to observe certain duties of care towards him. The investment firm is, after all, ideally placed to use its expertise. Its fund of knowledge is bound to be superior to that of an investor, particularly a retail investor. Nor is this any different where the investment firm acts purely as the investor’s contractual counterparty. In such cases, the investor is reasonably entitled to expect the investment firm to observe the same duty of care that would apply if it were providing an execution-only service. Moreover, the distinction between dealing on own account (principal dealing) on one hand and trading on behalf of the client (and other forms of investment service) on the other is tenuous, arbitrary, and easy to manipulate. This is all the more so where a contractual clause providing that an investment firm is acting solely as contractual counterparty is claimed to apply even where an employee of the investment firm advises the investor, contrary to the terms of the agreement. Clearly,

MiFID II also provides no practicable criterion. Indeed, to achieve an adequate level of investor protection MiFID II resorts to the artifice of reclassifying certain types of dealing on own account as acting on behalf of the client. Finally, both the UK Government (in response to the Kay Review) and the Dutch Supreme Court take the view that duties of care must also apply where an investment firm acts solely as an investor’s contractual counterparty. Under a future MiFID III, an investment firm which acts solely as contractual counterparty should be required to observe the same duty of care as applies in the case of an execution-only service.

1

Directive 2004/39/EC, OJ L 145, 30 April 2004, pp. 1–47 (MiFID I); Commission Directive 2006/73/EC, OJ L 241, 2 September 2006, pp. 26–58 (MiFID I Implementing Directive); Commission Regulation (EC) No. 1287/2006, OJ L 241, 2 September 2006, pp. 1–25 (MiFID I Implementing Regulation). 2 Directive 2014/65/EU, OJ L 173, 15 May 2014, pp. 349–496 (MiFID II); Regulation (EU) No. 600/2014, OJ L 173, 15 May 2014, pp. 84–148 (MiFIR). See Chapter 1, n. 2. 3 Recital 2 in the preamble to MiFID I; Recital 70 in the preamble to MiFID II. 4 Where an order is received and transmitted to another investment firm, this does not constitute carrying out a transaction in financial instruments. In such a case it is the other investment firm which carries out the transaction in one of the ways described in the main body of this chapter’s text (unless that other investment firm receives the order and then transmits it to yet another investment firm). This does not cause problems in investor protection. Like the carrying out of transactions in financial instruments on behalf of a client, the reception and transmission of orders constitutes an investment service. In both cases, therefore, there is a fairly high level of investor protection. 5 ‘Regulated market’ is defined in Article 4(1), point (14) MiFID I, and Article 4(1), point (21) MiFID II. 6 Turquoise is majority owned by the London Stock Exchange Group in partnership with the user community (see ). Multilateral trading facility (MTF) is defined in Article 4(1), point (15) MiFID I and Article 4(1), point (22) MiFID II. MTFs perform the same function in an economic sense as regulated markets, but are subject to supervision rules that differ to some extent from those for regulated markets. As operating an MTF constitutes an investment activity, the operator is deemed to be an investment firm and is subject to the MiFID rules that apply to investment firms. As operating a regulated market does not constitute an investment service or investment activity, the operator is not deemed to be an investment firm (in any event in this capacity). The operator is admittedly subject to MiFID rules, namely the rules governing the operation of a regulated market, but these are not the same as those applicable to investment firms.

7

See the definition of ‘execution venue’ in Article 44(1), in fine, MiFID I Implementing Directive: ‘a regulated market, an MTF, a systematic internaliser, or a market maker or other liquidity provider or an entity that performs a similar function in a third country to the functions performed by any of the foregoing’. In Article 64(1), in fine, Draft Commission Delegated Regulation, C (2016) 2398 final, 25 April 2016 this definition has been expanded so as to include OTFs. 8 It is also conceivable that an investment firm which executes an order for a client is itself the operator of an MTF or even of a regulated market. In such a case, no external party is involved in the carrying out of the transaction because the MTF or regulated market forms part of the investment firm’s own organization. 9 Naturally, there may also be a situation in which two or more parties act as seller, each contributing part of the 100 Shell shares. 10 See Article 4(1), point (5) MiFID I. The definition has been expanded in MiFID II (Article 4(1), point (5)). On this subject see Section IV.5. 11 See Section III below. 12 See D. Busch, Indirect Representation in European Contract Law (PhD Utrecht), (The Hague: Kluwer Law International, 2005), pp. 5–10 (the Netherlands), 75–9 (Germany), 128–33 (England and Wales). 13 See ibid., pp. 10–15 (the Netherlands), 79–83, 100–5 (Germany). As regards stockbrokers, see also L. D. van Setten, De commissionair in effecten (PhD Utrecht), (Deventer: Kluwer, 1998). 14 See the former Articles 76–85a of the Commercial Code (Wetboek van Koophandel). 15 See on commission agency Peter G. Watts, Bowstead & Reynolds on Agency, 20th edn, (London: Sweet & Maxwell, 2014), § 1-021; Busch (n. 12), 168–73. 16 See on undisclosed agency Watts (n. 15), § 8-068–8-082; Busch (n. 12), 133–6 and 139–68. 17 Article 4(1), point (6) MiFID I. Although Article 4(1), point (6) MiFID II contains the same definition, fewer cases are designated as dealing on own account under MiFID II. On this point see Section IV.5 below. 18 See Section III below. 19 For detailed accounts of systematic internalization, see: Guido Ferrarini and Fabio Recine, ‘The MiFID and Internalisation’, in Guido Ferrarini and Eddy Wymeersch (eds), Investor Protection in Europe: Corporate Law Making, the MiFID and Beyond (New York: OUP, 2006), p. 235; Johannes Köndgen and Erik Theissen, ‘Internalisation under MiFID: Regulatory Overreaching or Landmark in Investor Protection’, ibid., p. 271; S. Mutschler, Internalisierung der Auftragsausführung im Wertpapierhandel. Eine rechtliche und ökonomische Betrachting unter besonderer Berücksichtigung der Richtlinie 2004/39/EG des Europäischen Parlaments und des Rates vom 21 April 2004 über Märkte für Finanzinstrumente (PhD Universität Freiburg), (Baden-Baden: Nomos Verlag, 2007); A. Stefanski, Eigenhandel für andere. Market-making und Internalisierung im deutschen,

europäischen und US-amerikanischen Recht (PhD Universität Bonn), (Baden-Baden: Nomos Verlag, 2009). 20 See explicitly Recital (69) MiFID I Implementing Directive: ‘Dealing on own account with clients by an investment firm should be considered as the execution of client orders, and therefore subject to the requirements under Directive 2004/39/EC and this Directive and, in particular, those obligations in relation to best execution.’ This is repeated in Recital (103), Article 64(1), in fine, Draft Commission Delegated Regulation, C (2016) 2398 final, 25 April 2016. See on the duty of care Section III below. 21 A systematic internalizer quotes a price at which it is prepared to carry out a transaction in listed shares internally, provided that a liquid market exists in these shares, and makes this quote public on a reasonable commercial basis and in a manner which is easily accessible. In the case of shares to which there is not a liquid market, systematic internalizers must disclose quotes to their clients on request. See Article 27(1), first paragraph, MiFID I. In MiFID II these transparency obligations apply not only to dealing in shares but also to all kinds of other financial instruments. See Article 14(1) and Article 18(1) MiFIR. 22 In Article 4(1), point (7) MiFID I ‘systematic internaliser’ is defined as ‘an investment firm that, on an organised, frequent and systematic basis, deals on own account by executing client orders outside a regulated market or an MTF’. Article 4(1) (20) MiFID II contains a comparable definition. The criteria for determining whether an investment firm is a systematic internalizer are set out in Article 21 of the MiFID I Implementing Regulation. In Draft Commission Delegated Regulation, C (2016) 2398 final, 25 April 2016, the criteria for determining whether an investment firm is a systemic internalizer are spelled out in more detail, distinguishing between systemic internalizers for (1) equity and equity-like instruments (Article 12), (2) bonds (Article 13), (3) structured finance products (Article 14), (4) derivatives (Article 15), and (5) emission allowances (Article 16). 23 Investment firms which, either on own account or on behalf of clients, conclude transactions in shares admitted to trading on a regulated market outside a regulated market or MTF are required to make public the volume and price of those transactions and the time at which they were concluded. See Article 28(1) MiFID I. In MiFID II these posttrade disclosure obligations apply not only to shares but also to all kinds of other financial instruments. See Article 20(1) and Article 21(1) MiFIR. The execution of transactions outside a regulated market or MTF in fact comprises not only execution by means of internalization but also the matching of opposite client orders (agency crosses). See Section II.5 below. 24 The other sub-category of Insichgeschäft concerns serving two masters. On this subject, see Section II.5 below. 25 Watts (n. 15), § 6-064. 26 ibid, § 6-065. 27 ibid, § 6-066. 28 For a detailed comparison of Insichgeschäft and the English concept of self-dealing, see: S. Festner, Interessenkonflikte im deutschen un englischen Vertretungsrecht (PhD MPI

Hamburg), (Tübingen: Mohr Siebeck, 2006), p. 69 ff. 29 See for a treatment of Article 7:416 DCC in relation to transactions in financial instruments D. Busch, Vermogensbeheer (Monografie BW B8), (Deventer: Kluwer, 2014), § 17.3.2. 30 Naturally, it is also quite possible that where an investment firm is instructed to buy 100 Shell shares it is able to match this order with two or more orders to sell Shell shares. 31 See Section III below. 32 See, for example, Article 20(6), second paragraph, MiFID II. 33 See Recital 9 in the preamble to MiFIR. See also Section III below. 34 The other subcategory of Insichgeschäft concerns cases in which the representative acts as contractual counterparty in the transaction. On this point see Section II.4 above. 35 Watts (n. 15), § 6-048. 36 ibid. 37 For a detailed comparison of Insichgeschäft and the English concept of self-dealing, see: Festner (n. 28), 249. 38 See for a treatment of Article 7:417 DCC in relation to transactions in financial instruments Busch (n. 29), § 17.3.3. 39 Article 2(1)(d) MiFID I. 40 See Article 2(1)(d) and (j), MiFID II. 41 Article 4(1), point (10) MiFID I and Article 4(1), point (9) MiFID II. 42 For the European Commission’s letter, see: Working Document ESC-07-2007, Commission answers to CESR scope issues under MiFID and implementing directive (Appendix to CESR, Best Execution under MiFID, Questions & Answers, May 2007, CESR/07-320). 43 Court of Session 21 August 2012 [2012] CSOH 133. 44 Now a Justice in the UK Supreme Court. 45 Initially, the claim for damages was also based on fraudulent misrepresentation, but this ground was retracted in the course of the action (see para. 8 of the judgment). Nonetheless, Lord Hodge did consider this ground (see paras 85–93 of the judgment). I will not give any further consideration to this retracted ground here. 46 See paras 1–8 of the judgment. 47 Merely a synonym here for contractual counterparty. See about the confusing term ‘execution only’ Chapter 1, n. 17. 48 See para. 29 of the judgment. 49 Since 1 April 2013 this provision can be found in section 138D of the Financial Services and Markets Act (FSMA) 2000. 50 See paras 31, 40, 43, and 44 of the judgment. 51 See paras 45–62 of the judgment. 52 See paras 63–67 of the judgment.

53

Dordrecht District Court, 29 February 2012, FR 2012/48 (Bugro v Rabobank), ground

4.17. 54

See M.-P. Weller, Die Dogmatik des Anlageberatungsvertrags. Legitimation der strengen Rechtsprechungslinie von Bond bis Ille ./. Deutsche Bank, ZBB (2011), pp. 191–9, at p. 193 (left column). 55 See Article 6:248(2) DCC and Article 6:233(a) in conjunction with Article 6:235(1) DCC. Cf. D. Busch, ‘Best Execution’, in D. Busch and C. M. Grundmann-van de Krol (eds), Handboek beleggingsondernemingen (Deventer: Kluwer, 2009), pp. 715–17. 56 Gorham v British Telecommunications plc [2000] 1 WLR 2129, 2141 (CA) per Pill LJ (‘The courts can be expected to attach considerable weight to the content of codes […] but are not excluded from making their own assessment of a situation.’); Seymour v Ockwell [2005] EWHC 1137 (QB), 77 per Judge Havelock-Allan QC. On this subject, see for example L. D. van Setten and T. Plews, Chapter 11 (England & Wales) in D. Busch and D. A. DeMott (eds), Liability of Asset Managers (Oxford: OUP, 2012), §§ 11.24–11.27. 57 As regards the civil duty of care towards retail investors, see inter alia: HR 23 March 2007, NJ 2007/333, annotated by Mok (ABN AMRO v Van Welzen) (duty of care in relation to trade in options); HR 5 June 2009, JOR 2009/199, annotated by Lieverse (Treek v Dexia Bank Nederland) (duty of care in relation to offering of securities lease products); HR 24 December 2010, NJ 2011/251, annotated by Tjong Tjin Tai (Fortis Bank v Bourgonje) (duty of care of portfolio manager); HR 3 February 2012, AA (2012) 752, annotated by Busch (Coöperatieve Rabobank Vaart en Vecht UA v X) (duty of care when providing investment advice). See for further references to Dutch Supreme Court case law: Busch (n. 29), § 15.1. 58 Albeit a decision of one judge in the Outer House, that is the lower house of the Court of Session. Only a decision of the Inner House would bind a Scottish judge in future. 59 EU CoJ 30 May 2013, no. C-604/11, AA (2013) 663, annotated by Busch; JOR 2013/274, annotated by Busch (Genil 48 SL and others v Bankinter SA and others). 60 The effectiveness principle has been explicitly codified in MiFID II (unlike MiFID I). Article 69(2) MiFID II provides, in fine: ‘Member States shall ensure that mechanisms are in place to ensure that compensation may be paid or other remedial action be taken in accordance with national law for any financial loss or damage suffered as a result of an infringement of this Directive or [MiFIR]’. 61 Article 4(1)(5) MiFID II. 62 See HR 5 June 2009, JOR 2009/199, annotated by Lieverse (Treek v Dexia Bank Nederland), ground 5.2.1. 63 The Kay Review of UK Equity Markets and Long-Term Decision Making (Final Report July 2012), p. 65. 64 Ensuring Equity Markets Support Long-Term Growth. The Government response to the Kay review (November 2012), para. 2.8. 65 Government response (n. 64), para. 3.35.

10 MIFID II/MIFIR’S REGIME FOR THIRDCOUNTRY FIRMS Danny Busch and Marije Louisse

I. General II. 1. 2. 3. 4. 5.

Scope of MiFID II/MiFIR’s Regime for Third-Country Firms Third-Country Firms Investment Services and Investment Activities Selling and Advising Clients in Relation to Structured Deposits Data-Reporting Services Operating Trading Venues

III. 1. 2. 3. 4. 5. 6.

Eligible Counterparties and Per Se Professional Clients General Equivalence Decision by the Commission Registration with ESMA Continuing Obligations for Registered Third-Country Firms Transitional Regime Withdrawal of Registration by ESMA/Equivalence Decision by the Commission 7. No Equivalence Decision by the Commission

IV. 1. 2. 3. 4.

Retail Clients and Opt Up Professional Clients General Exercise of Member State Option Non-Exercise of Member State Option No European Passport!

5. Resolution of a Branch of a Third-Country Firm V. Retail Clients, Professional Clients, and Eligible Counterparties 1. After the Adoption of an Equivalence Decision and Exercise of the Member State Option 2. If no Equivalence Decision is in Effect 3. If the Member State Option has not been Exercised VI. 1. 2. 3.

Initiative Test Initiative Test Under MiFIR Initiative Test Under MiFID II Own Exclusive Initiative

VII. Conclusion

I. General [10.01] MiFID II/MiFIR introduces major changes for third-country firms —that is, investment firms and credit institutions, providing investment services and performing investment activities, established outside the EU/EEA. Under MiFID I, the position of such firms is a purely national matter. Each country within the EU/EEA has to decide for itself on the access of a third-country firm to its own national market. As the national regimes differ quite markedly from one another, the overall picture is fragmented.1 Consequently, a third-country firm cannot gain access to the European market all at once, and must instead apply to each country individually for access to its national market on the basis of the different requirements.2 This was considered by the Commission to damage the functioning of the single market as well as to create additional costs for these firms.3 [10.02] Under MiFID II/MiFIR, the position of third-country firms will no longer be a purely national matter. MiFID II/MiFIR now itself sets the parameters. It introduces a common regulatory framework that should harmonize the existing fragmented framework for the provision of services by third-country firms, ensure certainty and uniform treatment of thirdcountry firms accessing the EU/EEA, ensure an assessment of effective

equivalence by the Commission, and provide for a comparable level of protection to clients in the EU/EEA receiving services from third-country firms.4 However, as will be seen below, the legislation is unfortunately complex and does not provide for full harmonization. The Commission and the Member States found it hard to agree on all aspects of the regime for third-country firms. [10.03] This chapter is structured as follows. We first discuss the scope of MiFID II/MiFIR’s regime for third-country firms (Section II). Subsequently, we explain MiFIR’s regime for third-country firms (Section III) and MiFID II’s regime for third-country firms (Section IV). In Section V we discuss the friction between both regimes when a third-country firm provides investment services to eligible counterparties, professional clients, and retail clients. In Section VI we take a closer look at the initiative test. The Conclusion (Section VII) includes a diagram of the provisions of MiFID II/MiFIR’s third-country regime.

II. Scope of MiFID II/MiFIR’s Regime for ThirdCountry Firms 1. Third-Country Firms [10.04] Both the MiFID II third-country regime (Articles 39 up to 43 MiFID II) and the MiFIR third-country regime (Articles 46–49 MiFIR) apply to ‘third-country firms’.5 A third-country firm is a ‘firm that would be a credit institution providing investment services or performing investment activities or an investment firm if its head office or registered office were located within the Union’.6 A third-country firm is therefore a party with its head office or registered office outside the EU/EEA that would have qualified as a credit institution or investment firm if it had its head office or registered office in the EU/EEA.

2. Investment Services and Investment Activities [10.05] Under the MiFID II/MiFIR’s third-country regime, third-country firms may provide investment services or perform investment activities with or without any ancillary services.7 The MiFID II/MiFIR’s thirdcountry regime hence does not apply if a third-country firm provides any other services, such as selling, and advising clients in relation to, structured deposits or data-reporting services. We refer to Sections II.3 and II.4. [10.06] Different regimes apply for the provision of investment services depending on the type of client to whom the investment services are provided. The MiFID II third-country regime applies to the provision of investment services to retail and opt up professional clients (i.e. clients who have obtained professional client status by opting up)8 and the MiFIR thirdcountry regime applies to the provision of investment services to eligible counterparties and per se professional clients (i.e. with the exception of clients who have obtained professional client status by opting up).9 For further details of the regimes we refer to Sections III and IV. [10.07] The question is what regime applies to the performance of investment activities, since there is no client involved in the performance of such activities. If, for example, the third-country firm provides investment services to eligible counterparties and per se professional clients and performs investment activities, it could be argued from a pragmatic point of view that the third-country firm may perform such investment activities in compliance with the regime that applies to the provision of investment services to eligible counterparties and per se professional clients (the MiFIR third-country regime). This would entail that the third-country firm can register itself with ESMA, if the Commission has taken an equivalence decision, and can subsequently perform the investment activity (e.g. dealing on own account) throughout the EU/EEA.10 But what if the third-country firm intends to perform investment activities only, without providing any investment services, in the EU/EEA? What regime would then apply? Should the firm obtain an ESMA registration in accordance with the MiFIR third-country regime or should it open a branch, if the Member State in which the third-country firm intends to perform investment activities has

opted for the branch requirement, in accordance with the MiFID II thirdcountry regime? [10.08] In order to answer these questions it could be relevant to determine when investment activities are actually performed in the EU/EEA. For the provision of investment services, this can be determined by applying the initative test as set out in Articles 42 MiFID II and 46(5), third paragraph, MiFIR (see Section VI). As there are no clients involved in the performance of investment acitvities, the initiative test seems not to be of much help here. Articles 42 MiFID II and 46(5), third paragraph, MiFIR do however also contain a reference to investment activities.11 One could therefore argue that investment activities are performed in the EU/EEA when the counterparty is located in the EU/EEA (i.e. if the investment activity concerns dealing on own account) or when the multilateral trading facility (MTF) or organized trading facility (OTF) that is operated by the third party is open for members or participants that are located in the EU/EEA, and that it should therefore be possible to rely on the iniative test, if investment activities are performed on the exclusive initative of such counterparty or member/participant.12 Subsequently, one could argue that the identity of such counterparty or member/participant determines which third-country regime applies. The European legislator has, however, not provided any further guidance in that respect.

3. Selling and Advising Clients in Relation to Structured Deposits [10.09] MiFID II contains a new requirement in Article 1(4) for investment firms and credit institutions to comply with certain provisions of MiFID II when selling or advising clients in relation to structured deposits.13 Article 1(4) MiFID II does not contain a reference to thirdcountry firms,14 and as such this new requirement seems not to apply to third-country firms that sell or advise clients in relation to structured deposits. The MiFID II/MiFIR’s third-country regime also does not provide for a basis to apply Article 1(4) MiFID II to third-country firms, since the MiFID II/MiFIR’s third-country regime only applies to the provision of

investment services and performance of investment activities.15 Structured deposits do not qualify as financial instruments. Any services or activities that are conducted by a third-country firm in relation to structured products therefore do not qualify as the provision of investment services or the performance of investment activities. As a result, we are inclined to say that selling, and advising on, structured products is not an activity that is included in the scope of MiFID II/MiFIR’s third-country regime.16 This would have as a consequence that each Member State can set its own requirements for access to its market by third-country firms wishing to sell and advise clients in relation to structured deposits in its jurisdiction. [10.10] The wording of the initiative test under MiFID II/MiFIR17 is, however, somewhat confusing in this respect, since it includes a reference to investment products (which term includes structured deposits). The reference to investment products in the wording of the initative test seems to indicate that marketing investment products would be a regulated activity under MiFID II/MiFIR, if the third-country firm cannot rely on the initiative test. As said, Articles 1(4) and 39(1) MiFID II and Article 46(1) MiFIR indicate otherwise. Hopefully, this will be clarified in future amendments of MiFID II/MiFIR.

4. Data-Reporting Services [10.11] MiFID II introduces supervision on a new category of entities, namely data-reporting services providers.18 The provision of data-reporting services will be subject to prior authorization by the home Member State’s competent authority.19 The home Member State is the Member State where the data-reporting services provider has located its registered office or head office.20 The authorization will be valid for the entire EU/EEA and will allow the data-reporting services provider to provide the services for which it has been authorized throughout the EU/EEA. [10.12] MiFID II does not contain a third-country regime for datareporting services providers. It therefore seems that data-reporting services providers from third countries will have to establish a subsidiary in the

EU/EEA that will subsequently have to obtain an authorization to be able to provide these services in the EU/EEA, unless a national regime provides otherwise. [10.13] Member States may allow an authorized investment firm operating a trading venue (MTF or OTF) to provide data-reporting services subject to the prior verification of its compliance with Title V of MiFID II.21 It could therefore be that a branch of a third-country firm that has obtained an authorization pursuant to Article 39 of MiFID II may also provide such data-reporting services throughout the entire EU/EEA.22 This seems, however, not to be expressly intended under MiFID II.

5. Operating Trading Venues [10.14] Trading venues are regulated markets, MTFs, or OTFs.23 The operation of an MTF or OTF is an investment activity that may be performed by a third-country firm under the MiFID II/MiFIR’s thirdcountry regime. MiFID II/MiFIR’s third-country regime does not cover market access of third-country (operators of) regulated markets. Such market access will continue to be a purely national matter.24 [10.15] Notwithstanding the foregoing, MiFIR contains several provisions that apply to ‘third-country trading venues’, a term that is not defined in MiFIR. It is our understanding that the term refers to (operators) of trading venues that are established outside the EU/EEA.25 This could be thirdcountry firms that operate an MTF or OTF, or (operators of) regulated markets established outside the EU/EEA. The following provisions of MiFIR apply to such third-country trading venues: 1. A third-country trading venue may request access to a central counterparty (CCP) established in the EU/EEA, only if the Commission has adopted an equivalence decision relating to that third country. In accordance with Article 28(4) MiFIR the Commission may adopt an equivalence decision, if the legal and supervisory framework of a third country ensures that a trading venue authorized in that third country complies with legally binding requirements which are equivalent to the requirements for trading venues, resulting

from MiFIR, MiFID II, and MADII/MAR,26 and which are subject to effective supervision and enforcement in that third country.27 2. A third-country trading venue is only permitted to make use of the access rights in Articles 35 to 36 of MiFIR if the Commission has adopted a decision that the legal and supervisory framework of the third country is considered to provide for an effective equivalent system for permitting CCPs and trading venues authorized under foreign regimes (e.g. EU/EEA CCPs and trading venues) access to CCPs and trading venues established in that third country.28 3. Third-country trading venues may only request a licence and the access rights in accordance with Article 37 of MiFIR if the Commission has adopted a decision that the legal and supervisory framework of that third country is considered to provide for an effective equivalent system under which CCPs and trading venues authorized in foreign jurisdictions (e.g. EU/EEA CCPs and trading venues) are permitted access on a fair, reasonable, and non-discriminatory basis to: (a) relevant price and data feeds and information of composition, methodology, and pricing of benchmarks for the purposes of clearing and trading, and (b) licences, from persons with proprietary rights to benchmarks established in that third country.29 [10.16] Finally, we note that financial counterparties (as defined in EMIR30) and certain non-financial counterparties may conclude certain transactions with other such financial or non-financial counterparties in derivatives pertaining to a class of derivatives that has been declared subject to the trading obligation in accordance with the procedure set out in Article 32 MiFIR and listed in the register referred to in Article 34 MiFIR on thirdcountry trading venues, provided that the Commission has taken an equivalence decision and provided that the third country provides for an equivalent system for recognition of EU/EEA trading venues.31

III. Eligible Counterparties and Per Se Professional Clients

1. General [10.17] The regime for the provision of investment services to eligible counterparties and per se professional clients following an equivalence decision by the Commission is laid down in Articles 46–49 of MiFIR and is therefore directly applicable in the Member States. In the absence of an equivalence decision by the Commission, the provision of investment services to eligible counterparties and per se professional clients is, however, still a national matter (Article 46(4) MiFIR).

2. Equivalence Decision by the Commission [10.18] The Commission may adopt an equivalence decision in relation to a third country stating that (i) the legal and supervisory arrangements of that third country ensure that third-country firms authorized in that third country comply with legally binding prudential and business conduct requirements which have equivalent effect to the requirements set out in MiFID II/MiFIR and CRD IV/CRR32 (for this purpose it examines the authorization process, supervision, enforcement, capital requirements, governance, organizational requirements, conduct of business rules, and rules on market transparency and integrity) and (ii) the legal framework of that third country provides for an effective equivalent system for the recognition of investment firms authorized under third-country legal regimes (Article 47(1) of MiFIR). As a result, the equivalence decision by the Commission is not only an assessment in respect to the equivalence of the legal systems of the EU/EEA and third-country legal regimes, but also in respect to reciprocity (mutual recognition). [10.19] Taking into account that the Commission may adopt an equivalence decision, it is our understanding that there is no obligation for the Commission to adopt an equivalence decision, if there is equivalence and reciprocity in respect to a certain third country.33

A. Reciprocity Test

[10.20] Under MiFID I, reciprocity already was acknowledged as an important factor in the relations with third countries. According to Article 15 of MiFID I, Member States should inform the Commission of any general difficulties that their investment firms encounter in establishing themselves or providing investment services, and/or performing investment activities in any third country. Whenever it appeared that a third country did not grant EU/EEA investment firms effective market access comparable to that granted by the EU/EEA to third-country firms from that third country, the Commission could arrange for negotiations with the third country with a view to obtaining comparable competitive opportunities. If the conditions of effective market access were not fulfilled and EU/EEA investment firms were not granted national treatment affording the same competitive opportunities as were available to domestic investment firms, the Commission could decide that the competent authorities of the Member States must limit or suspend their decisions regarding requests pending or future requests for authorization and the acquisition of holdings by direct or indirect parent undertakings governed by the law of the third country in question for a maximum period of three months (a period which could be extended by the Commission). Such measures could then be used to exert pressure on the negotiations. Although Article 15 is withdrawn under MiFID II/MiFIR, reciprocity still is of importance in the decision by the Commission to adopt an equivalence decision under Article 47 of MiFIR (we refer to paragraph 10.18).

B. Equivalence Test [10.21] While the importance of reciprocity was stressed by many of the banks and brokers in their replies to the public consultation on MiFID II/MiFIR, because of possible increased competition with no reciprocity, other market parties (also including asset managers, funds, exchanges, and market operators) were mostly concerned about a ‘strict equivalence’ test for third-country firms.34 This could be viewed as protectionist and, as a result, could threaten access of EU/EEA investment firms to third countries.35

[10.22] A ‘strict equivalence’ test would entail that the Commission, instead of focusing on the result, would focus on specific means to achieve equivalence. To pass the strict equivalence test the EU framework for MiFID II services should be replicated in third countries. This is unrealistic. The concerns on the ‘strict equivalence’ test have been taken into account by the realization of MiFIR. The final text of Article 47 now reads that the prudential and business conduct framework of a third country may be considered to have equivalent effect where it fulfils all the conditions of Article 47(1) MiFIR, where the Commission Proposal stated that the framework may be considered equivalent where it fulfils all the conditions of (at that time) Article 37(1) MiFIR.36

C. The Equivalence and Reciprocity Test in Other EU/EEA Legislation [10.23] The fact that MiFID II/MiFIR regulates the position of thirdcountry firms reflects a European trend. EMIR, the CRA Regulation,37 and the AIFMD38 all deal with the position of parties (CCPs, credit rating agencies (CRAs), managers of alternative investment funds (AIFMs), and depositaries) established in a third country.39 These regimes all contain an equivalence and reciprocity—or similar—test, though the scope of and assessment for each of such regimes differ. [10.24] Under EMIR, a CCP established in a third country may, for example, provide clearing services to clearing members or trading venues established in the EU/EEA only where that CCP is recognized by ESMA.40 One of the conditions for such recognition is that the Commission has adopted an implementing act determining that the legal and supervisory arrangements of a third country ensure that CCPs authorized in that third country comply with legally binding requirements which are equivalent to requirements laid down in Title IV of EMIR, that those CCPs are subject to effective supervision and enforcement in that third country on an ongoing basis, and that the legal framework of that third country provides for an effective equivalent system for the recognition of CCPs authorized under

third-country legal regimes.41 Currently, the Commission has adopted ten equivalence decisions following the technical advice of ESMA.42 [10.25] The CRA Regulation contains an equivalence test for CRAs. In accordance with Article 5(6) of the CRA Regulation, the Commission may adopt an equivalence decision stating that the legal and supervisory framework of a third country ensures that CRAs authorized or registered in that third country comply with legally binding requirements which are equivalent to the requirements resulting from the CRA Regulation and which are subject to effective supervision and enforcement in that third country and the regulatory regime in that third country prevents interference from the supervisory authorities and other public authorities in the content of credit ratings and methodologies. Currently, the Commission has adopted nine equivalence decisions following the technical advice of ESMA.43 [10.26] The AIFMD contains a ‘same effect’ test for depositaries. In the case of a third-country alternative investment fund (AIF), the depositary can be in the third country in which the AIF is domiciled, inter alia, if the depositary is subject to effective prudential regulation, including minimum capital requirements and supervision which have the same effect as EU/EEA law, and are effectively enforced.44 The Commission will adopt implementing acts, stating that the prudential regulation and supervision of a third country have the same effect as EU/EEA law and are effectively enforced.45 Currently, no such implementing acts have been adopted.46 In addition, Article 37 of the AIFMD provides for the authorization of nonEU/EEA AIFMs intending to manage EU/EEA AIFs and/or market AIFs managed by them in the EU/EEA.47 Such a non-EU/EEA AIFM has to comply with the AIMFD, unless such compliance is incompatible with the law to which the non-EU/EEA AIFM and/or the non-EU/EEA AIF marketed in the EU/EEA is subject and the non-EU/EEA AIFM can demonstrate that: (i) it is impossible to combine such compliance with compliance with a mandatory provision in the law to which the nonEU/EEA AIFM and/or the non-EU/EEA AIF is subject; (ii) the law to which the non-EU/EEA AIFM and/or the non-EU/EEA AIF is subject provides for an equivalent rule having the same regulatory purpose and offering the same level of protection to the investors of the relevant AIF; and (iii) the non-EU/EEA AIFM and/or the non-EU/EEA AIF complies

with the equivalent rule referred to in point (ii).48 ESMA will develop draft regulatory technical standards on the conditions under which the law to which a non-EU/EEA AIFM or non-EU/EEA AIF is subject is considered to provide for an equivalent rule having the same regulatory purpose and offering the same level of protection to the relevant investors.49 We note that the ‘equivalence test’ pursuant to Article 37 AIFMD only relates to the applicability of the AIFMD and not to the authorization requirement. In any case, there is an authorization requirement for the non-EU/EEA AIFM, but the scope of the provisions of the AIMFD that apply to such a non-EU/EEA AIFM may differ, depending on the outcome of the ‘equivalence test’.

D. Will there ever be Equivalence Decisions under Article 47 of MiFIR? [10.27] Although the Commission has adopted equivalence decisions under EMIR and the CRA Regulation, there is no guarantee that the Commission will ever adopt an equivalence decision under Article 47 of MiFIR. The equivalence assessment for each of these regimes is different. Under the CRA Regulation, equivalence should be established with regard to the requirements resulting from the CRA Regulation, while under EMIR these are the requirements laid down in Title IV of EMIR. The equivalence assessment under EMIR and the CRA Regulation is rather limited, if you consider that for an equivalence decision to be taken under MiFIR, the third-country requirements should be equivalent to MiFID II/MiFIR and CRD IV/CRR. It could therefore take a considerable amount of time before the Commission will reach an equivalence decision under Article 47 of MiFIR, if indeed it ever comes to that point. [10.28] The Commission should initiate the equivalence assessment on its own initiative. Member States should be able to indicate their interest that a certain third country is or certain third countries are subject to the equivalence assessment carried out by the Commission, without such indications being binding on the Commission to initiate the equivalence process. When initiating those equivalence assessments, the Commission should be able to prioritize among third-country jurisdictions, taking into account: the materiality of the equivalence finding to EU/EEA firms and

clients, the existence of supervisory and cooperation agreements between the third country and the Member States, the existence of an effective equivalent system for the recognition of investment firms authorized under foreign regimes, and the interest and willingness of the third country to engage in the equivalence assessment process.50

3. Registration with ESMA [10.29] Following an equivalence decision in relation to a third country, a third-country firm that is authorized in that third country may provide investment services or perform investment activities with or without ancillary services to eligible counterparties and per se professional clients established throughout the EU/EEA where it is listed in the register of thirdcountry firms kept by ESMA. If cumulative conditions 1 to 3 below are fulfilled, ESMA will have to register the third-country firm.51 1. The Commission has adopted an equivalence decision in respect of the third country where the third-country firm is authorized. 2. The third-country firm is authorized in the jurisdiction where its head office is established to provide the investment services or activities to be provided in the EU/EEA and is subject to effective supervision and enforcement ensuring full compliance with the requirements applicable in that third country.52 3. ESMA has established a cooperation arrangement with the regulator(s) in the third country in respect of which the Commission has adopted an equivalence decision. A cooperation arrangement must specify at least: (i) the mechanism for the exchange of information between ESMA and the regulator(s) in the third country (including access to all information regarding the non-EU firms authorized in the third country that is requested by ESMA); (ii) the mechanism for prompt notification to ESMA where a third-country regulator deems that the third-country firm is infringing the conditions of its authorization or other law to which it is obliged to adhere; (iii) the procedures for coordination of supervisory activities (including on-site inspections).53

[10.30] In order for the applicant third-country firm to be registered with ESMA, it has to provide its contact details, national and international identification codes, proof of its authorization to provide investment services in its home country (through a written declaration issued by the competent authority of its home country), contact details of the competent authority of the home country, and the investment services to be provided and/or investment activities to be performed in the EU/EEA, together with any ancillary services.54 [10.31] ESMA has to assess within thirty working days whether the application is complete. If the application is not complete, the applicant third-country firm will have the possibility to provide additional information before a deadline set by ESMA. Within 180 working days of the submission of a complete application, ESMA will inform the applicant third-country firm whether the registration has been granted or refused. After a third-country firm has submitted a complete application, it can therefore take up to thirty-six weeks (about nine months) before the registration will be granted. To compare: an applicant for a MiFID II licence will be informed within six months of the submission of a complete application whether or not authorization has been granted.55 The question therefore arises of whether MiFID II/MiFIR’s third-country regime actually provides for effective market access.56 [10.32] Once a third-country firm has been registered by ESMA, it may provide investment services or perform investment activities to eligible counterparties and per se professional clients throughout the EU/EEA. The registration by ESMA hence replaces the authorizations from individual Member States and functions as a European passport.

A. Only on a Cross-Border Basis? [10.33] The question arises of whether a third-country firm that is registered with ESMA may conduct its activities only on a cross-border basis, thus without opening a branch; or also with a branch in one or more Member States. Article 46(1) MiFIR provides that a third-country firm ‘may’ operate within the EU/EEA without the establishment of a branch.

But ‘must’ it do so? Or does the rule contained in Article 46(1) MiFIR also permit a third-country firm to operate from a branch in the EU/EEA (rather than on a cross-border basis)? Our inclination would be to interpret Article 46(1) MiFIR as meaning that the regime it describes may be operated only on a cross-border basis and not from a branch in the EU/EEA. Article 47(3) MiFIR, however, seems to enable third-country firms to provide investment services or perform investment activities through a branch that has obtained an authorization under Article 39 MiFID II. Only in such a case can the competent authority of the home Member State of the branch exercise effective supervision. We will return to this subject in Section V.1.

4. Continuing Obligations for Registered ThirdCountry Firms [10.34] Third-country firms that are registered with ESMA no longer come within the scope of MiFID II/MiFIR and CRD IV/CRR.57 Member States may therefore not impose any additional requirements on such thirdcountry firms in respect of matters covered by MiFID II/MiFIR and CRD IV/CRR. This is a consequence of the equivalence decision by the Commission in which it has established that the legally binding prudential and business conduct requirements of the third country have equivalent effect to the requirements set out in MiFID II/MiFIR and CRD IV/CRR. This does not mean that a third-country firm is completely out of scope of any European legislation. A third-country firm that is registered with ESMA still has to comply with the following obligations: 1. Before providing any investment service, it has to inform its clients that are established within the EU/EEA: (a) that it is not authorized to provide services to clients other than eligible counterparties and per se professional clients, and (b) that it is not subject to supervision in the EU/EEA. The third-country firm must indicate the name and address of the competent authority or authorities responsible for supervision in the third country. This information must be provided in a durable medium and ‘in a prominent way’. The information has to be provided in English or in the official language of the Member State where the services are to be provided.58

2. Before services are provided or activities performed to an eligible counterparty or per se professional client established in the EU/EEA, the third-country firm must offer to submit any disputes relating to those services or activities to the jurisdiction of a court or arbitral tribunal in a Member State.59 3. The third-country firm has to inform ESMA, within thirty days, of any change of the information that has been provided in the application for registration with ESMA.60 [10.35] In addition, the third-country firm might have to comply with other European legislation, such as the prospectus requirement under the Prospectus Directive, if it issues securities to the public in the EU/EEA or such securities are admitted to listing in the EU/EEA, or the market abuse rules under MAD II/MAR.61

5. Transitional Regime [10.36] Third-country firms that already provide investment services or perform investment activities to eligible counterparties and per se professional clients established in the EU/EEA in accordance with national regimes will be able to continue to provide such services or perform such activities until three years after the adoption by the Commission of an equivalence decision in relation to the relevant third country.62

6. Withdrawal of Registration by ESMA/Equivalence Decision by the Commission [10.37] ESMA will withdraw the registration of a third-country firm if the following conditions are met. 1. ESMA has well-founded reasons based on documented evidence to believe that, in the provision of investment services and activities in the EU/EEA, the third-country firm (a) is acting in a manner which is clearly prejudicial to the interests of investors or the orderly

functioning of markets, or (b) has seriously infringed the provisions applicable to it in the third country and on the basis of which the Commission has adopted the equivalence decision;63 2. ESMA has referred the matter to the regulator in the third country and the regulator: (a) has not taken the appropriate measures needed to protect investors and the proper functioning of the markets in the EU/EEA, or (b) has failed to demonstrate that the third-country firm concerned complies with the requirements applicable to it in the third country; and 3. ESMA has informed the third country regulator of its intention to withdraw the registration of the third-country firm at least thirty days before the withdrawal.64 [10.38] If ESMA withdraws the registration of a third-country firm, ESMA must inform the Commission without delay and publish its decision on its website.65 [10.39] It should be noted that it is always the Commission (and not ESMA) that assesses whether the conditions under which an equivalence decision has been adopted ‘continue to persist in relation to the third country concerned’.66 Naturally, there is to some extent a link between the withdrawal of individual third-country firms from the register by ESMA and the withdrawal of an equivalence decision by the Commission, because if two or more third-country firms from a given third country are withdrawn from the register by ESMA, this would seem to indicate that something is wrong with the quality of the local regulatory oversight. And this may in turn be a reason for the Commission to carry out further investigation and, ultimately, to withdraw the equivalence decision. [10.40] If the Commission withdraws an equivalence decision in relation to a third country, any firm authorized in that third country may no longer use the rights under Article 46(1) MiFIR.67 Such third-country firms may then again only provide investment services or perform investment activities to eligible counterparties and per se professional clients that are established in the EU/EEA in accordance with the national regimes of the Member States.68

7. No Equivalence Decision by the Commission [10.41] In the absence of an equivalence decision in relation to a third country (because the Commission has not (yet) adopted such a decision or has actually withdrawn a decision69), a Member State may itself determine whether and, if so, on what conditions a third-country firm from that third country may provide investment services and/or perform investment activities (with or without ancillary services) to eligible counterparties and per se professional clients in its jurisdiction.70 In such cases a third-country firm could end up complying with differing national regimes. As an example, we discuss the national regimes of the Netherlands and the UK.71 [10.42] In the Netherlands the national regime for servicing eligible counterparties and per se professional clients varies depending on the state where the third-country firm is established. Third-country firms established in Australia, the United States, or Switzerland are exempt from the licence obligation for the provision of investment services and performance of investment activities under Article 10 of the Dutch Financial Supervision Act (Wft) Exemption Regulation (Vrijstellingsregeling Wft). Besides the exemption from the authorization requirement, these third-country firms are also partially exempt from the conduct of business supervision by the Netherlands Authority for the Financial Markets (Stichting Autoriteit Financiële Markten, AFM)72 and almost completely so from prudential supervision.73 Third-country firms established in a country, other than Australia, the United States, or Switzerland, that intend to provide investment services to eligible counterparties and per se professional clients on a cross-border basis or through a branch must apply for full authorization under Article 2:96 of the Dutch Act on Financial Supervision (Wet op het financieel toezicht, Wft). The conduct of business rules and prudential rules in the Wft apply in full. Where a third-country firm is authorized under Article 2:96 Wft it will not be able to benefit from any European passporting rights. Another option would be to establish a Dutch subsidiary that applies for full authorization under Article 2:96 Wft.74 [10.43] Under the current UK regime, third-country persons are subject to the same general prohibition as UK persons against providing investment services or performing investment activities in the UK without

authorization. A third-country person can provide investment services or perform investment activities to eligible counterparties and per se professional clients in the UK by establishing a UK subsidiary, which could apply to the Financial Conduct Authority (FCA) or Prudential Regulation Authority (PRA) for authorization, or through an authorized UK branch. The assessment for the authorization of such UK branch is primarily of a prudential nature (of the third-country firm and its home country’s prudential standards), and cooperation with the third-country jurisdiction in which the firm is authorized is an important factor. Where a third-country firm with a UK branch is authorized it will not be able to benefit from any European passporting rights. No exemption applies to third-country firms with their seat in designated third countries. The UK however does apply the ‘overseas persons’ exclusion.75 We refer to paragraph 10.70 for more information on the ‘overseas persons’ exclusion. [10.44] Although national regimes apply in the absence of an equivalence decision by the Commission, some provisions of MiFIR still apply. Also, in the absence of an equivalence decision, a third-country firm will have to inform its clients that are established within the EU/EEA, before providing any investment service: (a) that it is not authorized to provide services to clients other than eligible counterparties and per se professional clients, and (b) that it is not subject to supervision in the EU/EEA. The third-country firm must also indicate the name and address of the competent authority or authorities responsible for supervision in the third country and offer to submit any disputes relating to those services or activities to the jurisdiction of a court or arbitral tribunal in a Member State.76

IV. Retail Clients and Opt Up Professional Clients 1. General [10.45] MiFID II contains a Member State option regarding the position of a third-country firm wishing to provide investment services and/or perform investment activities (with or without ancillary services) in a

Member State to retail clients or to opt up professional clients. The Member State concerned may require such third-country firm to establish a branch in its territory.77 Where a Member State considers that the appropriate level of protection for its retail clients or opt up professional clients can be achieved by the establishment of a branch by the third-country firm, it is considered appropriate to introduce a minimum common regulatory framework at EU/EEA level with respect to the requirements applicable to those branches and in light of the principle that third-country firms should not be treated in a more favourable way than EU/EEA firms.78 This regulatory framework is laid down in Articles 39–43 MiFID II.

2. Exercise of Member State Option [10.46] If a Member State exercises the Member State option, MiFID II prescribes the following procedure. The branch must acquire a prior authorization from the competent regulator of such Member State.79 The third-country firm should submit its application for authorization to the regulator of the Member State where it intends to establish a branch.80 When making the application, the third-country firm must provide the regulator with certain information to ensure that it can adequately assess the application.81 [10.47] The regulator will grant the authorization if the cumulative conditions referred to below at 1–7 are fulfilled.82 This constitutes maximum harmonization: the Member States may not impose any additional requirements on the organization and operation of the branch in respect of the matters covered by MiFID II.83 Nor may a Member State treat any branch of third-country firms more favourably than firms from the EU/EEA.84 1. The provision of services for which the third-country firm requests authorization is subject to authorization and supervision in the third country where the firm is established and the requesting firm is properly authorized, whereby the competent authority pays due regard to any recommendations of the Financial Action Task Force (FATF) in

2.

3. 4. 5.

6. 7.

the context of anti-money laundering and countering the financing of terrorism.85 Cooperation arrangements, which include provisions regulating the exchange of information for the purpose of preserving the integrity of the market and protecting investors, are in place between the regulator in the Member State where the branch is to be established and the regulator of the third country where the firm is established.86 There is sufficient initial capital at the branch’s free disposal.87 One or more persons are appointed to be responsible for the management of the branch and they all comply with the governance requirements laid down in Article 88 and Article 91 CRD IV.88 The third country where the third-country firm is established has signed an agreement with the Member State where the branch is to be established, which: (a) fully complies with the standards laid down in Article 26 OECD Model Tax Convention on Income and on Capital, and (b) ensures an effective exchange of information in tax matters, including, if any, multilateral tax agreements.89 The firm belongs to an investor-compensation scheme authorized or recognized in accordance with the ICS Directive.90 The branch must (insofar as is relevant) be able to comply with various MiFID II provisions: (a) organizational requirements; (b) algorithmic trading; (c) trading process and finalization of transactions in an MTF and an OTF; (d) specific requirements for MTFs; (e) specific requirements for OTFs; (f) various conduct of business obligations (conflicts of interest; general duty to act honestly, fairly, and professionally; provision of adequate information; KYC rules; best execution; client order handling; transactions with eligible counterparties); (g) market transparency and integrity; (h) the transparency rules for trading venues; and (i) transaction reporting rules.91

[10.48] The Netherlands intends to exercise the discretion to apply the MiFID II regime of Article 39, since it could enhance investor protection. If a branch were established in the Netherlands, the regulators (AFM and Dutch Central Bank92) would be better placed to supervise compliance with the applicable rules by a third-country firm. The requirement of a branch

also means that the branch’s records and accounts would be better accessible to the regulators. Clients of the third-country firm would also have a contact point in the Netherlands to which they could apply if they have questions or complaints about the service. Moreover, it would be easier for the regulators to contact those persons responsible for the day-today running of the branch of the third-country firm in order to address matters relevant to supervision. In such a case, the prudential supervision exercised by DCB can also focus on the branch established in the Netherlands rather than on the third-country firm as a whole. If the Netherlands exercises the discretion to apply the MiFID II regime of Article 39 MiFID II, the current exemption for third-country firms established in Australia, the US, or Switzerland providing investment services to retail clients and opt up professional clients from the licence obligation as a thirdcountry firm under Article 10 of the Wft Exemption Regulation will be withdrawn.93

A. Withdrawal of Branch Authorization [10.49] The regulator may withdraw an authorization it has granted to a branch of a third-country firm in the event of one or more of the following situations: 1. The third-country firm: (a) does not use the authorization within twelve months, (b) expressly renounces the authorization, or (c) has provided no investment services or performed no investment activity for the preceding six months, unless the Member State concerned has provided for the authorization to lapse in the cases referred to at (a)–(c) (and withdrawal is therefore not necessary).94 2. The third-country firm has obtained the authorization by making false statements or by any other irregular means.95 3. The third-country firm no longer meets the conditions under which authorization was granted.96 4. The third-country firm has seriously and systematically infringed the provisions adopted pursuant to MiFID II governing the operating conditions for investment firms and applicable to third-country firms.97

5. The third-country firm falls within any of the cases where national law, in respect of matters outside the scope of MiFID II, provides for withdrawal.98

3. Non-Exercise of Member State Option [10.50] If a Member State does not exercise the Member State option, MiFID II merely provides one thing: the Member State may not require a third-country firm to establish a branch, if the firm wishes to provide investment services and/or perform investment activities (with or without ancillary services) in the Member State concerned to retail clients or opt up professional clients. After all, the harmonized requirements described in Section IV.2 apply in such a case. Whether and, if so, on what conditions a third-country firm may provide investment services and/or perform investment activities (with or without ancillary services) to retail clients or opt up professional clients in a Member State that has not exercised the Member State option depends on the national regime of the Member State concerned. [10.51] At the time of writing this chapter, the UK government was minded not to exercise the discretion to make use of the Member State option of Article 39 MiFID II. The reasons for not exercising this discretion were that the current regime has the virtue of being sufficiently tailored to client types and to the risks in question and balances the need to maintain investor protection, market integrity, and financial stability, while remaining open to business internationally.99

4. No European Passport! [10.52] Whether or not a Member State exercises the Member State option, an authorization does not constitute a European passport.100 If Member State A has exercised the Member State option and granted an authorization to a branch in its territory of third-country firm X, the authorization does not qualify as a European passport. In short, if thirdcountry firm X subsequently wishes to provide investment services and/or

perform investment activities (with or without ancillary services) in Member State B to retail clients and opt up professional clients, and Member State B has also exercised the Member State option, third-country firm X will also have to establish a branch in Member State B for which application must be submitted to the regulator in Member State B for separate authorization. However, the advantage of the MiFID II regime in comparison with the present, pre-MiFID II position is that the requirements to be imposed in respect of an authorization for a branch of a third-country firm have been harmonized. Consequently, the requirements to be met by third-country firm X in Member State B should not differ from those that apply in Member State A. It might be possible that Member State B will, in practice, rely on the assessment that has been made by Member State A. Third-country firm X will, however, be subject to the supervision of the competent authorities in both Member State A and Member State B.101 [10.53] As indicated in Section IV.3, if Member State B has not exercised the Member State option, the requirement that third-country firm X should also establish a branch in Member State B does not apply. Whether and, if so, on what conditions third-country firm X may provide investment services and/or perform investment activities (with or without ancillary services) in Member State B to retail clients and opt up professional clients will depend on the national regime of Member State B, although Member State B cannot demand the establishment of a branch since in such circumstances the harmonized requirements will apply (see Section IV.2). In comparison with the situation described in the previous paragraph, namely where Member State B has exercised the Member State option, this has the disadvantage that the requirements to be met by third-country firm X may differ between Member State B and Member State A, which is precisely the same as in the present pre-MiFID II position. [10.54] It remains to be seen which Member States will exercise the Member State option. It can in any event be inferred from the fact that this is a Member State option that during the negotiations on MiFID II no agreement could be reached on a regime in which a third-country firm which wishes to provide investment services and/or perform investment activities (with or without ancillary services) in a Member State to retail clients and opt up professional clients is by definition obliged to open a

branch in that Member State and apply for an authorization for the establishment of that branch.

5. Resolution of a Branch of a Third-Country Firm [10.55] The EU Bank Recovery and Resolution Directive (BRRD)102 lays down rules and procedures relating to the recovery and resolution of branches of third-country institutions (that is, credit institutions and investment firms within the meaning of Article 2(1)(2) and 2(1)(3) BRRD) —the so-called Union branches.103 Branches of such third-country institutions that are authorized pursuant to Article 39 MiFID II fall within the scope of the BRRD. Although a material discussion on the recovery and resolution of Union branches falls outside the scope of this chapter, we note the following. Article 96 BRRD requires that Member States ensure that resolution authorities have the powers necessary to act in relation to a Union branch that is not subject to any third-country resolution proceedings or that is subject to third-country resolution proceedings and one of the circumstances referred to in Article 95 applies (as a result of which recognition or enforcement of third-country resolution proceedings may be refused). Such powers, inter alia, include the exclusion of certain contractual terms in early intervention and resolution.104 Resolution authorities may exercise their powers where the resolution authority considers that action is necessary in the public interest and one or more of the following conditions are met: 1. The Union branch no longer meets, or is likely not to meet, the conditions imposed by national law for its authorization and operation within that Member State and there is no prospect that any private sector, supervisory, or relevant third-country action would restore the branch to compliance or prevent failure in a reasonable timeframe. 2. The third-country institution is, in the opinion of the resolution authority, unable or unwilling, or is likely to be unable, to pay its obligations to EU/EEA creditors, or obligations that have been created or booked through the branch, as they fall due and the resolution authority is satisfied that no third-country resolution proceedings or

insolvency proceedings have been or will be initiated in relation to that third-country institution in a reasonable timeframe. 3. The relevant third country authority has initiated third-country resolution proceedings in relation to the third-country institution, or has notified the resolution authority of its intention to initiate such a proceeding. [10.56] Where a resolution authority takes action in relation to a Union branch, it shall have regard to the resolution objectives and principles and the requirements relating to the application of the resolution tools.105 Union branches will have to contribute to the national resolution financing arrangements.106

V. Retail Clients, Professional Clients and Eligible Counterparties [10.57] Naturally, a third-country firm may wish to provide investment services and/or perform investment activities (with or without ancillary services) in the EU/EEA to: (a) retail clients, (b) professional clients (per se professional clients and/or opt up professional clients), and (c) eligible counterparties. Article 47(3) MiFIR provides for a—somewhat complicated —regime in that respect that can only be used by a third-country firm, if (i) the Commission has adopted an equivalence decision in relation to the third country where the third-country firm is established, and (ii) the Member State in which the third-country firm intends to be active has exercised the Member State option.

1. After the Adoption of an Equivalence Decision and Exercise of the Member State Option [10.58] If: (i) a third-country firm has an authorized branch in a Member State (this will necessarily be a Member State which has exercised the Member State option), and (ii) the Commission has adopted an equivalence decision in relation to the third country concerned, the third-country firm

may then provide investment services and perform investment activities covered by the branch authorization, in Member States other than that in which the branch is established,107 to eligible counterparties and per se professional clients, without the need to establish a branch in those other Member States.108 The branch remains subject to supervision by the regulator in the Member State in which it is established. Nonetheless, the regulator of the Member State in which the branch is established and the regulator of the host Member State may conclude ‘proportionate cooperation agreements’ to deliver the appropriate level of investor protection.109 [10.59] This is hardly surprising since, if the third-country firm wishes to provide services solely to eligible counterparties and per se professional clients within the EU/EEA, and it has registered itself with ESMA following an equivalence decision by the Commission in relation to the third country concerned, the third-country firm could do so even without establishing a branch. Nor should things be any different just because the third-country firm has to establish a branch in a Member State, because: (a) it provides services in that Member State to retail clients and opt up professional clients, and (b) that Member State has exercised the Member State option. [10.60] A potential difference might be that a third-country firm that makes use of Article 47(3) MiFIR does not have to register itself with ESMA to be able to provide investment services and perform investment activities to eligible counterparties and per se professional clients established throughout the EU/EEA, where this would be necessary if the third-country firm makes use of Article 46(1) MiFIR. However, the text of Article 47(3) MiFIR is not clear in this respect. It could be that a thirdcountry firm can choose. Either it can register itself with ESMA and provide services from the third country on a cross-border basis to eligible counterparties and per se professional clients within the EU/EEA (Article 46(1) MiFIR) or it can provide services to eligible counterparties and per se professional clients from an authorized branch in the EU/EEA on a crossborder basis without a registration with ESMA (Article 47(3) MiFIR).

[10.61] Naturally, everything would be different if the third-country firm also wishes to provide services in a Member State other than the one in which the branch is established to retail clients and opt up professional clients. No passport applies in that respect. If that other Member State has also exercised the Member State option, the third-country firm must also have an authorized branch there. If that other Member State has not exercised the Member State option, that Member State may not require the firm to establish a branch, but may impose other requirements under the national regime applicable there. We refer to Section IV.4.

A. Applicability of MiFID II/MiFIR Requirements [10.62] Can additional requirements be imposed on a third-country firm that provides investment services to eligible counterparties and per se professional clients through an authorized branch (that is, in accordance with Article 47(3) MiFIR)? In other words, does Article 46(3) MiFIR apply if a third-country firm provides investment services to eligible counterparties and per se professional clients throughout the EU/EEA from an authorized branch? If that is the case, the third-country firm would not be subject to MiFID II/MiFIR and CRD IV/CRR and the branch would not be subject to effective supervision in the EU/EEA (at least, insofar as investment services are provided to eligible counterparties and per se professional clients). This seems undesirable. As a result, it seems to be intended that a third-country firm may provide investment services to eligible counterparties and per se professional clients throughout the EU/EEA from an authorized branch only if the third-country firm is not registered with ESMA, since Article 46(3) MiFIR only applies where a third-country firm is registered with ESMA. We are therefore inclined to say that if a third-country firm provides investment services to eligible counterparties and per se professional clients throughout the EU/EEA from an authorized branch Article 41(2) MiFID II should be complied with.

2. If no Equivalence Decision is in Effect

[10.63] If no equivalence decision is in effect in relation to a third country (because the Commission has either not yet adopted such a decision or withdrawn a decision110), the European passport for the provision of investment services to eligible counterparties and per se professional clients in accordance with Article 47(3) MiFIR is not available for a third-country firm that has established an authorized branch in a Member State that has exercised the Member State option. In addition, it cannot register itself with ESMA. In such cases, a third-country firm is therefore once again put in the position of having to contend with the different national regimes for the provision of investment services to eligible counterparties and per se professional clients. See Section III.7 above. No passport would be available.

3. If the Member State Option has not been Exercised [10.64] If a Member State has not exercised the Member State option, the European passport for the provision of investment services to eligible counterparties and per se professional clients in accordance with Article 47(3) MiFIR is not available for a third-country firm, since this would require the establishment of an authorized branch in accordance with Article 39 MiFID II. If a Member State has not exercised the Member State option, it would not be possible to open an authorized branch in accordance with Article 39 MiFID II in such Member State. The provision of investment services to retail clients and opt up professional clients would in that case only be possible on a cross-border basis and would be regulated under the national regime of the Member State. Even if a Member State has not exercised the Member State option, it would still be possible for a thirdcountry firm to register with ESMA, if the Commission has taken an equivalence decision in relation to the third country in which the thirdcountry firm is established. Such registration functions as a European passport.

VI. Initiative Test

1. Initiative Test Under MiFIR [10.65] Article 46(5), third paragraph, MiFIR reads as follows: Member States shall ensure that where an eligible counterparty or [per se professional client] established or situated in the [EU/EEA] initiates at its own exclusive initiative the provision of an investment service or activity by a third-country firm, this Article does not apply [emphasis added] to the provision of that service or activity by the third-country firm to that person including a relationship specifically related to the provision of that service or activity. An initiative by such clients shall not entitle the third-country firm to market new categories of investment product or investment service to that individual.

[10.66] Article 46(5), third paragraph, MiFIR introduces an initiative test. The provision of investment services or activities to eligible counterparties and per se professional clients established or located in the EU/EEA is not subject to Article 46 MiFIR, if such eligible counterparty or per se professional client at its own exclusive initiative initiated the provision of an investment service or activity by the third-country firm. A client’s individual initiative does not entitle a third-country firm to market new categories of investment products or investment services to that individual.111 [10.67] The scope of the provision is not crystal clear. What is clear, however, is that if an eligible counterparty or a per se professional client established in the EU/EEA initiates at its own exclusive initiative the provision of an investment service or activity by a third-country firm and the Commission has adopted an equivalence decision in relation to that third country, there is no need to apply to ESMA for registration. But, does the initiative test also apply in relation to third-country firms in respect of which no equivalence decision is in effect (see Section III.7)? Purusant to Article 46(5), third paragraph, MiFIR ‘this Article does not apply’ if the initiative test is passed. However, Article 46 MiFIR only covers the situation where the Commission has adopted an equivalence decision in relation to a third country. Although Article 46(4), fifth paragraph, MiFIR admittedly provides that Member States may allow third-country firms to provide investment services or perform investment activities in their territories, if the Commission has not yet adopted an equivalence decision in relation to that third country or such a decision is no longer in effect, the

result of Article 46(5), third paragraph, MiFIR, if interpreted literally, is merely that this option for Member States ‘does not apply’ if the initiative test is passed. Undoubtedly it was not so intended. Although the literal text therefore creates an element of doubt, the initiative test should, in our opinion, be interpreted as meaning that the Member States must ensure that it is always available for third-country firms in relation to the provision of investment services or activities to eligible counterparties and per se professional clients established in the EU/EEA, regardless of whether the Commission has adopted an equivalence decision in relation to the third country concerned. Recital 43 MiFIR supports this interpretation as it refers to third-country firms generically. Another outcome would also be unfeasible, since the initiative test would be especially important for the third-country firms that are established in third countries for which the Commission has not adopted an equivalence decision yet. Another question would be whether the provisions of MiFID II/MiFIR, other than in respect to market access, apply if a third-country firm provides investment services based on the initiative test. This seems not to be the case, since Recital 43 MiFIR states that services that are provided on the exclusive initiative of a person should not be deemed as provided in the territory of the EU/EEA.112

2. Initiative Test Under MiFID II [10.68] Article 42 MiFID II reads as follows: Member States shall ensure that where a retail client or [opt up professional client] established or situated in the [EU/EEA] initiates at its own exclusive initiative the provision of an investment service or activity by a third-country firm, the requirement for authorization under Article 39 shall not apply [emphasis added] to the provision of that service or activity by the third-country firm to that person including a relationship specifically relating to the provision of that service or activity. An initiative by such clients shall not entitle the third-country firm to market otherwise than through the branch, where one is required in accordance with national law, new categories of investment products or investment services to that client.

[10.69] This provision also introduces an initiative test. The provision of investment services or activities to retail clients and opt up professional clients established or situated in the EU/EEA is not subject to the

requirement for authorization under Article 39 MiFID II, if such retail client or opt up professional client at its own exclusive initiative initiated the provision of an investment service or activity by the third-country firm. As is the case with Article 46(5), third paragraph, MiFIR, an individual initiative by a client does not in itself give a third-country firm the right to market new categories of investment products or investment services to the person concerned.113 The scope of the provision is once again not crystal clear. As reference is made to ‘the requirement for authorization under Article 39’, the provision, if interpreted literally, applies only to Member States which have exercised the Member State option and made it obligatory for a third-country firm to establish a branch if it wishes to provide investment services and/or perform investment activities (with or without ancillary services) in the Member State concerned to retail clients and to opt up professional clients. By this reasoning, Member States which have not exercised the Member State option would not have to apply the initiative test introduced in Article 42 MiFID II. However, they could introduce their own initiative test under a national regime. This strikes us as a less than desirable interpretation of Article 42 MiFID II. It would therefore benefit legal certainty if the initiative test introduced by Article 42 MiFID were to be applicable irrespective of whether a Member State has exercised the Member State option. Recital 111 of MiFID II supports this interpretation as it refers to third-country firms generically. [10.70] Some Member States might, however, think differently. One of the considerations for the UK government not to exercise the discretion to apply the regime of Article 39 MiFID II is that the UK’s ‘overseas persons’ exclusions would be substituted with the narrower concept of reverse solicitation for retail and opt up professional clients. The exclusion for ‘overseas persons’ includes exclusions for particular investment services and activities carried on in the context of a ‘legitimate approach’ or carried on ‘with or through’ an authorized or exempt UK person. The concept of ‘legitimate approach’ is linked to the UK restriction on making financial promotions. It allows cross-border business to be carried out on the basis that the overseas person has not solicited in any way the particular service to be provided to the UK client (which is similar to the MiFID II and MiFIR concept of ‘own exclusive initiative’), and also permits certain limited forms of client solicitation along the lines of exemptions in the FSMA

Financial Promotions Order 2005. These exemptions include, for example, certain types of solicitations made to certified high net worth individuals and previously overseas customers—both of which might be categorized as retail clients under MiFID II.114 [10.71] Lastly, we note that it seems that EU/EEA investment firms cannot rely on the initiative test when providing investment services to clients in other EU/EEA Member States, taking into account that Article 42 MiFID II only refers to third-country firms.

3. Own Exclusive Initiative [10.72] For the initiative test under both MiFID II and MiFIR, it should be determined whether the investment services are provided at ‘the own exclusive initiative’ of the client that is established or situated in the EU/EEA. The recitals of MiFID II and MiFIR give some further guidance in that respect. [10.73] It can be derived from Recital 111 MiFID II that: 115 [w]here a third-country firm provides services at the own exclusive initiative of a person established in the Union, the services should not be deemed as provided in the territory of the Union. Where a third-country firm solicits clients or potential clients in the Union or promotes or advertises investment services or activities together with ancillary services in the Union, it should not be deemed as a service provided at the own exclusive initiative of the client.

[10.74] Recital 43 MiFIR contains similar wording, although Recital 43 MiFIR clarifies that the provisions of MiFIR regulating the provision of services or undertaking of activities by third-country firms should not affect the possibility for EU/EEA investment firms or credit institutions to receive investment services or activities from a third-country firm at their own exclusive initiative or for a client to receive investment services from a third-country firm at their own exclusive initiative through the mediation of such a credit institution or investment firm.

[10.75] As a result, where a third-country firm solicits clients or potential clients in the EU/EEA or promotes or advertises investment services or activities together with ancillary services in the EU/EEA, it should not be deemed as a service provided at the own exclusive initiative of the client, as a result of which the third-country firm cannot rely on the initiative test of Article 42 MiFID II or Article 46(5), third paragraph, MiFIR.

VII. Conclusion [10.76] MiFID II/MiFIR introduce a common regulatory framework that should harmonize the existing fragmented framework for the provision of services by third-country firms, ensure certainty and uniform treatment of third-country firms accessing the EU/EEA, ensure an assessment of effective equivalence by the Commission, and provide for a comparable level of protection to clients in the EU/EEA receiving services by thirdcountry firms. But as we have seen, the legislation is unfortunately complex and does not provide for full harmonization. [10.77] Although the basics of the third-country regime under MiFID II/MiFIR seem to be set, the devil is in the detail. There is still a considerable lack of clarity in relation to the scope of the MiFID II/MiFIR’s third-country regime, for example where it concerns investment activities, and activities other than investment services and investment activities (such as selling and advising in relation to structured deposits), and the initiative test. In addition, the concurrence between MiFID II and MiFIR, if a thirdcountry firm provides investment services to eligible counterparties, professional clients, and retail clients, raises interesting questions that are not easily answered. [10.78] Table 10.1 provides a summary that may assist in finding a way through the provisions of the third-country regime under MiFID II/MiFIR.

Table 10.1: Provisions of the third-country regime under MiFID II/MiFIR

Retail clients and opt up professional clients

Eligible counterparties and per se professional clients

Provision of services through a branch

Member State has exercised the Member State option (Article 39(2) MiFID II) This is possible, if the branch has obtained authorization from the regulator of the Member State in which the branch is established. Member State has not exercised the Member State option We deem this not to be possible.

Commission has taken an equivalence decision (Article 47(3) MiFIR) We deem this to be possible, if the thirdcountry firm has an authorized branch in accordance with Article 39 MiFID II. Commission has not taken an equivalence decision (Article 46(4) MiFIR) This is only possible if this is in accordance with the applicable national regime of a Member State.

Provision of crossborder services

Member State has exercised the Member State option (Article 39(1) MiFID II) This is not possible. Member State has not exercised the Member State option This is only possible if this is in accordance with the applicable national regime of a Member State.

Commission has taken an equivalence decision (Article 46(1) MiFIR)This is possible, if the third-country firm is registered with ESMA. Commission has not taken an equivalence decision (Article 46(4) MiFIR) This is only possible if this is in accordance with the applicable national regime of a Member State.

Provision of services through the

This is possible (Article 42 MiFID II).

This is possible (Article 46(5), third paragraph, MiFIR).

initiative test Provision of services throughout the EU/EEA (European Passport)

1

Member State has exercised the Member State option (Article 39(1) MiFID II) This is not possible, unless the third-country firm establishes an EU/EEA subsidiary that obtains an authorisation that can subsequently be passported. Member State has not exercised the Member State option (Articles 34 and 35 MiFID II) This is not possible, unless the third-country firm establishes an EU/EEA subsidiary that obtains an authorisation that can subsequently be passported.

Commission has taken an equivalence decision (Article 46(1) MiFIR / Article 47(3) MiFIR) This is possible, if the third-country firm (i) is registered with ESMA or (ii) has an authorized branch in accordance with Article 39 MiFID II and has fulfilled the applicable information requirements under Article 34 MiFID II. Commission has not taken an equivalence decision (Articles 34 and 35 MiFID II) This is not possible, unless the thirdcountry firm establishes an EU/EEA subsidiary that obtains an authorisation that can subsequently be passported.

Cf. Recital 41 MiFIR, which talks of highly differentiated and fragmented regimes, and Commission Staff Working Paper accompanying the document Proposal for a Directive of the European Parliament and of the Council on Markets in financial instruments (Recast) and the Proposal for a Regulation of the European Parliament and of the Council on Markets in financial instruments, 20.10.2011, SEC (2011) 1226, final (Impact Assessment MiFID II), p. 114. 2 Cf. Recital 28 MiFID. However, the third-country firm can of course decide to establish a subsidiary in the EU/EEA, in which case the subsidiary can make use of the European passport. 3 Cf. Impact Assessment MiFID II (n. 1), p. 15. 4 Cf. Recital 41 MiFIR.

5

Article 1(1) MiFID II and Article 1(1)(f) MiFIR. Article 4(1)(57) MiFID II and Article 2(1)(42) MiFIR. 7 Article 46(1) MiFIR and Article 39(1) MiFID II. 8 In accordance with Section II of Annex II to MiFID II. 9 In accordance with Section II of Annex II to MiFID II. 10 We note that if such investment activity falls outside the scope of MiFID II/MiFIR, the third-country firm could perform such investment activity without complying with the MiFID II/MiFIR’s third-country firm regime at all. It can however be read in Chapter 2 that the scope of the exemptions for dealing on own account has been narrowed by MiFID II. 11 This comment equally applies for underwriting activities. Also in that case, no typical client relationship exists between a third-country firm and an investor, since the underwriting activities would typically be performed for an issuer. 12 In the Netherlands, the Netherlands Authority for the Financial Markets (Stichting Autoriteit Financiële Markten, AFM) has published an interpretation on the question of when activities of trading platforms (that is, regulated markets and MTFs) established abroad qualify as holding a market in financial instruments in the Netherlands (Interpretatie Beheren of exploiteren v.e. gereglementeerde markt of het exploiteren v.e. multilaterale handelsfaciliteit in Nederland, 15 March 2007). To answer this question the AFM assesses if the trading platform is active in the Netherlands. Previously, the AFM qualified the activities in the Netherlands of a foreign trading platform by using the criterion ‘the placing of trading screens’ by this trading platform. Due to the increased role of the Internet in the access to a trading platform, this criterion was deemed to be no longer appropriate. The AFM has therefore developed new assessment criteria. The AFM qualifies a trading platform, established abroad, as being active in the Netherlands if both of the following criteria are met: (i) the trading platform offers direct access (without making use of a local broker) to Dutch parties (connectivity test), and (ii) the trading platform (plans to) actively approach and market its activities to Dutch professional parties (initiative test). If only the connectivity test provides a positive outcome the following three factors could indicate, according to the AFM, that the foreign-based trading platform is active in the Netherlands: (i) the (relative) number of Dutch professional parties active on the platform; (ii) the relative share of Dutch professional parties on the total volume of the platform; and (iii) the impact of the platform on the Dutch capital markets. It is to be seen whether this interpretation will continue to be in place under MiFID II/MiFIR. 13 Article 1(4) MiFID II. See also Chapter 2. 14 You could say that third-country firms are investment firms and therefore Article 1(4) MiFID II applies to third-country firms. This seems, however, not to be in line with the systematics of MiFID II that includes a specific definition of third-country firms. 15 Article 46(1) MiFIR and Article 39(1) MiFID II. We also refer to Article 1(2)(b) MiFID II and Article 1(1)(f) MiFIR. 16 If structured deposits qualify as PRIIPs (Packaged Retail and Insurance-based Investment Products), the third-country firm should however act in accordance with the Regulation (EU) No 1286/2014 of the European Parliament and of the Council of 26 6

November 2014, OJ L 352, pp. 1–23 (PRIIPs Regulation), when acting as a PRIIP manufacturer or person selling a PRIIP, irrespective of whether it falls under MiFID II/MiFIR’s regime for third-country firms (Article 2 PRIIPs Regulation). 17 Article 42 MiFID II and Article 46(5), third paragraph, MiFIR. 18 Article 1(1) MiFID II. See also Chapter 2. 19 Article 59 MiFID II. 20 Article 4(1)(55) MiFID II. 21 Article 59(2) MiFID II. 22 Cf. Article 60(2) MiFID II. 23 Article 4(1)(24) MiFID II. See Article 4(1)(21), (22), and (23) MiFID II for the definitions of regulated market, MTF, and OTF. 24 Article 44 MiFID II. 25 See Article 38 MiFIR. 26 Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014, OJ L 173/179 and Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16 April 2014, OJ L 173/1. 27 Article 38(1) MiFIR. 28 Article 38(1) MiFIR. The equivalence decision should be taken in accordance with Article 38(3) MiFIR. 29 Article 38(2) MiFIR. The equivalence decision should be taken in accordance with Article 38(3) MiFIR. 30 Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012, OJ L 201, pp. 1–59. 31 Articles 28(1) and (4) MiFIR. We also refer to Article 23(1) MiFIR that allows investment firms to comply with the trading obligation by trading on a third-country trading venue that is assessed equivalent in accordance with Article 25(4)(a) MiFID II. 32 Strictly speaking Article 47(1) MiFIR does not refer to CRR. It is however our understanding that Article 47(1) MiFIR should be read as if it includes a reference to CRR. 33 We also refer to Recital 41 MiFIR. 34 Impact Assessment MiFID II (n. 1), p. 300. 35 ibid, pp. 304, 310. 36 Article 37(1) of the Commission Proposal. We also refer to Recital 41 MiFIR. 37 Regulation (EC) No 1060/2009 of the European Parliament and of the Council of 16 September 2009, OJ L 302, pp. 1–31, as amended. 38 Directive 2011/61/EU of the European Parliament and of the Council of 8 June 2011, OJ L 174/1. 39 Cf. N. Moloney, EU Securities and Financial Markets Regulation (Oxford: Oxford University Press, 2014), p. 403. 40 Cf. Article 25(1) EMIR. 41 Cf. Article 25(6) EMIR.

42

The Commission has determined that ten countries (Australia, Canada, Hong Kong, Japan, Mexico, Singapore, South Africa, South Korea, the US, and Switzerland) have equivalent regulatory regimes for CCPs to the EU/EEA. 43 The Commission has determined that nine countries (Argentina, Australia, Brazil, Canada, Hong Kong, Japan, Mexico, Singapore, and the US) have equivalent regulatory regimes for CRAs to the EU/EEA. 44 Article 21(5) AIFMD in conjunction with Article 21(6) AIFMD. 45 Article 21(6), last paragraph, AIFMD. 46 See also ESMA/2016/1140, Advice – ESMA's advice to the European Parliament, the Council and the Commission on the application of the AIFMD passport to non-EU AIFMs and AIFs (18 July 2016), para. 22. 47 Article 37 is currently not yet applicable in the Member States. This requires a delegated act from the Commission, which will be taken following positive advice from ESMA in accordance with Article 67 AIFMD. At the time of writing this chapter, ESMA has advised the Commission on the application of the passport to non-EU/EEA AIFMs and AIFs in twelve jurisdictions (ESMA (n. 46)). 48 Article 37(2) AIFMD. 49 Article 37(23)(b) AIFMD. 50 Recital 41 MiFIR. 51 Articles 46(1) and (2), opening words, in conjunction with Article 47 MiFIR. As regards the procedure to be followed by a third-country firm with ESMA, see Article 46(4), first and third paragraphs, MiFIR. See also Commission Delegated Regulation (EU) …/... of 14.7.2016 supplementing Regulation (EU) No 600/2014 of the European Parliament and of the Council with regard to regulatory technical standards concerning the information for registration of third country firms and the format of information to be provided to the clients (C(2016)4407 final) for the information that is necessary for the registration At the time of writing this chapter, this delegated act had not yet entered into force. 52 Article 46(2)(b) MiFIR. 53 Article 46(2)(c) in conjunction with Article 47(2) MiFIR. 54 Commission Delegated Regulation (n. 51), Articles 1 and 2. 55 Article 7(3) MiFID II. 56 The relevance of this question also relates to the element of reciprocity. If a third country assesses that third-country firms established in its territory do not have effective market access to the EU/EEA, it may decide not to grant effective market access to EU/EEA investment firms willing to access the market in such third country. 57 Article 46(3) MiFIR. The scope of MiFID II also determines the applicability of the requirements on the basis of CRD IV and CRR. We refer to Chapter 2. 58 Article 46(5), first and second paragraphs, MiFIR. Commission Delegated Regulation (n. 51), Article 3. 59 Article 46(6) MiFIR. 60 Commission Delegated Regulation (n. 51), Article 2(1).

61

Although the equivalence test by the Commission also includes whether the prudential and business conduct framework of the third country ensures market transparency and integrity by preventing market abuse in the form of insider dealing and market manipulation, we would expect that the market abuse rules under MAD II/MAR also apply to third-country firms that are registered with ESMA. 62 Article 54(1) MiFIR. 63 Articles 49(1)(a) and (b) MiFIR. 64 Articles 49(1)(c) and (d) MiFIR. 65 Article 49(2) MiFIR. 66 Article 49(3) MiFIR. 67 Article 47(4) MiFIR. 68 Article 46(4), last paragraph, MiFIR. 69 Article 49(3) MiFIR. 70 Article 46(4), last paragraph, MiFIR. 71 Following a Brexit (see Chapter 1, paragraph 1.56), the UK may qualify as a third country itself, depending on the arrangements be agreed upon between the EU and the UK. In such a case, any investment firm that has its head office or registered office in the UK, will qualify as a third-country firm and will only have access to the EU/EEA under the MiFID II/MiFIR’s third-country regime. As it is foreseen that the process of Brexit will take at least two years, MiFID II/MiFIR will already have entered into force in the UK prior to Brexit. After Brexit, the UK could, however, decide to no longer comply with MIFID II/MiFIR, again depending on the agreements established between the EU and the UK. The contents of such agreements are currently not yet known. 72 Article 35 Wft Exemption Regulation currently provides for a partial exemption from the conduct-of-business supervision. It is, however, intended that third-country firms that are exempt from the licence obligation under Article 10 of the Wft Exemption Regulation will be fully exempted from the conduct-of-business supervision after entry into force of MIFID II/MiFIR (Explanatory Notes to the Netherlands Draft Legislative Proposal for Implementation of MiFID II, p. 12). 73 With the exception of Articles 3:5, 3:6, and 3:7 Wft, see Article 18 Wft Exemption Regulation. 74 In the Netherlands the same national regime applies to the provision of investment services to retail clients and opt up professional clients, though there is then no exemption from the prudential requirements and the conduct of business rules are more extensive. 75 HM Treasury, Transposition of the Markets in Financial Instruments Directive II, March 2015, pp. 9–10. The same national regime applies for the provision of investment services to retail clients and opt up professional clients. 76 Article 46(6) MiFIR. 77 Article 39(1) MiFID II. 78 Recital 109 MiFID II. 79 Article 39(2), opening words, MiFID II.

80

Article 39(3) MiFID II. See Article 40 (Obligation to provide information) MiFID II. 82 Article 39(2), opening words, and Article 41(1), opening words, MiFID II. 83 Article 41(2), second paragraph, first part of the sentence, MiFID II. 84 Article 41(2), second paragraph, second part of the sentence, MiFID II. 85 Article 39(2)(a) MiFID II. 86 Article 39(2)(b) MiFID II. 87 Article 39(2)(c) MiFID II. 88 Article 39(2)(d) MiFID II. 89 Article 39(2)(e) MiFID II. 90 Article 39(2)(f) MiFID II; Directive 97/9/EC of the European Parliament and of the Council of 3 March 1997, OJ L 84/22. 91 Article 41(1)(b) in conjunction with (2), first paragraph, MiFID II. 92 De Nederlandsche Bank NV, DCB. 93 Explanatory notes to the Dutch Draft Legislative Proposal for Implementation of MiFID II, p. 11. 94 Article 43(a) MiFID II. 95 Article 43(b) MiFID II. 96 Article 43(c) MiFID II. 97 Article 43(d) MiFID II. 98 Article 43(e) MiFID II. 99 HM Treasury, Transposition of the Markets in Financial Instruments Directive II, March 2015, p. 11. 100 In the interests of completeness, we note that in the Commission Proposal for MiFID II (Proposal for a Directive of the European Parliament and of the Council on markets in financial instruments repealing Directive 2004/39/EC of the European Parliament and of the Council (Recast), 20.10.2011, COM/2011/0656 final) it was envisaged that the thirdcountry firm should be able to provide services in other Member States through the authorized and supervised branch, subject to a notification procedure (Recital 73 Commission Proposal MiFID II, Article 44 Commission Proposal MiFID II). This possibility for a European passport was however repealed under MiFID II. The only option under MiFID II for third-country firms wishing to obtain a European passport for the provision of investment services to retail client and opt up professional clients is to establish an EU/EEA subsidiary. 101 Cf. Article 41(2) MiFID II. 102 Directive 2014/59/EU of the European Parliament and of the Council of 15 May 2014, OJ L 173, pp. 190–348. 103 Articles 1(1)(e) and 2(1)(89) BRRD. 104 Article 96(1), second paragraph, BRRD. 105 Article 96(3) BRRD. 81

106

Article 103(1) BRRD. If a third-country firm can operate from a branch within the EU/EEA on a crossborder basis throughout the EU/EEA, we see no reason why it should be prevented from providing services from the branch to eligible counterparties and per se professional clients in the Member State where the branch itself is established. 108 See Article 47(3), first paragraph, MiFIR. However, the third-country firm must fulfil certain information requirements in relation to the regulator in the Member State where the branch is established, for example by (a) indicating in which other Member States it wishes to perform its activities, and (b) submitting a programme of operations. See Article 47(3), first paragraph, last sentence, MiFIR in conjunction with Article 34 MiFID II. 109 Article 47(3), second paragraph, MiFIR. 110 Article 49(3) MiFIR. 111 Article 46(5), third paragraph, MiFIR. 112 In the Netherlands, this interpretation seems to be followed (see Section 1:19c of the Netherlands Draft Legislative Proposal for Implementation of MiFID II). 113 Article 42, last sentence, MiFID II. 114 HM Treasury, Transposition of the Markets in Financial Instruments Directive II, March 2015, pp. 9–11. 115 Besides Recital 111 MiFID II, Recital 85 MiFID II sets out that: 107

a service should be considered to be provided at the initiative of a client unless the client demands it in response to a personalized communication from or on behalf of the firm to that particular client, which contains an invitation or is intended to influence the client in respect of a specific financial instrument or specific transaction. A service can be considered to be provided at the initiative of the client notwithstanding that the client demands it on the basis of any communication containing a promotion or offer of financial instruments made by any means that by its very nature is general and addressed to the public or a larger group or category of clients or potential clients. This recital does however not relate to the initiative test, but clarifies one of the conditions for not carrying out the suitability test, that is, that the service is provided at the initiative of the client or potential client (Communication from the Commission to the European Parliament pursuant to the second subparagraph of Article 251 (2) of the EC Treaty concerning the common position of the Council on the adoption of a Directive of the European Parliament and of the Council on markets in financial instruments, markets, amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC, 12.1.2004, COM (2004) 15, p. 16).

PART III

TRADING

11 GOVERNANCE AND ORGANIZATION OF TRADING VENUES The Role of Financial Market Infrastructure Groups Guido Ferrarini and Paolo Saguato

I. Introduction II. 1. 2. 3. 4. 5. III. 1. 2. IV. 1. 2. 3. 4. 5.

Trading Facilities: Concept and Regulatory Framework An Introduction to the Scenario of Trading Facilities The Regulatory Framework of EU Trading Venues Regulated Markets MTFs and OTFs Non-Discriminatory Access to Trading Venues and to CCPs The EU Scenario of Trading Venues The Evolution of the Trading Venues Market A Case Study of Six FMI Groups FMI Groups in an MiFID II World Formation of Pan-European Groups The OTF in FMI Groups Challenges of FMI Groups to Regulators MiFID II’s Approach to FMI Groups Consolidated Supervision?

V. Conclusion

I. Introduction [11.01] Ten years after its adoption and six years after the 2008 global financial crisis, the Markets in Financial Instruments Directive (MiFID I)1 underwent significant revisions. The reform spirit embraced by European institutions resulted in the adoption of two pieces of Level 1 legislation, a ‘revised’ Markets in Financial Instruments Directive (MiFID II)2 and a Markets in Financial Instruments Regulation (MiFIR),3 which intend to provide an updated regulatory framework for European financial markets so they will be up to the challenges posed by financial and technological innovation as well as the financial crisis.4 Market structures and market infrastructures are two of the areas touched by MiFID II.5 Faced with two compelling needs—to strengthen transparency and resilience in the over-the-counter (OTC) markets,6 and to increase efficiency and competition in the always evolving financial markets— European policymakers were induced to reconsider and reform the panorama of trading venues and the organizational environment in which they operate. [11.02] This chapter looks into the reformed regulatory framework for trading venues under MiFID II and how the trading venues scenario has evolved over the past few years. It is structured as follows. Section II introduces the concept of trading facilities and trading venues and provides an overview of the reforms brought about by MiFID II in the trading venues scenario. Regulated markets (RMs), multilateral trading facilities (MTFs), and organized trading facilities (OTFs) are the designated trading venues in the EU financial markets. Section III presents six case studies: the London Stock Exchange Group, the Deutsche Börse group, the Euronext group, the Bolsas y Mercados Españoles group, the NASDAQ OMX Nordic Exchange, and the BATS group. Each case study looks at the organizational structure of the trading venue, its governance, and the coexistence and relationship—within the same group—of trading venues and central clearing counterparties (CCPs). Section IV then critically and systemically evaluates the role of MiFID II in reorganizing the framework of trading venues and considers whether MiFID II’s regulatory and supervisory tools

are effective in tackling the emergence of Financial Market Infrastructure (FMI) groups. Finally, Section V concludes.

II. Trading Facilities: Concept and Regulatory Framework 1. An Introduction to the Scenario of Trading Facilities [11.03] A trading facility is the place where buyers and sellers meet to trade financial instruments (either securities or derivatives).7 The concept of a trading facility has become more fluid in recent decades due in part to the need to create marketplaces suitable for market participants’ evolving needs and in response to the impact that technological innovation is now playing in re-shaping the financial system.8 [11.04] Generally, trading facilities group together different categories of platforms providing trading services.9 Looking at two main features in the structure of trading facilities—first whether trading is bilateral or multilateral; and second whether the facility is regulated or unregulated— we can pinpoint three groups of facilities: trading venues, systematic internalizers, and alternative finance platforms.10 [11.05] A trading venue means an RM, an MTF, or an OTF;11 it is a multilateral regulated facility that provides a location for brokers and traders to meet and execute their trades, thus reducing traders’ searching costs for potential counterparties.12 In addition, a trading venue publicizes the related information concerning the prices, volumes, and often times the parties of the executed transactions, thus providing trade transparency; plus it offers quotes, namely the prices at which market participants are willing to sell (‘ask quotes’) and to buy (‘bid quotes’) financial instruments.13 These are the core activities that characterize any trading venue. Only some authorized trading venue operators provide additional services, such as listing of securities, the regulation and monitoring of issuers and broker-

dealers, and market supervision.14 Moreover, post-trade services are provided either directly by firms within the venue’s group or by institutions linked to them, such as clearing and settlement agents and CCPs, which facilitate the clearing and settlement of the executed transactions.15 [11.06] Systematic internalizers are regulated trading facilities operated by an investment firm that, ‘on an organized, frequent and substantial basis, offers bilateral trading by dealing on its own account (i.e. being a contractual part of the deal)’.16 [11.07] Moreover, technological innovation and regulatory arbitrage have contributed to the development of a vast array of alternative trading (electronic) platforms and systems operated by investment firms in a framework of bilaterally concluded financial transactions that fuel the OTC markets.17 These platforms are also found in primary markets, where they offer the facilities for crowdfunding and other alternative finance transactions (such as invoice trading).18 [11.08] Generally speaking, trading facilities have given rise to two main types of markets: public markets and private markets. Public markets are formal (being subject to a regulatory framework), multilateral (operating as an open marketplace where buyers’ and sellers’ trading interests meet), non-discretional (as they execute trades applying pre-set rules or parameters), and transparent (being subject to pre- and post-trade transparency). Conversely, private markets are informal (not being subject to a regulatory framework), bilateral (in that the venue enters into a trade with a client), discretionary (leaving the venue the authority to admit to the platforms and to the trades), and dark (not being subject to transparency requirements).19 Despite all being trading venues, RMs, MTFs, and OTFs create different markets. [11.09] RMs, traditional exchanges, and multilateral platforms more generally (such as MTFs), are usually associated with the first type of markets: public markets. [11.10] The second type of market, the private market, is generally linked to alternative trading (electronic) platforms and OTC markets, where

investment firms provide discretionary and bilateral (and in some instances multilateral) trading services, and also ‘dark pools’. This clear classification, however, has become more blurred due to the creation either by market actors or by regulatory intervention of a range of trading facilities that consequently contributed to the evolution of a spectrum of hybrid markets, which have evolved from the tuning at different levels of transparency regime, access rules, regulatory oversight, organizational aspects, etc.20 [11.11] Having at the antipodes public markets and private markets, hybrid markets are mixed organizational models of trading facilities which combine, to varying degrees, elements of both public and private markets. OTFs, for instance, because of the peculiarities of their architecture, create semi-public markets. OTFs are multilateral and formal venues, subject to pre- and post-trade transparency, but also subject to discretion in the execution of trades and therefore owing certain conduct of business duties to their clients including the ‘best execution’ requirement.21 Similarly SIs, because of their bilateral structure, lie in the private market sphere, although the increased regulation of their transparency regime shifted them away from the traditional private market OTC paradigm to a semi-private sphere. [11.12] The governance of trading venues has evolved substantially following the technological and competitive developments of globalization.22 First, exchanges, run in the form of mutual enterprises owned by their members, were generally demutualized, while public exchanges—trading venues open to all interested parties and run by a public entity governed by administrative law, and often enjoying broad selfregulatory powers—were mainly privatized.23 The resulting firms became public companies with diffuse shareholders or were merged into or acquired by other exchanges.24 Second, exchanges, mainly focused on the development of electronic platforms and the offer of liquidity services, became more similar to their competitors, such as MTFs and internalizing firms. They correspondingly became less (or poorly) incentivized to perform self-regulatory activities, to the extent that this generated positive externalities to competing platforms, and powerful conflicts of interest were created between exchanges’ regulatory and business activities.25

[11.13] Regulators highlighted these conflicts as problematic and as sufficient to justify regulatory reform.26 However, they also took the opportunity to expand their regulatory turf and to undertake some of the traditional rulemaking and monitoring functions of exchanges. In 2000, for example, the UK Listing Authority was transferred from the demutualized London Stock Exchange to the Financial Services Authority (FSA), now Financial Conduct Authority (FCA).27 Leaving the Listing Authority to a for-profit firm like the LSE – competing in the international market for trading services – would have raised concerns as to its independence and incentives to efficiently perform the relevant function. [11.14] More generally, two types of re-regulation followed the privatization and demutualization of exchanges. First, the governance of trading venue operators has generally been regulated to reflect the policy view that exchanges and other trading venues are firms offering transaction services to intermediaries and investors in a competitive setting.28 Both in the US and the EU, the various types of trading venues (including alternative trading systems and MTFs) are subject to rules that are increasingly similar across the sector given that these venues perform similar functions.29 Second, the role of exchanges in the regulation and supervision of listed issuers has been reduced and replaced by public regulation and supervision, which has correspondingly widened its scope of application.30

2. The Regulatory Framework of EU Trading Venues [11.15] The application of MiFID I31 to the EU’s equity trading markets in November 2007 heralded a new era for the EU’s financial markets.32 MiFID I’s securities trading rules were designed to reshape the EU trading market. The abolition of national ‘concentration rules’ in particular, while contributing to the fragmentation of securities markets, enhanced competition between trading venues in the EU, harnessing industry innovation and technological advances. Moreover, the financial crisis experience ‘largely vindicated’ MiFID I’s design.33 However, the European Commission highlighted four difficulties in the post-MiFID I scenario. (i)

The benefits of competition were not flowing efficiently to all market participants and had not always been passed on to end users, and market fragmentation had made the trading environment more complex and opaque. (ii) MiFID’s classification model had been outpaced by innovation. (iii) The financial crisis had exposed weaknesses in the regulation of nonequity instruments. (iv) Finally, rapid innovation and increasing market complexity called for higher levels of investor protection. The Commission thus sought a safer, sounder, more transparent, and more responsible financial system.34 [11.16] Toward this result, MiFID II brings about a structural reform in the design of multilateral trading systems—defined as the platforms or systems where buying and selling interests in trading financial instruments come together. It updates the trading venues classification, extending the regulatory perimeter around trading facilities operating in the OTC markets, and applies the same set of rules to this wider set of venues.35 A driving concern appears to be the ‘future-proofing’ of MiFID against changes to the nature of organized trading and to address current and potential regulatory arbitrage risks.36 MiFID II, while retaining the same taxonomy of multilateral trading venues operated by MiFID I, introduces a third one, the OTF, designed to capture all those non-equity trades that occur outside RMs and MTFs. At the same time RM and MTF rule-books have been aligned, as they ‘represent the same trading functionality’.37 MiFID II also introduces broader rules in terms of governance structure of the venues.38 The next sections analyse the regulatory framework of the three organized trading venues classified by MiFID II, and finally present what is a novelty in the regulation of the EU financial markets, namely the rule on nondiscriminatory access to CCPs and trading venues.

3. Regulated Markets [11.17] RMs represent the archetype of a public market. In practice, they encompass the main stock and derivatives exchanges. RMs are multilateral and non-discretionary trading venues, where buyers and sellers can trade in accordance with preset rules, and where trades conform to transparency

requirements.39 Market operators of RMs are the main target of MiFID II attention.40 Article 44(1) in particular makes a distinction between the market operator and the systems of the regulated market, both of which must comply with the relevant RM provisions. Para. (2) of the same Article specifies that the market operator performs the tasks relating to the organization and operation of the RM under the supervision of the competent authority. As a result, both market operators and the systems run by the same are subject to regulation and supervision.

A. Corporate Governance [11.18] In comparison with MiFID I, MiFID II has enhanced the governance requirements of market operators, aligning them with those applicable to investment firms, which in their turn track the governance requirements foreseen by the Capital Requirements Directive of 2013 (CRD IV, Articles 88 and 91) for banks. Members of the management body of a market operator must be of sufficiently good repute and possess sufficient knowledge, skills, and experience to perform their duties, and their composition shall reflect an adequately broad range of experience (Article 45(1)). In addition, all members shall commit sufficient time to perform their functions in the market operator, while the management body as a whole shall possess adequate collective knowledge, skills, and experience to be able to understand the market operator’s activities, including the main risks (Article 45(2)(a) and (b)).41 For this purpose, market operators shall devote adequate human and financial resources to the induction and training of their management body (Article 45 (3)). Furthermore, each member of this body shall act with honesty, integrity, and independence of mind to effectively assess and challenge the decisions of the senior management and to effectively monitor decision-making (Article 45(2)(c)). In addition, market operators which are significant in terms of size, internal organization, scope, and complexity must establish a nomination committee within the management body, to perform the tasks relating to the appointment of new members and the assessment of the board that are typical of such a committee both in financial and non-financial firms (Article 45(4)).42

[11.19] The overarching governance requirement for market operators asks their management body to define and oversee the implementation of the governance arrangements that ensure effective and prudent management of an organization, including the prevention of conflicts of interest, in a manner that promotes the integrity of the market (Article 45(6)). This requirement shows the two main objectives of corporate governance regulation in this area. On one side, (supervisory) boards should exercise their oversight on the management of the organization—including the market operator and the systems of the regulated market—and ensure that it is performed efficiently and with the prudence required in all financial institutions. On the other side, regulated markets are a particular type of institution, creating risks of a special type which are mainly connected with the need to ensure fair and orderly trading and the integrity of markets. This explains the focus on conflicts of interest, which can arise out of the multiplicity of parties involved in the organized trading of financial instruments, and on the resilience of trading systems. Both conflicts of interests and trading systems’ failures may not only jeopardize market integrity, but also threaten the financial system as a whole. The management body should therefore ensure that there are risk-management systems in place apt to minimize the risks at issue. [11.20] MiFID II also requires the ownership of market operators to be regulated (Article 46). First, the persons who are in a position to exercise, directly or indirectly, significant influence over the management of the RM should be suitable. Second, information regarding the ownership of the RM and/or market operator should be provided to the competent authority, including information as to any transfer of ownership which gives rise to a change in the identity of the persons exercising significant influence over the operation of the RM. Third, the competent authority should refuse to approve changes to controlling interests of the RM and/or market operator where there are objective grounds for believing that they would pose a threat to the sound and prudent management of the regulated market.

B. Organization

[11.21] Consistently with what we have seen for corporate governance in the preceding subsection, MiFID II defines some of the main tasks of the organizational structure of an RM as follows: (a) to identify conflicts of interest and manage their potential adverse consequences; (b) to manage the risks to which the RM is exposed; (c) to have arrangements for the sound management of the technical operations of the system, including contingency arrangements; (d) to have rules and procedures that provide for fair and orderly trading and efficient execution of orders; (e) to facilitate the efficient and timely execution of transactions; and (f) to have sufficient financial resources to facilitate its orderly functioning (Article 47(1)). The Directive takes specific care of the resilience of trading systems under Article 48(1), requiring an RM to have in place effective systems, procedures, and arrangements to ensure its trading systems are resilient, have sufficient capacity to deal with peak order and message volumes, are able to ensure orderly trading under conditions of severe market stress […] and are subject to effective business continuity arrangements to ensure continuity of its services if there is any failure of its trading systems.43

Other parts of the RM regime include aspects which are traditional in exchange regulation, such as admission of financial instruments to trading (Article 51), suspension and removal of financial instruments from trading (Article 52), access of members or participants to an RM, and market monitoring requirements (Article 54).

4. MTFs and OTFs [11.22] MiFID II is innovative in aligning the regulation of MTFs and RMs. However, the most radical reform in the trading venues architecture is the introduction of a new category,44 the OTF.45 The latter trades in nonequity assets—mainly derivatives and bonds—and is designed to capture all non-RM/MTF trading on organized venues, other than ad hoc bilateral trading between counterparties, which does not take place in an organized venue.46

A. Definitions [11.23] An MTF is a multilateral trading system operated by an investment firm or a market operator.47 The operation of an MTF is an investment service included in the lists of services and activities found in Annex I, Section A, MiFID II. Therefore, the operation of an MTF is subject to authorization both for investment firms and market operators.48 [11.24] An OTF is also a multilateral trading system or facility, the operation of which by either an investment firm or a market operator is an investment service requiring authorization. For an OTF to be authorized a detailed explanation must be provided to the competent authority of why the system cannot be operated as an RM, an MTF, or an SI.49

B. Corporate Governance [11.25] Like all investment firms,50 an investment firm operating either an OTF or an MTF must comply with the corporate governance requirements provided by CRD IV, which are similar to those already considered for RMs (Section II.3.A above).51 First, the firm’s governance arrangements should ensure the effective and prudent management of the same, including the prevention of conflicts of interest.52 Second, a nomination committee must be appointed in those institutions which are significant in term of their size, internal organization, and the nature, scope, and complexity of their activities.53 Third, members of the management body must possess sufficient reputation and knowledge, skills, and experience to perform their duties, and are subject to limitation on the number of directorships they can simultaneously hold.54 In addition, like all investment firms, those operating MTFs or OTFs must inform and notify the competent authority about the identities of their shareholders or members, whether direct or indirect, natural or legal persons, who have qualifying holdings and the amount of those holdings.55

C. Organization

[11.26] The organizational requirements of investment firms also apply to those firms operating an MTF and/or an OTF.56 Additional organizational requirements are either common to MTFs and OTFs or specific for each type of venue. [11.27] The common requirements include establishing transparent rules and procedures for fair and orderly trading, for the admission of financial instruments to trading, and for access to the trading facility, and identifying and managing the potential adverse consequences of any conflict of interest.57 Moreover, the requirements concerning systems resilience, circuit breakers, and tick sizes stated by MiFID II for RMs (Section II.3.B above), also apply to MTFs and OTFs.58 [11.28] Specific requirements for MTFs include the establishment and implementation of non-discretionary rules for the execution of orders in the trading system.59 In addition, investment firms and market operators operating an MTF must have arrangements in place to manage the risks to which the system is exposed and to facilitate the efficient and timely finalization of transactions.60 Moreover, firms operating an MTF should not execute client orders against their proprietary capital nor engage in matched principal trading.61 [11.29] The specific requirements for OTFs take care of keeping the multilateral character of these trading systems. First, firms operating an OTF are precluded from executing client orders against their proprietary capital or the capital of any entity that is part of the same group.62 Second, they are allowed to engage in matched principal trading only with respect to some instruments and only where the client has consented to this process.63 Third, dealing for own account other than matched principal trading is allowed to firms operating an OTF only for sovereign debt instruments for which there is not a liquid market.64 Fourth, the operation of an OTF and of a systematic internalizer cannot take place within the same legal entity.65 However, orders are executed on an OTF on a discretionary basis. This marks an important difference from MTFs, which must have nondiscretionary rules for the execution of orders. The discretion of the OTF operator is exercised when deciding to place or retract an order on the OTF

and/or when deciding not to match a specific client order with other orders available in the system at a given time.66

5. Non-Discriminatory Access to Trading Venues and to CCPs [11.30] Title VI of MiFIR is particularly important in EU financial markets regulation.67 It covers clearing services of traded financial instruments and access to CCPs and trading venues. More precisely, a CCP shall accept to clear financial instruments on a non-discriminatory and transparent basis, regardless of the trading venues at which a transaction is executed.68 Similarly, a trading venue shall provide trade feed on a nondiscriminatory and transparent basis upon request to any authorized CCP that wishes to clear transactions in financial instruments that are concluded at a trading venue.69 [11.31] Despite being only tangential to the direct regulation of trading venues, these provisions are very relevant considering the evolution of the exchange business, where in many instances, as Section IV underlines, trading venues and CCPs are a part of larger financial groups wherein a parent company controls firms operating in both the trading and posttrading businesses.70 As MiFID I intended to increase competition in the trading markets, at that point in time concentrated in a handful of national exchanges, MiFID II and MiFIR intend to guarantee the competitiveness and non-discriminatory access to infrastructural services, given that exchanges have morphed into conglomerates of trading and post-trading services providers and also in view of the growing trading and clearing market for derivatives.71

III. The EU Scenario of Trading Venues [11.32] This section explores how trading venues have evolved into FMI groups over the past few years and looks at the current structure of the FMI

groups in the EU. In doing so, we first provide a general overview on the business evolution in the trading section and conclude with six case studies on the largest FMI groups in the EU.

1. The Evolution of the Trading Venues Market [11.33] Over the last twenty years the trading industry has undergone significant reforms and transformations that have contributed to reshape the system of trading venues.72 Three main factors contributed to the reorganization of the trading industry: regulation, competition, and technology. Each of these factors, at some point in time, caused industry members to evolve their business structure. [11.34] The first signs that the trading industry had significantly evolved from its historic models appeared when traditional stock exchanges changed their corporate structure: from private companies to publicly owned entities, from member-owned mutual firms to investor-owned for-profit firms.73 Opening up equity capital to investors made it easier for firms to raise fresh resources for investments. These investments became necessary to keep pace with technological innovation and the increasing competition in the markets fostered by milestone reforms such as ISD,74 MiFID I,75 and now MiFID II.76 [11.35] To face the challenges posed by the heightened active competition of new platforms, which were able to increase market shares by exploiting technological innovation and by benefiting from the opening of the market, exchanges began a season of strong corporate investments. In the last twenty years, the market for corporate control of firms operating in the trading and post-trading services boomed.77 Through mergers and acquisitions, the champions among the traditional exchanges created big conglomerates of FMIs and financial services providers. This consolidation of the ‘exchange’ industry led to the growth of FMI groups, where a group parent company has the control, via vertical or horizontal silo structure, of different types of venues (both for cash instruments and derivatives instruments), central clearing houses, settlement and custody services providers, and market information services providers.

2. A Case Study of Six FMI Groups [11.36] This section explores the scenario of trading venues in the EU. More specifically, we examine six case studies of FMI groups to show how trading venues have evolved in their organizational structure in the last twenty years creating FMI groups wherein RMs, MFTs, and CCPs coexist.78 Our case studies are: the London Stock Exchange Group (LSEG), the Deutsche Börse group (Deutsche Börse), the Euronext group (Euronext), the Bolsas y Mercados Españoles group (BME), the NASDAQ OMX Nordic Exchange (Nasdaq Nordic), and the BATS group (BATS).

A. The London Stock Exchange Group [11.37] LSEG, a UK incorporated company, is a diversified international group that operates a broad range of international equity,79 bond, and derivatives markets, including the London Stock Exchange,80 Borsa Italiana,81 MTS (Europe’s leading fixed income market),82 and Turquoise (one of Europe’s largest MTFs).83 Post-trading and risk management services are also a significant and growing component of the LSEG’s business. Cassa di Compensazione e Garanzia (CC&G)84 and LCH.Clearnet Group85 are the clearing houses operating in the group, while Monte Titoli86 and globeSettle SA87 provide settlement and custody services. The current structure of LSEG is the result of ten years of active investment campaigning. After failed attempts to merge with Deutsche Börse in 2000 and with NASDAQ in 2004,88 LSE merged with Borsa Italiana in 2007. The merger with the Italian stock exchange and the appointment of a former investment banker as CEO of the new group invigorated the investment campaign of the newly created LSEG. Starting from the acquisition of one of the largest European MTFs—Turquoise—in 2009, in 2012 LSEG expanded its post-trading information business by acquiring FTSE International Limited and then completing the creation of a vertical silo of trading and post-trading services by acquiring the majority stake in LCH.Clearnet Group Limited—one of the largest European CCPs. Finally, in 2014, Frank Russell Company become part of LSEG, expanding the business of the group to global asset management.89 In March 2016 LSEG

and Deutsche Börse negotiated a merger agreement, which—if approved by the competent regulatory and competition authorities—would result in the creation of one of the largest FMI groups in the world. A group that— among other business lines—would include five main trading venues (LSE Main Market and AIM, Borsa Italiana, Frankfurter Börse, Turquoise); two of the largest CCPs in Europe (LCH.Clearnet and Eurex Clearing); Clearstream, a leading securities and collateral depository; and three main information providers, FTSE, Russell Group, and Stoxx.90 [11.38] If we break down the revenue of the LSEG, we see that ‘capital market income’, here including both primary and secondary markets, accounts for 23 per cent of the £1,418.6m total income for 2015.91 Revenue generated by primary markets—Main Market, AIM, and Borsa Italiana (on the equity side), and Exchange Trade Funds (ETFs) and Exchange Traded Products (ETPs) market (for exchange traded products)—amounts to 26.8 per cent of the capital markets revenue; while 73.2 per cent of revenue comes from the secondary markets—equity (59.2%), derivatives (8.5%), and fixed income (32.3%).92 Post-trading services offered by the group— including both CC&G, Monte Titoli, and LCH.Clearnet—represent 33 per cent of the global income; information services represent 37 per cent; and the remaining 7 per cent comprises technology services.93 [11.39] LSEG is a publicly traded company on the London Stock Exchange. As of 2 August 2016 LSEG had four notified substantial shareholders: Qatar Investment Authority (10.3%), BlackRock Inc (6.9%), Lindsell Train Limited (5%), and Veritas Asset Management LLP (3.0%). In general terms, 79 per cent of LSEG’s capital at as 8 June 2015 was held by institutional investors and 8 per cent by market participants and hedge funds.94 [11.40] In its 2014 report, LSEG identified the broad scope of MiFID II and the issues related to the access provisions affecting trading venues, CCPs, and benchmark providers, and acknowledged that is likely that almost all aspects of group operations will be affected to some degree by the MiFID II implementation. However, while in the process of waiting for the implementation of the Level 2 regulation, they cannot quantify the impact of the reforms on the group’s activities. The areas of biggest

regulatory development that will affect the group are the market structure measures aimed at promoting integration, competitiveness, and efficiency in the financial markets, namely the rules on non-discriminatory access to trading venues and CCPs.95

B. Deutsche Börse Group [11.41] Deutsche Börse AG is Europe’s largest exchange group for capitalization and it is an integrated provider of products and services covering the entire process chain of securities and derivatives trading and post trading.96 [11.42] Trading services (equity and derivatives) within the group are offered by the Frankfurter Börse, the German stock exchange, which operates through the Xetra electronic trading platform; and by the Eurex group, fully owned by Deutsche Börse, which operates a Eurex derivatives exchange, two MTFs (Eurex Bonds and Eurex Repo), and ISE, a US-based derivative exchange. [11.43] Of the group’s €2,367m total income for 2014, 51 per cent was represented by trading and clearing revenue (of which 8% represented the Xetra equity trading segment and 43% the Eurex derivative trading and equity and derivatives clearing); 32 per cent of the total income derived from the post-trading services offered by Clearstream (the fully owned posttrade settlement and custody provider). The commercialization of market data and market infrastructure technology accounted for 17 per cent of the group’s total income. [11.44] As for its shareholder structure, 94 per cent of shares in Deutsche Börse are held by institutional investors and 6 per cent by private investors.97 [11.45] Deutsche Börse fully owns Eurex group and its subsidiary Eurex Clearing AG, the group-owned clearing house, which exclusively carries out the central counterparty service of the group’s trading venues. Eurex Clearing is designated to clear all transactions executed on the group’s

exchanges and MTFs (general clearing conditions), but also provides clearing services to a non-group market, the Irish stock exchange.

C. Euronext Group [11.46] Euronext is a pan-European exchange group, offering trading services with multiple markets for equities, fixed income securities, and derivatives.98 Euronext holding was created in 2000 after the merger of the Dutch, French, and Belgian exchanges into a single cross-border entity. In 2002 the Portuguese stock exchange and the London-based London International Financial Futures and Options Exchange (LIFFE—a derivatives market) were acquired by Euronext. In 2007 the New York Stock Exchange—interested in expanding in the derivatives trading segment of Euronext—merged with Euronext, creating the first transatlantic exchange, called NYSE Euronext. In 2012, the Intercontinental Exchange (ICE), the leading network of regulated exchanges and clearing houses for financial and commodity derivatives markets, interested in LIFFE and in expanding its share in the European derivatives markets, acquired NYSE Euronext. Two years later, Euronext was spun off from ICE (with the exception of LIFFE) and listed again in Europe, reviving as a stand-alone exchange group. During the initial public offering of the new Euronext, a consortium of eleven investors took the lead in acquiring a significant holding in the new company (33.36%).99 [11.47] Differently from LSEG and Deutsche Börse, which developed their business as integrated vertical silos, Euronext grew via a horizontal integration, by focusing on trading activities on different products (both cash and derivatives) in different jurisdictions, but outsourcing the posttrading services. Of the €458.5m revenues in 2014, 50 per cent came from the equity and debt (cash) markets (14% listing fees and 36% cash trading); the derivatives trading markets represented a modest 10 per cent of the total income, with post-trading clearing and settlement services reaching 13 per cent. The sale of market data and indexes accounted for 20 per cent of all the revenues.100

[11.48] The post-trading services within Euronext are outsourced from the group to major CCPs. Clearing is provided by LCH.Clearnet SA (part of the LSEG), while settlement is offered by Euroclear (the leading custody and settlement provider), and, for Euronext Lisbon, by Interbolsa, the fully owned Portuguese Central Securities Depository.101

D. Bolsas y Mercados Españoles Group [11.49] In Spain, BME manages all the trading and post-trading infrastructures of the country, except for the futures exchange—Sociedad rectora del mercado de futuros del aceite de oliva SA.102 [11.50] BME is structured as a vertical silo; the BME holding group was founded in 2001 in order to consolidate the four regional exchanges (Bolsa de Bilbao, Bolsa de Valencia, Bolsa de Madrid, and Bolsa de Barcelona); the four stock exchanges’ governing companies also own 25 per cent each of Sociedad de Bolsas, SA, which manages and operates the Spanish electronic trading platform (SIBE).103 BME, in addition to being the operator of the four regional exchanges, is the market operator of two MTFs: Mercado Alternativo Bursátil (the market segment for SMEs) and Latibex (the market for European investors to trade euro-denominated Latin American stocks).104 [11.51] Furthermore, looking at the non-equity venues, the Mercado de Renta Fija, SA operates a regulated market and two MTFs in the fixedincome segment, while in the derivatives area BME is active through the MEFF Exchange, which manages the derivatives exchange. [11.52] Post-trading clearing services (for the moment only for derivatives and repos) are provided by BME clearing, SA, the group’s fully owned clearing house, which was set up in 2013 after the implementation of EMIR, and the consequential segregation of trading activities and clearing activities in two separate legal entities. Finally, settlement services are managed by Iberclear, which operates as a CSD for equities and bonds and offers trade reporting services.105

E. NASDAQ OMX Nordic Exchange [11.53] Nasdaq Nordic is a fully owned subsidiary of Nasdaq Inc, and manages and integrates through a single platform the stock exchanges of the Nordic and Baltic countries.106 It operates these venues as well as MTFs under a single rulebook for listing, equity, and derivatives trading; and offers, through First North, a growth market, designated for small and growing companies.107 [11.54] Nasdaq Nordic is also active in the management of post-trading infrastructures: it operates CSDs in Estonia, Latvia, Lithuania, and Iceland, and furthermore runs the group’s clearing house, Nasdaq Clearing AB, an investment company incorporated under Swedish law, which provides clearing services to Nasdaq Nordic markets. Every Nasdaq national market is authorized and supervised by the competent national authority. Nasdaq Inc. also controls Nasdaq NLX, a London-based MTF for trading interest rates derivatives. The trades executed on NLX are cleared trough LCH.Clearnet.108

F. BATS Group [11.55] BATS Global Markets is an international trading group specializing in the management of US and European exchange infrastructures.109 BATS is a privately held company owned by a consortium of banks and other investors.110 Founded in 2005, BATS is recognized and supervised by the SEC and provides trading services in the US as a registered national securities exchanges for the US equity markets.111 In 2011, BATS acquired Chi-x, the first established European MTF, which merged with BATS European MTF and became BATS Chi-x, a fully owned subsidiary of BATS Global Markets, which is currently one of the biggest European equity markets. [11.56] BATS Chi-x maintained the qualification of MTF until 2013, when it was recognized as an RM by the FCA, becoming the first panEuropean RM, offering trading services across fifteen European national markets over one platform and under one rule book.112

[11.57] In January 2012, BATS Chi-x adopted a user-driven model for clearing activities, providing CCPs’ interoperability options to its customers. Traders on BATS Chi-x may choose one of three different counterparties to clear their trades: LCH.Clearnet (the LSE majority-owned CCP), SIX x-clear (the CCP owned by SWZ, the Swiss stock exchange), and EuroCCP NV (where Bats Chi x holds an ownership interest of 25% and is also represented on the board).113

IV. FMI Groups in an MiFID II World 1. Formation of Pan-European Groups [11.58] The scenario that emerges from the analysis of the European trading venues is the result of market trends towards integration—at a firm level—and consolidation—at a geographical level. [11.59] After the de-mutualization and privatization of the historical exchanges, trading venues became active players in the markets. By going public and attracting new investors, venues were able to raise funds and expand their lines of business. New venues were also created, often at the initiative of banks and other intermediaries, with the effect of increasing competitive pressure on the incumbent exchanges. Regulation supported this market transition and played a central role in shaping the environment for structural changes. The injection of competition into the trading venues scenario and into the clearing markets operated by MiFID I114 and EMIR,115 and the necessity of stock exchanges to foster greater efficiency in their activities, resulted in trading venues updating their business models, structures, and governance, and integrating their trading and post-trading services. [11.60] Facing competition from new trading markets players, venues soon joined the new and growing segment of alternative trading systems— either by directly setting up their own MTFs or by acquiring existing ones.

Furthermore, they horizontally integrated cash and derivatives trading platforms. [11.61] In an alternative process, venues vertically integrated their trading activities with the post-trading businesses of CCPs and CSDs, offering their clients the whole chain of trading and post-trading services, and actual FMI groups. [11.62] In the aftermath of the financial crisis and the introduction of mandatory central clearing for OTC derivatives, the business structure of these groups shifted significantly. The FMI groups realized the potential of the clearing business—especially for derivatives—and, facilitated by the non-discriminatory access and interoperability policies enforced by the authorities, made their clearing activities one of the most profitable lines of their groups. Similarly, by taking advantage of their position as FMI operators they expanded their market information (indexes, market data statistics, real time data, regulatory news services, etc.) and technology businesses (trading platforms, settlement software, etc.). [11.63] In fact, it seems at the European (and in some ways also at the transatlantic) level, FMI groups have morphed into pan-European ventures, offering cross-border trading and post-trading services across different jurisdictions. [11.64] With this background in place, this section moves to question the current structure of the trading industry. First, MiFID II has impacted the industry by fostering competition with the introduction of a new class of multilateral trading venues, the OTF. How will this reform impact the current structure of the EU securities and derivatives markets? Second, trading venues have evolved into FMI groups with a systemic role; so it is important to consider what risks they pose to the financial system and what tools the authorities have to monitor them.

2. The OTF in FMI Groups

[11.65] MiFID II will reshape the trading landscape by introducing a new actor, the OTF. The set of rules on OTFs (Title III MiFID II), mandatory trading of derivatives (Title V MiFIR), and non-discriminatory access to CCPs and trading venues (Title VI MiFIR), represents the ‘crisis-driven’ pieces of regulation in the MiFID Review and completes the reform of the OTC derivatives market started by EMIR. [11.66] EU regulators, implementing the international guidelines set by the G20 in Pittsburgh in September 2009,116 which were then expanded by the Financial Stability Board to mandatory trading for OTC derivatives,117 opted for the creation of a more flexible form of venue as a marketplace for eligible derivatives (OTF trading activities would also include bonds, structured finance products, and emission allowances).118 The market operator or investment firm managing an OTF will have to provide the competent authority with a detailed explanation of why the trading system cannot be operated as an RM, MTF, or SI.119 [11.67] At the current stage, it is difficult to assess how markets are going to react to this new trading venue and how OTFs will be able to capture a relevant share of OTC trades. It is worthwhile to note that the International Swap and Derivatives Association (ISDA) welcomed the establishment of the OTF category.120 Furthermore, in the US, the Dodd–Frank Act introduced a new player in the trading landscape.121 Essentially, the Act established the swap execution facility (SEF)—similar to the European OTF—to provide trading services for derivatives. In the US, since the introduction of the secondary level regulation by the Commodity Futures Trading Commission, twenty-three SEFs have registered.122 [11.68] Because of the relevant role that derivatives clearing plays in FMI groups, it is very likely that the existing groups, which in many instances already offer trading services for more standardized derivatives on RMs or MTFs, will expand their trading business to the operation of OTFs. From the end of 2017, eligible derivatives will be required to be centrally cleared. At that time, centrally cleared derivatives that are today mainly traded and executed OTC will have to find a regulated locus for trading. The growth of OTFs within the current FMI groups is a very likely outcome.

3. Challenges of FMI Groups to Regulators [11.69] The trading venues environment has evolved a great deal over the last twenty years. From monolithic national stock exchanges, firms evolved into cross-border multi-business entities. The trading business has been integrated with post-trading services like clearing, settlement, and custody. Firms have expanded their market shares in the data and information market and in the IT trading and post-trading services. Having a look at the lines of business that primarily contribute to the profits of FMI groups, the role of the traditional trading business is generally relatively modest, compared to the growing weight of clearing activities and information and technologies services. [11.70] FMI groups are cross-border multi-business entities with a silo structure—either vertical or horizontal—and mostly run as unitary enterprises. The parent company—in the majority of the instances a listed company—is either a holding or a market operator, owning a controlling participation or full ownership of different lines of business, each operated as separate legal entities, often located in multiple jurisdictions. Primary listing markets, trading venues (both RMs and MTFs), CCPs, CSDs, and information services and IT services companies all coexist in the FMI group, where the parent company pulls the strings, setting the strategy and business plans for the whole group. [11.71] The evolution of stock exchanges into FMI groups is a relevant trait of the evolution of financial markets in the last twenty years and raises important challenges to financial regulators and competition authorities. First, FMI groups can potentially distort competition in the chain of trade and post-trade services. Increasing the clearing costs for transactions not traded in the group’s venues, or even blocking the access to clearing for the very same transactions, requiring contracts traded in the group’s venues to be centrally cleared in the group’s CCP, are examples of how FMI groups might distort competition in the financial markets. [11.72] Second, FMI groups are systemic institutions, not just for their size, but mainly for their structural/social role in providing a primary service for the smooth functioning of the financial markets. A situation of

financial distress in an FMI group might cause systemic disruption and undermine financial stability.

4. MiFID II’s Approach to FMI Groups [11.73] Does MiFID II acknowledge the potential risks connected to the activities of FMI groups? Do MiFID II and MiFIR (or EMIR) offer specific ‘tools’ to deal with FMI groups? The answer to these important questions is rather short. MiFID II does not explicitly take the FMI group into account. Trading venues are regulated as individual entities, which may or may not be operated within groups. The same is true for CCPs under EMIR. The fact that trading and post-trading services can give rise to a common enterprise within FMI groups is left in the background of EU financial regulation. [11.74] This could be explained historically, by arguing that trading and clearing have always been regulated differently; the first within securities regulation, the second more in the domain of banking and financial regulation because of the systemic implications of clearing activities. It is also possible that because the regulators are different—securities authorities for trading services, central banks for clearing and similar services—they want to keep their regulatory turfs separate. An additional explanation could be that governments and regulators want to avoid the joint exercise of trading and post-trading services within an individual enterprise, as this could cause mutual contamination with serious systemic consequences. From this perspective, keeping the relevant activities separate, under different regulatory and supervisory umbrellas, is a kind of ring-fencing of each area of activity, clearing in particular, which is more likely a cause of systemic concern. [11.75] However, the present EU regulation is not totally blind to the existence of FMI groups, and includes some tools which may prevent some types of contamination amongst trading, post-trading, and other activities when exercised in a group context. As shown in the following subsections, three sets of rules allow, to varying degrees, for the partial tackling of some of the potential risks underlying FMI groups, despite the fact that they are not specifically considered by MiFID II or MiFIR.

A. Conflicts of Interest [11.76] The first set of rules looks into the conflicts of interest that might arise in those situations where an investment firm or market operator runs an MTF or an OTF. In this scenario, the market operator or investment firm shall have arrangements in place to clearly identify the conflicts of interest between themselves, their owners, and the trading venue and its sound functioning; and manage the potential adverse consequences of these conflicts for the operation of the relevant venues or for their members or participants or users.123 This will require the individual firm within an FMI group to ring-fence the operation of its trading venue or venues, so as to avoid contamination of their sound functioning from conflicting interests found either in the same firm or within the FMI group to which the firm belongs. Similarly, RMs could be subject to the same set of rules on conflicts of interest prevention, which require clear identification and management of the potential adverse consequences of any conflict of interest between the RM, its owners, or its market operator, and the sound functioning of the RM. Assuming that in the same FMI group there are all three types of trading venues, the group’s market operators operating RMs, MTFs, and OTFs (and investment firms only for MTFs and OTFs) shall have in place effective arrangements to prevent and/or manage conflicts of interest arising between these different venues, their owners, and operators.124 However, no specific provisions of MiFID II look into the dynamics of conflicts of interest between the trading and post-trading activities within the FMI group.

B. Transparency of Ownership and Suitability of Shareholders [11.77] The second set of rules targets transparency in the shareholders structure. Regulators impose a disclosure regime on venues with regard to the ‘persons’ exercising particular influence on the management, qualifying shareholders, and connections between trading venues. The investment firms and market operators operating an MTF or an OTF are required to provide the competent authority with a detailed description of the functioning of the MTF or OTF, including any links to or participation by a

regulated market, an MTF, an OTF, or a systematic internalizer owned by the same investment firm or market operator, and a list of their members, participants, and/or users.125 Investment firms operating MTFs or OTFs as well as market operators of RMs have to disclose the competent authority all information regarding the ownership of the trading venue and of the investment firm/market operator.126 [11.78] Moreover, in the case of RMs, the persons who are in a position to exercise, directly or indirectly, significant influence over the management of the venue must be suitable.127 The same requirement applies to an investment firm intending to operate an MTF or OTF.128 Similar provisions have an impact on group structures, particularly when the venue is operated by a subsidiary of an FMI group and the executives of the parent company influence its management by determining the strategies and taking other key decisions.

C. Rules on Access [11.79] The third set of rules spots the risks connected to potential distortive effects that having an FMI group might create in the trading and post-trading market. These risks were exacerbated by the post-crisis reform of the OTC markets, which introduced mandatory trading and clearing for derivatives. EU policymakers addressed this point by focusing on the access to trading venues and CCPs.129 As mentioned earlier, these rules intend to foster competition between venues and between CCPs and support nondiscriminatory access to essential infrastructural financial services.130

5. Consolidated Supervision? [11.80] While addressing competition issues, MiFID II does not include provisions for the prudential regulation and supervision of FMI groups. As argued above, trading venues (and CCPs) are regulated as ‘ring-fenced’ stand-alone legal entities, each of them subject to prudential regulation and supervision. The trading venues within an FMI group, as well as their

market operator or investment firm operating the venue, are regulated under the national rules implementing MiFID II. When based in different countries, the regulators of these countries will supervise the trading venues and their operators under the rules implementing the ‘Obligation to cooperate principle’ stated in the Directive (Article 79).131 The same holds for CCPs under EMIR. As a result, FMI groups find themselves subject to multiple regulators, each supervising one or more of the entities composing the group. [11.81] Despite their systemic relevance as integrated enterprises, FMI groups are not subject to consolidated supervision. Regulators may, in practice, take a holistic view of their activities and supervise individual entities as part of a larger group, as they are well advised to do. However, this approach is not mandated by EU law, even though fragmented provisions of MiFID II and MiFIR hint at the need to consider individual entities operating trading venues as linked to other entities within a group, which may create the potential for conflicts of interest in the management of trading venues or for external influence in the strategy and business plans of individual entities in the FMI group. Also, their internal control and risk management systems should be centralized under company law and best practices of corporate governance, but similar requirements for an integrated risk-management system at group level are not foreseen by financial regulation. [11.82] A comprehensive consolidated approach to the supervision and regulation of FMI groups is lacking. The cross-border reach and the multibusiness span of FMI groups’ activities challenge the current regulatory and supervisory architecture of the European financial markets. This could generate problems, to the extent that the crisis of an entity within the FMI group could reverberate onto other entities of the same group, which also highlights the possible need for crisis-management systems, including recovery and resolution plans, at the FMI group’s level. After surveying the financial industry’s regulatory landscape, we spotted two regulatory responses to cross-border and multi-business ventures that might be considered as possible models to deal with the FMI groups phenomenon: the supervisory colleges of CCPs in EMIR and the Financial Conglomerates Directive.

[11.83] Under EMIR, when a CCP’s members are established in different member States and the potential default of a CCP might have cross-border effects, all relevant competent authorities and ESMA are involved not just in the supervision, but also in the authorization process of the CCP, with ESMA playing a coordination role in the ‘college of supervisors’, guaranteeing the consistent and correct application of EU law.132 The composition of the colleges reflects in some ways the indirect acknowledgement of EU regulators of the infrastructural role of CCPs. Supervisors of the CCP sit together with the supervisors of the entities which operate with the CCP: clearing members, trading venues, interoperable CCPs, and central securities depositories.133 [11.84] The Financial Conglomerates Directive provides the first comprehensive supplementary supervisory framework for financial groups operating multiple businesses in different sectors of the financial markets and across borders.134 Recognizing the systemic role of financial conglomerates, the Directive sets in place mechanisms to ‘enhance the prudential soundness and effective supervision of financial conglomerates’.135 To achieve this result the Directive builds a supplementary supervisory framework for financial conglomerates. The financial conglomerates regime has three peculiarities that make it a feasible regulatory solution to be considered for FMI groups. First, the Directive introduces a supplementary layer of supervision for financial conglomerates that is built on top of—and does not substitute—the sectorial supervisory and regulatory regimes of the individual financial activities within the conglomerate. Second, the Directive acknowledges that even though financial conglomerates are often managed on a business line basis, a crucial role is played by the holding company, whose management should be monitored and subject to specific conduct requirements. Finally, coordination should be ensured among the regulatory and supervisory entities involved in the oversight of the financial conglomerate, with an authority acting as a ‘coordinator’.136 [11.85] As financial conglomerates, FMI groups are multi-business and cross-border entities playing a systemic role in the financial markets; as CCPs, the business of FMI groups directly affects many stakeholders’ interests. For these reasons, prudential supervision of FMI groups on a

group-wide basis would be desirable to more effectively assess: the financial stability of the group; risk concentration and intra-group transactions; and internal risk-management processes and sound management at a conglomerate level. Furthermore, more coordination among authorities responsible for the supervision of the sectorial businesses within the conglomerate should be achieved. Two paths could be followed: the college approach—adopted for the supervision and authorization of CCPs under EMIR—has the benefit of creating a single forum where multiple regulatory and supervisory agencies meet. The coordinator solution—adopted for financial conglomerates—empowers the competent authority of the most relevant entity of the group to coordinate the gathering and dissemination of information on the conglomerate and to oversee and assess the financial situation, compliance, organization, and internal control of the financial conglomerate. A mix solution, which would combine elements of both the college and coordinate regimes, would be a desirable regulatory solution to the current regulatory gap, and a stronger role of ESMA could be envisioned.

V. Conclusion [11.86] MiFID II brings modest changes to the EU landscape of trading venues. The newly introduced OTFs are going to be the reference venues for a significant portion of derivatives trading in years to come. The RM and MTF regimes have been aligned, and specific provisions strengthen the governance of the venues and their operators. However market dynamics are already challenging the MiFID II regulatory framework for the governance and organization of trading venues. Trading venues have developed over the last twenty years into FMI groups that provide both trading and post-trading services. These new conglomerates test the capacity of the current regulatory and supervisory regime of financial markets—and MiFID II itself—to oversee their activities and to guarantee competition and stability in the trading and post-trading industry. MiFID II does not explicitly take FMI groups into account—trading venues are regulated as individual entities, which may or may not be operated within a group. In MiFID II, only three sets of rules address, to varying degrees,

some of the potential risks underlying FMI groups: conflicts of interest between firms operating RMs, MTFs, and OTFs; transparency of ownership of trading venues and suitability of trading venues’ shareholders; and, finally, non-discriminatory access to trading and post-trading services. However, prudential regulation and supervision of the FMI group has not been included in the MiFID review process. This regulatory gap might be a threat to the stability of financial markets, and regulators should consider a regulatory intervention to fill it. The experience of the regulatory and supervisory colleges of CCPs under EMIR and the regulatory framework of the Financial Conglomerates Directive could be two possible ways to strengthen the oversight of FMI groups.

1

European Parliament and Council Directive 2004/39/EC of 21 April 2004 on markets in financial instruments amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC (MiFID I) [2004] OJ L 145/1; see also European Commission Directive 2006/73/EC of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive [2006] OJ L241/26; European Commission Regulation (EC) 1287/2006 of 10 August 2006 implementing Directive 2004/39/EC of the European Parliament and of the Council as regards record-keeping obligations for investment firms, transaction reporting, market transparency, admission of financial instruments to trading, and defined terms for the purposes of that Directive [2006] OJ L 241/1. 2 European Parliament and Council Directive (EU) of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU (MiFID II) [2014] OJ L173/349. 3 European Parliament and Council Regulation (EU) Regulation 600/2014 of 15 May 2014 on markets in financial instruments and amending Regulation (EU) No 648/2012 (MiFIR) [2014] OJ L174/84. 4 Both pieces of Level 1 legislation provide the regulatory architecture for the EU internal financial market. However, an important role is going to be played in the coming years by ESMA in the adoption of extensive Level 2 delegated and implementing acts and technical standards. 5 The third regulatory pillar of the post-crisis reform of the European financial market infrastructure is the European Market Infrastructure Regulation (EMIR); European Parliament and Council Regulation (EU) 648/2012 of 4 July 2012 on OTC derivatives, central counterparties and trade repositories [2012] OJ L201/2; for a comprehensive

analysis of financial market infrastructures see Guido Ferrarini and Paolo Saguato, ‘Regulating Financial Market Infrastructures’ in Niamh Moloney, Eilis Ferran, and Jennifer Payne (eds) The Oxford Handbook of Financial Regulation (Oxford: OUP, 2016). 6 EMIR is the piece of legislation that primarily focuses on the regulation of OTC derivatives markets. 7 See Jonathan Macey and Maureen O’Hara, ‘From Markets to Venues: Securities Regulation in an Evolving World’ (2005) Stanford Law Review 58, 563; Guido Ferrarini and Niamh Moloney, ‘Reshaping Order Execution in the EU and the Role of Interest Groups: From MiFID I to MiFID II’ (2012) EBOR 13, 557, 565–71. 8 See Jonathan Macey and Maureen O’Hara, ‘Regulating Exchanges and Alternative Trading Systems: A Law and Economics Perspective’ (1999) 28 Journal of Legal Studies 17; on the role of technologies in reshaping the financial services industry see Chris Brummer, ‘Disruptive Technology and Securities Regulation’ (2015) Fordham Law Review 84, 977. 9 For the purpose of this study, we investigate trading venues as defined by MiFID II, namely RMs, MTFs, and OTFs, which are all multilateral and formal trading facilities. 10 MiFID II recognizes two groups of trading facilities: trading venues (including RMs, MTFs, and OTFs) and systematic internalizers. 11 MiFID II, Article 4(1)(24). 12 See Daniel R. Fischel, ‘Organized Exchanges and the Regulation of Dual Class Common Stock’ (1987) University of Chicago Law Review 54, 119, 121–3 (arguing that there is little fundamental difference between the economic role of a stock exchange and that of an ordinary shopping centre or flea market); Daniel R. Fischel and Sanford J. Grossman, ‘Customer Protection in Futures and Securities Markets’ (1984) Journal of Futures Markets 4, 273 (making similar statements for futures exchanges); Ruben Lee, What is an Exchange? The Automation, Management, and Regulation of Financial Markets (Oxford: OUP, 1998), pp. 8–33. 13 See J. Harold Mulherin, Jeffry M. Netter, and James A. Overdahl, ‘Prices are Property: The Organization of Financial Exchanges from a Transaction Cost Perspective’ (1991) Journal of Law and Economics 34, 591, arguing that a financial exchange is a firm that creates a market in financial instruments; its product is accurate information as reflected in prices. 14 See Robert Coase, The Firm, the Market and the Law (Chicago: University of Chicago Press, 1988), p. 9, arguing that ‘exchanges, often used by economists as examples of a perfect market and perfect competition, are markets in which transactions are highly regulated and suggesting that “for anything approaching perfect competition to exist, an intricate system of rules and regulations would normally be needed”’. For a defence of exchanges as effective self-regulators, see Paul G. Mahoney, ‘The Exchange as Regulator’ (1997) Virginia Law Review 83, 1453. 15 Larry Harris, Trading and Exchanges: Market Microstructure for Practitioners (New York: OUP, 2003), pp. 35–6. 16 MiFID II, Article 4(1)(20).

17

Examples of alternative trading (electronic) platforms are broker crossing networks (dark pools), order matching systems, etc. 18 See Chapter 23 in this volume. 19 This venue classification is based on Guido Ferrarini and Niamh Moloney,(n. 7), 584 . 20 See Guido Ferrarini and Paolo Saguato, ‘Reforming Securities and Derivatives Trading in the EU: From EMIR to MiFIR’ (2013) JCLS 13, 319. 21 MiFID II, Article 20(6)(8). 22 See Alberto Cybo-Ottone, Carmine Di Noia, and Maurizio Murgia, ‘Recent Developments in the Structure of Securities Markets’ (2000) Brookings-Wharton Papers on Financial Services 223; Jonathan Macey and Maureen O’Hara, ‘Globalisation, Exchange Governance, and the Future of Exchanges’ (1999) Brookings-Wharton Papers on Financial Services 1. 23 On the choice between members’ ownership/governance and investor ownership/governance in this area see Oliver Hart and John Moore, ‘The Governance of Exchanges: Members’ Cooperatives versus Outside Ownership’ (1996) Oxford Review of Economic Policy 12, 53. See also Guido Ferrarini, ‘Stock Exchange Governance in the European Union’ in Morten Balling, Elizabeth Hennessy, and Richard O’Brien (eds) Corporate Governance, Financial Markets and Global Convergence (Dordrecht: Kluwer Academic Publishers, 1998). 24 See Ruben Lee, Running the World’s Markets: The Governance of Financial Infrastructure (Princeton, NJ: Princeton University Press, 2011), pp. 169–200. 25 See Macey and O’Hara (n. 4), 582, arguing that ‘when exchanges engage in selfregulation they generate and enforce rules that directly affect their own commercial interests’. 26 See International Organization of Securities Commissions (IOSCO), ‘Regulatory Issues Arising from Exchange Evolution’. Consultation Report (2006) 7, arguing that ‘the move by many exchanges to a for-profit business model, together with increased competition in the provision of market services in most markets, raises a number of questions about the appropriate regulatory role of exchanges. These issues run from the compatibility of for-profit operation with public interest objectives to the adequacy and efficiency of regulation’. 27 See FSA, Review of the Listing Regime (2003). 28 See Ferrarini (n. 23), 139 ff. 29 See John C. Coffee, Jr and Hillary A. Sale, Securities Regulation (12th edn, St Paul, MN: Foundation Press, 2012); Niamh Moloney, EC Securities Regulation (2nd edn, Oxford: OUP, 2008), pp. 763 ff. For a comparative survey, see IOSCO, Regulatory Issues Raised by Changes in Market Structure. Consultation Report (2013), noting that in most jurisdictions similar rules apply for exchange trading market systems as ‘market places’. However, considerable differences appear to remain as regards OTC trading. For instance, several jurisdictions stated that most of the trading which takes place OTC is not subject to any pre-trade transparency requirement or fair access rule.

30

The above three paragraphs draw on Ferrarini and Saguato (n. 5), 576–7. The Markets in Financial Instruments Directive (MiFID) 2004/39/EC, OJ 2004 L 145/1, MiFID Level 2 Commission Directive 2006/73/EC, OJ 2006 L 241/26, and MiFID Level 2 Commission Regulation (EC) No 1287/2006, OJ 2006 L 241/1. 32 See Ferrarini and Moloney (n. 7), 560; Ferrarini and Saguato (n. 5), 580. 33 Commission Staff Working Paper Impact Assessment Accompanying the document Proposal for a Directive of the European Parliament and the Council on Market in Financial Instruments and he Proposal for a Regulation of the European Parliament and of the Council on Markets in Financial Instruments (COM(2011) 656 final), (SEC(2011) 1227 final), Brussels, 20.10.2011 SEC (2011) 1226 final, p. 5. 34 See Ferrarini and Saguato (n. 5), 576; Ferrarini and Saguato (n. 20), 339; See House of Lords, European Union Committee, ‘MiFID II: Getting it Right for the City and EU Financial Services Industry’ (2012) available at

(offering a detailed critical assessment of MiFID II). 35 See Ferrarini and Moloney (n. 7), 586. 36 MiFIR, Recital 3. 37 ibid., Recital 6. 38 For a primer on the MiFID II reforms to trading venues see Nis Jul Clausen and Karsten Engsig Sørensen, ‘Reforming the Regulation of Trading Venues in the EU under the Proposed MiFID II: Leveling the Playing Field and Overcoming Fragmentation?’ (2012) 12 European Company and Financial Law Review 275. 39 An RM is a multilateral system operated and/or managed by a market operator, which brings together or facilitates the bringing together of multiple third-party buying and selling interests in financial instruments—in the system and in accordance with its nondiscretionary rules—in a way that results in a contract, in respect of the financial instruments admitted to trading under its rules and/or systems, and which is authorized and functions regularly and in accordance with Title III of this Directive, MiFID II, Article 4(1) (21). 40 When RMs are operated or managed by a market operator other than the RM itself, MiFID II empowers Member States to determine how different obligations are allocated between the RM and the market operator, MiFID II, Article 44(1)(1). 41 Article 45(2)(a) sets some minimum limits on the amount of directors’ positions a member of the management body of market operators can hold. Specifically a director shall not hold at the same time positions exceeding more than one of the following combinations: (i) one executive directorship with two non-executive directorships; (ii) four non-executive directorships. However, executive or non-executive directorships held within the same group where the market operator owns a qualifying holding shall be considered to be one single directorship. 42 The four functions of the nomination committee can be summarized here as: (a) identify and recommend, for the approval of board or for approval of the general meeting, 31

candidates to fill management body vacancies (in doing so the nomination committee shall decide on a target for the representation of the underrepresented gender in the management body and prepare a policy on how to increase the number of the underrepresented gender in the management body in order to meet that target); (b) periodically, and at least annually, assess the structure, size, composition, and performance of the board, and make recommendations to the management body with regard to any changes; (c) periodically, and at least annually, assess the knowledge, skills, and experience of directors and of the board collectively, and report to it accordingly; (d) periodically review the policy of the board for selection and appointment of senior management and make recommendations to the board, MiFID II, Article 45(4). 43 The following paragraphs of Article 48 further specify the requirements for systems resilience and circuit breakers in electronic trading. 44 The consultation phase on the introduction of the new OTF classification was massive in scale (some 4,200 responses were received by the EU Commission) and was generally hostile, both from the trading/platform/exchange and the OTC sectors. Criticism from the OTC sector, at risk of being pulled into the classification, included the classification’s breadth and lack of clarity, the failure to exclude traditional broker dealing activity, the focus on current venue types rather than one core functionality, the danger of a proliferation of different OTF venues, the risk to flexibility, the potential risk to bilateral trading, the risk of unintended consequences, and the dangers of disproportionality, as the OTC sector was already regulated under the MiFID investment firm regime. Trading platforms/exchange sector concerns included whether a new classification—subject to a lighter regime—was an appropriate means of dealing with OTC trading, and why using existing classification to capture trading with the same functionality was not taken into account. 45 An OTF is a multilateral system which is not a regulated market or an MTF and in which multiple third-part buying and selling interests in bonds, structured finance products, emission allowances, or derivatives are able to interact in the system in a way that results in a contract in accordance with Title II of the Directive, MiFID II, Article 4(1)(23). 46 MiFIR, Recital 7. OTFs include: broker crossing systems, which can be described as internal electronic matching systems operated by an investment firm which execute client orders against other client orders. The new category also encompasses systems eligible for trading clearing-eligible and sufficiently liquid derivatives. It shall not include facilities where there is no genuine trade execution or arranging taking place in the system, such as bulletin boards used for advertising buying and selling interests, other entities aggregating or pooling potential buying or selling interests, electronic post-trade confirmation services, or portfolio compression, which reduces non-market risks in existing derivatives portfolios without changing the market risk of the portfolios. 47

An MTF is ‘a multilateral system, operated by an investment firm or a market operator, which brings together multiple third-party buying and selling interests in financial

instruments—in the system and in accordance with non-discretionary rules—in a way that results in a contract in accordance with Title II of this directive’, MiFID II, Article 4(1) (22). 48 MiFID II, Article 5(1) and (2). The first paragraph includes the general requirement for authorization of investment services. The second paragraph specifies: ‘By way of derogation from paragraph 1, Member States shall authorise any market operator to operate an MTF or an OTF, subject to the prior verification of their compliance with this Chapter’. 49 MiFID II, Article 20(7), (2). 50 MiFID II, Article 9. 51 The Capital Requirement Directive (CRD IV)—Directive on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, 2013/36/EU, OJ 2013 L 176/338. 52 CRD IV, Article 88(1) further specifying that those arrangements should comply with principles relating to the management body’s responsibility for strategic objectives, risk strategy, and internal governance; integrity of accounting and financial reporting systems; process of disclosure and communications; and effective oversight of senior management. Furthermore, the chairman of the management body must not exercise simultaneously the functions of a chief executive officer, unless justified by the institution and authorized by competent authorities. 53 CRD IV, Article 88(2), further specifying the tasks of the nomination committee. 54 CRD IV, Article 91(1). Moreover, they should act with honesty, integrity, and independence of mind to effectively monitor management decision-making: Article 91(2). 55 MiFID II, Article 10. 56 See MiFID II, Article 19(1), making reference to Article 16 on the organizational requirements of investment firms in general. 57 MiFID II, Article 18(1)–(4). 58 MiFID II, Article 18(5), requiring compliance with Articles 48 and 49. 59 MiFID II, Article 19(1). Para. 2 of this Article further requires that the rules governing access to an MTF comply with the requirements established under Article 53(3) for RMs. 60 MiFID II, Article 19(3). 61 MiFID II, Article 19(5). 62 MiFID II, Article 20(1). 63 MiFID II, Article 20(2) making reference to bonds, structured finance products, emission allowances, and certain derivatives. For the definition of ‘matched principal trading’ see Article 4(1), no. 38: a transaction where the facilitator interposes itself between the buyer and the seller to the transaction in such a way that it is never exposed to market risk throughout the execution of the transaction, with both sides executed simultaneously, and where the transaction is concluded at a price where the facilitator makes no profit or loss, other than a previously disclosed commission, fee or charge for the transaction.

64

MiFID II, Article 20(3). MiFID II, Article 20(4), further specifying what follows: ‘An OTF shall not connect with a systematic internaliser in a way which enables orders in an OTF and orders or quotes in a systematic internaliser to interact. An OTF shall not connect with another OTF in a way which enables orders in different OTFs to interact’. 66 MiFID II, Article 20(6). 67 Title V of MiFIR (Articles 28–34) sets the rule for mandatory trading of eligible derivatives on trading venues. 68 MiFIR, Article 35(1). 69 MiFIR, Article 36(1); ESMA, Final Report Draft Regulatory and Implementing Technical Standards MiFID II/MiFIR, 28 September 2015, ESMA/2015/1464 pp. 275–96. 70 EMIR in its Title V touches on another important aspect of the post-trading industry and how it might affect the cost of transacting. Title V discusses ‘Interoperability arrangements’—those between two or more CCPs that involve a cross-system execution of transactions, see EMIR Article 2(12). An interoperability arrangements kicks in when in a trade, the two parties choose two different CCPs to clear the transaction. The interoperability agreement links the two CCPs that ‘interoperate in managing the risk and settlement obligations arising between the parties. Interoperability is the key to archive more competition and reduce clearing fees in the post-trading services. 71 MiFIR, Recital 40; EMIR, Recital 34. 72 For a comprehensive and comparative analysis of the evolution of financial exchanges, see Francis A. Lee, Financial Exchanges: A Comparative Approach (New York: Routledge, 2012). 73 The only example in Europe of an exchange group still structured as an unlisted public limited company owned by its members and users is the Swiss Exchange Group SIX Group Ltd, see . 74 Investment Services in the Securities Field Directive 1993/22/EEC OJ L 141/27. 75 MiFID I. 76 MiFID II. 77 See Lee (n. 24). 78 As of August 2016 there are 102 authorized RMs in the EU and 150 MTFs—no OTFs have been yet authorized in the EU. See ESMA database of RMs, MTFs, SIs, and CCPs. . 79 . 80 . 81 . 82 . 83 . 65

84

. 85

The LSE Groups owns circa 57 per cent of LCH.Clearnet Group Ltd, with the remainder owned by its users; see . 86 . 87 . 88 Lee (n. 72),170–7; 189–94. 89 . 90 See LSEG and Deutsche Börse, ‘Merger of Equals: Creating a Leading Europe-Based Global Markets Infrastructure Group’ (16 March 2016) available at ; Philip Stafford, ‘Market Casts Eye over Deutsche Börse–LSE Tie-Up Details’ (16 March 2016) Financial Times, available at ; Philip Stafford and Arash Massoudi, ‘London Stock Exchange and Deutsche Börse Unveil Deal’ (16 March 2016) Financial Times, available at . Both LSEG and Deutsche Börse have clarified that the Brexit vote will not derail their deal; however, German regulators and politicians have expressed concerns about the plan to move the new group’s headquarters to London: James Shotter and Philip Stafford, ‘German Regulator Casts Doubts on LSE Deutsche Börse Merger’ (28 June 2016), Financial Times, available at . 91 See London Stock Exchange Group plc, Annual Report December 2015, available at . 92 ibid., 8–9. 93 ibid., 8–9. 94 . The remaining 13 per cent is held by the Qatar Investment Authority (10%) and other investors (3%). 95 London Stock Exchange Group plc (n. 91), 12. 96 . 97 . 98 . 99 The members of the consortium are Euroclear, BNP Paribas, BNP Paribas Fortis, Société Générale, Caisse des Dépôts, BPI France, ABN Amro, ASR, Banco Espirito Santo, Banco BPI, and Belgian holding public company SFPI (Société Fédérale de Participations

et d’Investissement). As of 31 December 2015, 65.76 per cent of the capital was publicly traded: see . 100 Euronext, 2014, A Year in Review, available at https://www.euronext.com/en/abouteuronext, p12. 101 ibid., 17; and . 102 . 103 See . 104 ibid. 105 Together with Clearstream, Iberclear launched the joint venture Regist-TR, a company incorporated in Luxembourg, which provides EMIR compliance trade reporting services to a wide range of European customers. 106 Nasdaq Nordic Ltd owns the Copenhagen, Stockholm, Helsinki, Oslo, Reykjavik, Tallinn, Riga, and Vilnius Stock Exchanges; see . 107 See . 108 See . 109 See . 110 BATS Global Markets’ owners are Bank of America Merrill Lynch, Citadel, Citigroup, Credit Suisse, Deutsche Bank, Goldman Sachs, Instinet, JPMorgan, KCG Holdings, Lime Brokerage, Morgan Stanley, Spectrum Equity, TA Associates, Tradebot Systems, and Wedbush; see . 111 See . 112 See . 113 See . 114 MiFID I, Title II, introducing the new MTF; and Article 34, opening up the clearing services industry to competition by regulating the access to clearing services. 115 EMIR, Article 7, on access to CCPs; and Title V on interoperability arrangements. 116 G20, ‘G20 Leaders’ Statement: The Pittsburgh Summit (2009)’, available at . 117 FSB, Implementing OTC Derivatives Market Reforms (25 October 2010), available at . 118 The Commission did not provide specific justifications on why it opted to introduce a new trading venue into the landscape rather than pursuing a restructuring of the existing trading facilities (RMs, MTFs, and SIs). The main rationales were the need to capture new trading facilities, reduce regulatory arbitrage, and respond to technological innovation; see European Commission, Public Consultation, Review of the Market in Financial Instruments Directive (MiFID) (2010). See Ferrarini and Moloney (n. 7), 584; 591–3 (also presenting the case of the Ferber Report that questioned the utility of the OTF classification

and focused on the existing classification between multilateral trading (MTF, RM) and bilateral trading (SI) as the regulatory path to be followed). 119 MiFID II, Article 20(7). 120 ISDA (the International Swaps and Derivatives Association), however, harboured some reservations on the promotion of multilateral trading systems above bilateral trading systems; see ISDA, MiFID/MiFIR: The OTF and SI regime for OTC derivatives (April 2012), available at . 121 Dodd–Frank Wall Street Reform and Consumer Protection Act, Pub L No 111-203, 124 Stat 1376 (2010), § 733. 122 See List of Trading Organizations—Swap Execution Facilities (SEFs), available at . 123 MiFID II, Article 18. 124 MiFID II, Articles 18 and 47. 125 MiFID II, Article 18(11); see ESMA (n. 69), 314–16. 126 MiFID II, Article 10(1); Article 46(2). 127 MiFID II, Article 46(1). 128 MiFID II, Article 10(1). 129 See Section II.5; EMIR, Articles 7–8, MiFID II, Articles 36–38, and MiFIR, Articles 35–36. 130 A tangential aspect to the access to CCPs and trading venues is the interoperability agreement between CCPs, namely the arrangement between two or more CCPs that involves a cross-system of execution of transactions; see EMIR, Articles 51–54. 131 MiFID II, Article 79(2), regulating the establishment of proportionate cooperation arrangements between competent authorities of trading venues operating cross-border and whose operations are of substantial importance to the functioning of the securities market and the protection of the investors of different Member States. For the implementing technical standards to establish standard form templates and procedures for the cooperation agreements, see ESMA, ESMA, Final Report Draft Regulatory and Implementing Technical Standards MiFID II, 11 December 2015, ESMA/2015/1858, pp. 7–11, 34–56. 132 See EMIR, Recital 52; EMIR also recognizes the role of colleges as mechanisms to ‘avoid divergent national measures or practices and obstacles to the proper functioning of the internal market’. 133 ibid., Article 18 (on the composition and role of colleges). 134 Directive 2002/87/EC of the European Parliament and the Council of 16 December 2002 on the supplementary supervision of credit institutions, insurance undertakings and investment firms in a financial conglomerate and amending Council Directives 73/239/EEC, 79/267/EEC, 92/49/EEC, 92/96/EEC, 93/6/EEC, and 93/22/EEC, and Directives 98/78/EC and 2000/12/EC of the European Parliament and of the Council [2003] OJ L 35/1 as amended by Directive 2005/1/EC of the European Parliament and of the Council of 9 March 2005 [2005] OJ L 79/9;
Directive 2008/25/EC of the European

Parliament and of the Council of 11 March 2008
[2008] OJ L 81/40; Directive 2010/78/EU of the European Parliament and of the Council of 24 November 2010 [2010] OJ L 331/120; Directive 2011/89/EU of the European Parliament and of the Council of 16 November 2011 [2011] OJ L 326/113; Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 [2013] OJ L 176/338. 135 Financial conglomerates are defined as ‘financial groups which provide services and products in different sectors of the financial markets’, ibid., Recital 2. 136 Chapter II of the Directive contains the rules of the supplementary supervision of financial conglomerates: capital adequacy (Article 6); risk concentration (Article 7); intragroup transactions (Article 8); internal control mechanisms and risk management processes (Article 9); stress testing (Article 9b); and ‘coordinator’ (Article 10).

12 EU Financial Governance and Transparency Regulation A Test for the Effectiveness of Post-Crisis Administrative Governance Niamh Moloney

I. Transparency Regulation and EU Financial Governance: Why It Matters II. Transparency Regulation and EU Financial Governance 1. The Purpose of Transparency Regulation 2. The MiFIR Transparency Regime III. MiFIR, Transparency, and Supervisory Administrative Governance IV. MiFIR, Transparency, and Administrative Regulatory Governance V. Conclusion

I. Transparency Regulation and EU Financial Governance: Why It Matters

[12.01] It is axiomatic that, in the wake of the far-reaching crisis-era reforms which took place over 2008–14, the ‘single rule-book’ which now governs EU financial markets has become wider, deeper, more technical, and more complex, and that the EU’s related ascendancy over its Member States with respect to financial market rule-making has become almost total.1 The scale of this recasting of EU financial market regulation is well illustrated by the new trading transparency regime,2 which will apply to EU financial markets once the Markets in Financial Instruments Directive II 2014 and Markets in Financial Instruments Regulation 20143 apply to those markets, in 2019.4 As noted in Section II below, transparency rules govern the disclosures on trading activity which must be made available by those market actors subject to transparency requirements; these actors are typically trading venues, in respect of multilateral trading on organized venues, and banks/investment firms of various types, in respect of bilateral, off-venue trading between counterparties. [12.02] The new trading transparency rules, which are contained in MiFIR and which will accordingly apply without further implementation by the Member States, exemplify the recent crisis-era recasting of EU financial markets regulations in a number of respects. First, they take the form of a Regulation and will apply directly in the Member States, thereby delivering maximum harmonization in this area. National discretion, save with respect to various waivers, has been removed. So too, accordingly, has the need for local consultation on and impact assessment of the new transparency regime;5 there are, accordingly, only very limited corrective mechanisms available at national level (save for the various waivers available) through which regulatory error or unintended consequences can be addressed. Second, and by stark comparison with the transparency regime contained in the precursor Markets in Financial Instruments Directive I 2004 (MiFID I),6 the new transparency rules apply to a significantly wider range of financial instruments and derivatives. In a major change, the new rules apply to nonequity asset classes as well as to the equity asset classes covered by MiFID I.7 In addition, a materially wider range of market actors and venues will be subject to the new rules; in particular, a new form of trading venue classification, the Organized Trading Facility (OTF), is deployed by MiFIR to attach non-equity transparency rules to those trading venues on which

standardized derivatives have traditionally traded.8 Third, the new transparency regime is significantly more detailed and granular than the precursor MiFID I regime. The MiFIR transparency regime will ultimately be composed of a series of components: (i) the Level 1 rules set out in MiFIR9—which contains many examples of the highly detailed legislative rule-making which is somewhat at odds with the high-level, norm-setting quality associated with Level 1 but which is a feature of crisis-era Level-1 rule-making; (ii) an administrative rule-book of vast scale—composed of Level 2 Binding Technical Standards (BTSs), adopted by the Commission but proposed by the European Securities and Markets Authority (ESMA), and also composed of ‘standard’ Level 2 administrative rules, adopted by the Commission and in relation to which ESMA provides Technical Advice; and (iii) a raft of soft Level 3 guidance and similar measures, adopted by ESMA. The Level 2 and Level 3 elements of the transparency rulebook are still under construction, although the end is in sight. Over summer 2016 the Commission began to adopt level 2 administrative rules and BTSs based on ESMA’s earlier BTS proposals and Technical Advice.10 In September 2015 ESMA had presented its lengthy proposals for BTSs on the transparency regime to the Commission.11 Earlier, in December 2014 ESMA had presented the Commission with its extensive Technical Advice on the relevant Level 2 transparency rules.12 Fourth, and finally, the new regime is characterized by a level of technical complexity which can confound all but the most expert. ESMA’s September 2015-proposed BTSs covered, for example, the empirical calculations governing when an interest rate derivative is liquid for the purposes of transparency regulation.13 [12.03] In its design, detail, and breadth, the MiFIR transparency regime exemplifies recurring substantive features of the EU’s crisis-era reforms which have been extensively charted.14 Accordingly, and given in particular that the new regime displays an intensely granular quality and is strongly characterized by immense technical complexity, it generates something of a challenge for the regulatory lawyer. How best to examine the MiFIR transparency regime? [12.04] It is beyond question that, functionally, the new transparency regime is of major importance to the EU financial market given its strong

market-shaping quality. The delay to the application date of MiFID II/MiFIR from 2017 to 2018 was in large part driven by the demands of the related implementation process, including with respect to the construction of the new systems required to ensure market compliance with the new transparency regime. But MiFIR-driven change is not likely to be only mechanical or operational in nature: the MiFIR transparency regime will bring fundamental changes to how trading in the EU takes place. Measures of a market-shaping orientation are, of course, not a new feature of EU trading market regulation. As has been widely discussed in the literature, MiFID I was a classic market-shaping measure in that it sought to change EU market microstructure by using law to liberalize share trading in the EU and to reallocate the benefits of share trading from the major trading venues (which benefited from a ‘concentration’ rule which allowed share trading to be funnelled to the major stock exchanges) and across a wider range of execution venues.15 Like MiFID I, MiFID II/MiFIR also has marketshaping ambitions. But, and reflecting the crisis-era reform context and the related driving concern to increase transparency on market activity generally, it has a more prescriptive orientation, particularly with respect to share trading. Over the MiFID II/MiFIR negotiations, the European Commission and European Parliament sought to reduce the volume of overthe-counter (OTC) share trading, in part as EU transparency rules do not currently (under MiFID I) apply to such trading. This trading can, accordingly, take place ‘in the dark’ and not contribute to price formation.16 While the Council adopted a more liberal approach to OTC share trading over the MiFID II/MiFIR negotiations, it was also concerned to ensure that trading in shares, to the extent possible, took place on open, transparent, and regulated platforms.17 The related political/institutional compromise on share trading, reflected in MiFID II/MiFIR, includes a requirement for all trading in shares to take place on organized venues (MiFID II, Article 25): an investment firm must ensure that the trades it undertakes in shares admitted to trading on a regulated market, or traded on a trading venue, must take place on a regulated market, multilateral trading facility, systematic internalizer, or equivalent third-country venue.18 Only those share trades which are non-systematic, ad hoc, irregular, and infrequent, or are carried out between professional counterparties and do not contribute to price discovery, are exempt from this requirement, which seeks to move standard share trading onto organized trading venues.

[12.05] But the market-shaping effects of MiFIR may also take the form of unintended consequences. Chief among these is the potential generation of regulatory incentives to market-makers to decrease liquidity supply in the non-equity markets. There are accordingly contradictions within the MiFIR/MiFID II regime, as much of the regime is otherwise directed to ensuring that the supply of liquidity from market-makers is appropriately managed by trading venues.19 In particular, the new transparency requirements may increase the market impact/position risks carried by those supplying liquidity to the non-equity markets, increase their costs, and thereby create incentives to reduce dealing/liquidity supply activities. The new transparency rules are therefore increasingly being associated with the creation of potentially significant and unpredictable market-shaping effects deriving from their potential to contract trading across a wide range of asset classes previously not subject to transparency regulation, and to generate, as a result, related risks to market liquidity (see Section II below on the interaction between liquidity and transparency regulation).20 The new bond market transparency rules, for example, may have unforeseen effects arising from the uncertain nature of their interaction with the wider market restructuring which is currently reshaping bond market trading (this reshaping is being driven by a range of factors, including the higher capital charges being imposed on market-making activities and the related movement by market-makers from principal to agency trading) and which is being associated with a thinning of bond market liquidity and with greater volatility.21 A concern for bond market liquidity is beginning to seep into EU financial markets policy. The Commission’s September 2015 Capital Markets Union Action Plan, which is currently framing capital market policy development in the EU, notes market concern in relation to the risks to liquidity in secondary bond market trading, and the related risks to the EU economy arising from any related future contraction in the primary issuance market and in higher borrowing costs for firms. Although the Commission has asserted that the new MiFIR transparency regime should increase the attractiveness of the EU capital market, it has also committed to monitoring developments in this area.22 MiFIR itself acknowledges the uncertainties associated with the new regime and supports related review: Article 52 requires that the Commission, after consulting ESMA, report to the European Parliament and Council on the impact in practice of the new transparency regime.

[12.06] The MiFIR transparency regime accordingly illustrates the extent of the market-shaping ambition of current EU financial market regulation, as well as its potential to operate as an agent of unforeseen effects. EU financial market regulation has, of course, long been (and continues to be) concerned with facilitative market-making/construction and liberalization, as is clear from the Capital Markets Union agenda.23 But the marketshaping potential of EU financial market regulation is now immense. MiFIR forms part of a regulatory continuum which includes the 2013 Capital Requirements Directive IV/Capital Requirements Regulation,24 which is reshaping banks’ business models as it impacts on bank balance sheets and lending practices,25 and the 2012 European Market Infrastructure Regulation,26 which is restructuring the organization of derivatives markets in the EU. Empirical observation of the impact of MiFIR will hence be of acute importance, and the MiFIR transparency regime will similarly provide a rich data source for analyses of a functionalist/law and finance orientation. The extensive scholarship which has followed from empirical observation of the impact of the MiFID I equity market transparency regime on equity markets in the EU27 suggests that MiFIR will generate a powerful case study for examination of the relationship between law and markets. [12.07] From a more institutionalist perspective, the MiFIR transparency regime exposes the persistence of the deep-rooted political tensions which have long accompanied the EU single financial market project, and how political preferences continue to shape EU financial governance.28 At present, these tensions and preferences can primarily be associated with the uneasy relationship between single market and euro-area financial governance; a relationship which is now highly unstable and unpredictable in the wake of the 23 June 2016 ‘Brexit’ decision in the UK.29 But the current euro area/single market tensions—however they evolve following the removal of the UK from the single market/euro area relationship—are simply the most recent expression of persistent and entrenched political differences across the Member States as to how the single financial market should be governed and of the related institutional divergences which shape these differences.30 In some respects, and particularly at Level 2, the MiFIR transparency regime is almost scientific in its dependence on empirical data and assessment. In others, however, and particularly at Level 1, it represents

a classic EU ‘hodge podge’ of political compromises. In particular, the transparency negotiations witnessed serious clashes between those Member States more supportive of facilitating market preferences and concerned as to the market-shaping effects of extending transparency requirements and those Member States concerned to extend transparency requirements and to ensure that as much trading as possible takes place on organized, transparent venues. The clashes have left their mark on the waivers which are available from the transparency regime, and which became the battleground for both positions. Complex and often peculiarly granular in their design, the waivers represent the means through which compromise was achieved and particular national preferences in relation to market microstructure protected. [12.08] The focus of this short critique, however, is on the administrative governance implications of the new transparency regime. First, although the MiFIR transparency regime certainly exemplifies the extent to which EU financial market regulation has become centralized and how the notion of a ‘single rule-book’ has become embedded within EU financial system governance, MiFIR also deploys administrative supervisory governance strategies to entrench regulatory governance. Second, and with respect to regulatory governance, the MiFIR transparency regime exposes the extent to which, some eight years on from the financial crisis, the administrative governance system which supports the EU financial system may come under pressure with respect to rule-making. In particular, the MiFIR transparency regime generates important but potentially intractable challenges with respect to the ability of the current administrative process to revise and suspend rules nimbly. [12.09] First, however, this chapter considers the nature of transparency regulation, the distinctiveness of the EU’s approach to transparency, and the main features of the MiFIR transparency regime.

II. Transparency Regulation and EU Financial Governance

1. The Purpose of Transparency Regulation [12.10] Transparency regulation governs the mandatory disclosure of the price, volume, and transaction information produced by trading venues and, under certain conditions, from bilateral trades between trading counterparties, and the availability of such disclosures to the market on a real-time basis. These disclosures, particularly in the equity markets, support price formation and, thereby, liquidity. But they also perform a number of related functions. In a transparent marketplace, potential traders can see all the orders entering the market and the transactions already completed, and can accordingly monitor the execution process.31 Transparency rules can also address fragmentation risk (which arises where trading in an instrument splits across multiple venues) as they ‘tie together’ execution data from different venues and support price formation, the pooling of liquidity, and the achievement of best execution. Transparency requirements have, in addition, a supervisory dimension: they support supervisors in monitoring the nature of trading and in detecting emerging risks, including with respect to market abuse, and, as they allow supervisors to monitor liquidity levels, market stability.32 [12.11] Transparency rules are a long-standing element of trading/tradingvenue regulation. They have been most strongly associated with the equity markets and, accordingly, with the price formation process and with the support of resource allocation: deep secondary market liquidity is associated with reductions in the cost of capital in the primary issuance markets.33 They are increasingly, however, becoming associated with the support of financial stability. The financial crisis reform agenda, reflecting the benefits which flow from the standardization of bespoke riskmanagement instruments,34 has led to market infrastructures being deployed as regulatory proxies to achieve financial stability35 and has also seen transparency rules become a device for supporting stability in the bond and derivatives markets, and a means for managing risk. [12.12] But there is a trade-off between transparency requirements and liquidity. Transparency requirements are acutely sensitive to the different trading functionalities which characterize trading venues,36 and can prejudice liquidity in some circumstances. In particular, in quote-driven or

dealer trading venues, in which dealers take on principal risk, trading reacts to action by dealers. The publication of trading information accordingly generates liquidity risks for dealers, as the disclosure of an open position can lead to strategic behaviour by other traders and expose the dealer to market impact/position risk. The liquidity risks can be acute with pre-trade transparency disclosures. From the trader perspective, pre-trade transparency rules carry market impact risks, particularly for large orders, as the market may move against the order as it is executed (a large sell order may drive the market down). Multilateral trading venues do not, however, carry direct risk from pre-trade transparency rules, as these platforms facilitate the interaction of different orders and do not trade directly or put their capital at risk. They face, however, risks to their business model should they be unable to provide some degree of dark (undisclosed) trading to meet traders’ needs. As is the case with MiFIR, regulatory systems typically provide regulatory waivers from transparency requirements and allow, to some extent at least, ‘dark’ trading. But dealerbased venues and bilateral trading relationships (where they are subject to pre-trade transparency requirements) face sharper risks. Where dealers execute orders against their proprietary instruments or capital and thereby take on principal risk, they become subject to market impact risk (as their trading position is exposed to the market and their capital is at risk). Dealers’ positions could be systematically undermined and it could become uneconomic for them to offer execution services. Liquidity could accordingly suffer should dealers become less willing to take large positions. The current global regulatory concern as to the volatile nature of bond market liquidity37 underscores the risks which can flow from unintended consequences arising from transparency rules. [12.13] Transparency requirements are also sensitive to the type of instrument traded and subject to transparency rules. The equity markets are typically associated with multilateral, order-book trading, with high levels of trading activity and liquidity, and accordingly with significantly lower risks in terms of pre-trade transparency. Bond markets, however, operate with different transparency and price formation dynamics.38 Bond markets are considerably less liquid than the equity markets and experience much thinner trading patterns. Trading is typically concentrated at the time of issue and over the months immediately following as the issue is re-

distributed; trading typically then thins very significantly until the period shortly before the issue matures, when taxation and other drivers can prompt stronger trading. Dealer-based trading models tend to dominate, although electronic multi-dealer platforms have developed, particularly in the sovereign and corporate bond markets. Traders are typically professional. Pricing is affected by a range of factors beyond trading information, including macroeconomic conditions (particularly for government bonds) and credit risk. Transparency is not a guarantee of liquidity; price transparency does not always indicate the availability of counterparties available to trade at the indicated price. Transplantation of the equity market transparency model can therefore generate liquidity risks, given the potential risks to dealers.39 But liquidity risks are not distinct to the bond markets; trading in some equities can also be highly illiquid, particularly with respect to SME (small and medium-sized enterprise) shares and the shares of small capitalization (‘small cap’) issuers.

2. The MiFIR Transparency Regime A. Approach [12.14] By contrast with MiFID I, which limited transparency requirements to the equity markets and which contained extensive exemptions and waivers, MiFIR adopts a maximalist approach to transparency. It imposes detailed pre- and post-trade transparency rules on organized trading venues (multilateral trading) and on certain forms of bilateral trading between counterparties (pre-trade transparency rules apply to certain forms of bilateral trading by ‘systematic internalizers’; post-trade transparency rules apply to all forms of bilateral trading between counterparties). MiFIR also applies transparency rules to a wide range of equity and non-equity asset classes. [12.15] The most extensive transparency requirements apply to the three forms of ‘trading venue’ for multilateral trading which are established under the MiFID II/MiFIR venue classification system: regulated markets (RMs) (MiFID II, Article 4(1)(21)); multilateral trading facilities (MTFs) (MiFID II, Article 4(1)(22)); and organized trading facilities (OTFs) (MiFID II, Article 4(1)(23)). RMs, MTFs, and OTFs are all different forms

of MiFID II/MiFIR organized multilateral venues on which third-party buying and selling interests in instruments interact in a manner which results in a contract. The major point of distinction in the classifications arises between MTFs/RMs and OTFs (MTFs and RMs are functionally almost identical for the purposes of trading/trading venue regulation). By contrast with RMs and MTFs, OTFs may not admit equity securities to trading – they can only support trading in bonds, structured finance products, emission allowances, or derivatives. Also by contrast with RMs and MTFs, trading on OTFs is discretionary in that the venue can intervene to support trading; this is not permissible for MTFs or RMs which are required to apply non-discretionary rules to trading (for this reason, OTFs are not permitted to trade equity securities, as the application of a discretionary trading process would lead to poor quality transparency disclosures and related blurred signals as to the pricing sentiment for equity securities). Distinct rules accordingly apply to OTFs to protect trading against the risks of discretionary intervention by the venue (MiFID II, Article 20). The particularities of the OTF trading venue classification reflect its distinct role. The classification is designed to capture the features of the specialized multilateral venues (typically dealer-based) which trade derivatives—in the derivatives segment venues are typically empowered to intervene to support the stability of the venue by providing liquidity, given that liquidity in derivative instruments can be volatile—and so to implement the G20 commitment that trading of standardized OTC derivatives move to ‘exchanges’ or ‘electronic trading platforms’ (the derivative trading obligation is set out in MiFIR, Articles 28–34).40 RMs, MTFs, and OTFs are all, however, subject to the same set of transparency rules. [12.16] Bilateral/OTC trading between counterparties is subject only to post-trade transparency requirements under MiFIR. This design decision reflects the need to protect the pre-trade position of counterparties who trade bilaterally and take proprietary positions. By contrast, multilateral trading venues do not take proprietary positions but support the interaction of third-party trading interests. MiFIR reflects, however, the blurring of functions between bilateral and multilateral functions which can arise where bilateral trading is heavily based on systems, by applying a distinct pretrade transparency regime to ‘systematic internalizers’ (SIs). An SI is an

investment firm which, on an organized, frequent, systematic, and substantial basis, deals on own account (bilaterally) by executing client orders outside an RM, MTF, or OTF, and without operating a multilateral system (MiFID II, Article 4(1)(20)). The distinct pre-trade transparency treatment for SIs, despite the bilateral functionality of SI trading, is a legacy from MiFID I. MiFID I was primarily focused on share trading and was, as noted above, a classic market-shaping measure. It repealed the previous ‘concentration’ rule which allowed Member States to require that share trading take place on the major trading venue in each Member State, and so brought different forms of execution venue into competition, notably OTC brokers and multilateral, organized trading venues.41 The related MiFID I regulatory scheme (which will apply until 2018) reflects the very bitter negotiations between the Member States as to how a regulatory level playing field could be constructed for the different forms of competing trading venues. In particular, the off-venue, bilateral OTC brokerage segment was fiercely opposed to being subject to the same transparency requirements as applied to multilateral trading venues, given that bilateral venues traded on their own capital and faced significant market risk from exposing their trading positions. The venue segment, however, argued that significant volumes of trading were passing through the OTC segment, that the automation and systematization of bilateral trading functions in the OTC sector could lead to systems which were very close in practice to the functionality of trading venues and which did not generate market risk for OTC brokers, and that a competitive advantage would be conferred on the OTC segment were it not regulated in the same manner as organized trading venues. The complicated compromise, which reflected sharp political interests, included distinct pre-trade transparency treatment for investment firms which executed client orders systematically, including by means of distinct proprietary systems. [12.17] Overall, MiFIR’s regulatory design, while not always entirely clear given the influence of political compromise, has been shaped by a driving concern to protect liquidity, particularly in non-equity asset classes, given the uncertain impact of MiFIR. As is noted in the following paragraphs, exemptions, waivers, suspensions, and calibrations, designed to protect liquidity, are a recurring feature of the new transparency regime. Instruments for which there is not a liquid market, for example, benefit

from a series of exemptions. Whether or not these mechanisms, which are the product of difficult political and institutional negotiations and of compromise, will prove to be operationally effective in protecting liquidity is in large part dependent on ESMA’s ability to design empirically robust administrative rules. The Level 1 mechanisms designed to protect liquidity are to be amplified by highly detailed administrative rules, either proposed by ESMA (where they take the form of BTSs) or in relation to which ESMA provides the Commission with Technical Advice (where they take the form of standard administrative rules). ESMA is also responsible for the large number of complex technical calculations required to determine liquidity levels in different asset classes on an ongoing basis.

B. Transparency Requirements: Trading Venues [12.18] The pre-trade transparency regime for equity and equity-like instruments, and for RMs, MTFs, and OTFs, requires that the market operator/investment firm operating the venue in question make public current bid and offer prices and the depth of trading interest at those prices which are advertised through their systems for shares, depositary receipts, exchange-traded funds (ETFs), certificates, and similar financial instruments (MiFIR, Article 3(1)). By contrast, MiFID I applies only to shares admitted to trading on a regulated market. The exact nature of the transparency obligation in practice is tailored to reflect the distinct trading functionality the venue operates (including order-book, quote-based, hybrid, and periodic auction systems) (MiFIR, Article 3(2)). The detail of the transparency regime will be set out in Level 2 administrative rules to be adopted by the Commission. [12.19] The pre-trade equity/equity-like transparency regime for these organized trading venues is subject to a series of waivers designed to allow venues to support ‘dark’ trading, given the efficiencies which dark trading can bring, notably for institutional investors and for market liquidity generally, particularly where the trades in question are vulnerable to market impact/liquidity risk and where the danger of price formation in the equity markets being damaged can be reduced or controlled. Four waivers are available (MiFIR, Article 4): the ‘reference price’ waiver, which is

available for (in effect) passive systems which match orders on the basis of a price derived from another trading venue but which do not engage in price discovery; the ‘negotiated trade’ waiver, which is available for systems which formalize negotiated transactions which are made within specified price parameters and which are designed to deal with special orders; the ‘large-in-scale’ waiver, which is available for large orders which are at risk of price impact; and the ‘iceberg’ orders waiver, which is available for orders which are held in an order-management facility before being disclosed to the market. The operational detail of these waivers will be governed by Level 2 administrative rules. While all of these waivers are currently available under MiFID I, their scope has been restricted under MiFIR, reflecting a pre-crisis and crisis-era concern as to the scale of ‘dark’ equity trading and the potential risk to price formation, as well as significant Member State concern. This concern was, however, concentrated in certain Member States only, and so led to difficult Council negotiations given the concern of other, more facilitative Member States to support dark trading, particularly by institutions, in order to protect market liquidity. [12.20] The complex negotiations within the Council led to detailed conditions being imposed on the reference price waiver and on the negotiated trades waiver; to these two sets of waivers being subject to an overall ‘volume cap’; and to a new ESMA-driven process for the approval of waivers. The conditions imposed on the reference price waiver, for example, include requirements governing the venue from which the reference price is derived, while the negotiated trades waiver is now subject to rules governing the instruments which are eligible (including ‘illiquid’ instruments—these are not subject to the volume cap), as well as to rules governing how ‘negotiated trades’ can be executed and the price set. The new volume cap, which applies to both waivers, requires that the percentage of trading in a financial instrument carried out on a trading venue under these two waivers must be limited to 4 per cent of the total volume of trading in that financial instrument on all trading venues across the EU over the previous twelve-month period (MiFIR, Article 5). In addition, overall EU trading in the financial instrument and carried out under the two waivers must be limited to 8 per cent of the total volume of trading in the instrument on all trading venues across the EU over the previous twelvemonth period (MiFIR, Article 5). As outlined in Section III, a new

procedure, led by ESMA, applies to the process under which national regulators can choose to apply certain waivers. [12.21] Post-trade transparency requirements also apply to equity and equity-like trading on RMs, MTFs, and OTFs: these organized trading venues must make public the price, volume, and time of transactions executed in in-scope instruments (MiFIR, Article 6). The detail of this regime will be set out in the related Level 2 regime. Deferrals are available for post-trade transparency requirements in order to address the liquidity and position risks associated also with pre-trade transparency. These are available in particular for large-in-scale transactions which generate significant market impact risk for liquidity providers where a position is being unwound (MiFIR, Article 7). [12.22] In a significant change from MiFID I, organized trading venues are also subject to pre-trade transparency requirements in relation to nonequity asset classes (MiFIR, Article 8). Market operators and investment firms operating a trading venue must make public current bid and offer prices and the depth of trading interest at those prices for bonds, structuredfinance products, emission allowances, and derivatives traded on a trading venue. The transparency requirements do not apply, however, to derivative transactions by non-financial counterparties which are objectively measurable as reducing risks relating to the commercial activity or treasury financing activity of the non-financial counterparty (or its group). The detailed transparency rules will be contained in Level 2 rules and so will reflect ESMA’s technical input. As with the equity regime, the rules will be calibrated to reflect the different types of trading venue in non-equity markets, including order book, quote-driven, hybrid, period auction trading, and voice broking systems (MiFIR, Article 8). And in another reflection of the equity regime, a waiver system applies (which will be governed by detailed Level 2 rules), although with respect to non-equity asset classes the liquidity risk is particularly acute, not least given uncertainties as to the impact the new transparency regime will have on the primarily dealer-based systems which operate in the non-equity markets. [12.23] Waivers are available for three types of order (MiFIR, Article 9): ‘large-in-scale’ orders and orders held in an order-management (‘iceberg’)

facility; orders with respect to actionable indications of interest in requestfor-quote and voice-trading systems, where the orders are above a size specific to the instrument, and where the size would expose liquidity providers to undue risk and takes into account whether the relevant participants are retail or wholesale traders; and orders in relation to derivatives which are not subject to the trading obligation which applies under MiFID II/MiFIR and to other financial instruments for which there is not a liquid market. By contrast with the equity waivers, some element of transparency is required in relation to certain waivers. Where a waiver is granted in respect of request-for-quote and voice-trading systems, the market operator or investment firm must make public at least indicative pretrade bid and offer prices which are close to the price of the trading interests advertised through the relevant system (MiFIR, Article 8). An ESMA-led waiver-approval process applies to these waivers, as it does for the equity waivers (see Section III). [12.24] In an indication of the scale of the liquidity risks attendant on the new non-equity market regime, a widely cast suspension power is available. In effect, national supervisors can suspend the non-equity market transparency requirements where liquidity is compromised (MiFIR, Article 9(4)). The national supervisor responsible for one or more trading venues on which a class of bond, structured-finance product, emission allowance, or derivative is traded may, where the liquidity of the asset class in question falls below a ‘specified threshold’, temporarily suspend pre-trade transparency requirements. The suspension is valid for an initial period not exceeding three months and may be renewed for further three-month periods if the conditions for the suspension continue to be met. The operationally critical ‘specified threshold’ is to be based on objective criteria, specific to the market for the asset class concerned, and will be governed by administrative BTSs at Level 2. In addition, the supervisor must notify ESMA before it takes action, and ESMA must adopt an opinion on whether the suspension is justified. [12.25] Post-trade transparency requirements also apply in the non-equity markets. Market operators and investment firms operating a trading venue must make public the price, volume, and time of the transactions (MiFIR, Article 10). As is the case with the equity regime, deferred publication is

permitted, subject to the conditions which apply, and is designed to address transactions which are large in size and transactions in illiquid instruments (MiFIR, Article 11); national supervisors may also stagger the disclosures required over a period of time and permit aggregated disclosures. In addition, national supervisors may suspend the post-trade transparency requirements where liquidity falls below a specified threshold.

C. Transparency Requirements: Bilateral Trading [12.26] Outside RMs, MTFs, and OTFs, all bilateral/OTC trading by investment firms, either on own account or on behalf of clients, is subject only to post-trade transparency requirements. Investment firms which conclude transactions in shares, depositary receipts, ETFs, certificates, and other similar financial instruments traded on a trading venue must make public the volume and price of those transactions and the time at which they are concluded. This regime maps that which applies to trading venues; a similar deferral regime also applies (MiFIR, Article 20). Similarly, with respect to non-equity asset classes, all investment firms which, either on own account or on behalf of clients, conclude transactions in bonds, structured-finance products, emission allowances, and derivatives traded on a trading venue must make public the volume and price of those transactions and the time at which they were concluded; deferrals are available (MiFIR, Article 21). [12.27] A distinct and complex regime (governing pre-trade transparency only) applies to a particular segment of the bilateral/OTC market and with respect to equity, equity-like, and non-equity asset classes: the SI segment —and in relation to the instruments in which an SI declares itself to be an SI (the obligation is instrument specific but a firm must declare as an SI where, in relation to the instrument in question, it meets the features of an SI) (MiFIR, Articles 14–15 and 18). The pre-trade equity/equity-like transparency regime for SIs is similar (albeit with some nuances) to the MiFID I regime for SIs; the non-equity pre-trade transparency regime is entirely new, although it reflects the main design features of the non-equity transparency regime for trading venues.

[12.28] With respect to equity/equity-like transparency, the main requirement imposed on SIs is to publish firm quotes, with respect to shares, depositary receipts, ETFs, certificates, and other similar instruments (for which they are SIs) (MiFIR, Article 14(1)). In order to protect SIs against position risk (as they trade bilaterally), and reflecting the design of the 2004 MiFID I regime, the obligation applies only in relation to instruments where there is a liquid market in those instruments (MiFIR, Article 14(1)). Where the market in the instrument in question is not liquid, the SI is to provide quotes on request to clients. In a further protection for SIs, the pre-trade transparency obligation applies only where the SI is dealing in sizes up to ‘standard market size’ (MiFIR, Article 14(2)).42 Where an SI chooses to deal only in sizes above the standard market size, it will not be required to disclose its quotes pre-trade. While an SI can decide the sizes for which it will quote, the minimum quote size is subject to conditions: it must be at least the equivalent of 10 per cent of the standard market size of the instrument in question (MiFIR, Article 14(3)). Each quote, for a particular instrument, must include a firm bid and offer price for a size or sizes up to the standard market size for the class of instrument in question; the quote must also reflect ‘prevailing market conditions’ for the instrument in question (MiFIR, Article 14(3)). Quotes must be made public on a regular and continuous basis during normal trading hours: they may be updated at any time and may be withdrawn where exceptional market conditions exist (MiFIR, Article 15(1)). [12.29] A series of obligations governs how SIs must publish and execute their quotes. These include that SIs are required to execute orders received from clients in relation to financial instruments for which they are SIs at the quoted prices at the time of the reception of the order. In justified cases, SIs can execute those orders at a better price than the quoted prices, as long as the price falls within a public range close to market conditions (MiFIR, Article 15(2)). An SI can also offer price improvement to professional clients without being subject to these requirements where, inter alia, the order is subject to conditions other than the market price (MiFIR, Article 15(3)). These rules governing changes to quotes (or price improvement by the SI) are designed to reflect the risks which SIs face as bilateral traders and how they are managed. In particular, in order to protect their trading positions, SIs can widen their quote spread and so offer price improvement

to clients on the published quote. But while counterparties to the SI (including retail investors) can benefit accordingly from price improvement, price formation can be compromised as the pre-trade transparency data published may not be accurate as regards pricing sentiment. Accordingly, while price improvement is permitted under MiFIR, it is subject to conditions which are designed to mitigate the risks. [12.30] Similarly, SIs are permitted to protect access to their quotes, given the risks to their trading positions. Under MiFIR Article 17, SIs can decide, on the basis of their commercial policy, in an objective, non-discriminatory manner, and in accordance with clear standards, which clients can access their quotes. SIs are not, accordingly, cast as market-makers under MiFIR’s transparency scheme, and can control access to their quotes. An SI may similarly refuse to enter into or discontinue business relationships with clients on the basis of commercial considerations (including in relation to a client’s credit status, counterparty risk, and settlement risk). As with all aspects of the SI regime and the transparency regime more generally, detailed Level 2 rules will govern the details of the regime. [12.31] Finally, SIs (and all MiFID II/MiFIR investment firms) are under an obligation (unless instructed otherwise by the client) to ‘take measures to facilitate’ the earliest possible execution of a ‘client limit order’ (by making the order public immediately in a manner that is easily accessible to other market participants) in respect of shares admitted to trading on a regulated market or traded on a trading venue, where that order is not immediately executed under ‘prevailing market conditions’ (MiFID II, Article 28(2)). This order-routing rule forms part of the matrix of transparency rules in that is designed to ensure that client limit orders, which are particularly informative for the price formation process as they contain a specified price limit (MiFID II, Article 4(1)(14)), are exposed to the market as soon as possible. [12.32] SIs are in addition subject to pre-trade transparency rules governing non-equity asset classes. The non-equity regime is similar in many respects to the SI equity regime, particularly in its concern to protect SIs from position risk, but has a number of distinct features designed to reflect the increased risks to liquidity arising from non-equity asset class

transparency requirements. MiFIR Article 18(1), which is the core obligation, requires an SI to make public firm quotes in bonds, structuredfinance products, emission allowances, and derivatives traded on a trading venue (as long as the firm is an SI for these instruments and there is a liquid market for these instruments) where three conditions are met. The SI must be prompted for a quote by a client; the SI must agree to provide a quote; and the SI must deal in sizes below the size at which the SI, as a liquidity provider, becomes exposed to undue risk.43 With respect to the quotes which must be provided, these must allow the SI to meet its best execution obligations, and also reflect prevailing market conditions in relation to the prices at which transactions are concluded for the same or similar instruments on a trading venue (MiFIR, Article 18(9)). An SI can update its quotes at any time and can also withdraw its quotes under exceptional market conditions (MiFIR, Article 18(3)). As with the equity regime, an access to quotes regime applies. While published quotes must be available to other clients of the SI, SIs are permitted to decide, on the basis of their commercial policy, in an objective, non-discriminatory way and in accordance with clear standards, the clients to whom they give access to their quotes; SIs may also refuse to enter into or discontinue business relations, on the same conditions as apply in relation to the equity regime (MiFIR, Article 18(5)). Price improvement is also permitted. In justified cases, an SI may execute a client order at a better price as long as the price falls within a public range close to market conditions (MiFIR, Article 18(9)). [12.33] Reflecting the more acute liquidity risk which arises in the nonequity markets, the Article 18 quote-publication and access requirements for SIs are disapplied where the instrument in question falls below the liquidity threshold which governs the suspension of pre-trade transparency requirements for non-equity asset classes in the trading venue space (MiFIR, Article 18(6)). Otherwise, an SI must enter into transactions with any client to whom the quote is made available under the published conditions, as long as the quoted size is at or below the size at which the SI, as a liquidity provider, would be exposed to undue risk (MiFIR, Article 18(6)). An SI can also establish (non-discriminatory and transparent) limits on the number of transactions it undertakes to enter into with clients in relation to any given quote (MiFIR, Article 18(7)).

III. MiFIR, Transparency, and Supervisory Administrative Governance [12.34] Once the Level 2 process which governs the adoption of administrative rules completes, the MiFIR transparency rule-book will form a regulatory regime of vast breadth and depth. But operational supervisory decision-making by national supervisors will, in addition, become subject to EU-level controls and processes. [12.35] The waivers which are available from the transparency requirements for equity and non-equity trading can be regarded as the key battleground on which national supervisors (and the Member States) will seek to protect national territory, in that the waivers afford national supervisors the possibility to apply tailored treatment to distinct national trading systems and practices. With respect to the non-equity markets, the general suspension power given to national supervisors to lift transparency requirements where liquidity in non-equity markets is compromised will similarly make available to supervisors a lever of immense operational value in addressing local market risks. But in each case, operational supervisory decision-making, already constrained by the detailed Level 2 rules which will govern the waivers and suspension powers, must take place within an EU-governed process. [12.36] In one of the key changes from MiFID I, ESMA is now empowered to oversee national supervisory decision-making with respect to equity and non-equity pre-trade transparency waivers. A degree of convergence has already been taking place in this area. Under MiFID I, waiver decisions in relation to the equity transparency rules are currently at the discretion of national supervisors, in accordance with the relevant MiFID I conditions. But under voluntary arrangements agreed by national supervisors, proposed supervisory waiver decisions are notified to ESMA which then adopts a common position across its member supervisors on the waiver. While many waivers have proved uncontroversial, in some cases it has proved difficult to reach a common ESMA position, particularly with respect to the reference price and negotiated trade waivers which were controversial over the MiFIR negotiations.44 This informal process has now

been formalized. Before it can grant a pre-trade transparency waiver from the equity market rules under Article 4, a national supervisor must (not less than four months before the waiver is intended to take effect) notify ESMA and the other national supervisors of its proposed waiver decision, and provide an explanation regarding the waiver. Within two months, ESMA must issue a non-binding opinion to the national supervisor, which assesses the waiver’s compatibility with MiFIR. In addition, where the national supervisor grants the waiver and another supervisor disagrees, the granting supervisor may refer the waiver to ESMA, and ESMA may exercise its binding mediation powers (under 2010 ESMA Regulation,45 Article 19). [12.37] The application of the Article 4 equity market waivers is to be monitored by ESMA, who is to report annually to the European Commission on their application in practice; ESMA is also to review all waivers and to issue an opinion to the national supervisors concerned on the continued compatibility of the waivers with MiFIR (Article 4(4) and (7)). [12.38] While ESMA does not have the same set of powers with respect to the deferrals available from the post-trade equity market regime, it must monitor the application of the deferred publication rules and report annually on their application to the European Commission. In addition, where a national supervisor disagrees with a deferral decision by another supervisor, the matter can be referred to ESMA for binding mediation (MiFIR, Article 7). [12.39] The pre-trade transparency waiver regime for non-equity asset classes is subject to the same ESMA-based waiver process as applies to pretrade equity market transparency requirements (MiFIR, Article 9), while the post-trade deferral regime for non-equity asset classes is similarly subject to ESMA monitoring and Commission reporting requirements (MiFIR, Article 10). In addition, and with respect to the general power conferred on national supervisors to suspend the pre-trade transparency regime for non-equity asset classes where liquidity is compromised (MiFIR, Article 9), a national supervisor must notify ESMA before it takes suspensive action and ESMA must issue an opinion on whether the suspension is justified. A similar notification/opinion requirement applies in relation to the general power

given to national supervisors to suspend post-trade transparency requirements when liquidity is compromised (MiFIR, Article 10). [12.40] A national supervisor is not required to follow ESMA’s opinion on a proposed waiver or suspension, whether in relation to the equity or non-equity market transparency rules; ESMA’s opinion is of non-binding quality. During the MiFID II/MiFIR negotiations, some Member States had proposed that ESMA be empowered to take a binding decision on waivers, but this position was not supported by the Council given in particular the concern of other Member States as to the validity of such a power in light of the constraints imposed on agency action by the Meroni doctrine.46 [12.41] A degree of discretion is appropriate and reasonable in this area, not least given the uncertainty associated with the new transparency rules. But concessions designed to ease the application of the new regime, and to provide a safety valve which releases the pressures which unintended consequences can generate, can lead to fragmentation, uneven transparency levels, and legal certainty risks. Lessons in this regard can be drawn from the concerns which proportionality requirements are currently generating across the crisis-era single rule-book. Proportionality devices are deployed across the single rule-book and are typically used to mitigate the risk of rules—particularly entirely new rules—applying in a disproportionate, inefficient, and costly manner. Proportionality devices typically require that rules be applied in a proportionate manner, usually in relation to particular categories of market actor. Proportionality devices can, however, lead to differences in national interpretation and to related transaction cost risks, and are increasingly coming under institutional scrutiny.47 The European Banking Authority, for example, has called for the ambiguities generated by the proportionality requirement applicable to the CRD IV remuneration rules—in particular in relation to whether the proportionality requirement permits the disapplication of certain remuneration rules to certain actors— to be addressed.48 [12.42] The risks inherent in the different waivers, deferrals, and suspensions are mitigated, however, by the conditions which govern their application by national supervisors. Highly detailed Level 2 administrative rules will amplify the conditions under which the equity/non equity waiver,

deferral, and suspension powers will apply. The likely restrictive nature of these rules, reflecting the mandates for administrative rule-making set out in MiFIR, is likely to reduce significantly the risk that the different waiver, deferral, and suspension powers will lead to national supervisors acting to protect national markets, to the prejudice of overall EU market transparency. ESMA has acknowledged market concern that the suspensive power available to national supervisors in respect of the non-equity pre- and post-trade transparency requirements where liquidity is compromised will, in practice, be very limited. In its initial Discussion Paper on the suspensive power, ESMA suggested that this power be used only in exceptional market circumstances, and that the governing liquidity threshold be set in such a way as to avoid unnecessary fluctuations in transparency requirements and to ensure a level playing field across the EU. The conditions which ESMA has proposed concerning the application of the suspensive power by national supervisors are, accordingly, restrictive. ESMA has, however, argued that the conditions reflect the intention of the co-legislators to provide a suspensive instrument which operates only in extraordinary circumstances, and which supports a stable environment and avoids unnecessary fluctuations in transparency requirements.49 [12.43] In parallel, the new supervisory process governing waiver, deferral, and suspension decisions by national supervisors, and the requirement for an ESMA opinion and for ESMA monitoring, provides a further means through which divergences in supervisory approaches in the application of the waiver/suspension regime can be addressed. Much depends, of course, on ESMA’s ability to adopt an opinion which runs counter to the wishes of the supervisor in question, where such an ESMA opinion is warranted. The new waiver and suspension powers are likely to be sensitive50 and ESMA can be expected to tread carefully when reviewing national supervisors’ decisions in this area. There are, however, indications that the capacity of ESMA’s decision-making body, its Board of Supervisors (composed of the national supervisors), to take decisions in a hierarchical manner, and to express an opinion which runs counter to the interests of a member supervisor, is strengthening. Under the 2012 Short Selling Regulation51 ESMA is empowered to provide an opinion on shortselling measures adopted by national supervisors. January 2016 saw ESMA,

for the first time, issue a negative opinion in which it found emergency action by the Greek supervisor not to be appropriate or proportionate.52 [12.44] The new MiFIR transparency regime accordingly fuses centralized supervisory and regulatory governance techniques in order to ensure the new rules are sufficiently nimble and can respond to local market conditions, but that, at the same time, pan-EU transparency is not compromised. The transparency regime does not confer decision-making authority on ESMA. ESMA’s powers are not of the same binding effect as the new product-intervention powers conferred on ESMA under MiFIR, for example. But by providing ESMA with a means for signalling peer supervisor approval or disapproval of local supervisory action, the new procedures governing waivers, deferrals, and suspensions should support supervisory dialogue and learning, and minimize any risk that the regime leads to unevenness in the application of transparency requirements.

IV. MiFIR, Transparency, and Administrative Regulatory Governance [12.45] The MiFIR transparency regime is, notwithstanding the extent to which it has been shaped by long-standing political interests and preferences, heavily based on administrative regulatory governance. The administrative rule-making process which supports EU financial governance, and which relies heavily on ESMA’s technocratic expertise (albeit that rule-adoption power remains with the Commission), is now battle hardened, following the intense administrative rule-making period during which the crisis-era legislative (Level 1) rule-book was amplified, over 2011–14 in particular.53 But the administrative regulatory governance required by the MiFIR transparency regime is of a different order to the administrative regulatory governance required of earlier crisis-era measures. MiFIR demands data-intense, complex, and highly granular calculations and rule design, and requires significant empirical and datagathering/interrogation capacity. It poses a significant challenge to the administrative governance system which supports EU financial governance and which has been primarily preoccupied, until now, with the construction

of relatively standard administrative rules (even if often of a detailed and complex nature). [12.46] Experience so far with the MiFIR transparency regime suggests, however, that the institutional apparatus which supports administrative regulatory governance is operating relatively well, particularly with respect to ESMA’s role. It has become clear that, with ESMA, the EU has available to it the technocratic administrative capacity without which a nuanced and calibrated transparency regime, which mitigates liquidity risks, could not be effectively delivered. This is apparent across two dimensions in particular. [12.47] First, ESMA has played a pivotal role in the construction of the highly detailed administrative rule-book which, informed by extensive datagathering and empirical assessment, is to amplify and calibrate the MiFIR transparency regime. For the most part, the administrative transparency rules which will amplify and calibrate the MiFIR transparency regime are in the form of BTSs and so were to be proposed by ESMA and adopted by the Commission, subject to the Commission’s powers to request revisions and to reject ESMA’s proposals under the 2010 ESMA Regulation which governs the adoption of BTSs. ESMA was charged with, inter alia, developing the highly technical details of the transparency regime; ensuring the administrative rules are appropriately calibrated to the different asset classes subject to the MiFIR transparency regime; ensuring the asset classes deployed are sufficiently granular; designing the operationally critical conditions which will govern the different exemptions, waivers, deferrals, and suspensions; and performing a series of volume/liquidity calculations on which the application of the transparency regime depends. ESMA’s BTS proposals for the MiFIR administrative transparency rules were published in September 2015. They set out, in highly granular detail, the transparency requirements to be imposed on trading in equity and non-equity asset classes, and contained a wealth of supporting detailed analysis and granular empirical evidence. The proposed non-equity market transparency rules, for example, were accompanied by detailed analysis of liquidity in the bond, structured finance products, securitized derivatives, interest rate derivatives, foreign exchange derivatives, credit derivatives, equity derivatives, commodity derivatives, exotic derivatives, contracts for difference, and emission allowances asset classes.54

[12.48] Second, ESMA is to establish (on its own behalf and on behalf of its member supervisors) a very large number of liquidity thresholds and other key indicators relating to the liquidity levels of assets: the outcomes of these calculations will determine how the transparency rules apply in particular circumstances. To take one example, ESMA will be responsible for monitoring the volumes of dark equity trading which govern the important ‘volume cap’ (MiFIR, Article 5) which applies to certain of the waivers which apply to the pre-trade transparency regime for equity markets. In addition, and under the complex and massive ‘FIRDS’ project, ESMA will perform a number of operationally complex liquidity/transparency related calculations, and will be required to gather, interrogate, and assess very substantial volumes of trading data, on behalf of a number of its member supervisors.55 Overall, MiFIR has required ESMA to construct a central system which will collect data on financial instruments from some 300 trading venues across the EU, calculate transparency and liquidity thresholds, and publish the related data in one central location.56 [12.49] ESMA accordingly provides the EU with a significant technical capacity which augurs well for the effective amplification of the MiFIR regime and for the mitigation of the risks to liquidity which the new regime poses. But ESMA was in some respects significantly handicapped in developing its proposals for BTSs and related calculation models and systems. The MiFIR transparency regime for non-equity asset classes is new, and few precedents were available to ESMA. In addition, ESMA was restricted by the dearth of the market/trading data (particularly on nonequity asset classes) on which the effective development of the new transparency regime depends. Data on trading activity and liquidity in the non-equity asset classes is still limited and fragmentary. Reporting to trade repositories on derivatives market trading activity has been very significantly strengthened following the coming into force of the 2012 EMIR which provides for mandatory trade reporting. But trade repository data was only just becoming available as ESMA commenced the process of developing MiFIR administrative rules, and difficulties still abound with respect to the quality of trade repository data.57

[12.50] Nonetheless, initial indications suggest that ESMA surmounted many of the difficulties and that the related administrative governance regime is working relatively well. Certainly, the MiFIR process has benefited from ESMA’s by now extensive experience with rule development, stakeholder engagement, and institutional coalition-building. For example, ESMA’s proposed administrative rules on the MiFIR transparency regime, set out in its September 2015 BTS proposals, emerged from an extensive process of stakeholder consultation, which included detailed initial Discussion and Consultation Papers.58 The ESMA proposals also reflected empirical evidence (to the extent it was available), the application of models designed and constructed by ESMA, and a detailed cost–benefit assessment.59 The proposed administrative rules also reflected an iterative process of engagement between ESMA and the industry. The September 2015 BTS proposals provide extensive evidence on ESMA’s thought process in relation to transparency regulation, and explain ESMA’s response to the extensive stakeholder feedback on its earlier Discussion and Consultation Papers, both generally and with respect to particular asset classes. What emerges is a clear ESMA concern to reflect industry concerns, where sufficiently evidenced, to follow an evidence-based approach, and not to impose preset determinations as to how transparency rules should operate. To take one example, with respect to the determinations as to whether particular non-equity asset classes are liquid (which determinations govern the application of certain exemptions, deferrals, and waivers, as well as how the transparency regime applies), stakeholders expressed concern as to the quality of the trade repository data being used by ESMA to make liquidity assessments, and in particular as to the overly short data collection period of the sample used, and the lack of granularity in the data. While stakeholders acknowledged that ESMA had ‘cleaned’ the data before performing the liquidity calculations, they remained concerned that the data was not of sufficient quality to support the required calculations. Stakeholders also expressed concerns that the level of granularity deployed by ESMA in assessing the liquidity of and transparency requirements for particular asset classes was not high enough and did not sufficiently reflect market practice.60 In response, ESMA reflected some (but not all) of the industry’s concerns, adopting, for example, a more granular and asset-class-specific approach to the assessment of liquidity,61 including with respect to bond liquidity,62 and to

the application of transparency requirements, and providing for the regular reassessment of liquidity thresholds.63 ESMA also made very significant efforts to ‘clean’ trade repository data and to work with key stakeholders in areas where there were particular difficulties with data, although it acknowledged that difficulties remain with the quality of trade repository data.64 [12.51] With respect to institutional relations, ESMA and the Commission have proved themselves adept in delivering informal enhancements to the process through which administrative rules are adopted. In an attempt to minimize the risks of ‘bottlenecks’ emerging, particularly where the Commission identifies legal mandate difficulties, ESMA and the Commission agreed to an ‘early legal review’ process for MiFID II/MiFIR.65 This informal procedural innovation is designed to ensure that legal mandate and other difficulties which may obstruct the adoption by the Commission of an administrative rule proposed or advised by ESMA are flagged at an early stage (the Commission’s prerogative with respect to the rejection, revision, and adoption of BTSs remains unchanged).66 [12.52] Whether or not ESMA’s approach to the administrative rules which will operationalize the new transparency regime was technically robust and has minimized the risk of unintended consequences remains to be seen. But it can be suggested with a reasonable degree of confidence that ESMA’s proposals emerged from an extensive, careful, and empirically driven design process, which involved close engagement with key stakeholders, and that ESMA has shown it could reach an agreed position, across its member supervisors, on a highly complex and often controversial set of proposed BTSs. It can also be suggested that, overall, ESMA’s approach was pragmatic, and not ideological.67 [12.53] As the constitutional location of administrative rule-making power, the Commission was empowered to adopt, revise, or reject ESMA’s September 2015 BTS proposals, in accordance with the procedures set out in the 2010 ESMA Regulation. For the most part, Commission/ESMA relations with respect to the BTS adoption process generally appear to be stable, and a productive dynamic appears to have developed over time. While tensions have arisen occasionally where the Commission has rejected

or revised ESMA’s BTS proposals, these episodes are, by and large, rare.68 The MiFIR process, given its scale and sensitivity had the potential to place considerable pressure on ESMA/Commission relations. In March 2016 ESMA made public its response to a letter from the Commission requesting revisions to ESMA’s September 2015 BTS Proposals. The Commission’s revisions had not at that time been made public, but ESMA’s response suggested that, while the Commission was broadly supportive of its proposals, a number of revisions had been requested with respect to the liquidity assessments required under the regime and other elements of the pre-trade transparency regime for bonds and derivatives;69 the industry response suggested that the Commission was minded to weaken some of ESMA’s proposals, notably by expanding the range of bonds deemed to be illiquid and so to benefit from lighter treatment.70 Since then, however, the Commission’s relevant BTSs, adopted over summer 2016, suggest that most of ESMA’s proposals were accepted, although some amendments were requested by the Commission in relation to ESMA’s proposals with respect to non-equity market transparency, notably with respect to the phase-in of the regime. [12.54] The ability of the European Commission to reject and/or revise BTS proposals which have been through an intensive and extensive ESMA design and consultation process is an integral element of the administrative rule-making process and protects the Commission’s institutional prerogatives. But, and as has frequently been noted by stakeholders,71 revisions and rejections by the Commission risk undermining ESMA’s technical capacity and short-circuiting the extensive rule-development and consultation process which ESMA engages in—particularly as the Commission stage of the BTS adoption process is relatively opaque. The treatment of ESMA’s BTS proposals, however, augurs relatively well for administrative regulatory governance. [12.55] Further enhancements may be required after the endorsement process is complete. It is unlikely that calibrations and reforms to the new transparency regime will not be required, even allowing for the ability of the market to adjust to new regulatory realities. Here, however, administrative regulatory governance may prove deficient. The Commission is required to engage in an assessment of the impact of MiFIR.

But more immediate action may be required in the form of the suspension of particular BTSs—on a pan-EU basis—where serious unintended consequences arise, particularly as regulatory error in the construction of the new regime, or regulatory reliance on poor quality empirical data, cannot be ruled out. Whether or not, and how, a BTS could be suspended is not, at the moment, clear, although the need for such a suspensive power, notably in relation to the suspension of a BTS mandating that certain derivatives be cleared through a central clearing counterparty (CCP) in the event of a CCP failing, has been canvassed by ESMA.72 As the constitutional location of administrative rule-making power, the Commission appears to be the natural site for such a suspensive power. But multiple difficulties arise, including with respect to how best to engage ESMA, as the location of technical expertise (and, arguably, better able to act speedily), and how to engage the Council and European Parliament, who exercise oversight power over the BTS-adoption process. Were such a suspensive power to be, alternatively, located within ESMA, this would imply the exclusion of the European Commission, Parliament, and Council, all of which would have been formally engaged in the original BTS adoption process. The Meroni difficulties are also significant, given the degree of discretion such a suspensive power would confer on ESMA. The need for some form of suspensive power may, however, become pressing, if significant unintended consequences emerge, and rapid remedial action is required.

V. Conclusion [12.56] The MiFIR transparency regime is likely to bring significant, market-shaping change to the EU financial market. It will also likely prove to be revealing as to the effectiveness of the administrative governance apparatus which supports EU financial governance, and on which the successful amplification, calibration, and application of the extensive, complex, and risk-prone transparency regime depends. [12.57] The MiFIR transparency regime includes supervisory as well as regulatory administrative governance and in this respect heralds a

significant intensification in the level of EU control over operational supervisory decisions. Initial indications suggest, however, that ESMA is likely to prove sensitive and pragmatic in how it wields its new supervisory powers with respect to the MiFIR transparency regime, and that benefits in terms of transaction cost reduction may follow. Regulatory administrative governance may be less successful, however, with respect to the MiFIR transparency regime. Certainly, ESMA provides the EU with a significant technical capacity with which to address the rule-design complexities of transparency regulation, and initial indications augur well with respect to the ESMA rule-design process. But given the uncertain impact of the new regime, difficulties may lie ahead as the administrative process does not yet contain a procedure for the nimble revision and suspension of rules which prove to have destabilizing effects on markets.

1

For discussion see, e.g., David Howarth and Lucia Quaglia, ‘Banking Union as Holy Grail: Rebuilding the Single Market in Financial Services, Stabilizing Europe’s Banks, and “Completing” Economic and Monetary Union’ (2013) Journal of Common Market Studies 51, 103 and Eilís Ferran, ‘Crisis-driven Regulatory Reform: Where in the World is the EU Going?’ in Eilís Ferran, Niamh Moloney, Jennifer Hill, and John C. Coffee (eds), The Regulatory Aftermath of the Global Financial Crisis (Cambridge: Cambridge University Press, 2012), p. 1. 2 For discussion of the new transparency regime see Niamh Moloney, EU Securities and Financial Markets Regulation (Oxford: Oxford University Press, 2014), Chapter V; Guido Ferrarini and Paolo Saguato, ‘Reforming Securities and Derivatives Trading in the EU: From EMIR to MIFIR’ (2013) Journal of Corporate Law Studies 13, 319; and Nils Clausen and Karsten Sørensen, ‘Reforming the Regulation of Trading Venues in the EU under the Proposed MiFID II: Levelling the Playing Field and Overcoming Market Fragmentation’ (2012) European Company and Financial Law Review 9, 275. 3 Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU (recast) [2014] OJ L173/349 (MiFID II); and Regulation (EU) No. 600/2014 of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Regulation (EU) No. 648/2012 [2014] OJ L173/84 (MiFIR). 4 The co-legislators have agreed to an extension of the current date of application of MiFID II/MiFIR from January 2017 to January 2018. 5 The UK Financial Conduct Authority (FCA), which is charged with much of the MiFIR implementation process, has decided, given the direct applicability of MiFIR, not to engage in consultation on much of MiFIR and to carry out only a limited impact

assessment, directed to the small number of areas where there is a degree of national discretion: FCA, ‘Markets in Financial Instruments Directive II Implementation, Consultation Paper I’, CP 15/43 (2015) 8. 6 Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC, OJ [2004] L145/1, 1-44. 7 ESMA has described the new regime as encompassing ‘an exponential increase in the number of instruments under pre- and post-trade transparency obligations’: ESMA, ‘Note on MiFID/MiFIR Implementation: Delays in the Go-Live Date of Certain MiFID Provisions’, ESMA/2015/1514. 8 The UK FCA has noted that the introduction of the OTF classification means that ‘many transactions currently categorized as off-venue will come within a multilateral trading environment’ and that overall market transparency should accordingly increase: FCA, ‘Markets in Financial Instruments Directive II Implementation, Consultation Paper I’, CP 15/43 (2015) 15. 9 The transparency regime is set out in MiFIR, Articles 3–22. 10 For an overview of the current state of adoption of the many Binding Technical Standards required, see . 11 European Securities and Markets Authority, ‘Final report: Draft Regulatory and Implementing Technical Standards MiFID II/MiFIR’, ESMA/2015/1464 (28 September 2015, ‘BTS Proposals’). 12 European Securities and Markets Authority, ‘Final Report: Technical Advice to the Commission on MiFID II and MiFIR’, ESMA/2014/1569 (19 December 2014, ‘Technical Advice’). 13 ESMA, ‘BTS Proposals’ (n. 11), 104–8. 14 See, e.g., on the alternative investment fund manager reforms Eilís Ferran, ‘After the Crisis: The Regulation of Hedge Funds and Private Equity in the EU’ (2011) European Business Organization Law Review 12, 379. 15 From the extensive literature see, e.g.: Lucia Quaglia, Governing Financial Services in the European Union. Banking, Securities, and Post-trading (London and New York: Routledge, 2010); Ryan Davies, Alfonso Dufour, and Brian Scott-Quinn, ‘The MiFID: Competition in a New European Equity Market Regulatory Structure’, in Guido Ferrarini and Eddy Wymeersch (eds), Investor Protection in Europe: Corporate Law Making, the MiFID and Beyond (Oxford: Oxford University Press, 2006), p. 163; Barbara Alemanni, Giuseppe Lusignani, and Marco Onado, ‘The European Securities Industry: Further Evidence on the Roadmap to Integration’, in ibid., p. 199; and Guido Ferrarini and Fabio Recine, ‘The MiFID and Internalisation’, in ibid, p. 235. 16 See, e.g., European Commission, ‘Commission Staff Working Paper: Impact Assessment’ (20 October 2011), SEC/2011/1226, 36–7.

17

Council of the European Union, ‘Cyprus Presidency Progress Report on MiFID II/MiFIR’ (13 December 2012), Council Document 16523/12. 18 The different venue classifications are outlined in Section II. 19 MiFID II, e.g., imposes new requirements on trading venues with respect to how they engage with market-makers and in relation to the obligations imposed on market-makers with respect to liquidity: MiFID II, Article 48. 20 For an assessment see ICMA Secondary Market Practices Committee, ‘The Current State and Future Evolution of the European Investment Grade Corporate Bond Second Market: Perspectives from the Market’, November 2014. 21 See recently OICV-IOSCO, ‘IOSCO Securities Markets Outlook, 2016’ (March 2016) 30–7 22 European Commission, ‘Action Plan on Building a Capital Markets Union’, COM (2015) 468, 13–14 . 23 See further Niamh Moloney, ‘Capital Markets Union: “Ever Closer Union” for the EU Financial System’ (2016) European Law Review 307. 24 Council Directive 2013/36/EU of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC [2013] OJ L176/338 (CRD IV); and Regulation (EU) No. 575/2013 on prudential requirements for credit institutions and investment firms (CRR) [2013] OJ L176/1. 25 For an early assessment see European Banking Authority, ‘Overview of the Potential Implications of Regulatory Measures for Banks’ Business Models’ (9 February 2015). 26 Regulation (EU) No. 648/2012 of 4 July 2012 on OTC derivatives, central counterparties, and trade repositories [2012] OJ L201/1, 1-59. 27 From the extensive discussions see e.g.: Bashir Assi and Diego Valiante, ‘MiFID Implementation in the Midst of the Financial Crisis’, European Capital Markets Institute Research Report No. 6 (2011); Bahram Soltani, Huu Minh Mai, and Meriem Jerbi, ‘Transparency and Market Quality: An Analysis of the Effect of MiFID on Euronext’ (2011), available at ; Valter Lazzari (ed.), Trends in the European Securities Industry (Milan: Egea, 2011); and Giovanni Petrella, ‘MiFID, Reg NMS and Competition Across Trading Venues in Europe and the USA’ (2010) Journal of Financial Regulation and Compliance 18, 257. For a summary of the data see Guido Ferrarini and Niamh Moloney, ‘Reshaping Order Execution in the EU and the Role of Interest Groups: From MiFID I to MiFID II’ (2012) European Business Organization Law Review 13, 557. 28 For an example see Daniel Mügge, ‘The Political Economy of Europeanized Financial Regulation’ (2013) Journal of European Public Policy 20(3), 458. 29 Prior to the Brexit decision the European Council had reached agreement on a ‘New Settlement’ governing the relationship between the UK and the EU, including the relationship between the single market and the euro area (the agreement is now null and

void): Decision of the Heads of State or Government Meeting Within the European Council, ‘Concerning a New Settlement for the United Kingdom with the European Union’, European Council Meeting, 18 and 19 February 2016 (EUCO 1/16), Annex 1. 30 Initially charted in the varieties of capitalism literature and now assessed across multiple dimensions, including with respect to the extent to which bank-based economic systems incorporate market-funding mechanisms. See Iain Hardie and David Howarth (eds), Market-based Banking and the International Financial Crisis (Oxford: Oxford University Press, 2013). 31 See, eg, Ruben Lee, What is an Exchange? The Automation, Management and Regulation of Financial Markets (Oxford: Oxford University Press, 1998) 256. 32 IOSCO, ‘Mitigating Systemic Risk. A Role for Securities Regulation. Discussion Paper’ (2011) 19 and 41. 33 See, e.g., Paul Mahoney, ‘The Exchange as Regulator’ (1997) Virginia Law Review 83, 1453. 34 Robert Merton, ‘A Functional Perspective of Financial Intermediation’ (1995) Financial Management 24, 23 and Ron Gilson and Charles Whitehead, ‘Deconstructing Equity: Public Ownership, Agency Costs and Complete Capital Markets’ (2009) Columbia Law Review 108, 231. 35 See Guido Ferrarini and Paolo Saguato, ‘Financial Market Infrastructures’, in Niamh Moloney, Eilís Ferran, and Jennifer Payne (eds), The Oxford Handbook of Financial Regulation (Oxford: Oxford University Press, 2015), p. 568. 36 Such as whether a trading venue is based on a ‘central order book’ in which orders interact and are matched according to the venue’s algorithms, or on dealers providing liquidity and trading on a principal basis. 37 Market liquidity is currently on the radar of the Financial Stability Board (FSB). Its 2016 Work Programme includes consideration of the factors affecting market liquidity and the impact of regulatory reform on liquidity: Financial Stability Board, ‘FSB Chair’s Letter to G20 Ministers and Governors on Financial Reforms—Progress on the Work Plan for the Hangzhou Summit’, 22 February 2016 . 38  See, e.g., ICMA Secondary Market Practices Committee, ‘The Current State and Future Evolution of the European Investment Grade Corporate Bond Second Market: Perspectives from the Market’, November 2014. 39 As was repeatedly noted during the MiFIR development period. The UK Financial Services Authority (FSA) (now Financial Conduct Authority), e.g., warned that the imposition of transparency requirements on the trading of non-equity asset classes required careful calibration if adverse impacts on liquidity were to be avoided: FSA, ‘The FSA’s Markets Regulatory Agenda’ (2010) 33–4. 40 G20 Leaders, ‘Leaders’ Statement’, Pittsburgh Summit, 24–25 September 2009. 41 See further Moloney (n. 2), 435–44. 42 The standard market size calculation is governed by MiFIR, Article 14(4) and (5).

43

The detailed requirements which apply to this condition are governed by the MiFIR Article 9(5) waiver for trading venues (in relation to non-equity trading pre-trade transparency requirements) and are concerned with liquidity. 44 ESMA, ‘Waivers from Pre-Trade Transparency. CESR Positions and ESMA Opinion’ (2012) (ESMA/2012/206). 45 Regulation (EU) No. 1095/2010 [2010] OJ L 331/84. 46 Under the Meroni doctrine, significant limits apply to the exercising of discretion by EU agencies: Case 9/56 Meroni v High Authority [1957–1958] ECR 133. The extent of ESMA’s permitted discretion has recently been considered by the Court of Justice of the EU in its seminal ‘Short-Selling’ ruling: Case C-270/12 UK v Council and Parliament, 22 January 2014. See, e.g., Miroslava Scholten and Marloes Van Rijsbergen, ‘The ESMA Short Selling Case. Erecting a New Delegation Doctrine for the EU upon the Meroni– Romano Remnants’ (2014) Legal Issues of Economic Integration 41, 289 and Merijn Chamon, ‘The Empowerment of Empowerment of Agencies under the Meroni Doctrine and Art. 114 TFEU: Comment on the UK v Parliament and Council (Short Selling) and the Proposed Single Resolution Mechanism’ (2014) European Law Review 39, 380. 47 For example, the concern expressed by the January 2016 Balz Resolution: European Parliament, ‘Resolution on Stocktaking and Challenges of the EU Financial Services Regulation: impact and the way forward towards a more efficient and effective EU framework for Financial Regulation and a Capital Markets Union’, 19 January 2016 (P8_TA-PROV(2016)0006), para 58. 48 The Opinion calls for legislative revisions to CRD IV to make clear the scope of the proportionality mechanism (EBA/Op/2015/25). 49 ESMA, ‘BTS Proposals’ (September 2015) 158–62. 50 The UK FCA has indicated that it proposes to apply the non-equity market waivers, given that non-equity markets ‘are often characterised by lower and episodic liquidity’, and that it proposes to also apply the equity market waivers, as they are important in ensuring an appropriate balance between liquidity and transparency: FCA, ‘Markets in Financial Instruments Directive II Implementation, Consultation Paper I’, CP 15/43 (2015) 23 and 22. 51 Regulation (EU) No. 236/2012 of the European Parliament and of the Council of 14 March 2012 on short selling and certain aspects of credit default swaps, [2012] OJ L86/1. 52 ESMA, Opinion, 11 January 2016 (ESMA/2016/28). 53 For discussion see Moloney (n. 2), 898–929. 54 ESMA, ‘BTS Proposals’ (September 2015) 54–135. 55 ESMA, ‘Note on MiFID/MiFIR Implementation: Delays in the Go-Live Date of Certain MiFID Provisions’, ESMA/2015/1514 (2015) 3. 56 ESMA Chairman Maijoor, ‘Statement to the ECON Committee’, 14 September 2015 (ESMA/2015/1349). 57 FSB, ‘9th Progress Report on Implementation of OTC Derivatives Market Reforms’ (July 2015), noting challenges in relation to regulatory access to and the usability of the

data held by trade repositories. 58 ESMA, ‘Discussion Paper on MiFID II/MiFIR’ (29 August 2014), ESMA/2014/548; ESMA, ‘Consultation Paper: MiFID II/MiFIR’, ESMA/2014/1570 (19 December 2014); and ESMA, ‘Addendum Consultation Paper: MiFID II/MiFIR’ (19 February 2015), ESMA/2015/319. 59 Annexed to the ESMA, ‘BTS Proposals’ (September 2015) Annex II. 60 ESMA, ‘BTS Proposals’ (September 2015) 57. 61 The bond option sub-asset class, e.g., includes sub-classes based on two criteria (the underlying bond/bond future, and time to maturity): ESMA, ‘MiFID II Briefing on Investment Firms Topics’, ESMA/2015/1469 (28 September 2015). 62 ESMA’s approach to bond market liquidity changed over the BTS development process to reflect stakeholder feedback (including from the European Parliament). ESMA ultimately proposed adopting an instrument-by-instrument approach which is designed to more appropriately reflect liquidity levels on the bond markets: ESMA, ‘BTS Proposals’ (September 2015) 94–7; and ESMA Chairman Maijoor, ‘ECON Scrutiny Speech—MiFID 2’, 10 November 2015 (ESMA/2015/1639). 63 Including with respect to, e.g., derivatives and emission allowances, in relation to which liquidity assessments will be made annually: ESMA September 2015 BTS Proposals, 58. Under ESMA’s proposals bond liquidity is to be re-assessed at the end of every quarter, based on the previous three months’ activity (ESMA, ‘MiFID II Briefing on Investment Firms Topics’, ESMA/2015/1469). 64 Verena Ross (ESMA Executive Director), ‘Keynote Speech at IDX 2015’ (9 June 2015) ESMA/2015/921. 65 Verena Ross (ESMA Executive Director), ‘Keynote Speech at IDX 2015’ (n. 64), noting that ‘given that MiFID II is of a size unprecedented in terms of number and volume of technical standards’, ESMA and the Commission considered it important for proposed BTSs to be legally reviewed before formal and final submission to the Commission. 66 Letter from ESMA Chairman Maijoor to Director General Faull, 15 May 2015. 67 ESMA Executive Director Ross has highlighted that ‘ESMA’s approach is not solely to maximize transparency for its own sake. Our approach is to determine the appropriate amount of transparency that benefits market functioning, while at the same time respecting [the] co-legislators’ intention to increase transparency’: Verena Ross (ESMA Executive Director), ‘Keynote Speech at IDX 2015’ (n. 64). 68 See further Moloney (n. 2), 921–9. 69 ESMA, Letter to Director General, DG Financial Stability, Financial Services, and Capital Markets Union, 21 March 2016 (ESMA/2016/404). 70 A. Mooney, ‘Asset Managers Cheer Rejection of EU Bond Rules’ Financial Times, Fund Management Supplement (21 March 2010), 5. 71 See, e.g., European Commission, ‘Commission Staff Working Document: Report from the European Commission to the European Parliament and Council on the operation

of the European Supervisory Authorities (ESAs) and the European System of Financial Supervision (EFS)’, SWD (2014) 261 (8 August 2014). 72 ESMA, ‘EMIR Review No 1: Review on the use of OTC derivatives by non-financial counterparties’, ESMA/2015/1251 (13 August 2015).

13 SME Growth Markets Rüdiger Veil and Carmine Di Noia

I. Introduction II. Existing Alternative Markets for SMEs in Europe 1. Overview 2. Regulatory Approaches: Unity and Diversity III. 1. 2. 3. 4. 5.

SME Growth Markets under MiFID II Regulatory Aim Concept What will the Future Regime under MiFID II look like? Further Requirements under the MAR Criticism

IV. Alternative Disclosure Obligations? 1. Alternative (1): Disclosure of Key Operating Milestones 2. Alternative (2): Disclosure of Current Event Reports V. What Role for the ‘SME Growth Market’ Label? Some Reflections on Alternative Scenarios VI. Conclusion

I. Introduction

[13.01] In recent years, alternative trading venues for small and mediumsized enterprises (SMEs) have emerged in all the important financial markets across Europe. These venues often admit the trading of stocks but, in some remarkable cases, they admit the trading of bonds, too.1 They differ from regulated markets (RMs) by imposing lower regulatory requirements and are therefore more cost-efficient for SMEs. It is, however, astounding how different the regulation of these alternative markets is. In part, this is because the 2003–2007 EU legislation on capital markets, especially the Market Abuse Directive (MAD)2 and Transparency Directive (TD),3 is limited to imposing that the Member States set rules for RMs.4 Alternative markets, which are multilateral trading facilities (MTFs) pursuant to the Markets in Financial Instruments Directive (MiFID I), are not therefore subject to these EU directives. Thus, the Member States and the stock exchange operators have largely been free to set their own rules for these markets. [13.02] The regulatory requirements for alternative capital markets in Europe cannot be discussed in detail here. However, this chapter depicts the different regulatory approaches for alternative markets in the United Kingdom, Italy, the Nordic States, France and Germany.5 Subsequently, the chapter will discuss the changes that alternative markets are likely to undergo due to the implementation of several capital markets law reforms until 2018. The focus is set on the changes introduced by the MiFID II, which Member States will adopt by 3 January 2018, and by the Market Abuse Regulation (MAR),6 which has been applicable since 3 July 2016. [13.03] The MiFID II introduces the category of ‘SME growth market’. The EU legislator’s intention is to make alternative capital markets more attractive with this label. Thus, in the following it will be analysed which European rules will apply to the SME growth markets and whether this new category actually does increase the attractiveness of alternative markets.

II. Existing Alternative Markets for SMEs in Europe

1. Overview [13.04] In most major EU Member States, alternative capital markets for SMEs are already in place. The most important of these is the Alternative Investment Market (AIM) run by the London Stock Exchange (LSE). It opened on 16 June 1995. The AIM is an MTF as defined in the MiFID I and a prescribed market according to the Financial Services and Markets Act 2000 (FSMA).7 Its purpose is to serve as a market for small and highgrowth enterprises. In April 2016, 1,013 companies (of which 824 were from the UK and 189 international) were admitted to trading. [13.05] The AIM Italia is a market run by Borsa Italiana, which aims at small and medium-sized Italian enterprises with high-growth potential. It was established by merger with the MAC (Mercato Alternativo del Capitale)8 on 1 March 2012 and is an MTF. Borsa Italiana is owned by the LSE. This explains why most of the rules for the Aim Italia conform to those for the AIM provided by the LSE. At the end of April 2016, 74 companies were admitted to trading. In Italy there is also an MTF segment (named Extra MOT pro) dedicated to companies issuing bond (so-called ‘mini-bonds’) open only to qualified investors. Euronext has also established alternative trading facilities in Brussels, Lisbon, and Paris. These alternative markets, operated under the trademark Alternext, are mainly aimed at SMEs and offer a more flexible legal framework than that of the RMs. In all those three jurisdictions, Alternext is an MTF.9 In May 2016, 184 companies were listed at the three trading venues.10 Previously, Alternext operated an Amsterdam-based MTF, which ceased its operations in 2014. However, in April 2015 Euronext announced that it would be launching a Belgian–Dutch SME exchange. This new MTF of the Euronext subsidiary Enternext aims to cater to the growing need of SMEs to obtain non-bank funding.11 [13.06] Alternative trading facilities in Denmark, Finland, Iceland, and Sweden are organized under largely uniform rules of the trademark First North. In Denmark NASDAQ OMX Copenhagen A/S operates First North Denmark, in Finland NASDAQ OMX Helsinki Ltd operates First North Finland, in Iceland NASDAQ OMX Iceland hf operates First North Iceland and in Sweden NASDAQ OMX Stockholm AB has established First North

Sweden and First North International. All of these markets are MTFs.12 Within the Swedish First North there is a special segment which is called First North Premier Segment (FNPS). In regards to this segment, stricter information requirements (e.g. regarding closely related party transactions) apply than to the ‘general segment’ of First North. [13.07] In Germany, several alternative trading facilities exist.13 The most important one—the Entry Standard—was established by the Deutsche Börse AG on 25 October 2005. The AIM in London and Alternext in Paris served as models.14 Currently, around 180 companies are admitted to trading in the Entry Standard. According to practitioners, the Entry Standard has proven its value as an alternative trading facility for SMEs in Germany.15 Another important market is the m:access, which was created by Börse München and currently includes stocks of around sixty companies.

2. Regulatory Approaches: Unity and Diversity A. Admission to Trading [13.08] A mix of regulations by the market operators as well as (national) statutory provisions characterizes all alternative markets. All these markets share a regulatory approach that requires the issuer to have an adviser for the application to admission and the subsequent listing time. For instance, at the AIM run by the LSE, an issuer has to appoint a so-called nominated adviser.16 The advisers’ duties and obligations during the admission process and trading are set out in detail in the regulations of the market operators (rule-books). Pursuant to these regulations, the adviser carries out the duties which otherwise are performed by the market operators, such as to check whether the issuer complies with the requirements for the admission to trading of securities. [13.09] Admission requirements differ from one market to another. The most liberal one in this respect is the AIM. On one hand, this is evident by the fact that no percentage of freely tradeable shares (free float) is required for the admission process. On the other hand, several facilitations concerning the admission document exist in favour of the issuer (in

contrast, the German Entry Standard is the only alterative market requiring a prospectus issued in accordance with European law). As long as the issuer is not legally bound to publish a prospectus,17 it is sufficient, in accordance with the market operator’s regulations, to draw up and publish a document informing about the issuer and its securities (admission document). Requirements regarding the content of the document are less stringent that those mandated by the prospectus regulated by the Prospectus Directive (PD).

B. Requirements for Being Public [13.10] All markets tackle the issue of information asymmetries between investors and issuers by means of disclosure obligations. The periodic disclosure obligations show large conformity in the way of reporting: most market regulations require an issuer to draw up and publish an annual as well as a semi-annual financial report. Facilitations are granted with the accounting standards the issuer shall apply. Apart from the premium segment of First North, the issuer is entitled to draw up its report according to national accounting law. Thus, the issuers are not obliged to apply the IFRS. This flexibility is a big advantage of alternative markets as the application of IFRS is particularly burdensome for SMEs. None of the alternative markets requires the publication of quarterly financial reports. [13.11] The focus of ad hoc disclosure is on the obligation to publish information relevant to the market without undue delay. Before the application of the Market Abuse Regulation on 3 July 2016, the market operators have pursued this aim by mandating general disclosure obligations for ‘price-sensitive information’.18 Furthermore, disclosure obligations are set forth for specific events and circumstances.19 Thus, there are general and special ad hoc disclosure obligations. Interestingly, most market operators did not address the problem that disclosure of future information can prejudice the legitimate interests of issuers, as it is known by the obligation to disclose inside information for issuers on RMs. If applied literally, the rules for alternative markets would not allow a delay in disclosure, despite the issuer’s justified interest.

[13.12] Furthermore, several market operators have stipulated that corporate transactions, such as substantial transactions and fundamental changes of business, shall be disclosed.20 In most alternative markets, save for the Entry Standard and m:access, transactions between related parties shall be reported. In the United Kingdom, this disclosure obligation is regulated in detail,21 while most other market regulations only have a general disclosure obligation.22 [13.13] Moreover, it is common that investors within alternative markets shall disclose the acquisition or the disposal of shares (as required by the Transparency Directive for major shareholdings in companies with shares traded on RMs). The legal bases for these obligations to provide information are set forth in statutory provisions of national capital market laws. In this sense, most alternative markets do not differ from the RMs.23 These disclosure obligations have a relatively long tradition in Europe’s RMs. Primarily, they fulfil capital market law purposes as they tackle information asymmetries; however, they also contribute to evincing potential takeovers at an early stage. For issuers listed on alternative markets, disclosure of major shareholdings is generally not tied to any direct significant costs. This is because the issuer is merely obliged to publish the information it received from the investor. Where non-fulfilment of the obligation to provide information is not sanctioned with the loss of (voting) rights, the issuer sees no reason to check for the accuracy of the information transmitted to it. Therefore, in this respect they do not have to bear any relevant costs. [13.14] With the approach of using market regulations to also tackle company law issues—for example the necessity of an approval by the general assembly for fundamental measures24—the AIM in the United Kingdom distinguishes itself from other alternative markets operating in Europe. None of the other market operators includes any specific company law provisions in their rules and regulations.

III. SME Growth Markets under MiFID II

1. Regulatory Aim [13.15] With Directive 2014/65/EU (MiFID II) the EU legislator created the new category of ‘SME growth market’. This is intended to be a new quality certificate for alternative trading facilities: The ‘SME growth market is a new sub category […] within the MTF category’,25 and the new legislative recognition ‘should raise their visibility and profile and aid the development of common regulatory standards in the Union for those markets’. 26 The operators of markets aimed at smaller and medium-sized issuers will have the option ‘to choose to continue to operate such a market in accordance with the requirements under the MiFID II without seeking registration as an SME growth market’.27 Issuers who are SMEs ‘should not be obliged’ to apply to have their ‘financial instruments admitted to trading on an SME growth market.’28

2. Concept [13.16] According to Article 33(1) MiFID II, Member States shall provide that the operator of an MTF may apply to its home competent authority to have the MTF registered as an SME growth market. The home competent authority may register the MTF as an SME growth market if it receives the application referred to in Article 33(1) MiFID II and the requirements set out in Article 33(3) MiFID II are complied with in relation to the MTF (Article 33(2) MiFID II). [13.17] One essential requirement is that ‘at least 50% of the issuers whose financial instruments are admitted to trading on the MTF are SMEs at the time when the MTF is registered as an SME Growth Market and in any calendar year thereafter’ (Article 33(3)(a) MiFID II). Thus, the term SME bears crucial significance and is defined in MiFID II as a company, which ‘had an average market capitalisation of less than EUR 200 million on the basis of end-year quotes for the previous three calendar years’ (Article 4(1) No. 13 MiFID II).29 All current alternative trading facilities (as described above) could indeed be registered as SME growth markets. Even though all markets have listed companies which have a market

capitalization much higher than €200 million, the majority of issuers have a smaller market capitalization.30

3. What will the Future Regime under MiFID II look like? [13.18] The MiFID II only provides a framework for SME growth markets. This directive focuses on investor protection on the SME growth market. Particularly important is that ‘appropriate criteria are set for initial and ongoing admission to trading of financial instruments of issuers on the market’ as well as ‘on initial admission to trading’, that ‘there is sufficient information published’ […] in an admission document or prospectus, and that ‘there is appropriate ongoing periodic financial reporting by or on behalf of an issuer’ (Article 33(3) MiFID II). [13.19] The Commission is empowered to adopt more specific provisions by delegated acts. Therefore, it shall ensure that: the measures shall take into account the need for the requirements to maintain high levels of investor protection to promote investor confidence in those markets while minimising the administrative burdens for issuers on the market and that deregistrations do not occur nor shall registrations be refused as a result of a merely temporary failure to meet the conditions set out […].

(Article 33(8) MiFID II) [13.20] In December 2014, after having consulted with stakeholders, ESMA handed over its technical advice to the Commission.31 It contains principle-based advice for the adoption of delegated acts. In this chapter not every technical aspect for an SME growth market regime can be addressed. However, the most defining principles, which ESMA recommended to the Commission, shall be pointed out:32 • first, ESMA follows the general notion that ‘an SME growth market should not be required to have rules that impose greater burdens on issuers than those applicable to RMs’;

• ESMA acknowledges the different rules governing alternative trading facilities and declares itself in favour of the EU law not being allowed to thwart these national developments; • according to ESMA, an SME growth market should not be required to have rules prescribing the use of IFRS but rather it should also allow issuers to apply national accounting standards; • issuers wishing to have their shares admitted for trading on the SME growth market shall publish either a prospectus, which meets the requirements of the Prospectus Directive, or an adequate admission document; and • issuers on a SME growth market shall publish an annual and a semiannual financial report. [13.21] In its draft delegated regulation, the Commission has largely followed ESMA’s technical advice. Article 78(2)(c) of the regulation requires that admission to trading on an SME growth market can only take place if the concerned issuer has published either a prospectus under the Prospectus Directive 2003/71/EC or an appropriate admission document drawn up under its responsibility that specifies whether an approved or review process was performed and, if so, by whom. The rule at the same time mandates, however, that the admission document be subject to an appropriate review on its completeness, consistency, and comprehensibility. [13.22] The SME growth market shall set the minimum contents of the admission document so that investors can make an informed assessment of the financial position and prospects of the issuer, and of the rights attaching to its securities. The Commission delegated regulation already establishes that an issuer’s assessment on the adequacy of its working capital, or on the plans to provide additional working capital, shall be among the information included in the admission document (Article 78(2)(d) and (e)). [13.23] Finally, SME growth market issuers are required to publish annual and semi-annual financial reports, respectively within six months after the end of each financial year and within four months after the end of the first six months of each financial year (Article 78(2)(g) Commission delegated regulation).33 Whether such reports should be drafted according to the International Financial Reporting Standards or local financial reporting

standards is up to the SME growth market operator to determine (Recital 114 Commission delegated regulation).

4. Further Requirements under the MAR [13.24] The Market Abuse Regulation (MAR), which has been applicable since 3 July 2016, also governs financial instruments which are traded on an MTF.34 Thus, the prohibition of insider trading and the prohibition of market manipulation will apply to all alternative trading facilities. However, this will not result in any significant changes, since the Member States already provide for the prohibition of insider trading and for the prohibition of market manipulation on all alternative trading facilities, at least as long as the relevant financial instruments are admitted to trading upon their issuers’ request or approval.35 [13.25] For the first time, indeed, there will be legal ad hoc disclosure obligations, which will be subject to supervision by National Competent Authorities (NCAs). So far, only regulations of market operators provide for disclosure obligations; moreover, their contents differ from those MAR provides for. On top of this, companies whose financial instruments are traded on MTFs have been subject to the MAR regime on insiders’ lists and managers’ transactions since 3 July 2016. Formally, the MAR regime for insider lists is lighter for such companies, as they are not required to draw up and update the list on an ongoing basis. However, they still have to identify persons with access to inside information and inform them about the consequent legal and regulatory duties, and, most importantly, they still have to provide the competent authority with an insider list upon request (Article 18(6) MAR). This last requirement greatly reduces the regulatory benefits stemming from the exemptions, as SME growth market issuers are likely to keep fully fledged insider registers with a view to being able to meet possible requests by competent authorities. In its draft implementing technical standards, ESMA has proposed reducing the regulatory burden upon SME growth market issuers. Those issuers are required to file with the authority, upon this latter’s demand, a smaller set of data than the one populating insider lists maintained by ordinary issuers.36 In spite of this regulatory effort, compliance with rules on insider lists will also remain

onerous for SME growth market issuers, as the bulk of the standard regime remains in place. [13.26] As for supervision, only the nominated adviser (on the AIM), whom the issuer has to appoint prior to admission of its securities to the alternative market, checks whether regulations are complied with. In the future, this will be subject to supervision by NCAs.

5. Criticism [13.27] In order to answer the question whether the SME growth market under MiFID II will be more attractive to issuers and investors than the current mostly self-regulated alternative trading facilities, first the new information requirements under the MAR shall be reviewed. An ‘issuer shall inform the public as soon as possible of inside information which directly concerns’ it (Article 17(1) MAR). Furthermore, ‘persons discharging managerial responsibilities [as well as persons closely associated with them] shall notify the issuer [and the NCA] of every transaction conducted on their own account relating to the shares or debt instruments of that issuer [or to derivatives or other financial instruments linked thereto]’ (Article 19(1), (4) MAR). The issuer shall then perform the publication of the directors’ dealings. [13.28] The problem, however, is that, when putting these disclosure obligations into legal practice, many issues arise which require expensive advice from lawyers. One example is the obligation to disclose inside information publicly. An obligation to disclose applies whenever information becomes inside information. To assess this issue correctly is a highly complex task. First, circumstances that ‘may reasonably be expected to come into existence’ must be considered as inside information as well (Article 7(2) MAR). It is quite a challenging task to assess whether a future event may reasonably occur. A further problem arises with regard to the price relevance as a further element of the term ‘inside information’: how does a reasonable investor assess the relevance of information? Finally, the right to delay publication of inside information (see Article 17(4) MAR) is

complicated and, due to a number of difficult interpretational issues, costly too.37 [13.29] The costs for such legal advice are sustainable for big enterprises; for SMEs, they may rise to prohibitive amounts.38 In addition, high pecuniary sanctions can be imposed upon infringement. An issuer can be sanctioned with up to 2 per cent of the total annual turnover of the legal group if it violates the obligation to disclose inside information.39 It can be concluded that the disclosure regime on a SME growth market is quite similar to the one on RMs and is particularly burdensome for SMEs. [13.30] The application of the new market abuse law might lead to even more disadvantages. Up until now, the issuers used the disclosure obligation set in the regulations of alternative trading facilities to inform the capital market about the company’s developments.40 This is possible, as these disclosure obligations are rather general, leave room for interpretation, and thus provide sufficient flexibility. Moreover, so far none of the NCAs ensure that information is only published when it is covered by the disclosure obligations. [13.31] In the future, this flexibility will be limited, as the MAR strengthened the duty to disclose inside information (Article 17(1) MAR). This term is defined precisely and requires the price relevance of information (see Article 7 MAR). Additionally, it has been explicitly stated, that the ‘issuer shall not combine the disclosure of inside information to the public with the marketing of its activities’ (Article 17(1) MAR). The NCAs will have to monitor the compliance with these regulations. In any case the issuer retains the possibility to disseminate information which is not pricesensitive (i.e. commercial or marketing information) provided that it is not disseminated through an OAM (officially appointed mechanism). [13.32] In light of the above, the question arises of whether market operators of alternative trading facilities may set out additional regulations for SME growth markets. Comparative legal research among established alternative trading facilities reveals that the operators provide a variety of disclosure obligations. It seems reasonable to maintain such obligations, when operators register their trading facilities as SME growth markets.

However, is this permissible? Investment firms and market operators operating SME growth markets are explicitly enabled to impose additional requirements to those set forth by MiFID II and its implementing measures (see Article 33(4) MiFID II), including issuers’ disclosure duties under MAR (Article 33(3)(e) MiFID II). However, concerns arise, since the MAR pursues maximum harmonization thus prohibiting stricter rules than those laid down in European law. This is justified by the objective of achieving a legal level playing field in the Union.41 It will prevent a regulatory and supervisory arbitrage.42 [13.33] Therefore, the national legislator, for one, is barred from setting out additional obligations within the regulatory scope of Article 1 MAR. The approach to maximum harmonization prevents the formerly prevalent gold plating. Nevertheless, it has not yet been clarified whether the market operator may at least set such regulations. The wording of MAR’s Recital, which refers the uniformity issue solely to the ‘Union market abuse framework’ and the ‘market abuse requirements in the form of a regulation’,43 is in support of such interpretation. Thus, the approach of maximum harmonization should probably refer to statutory law only. Moreover, it is acknowledged that the maximum harmonization of periodic disclosure obligations under the Transparency Directive refers solely to statutory provisions of the Member States and not to the regulatory powers of market operators.44 This applies even more when taking into consideration that the Transparency Directive has been revised parallel to the MAD. However, such interpretation is not certain. It can be argued that the MAR pursues ‘more clearly defined rules’ and the removal of ‘significant distortions of competition’45 and therefore bars market operators from setting additional and tailored disclosure obligations.

IV. Alternative Disclosure Obligations? [13.34] Numerous alternative disclosure obligations could be considered for SME capital markets, which enable the issuer to both inform the market participants of recent developments in their company and provide investors with access to relevant information. Two sets of regulations will be outlined

here, which have been tested regarding SMEs in the US and in New Zealand.

1. Alternative (1): Disclosure of Key Operating Milestones A. Concept [13.35] Reporting on key operating milestones (KOMs) is well known from US capital markets.46 There, it mainly plays a part in primary market publicity. In New Zealand, this reporting is designed to be an ongoing requirement and is utilized for the secondary market. [13.36] The concept of KOMs requires a company to be aware of its own business plan and to inform the capital market on whether or not it has reached the target it set therein. The starting point is that a company listed on the New Zealand-based NXT ‘must have key operating Milestones (KOMs) and annual targets (ATs) […] in respect of each KOM’ (Rule 42). Firstly, a company must determine a KOM. This must be done pertaining to the object of the company and its fundamental business-policy strategies. Furthermore, a KOM has to be quantitative and measurable. For a service provider a KOM could for example be the number of customers, the average revenue per customer, or the average service costs per customer.47 By contrast, an AT is an expectation of performance by the company pertaining to one KOM. For example, the AT pertaining to ‘number of customers’ could be 1 million. [13.37] The NXT Market Rules oblige a company to review the AT and to inform the capital markets where necessary ad hoc and in any case periodically: • Thus the company must immediately issue a notification, when it expects to miss a target set for a KOM by more than 10 per cent (see Schedule 5A No. 9). • It may reset an AT for a KOM at any time if it has reasons to expect that it will miss a target set for the KOM, but if it does so, it is required to

inform the capital markets (Rule 47). • Furthermore, a company must report quarterly in a business update on its performance regarding the achievement of the targets set for the KOMs. • Finally, a company must annually review the target set for each KOM and publish no later than two calendar months after the end of the business year for each KOM the (if applicable amended) target (Rule 43). [13.38] The company must evaluate at least every twelve months whether a KOM meets the NXT standard.48 Essentially this is done to ensure that the self-defined KOMs continue to represent the business model of the company. If so, the directors of the company must issue a declaration confirming that the KOM meets the NXT standard (Rule 44). Should a KOM not meet the NXT standard, the company is required—in consultancy with its NXT Advisor—to reformulate the KOM and to determine a new target for reaching the KOM (Rule 45). The capital markets must be informed about this, too (Rule 46).

B. Evaluation [13.39] The transparency concept of KOMs appears especially convincing for SMEs that are growing.49 Furthermore, the capital markets have a disciplinary effect on the management and thus perform corporate governance functions.50 With SMEs it is also imaginable that the business model and its risks could be measured by four or five KOMs. [13.40] From a practical perspective, however, valid arguments could be raised against such disclosure obligations. Above all, it is questionable whether the corresponding rules can be applied reliably and whether related reports would provide any additional information.51 Additionally, it should be noted that there is no practical experience yet with milestone reporting on secondary markets.52 This disfavours a regulation by statutory provision and favours leaving it to the market operator to test the functionality of milestone-reporting.

[13.41] When assessing the functionality of a system based on dissemination of KOMs, a market operator will have to consider that the practice of ad hoc disclosure on SME growth markets has changed, because the rules of the MAR only allow issuers to publish through the official dissemination channels—for SME growth market issuers, the SME growth markets’ websites, which can also refer to a specific page of the issuers’ websites (Article 78(2)(h) Commission delegated regulation)— information that is price relevant. On one hand, periodic milestone-reporting could counteract a decrease in published information. On the other, investors would receive valuable information about the development of the company.

2. Alternative (2): Disclosure of Current Event Reports A. Concept [13.42] The obligation to make public current events is a further alternative disclosure concept that market operators could establish. In Europe, this can mainly be seen on the AIM. According to Rule 17, an issuer must, under some circumstances, issue notification without delay. These include amongst others the appointment and dismissal of directors, the change of the legal name, and the change of the registered office address. There is also a disclosure obligation regarding ‘any deals by directors’ and ‘relevant changes to any significant shareholders’. In addition, Rule 17 obligates the publication of any material change between the company’s actual trading performance or financial conditions and any profit forecast, estimate, or projection included in the admission document or otherwise made public on its behalf. Any deviation by more than 10 per cent is deemed to constitute a material change.53 [13.43] The obligations under US capital market law for so-called Regulation A offerings to publish current event reports have been developed even further. This is mainly because on AIM a disclosure obligation for inside information is set in addition to the disclosure obligations provided for in Rule 17,54 whilst in the USA such an ad hoc disclosure obligation does not exist. The disclosure obligations in the US concern certain corporate events, such as (i) fundamental changes to the

business policy because of an agreement, (ii) opening of any insolvency proceedings, (iii) material modifications to rights of security holders, (iv) change of the certifying accountant, (v) non-reliance on previously issued financial statements, (vi) changes in control of the issuer, (vii) departure of certain persons from the management of the issuer, and (viii) sales of unregistered securities of the issuer.

B. Evaluation [13.44] The transparency requirements in the USA stipulate ad hoc disclosure obligations that are limited to current events. An issuer is not required to report on future events. It is therefore no wonder that the SEC assumes a disclosure obligation on current events will become relevant on average only once per year. The compliance costs caused by information that must be published ad hoc are deemed reasonable55 because the SEC assumes that an issuer generally does not require legal counsel to fulfil its obligation to publish current event reports. [13.45] The transparency concept provided in US law appears to be convincing for SME growth markets. It ensures that the market participants are informed about the most important developments in the company. The main advantage for the issuer is that this regime causes much lower costs compared to the disclosure obligations for inside information pursuant to the MAR. Therefore, when evaluating the MAR with the next reform, the European legislator should consider whether for SME growth markets an alternative transparency concept would be preferable. [13.46] However, this will only happen in future years. Until then, the market operators must reflect on whether they want to have explicit rules in place that require the publication of current events in addition to the obligation to publish insider information on both future and current events. The transparency obligation provided in the AIM rules could serve as a model in doing so.

V. What Role for the ‘SME Growth Market’ Label? Some Reflections on Alternative Scenarios [13.47] The certification role of admission to trading on a RM has long been recognized in the academic literature.56 While SME growth markets necessarily qualify as MTFs, they still retain a listing function when they select which financial instruments should be traded on their trading platform. For SME growth markets that decide on the admission to trading upon issuer request (primary listing MTFs) this results in a scrutiny of the applying issuers’ governance and/or financial instruments, so as to verify compliance with the requirements set forth in the applicable laws and in the MTF regulations.57 In this way, admission to trading shows investors that the relevant SMEs reach a minimum level of quality on some key features such as transparency and market liquidity. In spite of the risks inherent to the start-up phase or to the small size of traded companies, equity or debt should therefore come cheaper as far as investors perceive that information asymmetries are reduced and the trading venue operator ensures financial instruments are fairly priced. [13.48] In the MiFID II systems, not only do SME growth markets perform this listing function, but they are also themselves subject to a higher-level implicit certification by national competent authorities. These latter decide on the registration of SME growth markets after assessing compliance with the requirements of (national law implementing) Article 33 MiFID II. On top of the authorization to operate an MTF platform (Article 5 MiFID II), conferral of the label ‘SME growth market’ clarifies that the operator complies with further minimum standards and runs a trading venue specializing in financial instruments issued by SMEs. This upstream assessment of SME growth markets should enhance, downstream, SME growth markets’ ability to certify the quality of companies admitted to trading. [13.49] The long-term success of the ‘SME growth market’ trademark will crucially depend on the ability of this multi-level assessment system to signal high quality listing standards. Whether eligible MTFs will rush to display that label or will rather refrain from doing so by taking advantage of

their freedom not to seek registration as SME growth markets (Recital 134 MiFID II) is difficult to predict, as the market perception of operators’ credibility may depend on a number of variables. While any attempt to forecast the success of SME growth markets would be a high-risk exercise, the question does offer an opportunity to analyse some dynamics of the market for listing. [13.50] Broadly speaking, one can imagine two extreme scenarios, the real one possibly lying somewhere in between these two poles. In a first setting, operators of MTFs will avoid registration as SME growth markets as this would convey negative signals to potential investors, especially if the market was perceived as listing issuers that are unable to access funding through traditional channels. The second scenario is instead one where the label ‘SME growth market’ credibly signals a high quality market that compensates the inherent risks of start-up businesses with cost-effective regulation and supervision of key governance and/or information requirements. [13.51] With some unavoidable simplifications, some factors are likely to tilt towards one or the other scenario, in the light of SME growth market operators’ incentives to bear the costs of defining and enforcing strict listing requirements. Militating against the ability to credibly convey information on the quality of SME growth markets are operators’ incentives to admit as many companies as possible in order to benefit from higher revenues from admission and trading fees. Similar concerns have been raised with regard to the listing function performed by RMs,58 but the fear of losing the qualification of SME growth market in case the 50 per cent threshold is no longer met (Article 33(3)(a) MiFID II) may exacerbate the risk of a relaxed supervision on the admission criteria. [13.52] At the same time—coming now to the variables that may favour the success of SME growth markets—the negative reputational effects of a poor initial and ongoing scrutiny of admission standards may counteract the tendency to increase short-term revenues at the expense of long-term ones.59

[13.53] Equally difficult to evaluate are the effects of the competition among trading venues that the MiFID (I and II) regime aims to foster on SME growth market operators’ incentives. On one hand, secondary listing MTFs may free-ride on the SME growth markets’ primary listing activity, thus leading to suboptimal investments in the scrutiny of issuers’ quality.60 On the other hand, availability of secondary listing on other MTFs increases the liquidity of financial instruments first admitted to trading on primary listing SME growth markets. This may allow SME growth markets to freeride, in turn, on secondary listing: investors will be ready to provide issuers with cheaper capital in anticipation of the increased liquidity of their investment due to positive cross-network externalities,61 and SME growth markets operators can partially take advantage of this by increasing their admission fees.

VI. Conclusion [13.54] The introduction of SME growth markets under MiFID II is an important strategy to improve access to finance for SMEs in Europe. SME growth markets should be more flexible than RMs. However, this will not be the case. The regime for SME growth markets will consist of strict rules about insider trading and market manipulation that are subject to supervision by NCAs. To be sure, these parts of the regime are important to ensure investor confidence, and it is also a convincing strategy to protect investors by implementing a disclosure regime that ensures the publication of price-relevant information on SME growth markets. However, it is neither necessary nor recommendable to apply the respective disclosure obligations (as well as the similar regime for insiders’ lists and the identical one for managers’ transactions) under the MAR. Instead, a system based on current event reports is sufficient in order to tackle information asymmetries on SME growth markets. Therefore, the disclosure regime for SME growth markets in Europe should be reassessed with the aim of allowing market operators to experiment with alternative disclosure obligations on SME growth markets.

1

Article 77(2) of the Commission’s draft delegated regulation supplementing MiFID II as regards organizational requirements and operating conditions for investment firms (‘Commission draft delegated regulation’) sets specific criteria for assessing whether issuers having only non-equity instruments traded on a trading venue qualify as SMEs for the purposes of SME growth markets. 2 Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation (Market Abuse Directive, ‘MAD’), OJ L 96/16. 3 Directive 2004/109/EC of the European Parliament and of the Council of 15 December 2004 on the harmonization of transparency requirements in relation to information about issuers whose securities are admitted to trading on a regulated market and amending Directive 2001/34/EC (‘Transparency Directive’), OJ L 390/38. 4 The concept of an RM is defined in Article 4(1) No. 22 MiFID II; on this see R. Veil, ‘Capital Markets’, in R. Veil (ed.), European Capital Markets Law, 2nd edn (Oxford and Portland: Hart Publishing, 2016), § 7, paras 20–31. 5 Cf. for a detailed analysis R. Veil, Kapitalmarktzugang für Wachstumsunternehmen, (Tübingen: Mohr Siebeck, 2016), pp. 42–99. 6 Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16 April 2014 on market abuse (market abuse regulation) and repealing Directive 2003/6/EC of the European Parliament and of the Council and Commission Directives 2003/124/EC, 2003/125/EC and 2004/72/EC, OJ L 173/1. 7 Cf. M. Stuttard, ‘AIM’, in T. Nicholls and L. Graham (eds), A Practitioner’s Guide to the AIM Rules, 6th edn (London: Sweet & Maxwell, 2011), p. 2. 8 The MAC was a market for SMEs and only accessible for professional investors. Cf. P. Fioruzzi, E. de Nardis, N. B. Puppieni, and F. Saccone, ‘Italy’, in R. Panasar and P. Boeckman (eds), European Securities Law, 2nd edn (Oxford: OUP, 2014), para. 16.21. 9 Cf. M.-L. Tibi, ‘Alternext Rulebook 1.9’, in R. Panasar and P. Boeckman (eds), European Securities Law, 2nd edn (Oxford: OUP, 2014), para. 13.09. 10 See https://www.euronext.com/en/listings/alternext. 11 See Archie van Riemsdijk, ‘Euronext begint Nederlands-Belgische mkb-beurs – Update’, beurs.nl, 16 April 2015, . 12 The alternative markets are subject to (partially different) national legislation and to the First North Nordic-Rulebook. 13 Cf. for further alternative markets at regional exchanges in Germany, A. Meyer, ‘Das Konzept von Börsennotierung und Aktienplatzierungen’, in R. Marsch-Barner and F. A. Schäfer (eds), Handbuch börsennotierte AG, 3rd edn (Munich: C. H. Beck, 2014) § 7, para. 48. 14 Cf. M. Schlitt and S. Schäfer, ‘Der neue Entry Standard der Frankfurter Wertpapierbörse’ (2006) Die Aktiengesellschaft (AG), 147.

15

Cf. Meyer (n. 13), § 7 para. 48; L. F. Freytag and K. T. Koenen, ‘Neue Regeln für den Open Market’ (2011) Wertpapiermitteilungen (WM), 1594. 16 At other alternative markets, the role of advisers (‘listing partner’ at Alternext markets, ‘certified adviser’ at First North and ‘listing partner’ at the Entry Standard in Frankfurt) is similar. 17 This is the case if the issuer publicly offers stocks for sale (cf. Article 3(1) Prospectus Directive). 18 Cf. Rule 11 AIM Rulebook: An AIM company must issue notification without delay of any new developments which are not public knowledge concerning a change in (i) its financial condition, (ii) its sphere of activity, (iii) the performance of its business or (iv) its expectation of its performance, which, if made public, would be likely to lead to a substantial movement in the price of its AIM securities. 19

Cf. Rule 17 AIM Rulebook about ‘disclosure of miscellaneous information’, such as relevant changes to any significant shareholders, the resignation, dismissal, or appointment of any director, etc. 20 Cf. Rules 12–16 AIM Rulebook. 21 Cf. Rule 13 AIM Rulebook in connection with Annex 3. 22 At the AIM, the disclosure obligation is determined according to a gross assets test, profits test, turnover test, consideration test, and gross capital test. 23 In the UK, the statutory disclosure obligations regarding changes of major shareholdings (DTR 5 FCA Handbook) also apply to shares admitted to trading on the AIM. Cf. J. Bennett, ‘The Statutory Framework’, in Nicholls and Graham (eds) (n. 7), 101. 24 Cf. Rule 14 AIM Rulebook regarding reverse takeovers and Rule 15 AIM Rulebook regarding fundamental changes of business. 25 Recital 132 MiFID II. 26 Recital 132 MiFID II. 27 Recital 134 MiFID II. 28 Recital 134 MiFID II. 29 As per Article 77(1) of the Commission draft delegated regulation, the €200 million threshold for newly listed companies shall be calculated on the basis of the closing share price of the first day of trading for issuers admitted to trading for less than one year. For issuers admitted for more than one year or more than two years (but in any case for less than three years) the calculation is based on the last closing price at the end of the first year or, respectively, on the average of the last closing share prices of each of the first two years of trading. 30 Cf. Veil (n. 5), 53 regarding the AIM, p. 79 regarding the Swedish First North, and p. 88 regarding the AIM Italia.

31

ESMA, Technical Advice to the Commission on MiFID II and MiFIR. Final Report, 19.12.2014, ESMA/2014/1569. 32 Cf. ibid., p. 356 ff. 33 These deadlines are less strict than those set forth by Articles 4 and 5 Transparency Directive 2004/109/EC for issuers whose securities are admitted to trading on a RM (namely six months instead of four for the yearly report and four months instead of three for the half-yearly report). 34 Cf. Article 2(1) lit. (b) MAR. 35 France, Germany, Italy, Spain, Sweden, and the UK gold-plated the former MAD and provided prohibition on insider trading and market manipulation for MTFs in their national laws. 36 For instance, data concerning insiders’ personal address may be communicated only insofar as the SME growth market issuer knows them at the time of the authority’s request. 37 For an analysis see C. Di Noia and M. Gargantini, ‘Issuers at Midstream: Disclosure of Multistage Events in the Current and in the Proposed EU Market Abuse Regime’ (2012) European Company and Financial Law Review 9, 484, 504–6. 38 Similarly, G. Ferrarini and A. Ottolia, ‘Corporate Disclosure as a Transaction Cost: The Case of SMEs’ (2013) European Property Law Journal 2, 296, 305. 39 Cf. Article 30(2)(j)(ii) MAR for infringements of Article 17 MAR. 40 See in more detail Veil (n. 5), 64–70 regarding two issuers listed at the AIM and pp. 138–43 regarding two issuers listed at the Entry Standard. 41 See Recital 5 MAR. 42 See R. Veil, ‘Europäisches Insiderrecht 2.0: Konzeption und Grundsatzfragen der Reform durch MAR und CRIM-MAD’ (2014) Zeitschrift für Bankrecht und Bankwirtschaft (ZBB), pp. 85, 87. 43 See Recital 5 MAR. 44 See Article 3(Ia) TD; on this approach see R. Veil, ZGR 2014, pp. 544, 570. 45 See Recital 5 MAR. 46 The disclosure obligation has been part of the regime of so-called Regulation A offerings (Section 3 Securities Act and SEC Rules 251–63). 47 Cf. NXT, ‘Guidance Note: Key Operating Milestones’, February 2015, p. 4

48 The NXT standard is defined as follows in the ‘Guidance Note’: ‘The NXT standard means that an NXT company’s KOMs, when taken together, address the most significant factors by which the performance of the company’s or the company’s group’s business should be assessed and monitored and will result in understandable reporting for investors.’ 49 Cf. S. Franks, ‘NXZ gets key measure right in NXT market’, 19 September 2015, available at . 50 In more detail Veil (n. 5), 153 f.

51

In the USA, the SEC abolished alternative disclosure models as they would be a ‘lessuniform disclosure in a potentially unfamiliar format’. Cf. SEC, Amendments for Small and Additional Issues under the Securities Act (Regulation A), 25 March 2015, p. 95. 52 Currently, only three companies are listed at the NXT market with a market capitalization between $10.000 and $70.000 (see ). 53 Cf. M. Bell, ‘Continuing Obligations and Transactions’, in Nicholls and Graham (eds) (n. 7), 140. 54 Cf. Article 11 AIM Rulebook about the disclosure of price-sensitive information. 55 Cf. SEC, SEC Release No. 33-9741, ‘Final Rule: Amendments for Small and Additional Issues under the Securities Act (Regulation A), 25 March 2015, p. 234. 56 On the role of listing standards in a context of asymmetric information see e.g. T. Chemmanur and P. Fulghieri, ‘Competition and Co-Operation Among Exchanges: A Theory of Cross-Listing and Endogenous Listing Standards’ (2006) Journal of Financial Economics 82(2), 455–89. 57 Article 33(7) MiFID II also provides for secondary listing SME growth markets. Secondary listing is allowed only upon express or tacit approval by the issuer. Such approval may not lead to the application of further obligations relating to corporate governance or initial, ongoing, or ad hoc disclosure. 58 M. Kahan, ‘Some Problems With Stock Exchange-Based Securities Regulation’ (1997) Virginia Law Review 83(7), 1517; J. R. Macey, M. O’Hara, and D. Pompilio, ‘Down and Out in the Stock Market: The Law and Finance of the Delisting Process’ (2008) Journal of Law and Economics 68(4), 683 ff. 59 B. Steil, ‘Changes in the Ownership and Governance of Securities Exchanges: Causes and Consequences’, Brookings-Wharton Papers on Financial Services, 2002, 72 ff . 60 J. R. Macey and M. O’Hara, ‘From Markets to Venues: Securities Regulation in an Evolving World’ (2005) Stanford Law Review 58, 563 ff. 61 C. Di Noia, ‘The Stock-Exchange Industry: Network Effects, Implicit Mergers, and Corporate Governance’ (1999) Quaderni di Finanza Consob, 33.

14 DARK TRADING UNDER MIFID II Peter Gomber and Ilya Gvozdevskiy

I. Introduction II. 1. 2. 3.

Dark Pools: Motivation, Classification, and Related Literature Motivation Classification of Dark Pools Dark Trading in Academic Literature

III. Dark Pools under MiFID I 1. Venue Classification and Pre-Trade Transparency under MiFID I 2. Waivers to Pre-Trade Transparency IV. 1. 2. 3.

Dark Pool Regulation under MiFID II/MiFIR Waivers for Equity Instruments Volume Cap Mechanism Regulatory Technical Standards Relating to the Double Volume Cap Concept 4. Where can Investors Trade? Trading Obligations for Investment Firms 5. MiFID II: Ready Solutions

V. Summary

I. Introduction

[14.01] The implementation of MiFID I in November 2007 increased competition between traditional exchanges (regulated markets—RMs) and new players entering the European securities market (multilateral trading facilities—MTFs). Even though RMs still keep the larger share of turnover per national main index, the MTFs continually gained market share by providing pan-European trading services. In 2011, the two largest European MTFs, BATS Europe and Chi-X Europe (CXE), merged to become BATS Europe, which was granted the status of RM1 in 2013. Today, although the incumbent exchanges have managed to keep the largest market share in their respective national stocks and main indices, BATS Europe (CXE Book) has succeeded in becoming number two in the overwhelming majority of European national markets2 (see Figure 14.1), and is currently the largest European equity trading venue. The largest MTF in equity trading by market share is Turquoise.

Figure 14.1: Market share of national exchanges, CXE Book and Turquoise. (Source: authors’ representation based on data delivered by BATS Europe for the year 2015.)

[14.02] However, a significant part of turnover is still executed in the over-the-counter (OTC) markets, which are characterized by missing or limited transparency and pre-trade anonymity of both orders and traders. Having less regulatory supervision compared to RMs or MTFs, the OTC

markets and venues can deliver better execution conditions to investors by providing midpoint executions, lowering the spread and hiding the intentions and identity of investors behind the transaction. As a result, the market share in the OTC market was increasing over time and reached levels around 35 per cent in 2014 (see Figure 14.2). Gomber et al. claim that not only large institutional orders but also many executions of retail size take place on an OTC basis.3

Figure 14.2: Turnover market share report for January 2008 to July 2014. (Source: authors’ representation based on Thomson Reuters Market Share Reports January 2008 to July 2014.)

[14.03] The introduction of waivers to the transparency requirements of MiFID I permits trading venues not to show orders under certain conditions in the transparent (also called ‘lit’) order book. This has fostered the existence of ‘dark liquidity’, that is, orders that are hidden from market participants and therefore do not contribute to pre-trade transparency. [14.04] Originally, dark pools (often also called ‘crossing networks’) were created to enable institutional investors to execute large blocks of shares in

one lot on the market without adversely affecting their own execution price due to possible market impact. In addition, the identities of order-initiating parties remain hidden until execution. [14.05] On one hand, dark trading and dark pools are provided by RMs and MTFs which constitutes the so-called regulated dark pools. On the other hand, unregulated dark pools, which are not regulated as trading venues, can be found in the OTC market. [14.06] Regulated dark pools such as Liquidnet, Turquoise, and ITG are becoming more and more popular and are significantly gaining in turnover. In 2015, the total turnover of Liquidnet increased by 9.8 per cent compared to 2014,4 adding up to $133 billion. Turquoise MidPoint experienced a 26.7 per cent increase in turnover and a 22.9 per cent increase in volume for the same period. The value traded on ITG also rose by 45 per cent and volume by 25 per cent from December 2014 to December 2015.5 [14.07] MiFID II,6 the new directive for markets in financial instruments, and MiFIR,7 the respective regulation, which will have to be applied from January 2018,8 aim to shift significant parts of dark turnover to regulated lit markets by strengthening the transparency framework. A main instrument to achieve this is the introduction of the so-called double volume cap (DVC) regime imposed on transactions that are executed under the pretrade transparency waivers of MiFID I for negotiated transactions and reference price trades, such as trades at the midpoint of an RM. The regime introduces two thresholds valid on a per security basis that limit turnover under these pre-trade transparency waivers on individual venue and Union levels by 4 per cent and 8 per cent respectively. If they breach the limit, the respective venue or correspondingly all venues will be suspended from trading in that instrument under the two waivers for six months. [14.08] Against this background, the goal of this chapter is to analyse dark trading as well as the related regulatory requirements under MiFID I and MiFID II and to discuss the possible influence of the MiFID II/MiFIR implementation on dark trading and dark liquidity in Europe.

[14.09] The remainder of this chapter is structured as follows. Section II provides the motivation behind dark pool trading and a classification of dark pools. After that, it presents theoretical and empirical literature that deals with modelling and analysis of dark trading. Section III is devoted to MiFID I and contains the respective classification of execution venues and related transparency requirements. Section IV introduces the future dark trading regulation under MiFID II/MiFIR, including the volume cap mechanism and the trading obligation for equities. Moreover, this section presents the new market models and trading venues that are already set up against the background of the future regulation. Section V concludes.

II. Dark Pools: Motivation, Classification, and Related Literature [14.10] The concept of dark pools is closely related to the concept of dark liquidity. Dark liquidity implies orders with partial or missing pre-trade transparency, meaning that these orders are hidden from other market participants.

1. Motivation [14.11] Investors that are willing to trade large lots of securities may wish to hide respective orders from other market players for multiple reasons. The motivation behind the existence of dark pools is threefold.9 [14.12] First of all, trading in dark pools reduces the market impact of large orders. The concept of market impact (MI) relates to the implicit transaction costs which have to be borne by the party initiating the transaction. MI consists of two components: liquidity premium (LP) and adverse price movement (APM).10 LP corresponds to the half of the current bid–ask spread defining the costs of immediate liquidity demand. An order faces APM if it sweeps the order book and aggressively executes multiple limit orders on the opposite side of the order book. In dark pools and in OTC trading, transactions are mostly executed at a reference price often

represented by the midpoint of the most liquid RM for that security which prevents traders from facing implicit transaction costs. The second reason is price improvement based on the execution at the reference price. In this case, traders save (win) the difference between the best offer (bid) and midpoint if they initiate a buy (sell) transaction in comparison to the execution price on a lit market. [14.13] Finally, dark trading reduces information leakage. Compared to posting a large limit order in a transparent order book, the hidden orders do not reveal order limits or volume to the market. [14.14] The advantages of dark pools lead to increasing trading volumes on these venues. Figure 14.3 shows the path of total turnover generated by regulated dark pools in Europe from January 2008 to July 2014. During that time window, the turnover grew from €0.38 billion to €60 billion.11 The dashed line reflects the linear trend and slopes upwards with an average monthly increment of €0.7 billion.

Figure 14.3: Dark order book total turnover (solid) in billions of euros, and trend line (dashed). (Source: authors’ representation based on Thomson Reuters Market Share Reports January 2008 to July 2014.)

[14.15] According to the BATS Europe report, the proportion of dark pool turnover in total volume generated by RMs and MTFs fluctuated at around 7.5 per cent during 201512 (see Figure 14.4).

Figure 14.4: Proportion of dark pool turnover as a percentage of total volume. (Source: authors’ representation based on data delivered by BATS Europe for the year 2015.)

2. Classification of Dark Pools [14.16] Dark pools can be operated by RMs and MTFs. Furthermore, they appear in trading on an OTC basis. [14.17] RM- and MTF-operated dark pools are classified as regulated dark pools (see Figure 14.5). RM-sponsored dark pools were mainly introduced in response to the liquidity loss resulting from MTF activity. RMs operate dark order books in order to avoid market impact and provide better execution conditions for members and their customers. Examples of RM-operated dark pools are Xetra Midpoint of Deutsche Börse and Smartpool of Euronext. MTF-operated dark pools were originally designed to trade large lots of securities. For example, Liquidnet13 reached $1.8 million in average execution size. Another representative, ITG POSIT, founded in 1987, gains an average daily volume of 12.1 million shares with average trade size of 15,393 shares.14

Figure 14.5: Classification of dark pools.

[14.18] Dark orders also exist in the order books of RMs or MTFs as hidden, iceberg, or stop orders. They can be executed against other dark orders or combined with the orders of lit markets during auctions. [14.19] Unregulated dark pools are part of the transactions conducted on an OTC basis. OTC markets have no obligation to fulfil pre-trade transparency requirements. This type of dark pools is operated by full service brokers who provide their clients with a variety of services, including dark trading. OTC dark pools are often referred to as broker/dealer crossing networks (BDCN) which CESR,15 the predecessor of

ESMA, defined as ‘internal electronic matching systems operated by an investment firm that execute client orders against other client orders or house account orders’. This type of pools includes, for example, SigmaX (Goldman Sachs), Crossfinder (Credit Suisse), MS POOL (Morgan Stanley), and Super X (Deutsche Bank) (see Figure 14.5). [14.20] Figure 14.6 presents the order flows within a typical full service broker dark pool. There are two basic types of customer order flow: directed order flow and non-directed order flow. Directed order flow represents orders where customers explicitly specify their targeted execution venue. Here, the customer uses the broker ID for the purpose of direct market access (DMA), thereby routing orders through the broker system directly to RMs, MTFs, or other (regulated) dark pools. [14.21] Non-directed orders can be submitted either through the classic telephone calls or in an automated way via order management systems (OMS) of institutional customers applying mostly the FIX protocol or via internet brokerage facilities by retail customers. In case of non-directed orders, the broker has to decide where to place the order, for example by using either an internal order-routing engine or the discretion of a sales trader.

Figure 14.6: Full service broker dark pool (BDCN).

[14.22] By implementing a dark book, the broker enables his customers to directly access the dark book via electronic order entries applying the FIX protocol. Based on the reference prices of the (lit) RMs or MTFs, these dark books not only enable execution of these customer orders against each other (like in an RM or MTF dark pool) but they also increase execution probability by including proprietary orders of the broker (internalization), telephone orders received by the sales traders, and other (non-directed) retail or institutional orders. Thereby, these BDCNs leverage the diverse order flow that brokers receive and try to execute as much as possible of this order flow internally instead of routing it to RMs or MTFs. [14.23] Another type of unregulated dark pool is broker consortia dark pools, that is, liquidity pools operated by several brokers simultaneously; examples include BIDS (BIDS Trading LP) or LeveL (LeveL ATS).16

3. Dark Trading in Academic Literature

[14.24] The majority of academic literature devoted to dark pools analyses the effect of dark trading on liquidity and the price discovery process. Basically, this literature can be subdivided into theoretical and empirical papers.

A. Theoretical Papers [14.25] Theoretical papers model and analyse order flows of informed and uninformed (liquidity) traders that may decide between two execution venues where one is a regular exchange and another is represented by a crossing network (dark pool). In the models, informed traders are aware of the true value of the underlying security (asset, instrument) while the behaviour of the liquidity traders is supposed to be stochastic. Subject to a set of assumptions, theoretical papers discuss the resulting interaction and consequences of dark pool trading in terms of generated order flows, turnover, spreads, and liquidity changes on both lit markets and dark pools. [14.26] Hendershott and Mendelson17 claim that cost advantage and trading volume are the main determinants of crossing networks’ competitiveness. Along with the positive (liquidity) externality motivated by the presence of dark pools (crossing networks) on the market, the authors discover a negative crowding effect taking place when traders having low-liquidity preferences compete with high-liquidity-preference traders. This competition results in diminished execution probability for the traders having high-liquidity preferences. Thus, after reaching a certain critical mass, a further increase of order flow in crossing networks leads to a reduction in overall welfare. [14.27] Duffie et al.18 build a stochastic model that relies on marketmaking and asset pricing based on search and sequential bargaining processes. They discover that bid–ask spreads that investors face on the market can be tighter if investors are sophisticated, that is, if they have alternatives for immediate execution and opportunities to switch to another counterparty or simultaneously access multiple market-makers. Improved search alternatives foster market makers to give better prices.

[14.28] Degryse et al.19 construct a model of dynamic trader arrivals that can choose between the continuous dealer market (DM) and a crossing network (CN) operating under three types of informational setting: transparency, partial opaqueness, and complete opaqueness. The authors conclude that traders with a higher relative willingness to trade prefer the dealer markets expecting higher execution probability. Crossing networks attract traders that are willing to save half of the bid–ask spread and that would quit trading in the absence of this opportunity. In the complete opaqueness regime where DM and CN are isolated, the CN may contribute to the welfare if the execution probability is high but the value of the underlying asset is low. Coexistence of DM and CN creates higher welfare than isolated DM only for assets with large relative spread. The resulting positive effect offsets the negative crowding impact explained by Hendershott and Mendelson. [14.29] Buti et al.20 show that dark pools present a positive externality in terms of liquidity for traders. The introduction of dark pools to their model attracts order flows from the limit order book, tightening the spread for liquid stocks and widening the spread for illiquid stocks. At the same time, the overall trading turnover rises. The share of dark pools is higher for deeper markets and narrow spreads. [14.30] The paper of Ye21 extends the Kyle22 set-up which has one source of uncertainty in the model: execution price. Ye embeds an endogenous execution probability into the model, that is, the resulting market model accounts for both uncertainties: the price of execution and the probability of execution. The model contains three types of agents: informed traders that can choose between a lit exchange and a crossing network (dark pool), liquidity traders that trade for exogenous reasons, and market-makers quoting on the exchange. The orders of informed traders can be immediately executed on the lit exchange generating market impact or can be routed to the dark pool where the probability of execution decreases in the volume of the order. Price discovery within the model depends on the activity of liquidity traders. Increase of liquidity trading on the lit exchange improves price discovery while trading in the dark pool reduces it.

[14.31] Zhu23 investigates the impact of exchange based dark trading on the price discovery process. His model is based on strategic venue selection by informed traders and uninformed liquidity traders. The order flows of the informed traders are correlated with each other and with the value of the underlying asset. All informed traders place orders on the same market side intending to trade in the same direction. As a result, traders have to bear execution risks entering the dark markets. On the contrary, the order flows of the uninformed traders are uncorrelated and have lower execution risk and higher probability of execution in the dark pools. The authors conclude that the introduction of dark pools improves price discovery since informed traders choose to trade on the lit markets minimizing the risk of execution.

B. Empirical Papers [14.32] Empirical papers around dark trading investigate the reasons behind the market share distribution between OTC markets, dark pools, and lit markets and try to discover and explain the dependencies between dark and lit turnover as well as the impact of that relationship on market quality. [14.33] O’Hara and Ye24 investigate the topic accompanying dark trading: market fragmentation. They analyse off-exchange turnover and turnover generated by the lit markets. The authors discover that off-exchange venues execute 30 per cent of all equity volume and conclude that a higher level of fragmentation, that is, the existence of the off-exchange markets, results in lower transactions costs, faster execution, and better market quality. On the other hand, fragmentation contributes to short-term volatility. [14.34] Hatheway et al.25 investigate dark trading on the US market, applying a sample of 116 stocks listed on NYSE and NASDAQ. They distinguish three main properties of dark trading: exemption from the fair access requirement, pre-trade opacity, and sub-penny executions. The authors derive that dark trading attracts uninformed traders who cannot contribute to price discovery. The lit market, as a result, retains informed traders. This fact significantly disincentivizes liquidity providers to compete for order flow.

[14.35] Nimalendran and Ray26 use a unique proprietary data set from a large crossing network to examine the linkages between dark and lit venues. The authors find that trades on the crossing networks may cause information leakages. The generated information is short term and cannot be explained using fundamental analysis. Trades of liquid stocks, trades in dark pools, and trades of large blocks reveal less information to the open market. Further, the authors discover algorithmic activity that can match information related to the activity on lit venues with executions in crossing networks. This may lead to price discovery, despite the hidden nature of the crossing networks. [14.36] Gomber et al. analyse 66 months of intraday data and conclude that more than 50 per cent of the total trading volume in EURO STOXX 50 stocks happens in OTC markets.27 Most of the OTC trades are of retail size, while less than 5 per cent of trades are classified as large-in-scale (LIS). The share of LIS OTC trades steadily decreased from 15 per cent in 2008 to 2 per cent in 2013. Thus, the authors conclude that the reality of OTC trading in Europe does not reflect the regulatory intentions of MiFID I. [14.37] In a subsequent paper, Gomber et al. search for determinants of trading volume on lit and dark venues represented by OTC, dark pools, and systematic internalizers (SIs), studying the order-routing decisions of investors.28 They discover that the level of anonymity and immediacy across venues explains related market shares. During times of high adverse selection, investors choose lit venues operating continuous trading systems or dark venues with high levels of execution certainty, immediacy, and anonymity. On the contrary, un-informed investors prefer dark venues in order to achieve better execution prices. [14.38] Degryse et al. study the relation between dark trading and hidden orders on lit markets.29 They conclude that these are substitutes and negatively affect each other. Hidden orders will be preferred during the high-volume days. Both types of trading are negatively affected by the presence of algorithmic traders, while the effect of total traded volume, spread, and depth cannot be determined unambiguously.

III. Dark Pools under MiFID I [14.39] In order to detail the regulatory basis for dark pools in MiFID I and against the background of the European trading landscape, in the following the basic concepts of MiFID I (trading venue classification, pretrade transparency requirements, and the waivers concept) will be described and linked to the Directive’s economic impact in Europe after November 2007.

1. Venue Classification and Pre-Trade Transparency under MiFID I [14.40] MiFID I identifies three types of execution venues: regulated market (RM), multilateral trading facility (MTF), and systematic internalizer (SI). [14.41] An RM is a ‘multilateral system operated and/or managed by a market operator, which brings together or facilitates the bringing together of multiple third-party buying and selling interests in financial instruments […] in a way that results in a contract, in respect of the financial instruments admitted to trading under its rules and/or systems’.30 According to the ESMA database,31 there are 103 registered RMs as of January 2016. Former MTF BATS Europe,32 London Stock Exchange, and Deutsche Börse are the largest three RMs in European equity trading and executed 43 per cent of the lit market trading volume, with a total transaction volume of 5,194 billion in 2015 (see Figure 14.7). [14.42] An MTF is a ‘multilateral system, operated by an investment firm or a market operator, which brings together multiple third-party buying and selling interests in financial instruments […] in a way that results in a contract’.33 ESMA reports MTFs to be the largest group of trading venues (150 MTFs). Turquoise (an MTF being part of the London Stock Exchange Group) has a leading position in Europe with a turnover of €967 billion34 and market share of 8 per cent among the lit markets (Figure 14.7).

Figure 14.7: Lit order books turnover in billions of euros, and respective market share as a percentage of the total lit turnover. (Source: authors’ representation based on data delivered by BATS Europe for the year 2015.)

[14.43] An SI is an ‘investment firm which, on an organized, frequent and systematic basis, deals on own account by executing client orders outside a regulated market or an MTF.’35 There are only eleven registered SIs in Europe, with a total market share well below 2 per cent. [14.44] MiFID I does not explicitly provide a definition for OTC markets. The Directive mentions OTC only once in Recital 53 and specifies that the characteristics of related transactions are ad-hoc and irregular and are carried out with wholesale counterparties and are part of a business relationship which is itself characterised by dealings above standard market size, and where the deals are carried out outside the systems usually used by the firm concerned for its business as a systematic internaliser.36

[14.45] The pre-trade transparency requirements37 under MiFID I for RMs and MTFs depend on the market model of the respective trading venue. Four different types of market models are specified: continuous auction order book trading systems, quote-driven trading systems, periodic auction trading systems, and other models that cannot be classified into the previous categories or hybrid systems that possess properties of multiple market models.

[14.46] Continuous auction order book trading systems match orders at the best available prices on a continuous basis by means of an order book and a trading algorithm without human intervention. The execution venues under this type of market model have to reveal the aggregate number of orders and of the shares those orders represent at each price level, for the five best bid and offer price levels. [14.47] Quote-driven trading systems operate by matching incoming client orders against the quotes provided by market-makers on a continuous and mandatory basis. The quotes are the binding commitments to buy and sell the respective shares and indicate the price and volume of shares in which the registered market-makers are prepared to buy or sell. For this market model, MiFID I requires that the best quotes are revealed, that is, best bid and best ask quotes of each market-maker and the number of shares related to these quotes. [14.48] Periodic auction trading systems match orders on the basis of a periodic auction and a trading algorithm operated without human intervention. The system displays the indicative price that would satisfy the system’s trading algorithm and the volume that would be executable at that price. [14.49] The final type of market model includes hybrid systems that have properties of two or more of the types listed above and other trading systems with a price-determination mechanism of different nature. These systems shall reflect adequate information that is allowed by the price discovery mechanism. [14.50] An investment firm (IF) may execute client orders outside of the RM or MTF on an organized, frequent, and systematic basis. If in addition this activity is performed professionally on a non-discretionary and regular basis, the IF shall be classified as an SI according to MiFID I. SIs have to publish quotes on a regular and continuous basis in those shares admitted to trading on an RM for which they are SI and for which there is a liquid market.38 Article 22 of Regulation 1287/2006 defines liquid market as conditions when a share is traded on a daily basis with a free float exceeding €500 million.

[14.51] RMs, MTFs, and SIs have to publish the required information on a continuous basis. Data must be accessible as soon it is available and must remain accessible for all market participants until it is updated.39 Moreover, the data has to be revealed in as close to real time as possible.40

2. Waivers to Pre-Trade Transparency [14.52] There is no explicit definition of dark pools in MiFID I. However, the Committee of European Securities Regulators (CESR) specified a dark pool of liquidity to be ‘a trading facility where there is no pre-trade transparency, i.e. where orders are not publicly displayed based on pre-trade transparency waivers provided by MIFID’.41 Waivers, according to Article 29 (for MTFs) and Article 44 (for RMs) in MiFID I, enable operators of trading venues to abstain from the pre-trade transparency requirements for certain types of orders and market models given that an incoming order can be classified as one of the four following types. [14.53] Reference price (RP) systems42 are the systems where the execution price is determined in accordance with a reference price generated by another system, where that RP is widely published and is regarded by market participants as a reliable RP, for example the European Best Bid or Offer (EBBO)43 or primary best bid and offer (PBBO).44 In most reference price systems, eligible orders are executed against each other at the midpoint of the liquid reference market, mostly applying volume/time priority. [14.54] Figure 14.8 shows an example of the mechanism in reference pricing. Here, the midpoint of the lit market equals is average between the best bid and best offer (midpoint = €100). This midpoint is imported to the dark book as a reference price (see stream (a) of Figure 14.8). The dark book already contains two offers of different limit, quantity, and time stamp. Assume that at 9:03:00 a midpoint buy order with a volume of 30,000 shares and a limit of €101 enters the system (stream (b) of Figure 14.8). This midpoint order will be executed at the midprice of the reference market, that is, the execution price will be favourable for the investor since midpoint = €100 < €101 = best offer price. This triggers the following set of

executions (stream (c) of Figure 14.8) according to the volume/time priority: (1) 20,000 @ €100 (2) 10,000 @ €100. The dark executions at midpoint bring price improvements for both buyers and sellers relative to their order limits and relative to the price they would get on the reference market.

Figure 14.8: Reference price trading.

[14.55] The second type of waiver is applicable to negotiated trade (NT) systems45 which formalize negotiated transactions given that transactions take place at or within the current volume-weighted spread reflected in the order book or the quotes of market-makers or within a percentage of a suitable reference price. Alternatively, the transaction can be subject to conditions other than the current market price of the share. [14.56] The third type refers to orders that are held in an ordermanagement facility46 (OMF) maintained by an RM or MTF pending these orders being disclosed to the market. Such types of orders include, for example, iceberg orders where the peak (child order) of the order is shown to the market while the overall quantity (parent order) remains hidden and will be revealed successively as soon as the peaks are executed. Another example of an OMF order is a stop order which is activated only if the price

of an asset breaches a certain explicitly predefined price threshold. Traders use stop buy orders to follow a trend by buying an asset if the price exceeds the predefined price threshold and apply stop loss orders to limit losses by selling an asset when the price goes below the predefined price threshold. [14.57] Finally, in order to benefit from the large-in-scale (LIS) waiver47 the order will be large compared to one of a ‘normal’ market size (NMS) if it is at least of a size as displayed in Table 14.1, depending on the average daily turnover of the share in question. Table 14.1: Large-in-scale order size (in euros).

(Source: Regulation (EC) No. 1287/2006.)

[14.58] If an incoming order is subject to one of these waivers, the order may remain hidden. The regulated dark pools that apply these waivers attracted a turnover of €982 billion in 2015. The leading top five venues cover 79 per cent of the dark pool market (see Figure 14.9).

Figure 14.9 Dark order books turnover in billions of euros, and respective market share as a percentage of total dark turnover. (Source: authors’ representation based on data delivered by BATS Europe for the year 2015.)

IV. Dark Pool Regulation under MiFID II/MiFIR [14.59] MiFID I was introduced in 2004 and had to be applied from November 2007. The Directive aims to improve the competitiveness of EU markets and to ensure a higher degree of protection for investors. The evolution of markets resulted in the development of new trading systems that fall outside of MiFID I, which lead to the necessity of a review of the Directive.48 [14.60] The new Directive, MiFID II, and its accompanying Regulation, MiFIR, were published in June 2014. One important goal of MiFID II is to move trading of financial instruments into regulated platforms. The new market framework introduces a new category of trading venues called organized trading facility (OTF) for non-equity instruments to be traded on a multilateral platform. OTFs are broadly defined to capture new types of organized execution49 and are subject to similar regulatory requirements as MTFs.

[14.61] Furthermore, by the means of a trading obligation, the new directive ensures that investment firms performing internal matching will be authorized as MTFs or SIs. Non-systematic, ad hoc, irregular, and infrequent transactions as well as transactions that do not contribute to the price discovery process, however, can be further fulfilled on the OTC basis. Competition in trading and clearing of financial instruments is improved by a harmonized regime for non-discriminatory access to central counterparties (CCPs) which means that a trading venue has the right to non-discriminatory treatment in terms of how contracts traded on its platform are treated in terms of collateral requirements and netting of economically equivalent contracts and cross-margining with correlated contracts cleared by the same CCP, and non-discriminatory clearing fees.50

[14.62] Furthermore, MiFID II introduces controls for algorithmic and high-frequency trading techniques. Following Recital (64) of MiFID II, investment firms applying algorithmic and high-frequency trading techniques must ensure that their strategies are properly tested, cannot be used for market abuse, and do not create a disorderly market. In this context, trading venues have to ensure resiliency of trading systems and the presence of circuit breakers that are able to halt trading during turbulent price movements. [14.63] Finally, the directive establishes new regulations for trading transparency and dark trading. This will be the main focus for further discussion in this section. [14.64] MiFID II/MiFIR retains the basic concept of pre-trade transparency for shares of MiFID I: market operators and investment firms operating trading venues have to make public ‘current bid and offer prices and the depth of trading interests at those prices’ calibrated for different types of trading systems.51 However, the classes of instruments covered are significantly extended: Article (3)(1) of MiFIR requires the application of pre-trade transparency requirements not only for shares but also for depositary receipts, ETFs, certificates, and similar financial instruments. Article (8)(1) of MiFIR further extends this concept to apply for bonds, structured finance products, emission allowances, and derivatives. As in the

case of MiFID I, the competent authorities are able to waive pre-trade transparency requirements for specific market models and order types. [14.65] The key new regulations of MiFID II concerning dark trading are the so-called volume caps for equity instruments. Furthermore, MiFID II introduces a trading obligation for share trading that might significantly affect trading venue decisions and specifically the market share distribution between OTC trading, Sis, and trading venues. [14.66] Therefore in the following, first the waivers for equity instruments under MiFID II are outlined in comparison to MiFID I including important aspects concerning the respective Regulatory Technical Standards (RTS); the double volume cap mechanism is then described in detail—again including the debate on the respective RTS. Finally, the trading obligation as a means to curb OTC trading in shares is presented. As multiple initiatives have been launched in the securities trading industry over the past two years that are already anticipating the future dark pool regulation (specifically the volume cap mechanism) by setting up or proposing new trading venues or new trading functionalities, those initiatives will be described in Section IV.5.

1. Waivers for Equity Instruments [14.67] In MiFID II, the four types of pre-trade transparency waivers for equity trading that were already defined in MiFID I also apply (reference price (RP), negotiated transactions (NT), large-in-scale (LIS) and order management facilities (OMF)). They are defined in Article 4 of MiFIR. However, some details in their new design differ from MiFID I and will be detailed below. The Reference Price waiver is valid for: systems matching orders based on a trading methodology by which the price […] is derived from the trading venue where that financial instrument was first admitted to trading or the most relevant market in terms of liquidity, where that reference price is widely published and is regarded by market participants as a reliable reference price.52

[14.68] Compared to the Regulation 1287/2006 (MiFID I, Level 2), the definition of MiFIR includes requirements for the reference market, which must be the market of initial public offering or the most liquid market. The midpoint53 of the appropriate trading venue can serve as a reference price in continuous trading. Outside continuous trading, the respective opening or closing price can also be applied as a reference price.54 [14.69] The definition of NT in MiFIR was extended by adding transactions ‘in an illiquid share, depositary receipt, ETF, certificate or other similar financial instrument that does not fall within the meaning of a liquid market, and are dealt within a percentage of a suitable reference price’.55 Furthermore, all transactions related to this waiver have to fulfil the rules of the trading venue56 and may not abuse, or attempt to abuse, the market.57 [14.70] Trading under the RP waiver as well as the waiver for NT made within the current volume-weighted spread fall under restrictions of the double volume cap (DVC) mechanism which is the central concept of dark pools regulation in the context of MiFID II. DVC will be discussed in further details in the next subsection. [14.71] The last waiver proposed by MiFIR relates to the ‘orders held in an order management facility of the trading venue pending disclosure’58 and remains unaffected compared to MiFID I. Based on the consultation paper (CP1)59 containing proposals regarding delegated acts, and discussion paper (DP)60 with proposals for technical standards, both published in May 2014, followed by a Final Report (FR1)61 and Consultation Paper (CP2)62 in December 2014, ESMA developed multiple Regulatory Technical Standards (FR3)63 published in September 2015. Some important aspects of these RTS64 are highlighted below. [14.72] Following the concerns of stakeholders presented in the final report (FR3), the most relevant market in terms of liquidity mentioned in the definition of RP waiver shall be the trading venue with the highest turnover for the related instrument. The calculation of trading turnover shall

account for all relevant trading sessions and exclude the RP, NT, and LIS transactions.65 [14.73] Article 4(1)(b)(iii) related to the NT refers to conditions other than the current market price meaning that the underlying negotiated transactions shall not be reported under the waiver if they do not contribute to the price formation process.66 [14.74] The order-management facility waiver refers to orders that are exempted from pre-trade transparency pending disclosure to the market. Reserve or iceberg orders and stop orders are the examples explicitly discussed by ESMA. The minimum order size of stop orders and iceberg orders must be defined by the trading venue. The reserve orders must exceed €10,000.67 [14.75] As in MiFID I, orders eligible for the LIS waiver68 must exceed normal market size. However, in the proposed MiFID II RTS (see Table 14.2), LIS specifications differ from the classification provided by Regulation 1287/2006 in the context of MiFID I. The former smallest class for orders in shares of an average daily turnover (ADT) under €500,000 is now split into three classes with: (i) ADT < €50,000, (ii) €50,000 ≤ ADT < €100,000, and (ii) €100,000 ≤ ADT < €500,000. This new classification must define adequate trading transparency for less-liquid stocks with low turnover and enable them to benefit from the LIS waiver. Table 14.2: Large-in-scale order size (in euros) for shares and depositary receipts.

(Source: RTS Final Report (2015/1464))

[14.76] The LIS threshold for ETFs is set to be €1,000,000 regardless of their liquidity or the liquidity of their underlying.

2. Volume Cap Mechanism [14.77] In order to protect the price discovery process on public markets by limiting the amount of orders that are hidden in dark venues, MiFIR restricts the value of transactions executed under the RP and the NT waivers (Article 4(1)(a) and 4(1)(b)(i)) on an instrument-by-instrument basis. However, transactions under the LIS and OMF waivers remain unaffected by the double volume cap mechanism. [14.78] The turnover in dark venues is restricted by a DVC mechanism69 which states that: (a) the percentage of trading in a financial instrument carried out on a trading venue under those waivers shall be limited to 4 per cent of the total volume of trading in that financial instrument on all trading venues across the Union over the previous 12 months. (b) overall Union trading in a financial instrument carried out under those waivers shall be limited to 8 per cent of the total volume of trading in that financial instrument on all trading venues across the Union over the previous 12 months. [14.79] The first cap of the mechanism restricts the usage of waivers by individual trading venues. If in a specific equity instrument the proportion of total value traded under NT and RP waivers together exceeds 4 per cent of total trading volume in that instrument, the competent authority has to suspend that venue from trading under the NT and RP waivers for six months. [14.80] The second cap suspends all trading venues across the Union from trading under NT and RP waivers, if the cumulated proportion of total value traded under these waivers across the Union exceeds 8 per cent of total European trading volume in that instrument. [14.81] The volume cap mechanism only applies to those negotiated trades that were executed within the volume-weighted spread,70 which implies that transactions in illiquid shares, depositary receipts, ETFs, certificates, and transactions subject to conditions other than the current market price are not influenced by the volume cap restriction.

[14.82] ESMA must publish on a monthly basis the total trading volume within the Union for each financial instrument in the previous twelve months. In case of a threshold breach, the competent authority has to suspend trading under related waivers within two working days,71 both for individual venues if the 4 per cent threshold is breached and for all execution venues across the Union if the 8 per cent threshold is breached. [14.83] In addition, if the proportions of relevant trading volumes mentioned above reached (exceeded) the values of 3.75 per cent for the first restriction on a single venue level and 7.75 per cent for the second restriction valid for all venues across the Union, ESMA has to publish additional notification after the fifteenth day of a calendar month. [14.84] The DVC mechanism triggers additional burden and workload for venue operators in terms of processes and IT infrastructure that is required to track and publish the volumes on a reliable basis. The operators must have in place systems72 that enable the identification of all trades which have taken place on their venues under the NT and RP waivers. Furthermore, operators have to ensure that the trading volume does not, under any circumstances, exceed 4 per cent of total trading volume permitted under the NT and RP waivers in a specific financial instrument. [14.85] The Regulation requires ESMA to start the period for the publication73 of trading data on 3 January 2017.74

3. Regulatory Technical Standards Relating to the Double Volume Cap Concept [14.86] Article 5(9) of MiFIR requires ESMA to develop draft RTS in the context of the DVC. These draft RTS—which are delivered to the EU Commission for approval and final adoption—must detail the method for gathering and publishing the data required for calculating total trading volumes and market shares of trades executed under the NT and RP waivers.

[14.87] In its draft RTS,75 delivered in September 2015, ESMA proposed two sources of volume evaluation. The first consists of reports submitted by trading venues to respective competent authorities over a twelve-month rolling window period. The second source would be the data collected by consolidated tape providers (CTPs) that operate across the Union and collect post-trade information. The latter source will be used to validate the data provided by the venues. With respect to the frequency of the calculations and publications, ESMA suggested data reports on a bi-weekly basis in order to have time for correction of potential errors and to be able to publish notification concerning the 3.75 per cent and 7.75 per cent thresholds which will be available within five working days after the fifteenth day of the month. [14.88] ESMA is aware of the fact that the same transactions may cause repeated suspension from trading. However, if the venue continues to operate the lit market during the suspension time, the share of transactions under the waiver regime will decrease in relation to the total turnover. As a result, after six months the share will fall below the breached threshold and will be sufficient again for the trading venue to proceed with trading under waivers. [14.89] Concerning the alternatives of reporting the turnover values (price multiplied by quantity) versus the volume (total quantity only), the proposed text of technical standards requires the trading venues to publish turnover value data in the original currency. ESMA takes over the responsibility for currency conversion. [14.90] If venues have to report with respect to financial instruments for which there is less than twelve months of data available, ESMA provides no exception including newly listed assets. The data has to be reported even if the instruments were traded during limited time. [14.91] Moreover, the reported data has to employ unique identifiers for each trading venue. In case of RM dark pools the reported data shall contain two separate identifiers. This is essential in the context of the DVC mechanism. Thus, the templates prepared by ESMA for reporting purposes shall distinguish all trading venues authorized under MiFID II.

4. Where can Investors Trade? Trading Obligations for Investment Firms [14.92] Market transparency is a central concept of both MiFID I and MiFID II/MiFIR. Recital 6 of MiFIR argues that all possible types of trading systems, including BDCN, must be properly regulated and authorized as MTFs or SIs. [14.93] The new trading obligation in Article 23 MiFIR requires IFs to execute trades in shares only on RM, MTF, SI, or third-country alternatives unless the trades are non-systematic, ad hoc, irregular, and infrequent, or are carried out between eligible and/or professional counterparties, and do not contribute to the price discovery process.76 Article 23(2) MiFIR clarifies that an investment firm performing internal matching to execute client orders on a multilateral basis must be authorized as an MTF with all consequent compliance requirements. Thus, the trading obligation limits the capability of IFs to trade on an OTC basis and will shift volumes from OTC trading into regulated venues.

5. MiFID II: Ready Solutions [14.94] Dark trading opportunities are highly relevant for buy-side traders that try to execute large orders with minimum market impact. They will also expect their brokers to identify suitable order execution channels in a post-MiFID II world. Therefore, the securities trading industry will try to avoid transactions that might bring specific instruments into the threat of falling under the volume caps. Anticipating the future challenge of the cap regime, a lot of market participants and market operators in 2014 and 2015 announced or even launched solutions that allow dark execution for their customers not to be subject to the double volume cap regime. Among others these solutions include: (i) the BATS Europe Periodic Auction Book market model, (ii) the Plato Partnership project, (iii) the Volume Discovery Order introduced by Deutsche Börse, (iv) Turquoise’s Block Discovery service, and (v) the Mid-Price Pegged Order functionality of the London Stock Exchange.

[14.95] In September 2015 BATS Europe introduced the Periodic Auctions Book77 model, which operates independently from other order books of the market operator. Orders are executed through an auction functionality that is available through the whole trading day. Every auction has a random end and executions are only possible within the European best bid and offer price range. As for auctions, specific pre-trade transparency requirements have to be applied, though orders of this market model are exempted from the waiver regime. [14.96] The Plato Partnership project is a consortium of eight banks and seven fund managers78 created at the end of 2014. Plato is a not-for-profit dark pool venue for anonymous block trades aiming for cost reduction, market structure simplification, and better liquidity conditions.79 Currently, the consortium is in the process of identifying partners that will provide technology and market models to prevent the dark pool falling under the DVC. The partnership will use revenue generated from trading to fund academic research. [14.97] In the new Xetra release 16.0 (live from late 2015),80 Deutsche Börse presented the Volume Discovery Order (VDO) which extends the functionality of iceberg orders. Aside from the visible part (peak), a VDO requires an additional limit valid for the hidden part. The hidden volume is available for matching against other hidden parts of VDOs at the current midpoint, that is, the second limit serves as a barrier for the executable volume and will not be used for price discovery. As VDOs have a minimum executable quantity above LIS, it is exempted from pre-trade transparency under the LIS waiver and therefore does not fall under the DVC regime. [14.98] Turquoise introduced the Block Discovery81 service as an enhancement to the existing Midpoint Dark Book in October 2014. Clients that are willing to trade LIS orders can submit conditional orders also known as block indications to the system. In the case of a potential match, the system invokes the counterparties involved in the potential transaction to confirm their trading intentions by sending a Qualifying Block Order. If this succeeds, the orders are executed within a 1.5-second time interval with a randomized end.

[14.99] In late 2014, the London Stock Exchange (LSE) announced the new hidden Mid-Price Pegged Order functionality,82 which is designed to facilitate LIS dark trading, allowing clients to interact with lit and dark liquidity that exceeds the value of pre-defined minimum execution size. The Mid-Price Pegged Orders are matched either against other Mid-Price Pegged Orders or against other orders exceeding in volume the minimum execution size of the Mid-Price Pegged Order. [14.100] The examples show that market operators as well as buy- and sell-side firms are in the process of preparing for the future trading ecosystem under a MiFID II DVC regime. Specifically, they are launching new functionalities that help to protect the interests of institutional investors. Most initiatives implement this by providing market models and trading functionalities that avoid a classification of orders under the NT and RP waivers, and try to bring these under the umbrella of the LIS waiver, which is not affected by the DVC mechanism.

V. Summary [14.101] MiFID II, which has to be applied from 3 January 2018, aims to add more transparency to European securities trading by bringing turnover that is today executed on an OTC or dark pool basis into a more transparent framework. The directive retains the waiver regime proposed in MiFID I, which empowered competent authorities to waive the pre-trade transparency requirements for certain types of orders and market models. [14.102] Relative to today, after the implementation of MiFID II/MiFIR, the LIS and OMF waivers will be largely unchanged. However, the NT and RP waivers are subject to a new DVC threshold regime. The caps are imposed on the turnover under the waivers both at an individual venue and overall Union level. [14.103] If the relative share of turnover in a financial instrument under the RP and NT waivers on a single trading venue exceeds 4 per cent or total trading in all dark pools exceeds 8 per cent of overall trading in that instrument across the Union, the respective competent authorities will

suspend the venue or all dark venues within two working days from trading under the waivers for six months. [14.104] Market participants believe that the DVC regime and the trading obligation will move dark pool and OTC trading to the regulated and transparent venues. In 2014 the LSE conducted a study applying the rules to the trading volumes. Ninety-nine out 100 FTSE83 constituents were subject to suspension from dark trading for six months, which suggests that dark trading will be certain to decrease after MiFID II enforcement. [14.105] However, not all observers agree that dark turnover will necessarily diminish. Richard Semark, president of UBS MTF says that ‘volume caps applying to small “dark orders” are likely to increase the amount of larger dark trades’.84 Rebecca Healey from TABB Group adds that the DVC does not cover all types and venues of dark trading. Moreover, it will be very hard to verify the turnover data, leaving venues in fear of the ‘danger of unwittingly breaching the caps’.85 And some of the execution venues have introduced MiFID II-ready solutions that keep the large orders hidden and qualify these as LIS. Thus, large transactions may still stay in the dark. [14.106] Both regulators and market participants are sceptical about the success and effectiveness of the new DVC regime. Markus Ferber, the European Parliament’s rapporteur for MiFID II said: ‘My personal conviction is that the double volume cap will not function. Now it’s my obligation to make them functionable.’86 In this context, Rob Boardman, CEO of ITG Europe, stated that: ‘There is a lot of demand for dark trading as the numbers show. Even with the upcoming caps in MiFID II, the market will find solutions to continue dark trading because it is proven to lower market impact and provide better execution.’87 And the European Commission itself sees a necessity in investigating DVC effectiveness; MiFIR provides for an ESMA review of the waiver regime two years after its application.88

1

In the UK, Recognized Investment Exchange (RIE).

2

The only exception is Dublin (Ireland) where BATS Europe placed third after the Irish Stock Exchange and Turquoise MTF. 3 P. Gomber, S. Sagade, E. Theissen, M. C. Weber, and C. Westheide (2015), ‘The State of Play in European Over-the-Counter Equities Trading’ (2015) Journal of Trading, Spring. 4 Source: (2016). 5 J. Bakie, ‘Dark Pool Figures Show Growing Global Demand Despite Controversy’ (2016) The Trade, 19 January . 6 Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU, OJ L 173/349. 7 Regulation (EU) No. 600/2014 of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Regulation (EU) No. 648/2012, OJ L 173/84 (‘MiFIR’). 8 The go live date of MiFID II/MiFIR was initially scheduled for 3 January 2017. In February 2016, the European Commission published a press release (EC 2016a) proposing a one-year delay of the application date due to the ‘exceptional technical implementation challenges faced by regulators and market participants’. 9 Gomber and Pierron, ‘MiFID: Spirit and reality of a European financial markets directive’, Policy Platform White Paper, Frankfurt am Main, Goethe University/Celent (November 2010), available at: . 10 P. Gomber and U. Schweickert, ‘Der Market Impact: Liquiditätsmaß im elektronischen Wertpapierhandel’ (2002), Die Bank 7, 485–9. 11 Own analysis based on Thomson Reuters Market Share Reports for January 2008 to July 2014. 12 Own representation based on BATS Global Markets—Europe, ‘Pan-European Market Volume Summary: 2015’, available at . 13 Jacky Howe/Liquidnet Europe, ‘Press Release: Liquidnet Announces Best Ever Performance in European Equities in 2015’, 21 January 2016, available at . 14 ITG/J.T. Farley, ‘Press Release: ITG Releases December 2015 U.S. Trading Volumes’, , 11 January 2016. 15 Committee of European Securities Regulators, ‘CESR Technical Advice to the European Commission in the context of the MiFID Review and Responses to the European Commission Request for Additional Information’, 13 October 2010, available at: .

16

Furthermore, there are dark pool aggregators which collect fragmented liquidity from multiple dark pools and can route client orders to respective markets using automated trading algorithms if required. Theses aggregators like Spotlight (Fidessa), Tradebook (Bloomberg), and SuperX+ (Deutsche Bank) do not constitute dark pools as such and therefore are not included in our classification. 17 Terrence Hendershott and Haim Mendelson, ‘Crossing Networks and Dealer Markets: Competition and Performance’ (2000) Journal of Finance 55(5), 2071–115. 18 D. Duffie, N. Garleanu, and L. H. Pedersen, ‘Over-the-Counter Markets’ (2005) Econometrica 73(6), 1815–47. 19 H. Degryse, M. van Achter, and G. Wuyts ‘Dynamic Order Submission Strategies with Competition between a Dealer Market and a Crossing Network’ (2009) Journal of Financial Economics 91, 319–38. 20 S. Buti, B. Rindi, and I. M. Werner, ‘Diving into Dark Pools’ (17 November 2011). Charles A. Dice Center Working Paper No. 2010-10; Fisher College of Business Working Paper No. 2010-03-010. Available at SSRN: . 21 Ye, Mao, ‘A Glimpse into the Dark: Price Formation, Transaction Cost and Market Share of the Crossing Network’ (9 June 2011). Available at SSRN: . 22 A. S. Kyle, ‘Continuous Auctions and Insider Trading’ (1985) Econometrica 53(6), 1315–35. 23 Haoxiang Zhu, ‘Do Dark Pools Harm Price Discovery?’ (2014) Review of Financial Studies 27(3), 747–89. 24 M. O’Hara and M. Ye, ‘Is Market Fragmentation Harming Market Quality?’ (2011) Journal of Financial Economics 100(3), 459–74. 25 F. Hatheway, A. Kwan, and H. Zheng, ‘An Empirical Analysis of Market Segmentation on US Equities Markets’ (15 November 2014). Available at SSRN: . 26 M. Nimalendran and S. Ray, ‘Informational Linkages Between Dark and Lit Trading Venues’ (2004) Journal of Financial Markets 17, 230–61. 27 See Gomber et al. (n. 3). 28 P. Gomber, S. Sagade, E. Theissen, M. C. Weber, and C. Westheide, ‘Distinct Dark Markets and the Determinants of their Trading Volume’ (2015) SAFE Working Paper. 29 H. Degryse, G. Tombeur, and G. Wuyts, ‘Two Shades of Opacity: Hidden Orders versus Dark Trading’ (2015) available at . 30 Directive 2004/39/EC (‘MiFID’): Article 4(1)(2)(14). 31 The ESMA Registers portal provides web visitors with information concerning the European regulatory framework for investment firms and credit institutions, available at: (accessed August 2016).

32

BATS Europe operates two lit and two dark order books still providing the books of the former Chi-X MTF after its acquisition by BATS Europe in February 2011. 33 Directive 2004/39/EC (MiFID I): Article 4(1)(2)(15). 34 2015 figure; data source: BATS Europe. 35 Directive 2004/39/EC (MiFID I): Article 4(1)(2)(7). 36 Directive 2004/39/EC (MiFID I): Recital (53). 37 Commission Regulation 1287/2006 implementing Directive 2004/39/EC as regards record-keeping obligations for investment firms, transaction reporting, market transparency, admission of financial instruments to trading, and defined terms for the purposes of that Directive, OJ L 241/1, ANNEX II Table 1. 38 Directive 2004/39/EC: Article (27)(1). 39 Commission Regulation (EC) No. 1287/2006: Article (29)(1). 40 ibid: Article (29)(2). 41 CESR, ‘Questions and answers on MiFID: Common positions agreed by CESR Members in the area of the Secondary Markets Standing Committee’, CESR/10-591, 6 May 2010. 42 Commission Regulation (EC) No. 1287/2006: Article (18)(1)(a). 43 ESMA/2011/241g: Table IV. 44 BATS Europe Dark Pool, Separate & Safe Midpoint Order Book, available at:

(accessed August 2016). 45 Commission Regulation (EC) No. 1287/2006: Article (18)(1)(b). 46 ibid: Article (18)(2). 47 ibid: Article (20). 48 Markets in Financial Instruments Directive (MiFID II): Frequently Asked Questions; European Commission, Brussels, 15 April 2014. 49 Regulation (EU) (MiFIR) No. 600/2014: Recital (8). 50 ibid: Recital (28). 51 ibid: Article (3)(1). 52 ibid: Article (4)(1)(a). 53 ibid: Article (4)(2)(a). 54 ibid: Article (4)(2)(b). 55 ibid: Article (4)(1)(b). 56 ibid: Article (4)(3)(a). 57 ibid: Article (4)(3)(b). 58 ibid: Article (4)(1)(d). 59 ESMA, ‘Consultation Paper MiFID II/MiFIR’ (22 May 2014) ESMA/2014/549 (CP1). 60 ESMA, ‘Discussion Paper MiFID II/MiFIR’ (22 May 2014) ESMA/2014/548 (DP).

61

ESMA, ‘Final Report: ESMA’s Technical Advice to the Commission on MiFID II and MiFIR’ (19 December 2014) ESMA/2014/1569 (FR). 62 ESMA, ‘Consultation Paper: MiFID II/MiFIR’ (19 December 2014) ESMA/2014/1570 (CP2). 63 ESMA, ‘Regulatory technical and implementing standards—Annex I MiFID II / MiFIR’ (28 September 2015) ESMA/2015/1464 (FR3). 64 The final report involves nine chapters, each defining regulatory and technical implementing standards (RTS and ITS). RTS 1–RTS 5 relate to the transparency requirements for shares, depositary receipts, ETFs, certificates including standards on the volume cap mechanism, and the trading obligation for derivatives. RTS 6–RTS 12 cover microstructural issues providing standards for the IFs engaged in algorithmic trading, venues allowing algorithmic trading, fee structure, and tick size regime. RTS 13–16 address data publication defining standards on authorization, requirements and publication of transaction data, data disaggregation and access in respect of central counterparties, venues, and benchmarks. RTS 17–18 and ITS 19 provide requirements for trading venues in terms of admission of securities to trading. RTS 20–21 are devoted to commodity derivatives including criteria for ancillary activity. RTS 22–25 cover market data reporting and clock synchronization. Finally, RTS 26 and 27–28 address clearing of derivatives and best execution, respectively. 65 ESMA/2015/1464 (FR3), p. 18. 66 ibid, p. 23. 67 ibid, p. 24. 68 Regulation (EU) (MiFIR) No. 600/2014: Article (4)(1)(c). 69 ibid: Article (5). 70 ibid: Article (4)(1)(b)(i). 71 ibid: Article (5)(2) for single venue and MiFIR Article (5)(3) for all execution venues across the Union. 72 ibid: Article (5)(7). 73 ibid: Article (5)(8). 74 Updated from 3 January 2016; see EC (2016b) for details. 75 ESMA/2015/1464 (FR3). 76 Regulation (EU) (MiFIR) No. 600/2014: Article (23)(1). 77 BATS Europe, ‘Periodic Auctions Book: Lit Book Dedicated to Intra-Day Auctions’ (2016) . 78 Plato Partnership, ‘About’ (2015) . 79 Plato Partnership, ‘Process’ (2015) . 80 Michael Krogmann and Holger Pratt, ‘Press Release: Announcement of Xetra Release 16.0’ (28 April 2015)

. 81 London Stock Exchange Group, ‘Press Release: Turquoise Block Discovery™ Goes Live With Seven Major Brokers’ (20 October 2014) . 82 London Stock Exchange Group, ‘Press Release: London Stock Exchange Introduces New MiFID II-Ready Trading Enhancements for Block Trades’ (14 October 2014) . 83 Tim Cave, ‘Dark Pool Caps Provide 99 Problems for UK Blue Chips’, Financial News (Paris, 15 April 2015) , accessed 1 June 2016. 84 Philip Stafford, ‘Markets Planning for a World after the Day of the Mifid’, Financial Times (London, 13 October 2015) . 85 R. Healey, ‘Vision Note: MiFID II Double Volume Cap: Slam Dunk or Air Ball?’ (London, 24 November 2015) . 86 John Detrixhe and Jim Brunsden, ‘European Dark-Pool Equity Trading Limits Won’t Work, Ferber Says’, Bloomberg (London, 15 April 2015) . 87 John Bakie, ‘Dark Pool Figures Show Growing Global Demand Despite Controversy’, The Trade (London, 19 January 2016) . 88 Regulation (EU) (MiFIR) No. 600/2014: Article (4)(7).

15 Derivatives Trading, Clearing, Stp, Indirect Clearing, and Portfolio Compression Rezah Stegeman and Aron Berket

I. Introduction 1. Background 2. Derivatives II. 1. 2. 3. 4.

The Trading Obligation The General Obligation What is Exchange Trading? Scope Miscellaneous

III. 1. 2. 3. 4.

The Clearing Obligation The General Obligation What is Central Clearing? Scope Miscellaneous

IV. 1. 2. 3. 4. 5.

The Straight-Through-Processing Obligations The General Obligation What is STP? Scope The Obligations in More Detail Miscellaneous

V. 1. 2. 3. 4. 5. VI. 1. 2. 3. 4.

The Obligations in Respect of Indirect Clearing The General Obligation What is Indirect Clearing? Scope The Obligations in More Detail Miscellaneous The Obligations in Respect of Portfolio Compression The General Obligations What is Portfolio Compression? Scope The Obligations in more Detail

VII. A Few Final Remarks

I. Introduction 1. Background [15.01] Once the smoke cleared after the global financial crisis of 2007/2008 many commentators agreed that ‘whilst OTC derivatives did not cause the global financial crisis of 2007/2008, they likely contributed to amplifying various problems and provided channels for systemic risk to propagate’.1 With this in mind, in September 2009 the G20 leaders called on their finance ministers and central bank governors to reach agreement on an international framework of reform in various areas, including OTC derivatives markets: ‘All standardised OTC derivative contracts should be traded on exchanges or electronic trading platforms, where appropriate, and cleared through central counterparties by end-2012 at the latest. OTC derivative contracts should be reported to trade repositories. Non-centrally cleared contracts should be subject to higher capital requirements.’2 Most of these G20 reform commitments have been fulfilled through the enactment of the European regulations CRR,3 MAR,4 and EMIR5 and changes to the European directives CRD and MAD. The remaining implementation steps of the European legislator regard the trading obligation which is provided in

Title V of MiFIR.6 The European legislator has also decided to add further rules relating to derivatives in this part of the MiFID II package. These include (i) a clearing obligation, (ii) rules on the timing of acceptance for clearing (straight-through-processing (STP) obligations), (iii) rules on indirect clearing arrangements, and (iv) rules on portfolio compression. [15.02] In this chapter, we discuss these five provisions of Title V of MiFIR as they and their related RTS stand at the time of writing (31 August 2016).7 While each of the topics could be afforded a book in its own right, we have sought to address the key points of each in the room available in this volume.8

2. Derivatives [15.03] For those not too familiar with derivatives, a short introduction seems in order. A derivative can be defined as a financial contract derived from at least one underlying asset, meaning that its value depends on the value of such underlying asset.9 The underlying asset can be anything, such as a security, a currency, an interest rate, a credit risk, a commodity, etc. Derivatives typically provide for future performance of obligations to exchange the (benefit of the) underlying assets that are subject to possible future market fluctuations. They therefore allow the market risk of the underlying asset to be transferred from one party to another. Parties removing risk are considered to be using derivatives for hedging purposes, whereas parties acquiring risk are betting that their counterparty will be wrong about the future value of the underlying asset and are therefore considered to be using derivatives for speculative purposes.10 [15.04] Derivatives can be classified into two categories: (i) exchangetraded derivatives (ETDs, also referred to as ‘listed derivatives’), such as futures, and (ii) over-the-counter (OTC) derivatives, such as interest rate swaps. MiFIR defines ‘exchange-traded derivative’ as a derivative that is traded on a regulated market (RM) (or on a third-country market considered to be equivalent to an RM), and as such does not fall within the definition of an OTC derivative as defined in EMIR.11 Equally, the EMIR definition

referred to states that ‘OTC derivative’ or ‘OTC derivative contract’ is a derivative contract the execution of which does not take place on an RM (or on a third-country market considered as equivalent to an RM). From these definitions it can be deduced that, for the purposes of EMIR and MiFID II, derivatives traded on an MTF or an OTF qualify as OTC derivatives rather than ETDs. [15.05] A further, but related difference between ETDs and OTC derivatives, is their level of customization to the needs of their users. ETDs are highly standardized products and thus generally available off the shelf.12 OTC derivatives on the other hand are generally more tailor-made in order to match the user’s specific needs; they are bespoke transactions negotiated and agreed between parties.13 One could draw a comparison with a train and a taxi; one drops you off close enough to where you want to be, whereas the other brings you to your destination exactly. For instance, a party wanting to hedge the interest rate risk on a five-year loan that it is due to take out in seven months’ time may want to use an interest rate swap specifically customized to these dates. Standardized products, such as interest rate futures, may not perfectly match the client’s requirements, exposing them to residual interest rate risk that they are unprepared to take on.14

II. The Trading Obligation 1. The General Obligation [15.06] The first obligation that Title V MiFIR introduces, is the trading obligation in respect of certain derivatives. Under Article 28, certain market participants concluding certain standardized derivatives transactions must do so on an RM, an MTF, an OTF, or an equivalent third-country trading venue.15 This obligation follows directly from the G20 reform commitment to trade standardized OTC derivative contracts on exchange or electronic trading platforms where appropriate,16 in order to improve transparency and mitigate systemic risk in the derivatives markets.

[15.07] In implementing this G20 commitment, the Committee of European Securities Regulators (CESR)17 initially did not support an actual trading obligation for derivatives, but rather supported other incentives leading to derivatives being traded on a trading venue.18 However, the European Commission (‘Commission’) decided that trading certain derivatives on a trading venue should be mandated, since this would be beneficial in terms of transparency, competition, market oversight, and price formation.19 Enhanced liquidity, greater operational efficiency, and easy access for market participants had already been identified by CESR as potential benefits.20 Market participants, on the other hand, have argued that mandatory trading could also have a negative impact, since it would lead to reduced room for innovation and reduced liquidity and would not be consistent with users’ needs.21 Furthermore, market participants have argued that mandatory trading and related central clearing also lead to constraints, including on trading hours, membership, and costs. In particular, the potential margin requirements which are associated with trading on a trading venue (due to the related central clearing) are seen as burdensome for some (non-financial) market participants.22

2. What is Exchange Trading? A. Trade Life Cycle [15.08] Every derivative transaction follows a number of phases, which in its totality is commonly referred to as the ‘trade life cycle’.23 This life cycle starts even before a specific trade is actually made by two parties, since the pre-trade phase of the life cycle consists of parties establishing a trading relationship between each other (the ‘onboarding’ or ‘documentation’ phase).24 Subsequently, the trading phase commences. Parties can start initiating individual trades by submitting orders (either on a trading venue or bilaterally). If agreement is reached on the terms of a trade, the trade is executed. The trading phase differs depending on whether the derivative is traded bilaterally or on a trading venue. A bilateral trade is initiated by a party directly contacting an executing broker (EB) or dealer to agree on the terms of the derivatives contract. However, when trading derivatives on a

trading venue, parties typically do not contact each other directly, but rather connect to a trading venue (which brings together multiple third-party buying and selling interests), either by way of direct access or via a trading member (which may be their EB), where a match is ultimately created. Finally, the trade comes into the post-trade phase, which consists of clearing (discussed in Section III, below) and settlement.25

B. Trading Models [15.09] Currently, OTC derivatives are mainly traded in so-called quotedriven markets whereby dealers quote prices at which they would be willing to trade with their counterparts. A trade takes place when a counterparty (a client or another dealer) contacts the quoting dealer and they both agree to the deal—often the quoted price is merely indicative and may be improved upon through bilateral negotiation. In contrast to OTC derivatives, most ETDs are traded in so-called order-driven markets whereby orders are submitted to a central limit order book which lists all outstanding buy and sell orders. A trade is executed if it can be matched against an existing order in the book; if not, the order will join the list and wait for a new offsetting trade to arrive.26 [15.10] Most OTC derivatives are executed over the phone using socalled voice-broking, or through proprietary online platforms. However, trading of OTC derivatives on electronic trading venues is growing and there is a variety of electronic trading models for OTC derivatives. Some are currently in use, while others are still being developed, including singledealer platforms (SDPs), multi-dealer request for quote (RFQ) model, multi-dealer limit order book model, inter-dealer limit order book model, and central limit order book (CLOB) model. We will only discuss the CLOB model and the RFQ model, mainly as these models are required to be offered in the US by SEFs and we expect Europe to follow suit.27 i. Central Limit Order Book Model [15.11] In the CLOB model, which is typically used on futures exchanges, any market participant may publish bid and offer quotes, in this way supplying liquidity to the market.28 Other

market participants can click on such a quote on their screen to accept it. If a party accepts the quote the trade is executed. Sometimes the rules of the relevant trading venue allow the dealer to have a ‘short last look’, that is, a short period of time in which it can refuse the transaction. In theory, the CLOB model allows end-users to trade independently of any dealer, but directly with other end-users, provided the instrument traded on such trading venue is sufficiently liquid. ii. Multi-Dealer Request-for-Quote Model [15.12] The multi-dealer RFQ model is a variation on the classic RFQ model used in the OTC space, in which parties request dealers to provide quotes. The difference is that under the multi-dealer RFQ model, the quotes are requested via a trading venue. If a party has sent such a request to a trading venue, the trading venue will redirect this request to a finite number of dealers, who may then provide their respective quotes.29 The party requesting the quote is then in a position to trade with the dealer offering the best price, and may do so within a limited time window. Usually, the trading venue also shows the price offered on that trading venue under its CLOB model, as a result of which the party requesting the quote can then pick the best price provided as a result of the RFQ process or the price offered on that trading venue under the CLOB model.

3. Scope A. Entities in Scope [15.13] The trading obligation will apply to financial counterparties (FCs) and non-financial counterparties (NFCs) that meet certain requirements under EMIR, both as defined in EMIR. An FC is defined as ‘an investment firm, a credit institution, an insurance undertaking, an assurance undertaking, a reinsurance undertaking, a UCITS and, where relevant, its management company, an institution for occupational retirement provision or an alternative investment fund managed by an AIFM’.30 NFCs are undertakings established in the EU which do not qualify as FCs.31 They can be categorized by those having a rolling average position in derivative contracts over thirty days exceeding the clearing threshold (NFC+s),32 and

those that do not exceed this threshold (NFC-s). In calculating its positions, an NFC should only include all of the OTC derivative contracts entered into by the NFC or by other non-financial entities within the group to which the NFC belongs, which are not objectively measurable as reducing risks directly relating to the commercial activity or treasury financing activity of the NFC or of that group.33 The trading obligation will only apply if both counterparties are subject to the trading obligation. Consequently, if only one of the counterparties is subject to the trading obligation, such party may still trade the derivative over the counter with its counterparties that are not subject to the trading obligation. i. Third-Country Entities [15.14] The scope of the trading obligation is not limited to entities incorporated within the EU, but also extends to certain third-country entities in two specific situations.34 First, where European counterparties subject to the trading obligation enter into derivatives transactions with counterparties being either a third-country financial institution or a third-country entity that would be subject to the trading obligation if it were established in the EU. Second, where thirdcountry entities that would be subject to the trading obligation if they were established in the EU enter into derivative transactions, provided that the derivatives contract has a direct, substantial, and foreseeable effect within the EU or where the imposition of the trading obligation is necessary or appropriate to prevent the evasion of any provision of MiFIR. A derivative is considered to have such an effect if it is entered into (i) by a third-country entity having a qualifying guarantee from a European FC, or (ii) between two EU branches of third-country counterparties if they would be subject to the trading obligation if they were established in the EU.35 ii. Exemptions [15.15] The intra-group transactions exemption from the clearing obligation provided for by EMIR also applies in respect of the trading obligation under MiFIR.36 The European legislator has decided that imposing a clearing obligation with respect to these transactions would be too burdensome, since these transactions could be necessary to aggregate group risk, leading to more efficient risk-procedures and, as such, reducing systemic risk.37 The same rationale could be applied with respect to the trading obligation.

[15.16] EMIR also contains a temporary exemption for OTC derivative contracts that are objectively measurable as reducing investment risks directly relating to the financial solvency of pension scheme arrangements.38 It is acknowledged that entities operating pension scheme arrangements typically minimize their allocation to cash in their portfolios in order to maximize the efficiency and return for their policy holders (i.e. future pensioners).39 This could pose problems to such entities since variation margin (see paragraph 15.39) typically must be posted in cash. To avoid negatively impacting the income of future pensioners, the clearing obligation is postponed until a suitable technical solution for the transfer of non-cash collateral as variation margin is developed by Central Counterparties (CCPs). Since ETDs will have to be cleared under MiFIR (see Section III), the same rationale applies to the trading obligation, and therefore entities operating pension scheme arrangements are also (temporarily) exempted from the trading obligation.40

B. Derivatives in Scope [15.17] Before it can be considered whether a class of derivatives should be subject to the trading obligation, it must be determined whether the class of derivatives meets the requirements which EMIR applies for its clearing obligation. The Commission must determine per class of derivatives whether it is subject to the clearing obligation, following ESMA developing a draft technical standard to such end.41 ESMA must develop such a technical standard within six months of the adoption of RTS mandating a certain class of derivatives for clearing in accordance with EMIR.42 [15.18] Under EMIR, when deciding if a class of OTC derivatives should be subject to the clearing obligation, ESMA is required to take into account: (i) the degree of standardization of the contractual terms and operational processes; (ii) the volume and liquidity; and (iii) the availability of fair, reliable, and generally accepted pricing information.43 The overarching aim of this assessment is to reduce systemic risk. If a class of derivatives meets these requirements and, consequently, becomes subject to the EMIR clearing obligation, ESMA should determine whether such class of

derivatives should also become subject to the trading obligation, having regard to: (i) whether it is traded on a trading venue (the ‘venue test’); and (ii) whether it is sufficiently liquid (the ‘liquidity test’).44 i. Venue Test [15.19] For the venue test, ESMA must assess whether a class of derivatives is admitted to trading or is trading on at least one admissible trading venue. ESMA does not have to consider whether a class of derivatives is also accepted for clearing by at least one qualified CCP, as ESMA would have already established this at the time it determined that the class of derivatives should become subject to the EMIR clearing obligation. ii. Liquidity Test [15.20] For the liquidity test, ESMA must determine whether there is sufficient third-party buying and selling interest in the class of derivatives or a relevant subset thereof for such a class of derivatives to be considered sufficiently liquid to trade only on trading venues.45 The key component in this assessment is the ‘sufficient liquidity’ of a class of derivatives. When assessing the liquidity of a class of derivatives, ESMA is required to take into account: (i) the average frequency of trades, (ii) the average size of trades, (iii) the number and type of active market participants, and (iv) the average size of spreads.46 Further criteria to be taken into account by ESMA are set out in RTS.47 When assessing whether a class of derivatives is sufficiently liquid, ESMA must also take into account the anticipated impact of a trading mandate on liquidity and on the commercial activities of end users so as to ensure that the trading obligation does not de facto lead to an unintended ban on such class of derivatives. [15.21] The liquidity test has been made to align with other legislative regimes in three respects. First, ESMA may propose that a class of derivatives is only sufficiently liquid when traded below a certain size.48 Although the principle is comparable to the large-in-scale (LIS) thresholds under the transparency regime, ESMA has explicitly indicated that these thresholds will not by definition be the same, but will be assessed by ESMA on a case-by-case basis per class of derivatives.49 [15.22] Second, it has been argued that a class of derivatives can only be considered sufficiently liquid for the purpose of the trading obligation if the market is also considered sufficiently liquid for transparency purposes.

However, ESMA has decided to maintain maximum flexibility on this point, and has indicated that it does not consider itself to be bound by such a pre-determined approach.50 [15.23] Third, for a class of derivatives to become subject to the trading obligation, ESMA will have to assess its liquidity twice, that is, when deciding on the EMIR clearing obligation and when deciding on the trading obligation. These assessments are not fully aligned. Although it would simplify the procedure to have just one assessment, ESMA believes this would be neither desirable nor feasible because of the different regulatory aims of the EMIR clearing obligation and the trading obligation.51 [15.24] The impact of the trading obligation will largely depend on how ESMA determines whether a class of derivatives is sufficiently liquid, as well as the granularity of the classes of derivatives. Where ESMA adopts a more granular assessment, it will become less likely that the obligation will have unintended negative consequences. Regulators have indicated that ESMA seems to be open to a ‘pretty granular COFIA’ approach.52 As a starting point ESMA will take the same approach as it has taken under EMIR. As such, the first distinction will be made on the basis of classes of derivatives, that is, interest rate derivatives, credit derivatives, equity derivatives, foreign exchange derivatives, and commodity derivatives.53 These classes of derivatives will be further split into sub-classes, in which ‘any other characteristic required to identify one contract in the relevant class of OTC derivatives from another’ is taken into account.54 ESMA will create these classes and sub-classes by separating key characteristics from other characteristics. Key characteristics relate to those shared on the level of a class of derivatives (e.g. product type), whereas ‘other characteristics’ are shared at the level of a sub-class of derivatives (e.g. tenor). ESMA has indicated that it may take an even more granular approach where appropriate. iii. Alternative Process [15.25] There is also an alternative process that ESMA may adopt to determine whether a class of derivatives should be subject to the trading obligation.55 Under this provision, ESMA may also identify and notify the Commission of the classes of derivatives or individual derivative contracts which should be subject to the trading

obligation, but for which no CCP is authorized under EMIR or which do not pass the venue test. ESMA could use this power where economically equivalent OTC derivative contracts are created to circumvent the trading obligation.56 ESMA may adopt this approach at its own initiative, although it would have to conduct a public consultation first. When determining if a contract should be subject to the trading obligation, ESMA must consider certain criteria.57 This could lead to the situation that ESMA must perform the venue test when determining at its own initiative whether a derivative should be subject to the trading obligation, notwithstanding that such derivative has not passed the venue test. This could not have been the intention of the European legislator, and therefore it could be argued that in such circumstances ESMA must only perform the liquidity test.

4. Miscellaneous A. Timing [15.26] Under EMIR, the clearing obligation takes effect in a phased-in manner depending on the type of counterparty. Consequently, NFC+s have a longer period to comply with the clearing obligation than FCs. This also seems to be the approach under MiFIR with respect to the trading obligation.58 From the text of MiFIR, it is not clear whether there will be a further distinction in FCs, similar to the €8 billion threshold under EMIR. [15.27] Once it is set out in RTS that a certain class of derivatives is subject to the trading obligation, ESMA must continue to assess such class of derivatives. Should there be a change in circumstances which results in a material change to the results of the venue test and the liquidity test, ESMA has the power to amend, suspend, or revoke the existing RTS.59 Before making such a decision, ESMA may consult third-country authorities. If ESMA plans to make such an amendment, suspension, or revocation, it must present draft RTS to the Commission.60 Consequently, when a material change is noted by ESMA, it will still take a considerable amount of time before the legislation is changed, and before the relevant class of derivatives is no longer subject to the trading obligation. For this reason,

market participants have argued that ESMA should also have the power to temporarily suspend the trading obligation in exceptional circumstances.61 No such power is set out in MiFIR. Therefore, if such a material change occurs, and it is not swiftly remedied by amending, suspending, or revocating the relevant RTS, this could lead to a de facto ban on trading such a derivative until the lengthy legislative procedure is concluded. [15.28] ESMA will also be required to continuously assess classes of derivatives not subject to the trading obligation. ESMA will then have to assess whether such a class of derivatives poses systemic risk and whether regulatory arbitrage opportunities arise between derivatives subject to the trading obligation and derivatives which are not subject to the trading obligation, in which case ESMA should consider bringing such a class of derivatives into the scope of the trading obligation.62

B. Register [15.29] If a class of derivatives becomes subject to the trading obligation, ESMA will include such class of derivatives in a register specifying, ‘in an exhaustive and unequivocal manner’, the derivatives that are subject to the trading obligation.63 The register will be published on ESMA’s website. In this register, ESMA will not only include the classes of derivatives that are subject to the trading obligation, but also the trading venues where they are admitted to trading or traded and the dates from which the trading obligation takes effect.

C. Package Transactions [15.30] It is unclear whether package transactions (single transactions in which multiple financial instruments are traded) fall within the scope of the trading obligation if they include a derivative subject to the trading obligation. If package transactions of this kind are within scope, it would likely lead to a de facto ban on such transactions, since package transactions are typically not liquid enough to be traded on a trading venue. In its final

report ESMA has acknowledged that it may be desirable to exempt such transactions from the trading obligation where these are used to manage risks and to improve the resilience of financial markets.64 This would be good news for market participants, since package transactions tend to lead to reduced transaction costs and lower execution risk. While ESMA has not yet taken any decision in this regard, it has indicated that it will take the treatment of package transactions into account when deciding on specific classes of derivatives.65

D. Should ETDs Fall Within the Scope? [15.31] It has been argued that the scope of the trading obligation should be extended so as to not only include OTC derivatives, but also derivatives typically traded on RMs (i.e. ETDs).66 As part of the procedure for a class of derivatives becoming subject to the trading obligation, it should be subject to the EMIR clearing obligation. Since ETDs traded on RMs do not fall within the scope of the EMIR clearing obligation, as they are not OTC derivatives, they cannot be subject to the trading obligation under the standard process. The Article 32(4) route (discussed in paragraph 15.25) offers no solution either, since one of the criteria for ESMA to subject certain derivatives to the trading obligation at its own initiative is that the class of derivatives is not yet traded on a trading venue, which will by definition not be the case. The fear is that if these ETDs are outside the scope of the trading obligation, it could result in such ETDs being traded bilaterally, with market participants moving away from trading venues. This would obviously be contrary to the objective of the trading obligation of increasing transparency. ESMA has acknowledged this issue, but believes that the risk of ETDs moving off-venue is limited, due to the new transparency obligations67 and the qualitative thresholds applying under the new SI regime.68

III. The Clearing Obligation

1. The General Obligation [15.32] The second obligation that Title V MiFIR introduces is the clearing obligation of ETDs. Under Article 29(1) operators of RMs must ensure that all transactions in derivatives that are concluded on their RM are cleared by a CCP. [15.33] Before EMIR came into operative effect, there was no general obligation to clear, and once EMIR came into effect this obligation only extended to certain OTC derivatives.69 Under MiFIR, the scope of the CCP clearing obligation has been extended to ETDs.70 Recital 37 of MiFIR provides the following rationale for extending the CCP clearing obligation to ETDs: Regulation (EU) No 648/2012 lays down the criteria according to which classes of OTC derivatives should be subject to the clearing obligation. It prevents competitive distortions by requiring non-discriminatory access to CCPs offering clearing of OTC derivatives to trading venues and non-discriminatory access to the trade feeds of trading venues to CCPs offering clearing of OTC derivatives. As OTC derivatives are defined as derivative contracts whose execution does not take place on a regulated market, there is a need to introduce similar requirements for regulated markets under this Regulation. Derivatives traded on regulated markets should also be centrally cleared.

[15.34] It is unclear to us why the conclusion in the last sentence follows from the desire to extend non-discriminatory access rules71 to ETDs, especially since the latter rules apply to financial instruments (excluding OTC derivatives) rather than merely ETDs, and financial instruments are not brought within the ambit of the central clearing obligation. In our view, the more apparent rationale for applying a CCP clearing obligation to ETDs would be to ensure that OTC derivatives being driven to RMs would not escape central clearing (as the inherent risks sought to be mitigated under EMIR would still be present).

2. What is Central Clearing?

[15.35] As set out in paragraph 15.08 the trading phase of the life cycle of a derivatives transaction is followed by the first post-trading service, being the clearing service. Clearing is the process that occurs in between execution and settlement. The time lag can be far greater for derivatives than for securities (up to a few decades, whereas securities are typically settled within three days). During this time lag trades need to be processed, managed, monitored and ultimately prepared for settlement.72 These activities can be referred to as basic clearing services which are essential to the life cycle of a trade. However, since the clearing obligation refers to clearing by a CCP (also called ‘central clearing’), it is CCP services that are mandated by MiFIR, not basic clearing services. [15.36] Neither MiFIR nor MiFID II contains a definition of ‘clearing’. However, the term is defined in EMIR as ‘the process of establishing positions, including the calculation of net obligations, and ensuring that financial instruments, cash, or both, are available to secure the exposures arising from those positions’.73 This definition can be explained by considering the following main features of clearing, as set out below.

A. Novation [15.37] Central clearing typically involves trades between market participants being replaced with corresponding trades in such a way that a CCP is interposed between the relevant market participants through novation, thereby becoming the ultimate buyer to every sell trade and the ultimate seller to every buy trade (see Figure 15.1). Interposing the CCP between the market participants in this way is designed to minimize the adverse, potentially systemic, effects of a default (typically on insolvency) by a market participant. This is because the CCP has to perform its obligations to the ultimate counterparty on one end, whether or not the ultimate counterparty on the other end is in default. In this sense the CCP acts as a buffer between the original counterparties, thereby managing the counterparty credit risk inherent in open derivatives positions between execution and settlement.

Figure 15.1: A CCP interposed between the relevant market participants through novation.

B. Multilateral Netting [15.38] Generally, many counterparties will have offsetting positions at the end of the day. A beneficial consequence of replacing many counterparties with one CCP is that any offsetting positions can be netted off against each other. As a result of this ‘multilateral netting’ a CCP reduces its exposure against its counterparties. Compared to a non-centrally cleared situation the multiplicity of bilateral exposures between the original counterparties, and thus the gross counterparty credit risk in the market, is dramatically reduced.74 Figure 15.2 illustrates this.

Figure 15.2: Exposure of a non-centrally cleared situation versus ‘multilateral netting’.

C. A CCP’s Protective Measures [15.39] To mitigate the risks the CCP takes on by interposing itself between counterparties, the CCP has protective measures in place, typically in the form of financial safeguards, including by way of membership

criteria and margining. In particular, the CCP only accepts counterparties that satisfy strict membership criteria.75 This serves to reduce the likelihood of such a ‘clearing member’s’ default. If such a default were to occur, the CCP’s margining measures, which are provided either in the form of cash or securities, are intended to ensure that all obligations related to the open positions of the defaulting clearing member can be liquidated shortly after the default occurs. There are two types of margins that are implemented by the CCP; ‘initial margin’ (IM) and ‘variation margin’ (VM). The initial margin is set to cover the potential costs of replacing an open position (forward looking), and the variation margin is set to cover fluctuations in the value of an open position (backward looking). There are additional protective measures in what is called the ‘default waterfall’, but these measures are less relevant for this chapter and will therefore not be discussed.

3. Scope [15.40] As set out in paragraph 15.33 above, the MiFIR clearing obligation extends to ETDs. Contrary to the central clearing obligation under EMIR, the MiFIR clearing obligation applies to the operator of the RM, rather than to the counterparties to the derivatives contract. This approach makes sense as it will be easier for regulators to monitor compliance, and also because counterparties to ETDs do not know each other’s identity due to the anonymity that an RM provides. However, similar to the EMIR clearing obligation, the burden in terms of costs and margining is ultimately borne by the counterparties.

4. Miscellaneous A. Access to Central Clearing [15.41] Central clearing can be ‘direct’ in the sense that the parties to the trade are members of the CCP through which the trade is cleared (see Figure 15.3, top). It can also be ‘indirect’ in the sense that a counterparty to the trade is not a clearing member (CM), but employs a clearing member to

clear the trade (see Figure 15.3, second from top). In that case such counterparty is the client of a clearing member which then acts as its clearing broker (CB).76 This means that not only the CCP, but also the CB is interposed between the original counterparties. In the European principalmodel the client faces its CB as principal and the CB in turn faces the CCP as principal.77 This in effect creates identical back-to-back trades, but, for the most part, without ever establishing any principal or other relationship between the CCP and the client.

Figure 15.3: ‘Direct’ and ‘indirect’ central clearing.

[15.42] The term ‘indirect clearing’ is typically used to refer to a subsequent link in the clearing chain, consisting of a position between the client of a CB (the ‘direct client’) and the client of that client (the ‘indirect client’). In respect of ETDs this may be done for a variety of reasons. For the purposes of this chapter, we will only consider a type of indirect clearing commonly referred to as ‘client of client clearing’.78 Under this form of indirect clearing, a client’s EB which is not a clearing member of the relevant CCP may offer clearing services to its client as part of its business. The EB then acts as a clearing intermediary, and it is unlikely that the CB will know who the indirect client is (see Figure 15.3, third from top; and Step 4 in paragraph 15.53).79 This form of indirect clearing operates in as similar manner for OTC derivatives, albeit that the EB would be the original counterparty to the trade rather than the party providing clearing services and thus would end up on the other side of the CCP following the

novation. The role of the direct client would be performed by a direct client of a clearing member (see Figure 15.3, bottom).

B. Position of the CB [15.43] It is relevant to note that in the European principal-model it is the CB rather than the client that has a legal relationship with the CCP. Therefore, under the client’s cleared positions, there is only a counterparty risk of the CCP as against the CB. That is in accordance with the CCP’s need to limit its risks, as the CB is deemed to be more solvent than the client (whom is unknown to the CCP). The principal-model also has unintended consequences, because it does not correspond with the intended allocation of the counterparty risk between the client and the CB. For instance, there is counterparty risk of the client against the CB as a result of the legal relationship between them (i.e. the CB is the legal counterparty of the client). This is not intended by the client, because the client wishes to replace the original counterparty risk with counterparty risk as against the CCP (due to the CCP’s function to effectively guarantee the settlement of positions). Therefore, the client will want to eliminate the counterparty risk against the CB, which it may do by, for example, ensuring that its collateral does not become part of the bankruptcy estate of the insolvent CB.

C. Segregation at (Direct) Client Clearing i. Assets and Positions Protection [15.44] When a client posts IM to its CB, a counterparty risk arises between the client and the CB in respect of such IM. This follows from the fact that the title to the collateral is transferred from the client to the CB. Consequently, if the CB becomes insolvent, the client only has a contractual, unsecured claim for the return of the collateral value. As set out in the previous paragraph, this is not intended, so the client will want to eliminate this counterparty risk. [15.45] EMIR supports this position by imposing the use of certain account structures on CCPs and CBs.80 A CCP must offer two types of

account structure to its clearing members that offer client clearing services; one type of account in which the clearing member’s proprietary positions and related collateral are registered (the ‘house account’), and one type of account for the positions entered into by the CB for its clients and the collateral related to those positions (the ‘client account’). Since the client account is segregated from the CB’s house account at the CCP level, the client is protected from the insolvency of its CB, because other creditors cannot assert claims on the collateral posted to its client account.81 The client accounts can in turn be further distinguished into ‘omnibus segregated accounts’ (OSAs) and ‘individually segregated accounts’ (ISAs). An OSA is used for several clients (and therefore creates fellow customer risk (or risk of loss mutualization) amongst the OSA account holders) and an ISA is used just for one client (see Figure 15.4).82 An ISA could be more expensive to hold than an OSA due to the specific management it requires.

Figure 15.4: Use of portfolio compression to achieve multilateral netting. Figures 15.4, Figure 15.5, Figure 15.6, Figure 15.7

ii. Netting [15.46] There are many mutual collateral obligations between a CCP and its clearing members. However, the multilateral netting by a CCP provides for netting of these obligations into a net collateral

obligation.83 In client clearing this is also the case, although under EMIR the collateral posted to client accounts may not be netted against the collateral obligations for the CB’s own account.84 The degree to which netting is permitted depends on the type of client account chosen. In the case of an ISA, the collateral cannot be netted against collateral obligations other than those of the relevant client. With an OSA the CB is allowed to net the collateral against collateral obligations of other clients for whom the CB uses that OSA. Whether netting actually takes place in an OSA depends on the type of OSA. No netting will take place in the case of a ‘gross OSA’, but it will in a ‘net OSA’. The net collateral obligations of the CB towards the CCP will be less with a net OSA than with a gross OSA or an ISA. As a result the collateral obligations of the client to its CB could also be less and thus financially beneficial, providing a second motivation for clients to choose to accept the risks of a net OSA.

D. Default management; Post-Default Porting, and Leapfrog Payment [15.47] The whole purpose of segregation is to allow for porting and the leapfrog payment.85 In simple terms, porting is the transfer of a client’s positions (and associated collateral) registered in a client account at CCP level to another client account at CCP level of another CB (in case the original CB defaults (i.e. becomes insolvent)). This typically needs to occur within a short time frame.86 Whereas an OSA offers economic benefits as a result of netting, the ISA offers the benefit of being more eligible for porting.87 This is because OSAs have the disadvantage of a multitude of clients sharing them, making it difficult to port to another CB as it is unlikely that all these clients have the same back-up CB. If porting does not occur, for example because the client has not appointed a back-up CB, the CCP will close out the positions, recover any expenses (including hedging costs), and return any balance owed to the client under the closed-out positions directly to the client, thereby circumventing the bankruptcy estate of the insolvent CB.88 This is commonly referred to as the ‘leapfrog payment’.

[15.48] In conclusion, each type of account has its benefits and drawbacks. An ISA is more eligible for porting, but could be more expensive and lacks the netting benefit that an OSA offers; and an OSA creates fellow customer risk (or risk of loss mutualization) amongst the OSA account holders.

IV. The Straight-Through-Processing Obligations 1. The General Obligation [15.49] The third set of obligations that Title V MiFIR introduces is in relation to the straight-through-processing (STP) obligations. Under Article 29(2), CCPs, trading venues, and investment firms which act as CBs89 must have in place effective systems, procedures, and arrangements in relation to cleared derivatives to ensure that transactions in cleared derivatives are submitted and accepted for clearing as quickly as technologically practicable using automated systems. [15.50] The recitals of MiFIR are silent on the rationale for introducing this obligation. It is likely that the inclusion of this obligation relates to the fact that the requirement for central clearing is becoming increasingly mandated. As we saw in EMIR, the flipside of the clearing obligation was the introduction of regulation and supervision of the CCPs into which OTC derivatives were funnelled. The European legislator seems to pick up this quest for safety of the post-trading process where it left it at the time of finalizing EMIR, since it is now also able to include centrally cleared ETDs, noting: In order to manage operational and other risks, when transactions in cleared derivatives are submitted and accepted for clearing and to provide certainty to counterparties as soon as possible, it is important to determine whether a cleared derivative transaction will be accepted for clearing by a CCP at an early stage, and to the extent possible before the transaction is entered into, as well as the consequences of a CCP not accepting for clearing the derivative transaction submitted.90

[15.51] From this we conclude that the safety of central clearing is no longer sought only in CCP regulation as provided by EMIR, but now also in reducing the operational risk in the post-trading process in MiFIR.

2. What is STP? [15.52] There is no general definition of STP, nor is it defined in MiFIR, but the essence of it is that the execution and the clearing of a trade should be practically simultaneous processes, because the quicker a trade reaches the CCP from the time it was executed, and the fewer people involved in the process, the lower the risk of errors.91 This ideally requires a seamless electronic transmission of information and data, utilizing standardized technologies and infrastructure, and thus a full automation of the trading process.92 By eliminating trade failures, the efficiency of clearing and thus operational efficiency is increased.

A. ETDs [15.53] The existing clearing process for ETDs in Europe can be summarized as follows (as illustrated in Figure 15.5).93 • Step 1: An order is submitted to an EB by its client. • Step 2: The EB submits the order to the RM. All orders submitted to the RM are subject to validation. To prevent orders from being entered in error, both EBs and RMs implement ‘obvious error/fat finger’ and price level checks. Similarly, for risk mitigation purposes, many RMs have pre-trade risk limits (relating to order/position size) in place between CBs and their direct clients and EBs, including ‘kill’ buttons, which can be used to suspend the trading activity of a client or EB.94 • Step 3: If the order matches at the RM with an opposing order submitted by another EB for its client, the resultant trade is automatically submitted to a CCP immediately upon such execution. This means that no unmatched trades can arise.

• Step 4: Upon trade execution, a chain of linked back-to-back principalto-principal contracts are immediately created between (i) the CCP and the CB (of the EB, if the EB is a not a clearing member of the relevant CCP), (ii) the CB and the EB (if the EB is a not a clearing member of the relevant CCP), and (iii) the EB and the client.95 This means that there is immediate certainty of clearing.96 Where the client has not appointed its own CB, but its EB offers clearing services, its trade will remain cleared on this basis of indirect clearing (discussed in paragraph 15.42). Otherwise the following steps would follow. • Step 5a: Where a client and its EB have the same CB, the CB’s appropriate trade source system (‘middleware’) will assign the trade to the correct position keeping account at the CCP and will update the client’s account in the CB’s back office. This means that the contractual arrangements between the CB and the EB, as well as between the EB and the client, are terminated and replaced by a contract between the CB and the client. • Step 5b: Where the client and its EB do not have the same CB (or where the EB is a clearing member of the relevant CCP, but is not the CB of the client), the trade must be given up to another CB (where the client’s account is held), that is, entering the necessary allocation instructions into the clearing system.97 This is done by the CB’s middleware giving up the trade to the client’s CB. When the client’s CB’s middleware claims the trade and assigns it to the correct position keeping account at the CCP and updates the client’s account in the client’s CB’s back office, the contractual arrangements between the CCP and the EB’s CB, between the EB and its CB, and also between the EB and its client, are terminated and replaced with contracts between the CCP and the client’s CB as well as between the client and its CB.

Figure 15.5: Segregation at (direct) client clearing.

B. OTC Derivatives [15.54] The central clearing of OTC derivatives by clients through the intermediation of CBs has only been developed more recently and therefore is less mature than ETD clearing. The existing clearing process for OTC derivatives in Europe can be summarized as follows (and is illustrated in Figure 15.6), whereby we set out a scenario in which a trade is executed between a self-clearing EB and a client that clears through a CB.98 • Step 1: The client and its EB bilaterally (i.e. not on an RM) execute an OTC derivative transaction, which they intend to clear. In contrast to ETDs, for OTC derivatives the EB acts as a principal to the contract rather than as an agent. • Step 2: The client and/or its EB submit the details of that OTC derivative transaction to the agreed CCP (together with details regarding the CB of the client) via middleware. This marks the first step in the clearing process, where the consistency of the buyer’s and seller’s terms of trade are assessed in order to prevent any unintentional errors, that is, the trade confirmation. The trade details need to match in the

•

• •

•

•

middleware before they are sent on to the CCP. Any mismatched trades will remain in the middleware until the mismatch is resolved or the trade is terminated. Step 3: On receipt of the message from the relevant middleware, the CCP performs (i) a counterparty risk check to validate whether the relevant CBs are within their established risk limits, and (ii) product validation to ensure that the relevant transaction meets the relevant product eligibility criteria. Step 4: The CCP sends a request to clear the OTC derivative transaction to the CB via the request/consent flow system. Step 5: The CB will check that the trade falls within the client’s credit limits and will accept or reject the OTC derivative transaction for clearing on that basis (although other grounds may be put forward by the CB for rejection) and will then confirm acceptance with the CCP. Step 6a: Once the CCP has received an acceptance from the CB, and assuming its own checks on the EB and the CB in Step 3 were positive, the CCP will accept the trade for clearing. Subject to satisfaction of the requirements in the relevant CCP’s rulebook, upon acceptance the CCP will register two contracts: one between the CB and the CCP on behalf of the client, the other between the CCP and the EB. Under the clearing agreement between the CB and the client, a back-to-back transaction is deemed to come into existence between the CB and the client. Step 6b: The CCP will send a confirmation back to the middleware or the client that the trade has been cleared.

Figure 15.6: ETDs.

i. Contractual Time Frames [15.55] Under the existing process, there are no regulatory timing requirements with respect to OTC derivative clearing. Therefore, in theory (subject to the timing requirements set out in the relevant CCP’s rulebook, for example some CCPs stipulate that trades can only be submitted for clearing up to close of business on the day after execution (‘t+1’)) a trade could remain uncleared indefinitely if, for instance, the middleware trade submissions do not match or the trade is not accepted or is rejected by the CB. However, whilst there are no regulatory timing requirements, best practice guidelines indicate a time period of four hours from the point of execution to clearing. In addition, counterparties may agree to impose contractual time frames. The market standard execution agreement99 mandates the following time frames: 1. 150 minutes from the execution of the transaction for the client and the EB to complete Step 2 (the submission cut-off);

2. 90 minutes after the submission of the transaction to the CB for rejection/acceptance of the transaction (Step 5) with an absolute longstop date of 10:30 a.m. on the business day immediately following the day on which the transaction was executed (the clearing member cut-off);100 and 3. The CCP must reject/accept the transaction (Step 6a) before 6 p.m. if executed prior to 4 p.m. (assuming 7 p.m. as the latest time at which a derivative transaction may be submitted to a CCP and accepted for clearing on the same day), with an absolute longstop date of 12:00 p.m. on the business day immediately following the day on which the transaction was executed (the CCP cut-off). [15.56] We understand that, in a variation to the model described above, some CBs communicate their clients’ limits to the CCP and allow the CCP to perform the relevant client credit limit check on their behalf.

3. Scope [15.57] The STP obligations apply to ‘cleared derivatives’, which are defined as (i) all derivatives which are to be cleared pursuant to the clearing obligation set out either in EMIR or in MiFIR, and (ii) all derivatives which are otherwise agreed by the relevant parties to be cleared. Accordingly, the obligation applies to all derivatives submitted for clearing.101 A specific regime applies to derivatives traded on a trading venue (i.e. an RM, an MTF, or an OTF). Note that these are not simply ETDs, because derivatives traded on an MTF or an OTF qualify as OTC derivatives rather than as ETDs under MiFIR. In Section IV.4.A we discuss the rules applicable to these derivatives. A different regime applies to derivatives not entered into on a trading venue (i.e. an RM, an MTF, or an OTF) but on a bilateral basis. Note that these are not simply OTC derivatives, because derivatives traded on an MTF or an OTF also qualify as OTC derivatives, but fall within the aforementioned regime. We discuss the rules applicable to these derivatives in Section IV.4.B.

4. The Obligations in More Detail A. Cleared Derivatives Concluded on a Trading Venue i. Predetermined Information [15.58] To ensure that trades can be submitted and accepted for clearing as quickly as technologically practicable, the information (and format thereof) needed by a trading venue and a CCP from counterparties to perform their tasks must be predetermined and clearly set out in their respective rules.102 ii. Pre-Trade Checks [15.59] It is only after clearing has been completed that a counterparty faces the CCP rather than its original counterparty. Therefore, obtaining certainty on clearing at an early stage (and preferably even before trade conclusion) is generally considered a key element in reducing credit risk. The trading venue must give CBs the ability to provide their limits set for their clients.103 This allows the trading venue to check the orders placed against these limits, which is mandatory so as to limit the situations in which a concluded trade may not be accepted by the CB or the CCP. The pre-check of trades entered into electronically must be performed within sixty seconds of the receipt of the order by the trading venue. In relation to trades which are not entered into electronically, the check must be performed within ten minutes of receipt of the order.104 Subsequently, where the order is not within the limit, the trading venue must inform the client and the CB on a real-time basis for orders to be concluded electronically, and within five minutes following the pre-check for nonelectronic orders.105 The timing difference follows from the fact that when a trade is entered into electronically on a trading venue, its processing can be automated at a higher level and therefore its processing time can be much shorter than when it is not. iii. Time Frame for Submission by Trading Venue to CCP [15.60] The trading venue must submit trade information to the CCP (i) within ten seconds of conclusion when it is concluded electronically, and (ii) within ten minutes of conclusion when it is concluded non-electronically.106 iv. Timeframe for CCP Acceptance [15.61] The CCP must (i) accept or not accept a trade submitted for clearing within ten seconds of receiving the

trade information from the trading venue, and (ii) inform the CB and the trading venue of a non-acceptance on a real-time basis.107 The reason why the time granted to a CCP to decide on acceptance is the same for both electronically traded and non-electronically traded cleared derivatives is that the information is sent to the CCP in a pre-agreed electronic format in both cases. v. Non-Acceptance [15.62] Both the CB and the trading venue must inform the client of a non-acceptance as soon as possible after the CCP has informed them thereof.108 vi. Exemption [15.63] Where (i) the trading venue’s rules require its trading members to have a CB; (ii) the CCP’s rules ensure that a trade that is entered into is automatically and immediately cleared; and (iii) the trading venue’s rules provide that the trading member or its client becomes counterparty to the trade after it has been cleared, this is considered sufficient to identify whether the trade can be cleared by a CCP prior to conclusion.109 Therefore, the STP rules discussed above are not applicable in these automated clearing circumstances.

B. Cleared Derivatives Concluded on a Bilateral Basis i. Time Frame for Submission by Counterparties to CCP [15.64] The processing of trades entered into on a bilateral basis is usually less automated than the processing of trades entered into on a trading venue. Therefore, the time granted to counterparties to submit a trade entered into on a bilateral basis to a CCP is longer than the time granted for a trade concluded on a trading venue. To ascertain the starting time, the CB must obtain evidence from its client of the time of conclusion. Interestingly, it is the CB that must ensure that the trade information is submitted to the CCP by the counterparties within thirty minutes of conclusion.110 We would expect a CB only to be responsible for the information to be sent by its own client, rather than by both counterparties. [15.65] Considering the currently prevailing contractually permitted time (discussed in paragraph 15.54 for Step 2), this would mean a reduction of

120 minutes. ii. Time Frame for CB Acceptance [15.66] In order to manage the credit risks related to cleared trades, a CCP must allow a CB to review the trade details of its client and to decide whether to accept it. As the process between a CCP and a CB is usually automated, this process should require limited time. The CCP must provide the trade information to the CB within sixty seconds of the receipt of this information. The CB must perform its review and accept or not accept the trade within sixty seconds of receipt of the information from the CCP.111 [15.67] Considering the currently prevailing contractually permitted time (discussed in paragraph 15.54 for Step 4), this would mean a reduction of eighty-nine minutes. iii. Time Frame for CCP Acceptance [15.68] Ensuring that trades are submitted to clearing as quickly as technologically practicable does not therefore imply that all such trades will be accepted for clearing in all circumstances. Where such trades are not accepted for clearing, counterparties should have clarity on the treatment of those trades in order to hedge their risk.112 Once the CB has informed the CCP of its acceptance or non-acceptance, the CCP must (i) accept or not accept the trade within ten seconds of the receipt of the CB’s acceptance or non-acceptance,113 and (ii) in the case of a non-acceptance, inform the CB on a real-time basis. The CB must then inform its client as soon as it has been informed by the CCP.114 We would have expected a similar notification obligation on the CB in the case of its non-acceptance (discussed above). [15.69] Considering the currently prevailing contractually permitted time (discussed in paragraph 15.54 for Step 6a), this would mean a reduction of a few hours. iv. Exemption [15.70] Where the CCP’s rules (i) ensure the setting and maintenance of limits on a regular basis by a CB for its clients,115 and (ii) provide that a trade that is within such limits is cleared automatically by the CCP within sixty seconds of receiving the information from the counterparties, the STP rules discussed above on CB acceptance and CCP

acceptance (excluding those applicable in the case of non-acceptance) are not applicable in these automated clearing circumstances.116

C. Trades not Accepted by the CCP i. Cleared Derivatives Concluded on a Trading Venue Electronically [15.71] As the processing of a cleared derivative concluded electronically on a trading venue and submitted for clearing to a CCP requires limited time, the time for the market to move, and consequently for the value and the risk of the cleared derivative to change in between the order and the non-acceptance, is also very limited. Since the damage potentially suffered by counterparties whose trades are not accepted for clearing by the CCP is negligible, and in order to provide certainty to counterparties, cleared derivatives concluded electronically on a trading venue and not accepted for clearing by a CCP must be voided by the trading venue.117 ii. Cleared Derivatives Concluded on a Trading Venue NonElectronically or on a Bilateral Basis [15.72] As the processing of cleared derivatives other than those concluded electronically on a trading venue or on a bilateral basis usually takes longer, this period of time may be sufficiently long for the market to have moved, and consequently for the value and the risk of the trade to have changed significantly. Therefore, voiding the trade might not be the appropriate treatment for all nonacceptances. To provide certainty on the treatment of cleared trades other than cleared trades concluded electronically on a trading venue and not accepted by a CCP for clearing, the rules of the trading venue, and the contractual arrangements between the counterparties where appropriate, should clarify in advance how these trades are to be treated.118 iii. Technical or Clerical Problem [15.73] When a cleared trade is not accepted for clearing for technical or clerical reasons arising from the transmission of inaccurate or incomplete information (i.e. other than creditrisk-related reasons), the counterparties may still want to clear that trade. Parties can then decide to resubmit such a trade. Where counterparties agree to this, a second submission in the form of a new trade with the same economic terms can be made. This is considered to still amount to the

proper management of operational or other risks, as long as it is within one hour of the previous submission, and providing it allows the investigation and resolution of the non-credit reasons of why the trade was not accepted.119

5. Miscellaneous A. Frontloading [15.74] There are various reasons for OTC derivatives that have been entered into previously being submitted for clearing at a CCP (also known as ‘frontloading’ or ‘backloading’) at a later point in time. One is that the EMIR clearing obligation requires parties subject to it to have their bilateral OTC derivatives entered into during the so-called frontloading period cleared at the time the clearing obligation comes into effect. On a strict reading of the obligation of a CB to ensure that trade information is sent by the counterparties to the CCP within thirty minutes of the conclusion of the trade, frontloading would by definition lead to a breach of this obligation, since the relevant derivatives were concluded some time ago.

B. Consistency between the EU and the US [15.75] There is a lack of consistency between the EU and the US in relation to the time to accept or not accept a trade after conclusion for derivative on a trading venue and bilateral derivatives. Under the US rules, set by the CFTC on STP, the CCP has to accept or not accept a trade within ten seconds of such trade being concluded on an SEF (similar to a trading venue). Any other trade concluded bilaterally must be accepted or not accepted within sixty seconds. Under MiFIR, the CCP is required to accept or not accept the trade within ten seconds without consideration of whether the trade was concluded on a trading venue or bilaterally.120

V. The Obligations in Respect of Indirect Clearing

1. The General Obligation [15.76] The fourth set of obligations that Title V MiFIR introduces follows from the requirements regarding indirect clearing arrangements121 for ETDs. Under Article 30, indirect clearing arrangements with regard to ETDs are permissible provided that those arrangements do not increase counterparty risk and that they ensure that the assets and positions of the counterparty benefit from protection with equivalent effect to that referred to in Articles 39 and 48 EMIR (discussed in Sections III.4.C and D above).122 For the avoidance of doubt, there is no obligation to perform indirect clearing, but rather there are obligations that apply when performing indirect clearing. [15.77] The regulation of indirect clearing of OTC derivatives is already provided under EMIR. The effect of MiFIR is to extend the scope of the indirect clearing regime to ETDs. The rationale is identical to that of EMIR, being: An indirect clearing arrangement should not expose a CCP, clearing member, client or indirect client to additional counterparty risk and the assets and positions of the indirect client should benefit from an appropriate level of protection. It is therefore essential that any type of indirect clearing arrangements comply with minimum conditions for ensuring their safety. To that end, the parties involved in indirect clearing arrangements should be subject to specific obligations and indirect clearing arrangements should only be permissible provided that they meet the conditions defined in this Regulation. Such arrangements extend beyond the contractual relationship between indirect clients and the client of a clearing member that provides indirect clearing services.123

[15.78] An indirect clearing regime aiming to protect indirect clients is necessary because, although EMIR imposes segregation, porting, and default management rules on CCPs (including those clearing ETDs) and CBs, these only extend to the protection of the (direct) clients of CBs and therefore do not suffice in the case of indirect clients. [15.79] Incidentally, under EMIR, indirect clearing will be the only way for certain parties to comply with the EMIR clearing obligation, although to date there has been no offering of indirect clearing for OTC derivatives.124 However, as set out in paragraphs 15.42 and 15.53, in the ETD space, ETDs

are already cleared on the basis of indirect clearing where clients do not have a CB. Therefore the existing ETD arrangements will need to take account of the new rules applicable to indirect clearing under the Draft MiFIR ICA RTS,125 which are discussed in this section.

2. What is Indirect Clearing? [15.80] As set out in Section III.4.A, indirect clearing is essentially client clearing whereby the relevant client is not a (direct) client of a clearing member, but rather a client of a (direct or other indirect) client of a clearing member and thus a clearing member’s ‘indirect client’. The term ‘indirect client’ is defined as ‘the client of a client of a clearing member’.126 Accordingly, the terms ‘indirect clearing arrangement’ and ‘indirect clearing service arrangement’ are defined as ‘the set of contractual relationships between the central counterparty (CCP), the clearing member, the client of a clearing member and indirect client that allows the client of a clearing member to provide clearing services to an indirect client’. [15.81] The chain from the CCP up to the ultimate client (the ‘end indirect client’) may be longer and contain more than two intermediate entities (i.e. CB and (direct) client), though it may be limited by the Draft MiFIR ICA RTS (as further discussed in Section V.4.E).

3. Scope [15.82] The obligations in respect of indirect clearing apply to all actors in the chain other than the indirect client or, where the chain is longer (as further discussed in Section V.4.E), the end indirect client. The intermediate entity facing the end indirect client is subject to the obligations of a client as if it were a client.127 Furthermore, only the indirect clearing of ETDs is within scope, since the indirect clearing of OTC derivatives is already covered under EMIR.128

4. The Obligations in More Detail A. Structure of the Indirect Clearing Arrangements [15.83] EMIR requires CCPs to qualify as systems within the meaning of the Settlement Finality Directive (SFD)129 so that their clearing members, assuming that they are credit institutions or investment firms, qualify as participants within the meaning of the SFD.130 This results in the applicability of certain protections under the SFD for (direct) clients.131 According to Recital 3 Draft MiFIR ICA RTS, in order to ensure an equivalent level of protection to indirect clients as is granted to (direct) clients under EMIR, it is also necessary to ensure that the (direct) clients providing indirect clearing services are credit institutions or investment firms.132

B. Segregation [15.84] Because an indirect client can be confronted with the insolvency of both a CB and a client, both types of insolvency events must be addressed by the appropriate protection mechanisms. To that end the Draft MiFIR ICA RTS provide for segregation obligations in respect of CCPs, CBs, and (direct) clients. [15.85] At least the following two types of indirect client accounts must be opened and maintained by the CB and offered by the (direct) client to the indirect client:133 (1) an omnibus account with the assets and positions of the client held for the accounts of its indirect clients (essentially an omnibus indirect client account, ‘OICA’); and (2) an omnibus account with the assets and positions of the client held for the accounts of its indirect clients, where the CB must ensure that the positions of an indirect client do not offset the positions of another indirect client, and that the assets held for the account of an indirect client cannot be used to cover the positions of another indirect client

(essentially a gross omnibus indirect client account (GOICA), with the basic OICA therefore amounting to a net OICA). [15.86] Both the OICA and GOICA must correspond with an ISA held by the CB (at the level of the CCP) for the account of the client rather than an OSA of the client.134 Where the assets and positions of several indirect clients are managed by a CCP in a single ISA that corresponds with a GOICA, the CCP must keep separate records of the positions of each indirect client, calculate the margins in respect of each indirect client separately, and collect the sum on a gross basis.135 [15.87] It is possible for the CB to offer an ISA for indirect clients (essentially an individually segregated indirect client account, ‘ISICA’). However, this requires the CB to maintain separate accounts for each indirect client, which results in a greater operational burden to manage all of the separate accounts than if the indirect clients were managed in a single account.136 Requiring an ISICA in the case of indirect clearing services would multiply the number of necessary accounts and related operational processes, raising the complexity, cost, and operational risk of indirect clearing arrangements. However, a GOICA reduces the number of accounts to be opened and maintained and therefore the related number of operational steps to assess margin and settle collateral in these accounts. A GOICA reduces the cost and complexity compared to an ISICA, while still allowing the collateral and the positions of different indirect clients to be distinguished, thereby ensuring an equivalent level of protection to that of an ISICA. Although an ISICA could still be offered, the requirements applicable to the use of the GOICA would, at a minimum, need to be met.137 [15.88] Where the indirect client does not instruct the (direct) client of its choice of OICA, GOICA, or ISICA within a reasonable period of time, the client may provide indirect clearing services to this indirect client using an OICA or a GOICA.138

C. Default Management; Post-Default Porting and Leapfrog Payment [15.89] The CB must establish robust procedures to manage the default of a client that provides indirect clearing services.139 In respect of an OICA this means that the CB must:140 (1) ensure that the default management procedures allow the prompt liquidation of the assets and positions of indirect clients following the default of a client, which must include liquidating the positions of the indirect clients at the level of the CCP. The CB must include in the procedures the details of the communication from the CB to the indirect clients regarding the default of the client and the period of time during which the portfolios of the indirect clients will be liquidated; and (2) after the completion of the default management process for the default of a client, readily return to the client for the account of the indirect clients any balance owed from the liquidation of the assets and positions of the indirect clients by the CB (i.e. a leapfrog payment). [15.90] In respect of a GOICA this means that the CB must:141 (1) include in the default management procedures (i) the steps to conduct a transfer of the assets and positions (i.e. porting) held by the defaulting client for the account of its indirect clients to another client or CB; and (ii) the steps required to initiate the payment of the proceeds from a liquidation of the assets and positions of indirect clients to each of these indirect clients; (2) as a minimum, contractually commit itself to trigger the procedures for porting to another client or CB designated142 by all143 the indirect clients whose assets and positions are being ported, on the relevant indirect clients’ request and without the consent of the defaulting client; (3) ensure that the procedures allow the prompt liquidation of the assets and positions of indirect clients following the default of the client, which must include liquidating the positions of the indirect clients at

the level of the CCP, if porting has not taken place for any reason within a predefined transfer period specified in the indirect clearing arrangements. The CB must include in the procedures the details of the communication from the CB to the indirect clients regarding the default of the client and the period of time during which the relevant indirect client portfolios will be liquidated; (4) following the liquidation of the assets and positions of the indirect clients, at least, contractually commit itself to trigger the procedures for the payment of the liquidation proceeds to each of the indirect clients; and (5) after the completion of the default management process for the default of a client, and when the CB has not been able to identify the indirect clients or to complete the payment of the liquidation proceeds to each of the indirect clients, readily return to the client for the account of the indirect clients any balance owed from the liquidation of the assets and positions of the indirect clients by the CB (i.e. a leapfrog payment). [15.91] According to ESMA, the porting and leapfrog payment obligations are obligations of means rather than of results, since reference is made to procedures and their initiation rather than to actual completion.144

D. Information Flow [15.92] Considering the number of intermediate entities that may be involved in the context of indirect clearing, it is important that all links in the chain have the necessary information to satisfy their respective obligations. For example, in respect of a GOICA, the client must ensure that the CB has all the necessary information to identify the positions and the collateral value held for the account of each indirect client in the account on a daily basis, and the CB must pass this information on to the CCP. In turn, the CCP must use this information for calculating the margin requirements separately for each indirect client.145 [15.93] CBs should use information provided by clients for riskmanagement and margining purposes only and should prevent the misuse of

commercially sensitive information, through, for example, the use of effective barriers between different divisions of a financial institution to avoid conflicts of interest (i.e. Chinese walls).146

E. Extended Chains [15.94] As alluded to in paragraph 15.82, the chain between the CCP and the ultimate client (the end indirect client) may in practice be longer than the structure discussed above. However, the permitted length of the chain is restricted under MiFIR to a client of a client of the indirect client, amounting to two additional links to the chain (i.e. CCP, CB, (direct) client, indirect client, client of indirect client, and client of client of indirect client).147 [15.95] An extended chain is only permitted where:148 (1) the indirect client (and in the case of the longest permitted chain, the client of the indirect client) that provides clearing services is an authorized credit institution, investment firm, or equivalent thirdcountry credit institution or investment firm; (2) the transactions arising as part of the indirect clearing arrangement are exclusively transactions entered into by the end indirect client whose assets and positions are managed by the CB in an OICA; and (3) certain group relationship criteria relating to the parties in the chain are met.149 [15.96] Where these conditions are satisfied, the (direct) client is subject to the obligations as if it were a CB and the indirect client is subject to the obligations as if it were a (direct) client.150 In the case of the longest permitted chain, the indirect client is also subject to the obligations as if it were a CB, and the client of the indirect client is subject to the obligations as if it were a (direct) client.151

5. Miscellaneous

[15.97] The Draft MiFIR ICA RTS have been the subject of much criticism, especially expressed by ISDA and FIA Europe. The main points of concern expressed by these major trade organizations include the following.152

A. Concerns about Default Management i. Extent of Protection [15.98] In relation to (direct) client clearing, we suspect that porting and the leapfrog payment are generally enforceable in European jurisdictions on the basis of the following. In accordance with the Banks RWD,153 a European CB which is a credit institution may in principle be subject only to insolvency proceedings of the home Member State where it has its head office. This is, however, subject to Article 8 SFD which provides: ‘In the event of insolvency proceedings being opened against a participant in a system, the rights and obligations arising from, or in connection with, the participation of that participant shall be determined by the law governing that system’ (the ‘CCP System Law’). It is unclear to what extent these ‘rights and obligations’ are covered. In our view, Article 8 SFD should be interpreted widely according to its purpose, and consequently be considered to provide that the CCP System Law will govern all rights and obligations of a European CB, being a participant within the meaning of the SFD, that derive from its participation in the system and the operation of the contractual arrangements governing that system, as constituted by the rules, default rules, procedures, by-laws, and other arrangements of the CCP which are binding on its clearing members (the ‘CCP Rules’). In the case of the insolvency of a European CB and where the applicable insolvency law in accordance with Article 8 SFD does not apply so as to impair the effect of CCP Rules that facilitate porting or the leapfrog payment, porting and the leapfrog payment will generally be enforceable. [15.99] However, it is less clear whether porting and the leapfrog payment are generally enforceable in European jurisdictions in the case of indirect clearing. In particular, the question arises of whether a client qualifies as a participant (for which the above analysis applies) or as an ‘indirect participant’. The latter type of participant is defined in the SFD as a credit

institution with a contractual relationship with an institution participating in a system executing transfer orders which enables the above-mentioned credit institution to pass transfer orders through the system. This suggests that a client is an indirect participant rather than a (direct) participant. To ensure the SFD protection (also referred to in paragraph 15.83), the local legislator must consider the (direct) clients of CBs as participants as well. It is therefore interesting to see that the Draft MiFIR ICA RTS (similar to the EMIR ICA RTS) do not appear concerned about whether or not the Member State option in the SFD to consider an indirect participant (i.e. the direct client) as a participant has been exercised.154 ii. Obligations of Means [15.100] The porting obligation seems to rely on the concept of an ‘obligation of means’ as opposed to an ‘obligation of results’ to attempt to mitigate the risks related to porting by reducing the obligation on CBs considerably. This raises questions on what the obligation actually means and when the obligation is fulfilled. It is not clear whether the porting obligation requires the CB to agree to initiate porting when requested by the defaulting client’s indirect clients, or whether the CB would be permitted to deny such requests for legitimate reasons. [15.101] Likewise it is unclear what is required for CBs to satisfy this ‘obligation of means’ to try to make leapfrog payments. Furthermore, the MiFIR RTS cannot override third-country insolvency regimes, which may prohibit the making of leapfrog payments which bypass an insolvency administrator of a defaulting client.155 In addition, in the absence of a harmonized European insolvency regime, a leapfrog payment may be challenged by an insolvency administrator of a defaulting EU client.

B. Concerns about Information Flow [15.102] In general, applying the obligations to longer chains of intermediate entities will have the effect of exacerbating the other general issues that apply to a four-party chain, because of the greater number of parties involved and the exponential growth in the number of accounts required. It also becomes more likely that one or more of the parties in the chain is a non-EU entity, which may be subject to conflicting laws and

regulations. In addition, the greater degree of separation between the CCP and end indirect client makes operational issues such as information flows more difficult and subject to increased operational and systemic risk.156

C. Concerns about Territoriality [15.103] It is unclear whether the scope of the MiFIR ICA RTS is limited to EU indirect clients. Such a limitation would be consistent with ESMA’s mandate under Article 30 MiFIR to develop RTS ‘ensuring consistency’ with the provisions under the EMIR ICA RTS, as the EMIR ICA RTS are limited to entities subject to the EMIR clearing obligation,157 which has limited application to non-EU indirect clients. [15.104] Similarly, the MiFIR ICA RTS should also only apply in respect of ETDs cleared on an EU CCP and not seek to regulate third-country CCPs’ rules and arrangements for indirect clearing of ETDs. To the extent such third-country CCPs are recognized as a result of the equivalence process for third-country CCPs established under EMIR and MiFIR, the ‘equivalent’ rules of the third country would apply instead of the MiFIR ICA RTS. This is consistent with Article 30(1) MiFIR which states that indirect clearing arrangements are permissible provided that the assets and positions of the counterparty benefit from protection with ‘equivalent effect’ to that referred to in Articles 39 and 48 EMIR. ESMA has clarified that the requirements of EMIR Article 39 apply only to EMIR-authorized CCPs, whereas non-EU CCPs wishing to provide clearing services to EU clearing members or trading venues would instead be subject to the thirdcountry recognition procedure in Article 25 EMIR (including an equivalence assessment).158

VI. The Obligations in Respect of Portfolio Compression

1. The General Obligations [15.105] The fifth set of obligations that Title V MiFIR introduces is in relation to a transparency and record-keeping obligation when providing portfolio compression. For the avoidance of doubt, there is no obligation to compress portfolios. Under Article 31, (i) investment firms and market operators providing portfolio compression must make public through an approved publication arrangement (APA)159 the volumes of transactions subject to portfolio compressions and the time they were concluded within the time limits specified in Article 10 MiFIR, and (ii) investment firms and market operators providing portfolio compressions must keep complete and accurate records of all portfolio compressions which they organize or participate in. Those records must be made available promptly to the relevant competent authority or ESMA upon request. [15.106] Although, at first sight, this may appear to be a burden for parties providing portfolio compression, it is actually a light-touch regime. Since portfolio compression de jure entails counterparties entering into a new derivative, the investment firm or market operator providing the portfolio compression service would in principle be subject to all the obligations posed on entering into a derivative (e.g. best execution rules, the trading obligation, and the transparency requirements). Because not all of these rules are tailored, and thus appropriate, for these circumstances, investment firms and market operators are exempted from the obligations under Articles 8, 10, 18, 21, and 28 of MiFIR and the obligations under Articles 1(6) and 27 MiFID II whilst providing portfolio compression services.160

2. What is Portfolio Compression? [15.107] Portfolio compression is the process under which multiple derivatives contracts are terminated and replaced by other, typically fewer, derivatives contracts. There are several different reasons to engage in this activity, for instance to standardize the coupons and/or the coupons period of the derivative, to make the derivative(s) eligible for clearing, or to facilitate the management of the contract.161 Another reason could be the

termination of a derivative transaction, which involves a two-step process. If a party would like to terminate a derivative in this way, it should first enter into an offsetting transaction.162 This is an identical trade, with the exception that where the party entering into the offsetting transaction is a buyer under the initial trade it would be the seller, and vice versa. Second, the parties have to tear up the contracts of both the initial derivative transaction and the offsetting transaction. Parties are now in a position to do so, since the transactions are equivalent but opposite, that is, the economic result of having both transactions in place is the same as having none in place. [15.108] Portfolio compression could also be initiated as a way to achieve multilateral netting,163 in this way reducing operational costs and minimising counterparty risk, as well as reducing systemic risk. Figure 15.7 shows how this could be achieved. Obviously, this could only work if at least three parties were involved in the compression exercise. In addition, the derivatives should be similar to a certain extent (i.e. class, tenor, currency, etc.).

Figure 15.7: OTC Derivatives.

[15.109] ‘Portfolio compression’ is defined in MiFID II as: a risk reduction service in which two or more counterparties wholly or partially terminate some or all of the derivatives submitted by those counterparties for inclusion in the portfolio compression and replace the terminated derivatives with

another derivative whose combined notional value is less than the combined notional value of the terminated derivatives.164

While this definition is generally consistent with the overview provided above, it is in certain respects more limited, as described further in Section VI.3 below. [15.110] The role of portfolio compression has become increasingly important and prevalent in the market, particularly due to the post-crisis regulatory reforms, most notably EMIR and CRR/CRD. Under EMIR, parties with 500 or more uncleared OTC derivative contracts outstanding must regularly, and at least semi-annually, ‘analyse the possibility to conduct a portfolio compression exercise in order to reduce their counterparty risk’. If there is a possibility to conduct such a compression exercise, parties should do so or explain why it would be inappropriate to do so.165 Naturally, this leads to an increase in compressions of uncleared OTC derivatives portfolios. A second regulatory development triggering an increase in compressions is the leverage ratio, which is imposed on banks and certain investment firms under CRR/CRD, whereby the calculation of derivatives exposure is based on the gross notional amount. As a result, firms subject to these rules now seek to allocate capital to the transactions in their books in the most efficient way. This incentivizes banks to compress their (cleared and uncleared) portfolios where possible.

3. Scope A. Criteria [15.111] The requirements regarding portfolio compression are not applicable to all compressions, but only to those falling within the definition of ‘portfolio compression’ (see Section VI.2). From this definition it is evident that four criteria must be established for a portfolio compression exercise to qualify as portfolio compression for the purposes of MiFIR: (i) the portfolio compression should be exercised by an investment firm or market operator; (ii) a number of derivatives should be terminated and replaced; (iii) the exercise should reduce the combined

notional value of the portfolio; and (iv) the exercise should be performed by two or more parties. The requirements apply to both ETDs and OTC derivatives. i. Re (i) the Portfolio Compression should be Exercised by an Investment Firm or Market Operator [15.112] There are a number of different types of entities which provide portfolio compression services. Some of them fall within the scope of MiFID II, whilst others do not. In scope of the rules are investment firms and market operators. This means that other entities offering portfolio compression services, such as CCPs or trade repositories, are not subject to these rules. This does not mean that these entities are not allowed to provide portfolio compression services, but merely that they are out of scope. ii. Re (iii) the Exercise should Reduce the Total Notional Value of the Portfolio [15.113] As a result of the portfolio compression the combined notional value of the derivatives submitted for compression should be reduced. This should be assessed at the portfolio level.166 This does not necessarily mean that the notional value of the derivatives of every counterparty participating in the portfolio compression should be reduced. However, the notional value of the derivatives of at least one of the counterparties should be reduced. Furthermore, in any case, it should not lead to an increased notional value of the portfolio of one of the counterparties. Whether a reduction has taken place when compressing a portfolio should be assessed by comparing the combined notional value of the portfolio submitted for compression with the combined notional value of that same portfolio after the compression has taken place. If the combined notional amount of the portfolio submitted for compression is higher than the combined notional amount of the portfolio resulting from the compression, the exercise qualifies as reducing the notional value of the portfolio. [15.114] There may be circumstances where two or more counterparties decide to submit a number of derivatives to a risk reduction service to terminate these derivatives and replace them with another derivative which does not have a lower aggregate notional value than the terminated derivatives. Counterparties could decide to perform such an exercise, for

instance, to standardize the coupons and/or the coupons period, to make the derivative(s) eligible for clearing or to facilitate the management of the contract.167 Although there could be good economic reason for parties to perform such an exercise, this falls outside the scope of the definition of portfolio compression, and consequently outside the scope of the obligations relating to portfolio compression under MiFIR. iii. Re (iv) the Exercise should be Performed by two or more Parties [15.115] Portfolio compression can be performed by two or more counterparties, that is, it can be bilateral or multilateral. With multilateral portfolio compression, typically a third-party service provider provides the portfolio compression service. Multilateral portfolio compression could be more favourable to parties, since more counterparties, and therefore more transactions, are included, which potentially increases the reduction of notional value of the portfolios. [15.116] It has been argued that the criteria for bilateral and multilateral compression should be ‘aligned and sufficiently high level to allow the development of further compression services’.168 An example of such a compression service would be a third type of portfolio compression, whereby portfolio compression is performed by a single counterparty vis-àvis the CCP, which is also referred to as unilateral portfolio compression.

B. Requirements [15.117] In addition to the level 1 definition, MiFIR provides the Commission with the opportunity to further set out the ‘elements of portfolio compression’ in a delegated act.169 We believe that ‘elements of portfolio compression’ should be read as ‘requirements to benefit from the light-touch regime’. 170 These requirements are focused on the process that should be followed for a market operator or investment firm to benefit from the light-touch regime. The (draft) delegated act focuses in particular on (i) the steps of the process, (ii) the content of the agreement, and (iii) the legal documentation supporting the portfolio compression.171

i. Process [15.118] As set out above, portfolio compression can take place in many forms. ESMA advised recognizing this diversity by making a distinction in the process which should be followed under the delegated act between when parties have engaged a third-party service provider and when parties have not done so. Contrary to this advice, the Commission has decided in the (draft) delegated act not to make any distinction on this point. In this way, it has created a uniform process which should be followed in order to benefit from the light-touch regime. [15.119] If the investment firm or market operator wants to benefit from the light-touch regime, the participants in a compression exercise should at least take the following two steps before the portfolio compression process is initiated.172 First, the participants should specify their risk tolerance, including, at a minimum, a limit to counterparty credit risk, a limit to market risk, and a cash payment tolerance.173 This risk tolerance of the parties should be respected by the investment firm or market operator in the portfolio compression exercise. [15.120] Second, links should be established between the derivatives submitted by the participants. In addition, the investment firm or market operator should submit to each participant a compression proposal, in which the counterparties affected by the compression are identified, the related change to the notional value of the transaction is set out, and the variation of the combined notional amount compared to the risk tolerance is specified. [15.121] Subsequently, the investment firm or market operator may allow the participants to add derivatives eligible for termination or reduction.174 Adding such derivatives should lead to an adjustment of the compression to the risk tolerance set by the participants and the maximization of efficiency of the portfolio compression. [15.122] Finally, the participants should agree on the compression proposal, or a subsequent compression proposal. After this, the investment firm or market operator performs the compression exercise.175

ii. Legal Documentation [15.123] Before parties perform their first compression exercise, they should conclude an agreement which provides for portfolio compression and its legal effects. This agreement should, at a minimum, (i) identify the point in time at which each portfolio compression becomes legally binding,176 and (ii) include all relevant legal documentation describing how derivatives submitted for inclusion in the portfolio compression are terminated and how they are replaced by other derivatives.177

4. The Obligations in more Detail A. Transparency [15.124] When providing portfolio compression, investment firms and market operators are subject to a post-trade transparency requirement.178 Under this requirement, an investment firm or market operator should make certain information with respect to the compression exercise public through an APA, including (i) a list of derivatives submitted for inclusion in the portfolio compression, (ii) a list of derivatives replacing the terminated derivatives, (iii) a list of derivatives changed or terminated as a result of the portfolio compression, and (iv) the number of derivatives and their value (i.e. the volume of the derivatives).179 The value of the transaction should be expressed in terms of notional amount, thereby disregarding the actual marked-to-market value of the derivative. This makes sense, since the marked-to-market value changes over time and may depend on other counterparties.180 The information which has to be published should be reported per type of derivative and per currency. [15.125] In line with the standard post-trade transparency rules for derivatives,181 the market operator or investment firm should make the details of the portfolio compression public ‘as close to real-time as technically possible’, and, in any case, no later than the end of the following business day (t+1).182 Parties should make the required data public as of the confirmation by the third-party service provider or the counterparties that

the portfolio compression is legally binding, following the acceptance by all counterparties of the compression proposal.183

B. Record-Keeping [15.126] Investment firms and market operators providing portfolio compression must keep complete and accurate records of all compressions which they organize or participate in.184 The MiFID II package does not provide for any further indication as to how firms should comply with this requirement. For instance, it is not exactly clear which information should be kept, or for how long. There is also no power for the Commission to set out further rules in respect of this obligation. It could be argued that this provision should be read in line with the standard record-keeping requirements,185 although no guidance has been given in this respect. If this provision were to be read in line with those requirements, the records kept under this obligation should be kept for five years (as a minimum). [15.127] Upon request by ESMA or the relevant competent authority, these records should be made available by the relevant firm or market operator. This differs from the standard record-keeping obligation, under which the records should only be made available to the relevant competent authority, and the role of ESMA is limited. Another difference seems to be that the records should be made available ‘promptly’, whereas standard record-keeping requirements do not include any time frame. Whether this is an omission of the European regulator or a deliberate deviation is unclear. However, it could be argued that, notwithstanding this apparent difference, the difference in practice will be limited, since firms will generally provide these records shortly after request by the competent authority.

VII. A Few Final Remarks [15.128] Time will tell whether the new rules discussed in this chapter will impact on the derivatives markets as they stand today, and, if so, to what extent and in what ways.

[15.129] The opinions on whether the trading obligation will have an effect on markets are very fragmented. This makes it hard to predict the exact consequences of the trading obligation. However, looking at the US, according to a Bank of England Staff Working Paper, activity and liquidity in the USD and EUR plain vanilla IRS markets have increased following the Dodd–Frank trading mandate.186 Furthermore, it is apparent that the associated execution costs in these markets are reduced significantly. The causes of this reduction in execution costs are likely the shortening in dealer intermediation chains and more competitive pricing by dealers. Both are potentially attributed to mandatory exchange trading, as the results of the research suggest that inter-dealer market activity has reduced following the trading mandate entering into force. However, the researchers could not scientifically prove a causal link between the trading mandate and the reduction in inter-dealer trading. As such, the full impact of the trading mandate remains uncertain. [15.130] With respect to the clearing obligation and the STP obligation, we expect the impact of these obligations to be relatively limited as they seem to be in line with current market practices. However, it remains to be seen whether the mandatory reduced time frames do not increase the risk of errors. [15.131] This is probably different for the indirect clearing rules, since the level of segregation and the default management processes as contemplated by MiFIR in the ETD space are not part of current clearing processes. [15.132] In relation to the portfolio compression rules under MiFIR, the European legislator has sought to continue to stimulate portfolio compression through regulation. The light-touch regime seems to strike the right balance between, where possible and appropriate, encouraging parties to compress their portfolios on one hand and increasing transparency for the entire market on the other. It is therefore conceivable that portfolio compression will remain a much-used exercise for the foreseeable future.

1

Quote by J. Gregory Central Counterparties (West Sussex: Wiley Financial Services, 2014), p. 3; but other commentators include S. Henderson: ‘derivatives are not what got us

here, US home mortgages are. […] Derivatives generally have not played any significant role in the creation or prolongation of the crisis’ (S. Henderson, ‘Regulation of Credit Derivatives: To What Effect and for Whose Benefit?—Part 4’ (2009) JIBFL 6, 342). 2 Paragraph 13 of the Leaders’ statement, the Pittsburgh Summit, 24–25 September 2009, . 3 Directive 2013/36/EU on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms (OJ L 176/338, ‘CRD’) and Regulation (EU) No. 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No. 648/2012 (OJ L 176/1, ‘CRR’). 4 Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse (OJ L 173/179, ‘MAD’) and Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16 April 2014 on market abuse (market abuse regulation) and repealing Directive 2003/6/EC of the European Parliament and of the Council and Commission Directives 2003/124/EC, 2003/125/EC, and 2004/72/EC (OJ L 173/1, ‘MAR’). 5 Regulation (EU) No. 648/2012 of the European Parliament and of the Council of 4 July 2012 on OTC derivatives, central counterparties, and trade repositories (OJ L 201/1, commonly referred to by the acronym ‘EMIR’ of its initial working title ‘European Markets Infrastructure Regulation’. 6 Regulation (EU) No. 600/2014 of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Regulation (EU) No. 648/2012 (OJ L 173/84, ‘MiFIR’). 7 Article 33 MiFIR on the mechanism to avoid duplicative or conflicting rules is out of scope of this chapter. 8 For the same reason we generally do not discuss, nor refer to, any documents that were part of the legislative process preceding the present documents at the time of writing. 9 MiFIR provides for a less generic definition, and rather a descriptive one in Article 2(1)(29): Those financial instruments defined in point (44)(c) of Article 4(1) of Directive 2014/65/EU; and referred to in Annex I, Section C (4) to (10) thereto. These are (i) securities giving the right to acquire or sell any transferable securities or giving rise to a cash settlement determined by reference to transferable securities, currencies, interest rates or yields, commodities or other indices or measures; and (ii) (4) Options, futures, swaps, forward rate agreements and any other derivative contracts relating to securities, currencies, interest rates or yields, emission allowances or other derivatives instruments, financial indices or financial measures which may be settled physically or in cash; (5) Options, futures, swaps, forwards and any other derivative contracts relating to commodities that must be settled in cash or may be settled in cash at the option of one of the parties other than by reason of default or other termination event; (6) Options, futures, swaps, and any other derivative contract relating to commodities

that can be physically settled provided that they are traded on a regulated market, a MTF, or an OTF, except for wholesale energy products traded on an OTF that must be physically settled;(7) Options, futures, swaps, forwards and any other derivative contracts relating to commodities, that can be physically settled not otherwise mentioned in point 6 of this Section and not being for commercial purposes, which have the characteristics of other derivative financial instruments; (8) Derivative instruments for the transfer of credit risk; (9) Financial contracts for differences; (10) Options, futures, swaps, forward rate agreements and any other derivative contracts relating to climatic variables, freight rates or inflation rates or other official economic statistics that must be settled in cash or may be settled in cash at the option of one of the parties other than by reason of default or other termination event, as well as any other derivative contracts relating to assets, rights, obligations, indices and measures not otherwise mentioned in this Section, which have the characteristics of other derivative financial instruments, having regard to whether, inter alia, they are traded on a regulated market, OTF, or MTF. 10

There are other usages, such as arbitrage whereby a party is looking to benefit from a price difference, for example from the fact that the current buying price of an asset falls below the price specified in a futures contract to sell the asset. 11 Article 1(32) MiFIR and Article 2(7) EMIR. 12 This is necessary as they should be liquid enough to be regularly traded on the relevant trading venue. 13 However, it is possible that certain derivatives are available in both an exchangetraded and an OTC version. 14 N. Smyth and A. Wetherilt, ‘Trading models and liquidity provision in OTC derivatives markets’, in Research and analysis: OTC derivatives markets: trading models and liquidity, Bank of England Quarterly Bulletin 2011 Q4, p. 332. 15 Consequently, systemic internalizers (SIs) and other platforms not falling within the definition of these types of trading platforms (e.g. bulletin boards) do not qualify as exchanges for the purposes of the trading obligation (see also Recitals 8 and 27 MiFIR). 16 Recital 26 MiFIR. 17 The predecessor of the European Securities and Markets Authority (ESMA). 18 CESR, Technical Advice to the European Commission in the Context of the MiFID Review—Standardisation and Organised Platform Trading of OTC Derivatives, October 2010, CESR/10-1096, p. 12. 19 European Commission, Public Consultation on the Review of the Markets in Financial Instruments Directive (MiFID), 8 December 2010, pp. 12–13, . With respect to the benefits of transparency, some parties have remarked that the pre- and post-trade transparency rules already cover this sufficiently. It has also been argued that too much transparency could harm liquidity, and therefore, more transparency is not by definition beneficial from a liquidity perspective.

20

CESR (n. 18), 10–11. CESR, Feedback statement on its Technical Advice to the European Commission in the Context of the MiFID Review—Standardisation and Organised Platform Trading of OTC Derivatives, December 2010, CESR/10-1210, p. 11. 22 See paragraph 15.39 for more on these margin requirements. 23 K. Shaik, Managing Derivatives Contracts: A Guide to Derivatives Market Structures, Contract Life Cycle, Operations, and Systems (New York: Apress, 2014); R. Baker, The Trade Lifecycle (Chichester: John Wiley & Sons, 2010). 24 This will usually involve a type of master agreement, which includes provisions governing all individual trades. Typically, the ISDA Master Agreement or the FOA Futures and Options Agreement, for OTC derivatives and ETDs, respectively. Parties also enter into additional documentation, such as documentation relating to credit support, clearing, and/or give-ups. 25 Settlement refers to the legal obligations of the counterparties being discharged, so the actual delivery/payment. For derivatives this occurs either by physical delivery or cashsettlement at the expiration of the duration of the trade. It may also be achieved by the prior closing out of the position. 26 Smyth and Wetherilt (n. 14), 334. 27 Unlike the US, where Swap Execution Facilities (SEFs) are required to offer both models as a minimum, MiFIR does not mandate any particular model on trading venues. However, it is not inconceivable that European trading venues will offer at least these two models too. See E. Benos, R. Payne, and M. Vasios, ‘Centralized trading, transparency and interest rate swap market liquidity: evidence from the implementation of the Dodd-Frank Act’, Bank of England Staff Working Paper, No. 580, January 2016, p. 7. 28 Although the initial pick-up of the CLOB model was rather limited in the US, it is now starting to gain traction there. Whether this will continue will all depend on how much liquidity CLOBs are able to offer. 29 To achieve markets that are as efficient as possible, leading to the lowest price for the party requesting a quote, it seems preferable that the platform redirects quote requests to as many dealers as possible. However, since this will increase the costs dealers incur (which will ultimately be borne by their counterparties), redirecting quote requests to too many dealers could actually increase hedging costs. For this reason, in the US the US Commodity Futures Trading Commission (CFTC) has indicated that quote request will only have to be redirected to (a minimum of) five dealers (G. F. Peery, The Post-Reform Guide to Derivatives and Futures (Hoboken, New Jersey: Wiley Finance, 2012), p. 217). 30 Article 2(8) EMIR; all financial counterparties included in this definition should be authorized in accordance with the relevant European directive to fall within the scope of this definition. 31 Article 2(9) EMIR. 32 Article 10(1)(b) EMIR. Article 11 Regulation (EU) No. 149/2013; the clearing thresholds for the purpose of the clearing obligation are: (a) €1 billion in gross notional 21

value for OTC credit derivative contracts; (b) €1 billion in gross notional value for OTC equity derivative contracts; (c) €3 billion in gross notional value for OTC interest rate derivative contracts; (d) €3 billion in gross notional value for OTC foreign exchange derivative contracts; and (e) €3 billion in gross notional value for OTC commodity derivative contracts and other OTC derivative contracts not provided for under points (a) to (d). 33 Article 10(3) EMIR; see Article 10 Regulation (EU) No. 149/2013 for the criteria for establishing which OTC derivative contracts are objectively reducing risks. 34 Article 28(2) MiFIR. 35 Article 2 of the draft Commission Delegated Regulation of 13 June 2016 supplementing Regulation (EU) No. 600/2014 of the European Parliament and of the Council on markets in financial instruments with regard to regulatory technical standards on the direct, substantial and foreseeable effect of derivative contracts within the Union and the prevention of the evasion of rules and obligations. It is similar to the RTS adopted under Article 4(4) EMIR on the same topic (Recital 6 of this draft Commission Delegated Regulation). 36 Article 3 EMIR and Article 28(1) MiFIR. 37 Recital 38 EMIR. 38 Article 89 EMIR. 39 Recital 26 EMIR. 40 Article 28(1) MiFIR. 41 Article 32(1) MiFIR. 42 The first RTS mandating a class of derivatives to be cleared was adopted in August 2015. This means that ESMA should have decided on this by February 2016 at the latest. At the time of writing, no such decision has been published. This has not come as a complete surprise, since Verena Ross, executive director at ESMA, indicated in June 2015 that at that time ‘ESMA will not have data available to check how liquid the relevant derivative classes are post-imposition of the clearing obligation and will not be able to assess the arrival of the new organised trading facilities (“OTFs”). This means an important piece of the puzzle is missing at a time when ESMA has to make its first assessments’ (Verena Ross’s Keynote speech at IDX 2015, London, 9 June 2015 (ESMA/2015/921)). At present, it is not clear when ESMA will be in a position to decide on this. 43 Article 5(4) EMIR; these criteria are further set out in Article 7 Regulation (EU) No 149/2013. 44 Article 32(2) MiFIR. Whilst the US trading mandate under Title VII, Part II, Section 723, Paragraph 8 of the US Dodd–Frank Wall Street Reform and Consumer Protection Act is broadly similar to the EU regime, these regimes do differ in the process of determining whether a certain class of derivatives is subject to the trading mandate. In the US, the regulator has taken a ‘venue led’ approach, under which trading venues should notify the CFTC when a class of derivatives subject to the clearing obligation is ‘made available to trade’. In this notification the trading venue could argue why the class of derivatives should not be subject to mandatory trading. Subsequently, the CFTC may give market participants

the opportunity to present their views. The CFTC will then decide whether the relevant class of derivatives will be subject to the trading mandate. ESMA has also considered a similar venue-led approach, but ultimately felt that it was not in a position to propose such an approach, since this would not be in line with the text of Article 32 MiFIR. (See also: FIA and FIA Europe Special Report Series: Derivatives under MiFID II—Part 1, 22 January 2015, p. 2; and ESMA Discussion Paper on MiFID II/MiFIR, 22 May 2014, pp. 189–90.) 45 Article 32(2)(b) MiFIR. 46 Article 32(3) MiFIR. These factors have been heavily criticized, since they would not be a correct definition of the concept of liquidity. However, ESMA has made clear that it has not tried to redefine the concept of liquidity, but merely to define liquidity for the purpose of the trading obligation (R. Buenaventura in ESMA, ‘ESMA Open Hearing On MiFID II / MiFIR—19 February 2015 in Paris’ , accessed 1 June 2016). 47 The draft Commission Delegated Regulation of 26 May 2016 supplementing Regulation (EU) No. 600/2014 of the European Parliament and of the Council on markets in financial instruments with regard to regulatory technical standards on criteria for determining whether derivatives subject to the clearing obligation should be subject to the trading obligation (‘Draft Commission Delegated Regulation on the Trading Obligation’). Although these criteria are in line with the criteria for the definition of ‘liquid market’ for non-equities under the pre-trade transparency requirements, the thresholds and the approaches to assessing these criteria are not necessarily the same. 48 Article 32(3) MiFIR. 49 ESMA Discussion Paper on MiFID II/MiFIR, 22 May 2014, p. 192. 50 Recital 3 Draft Commission Delegated Regulation on the Trading Obligation. 51 ESMA Final Report on the Draft Regulatory and Implementing Technical Standards MiFID II/MiFIR, 28 September 2015, p. 188. 52 According to the Netherlands Authority for the Financial Markets (see read-out of AFM meeting, 28 October, ISDA). This seems to be consistent with the opinion of ISDA (see ISDA Presentation on MiFID II/MiFIR: Transparency & Trading Obligation by Fiona Taylor, Director at ISDA, September 2014). COFIA refers to ‘classes of financial instruments approach’. 53 See Article 2(6) EMIR: ‘a subset of derivatives sharing common and essential characteristics including at least the relationship with the underlying asset, the type of underlying asset, and currency of notional amount’. 54 ESMA (n. 44), 185. 55 Article 32(4) MiFIR. 56 ESMA Final Report on the Draft Regulatory and Implementing Technical Standards MiFID II/MiFIR, 28 September 2015, p. 190. However, it should be noted that such derivatives could become subject to the EMIR clearing obligation, and ultimately to the MiFID trading obligation as well.

57

Set out in Article 32(2) MiFIR. Article 32(1)(b) MiFIR. 59 Article 32(5) MiFIR. 60 In accordance with Article 32(1) MiFIR. 61 ESMA (n. 56), 191. 62 Article 28(2) MiFIR. 63 Article 34 MiFIR. 64 Recital 10 Draft Commission Delegated Regulation on the Trading Obligation; ESMA (n. 56), 190. 65 ISDA ‘Response to ESMA’s MiFID II/MiFIR Consultation Paper of December 19, 2014’, 2 March 2015 , pp. 124–5. 66 ESMA (n. 56), 190–1. 67 Title III MiFIR. 68 Article 4(1)(20) MiFID II and Articles 12–17 Draft Commission Delegated of 25 April 2016 supplementing Directive 2014/65/EU of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive. 69 Article 5(2) EMIR and, to date, Regulation (EU) 2015/2205 for G4 currencies IRS. 70 It is unclear to us why this obligation surfaces in MiFIR and was not included in EMIR immediately, but it is likely that the European legislator wanted to prioritize G20 commitments (only relating to OTC derivatives) first. 71 Provided in Articles 35 and 36 MiFIR. 72 T. Hasenpusch, Clearing Services for Global Markets: A Framework for the Future Development of the Clearing Industry (Cambridge: Cambridge University Press, 2011), p. 18. 73 Article 2(3) EMIR. 74 It can therefore be said that clearing services not only benefit individual market participants, but markets as a whole by increasing their efficiency. Hasenpusch (n. 72), 1. 75 These include requirements on, inter alia, capital, supervision (as a credit institution or investment firm), operational capabilities (e.g. to post margin at short notice), participation in the default management process, and contribution to the default fund. 76 A clearing member and a CB are one and the same entity, albeit from the perspective of the CCP and the client, respectively. 77 This is different from the American ‘agency model’, in which the CB (called an ‘FCM’) acts as agent and guarantor of the client such that, at least in theory, the client faces the CCP through its agent, the clearing member, which then guarantees the client’s performance to the CCP, see S. Renas, N. Melnick, and C. Davis, ‘OTC Derivatives Clearing: How the Agency and Principal Models Compare’ (1 March 2012), International Financial Law Review, . 58

78

Other structures are referred to as ‘affiliated client clearing’, ‘non-affiliated client clearing’, and ‘prime-broker clearing’. (See FIA Europe response to MiFID II/MiFIR Discussion Paper, 1 August 2014.) 79 The EB may not wish to introduce its client to the CB, for fear of losing the client to the CB. 80 Article 39 EMIR. 81 The segregation under EMIR only protects the collateral held in the client account, so there still remains a risk for the client during the (usually) short period between the moment the CB has received the collateral from the client and when the collateral is posted to the client account (‘transit risk’). There is also a risk for the client if the CB posts collateral to the client account that differs in kind and value from the collateral that is posted by the client (although some clients may opt for collateral transformation). 82 In practice there are various forms of OSAs and ISAs, distinguishable by their particular manner of segregation. For example, ISAs can be distinguished in value segregation, whereby only the value of the collateral is secured, and asset segregation, whereby the actual collateral instruments are secured. Note that the market use of these terms does not always align with the regulatory use in EMIR and CRD. 83 See paragraph 15.38. 84 Article 39(9) EMIR. 85 Article 48(5), (6), and (7) EMIR. 86 Anywhere between 24 and 48 hours, depending on the CCP and type of account. 87 It is even questionable whether ISA porting would actually work in a time of stress, particularly if the number of CBs in the market has consolidated and reduced to approximately eight, as is expected by some commentators. 88 Article 48(7) EMIR. 89 In accordance with Article 2(14) EMIR. 90 Recital 1 of the European Commission’s draft Delegated Regulation of 19 June 2016 supplementing Regulation (EU) No 600/2014 of the European Parliament and of the Council with regard to regulatory technical standards specifying the obligation to clear derivatives traded on regulated markets and timing of acceptance for clearing (‘Draft Commission Delegated Regulation on STP’). The European Parliament and the Council will consider this Regulation, and if neither of them objects it will enter into force twenty days after its publication in the OJ. It will apply from the date appearing in the second paragraph of Article 55 MiFIR. 91 Baker (n. 23), 72. 92 Hasenpusch (n. 72), 37 (cf. Accenture (ed.) ‘Leaving Safe Havens: The Accelerating Evolution of the European Exchange Landscape’, White Paper, Sulzbach/Frankfurt aM, 2001). 93 Derived almost verbatim from FIA Europe’s response to ESMA’s MiFID II/MiFIR Discussion Paper of 22 May 2014, pp. 163–82.

94

In contrast to the OTC model, invalid orders are rejected at this stage and no manual intervention is required. 95 At every stage of the post-trade process chain the contractual relationships are clear, supported by (i) CCP and trading venue rules governing contractual relationships arising on trade execution, (ii) industry standard clearing agreements, (iii) FIA industry standard give-up agreements, and (iv) FIA standardized CB/client agreements. 96 And an obligation on the relevant CB to meet whatever margin requirements the CCP may deem necessary to manage the risks of that position. Given that each order also either implicitly or explicitly contains the minimum necessary clearing information, the trade details will be immediately reflected in the applicable clearing system and, provided the trade has the applicable client details attached, middle office systems will ensure that the trade will immediately flow through to the client’s account at the CB without any manual intervention, subject to the capacity and throughput performance of the CCP’s and CBss systems. 97 By virtue of an EB having to appoint a CB, there is always a clearing member that is responsible for the clearing of the trade as soon as the trade has been executed. This CB would continue to be responsible for the clearing of the trade in place of the ‘receiving’ CB if, for example, the trade is not allocated correctly. 98 Derived almost verbatim from ISDA’s reply to ESMA’s MiFID II/MiFIR Discussion Paper of 22 May 2014, pp. 180–206. 99 ‘ISDA/FIA Europe Cleared Derivatives Execution Agreement’, 9 February 2016 . 100 We understand that some participants have infrastructure in place to ensure that this step is conducted on a much quicker timescale (i.e. within a matter of minutes). 101 Article 29(2) MiFIR. As we saw in paragraph 15.33 above, MiFIR extends the scope of the clearing obligation to ETDs. It is slightly puzzling to us why the STP obligation only relates to derivatives and not to all financial instruments, since all CCPs, not only those clearing OTC derivatives, are subject to EMIR. 102 Article 1 Draft Commission Delegated Regulation on STP. 103 Article 2(2) Draft Commission Delegated Regulation on STP. 104 Article 2(3)(a) and (b) Draft Commission Delegated Regulation on STP. 105 Article 2(4)(a) and (b) Draft Commission Delegated Regulation on STP. 106 Article 3(2) and (3) Draft Commission Delegated Regulation on STP. 107 Article 3(4) Draft Commission Delegated Regulation on STP. 108 Article 3(4) Draft Commission Delegated Regulation on STP. 109 Article 2(1) Draft Commission Delegated Regulation on STP. 110 Article 4(1) Draft Commission Delegated Regulation on STP. 111 Article 4(2) Draft Commission Delegated Regulation on STP. 112 Recital 9 Draft Commission Delegated Regulation on STP. 113 Article 4(3) Draft Commission Delegated Regulation on STP. 114 Article 4(5) Draft Commission Delegated Regulation on STP.

115

In accordance with the Draft European Commission Delegated Regulation on Algorithmic Trading. 116 Article 4(4) Draft Commission Delegated Regulation on STP. 117 Article 5(1) and Recital 10 Draft Commission Delegated Regulation on STP. 118 Article 5(2) and Recital 11 Draft Commission Delegated Regulation on STP. 119 Article 5(3) and Recital 12 Draft Commission Delegated Regulation on STP. 120 EACH’s response to ESMA’s consultation paper on MiFID II/MiFIR dated 19 December 2014. 121 Please refer to paragraph 15.80 for the definition of this term. 122 The protection referred to in Articles 39 and 48 EMIR covers segregation, portability, and default procedures. 123 Recital 1 Draft MiFIR ICA RTS and Recital 4 Regulation (EU) No. 149/2013 of 19 December 2012 supplementing EMIR (the ‘EMIR ICA RTS’). 124 First, there is only a limited number of CBs, due to the strict membership criteria of CCPs. Second, due to strict capital requirements rules under CRD, including the leverage ratio, CBs have limited capacity and consequently tend to limit their clearing services offering to clients with (profitable) high volumes and to preferred clients. This will probably leave the majority of counterparties that are subject to the EMIR clearing obligation without access to (direct) client clearing. 125 At the time of writing, the expected RTS specifying rules for the admissible types of indirect clearing service arrangements had not yet been published. However, ESMA’s final report (Draft regulatory technical standards on indirect clearing arrangements under EMIR and MiFIR, 26 May 2016, ESMA/2016/725) does provide draft RTS (the ‘Draft MiFIR ICA RTS’) which we have used as the basis for this section. This followed on from the ESMA’s consultation paper (Indirect clearing arrangements under EMIR and MiFIR, 5 November 2015, ESMA/2015/1628, ‘ICA Consultation Paper’). In the context of the development of the Draft MiFIR ICA RTS, the responses to its consultation raised a series of concerns that also relate to the EMIR ICA RTS. Consequently, ESMA decided to address these issues by proposing amendments to the EMIR ICA RTS at the same time as preparing the Draft MiFIR ICA RTS. In the ICA Consultation Paper ESMA emphasizes the need for consistency between the EMIR ICA RTS and the Draft MiFIR ICA RTS, but ESMA also raises the question of how the OCT derivatives market and the ETD market differ, in order to determine on what aspects different rules are needed. In FIA Europe’s response to ESMA’s MiFID II/MiFIR Discussion Paper of 22 May 2014, it expressed the view that the Draft MiFIR ICA RTS requirements should differ from the EMIR ICA RTS requirements, because of the differences between the OTC market and the ETD market. (These differences include, inter alia: (i) there is less necessity for porting mechanisms in relation to ETDs, because ETDs are by their nature highly liquid and standardized and therefore much easier to close out; (ii) OTC derivatives markets and ETD markets involve fundamentally different products and structures; and (iii) the current indirect clearing arrangements for ETDs are different from the arrangements under the EMIR ICA RTS, so if indirect clearing is not permitted under its current form, the client would potentially need

to put in place additional legal relationships (which consequently would lead to a higher administrative burden and costs).) We will not discuss any changes in comparison to the current EMIR ICA RTS, such as the fact that porting to another CB need no longer be supported in the case of a client default (Article 4(4) EMIR ICA RTS vs Article 4(7) Draft MiFIR ICA RTS). 126 Article 1(1) Draft MiFIR ICA RTS and Article 1(a) EMIR ICA RTS. 127 Article 6(4)(b), (5)(b), and (6)(b) Draft MiFIR ICA RTS. 128 See Chapter II of Regulation (EU) No. 149/2013. 129 Directive 98/26/EC on settlement finality in payment and securities settlement systems. 130 Here, we have not included all indirect clearing obligations, but limited ourselves to a high-level overview. For example, we do not discuss that as indirect clearing arrangements may give rise to specific risks CBs and CCPs must routinely identify, monitor and manage any material risks arising from facilitating indirect clearing services (Articles 3(3) and 4(8) Draft MiFIR ICA RTS). 131 Article 17(4) EMIR. 132 Or equivalent third-country credit institutions or investment firms. Article 2(1) Draft MiFIR ICA RTS. 133 Articles 4(2) and 5(1) Draft MiFIR ICA RTS. 134 Articles 4(4) and 5(5) Draft MiFIR ICA RTS. 135 Article 3(2) Draft MiFIR ICA RTS. 136 The increased complexity associated with the higher number of entities between the CCP and the indirect client compared to (direct) client clearing arrangements should be mitigated with requirements for an alternative and operationally simpler choice of account structures for indirect clearing arrangements than for (direct) client clearing arrangements (see Recital 4 Draft MiFIR ICA RTS). 137 Recital 6 Draft MiFIR ICA RTS. 138 Article 5(2) Draft MiFIR ICA RTS. 139 Article 4(5) Draft MiFIR ICA RTS. 140 Article 4(6) Draft MiFIR ICA RTS. 141 Article 4(7) Draft MiFIR ICA RTS. 142 That other client or CB is obliged to accept those assets and positions only where it has previously entered into a contractual relationship with the relevant indirect clients in which it has committed itself to do so. 143 The reason that all of the relevant indirect clients must agree (probably up front) on one and the same back-up entity relates to the fact that under these post-default circumstances an OSA can only be ported to one back-up entity. 144 Paragraphs 44–47 of Section 3.2 of the ICA Consultation Paper. 145 Article 5(4), 4(3), and 3(2) Draft MiFIR ICA RTS. 146 Article 4(8) and Recital 17 Draft MiFIR ICA RTS.

147

Article 6(3) Draft MiFIR ICA RTS. Article 6(1), (2), and (3) Draft MiFIR ICA RTS. 149 Please refer to Article 6(1)(c), (2)(c), and (3)(c)/(d) Draft MiFIR ICA RTS for them. 150 Article 6(4) and (5) Draft MiFIR ICA RTS. 151 Article 6(6) Draft MiFIR ICA RTS. 152 These concerns (except for the one on the extent of protection under paragraphs 15.98 and 15.99) are derived from the responses of ISDA and FIA Europe to the Draft RTS. Both responses can be found at: . 153 Directive 2001/24/EC on the reorganization and winding up of credit institutions (‘Banks RWD’). 154 According to Article 2(f) SFD: 148

A Member State may decide that, for the purposes of this Directive, an indirect participant may be considered a participant if that is justified on the grounds of systemic risk. Where an indirect participant is considered to be a participant on grounds of systemic risk, this does not limit the responsibility of the participant through which the indirect participant passes transfer orders to the system. (emphasis added) 155

Paragraph 36 of the ICA Consultation Paper. In relation to information flows in longer chains, several concerns arise. Proper exchange of information requires the operational procedures and communication within the chain to function in a synchronous manner. In order to achieve this, standard documentation will be essential because otherwise it would be very difficult to align the contracts of all market participants. As mentioned, with longer chains it becomes more likely that a non-EU entity is party to the chain, so this documentation needs also to ensure the cross-border exchange of information (Michael Beaton, ‘Indirect Clearing: In Search of One Template To Rule Them All’, Derivatives Risk Solutions Blog, 21 December 2015. Accessed at: ). 157 This follows from the answer to Question 18 of ESMA’s EMIR Q&A. 158 This follows from the answer to Question 12(j) of ESMA’s EMIR Q&A. 159 As defined in Article 2(1)(34) MiFIR and 4(1)(52) MiFID II. 160 Recital 8 MiFIR. However, MiFIR does not include an exemption for the other rules applicable when entering into a derivative, such as the EMIR clearing and reporting requirements. 161 ESMA Final Report on ESMA’s Technical Advice to the Commission on MiFID II and MiFIR, 19 December 2014, p. 442. 162 This method can also be used to partially terminate a derivative transaction by entering into an offsetting transaction with a lower notional amount than the initial 156

transaction. 163 Gregory (n. 1). 164 Article 2(1)(47) MiFIR. 165 Regulation (EU) No. 149/2013 of 19 December 2012 supplementing Regulation (EU) No. 648/2012 of the European Parliament and of the Council with regard to regulatory technical standards on indirect clearing arrangements, the clearing obligation, the public register, access to a trading venue, non-financial counterparties, and risk mitigation techniques for OTC derivatives contracts not cleared by a CCP, OJ L 52/11. 166 ESMA (n. 161), 445. 167 ESMA (n. 161), 442. 168 ESMA (n. 161), 441. 169 Article 31(4)(a) MiFIR. 170 A strict reading of Article 31(4) seems to suggest that these elements (under Article 31(4)(a)) should be specified ‘in such a way as to make use as far as possibly of any existing record-keeping, reporting or publication requirement’. We do not see how the elements could be specified as such. Therefore, it seems likely that this is only relevant with respect to Article 31(4)(b), which refers to ‘the information to be published pursuant to [Article 31(2)]’. 171 Recital 16 of the European Commission’s draft Delegated Regulation of 18 May 2016 supplementing Regulation (EU) No. 600/2014 of the European Parliament and of the Council with regard to definitions, transparency, portfolio compression and supervisory measures on product intervention and positions together with an annex setting out the data to be provided for the purpose of determining a liquid market for shares, depositary receipts, exchange traded funds (ETFs) and certificates (the ‘Draft Commission Delegated Regulation on Portfolio Compression’). 172 Article 17(4) Draft Commission Delegated Regulation on Portfolio Compression. 173 This was also advised by ESMA in its Technical Advice, as it considered it important that the risk management framework of the counterparties would be respected when compressing portfolios (see ESMA (n. 161), 443–4). 174 Article 17(5) Draft Commission Delegated Regulation on Portfolio Compression. 175 Article 17(6) Draft Commission Delegated Regulation on Portfolio Compression. 176 Article 17(2) Draft Commission Delegated Regulation on Portfolio Compression. 177 Article 17(3) Draft Commission Delegated Regulation on Portfolio Compression. 178 Article 31(2) MiFIR. 179 See Article 18(1) Draft Commission Delegated Regulation on Portfolio Compression for the full list of information which has to be made public. 180 ESMA (n. 161), 446. 181 The time limits under the standard post-trade transparency rule for derivatives are specified in Article 10 MiFIR. 182 Article 18(2) Draft Commission Delegated Regulation on Portfolio Compression; according to ESMA, if a third party is providing the service of portfolio compression, the

publication of data should be made by such third party within a few minutes, whereas the time frame may be more lenient if no such service provider is used, since the process may then be more manual and more time may be needed. ESMA (n. 161), 446 183 Several options were considered by ESMA in this respect. However, ultimately, ESMA has concluded that this was the correct approach (see ESMA Consultation Paper on MiFID II and MiFIR, 22 May 2014, p. 311). 184 Article 31(3) MiFIR. 185 Under Article 25 MiFIR. 186 Benos, Payne, and Vasios (n. 27).

16 Commodity Derivatives Antonella Sciarrone Alibrandi and Edoardo Grossule*

I. Introduction II. Features of the Commodity Derivatives Markets and Its Relation to the Crisis III. Overview of the Most Important EU Provisions on Commodity Derivatives IV. 1. 2. 3. 4. V. 1. 2. 3. 4. 5.

The Regulatory Perimeter of the MiFID II/MiFIR Package Preliminary Remarks The Enlargement of the Definition of Commodity Derivatives Changes in the Exemption Regime The Ancillary Activity Exemption The New Position Limits Regime An Introduction to the Position Limits Regime The Distinction between Financial and Non-Financial Entities The Hedging Exemption The Methodology of Position Limits’ Calculation A First Evaluation of the Position Limits Regime: EU versus USA

VI. A New Set of Intervention Powers for Authorities and Trading Venues VII. Conclusion

I. Introduction [16.01] The commodity derivative sector has clearly been influenced by a process of regulatory innovation in, amongst others, the framework of the MiFID Review. We may say that it has been one of the areas most affected by the new wave of regulation, where the legislators were able to introduce many new important features in the applicable regulatory regime. In fact, despite not being recent—commodity derivatives date back to the late nineteenth century and were the first derivatives that spread on US markets —the commodity derivative sector has undergone major innovations in the past few years.1 [16.02] On one hand, the vast majority of such derivatives has fallen within the scope of application of MiFID II and MiFIR and, in particular, of the new and more stringent general rules on trading (including pre-trading and post-trading) provided by them. On the other hand, commodity derivatives are now subject to specific provisions—also included in MiFID II/MiFIR—issued by EU policymakers to tackle the turbulences that affected commodity markets during and after the financial crisis. In this respect it must be emphasized that, in recent years, many economists and legal scholars have identified the lack of effective rules as one of the main causes of such turbulences. [16.03] It must also be underlined that the inclusion of the rules on commodities and commodity derivatives in the MiFID Review—rules which will be discussed in further detail in the following paragraphs—has been accompanied by the adoption of further EU provisions also applicable to them, included in other EU Directives and regulations, some of which, such as EMIR,2 are already in force. Such EU provisions will be outlined in Section III below, along with the various regulatory strategies and instruments adopted from time to time by the EU legislator. This wider perspective will help to better understand the regulatory framework where MiFID II and MiFIR are inserted and to evaluate the effectiveness of the legal package as a whole. An overview of all new regulatory measures will be offered in the following, where we will try to verify if these measures are part of a real European strategy on this matter, or if they are simply the

expression of fragmentary interventions and as such not perfectly coherent with one another. [16.04] The regulatory process is, however, still in progress. With regard to the MiFID Review, ESMA has not implemented the Level 2 measures yet. A final draft of the Regulatory Technical Standards (hereinafter, ‘RTS Draft’)3 was issued by ESMA in September 2015, but the European Commission, following criticism in the European Parliament, asked ESMA in March 2016 to revise some of its proposed drafting, especially that regarding two profiles of commodity derivatives (the ancillary exemption and the position limit regime). Apart from the MiFID Review, there are further EU provisions related to commodity derivatives which are in the process of being approved and will be adopted in the near future; among them, the Regulation of Financial Benchmarks, on which the Council, the Commission, and the European Parliament found an agreement in November 2015, seems to be particularly relevant. [16.05] Beside the rules applicable to all varieties of commodities, provisions governing specific categories of these latter have also been adopted (e.g. REMIT, with regard to energy commodity markets4). And the idea of introducing a differentiated disciplinary treatment based on the type of the commodity (in particular, agriculture) has recently again been proposed by the European Commission, asking ESMA, as mentioned above, for a modification of the RTS Draft on the matter. Moreover, it must be pointed out that some Member States (e.g. France) have already intervened, adopting specific measures related to agricultural commodities, somehow anticipating, in this field, the regulatory solutions proposed by MiFID Review. [16.06] Before turning to comprehensive investigation of the different European interventions on commodities, the causes of the recent turmoil in the physical and financial commodity markets, as well as the characteristics and functioning of these markets, will be briefly described in Section II below. Such analysis seems useful for the purposes of the present chapter because, in the wake of such turmoil, several international organizations (e.g. G20, IOSCO, etc.) have pointed the finger at the scarce regulation of the sector and have highlighted the need for sharper regulation.

[16.07] In light of these clear mandates, the European Union implemented the above-mentioned wide range of countermeasures in order to tackle the weakness showed by markets on commodity derivatives. More or less at the same time, US policymakers adopted measures aiming at implementing the G20 mandates, such as Title VII of the ‘Dodd–Frank Wall Street Reform and Consumer Financial Protection Act’ of 2010 (hereinafter ‘Dodd–Frank Act’). The same happened in almost all the major countries of the world (e.g. India). Such measures will also be considered in the present chapter and compared with those implemented by the EU, in order to show that, despite the fact that all such measures find their roots in the G20 commitments and IOSCO principles,5 there are significant differences among them, especially with regard to the tools chosen from time to time by rule-makers to achieve the purpose of stabilizing commodity markets and making them more reliable and less exposed to speculation. And it is also useful to underline that in any country where new and specific rules regarding commodity derivatives have been issued, the debate on the real effectiveness of these rules is still open and fierce. In short, this topic is still highly controversial and relevant.

II. Features of the Commodity Derivatives Markets and Its Relation to the Crisis [16.08] At an early stage, the commodity derivatives were designed to be used by buyers/sellers of physical commodities (i.e. all the links in the supply chain) to manage the typical risks of basic markets and, among them, the risk of an adverse change in prices, by passing these risks on to the subjects who were willing to take them (hedging).6 [16.09] Some of these derivatives, like futures and related options, have always been financial in nature, since they are standardized, traded on regulated markets, and settled by payment of a differential. Beside them, there are other derivatives—such as forwards and related options—with ‘tailored’ terms and conditions, which have been traded OTC—the largest derivative market, which was not subject to margin collateral, data reporting, or other requirements of regulated exchanges—and settled by

physical delivery of the underlying commodities. They also function as an important instrument for price indication (price discovery), because the buying and selling on regulated markets of such derivatives helps to determine the spot market prices of commodities.7 [16.10] At first, commodity derivatives did not attract a great deal of attention from legislators and academics, because of the small extent of the phenomenon and the limited involvement of the financial institutions. Over the last ten years, however, there has been a significant change in commodity derivatives markets that can be attributed to several factors: (1) since the economic crisis, high risks and low rates of investment returns on many assets (stocks, bonds, real estate, etc.) have prompted many (institutional) investors to turn to the commodity markets, which were, at that time, very profitable on average; (2) this triggered the increasing dominance of financial participants that have no reasons to be related to producing, trading, or selling physical commodities (among these participants, a big role is played by investment banks, hedge funds, pension funds, and ETF funds linked to indexes); (3) as a consequence of this new scenario, the amount of investments in this sector, and especially in OTC derivatives, rose sharply.8 [16.11] It must be emphasized that such change in the commodity derivatives markets has been made possible (and, perhaps, encouraged) by the lack of clear rules, allowing for a strong increase in speculative flows. In addition, as it is well known, during the crisis there was a sharp increase in the volatility of prices of commodities. [16.12] The impact of this phenomenon was large enough to raise a debate among academics—mainly in the economic field—and, as already said, led many international organizations (G20, IOSCO, FSB) to dedicate their attention to it. Leaving aside the mixed results of economic research with regard to the relationship between financial speculation and volatility of commodities spot prices, it is important to note that, in the academic literature and in the official documents on the topic prepared by international organizations, it is generally acknowledged that, among the main causes of unstable and unpredictable prices (the others being more strictly related to production factors like climate and country risks), there

are commodities markets failures; and that these markets’ failures mainly depend on three causes: (1) the difficulties of the operators accessing comprehensive information on financial products on commodities; (2) the potential irrational behaviour of operators under given circumstances that may cause an overestimation or underestimation of commodity prices; and (3) a lack of rules, which may create incentives for excessive risk-taking or abuse of position by operators. [16.13] More specifically it must be emphasized that the volatility of agricultural products has had serious social implications and led, for example, to the Arab Spring. Such social implications, in turn, raised the level of attention and concern of the main international organizations, leading to their requesting a better, more stringent regulation not only of agricultural commodity derivatives but of commodity derivatives in general. With regard to regulation, however, we must not ignore the fact that, with specific reference to agricultural commodities, such concerns were (and still are) closely connected with those relating to food security and scarcity, and their dramatic geopolitical implications.9 This reason might lead to a differentiation of the treatment of the regulations of agricultural derivatives in comparison to others, as has already happened in France with peculiar regard to the position limits regime, and, as recently proposed by the European Commission, with regard to the RTS Draft issued by ESMA. [16.14] Having said this, it is also important to note that several international organizations have highlighted the need to find appropriate regulatory answers capable of allowing efficient transactions at no excessive cost, and operations aimed at the covering of risks (the proper functions of derivatives), while at the same time preventing the possible adverse impact of financial speculation. Notably, the Special Task Force on Commodity Futures Markets, set up by the International Organization of Securities Commissions (IOSCO) in March 2009, presented a Report that highlighted the need for greater transparency of commodity derivatives markets, especially regarding OTC activity, and urged national regulators to review their powers to ensure that they were able to gather appropriate information on OTC and physical markets.10

[16.15] Following these recommendations, at the 2009 Pittsburgh summit, the G20 stated the aim to ‘improve the regulation, functioning, and transparency of financial and commodity markets to address excessive commodity price volatility’.11 In April 2010, the Financial Stability Board created the OTC Derivative Working Group which is playing a big role in transposing the G20’s agreed objectives into domestic regulations and has issued several reports on the implementation of the OTC derivatives markets reform. [16.16] To fulfil the mandates of these international organizations, legislators (in Europe but also in the USA, India, etc.) issued several statutory provisions applying to commodity derivatives.

III. Overview of the Most Important EU Provisions on Commodity Derivatives [16.17] The new EU regulatory and supervisory framework for commodity derivatives markets introduces many instruments, with both strengths and weaknesses, designed to protect the integrity, and the price discovery and hedging functions, of such markets. The enactment through different laws and numerous important ESMA technical standards—most of which still need to be adopted—results in a complex framework, which cannot yet be fully assessed. It also raises questions about the supervisory and enforcement capacities of the relevant authorities. [16.18] Given that complexity and the interplay amongst the different sets of rules, prior to turning to a more detailed analysis of the MiFID II/MiFIR package it seems useful to provide an overview of such rules by categorizing them on the basis of the regulatory strategies that have been adopted by the EU legislator from time to time. [16.19] But even before this, it seems important to underline that, as a whole, those rules shape the new regulatory framework applicable to commodity derivatives. Within this context, it is possible to recognize two main different areas of action by the European legislator.

(i) The first is the choice to apply the rules regarding investment services and markets to the commodity derivatives by including a larger number of such derivatives in the relevant notion of financial instruments. (ii) The second area of action relates to the response to the problem of financial speculation in the commodities markets with a special set of rules applicable to commodity derivatives, whose contents vary depending on the nature of the counterparty (financial, ‘FC’, or nonfinancial, ‘NFC’) and on the purpose of the operation (hedging or speculative). [16.20] The (conceptually) first of the aims pursued by EU legislation is to avoid excessive financialization of commodity derivatives markets and support orderly pricing and settlement conditions, to prevent marketdistorting positions, to ensure convergence between the prices of derivatives and the spot prices for the underlying commodity, and to impede market abuse. [16.21] The most important measures pursuing this aim are included in MiFID II. Such provisions establish limits on ex ante positions related to all commodity derivatives and prohibit participants from holding commodity derivatives contracts that are beyond a quantitative threshold or limit (Article 57). Position limits are not imposed on those who trade for hedging of price risks, which means that position limits are not imposed on nonfinancial or commercial participants, except when they trade for speculative reasons. Instead, position limits do apply to financial entities (this subject will be analysed in further detail in Section IV below). [16.22] The same aim—although from a complementary perspective— can be found in the MAR/CSMAD package, which lays down stricter provisions to prevent market abuse, defined as insider dealing, unlawful disclosure of inside information, and market manipulation in relation to both derivatives and spot markets.12 [16.23] MAR also prohibits manipulation of benchmarks, but in 2013 the Commission adopted a proposal for a specific regulation on indices used as benchmark in financial instruments and financial contracts, on which an

agreement between the European Parliament and the Council of the EU was reached in November 2015.13 [16.24] A second aim to be considered is the reduction of the typical risks arising from OTC derivative contracts, including those for commodity derivatives, by outlining a regulatory framework for such derivatives and, at a second stage, moving part of them to trading venues. [16.25] This aim was initially achieved by adopting Regulation EU 648/2012 on European Market Infrastructure (EMIR). EMIR requires standard derivative contracts to be cleared through CCPs, as well as margins for non-cleared trades, and establishes stringent organizational business conduct and prudential requirements for these CCPs.14 The legislation also wants financial firms trading OTC commodity derivatives to have clear trades, while non-financial firms will only become subject to the clearing requirement if their positions—to be calculated without taking into account derivatives having a hedging15 purpose—breach an aggregate threshold of €3 billion.16 [16.26] The aim mentioned above can also be found in the MiFID II/MiFIR package. In particular, Regulation EU No 600/2014 on Markets in Financial Instruments (MiFIR) introduces the obligation (see Article 28) that OTC derivative contracts, which must be cleared according to EMIR, have to be traded on trading venues (regulated markets (RMs), MTFs, or OTFs). Which cleared OTC contracts have to be traded is based on the criteria set out in MiFIR (Article 32) and the regulatory technical standards drafted by ESMA, and is mainly determined by two factors: whether there is sufficient buying and selling (liquidity) of that particular derivative, and whether a derivative is admitted to a trading venue. The trading obligation applies to financial counterparties and non-financial counterparties that are subject to the clearing obligation according to EMIR. On its website, ESMA is to publish which derivatives are subject to the trading obligation, on which venue and from what date onwards (Article 34). [16.27] A third aim of the EU provisions in question is the improvement of transparency and circulation of information on the formation of prices

and traded volumes, both on trading venues and OTC, also introducing reporting duties. [16.28] As a consequence, EMIR ensures that information on all European derivative transactions is reported to trade repositories and is accessible to supervisory authorities, including ESMA, to give policymakers and supervisors a clear overview of what is going on in the markets. [16.29] In addition, under MiFID II, each operator of a trading venue has to publish a weekly report with aggregate information about the trade in different commodity derivatives on its venue, except when the number of persons trading, and their open positions, are below a threshold that is to be set by the European Commission. Trading venue operators, as well as investment firms, have to provide supervisors, on at least a daily basis, with detailed reports about positions held (Article 58). In addition, ESMA is required to publish a centralized report on commodity derivatives trading, based on the weekly aggregate reports from the venues, at a specific time in the week. [16.30] A fourth aim of the EU provisions being discussed is to provide an adequate and effective enforcement of the regulatory framework outlined above, by granting ESMA, the National Competent Authorities (hereinafter, ‘NCAs’), and the managers of the trading venues specific powers and responsibilities, including product intervention. [16.31] This aim is pursued by MiFID II, which provides, inter alia, that NCAs are given access to detailed information from trading entities and trading venues. In addition, the national competent authorities have the power to intervene in the markets and to impose sanctions (as further detailed in MiFIR; see also MAR), including imposing position limits on an individual person. [16.32] Furthermore, MiFIR ensures that competent authorities have the powers to protect both commodity derivatives and spot markets. They may prohibit and restrict the marketing, distribution, and sale of any financial instrument or activity that threatens the orderly functioning and integrity of such markets, the stability of the financial system, and investor interests.

National authorities are therefore granted intervention powers (in addition to those provided by MiFID II), including when a derivative has a detrimental effect on the price-formation mechanism in the underlying spot market (see Article 42).17 Either if national authorities have not taken adequate action or have not taken it at all, or in other situations stipulated by MiFIR, ESMA is empowered (Article 40) to intervene temporarily to prevent, prohibit, or restrict, in the EU, the sale of financial instruments or financial activities. [16.33] Under MiFID II, national authorities must also report to, and cooperate with, ESMA and other relevant national authorities, including public bodies responsible for the oversight, administration, and regulation of agricultural and energy spot markets (Article 79). [16.34] Under MiFID II, operators of any commodity trading venue are always obliged to apply position-management controls. For instance, they must be able to verify if a trade is for speculative or for hedging purposes. They have the power to monitor and access all necessary information and documentation. They can require a trader to terminate or reduce a position, or to provide liquidity back into the market. [16.35] Eventually, and with specific reference to trade repositories, under EMIR ESMA has the authorization, supervisory, and enforcement powers over them in the EU, and must recognize those in third countries. [16.36] A fifth category of EU provisions aims at ensuring that financial institutions—more precisely, banks and investment firms—undergoing risks connected with proprietary trading in commodity derivatives, meet specific prudential requirements. Such provisions are laid out in the CRD IV/CRR package.18 [16.37] More specifically, CRR establishes uniform EU rules for very detailed prudential requirements to set aside capital buffers and riskmanagement requirements. This prudential regime applies to credit institutions as well as to investment firms, but is adapted for the latter as a function of the investment services they provide.

[16.38] It is worth noting that Articles 493(1) and 498(1) of CRR exempt ‘commodity dealers’19 from large exposures requirements and from own funds requirements respectively. Both exemptions will expire on 31 December 2017. This deadline will probably be extended in order to allow the Commission to finalize its proposal of an appropriate regime for the prudential supervision of investment firms in general (see Proposal of the Commission COM(2015) 648 final). [16.39] Besides the above, it is finally worth considering here that some rules on commodity derivatives are also to be found in the European provisions on collective investment schemes, more exactly in Directive 2014/91/EU on Undertakings for Collective Investment in Transferable Securities (UCITS V) and in Directive 2011/61/EU on Alternative Investment Fund Managers (AIFMD). Some of the rules introduced in this context are aimed at protecting investors (and have a very different perspective compared to the interventions described up to now); other rules aim instead at avoiding the situation where transactions put in place by collective investment schemes negatively affect the orderly functioning of commodity markets. [16.40] In particular, UCITS V, in aiming to protecting investors (especially those in retail), prohibits direct investment in commodity derivatives with the capital from these investors in the fund. Instead, UCITS funds are allowed to track a commodity index, if that index is composed of different commodities and is transparent.20 [16.41] AIFMD, in its turn, does not set a prohibition of direct investment in commodity derivatives, but lays down rules aiming to avoid transactions put in place by alternative fund managers negatively affecting the orderly functioning of commodity markets, by setting the reporting requirements and intervention powers of the competent authorities.21

IV. The Regulatory Perimeter of the MiFID II/MiFIR Package

1. Preliminary Remarks [16.42] As said above, the new regulatory framework introduced through the MiFID Review brings with it the increased attention of the EU legislator with regard to commodities and commodity derivatives. Most of these now fall within the scope of the application of the new rules, with the two kinds of effects described above. As a consequence, to exactly define the scope of application of this new set of rules, two things are needed. On one hand, it is important to understand which commodity derivatives are included within the definition of financial instruments and are subject to MiFID II/MIFIR; on the other hand, it is relevant to properly identify the perimeter of the exemptions set by MiFID II/MIFIR in relation to dealing in commodity derivatives. In fact, the MiFID Review also made significant changes to these exemptions by restricting their scope of application— compared to the number of exemptions made by MiFID, the ancillary exemption remains, in the end, the only useful one. [16.43] With regard to the scope of MiFID II and MiFIR, the RTS Draft, published by ESMA on 28 September 2015, is also to be taken into consideration. In particular, Article 2(4) of MiFID II requires ESMA to develop RTS to specify the criteria for establishing when an activity is to be considered as ancillary to the main business on a group level.22

2. The Enlargement of the Definition of Commodity Derivatives [16.44] It is worth recalling that Directive 93/22/EC (ISD) excluded commodity derivatives from its scope. As a consequence, firms dealing in such derivatives were not subjected to the Directive and could not benefit from the European passport. MiFID I entailed a significant change to the approach of the EU legislator, due to which some of such derivatives were (and still are) included in the notion of financial instruments that can be the object of investment services and activities carried out by investment firms and banks.

[16.45] In particular, MiFID I introduced specific definitions detailing which commodity derivatives are included in its scope.23 Definitions C5, C6, and C7, contained in Section C of Annex 1, list the derivatives with an underlying commodity to be qualified as financial instruments having financial features. While definition C5 concerns the derivatives contracts relating to commodities that must be settled in cash or may be settled in cash at the option of one of the parties, definitions C6 and C7 regard the derivatives that can be physically settled. In more detail, C6 mentions some derivatives traded on a regulated market or an MTF, while C7 includes all other derivatives relating to commodities, even if traded OTC, except those having commercial purposes. [16.46] Over the years, the new definitions have developed a series of differences due to their interpretations by national authorities, so that ESMA, in May 2015, was obliged to enact Guidelines24 in order to ensure a consistent application of the commodity derivatives definitions. Even in the presence of MiFID II, these Guidelines are important because the definitions they contain both refer to MiFID I—which will remain applicable until 3 January 2018—and are also at the basis of the EMIR Regulation already in force. In these Guidelines, ESMA has enlarged the definition scope, firstly saying that C6 applies to all commodity derivatives contracts, including forwards, providing that they can or must be physically settled, and they are traded on a regulated market and/or an MTF. Moreover, the Guidelines enlarge the meaning of ‘physically settled’ derivatives, including a broad range of delivery methods. Unfortunately, even these new Guidelines cannot be said to clarify things, especially with reference to the definition of ‘physically settled’. [16.47] With the MiFID Review, the commodity derivatives considered as financial instruments were further expanded with the aim of following the inputs of the G20 mandate mentioned above.25 In this revision, C6 and C7 have been implemented taking into account regulatory innovations in the field of derivatives in general. (i) In more detail, the types of commodity derivatives covered in Section C6 of Annex I to MiFID II now include not only those traded on a regulated market or an MTF but also those traded on an OTF, the new

trading venues designed by MiFID II, excluding those which must be physically settled. (ii) C7, instead, has been emended because of the EMIR provisions for OTC derivatives (in particular, the clearing obligation) and, contrary to the earlier version, it does not contain any reference to clearing and settlement through recognized clearing houses or to margin calls. (iii) The reason for both these extensions is the economic equivalency of these commodity contracts to financial instruments and the fact that they pose similar risks. [16.48] With regard to the definitions of commodity derivatives adopted by MiFID II, it is always worth noting that there are some specific provisions concerning the energy derivative contracts. [16.49] First of all, C6 excludes, from the financial instruments list, the wholesale energy products traded on an OTF that must be physically settled.26 The main reason for such an exclusion is that these contracts fall within the scope of Regulation EU No 1227/2011 on wholesale energy market integrity and transparency (REMIT), which introduced a specific legal framework in order to prevent market manipulations in this crucial field. As a consequence, they are subject to a level of regulation and supervision comparable with financial markets legislation and so their exclusion is justified as a proportional amendment to avoid unnecessary dual regulation (for this reason they are defined as ‘REMIT carve-out’).27 Without this exclusion, a dual regulation would increase both the regulatory burden and the cost of trading in the energy market, reducing liquidity and causing a negative impact on prices for energy consumers and on security of supply, undermining efforts to build a strong internal energy market.28 [16.50] In addition to the permanent exemption described above, in the MiFID package there are also two temporary exemptions. One of these is provided by Article 95 of MiFID II and exempts C6 energy derivative contracts from certain obligations under EMIR Regulation. The other one is stipulated in the ESMA RTS Draft, through an exclusion from C6 of the energy derivative contracts relating to coal or oil which are traded on an OTF and must be physically settled.29

3. Changes in the Exemption Regime [16.51] Besides the new definitions of commodity derivatives, MiFID II introduces important changes to the exemption regime in a restrictive sense, always with the aim of responding to the excessive speculation and risktaking problems, but at the same time not hindering the business for nonfinancial entities too much. More precisely, this change has been made with the objective of a narrower possibility that trade operators undertake auxiliary financial services related to commodity derivatives when offering their customers hedging instruments (for more clarification on the distinction between financial and non-financial entities, and between hedging and speculative derivatives, see further Section IV). [16.52] Under MiFID I, the presence of a number of exemptions is intended to cover commercial users and producers of commodities. In particular, beside the general exemption for intra-group activities, these entities have one of the following three exemptions at their disposal: (i) dealing on own account exemption: with some exceptions, set forth in favour of entities that do not provide investment services or perform investment activities other than dealing on own account; (ii) ancillary exemption: set forth in favour of entities trading financial instruments on their own account or providing investment services in commodity derivatives to clients of their main business, as far as this activity is ancillary to their main business considered at the group level, and provided that such main business is not the provision of investment services or banking services; (iii) specialization exemption: set forth in favour of ‘persons whose main business consists of dealing on own account in commodities and/or commodity derivatives’; not applicable, however, if such entities belong to a group whose main business consists in the provision of other investment or banking services included in reserved activities. [16.53] Under MiFID II, instead, the exemption for those who (i) deal on own account (including market-makers) in commodity derivatives, excluding persons who deal executing client orders, or (ii) provide investment services, other than dealing on own account, operates only

under the following three conditions (Article 2(1)(j), amending the previous Article 2(1)(i)): (i) for each of those cases individually and on an aggregate basis, this is an ancillary activity to their main business, when considered on a group basis, and that main business is not the provision of investment services within the meaning of this Directive or banking activities under Directive 2013/36/EU, or acting as a market-maker in relation to commodity derivatives; (ii) those persons do not apply a high-frequency algorithmic trading technique; and (iii) those persons notify annually the relevant competent authority that they make use of this exemption and, upon request, report to the competent authority the basis on which they consider that their activity of dealing on own account is ancillary to their main business. [16.54] The previous exemption (Article 2(1)(k) MiFID I), therefore, has been abolished. [16.55] As just said, the ancillary exemption cannot be applied when techniques of high-frequency algorithmic trading (HFT)30 are involved. These techniques have raised many doubts about their impact on the markets, including commodity markets. Faced with these problems, MiFID II introduces a number of regulatory provisions, with the aim of limiting the negative effects of HFT, and particular care is taken to prevent possible market manipulations. These measures comprise specific organizational requirements for both intermediaries and trading venues involved in the practice of the market, as well as a series of requirements designed to ensure greater transparency in these negotiations. With the same objective, it was also established that for those who trade using algorithmic techniques, the exemptions provided for by MiFID II do not operate; and this happens even to those who operate in commodity derivatives (Article 2(1)(j)). Regardless of the purpose of hedging or speculation, these investment firms will therefore apply position limits, in addition to the set of rules introduced in the general MiFID II for operators using these algorithmic trading techniques and to all the transparency requirements contained into the Directive.

4. The Ancillary Activity Exemption [16.56] Under MiFID I, in order to determine whether an activity could be considered ‘ancillary’ to the main business, there are no quantitative criteria. In general, the valuation of the activity is traditionally understood as excluding dealing on own account for speculative purposes in the case of commodity producers.31 For these reasons, and in the absence of quantitative criteria, market participants do not typically calculate the scale of their activities in specific instruments weighed in relation to the main business. [16.57] Moreover, under MiFID I, market participants are not subject to any obligation to notify competent authorities of the use of the ancillary exemption. As a consequence of this regulatory scheme, competent authorities have developed supervisory practices in order to qualify an activity as ‘ancillary’ to the main business. Usually, this evaluation is made on a case-by-case basis, thus such practices are not harmonized across the EU. [16.58] The new MiFID II regime implies some regulatory changes in this area.32 Here, the first key legislative modification introduces strict quantitative parameters, which will delineate which trades and volumes can be considered ‘ancillary’. Moreover, when these parameters are present, interested entities must give annual notification to the competent authority that they are making use of this exemption and, eventually, upon request, report to the competent authority the basis on which they consider this activity ancillary to their main business. [16.59] With regard to the quantitative parameters, in order to establish the proportion of ancillary and main businesses, both the size of trading activity and capital employed should be considered as main indicators. With regard to non-financial firms, these criteria, on one hand, confirm an effective exemption; on the other, they widen the scope for the Directive to including commercial users dealing in financial instruments ‘in a disproportionate manner’ compared with the level of investment in the main business.33

[16.60] In its consultation paper ESMA proposed two tests that need to be passed cumulatively in order for investment activities to be considered ancillary. The first test covers trading activity thresholds; the second test main business thresholds.34 (i) The trading activity threshold test compares the size of the firm’s trading activity to the size of the overall market trading activity in the EU (i.e. determining the market share of a given entity in a commodities derivatives class). In its final report ESMA considers that there is merit in applying the trading activity test at an asset classspecific level.35 In fact, ESMA highlights that commodity markets differ significantly in terms of size, number of market participants, level of liquidity, and other characteristics. As a consequence, different thresholds will apply for different asset classes in relation to the test on the size of the trading activity.36 ESMA has also noted that, for this test, only the activity undertaken for non-hedging purposes has to be taken into account. (ii) In relation to the main business threshold test, in a first proposal ESMA established that it was based on considering the ratio of the capital employed in carrying out the ancillary activity to the capital employed in carrying on the main business. In the final report ESMA states that it has reviewed and ‘fundamentally changed’ the approach in relation to the calculation, abandoning the capital employed test and taking into account the overall activity of a group’s main business without any further reductions (encompassing privileged transactions and transactions executed in an entity of the group authorized in accordance with MiFID II or the CRD IV). [16.61] ESMA also clarifies that the comparison of the ancillary activity with the main activity should be done by comparing all ancillary activities taken together against the main activity. Where a firm undertakes only one of the ancillary activities mentioned in Article 2(1)(j) (dealing on own account or providing investment services), it would have to undertake the ancillary test only on the basis of this individual ancillary activity. [16.62] It is worth noting that the Draft of RTS 20 has been criticized by the European Commission, which has called for a tighter definition of

capital in the ‘ancillary activity’ test to make sure that position limits cover all relevant actors.

V. The New Position Limits Regime 1. An Introduction to the Position Limits Regime [16.63] One of the major changes in the EU commodity derivatives regulation is undoubtedly the introduction of position limits. The objectives underlying the imposition of this measure are to diminish, eliminate, or prevent excessive speculation; to deter and prevent market manipulation, squeezes, and corners; as well as to ensure sufficient market liquidity for bona fide hedgers (in technical terms, the regulator fosters hedge–hedge and hedge–speculative transactions).37 In particular, Article 57 of MiFID II implemented this ex ante limit on positions that operators can hold on commodity derivatives. RMs, MTFs, and OTFs must apply the limits on commodity derivatives traded. [16.64] With regard to the position limits regime, the RTS Draft, published by ESMA on 28 September 2015, is also to be taken into consideration.38 In particular, Article 57(3) and (12) of MiFID II requires ESMA to develop RTS to specify the methodology for the calculation and application of position limits for commodity derivatives traded on trading venues and economically equivalent OTC contracts.39 [16.65] It is worth noting that not only MiFID II but also the American legislator—as it will be explained below—have introduced ex ante limits as regards the positions in commodity derivatives, not for hedging purposes, beyond which individual operators must not go.

2. The Distinction between Financial and NonFinancial Entities

[16.66] In the context of this new regulatory framework, one of the crucial aspects is represented by the exemption for non-financial entities from the mandatory position limits regime. More precisely, the position limits do not apply to positions held which are objectively measurable as reducing risks directly related to the commercial activity of such nonfinancial entity (hedging exemption). Because of that, the Directive introduces specific rules in order to differentiate such positions from any other trades. In relation to the above reasoning, the hedging exemption is not available to financial counterparties. [16.67] In this case, it first is important to distinguish clearly between ‘financial counterparty’ (FC) and ‘non-financial counterparty’ (NFC). Because MiFID II does not provide any definition of a non-financial entity, it seems appropriate to apply the notions given by EMIR, which introduces a distinction between FC and NFC.40 The definition of FC refers to the exercise of financial activities from subjects authorized by proper disciplines, like banks, insurance companies, investment fund managers, etc.: in general terms, it could correspond to the MiFID ‘investment firm’. NFC, instead, refers to ‘an undertaking established in the Union other than the entities referred to in points (1) and (8)’.41 In other words, for the purpose of the position limits regime a non-financial entity can be considered the same as a non-financial counterparty under EMIR. [16.68] NFC activities, with no systemic importance,42 can consist in the stipulation both of hedging derivative contracts and of non-hedging derivative contracts (these latter are included in the position limits regime). OTC derivatives contracts are however used by non-financial entities, mainly for the purpose of hedging commercial and treasury financing activities.43 [16.69] EMIR preferential treatment towards NFC aims to differentiate the applicable rules regarding clearing obligation, transparency obligation, etc. related and proportional to the subject status. To understand EMIR’s favourable disposition towards NFC against derivative risks, it can be useful to consider the advice of the European Systemic Risk Board, which says:

in a number of market segments, such as commodities markets, which were previously mainly accessed by non-financial corporations, the use of OTC derivatives for speculative, investment or trading purposes has become predominant. This is partly a response to the entry of financial institutions into these markets, the so-called financialization process. This calls for a prudent approach to the risks posed by the use of derivatives by non-financial corporations.44

3. The Hedging Exemption [16.70] Now that the definition of FC and NFC has been made clear, it is important to analyse the reasons why hedging derivatives held by nonfinancial or commercial users are not included in the position limits and the scope of the hedging exemption.45 To understand the reasons it is worth remembering that such limits, as confirmed by the hedging exemption, aim to contain (but not prohibit) the speculation (not removing liquidity on the market) in order to improve convergence between spot markets and derivatives markets. In other words, this measure favours two types of transactions: technically, the hedge–hedge and the hedge–speculative. [16.71] As regards the scope of the hedging exemption, it becomes crucial to define the meaning of ‘positions objectively measurable as reducing risks directly related to the commercial activity’, as outlined by Article 57(1) MiFID II.46 In this perspective, it could be useful to consider the standard developed for the purpose of EMIR. In that context, a derivatives contract is objectively measurable as reducing risks directly relating to the commercial activity of the non-financial counterpart when, by itself or in combination with other derivatives contracts, it fulfils at least one of the three criteria on which the hedging test is based.47 Even if the hedging exemption is used in EMIR in order to exclude the application of the stricter rules concerning OTC derivatives, in the consultation paper ESMA considers, as regards the hedging criterion, that the discrepancies between MiFID II/MiFIR and EMIR are not so relevant.48 In addition, ESMA, in its RTS Draft, implements this assumption, also widening the hedging definitions to make them applicable to traded derivatives contracts, too.

[16.72] Hence, ESMA’s efforts to give a common interpretation of these legal frameworks seem as reasonable as they are appropriate, considering potential negative impacts in terms of cost of compliance for market operators.49 [16.73] It must be underlined, thus, that the hedging exemption does not operate automatically but requires a notification from the NFC and regulatory approval by the competent authority. In this perspective, a specific procedure, by which interested non-financial entities may be exempted from the position limits regime, will be specified by ESMA at Level 2 legislation.50 The requirements for the hedging exemption procedure seem to be similar to the notification schemes used for the EMIR hedging exemption. At the moment, on this aspect, the ESMA outline under the scope of MiFID II is that the exemption is available only in respect of specific positions. In other words, the hedging exemption cannot be considered a universal exemption for certain types of persons, which exempts them from position limits for all activities undertaken in all commodity derivatives contracts. The basis for the notification is, instead, related to specific contracts. Thus, ESMA proposes that the notification is made to the competent financial authority of the relevant trading venue.51

4. The Methodology of Position Limits’ Calculation [16.74] The position limits regime, introduced by MiFID II, may be applied for each contract, both physically settled and cash settled, with respect to positions in commodity derivatives. The RTS 21—concerning the regulatory technical standard for position limits—introduces two different methodologies to the calculation of position limits for spot and other months’ contracts (Article 9 and following).52 In this regulatory proposal, the position limits in the spot month should be based on deliverable supply;53 by contrast, in other months, the limits should be based on total open interest (where there is no underlying deliverable supply for a commodity derivative, the spot month position limit should be based on open interest, too).

[16.75] According to the RTS Draft, spot month position limits for cash settled and physically settled commodity derivatives are based on a percentage of deliverable supply, and the baseline is set at the 25 per cent. In the RTS Draft, ESMA changed the earlier approach and adopted a more stringent formulation, allowing Member States’ competent authorities to increase the 25 per cent baseline position limit to 35 per cent, and to decrease it to 5 per cent of the deliverable supply (Article 9 RTS 21).54 Regarding the other months’ limits, both cash settled and physically settled commodity derivatives are based on a percentage of total open interest in the commodity contract excluding open interest in the spot month. Other months’ limits are calculated as 25 per cent of average annual open interest, expressed in lots in the underlying commodity (‘the baseline’). [16.76] ESMA also adopted the asymmetric spread for the other months, giving Member States’ competent authorities the power to adjust the baseline down to 5 per cent and only up to 35 per cent of open interest, taking into account the factors listed under Article 57(3)(a) to (g).

5. A First Evaluation of the Position Limits Regime: EU versus USA [16.77] Although there is scientific uncertainty about the existence of an effective link between financial speculation and increased volatility of spot prices of commodities, and the position limits strategy is not unequivocally seen as an effective tool against speculation, these kinds of rules are present today in various legal systems, among them, that of the USA. As a matter of fact, in this country, position limits have been present for a long time because they have been thought of as an effective tool to prevent market abuse and, especially, the risk of ‘cornering’ the market. In more detail, this tool had already been included in the US regulations exchange trade derivatives markets since the Commodity Exchange Act (CEA) of 1936, which explicitly pursued the goal of reducing the risk of distorting the market, authorizing the Commodity Futures Trading Commission (CFTC) to impose limits on the size of speculative positions in futures markets.55

[16.78] Following this streamlining but with the further aim to prevent the excessive financialization and speculation of commodity markets, the Dodd–Frank Act of 2010 decided to move the trading of derivatives to trading venues while extending the limits of positions to the new markets. The reasons for the implementation of such limits include Congress’s interest in curbing excessive speculation that would cause price volatility in a given commodity.56 Title VII, Section 737 of Dodd–Frank empowered the CFTC to establish position limits for certain commodity-based derivatives, other than bona fide hedge positions, that may be held by any person with respect to physical commodity futures and option contracts. This Section also covers the previously unregulated swaps market, including ‘swaps that are economically equivalent to contracts of sale for future delivery or to options on the contracts or commodities traded on or subject to the rules of a designated contract market’, and requires the establishment of ‘aggregate position limits involving these previously unregulated instruments’. It must be underlined that even before Dodd–Frank, most agriculturally based commodity markets already had federally mandated position limits, but now the new rule extends position limits to metal and energy commodities markets. Like the agriculturally based commodities, these new position limits on metals and energy will extend across multiple trading venues, in order to prevent the effects that one large trader may have on price fluctuations across exchanges on a single commodity.57 [16.79] To implement the Section cited above, in October 2011 CFTC first issued some rules on position limits concerning twenty-eight physical commodity futures contracts, and futures and swaps that were ‘economically equivalent’ to those contracts.58 But in September 2012, a US District Court vacated the rule and remanded it back to the CFTC, which issued a new one in November 2013. The comment period of this proposal was reopened a number of times in 2014 and in 2015. Recently, in February 2016, the CFTC Energy and Environmental Markets Advisory Committee (EEMAC) released a report on the CFTC’s position limits proposal saying the rule as proposed is unnecessary because it could harm liquidity and would create numerous practical challenges. [16.80] What has been said up to now shows that the strength of the debate means that it is very difficult to reach an agreement on this topic. On

one hand, there are those who want position limits, saying they would curb speculative trading by putting a cap on the number of derivatives contracts a trader can have in a given commodity. On the other hand, some have argued that limits threaten legitimate hedging and other risk-reduction strategies used by companies that have big exposure to commodity price swings, such as airlines and agricultural equipment makers. Moreover, they state that position limits harm the ability of traders to react to market shocks and could worsen volatility in commodity price swings. [16.81] The same divergence of opinions can also be found in Europe and is one of the reasons which urged the European Commission to criticize the RTS Draft proposed by ESMA. One of the several contentious elements of the RTS is the methodology which ESMA has suggested NCAs should use to formulate position limits. There is another area where the European Commission has indicated that ESMA needs to rethink its proposal: the RTS needs to show more sensitivity to different types of commodity, in particular to agricultural commodities which, according to the Commission, should attract lower limits for both spot and other months’ limits because of their high volatility.59 In order to support the changes proposed by the Commission, giving particular attention to derivatives with foodstuffs as an underlying, ESMA reviewed the RTS Draft and proposed lowering the position limits for these derivatives by 2.5 per cent (therefore, from 25% to 22.5%).60 Moreover, ESMA suggests that in circumstances where deliverable supply and open interest diverge significantly, the other months’ position limits should be adjusted accordingly. Finally, the definition of contracts traded ‘OTC only’ has been slightly widened in order to prevent circumventions of the position limits regime by trading OTC. [16.82] In addition to this, and in regard to agricultural commodities, position limits have been implemented in France since July 2015, in the context of a wider reform of banking activities. Under the Separation and Regulation of Banking Activities Act61 (which introduced Article 421-16-2 into the Code Monétaire et Financier), anyone involved in agricultural commodity derivatives must comply with the obligation to declare and limit their positions. In this process, the Autorité des Marchés Financiers (hereinafter, ‘AMF’) sets out rules in order to strengthen the regulation of the financial instruments markets when the underlying assets are

agricultural commodities. Under the new provisions, no person is authorized to hold a net position when it exceeds a limit set out by the AMF in a specific instruction in financial instruments (1) which are admitted to trading in a regulated market, or MTF, established in France;62 (2) whose underlying is an agricultural commodity; and (3) whose transactions are cleared through a CCP (Articles 580-1, and 580-2). In order to supervise the effectiveness of these new rules, the regulation imposes the requirement of a daily communication to the AMF on operators (Article 580-2). It must be underlined that MiFID II has introduced position limits inside a more extensive architecture. At the same time, the new position limits regime requires, from industries, robust internal procedures and reporting lines in order to monitor aggregated positions across the globe and net any economically equivalent positions in real time. In addition, MiFID II provisions (Article 69(2)(p)) also include, among supervisory powers granted to the Member States’ competent financial authorities, the general power to ‘limit the ability of any person from entering into a commodity derivative, including the introduction of limits on the size of a position any person can hold at all times’.

VI. A New Set of Intervention Powers for Authorities and Trading Venues [16.83] After the financial crisis, the EU regulator began to consider a spectrum of tools for protecting the integrity and proper functioning of markets. Under this new regulatory strategy, both ESMA and NCAs are charged by the legislator with a range of effective powers (those concerning trading venues can be found at end of this section). Among them, we can consider first the product intervention powers, and second the position limit controls. This policy change marks an important evolution in the EU regulatory strategy and can be considered the most interventionist form of retail market regulation.63 [16.84] It has a very important role in the MiFID Review and can also be effective with respect to the problem of excessive volatility in both spot and

derivative commodities markets. Under these new provisions (above all, Article 42 MiFIR), NCAs have the power to prohibit or restrict ‘the marketing, distribution or sale of certain financial instruments or derivatives that pose a threat to the orderly functioning and integrity of financial markets or commodity markets or to the stability of whole or part of the financial system within at least one Member State’, and those that may have ‘a detrimental effect on the price formation mechanism in the underlying market’.64 If NCAs have not taken an adequate action, or any action at all, or in other situations stipulated by MiFIR that could compromise the integrity of the underlying commodities market, ESMA is empowered (Article 40) to intervene temporarily to prevent, prohibit, or restrict the sale of financial instruments or financial activities, after consulting the competent supervisory and regulatory authorities of the potential spot markets.65 [16.85] The MiFID Review is not the first occasion in which the European legislator has applied the product intervention strategy to commodity derivatives. It had already happened with Directive 2014/91/EU on Undertakings for Collective Investment in Transferable Securities (UCITS V), where product intervention has been useful to shape the portfolio and to define the risk management rules which apply to mutual funds.66 Article 49 and following of the UCITS V state precisely which assets are eligible for UCITS fund managers and, regarding commodity derivatives, the policy is severe because such derivatives are not considered eligible products at all.67 The purpose of this regulatory regime is to protect investors.68 From this viewpoint, commodity derivatives are considered excessively risky and exposed to wide price fluctuations. By prohibiting funds to operate in the commodity derivatives market, however, the regulator introduces a limit on the possible distortive effects of speculation on the regular performance of these markets. [16.86] The new regulatory scheme has also introduced, as said above, new harmonized mandatory powers for NCAs to control trading on commodity derivatives markets, in the form of intrusive power to set and impose position limits with wide discretion powers of intervention. As specifically detailed in RTS 21 (mentioned earlier), NCAs must set limits based on the methodology developed by ESMA. NCAs have the power, in

accordance with this methodology, to review and reset position limits if significant market changes occur (e.g. in delivery supply or in open interest) on the basis of new market dynamics (Article 57(4)). [16.87] Moreover, in such exceptional cases, NCAs can also impose a strengthened regime of position limits, where the action is objectively justified and proportionate, taking into account the liquidity and the orderly functioning of the specific market (Article 57(13)). The authorities can also take position management action, in supervision and enforcement, to request information from any person regarding the size and purpose of a position or exposure, to limit the ability of any person to enter into a commodity derivative, by introducing limits on the size of a position in accordance with Article 57 (Article 69(2)(p)). [16.88] In addition to the extensive powers of national authorities, under the MiFID Review the role of ESMA has grown further. This tendency, developed step by step since the 1999 Financial Services Action Plan,69 is confirmed by both MiFID II and MiFIR. Under MiFIR (Article 45(1)), ESMA may require any person to reduce the size of a position, or eliminate it, and, ‘as last resort’,70 to limit the ability of a person to take position in a commodity derivatives transaction. Any decisions taken by ESMA in relation to these rules have the precedence over any other measures taken by the NCA under MiFID II (Article 69(2)(p)). [16.89] Lastly, the new intervention power allocation under the MiFID/MiFIR regime concerns the trading venues which trade commodity derivatives. These must put position management controls in place in order to monitor the open interest positions of a person, requiring the closing or reduction of a position holder, or providing liquidity to the market—at an agreed price and volume—with the scope of reducing potential negative effects of a dominant position (MiFID II, Article 57(8)–(10)).71 Given the relevant impact of this tool, the position management controls must be transparent and non-discriminatory. In this sense, the trading venues shall specify how they are applying these rules, in relation to the markets’ main characteristics. These controls are a mandatory part of the new control framework and will necessarily interact closely with the ESMA position

limits methodology and the relevant competent authority’s position limits regime.

VII. Conclusion [16.90] This deep analysis allows us to draw some final considerations. First of all, at this stage of the reforming process the new regulatory framework, which was one of the effects of the financial crisis, seems to be a little too complex and, sometimes, confusing. In particular, a lot of doubts have arisen in the EU as well as in the USA about the effective capability of this new set of rules, especially those relating to position limits, to preserve the commodity derivatives and commodity markets from the excessive financialization and the negative impact of speculation. [16.91] In particular, criticism has grown of general regulations in commodity derivatives which do not distinguish enough between specific kinds of commodities. The US regulation on commodity derivatives before Dodd–Frank, and even the recent choice made by the French legislator, demonstrate that a specific regulation could be useful in order to address some of the issues concerning the proposed level of position limits. To be more precise, if the objective of regulation is to curb speculation, the regulator should weigh the possibility of introducing reasonable volume limits based on some form of correlation between the notional amount of traded derivatives and the size of physical production. Moreover, with the same aim, more severe position limits related to particular kinds of commodities (e.g. agricultural) might be adopted. [16.92] In addition, the new set of rules also highlights that, given the highly technical aspects of commodity derivatives regulation, the EU legislator has conferred a wide range of powers to ESMA and to the NCAs with regard to the regulatory issue. This strategy, though, poses an important question related to the nature of the rules (RTS/ITS) issued especially by ESMA which, actually, is not only technical but has also a political meaning with all the well-known consequences for accountability. This is confirmed by what has recently happened between the European Commission and ESMA with reference to some RTS proposals and in

particular the ones relating to commodity derivatives, which the Commission rejected asking for tighter regulation in the field of ‘food’ derivatives.72 The RTS Final Draft on Position Limits, as well as on ancillary activities, has witnessed once more the need for a regulation that considers particular aspects related to specific categories of financial instruments, such as the concerns about speculation and possible impacts on food prices for agricultural commodity derivatives. [16.93] Finally, given the deep and rapid changes which have occurred in commodity market conditions over the past few years, a further question must be considered. What must be studied and verified is the current relevance and effectiveness of a regulatory framework—not even applied in Europe up to now—which was designed for a financial and economic scenario completely different from the one of today. As it is known, speculation was blamed for a run-up in food and energy prices that caused many political and social issues in several developing countries, posing a relevant ethical problem on the financialization of these markets. The recent slump in commodity prices is now producing some more problems for developing countries in terms of lower resources for both public and private investments. It will be interesting to discover if the set of rules contained in all the regulations discussed in this chapter will manage to solve the newly arising problems, or whether it is another telling example of ‘bubble laws’.73

*

Although this chapter is the result of joint work, Sections I, II, III, and IV should be attributed to Antonella Sciarrone Alibrandi, and Sections V, VI, and VII should be attributed to Edoardo Grossule. 1 J. Black, What is a Regulatory Innovation? in J. Black, M. Lodge, and M. Thatcher (eds) Regulatory Innovation: A Comparative Analysis (Elgar, 2005), pp. 1–15. 2 Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012 on OTC derivatives, central counterparties and trade repositories (known as ‘EMIR’: European Market Infrastructure Regulation). 3 ESMA, Final Report on the Draft Regulatory and Implementing Technical Standards MiFID II/MiFIR, 28 September 2015 (ESMA/2015/1464). In order to comply with the request made by the European Commission in March 2016, ESMA issued amended draft versions of the RTS both on position limits regime (RTS 21, amended version issued on 2 May 2016) and ancillary activities (RTS 20, amended version issued on 30 May). On this

discussion, see ESMA letters on position limits (ESMA/2016/402) and ancillary activities (ESMA/2016/403), at . See also, P. Stefford, ‘EU rejects MiFID II trading reforms’, at , 17 March 2016. 4 Regulation EU No 1227/2011 of the European Parliament and of the Council of 25 October 2011 on wholesale energy market integrity and transparency (known as ‘REMIT’). 5 G20, Leaders’ Statement, Pittsburgh Summit, in A Framework for Strong Sustainable and Balanced Growth, 2009, para. 12; see also, IOSCO, Principles for Regulation and Supervision of Commodities Derivatives Market, Final Report, 2011, 26–38. 6 C. Culp, ‘The Social Functions of Financial Derivatives’ in R. Kolb and J. Overdahl (eds) Financial Derivatives: Pricing and Risk Management (Hoboken, NJ: Wiley & Sons, 2010), p. 58. The need to hedge against unfavourable changes in prices has always been one of the primary needs of operators. Because of a lack of both transparency and predictability of prices in the commodity market, the European Commission emphasized ‘the social’ function of derivatives: cf. European Commission, A better functioning food supply chain in Europe, 8 October 2009 (COM(2009) 591), 8. On this point, for a general perspective, R. Kolb and J. A. Overdahl, Futures, Options and Swaps (5th edn, Malden, MA: Blackwell Publishing, 2007), pp. 3 et seq. 7 Kolb and Overdahl (n. 6), 3 and following. 8 M. Falkowski, ‘Financialization of Commodities’ (2011) Contemporary Economics IV, 4 and following. 9 European Commission (n. 6), 2 and following. The Commission underlined that ‘from mid-2007 to mid-2008, agricultural commodity prices rose sharply, which resulted in increased consumer food prices and higher inflation levels overall. […] Consumer food prices continued to increase and only started declining in May 2009 raising concerns about the functioning of the food supply chain.’ Also, for a specific analysis on price volatility, see European Commission, Food Prices in Europe (COM(2008) 821), 2 and following. 10 IOSCO (n. 5). In this sense, it could be useful to read in conjunction the measures of both MiFIR and EMIR, which operate decisive actions, especially with regard to the OTC derivatives market, which has always been characterized by opacity of trading and speculation. G. Ferrarini and P. Saguato, ‘Regulating Financial Market Infrastructure’ in N. Moloney, E. Ferran, and J. Payne (eds) The Oxford Handbook of Financial Regulation (Oxford: Oxford University Press, 2015), pp. 568, 580–2. As outlined by the authors, in fact: ‘[the trading infrastructure] contributes to the efficiency of financial market transactions and to the liquidity of markets, by executing orders in an orderly manner and providing pre-/post-trade information.’ 11 G20 (n. 5). 12 On this topic, K. Alexander and V. Maly, ‘The New EU Market Abuse Regime and the Derivatives Markets’ (2015) Law and Financial Market Review 9, 243–50. On MAD I, in general, G. Ferrarini, ‘The European Market Abuse Directive’ (2004) Common Market Law Review 41, 716–25. 13 See I. H.-Y. Chiu, ‘Financial Benchmarks: Proposing a Governance Framework Based on Stakeholders and the Public Interest’ (2015) Law and Financial Markets Review 9,

223–42. 14 P. Lucantoni, ‘Central Counterparties and Trade Repositories in Post-Trading Infrastructure Under EMIR Regulation on OTC Derivatives’ (2014) Journal of International Banking Law and Regulation, 29, 681 and following. E. Avgouleas, ‘Regulating Financial Innovation’ in Moloney, Ferran, and Payne (eds) (n. 10), 679. 15 Under Article 10, para. 3, EMIR, a derivative contract is for hedging purposes when it is ‘objectively measurable as reducing risks directly relating to the commercial activity or treasury financing activity of the non-financial counterparty or of that group’. 16 Cf. Articles 10 and 11 Commission Delegated Regulation (EU) No 149/2013, 19 December 2012 (EMIR Regime). In particular, for commodity derivatives the notional fixed threshold is €3 billion. 17 It is worth noting that, in cases where the physical agricultural markets are seriously affected (see Article 42(2)(f)), national competent financial authorities have to properly consult with public bodies competent for the physical agricultural markets (under Council Regulation (EC) No 1234/2007 of 22 October 2007 establishing a common organization of agricultural markets and on specific provisions for certain agricultural products (‘Single CMO Regulation’)). 18 Directive 2013/36/EU (CRD IV) of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC; Regulation (EU) No 575/2013 (CRR) of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012. 19 In this context commodity dealers are defined as ‘investment firms whose main business consists exclusively of the provision of investment services or activities in relation to the financial instruments set out in points 5, 6, 7, 9 and 10 of Section C of Annex I to Directive 2004/39/EC and to which Council Directive 93/22/EEC of 10 May 1993 on investment services in the securities field did not apply on 31 December 2006’. 20 For a deeper analysis on UCITS regime, in general, N. Moloney, EU Securities and Financial Markets Regulation (Oxford: Oxford University Press, 2014), p. 194 and following; C. P. Buttigieg, ‘The 2009 UCITS IV Directive: A Critical Examination of the Framework for the Creation of a Broader and More Efficient Internal Market for UCITS’ (2012) Business Law Journal 6, 453–78;; N. Moloney, How to Protect Investors. A Lesson from the EC and the UK (Cambridge: Cambridge University Press, 2010). 21 For a critical introduction to AIMFD framework, N. Moloney, EU Securities (n. 20), 269–315. 22 In particular see ESMA RTS 20 on this topic (see also n. 3). 23 It is worth noting that in the MiFID context ‘commodity’ means a commodity as defined in point (1) of Article 2 of Commission Regulation (EC) No 1287/2006 (MiFID implementing regulation). This regulation defines as a commodity ‘any goods of a fungible nature that are capable of being delivered, including metals and their ores and alloys, agricultural products, and energy such as electricity’.

24

ESMA, Guidelines on the application of the definitions in Sections C6 and C7 of Annex 1 of MiFID, 6 May 2015 (ESMA/2015/675). 25 MiFIR gives a definition of commodity derivatives in point (30) of Article 2(1). According to this definition, ‘commodity derivatives’ means ‘those financial instruments defined in point (44)(c) of Article 4(1) of Directive 2014/65/EU; which relate to a commodity or an underlying referred to in Section C(10) of Annex I to Directive 2014/65/EU; or in points (5), (6), (7) and (10) of Section C of Annex I thereto’. 26 These wholesale energy products are defined in Article 2, para. 4 of REMIT. 27 European Commission, Market in Financial Instruments Directive (MiFID II): Frequently Asked Questions, April 2014, 10 (MEMO/14/305). 28 A. Johnston and G. Block, EU Energy Law (Oxford: Oxford University Press, 2013); E. Blunsdon, ‘REMIT for Double Jeopardy?’ (2013) Utility Week, 2 August. 29 ESMA, Consultation Paper (MiFID II–MiFIR), 22 May 2014, 278–9 (ESMA/2014/549). In particular, the exemption for commodity derivatives on coal and oil is: (1) temporary, because it operates for 3.5 years after the MiFID II Directive was applied; (2) conditional, because the exemption can only be granted by the relevant competent authority; and (3) of limited extent, because it refers only to exhaustively listed EMIR provisions. In contrast to REMIT carve-out, it still leaves C6 energy derivatives contracts categorized in the scope of financial instruments, which entails being subject to other MiFID II provisions, position limits included. 30 E. Avgouleas, ‘Regulating Financial Innovation’ in Moloney, Ferran, and Payne (eds) (n. 10), 679–81. 31 An example in this sense is given by FCA, The Perimeter Guidance Manual, PERG 13.5 Exemptions from MiFID, released 5 April 2016, available at . The English Authority, in fact, clarified that: where, for example, a commodity producer buys or sells commodity derivatives for the purposes of limiting an identifiable risk of its main business, for instance in circumstances where the risk management exclusion in Article 19 of the Regulated Activities Order would apply, in our view this would qualify as ancillary for the purposes of this exemption. On the other hand, where a commodity producer deals on own account for speculative purposes, it is unlikely that this would be ancillary to the main business in the case of Article 2.1(i) MiFID. 32

ESMA, Discussion Paper MiFID II/MiFIR, 22 May 2014 (ESMA/2014/548), 386 and following, where ESMA observed that the exemptions currently available ‘are effectively carried over under Article 2(1)(j) of MiFID II in similar but not identical terms’. 33 Cf. Recital 20, MiFID II, which also specifies that ‘those criteria should at least take into consideration the need for ancillary activities to constitute a minority of activities at group level and the size of their trading activity compared to the overall market trading activity in that asset class’.

34

In the RTS 20 (also in the amended version issued on 30 May 2016) of the Annex I of the RTS Draft (399 and following), ESMA tests are designed as follows: (1) the first test (the ‘trading activity thresholds’) expresses the relationship between the ‘speculative’ activity and the overall EU market activity in each class of commodity derivatives. The reasons are that firms having a significant share of the market in a particular class of derivatives will not be allowed to benefit from the exemption as they should compete with other market participants on a level playing field (Article 2); (2) the second test will determine whether the investment activity is large in size, relative to what the entity does as its main business (Article 3, the ‘main business thresholds’). 35 The thresholds are: 4 per cent derivatives on metals; 3 per cent derivatives on oil and oil products; 10 per cent derivatives on coal; 3 per cent derivatives on gas; 6 per cent derivatives on power; 4 per cent derivatives on agricultural products; 15 per cent in relation to derivatives on other commodities, including freight and commodities referred to in Section C 10 of Annex I to Directive 2014/65/EU; 20 per cent in relation to emission allowances or derivatives thereof. 
 36 Cf. Recital 5 RTS 20 of the Annex I of the RTS Draft. 37 The taxonomy of derivatives contracts based on their economic purpose is well described by T. E. Lynch, ‘Gambling by Another Name: The Challenge of purely Speculative Derivatives’ (2012) Stanford Journal of Law, Business & Finance 17, 75 and following. 38 In particular, see ESMA RTS 21 on this topic (cf. also n. 3). 39 An OTC contract economically equivalent to an exchange-traded derivative (ETD) is considered a spot month contract when the commodity derivative traded on a trading venue to which it is equivalent is the spot month. 40 On this topic, see also RTS Draft, 352 and following. In particular, in the absence of a definition of ‘non-financial entity’, ESMA proposed that a non-financial entity should be considered as any entity which is not a financial institution under MiFID II or other relevant EU legislation and that the definition of a financial entity (in other words, the inverse of a non-financial entity) should include entities that are outside the EU that would be a financial entity (or a non-financial entity) under the various Directives and Regulations if their activities were performed in the EU. 41 Specifically, EMIR, Article 1(1) foresees that ‘This Regulation lays down clearing and bilateral risk-management requirements for over-the-counter derivative contracts, reporting requirements for derivative contracts and uniform requirements for the performance of activities of central counterparties (“CCPs”) and trade repositories.’ EMIR Article 2(8) foresees that: ‘financial counterparty’ means an investment firm authorised in accordance with Directive 2004/39/EC, a credit institution authorised in accordance with Directive 2006/48/EC, an insurance undertaking authorised in accordance with Directive 73/239/EEC, an assurance undertaking authorised in accordance with Directive 2002/83/EC, a reinsurance undertaking authorised in accordance with Directive

2005/68/EC, a UCITS and, where relevant, its management company, authorised in accordance with Directive 2009/65/EC, an institution for occupational retirement provision within the meaning of Article 6(a) of Directive 2003/41/EC and an alternative investment fund managed by AIFMs authorised or registered in accordance with Directive 2011/61/EU. 42

Regarding the NFC (without systemic importance) commitment, unlike FCs, they have solely to: (1) report at trade repository about derivative contracts; and (2) adopt some risk-mitigation techniques (lower than those of FCs), through the communication of their viability into derivative contracts. 43 Cf. Articles 10 and 11 Commission Delegated Regulation (EU) No 149/2013, 19 December 2012 (EMIR Regime). In particular, for commodity derivatives the notional fixed threshold is €3 billion. 44 ESRB, Advice of the European Systemic Risk Board of 31 July 2012, (ESRB/2012/2). 45 On this point see ESMA, Final Report (n. 3). 46 ESMA under the mandate of MiFID II (Article 57(12)) has developed RTS 21 to determine, inter alia, the criteria and methods to understand whether a transaction can be qualified as reducing risks directly relating to commercial activities. As said above, this RTS Draft is now under review in order to comply with the request made by the European Commission in March 2016. 47 These criteria, contained in Article 10 Commission Delegated Regulation (EU) No 149/2013, 19 December 2012 (EMIR Regime), include the derivatives contract capacity: (1) to cover the potential change in the value of objects of the non-financial counterparties’ main business; (2) to hedge the risks arising from fluctuation of interest rates, inflation rates, foreign exchange rates, or credit risk. Criterion (3) used by this Regulation refers to a qualification as a hedging contract pursuant to International Financial Reporting Standards (IFRS) adopted in accordance with Article 3 of Regulation (EC) No 1606/2002 of the European Parliament and of the Council. 48 EMIR refers to ‘treasury financing activity’ which was not included in MiFID II position limits framework. This discrepancy can be explained by the fact that commodity derivatives are not commonly used for the purpose of treasury financing. 49 In this sense, for the purposes of both MiFID II and MiFIR, Recital 21 MiFID II recommends that activities are deemed to be objectively measurable as reducing risks directly relating to the commercial activity or treasury financing activity and intra-group transactions should be considered in a consistent way with EMIR. 50 Article 57(12)(f) MiFID II entrusted ESMA in this sense. 51 ESMA (n. 32), 414. 52 In the context of RTS 21, ‘spot month’ means the commodity derivatives contract in relation to a particular underlying commodity whose maturity is the next to expire in accordance with the rules set by the trading venue. In the Recital 11 of RTS 21, ESMA clarifies what should be understood by ‘spot month’ relating to position limits regime. In that regard, a broad approach has been taken so that the spot month period does not need to

correspond to a time period which is ‘a month’ but rather it is specific to each commodity derivatives contract and is the contract next to expire, as determined by the rules of the relevant trading venue (so it could be three days, two weeks, three months, etc.). 53 By adopting this asymmetric spread, ESMA has diminished the size of position limits in comparison with the December 2014 RTS Draft methodology. 54 The exception is the spot month position limits for cash settled commodity derivatives with no deliverable supply (commodity derivatives listed under Annex I, Section C10, e.g. weather) which are based on a percentage of total open interest in the spot month. 55 More precisely, Core Principle 5 of Section 5(d) of the Commodity Exchange Act (CEA) requires designated contract markets to adopt speculative position limits or position accountability for speculators, where necessary and appropriate, to reduce the potential threat of market manipulation or congestion, especially during trading in the delivery month. 56 M. Greenberger, Will the CFTC Defy Congress’s Mandate to Stop Excessive Speculation in Commodity Markets and Aid and Abet Hyperinflation in World Food and Energy Prices?—Analysis of the CFTC’s Proposed Rules on Speculative Position Limits, University of Maryland, Legal studies research no 2011—20, at ; see also A. Notini, ‘The Validity of CFTC Position-Limit Rulemaking Under Dodd-Frank’ (2013) Suffolk Univ. Law Rev. 46, 192–3. 57 In this sense, ‘Report of the Finance & Transactions Committee’ (2011) Energy Law Journal 32, 331 and following. 58 These include grain and livestock futures, energy, and precious metals, with separate limits set for spot month (generally, 25% of deliverable supply), and non-spot month (generally, 10% of open interest in the first 25,000 contracts and 2.5% thereafter). 59 The challenge NCAs face in calculating and adjusting position limits has caused concern in the market. The NCAs do not have access to the market data that is required to adjust the baseline position limit and determine deliverable supply taking into account the particulars of each individual contract. They are regulatory bodies, not market participants, and are being asked to analyse market conditions for each individual commodity contract and set position limits; this involves economic rather than regulatory decisions. The FCA has expressed concern over its ability to set individual position limits for between 1,500 and 1,900 commodity contracts. 60 See Recital 12, ESMA, Opinion (Annex), Amended draft Regulatory Technical Standards on the methodology for the calculation and the application of position limits for commodity derivatives traded on trading venues and economically equivalent OTC contracts, May 2016 (ESMA/2016/668). In particular, this Opinion takes into consideration the European Commission letter to ESMA requesting amendments to RTS 21, published on 20 April 2016, available at . 61 L. 2013-672 of 26 July 2013, at . 62 Or in a non-French market, if such financial instruments imply a physical delivery in French territory (Article 580-2).

63

In this sense, N. Moloney, ‘The Legacy Effects of the Financial Crisis on Regulatory Design in the EU’ in E. Ferran, N. Moloney, J. G. Hill, and J. C. Coffee Jr. (eds) The Regulatory Aftermath of the Global Financial Crisis (Cambridge: Cambridge University Press, 2012), pp. 186 and following. 64 In more detail, the product intervention powers can be used only when identified risks cannot be managed through the exercise of ordinary supervision or enforcement powers due to existing rules. Moreover, the actions of the regulators must be proportionate to the identified risks, the level of sophistication of investors or market participants concerned, and the likely effects on investors and operators. In this sense, before putting any actions in place, it is necessary to consult the supervisory authorities of other Member States, in particular whether these interventions could have effects in other jurisdictions (cf. Article 42, para. 2(c)(d), MiFIR). 65 Authorities can be identified under Regulation No 1234/2007/EC. 66 For a more detailed analysis of this topic, see Moloney (n. 20). 67 This regime was laid down since the first UCITS version. Cf. CESR, CESR guidelines concerning eligible assets on investment by UCITS, CESR/07-044. Funds may only invest in structured instruments whose performance is linked to the market indices of commodities or commodity derivatives. This means that commodity index ETFs must be synthetic in the EU. More precisely, the provision comes from UCITS V Article 50(1), which lists the eligible assets, and Article 50(2), which specifies the exemptions. Commodity derivatives do not fall under these two provisions. In fact, the trading operations on commodity derivatives cannot occur even indirectly through shares of ETFs authorized for trading in commodity derivatives. 68 N. Moloney, ‘Regulating the Retail Markets’ in Moloney, Ferran, and Payne (eds) (n. 10), 761. 69 This aspect was specified by N. Moloney, ‘Law Making Risks in EC Financial Market Regulation After the Financial Services Action Plan’ in S. Weatherill (ed.) Better Regulation (Oxford: Hart Publishing, 2007), pp. 367 and following. 70 The exercise of these powers by ESMA is in fact strictly conditioned: (1) first, the ESMA measures have to address issues relating to the orderly functioning and integrity of commodity derivatives markets (or more generally to the financial market as a whole); then (2) EMSA should intervene only when NCAs have not made any or sufficient decisions. 71 Cf. ESMA (n. 32), 406. Even if ESMA is not mandated under MiFID II to develop technical standards in respect of this requirement and, therefore, position management controls applied by trading venues are not discussed further in the Discussion Paper, ESMA’s view is that this regime will operate in tandem with position limits set by NCAs. See also, Moloney (n. 20), 536. 72 In general, the relations between ESMA and the European Commission are to be considered stable. The Commission has normally tended to accept ESMA’s technical advice; a tendency justified by the extensive procedure of consultation to which ESMA’s projects are subject. Probably, the political delicacy of any matters concerning ‘food’ derivatives amplifies the natural conflict between policy goals and technical solutions. In

this perspective, the issues raised on Short Selling Regulation in relation to CDS on sovereign debt securities could be read. On this topic, see Moloney (n. 20), 906. 73 L. E. Ribstein, ‘Bubble Laws’ (2003) Houston Law Review 40, 77 and following.

17 Algorithmic Trading and High-Frequency Trading (HFT) Pierre-Henri Conac

I. Introduction II. Algorithmic Trading 1. Investment Firms 2. Trading Venues III. High-Frequency Trading 1. The Specific Regulation of High-Frequency Trading 2. The Specific Regulation to Prevent Market Abuses IV. Conclusion

I. Introduction [17.01] Algorithmic trading and high-frequency trading (HFT), which is a type of algorithmic trading, have developed to become major trading techniques in securities markets. However, these developments raise two serious regulatory issues: the first of market stability, for both of them; and the second of market abuse for high-frequency trading. [17.02] Therefore, algorithmic trading, mostly in its HFT form, has become subject to increased political and regulatory scrutiny in the

European Union. This led the European Securities and Markets Authority (ESMA) to adopt Guidelines dealing with the subject as early as 2011, and to the EU legislator also taking action on the topic. Both algorithmic trading and high-frequency trading are subject to specific rules in the new Directive of 2014 on the Markets in Financial Instruments Directive (MiFID II).1 [17.03] Algorithmic trading can be defined as a computer-driven process in which an algorithm replicates decisions previously made by traders and investors in relation to the composition, timing, format, and destination of orders. It is a piece of software which makes decisions without human intervention. Article 4(1)(39) of MiFID II provides a definition: ‘algorithmic trading’ means trading in financial instruments where a computer algorithm automatically determines individual parameters of orders such as whether to initiate the order, the timing, price or quantity of the order or how to manage the order after its submission, with limited or no human intervention.2

[17.04] Algorithmic trading has been developed since the 1980s as a consequence of the progress of computing power. Therefore, it is not a new development although its scale has certainly increased since the 1980s. [17.05] High-frequency trading is a subcategory of algorithmic trading. It refers to automated trading conducted at millisecond or microsecond speeds throughout the trading day. Article 4(1)(40) of MiFID II defines the ‘highfrequency algorithmic trading technique’ as an algorithmic trading technique characterised by: (a) infrastructure intended to minimise network and other types of latencies, including at least one of the following facilities for algorithmic order entry: co-location, proximity hosting or high-speed direct electronic access; (b) system-determination of order initiation, generation, routing or execution without human intervention for individual trades or orders; and (c) high message intraday rates which constitute orders, quotes or cancellations.

[17.06] The characteristics of HFT are speed and a reduced latency (thousands of trades in milliseconds or 1/1000 of a second); a location of the computer as close as possible to the computer of the market (also known as co-location) in order to benefit from higher speed; proprietary trading; a limited capital commitment since HFT traders tend to close all their positions at the end of the day and to commit limited capital; and a range of

long and short strategies with no regard to the direction of the markets. HFT firms usually trade in the direction of price changes and therefore follow the trend. HFT traders have no real interest in the underlying securities they trade in. [17.07] HFT is more recent than algorithmic trading and is a product of technological and regulatory developments. HFT activity benefited from increased computing power and improved telecommunications infrastructure, as well as falling processing costs. These have reduced delays in transmission of orders, also called latency, to the markets. From the regulatory side, HFT benefited from the decimalization of pricing in the United States in 2001, decided by the Securities and Exchange Commission (SEC). In addition, HFT has been made more attractive by the fragmentation of markets both in the US, through the development of alternative trading systems (ATS), and in the EU because of the end of the concentration rule on the regulated markets by the Directive 2004/39/EC of 2004 on markets in financial instruments (MiFID I) and the development of Multilateral Trading Facilities (MTFs). [17.08] HFT is a form of proprietary trading. Some firms have specialized on HFT, such as Getco and Tradebot in the US. Hedge funds such as Citadel or Renaissance Technologies are also active in HFT. Finally, there are also HFT trading desks of investment banks, for example Goldman Sachs or Citigroup. There are also specialized HFT firms in Europe, such as Optiver and others in Amsterdam. The Netherlands has developed itself into a location for several European HFT firms because of the historical existence of a derivatives market there. [17.09] The share of HFT in the volume of trades in stocks in Europe and in the US is significant. According to a 2014 ESMA report, in a study covering a sample of 100 stocks from nine EU countries, HFT (including by investment banks) accounted in May 2013 for 43 per cent for the lifetime of orders approach.3 Although the regulatory focus is mostly on stock exchanges because of the potential impact on retail investors, algorithmic trading and HFT do not exist only on those markets. These trading strategies can be used on any type of market such as the bond market. HFT is also used on foreign exchange markets.4

[17.10] Algorithmic trading and HFT have been subject to a public debate, especially in the US but also in Europe, about their supposed benefits and risks. As to algorithmic trading, it provides certain benefits. It removes the human emotion from trading—they are robot investors mimicking human reactions—and allows for immediate speed of reaction. However, there are also some risks. Algorithmic trading desks may face a significant amount of intraday risk if they do not have transparency and robust controls in place. Without such adequate controls, losses can accumulate and spread rapidly. Internal controls may not have kept pace with speed and market complexity. This type of trading also amplifies markets swings since all algorithms tend to take the same direction. Therefore, it can create more volatility and even amplify systemic risk. [17.11] The risks of detrimental effects from algorithmic trading on market stability were underlined by the 6 May 2010 ‘Flash Crash’ in the US. In thirty minutes, the Dow Jones Industrial Average dropped about 9 per cent with some stocks being listed for a few cents for a few minutes. A 2010 SEC study into the Flash Crash showed that control of algorithms could be lost and that there could be detrimental interactions between algorithms (notably at pre-opening or pre-closing).5 Although, the SEC study showed that HFT played a disrupting part, it also appears that it did not trigger the Flash Crash. However, the Flash Crash is not the only instance where algorithmic trading has lead to significant problems in one firm or general market instability. The October 1987 stock market crash was partly caused by dynamic portfolio insurance, which is a form of algorithmic trading. Dynamic portfolio insurance operates so that if the market falls beyond a certain point, sell orders are triggered automatically. Again, more recently, in August 2012 Knight Capital lost $440 million because of a computer glitch. Due to the severity of the loss compared to its capital, it had to be sold to another firm in December 2012. [17.12] As to high-frequency trading, it also has benefits. HFT firms argue that it contributes to market liquidity, mostly via market-making strategies, and reduces bid-ask spreads which benefits investors. However, detractors argue that both benefits are marginal since they only hold positions for milliseconds; the liquidity they provide is only ‘transitory’. Additionally in terms of benefits, high-frequency traders also contribute to

market efficiency via arbitrage strategies by providing some balance to prices across trading venues. [17.13] However, there are several risks associated with HFT. The first is the instability and unreliability of the order book because orders are entered on and removed at huge speeds. The instability of the order book leads to uncertainty in trading for the other market participants. The second risk is that it could push investors away to dark markets where HFT traders will, in principle, not have the same access to the order book. There is no academic consensus on HFT as regards its impact on market efficiency. Articles are not conclusive and support both views but it is clear that the contribution of HFT to capital formation is close to zero. Finally, it is argued that HFT could constitute market manipulation in some cases. [17.14] There has been a lively debate in the US on this precise issue of whether HFT is a form of market manipulation. Michael Lewis argued in Flash Boys, published in 2014, that HFT raises issues of front-running other investors.6 Charlie Munger, Warren Buffet’s business partner, publicly branded HFT as ‘legalized front-running’ and ‘basically evil’. It actually seems strange that HFT firms are also highly profitable7 and that some, like Virtu according to its 2014 IPO prospectus, seem never to lose money on almost any trading day. However, the SEC does not consider HFT a form of market abuse as such. Rather, it views it as a market structure issue linked to market stability. Because of this lack of action by the SEC, some investors led by Brad Katsuyama, a former trader and critic of HFT, have developed an alternative market called Investors Exchange (IEX).8 In IEX, the speed of orders will be reduced, using a ‘speed bump’ that slows orders by 350 millionths of a second, in order to remove the speed advantage of HFT traders. IEX has received SEC approval to become a registered national securities exchange in June 2016 despite intense lobbying by users of HFT strategies, such as Citadel, and exchanges. [17.15] There has also been a debate in the EU about the benefits and risks of algorithmic trading and, also, especially of HFT. However, unlike in the US, the debate has not become mainstream but has remained among regulators and the finance industry. European securities regulators have converging views on algorithmic trading but differ on HFT.

[17.16] There is a consensus among securities regulators in the EU on the fact that algorithmic trading can contribute to systemic risk due to speed and automation. There is also a consensus on possible remedies: licensing, controls at various levels, circuit-breakers … [17.17] However, there is strong disagreement among national regulators on HFT. Some European regulators have a positive attitude towards it. This is the case in, for example, the Netherlands, where several of the most active European high-frequency trading firms are based.9 Other supervisors are critical and tend to consider that HFT is a form of market abuse. This is the case, for example, in France and Germany. The position of the French AMF (Autorité des Marchés Financiers) is that efficiency gains are marginal while the risks of market abuse are high. As a consequence, France introduced restrictive legislation in 2012. Similarly, Germany introduced legislation on HFT in 2012 which requires HFT traders to hold a licence, includes rules applying to them as well as to trading venues, and qualifies some algorithmic trading practices as insider trading.10 [17.18] Because of these different views among the Board of Supervisors of ESMA, the Guidelines of 2012 on organizational requirements for investment firms and trading venues are rather limited.11 They include mostly organizational requirements and some provisions designed to address the risk of market abuse. The Securities and Markets Stakeholder Group (SMSG) of ESMA took in its advice a critical view of HFT, going so far as to suggest that minimum tick sizes could be set and implemented consistently on all European trading platforms (including OTC) under the authority of ESMA.12 [17.19] There has also been a debate at the international level. As a consequence of the May 2010 ‘Flash Crash’, the G20 raised the issue of the risk associated with algorithmic trading and this prompted regulators to act. The November 2010 G20 Seoul summit called on the International Organization of Securities Commissions (IOSCO) to develop by June 2011, and report to the FSB, recommendations to promote market ‘integrity and efficiency to mitigate the risks posed to the financial system by the latest technological developments’.13 IOSCO published in 2011 a Report on ‘Regulatory Issues Raised by the Impact of Technological Changes on

Market Integrity and Efficiency’.14 The report included five general, as it is usual with IOSCO, recommendations designed to reduce market instability. Those recommendations included, for instance, requirements to have in place suitable trading control mechanisms (such as trading halts, volatility interruptions, limit-up-limit-down controls, etc.) to deal with volatile market conditions, and to submit all order flow to appropriate controls, including automated pre-trade controls. One recommendation addressed the risk of market abuse that may arise as a result of technological developments. [17.20] MiFID II includes provisions on algorithmic trading (Part II) and also on HFT (Part III). Those provisions can be found in Articles 17 (algorithmic trading), 48 (systems resilience, circuit breakers, and electronic trading), 49 (tick sizes), and 50 (synchronization of business clocks) of MiFID II. The ESMA has had a strong influence on the content of the proposal of MiFID II. Therefore, many of those provisions reflect the 2012 ESMA Guidelines on systems and controls in an automated trading environment.15 Due to the highly technical nature of algorithmic trading and HFT, there were few changes between the proposition and the final version. The provisions of MiFID II include requirements on licensing of HFT traders and organization duties of investment firms and trading venues, but also reporting, documentation, and standards (e.g. algorithmic flagging, or ‘algo flagging’), and requirements on market microstructure (e.g. tick size) and liquidity provision (to provide market-making). However, there are no provisions on latency and fee structure, which are critical parameters, and no provisions on how to organize cross-market surveillance. [17.21] In September 2015, ESMA published the Draft Regulatory Technical Standards (RTS) and Technical Advice for Delegated Acts for Level 2 measures.16 Those Level 2 measures—especially draft RTS 6 (Regulation on the regulatory technical standards specifying the organizational requirements of investment firms engaged in algorithmic trading, providing direct electronic access and acting as general clearing members), 7 (Draft regulatory technical standards on organizational requirements of regulated markets, multilateral trading facilities and organized trading facilities enabling or allowing algorithmic trading through

their systems), 8 (Draft regulatory technical standards on market making agreements and market making schemes), 9 (Draft regulatory technical standards on the unexecuted orders to transactions), 10 (Draft regulatory technical standards on requirements to ensure co-location and fee structures are fair and non-discriminatory), 11 (Draft regulatory technical standards on the tick size regime for shares, depositary receipts and exchange traded funds), and 12 (Draft regulatory technical standards on the determination of a material market in terms of liquidity relating to trading halt notifications) —deal with requirements under Articles 17, 48, 49, and 50 of MiFID II. They cover firms that engage in algorithmic trading or HFT techniques, that provide direct electronic access (DEA), DEA users, general clearing members, and trading venues (RMs, MTFs, and OTFs). A Commission Delegated Regulation as regards organizational requirements and operating conditions for investment firms was published in April 2016 but is limited to defining algorithmic trading, HFT, and direct electronic access.17

II. Algorithmic Trading [17.22] The regulation of algorithmic trading is done through requirements applicable to (1) investment firms, since they are users of this technique, and to (2) trading venues.

1. Investment Firms MiFID II imposes requirements on (A) all investments firms that engage in algorithmic trading, and also special requirements on (B) those who provide special services.

A. Requirements for all Investment Firms [17.23] ESMA 2012 Guidelines adopted under MiFID I had provided that investment firms should manage the risks relating to those trading activities (Guideline 4: Organizational requirements for investment firms to promote

fair and orderly trading in an automated trading environment).18 MiFID II provides more precise rules. [17.24] MiFID II requires that an investment firm has ‘effective systems and risk controls to ensure that its trading systems are resilient and have sufficient capacity, are subject to appropriate trading thresholds and limits and prevent the sending of erroneous orders or the systems otherwise functioning in a way that may create or contribute to a disorderly market’.19 Article 15 of the ESMA draft RTS 6 sets out four mandatory pre-trade controls applicable for trading in all financial instruments: price collars, maximum order value, maximum order volume, and maximum messages limit. These systems should, according to Article 12 of the RTS 6, include a ‘kill function’ which allows investment firms to switch off access to the market in case of disruption. Investment firms should also stress test their systems to ensure that they are capable of withstanding increased order flows or market stresses (Article 10 of RTS 6), and also carry out penetration tests and vulnerability scans of systems (Article 18 of RTS 6). These required testing should take place before initial deployment and after substantial changes to the algorithm. They should have business continuity arrangements and pre-trade and post-trade controls of order flow. The requirement to have in place robust systems to ensure continuity and regularity in the performance of the investment firm are close to ESMA 2012 Guideline 2.20 [17.25] Article 5 of the ESMA draft RTS 6, in order to make sure that these requirements are applied, requires the senior management of the investment firm to designate a ‘responsible party’ to sign off the initial deployment or substantial update of an algorithmic trading system or strategy or trading algorithm. This responsibility cannot be outsourced. [17.26] Investment firms that engage in algorithmic trading should also notify the competent authorities of their home Member State and of the trading venue.21 The competent authority of the home Member State of the investment firm may require information on the algorithmic trading strategies, key compliance, and risk controls.

[17.27] In order to provide some market stability, investment firms that engage in algorithmic trading in order to pursue a market-making strategy are subject to certain requirements.22 The definition of pursuing a marketmaking strategy is provided in the Directive: when, as a member or participant of one or more trading venues, its strategy, when dealing on own account, involves posting firm, simultaneous two-way quotes of comparable size and at competitive prices relating to one or more financial instruments on a single trading venue or across different trading venues, with the result of providing liquidity on a regular and frequent basis to the overall market.

The definition is specified by the RTS in order to trigger the requirement that investment firms enter into a binding written agreement with the trading venue.23 Article 1 of the ESMA 2015 draft RTS 8 holds that investment firms shall be obliged to enter into a market making agreement where in execution of one market making strategy they post firm, simultaneous two-way quotes of comparable size and competitive prices when dealing on their own account in at least one financial instrument on one trading venue for at least 50% of the daily trading hours of continuous trading at the respective trading venue, excluding opening and closing auctions, for half of the trading days over a one month period (regular and frequent basis).

For the purposes of this paragraph (a) a firm quote shall include any orders and quotes that can be matched against an opposite order or quote under the rules of a trading venue; (b) quotes shall be deemed simultaneous two-way quotes if they are posted in a way that both sides are present in the order book at the same time; (c) two quotes shall be deemed of comparable size when their sizes do not diverge by more than 50% from each other; (d) quotes shall be deemed to have competitive prices where they are posted at or within the maximum bid-ask range set by the trading venue and imposed upon every investment firm signing a market making agreement. They should carry out this market activity continuously during a specified proportion of the trading venue’s trading hours except under exceptional circumstances, with the result of providing liquidity on a regular and

predictable basis to the trading venue. Finally, they should have in place at all times effective systems and controls to ensure that they fulfil their obligations under the agreement with the trading venue. [17.28] Some practices of investment firms create specific risks for market stability. Therefore, they are subject to special rules in MiFID II.

B. Direct Electronic Access and General Clearing Member [17.29] Direct electronic access (DEA) means an arrangement where a member or participant or client of a trading venue permits a person to use its trading code so the person can electronically transmit orders relating to a financial instrument directly to the trading venue.24 DEA includes arrangements which involve the use by a person of the infrastructure of the member or participant or client, or any connecting system provided by the member or participant or client, to transmit the orders (direct market access) and arrangements where such an infrastructure is not used by a person (sponsored access). Direct Market Access (DMA) is an arrangement through which an investment firm that is a member/participant or user of a trading platform permits specified clients (including eligible counterparties) to transmit orders electronically to the investment firm’s internal electronic trading systems for automatic onward transmission under the investment firm’s trading ID to a specified trading platform. Sponsored Access (SA) is an arrangement through which an investment firm that is a member/participant/ or user of a trading platform permits specified clients (including eligible counterparties) to transmit orders electronically and directly to a specified trading platform under the investment firm’s trading ID without the orders being routed through the investment firm’s internal electronic trading systems. [17.30] The ESMA 2012 Guidelines already identified those activities as a source of risk. The Guidelines included provisions on organizational requirements for regulated markets and multilateral trading facilities (Guideline 7, Organizational requirements for regulated markets and

multilateral trading facilities whose members/participants and users provide direct market access/sponsored access) as well as investment firms (Guideline 8, Organizational requirements for investment firms that provide direct market access and/or sponsored access) whose members/participants and users provide direct market access/sponsored access. Those Guidelines require that trading venues monitor their systems to minimize any potential disruption caused by these third parties, and investment firms must establish policies and procedures to ensure the trading of those clients complies with the rules and procedures of the relevant trading platforms. [17.31] Under MiFID II, DEA users will be regulated as investment firms.25 In addition, investment firms offering DEA will need to have effective systems and controls which ensure ‘a proper assessment and review of the suitability of clients using the service’, that ‘clients using the service are prevented from exceeding appropriate pre-set trading and credit thresholds’, that ‘trading by clients using the service is properly monitored’, and that ‘appropriate risk controls prevent trading that may create risks to the investment firm itself or that could create or contribute to a disorderly market or could be contrary to […] the rules of the trading venue’.26 Investment firms will remain responsible for ensuring that clients comply with the requirements of MiFID II and the rules of the trading venue. They have a duty to monitor the transactions. There is an obligation to have a binding written agreement and the investment firm retains responsibility. There is also a duty to keep records. The competent authority has the right to inspect these and cooperation is required between competent authorities. ESMA Draft RTS 6 fleshes out these requirements. [17.32] An investment firm that acts as a general clearing member for other persons is subject to similar requirements. It should have in place effective systems and controls to ensure clearing services are only applied to persons who are suitable and meet clear criteria, and that appropriate requirements are imposed on those persons to reduce risks to the investment firm and to the market.27 Article 16 of the ESMA Draft RTS 6 requires that investment firms monitor their clients’ exposure in as close to real-time basis as possible. The investment firm must also ensure that there is a binding written agreement between the investment firm and the person

regarding the essential rights and obligations arising from the provision of that service.

2. Trading Venues [17.33] MiFID II also applies requirements to trading venues. The scope is large since it covers regulated markets, multilateral trading facilities (MTFs) and also organized trading facilities (OTFs). Trading venues should act as a second line of defence, after investment firms, against market instability created by algorithmic trading. Therefore, some rules are similar to those (A) applicable to investment firms. However, others are specific to trading venues (B).

A. Organizational Requirements [17.34] Under MiFID II, regulated markets will need to have in place effective systems, procedures, and arrangements to ensure their trading systems are resilient, have sufficient capacity to deal with peak order and message volumes, are able to ensure orderly trading under conditions of severe market stress, are fully tested to ensure such conditions are met, and are subject to effective business continuity arrangements to ensure continuity of their services if there is any failure of their trading systems.28 Similar rules apply to MTFs and OTFs. ESMA draft RTS 7 (Draft regulatory technical standards on organizational requirements of regulated markets, multilateral trading facilities and organized trading facilities enabling or allowing algorithmic trading through their system) subjects those organizational requirements to a proportionality principle since trading venues are very diverse. The requirement to have in place robust systems to ensure continuity and regularity in the performance of the market is close to ESMA 2012 Guideline 1 adopted under MiFID I.29 [17.35] As to the capacity and resilience of trading venues, there should be a due diligence to ensure that all members or participants of trading venues meet certain predefined parameters: testing the capacity of members

or participants to access trading systems (test their own systems, conformance testing on members, minimum requirements for the testing environment), testing the members’ algorithms to avoid disorderly trading conditions, trading venue’s capacity, and ongoing monitoring and periodic review of the performance and capacity of the trading system. Those systems should be robust enough to ensure continuity and consistency of performance and able to generate alerts within five seconds of the event, which implies that they should be also automatic. [17.36] There should be means to ensure resilience of trading venues: business continuity, prevention of disorderly trading conditions (mechanisms to halt trading), pre-trade and post-trade controls, direct electronic access, and security. Appropriate criteria must be set and applied regarding the suitability of persons to whom such access may be provided and ensuring that the member or participant retains responsibility for orders and trades executed using that service.30 Trading venues will also have to set appropriate standards regarding risk controls and thresholds on trading through such access and be able to distinguish and if necessary to stop orders or trading by a person using direct electronic access separately from other orders or trading by the member or participant. [17.37] Regulated markets, MTFs, and OTFs that allow algorithmic market-making strategies must have written agreements with the investment firms in order to provide liquidity to the market on a regular and predictable basis.31 Trading venues will have to monitor and enforce compliance by investment firms. Trading venues are responsible for identifying ‘exceptional circumstances’ which allow the interruption of market-making activities. Article 3 of ESMA draft RTS 8 (Draft regulatory technical standards on market making agreements and market making schemes) provides a list of events, such as extreme volatility. [17.38] Those organizational requirements are similar to those which apply to investment firms. However, some provisions are specific to trading venues.

B. Trading Requirements

[17.39] The ESMA 2012 Guidelines adopted under MiFID I did not include precise requirements as to fair and orderly trading.32 The 2012 Guideline 3 (Organizational requirements for regulated markets and multilateral trading facilities to promote fair and orderly trading in an automated trading environment) only required that ‘regulated markets’ and multilateral trading facilities’ rules and procedures for fair and orderly trading on their electronic markets should be appropriate to the nature and scale of trading on those markets, including the types of members, participants and users and their trading strategies’. MiFID II goes further and provides more precise requirements. [17.40] First, trading venues must include systems that limit the ratio of unexecuted orders to transactions (OTR), in order to slow down the flow of orders, and that limit the maximum tick size that may be executed on the market.33 Second, they must also have transparent, fair, and nondiscriminatory rules on co-location.34 This requirement is essential in order to have equality of treatment among HFT traders. Article 2 of the ESMA draft RTS 10 (Draft regulatory technical standards on requirements to ensure co-location and fee structures are fair and non-discriminatory) requires the publication of a trading venue’s policies on its website. This should help assure a level playing field among HFT traders and also with other traders and investors. Finally, trading venues should have transparent, fair, and non-discriminatory fee structures that do not create incentives to place, modify, or cancel orders or to execute transactions in a way which contributes to disorderly trading conditions.35 To provide this transparency, Article 3 of the ESMA draft RTS 10 requires that trading venues publish their policies on their websites. [17.41] MiFID II is also focused on High-Frequency Trading.

III. High-Frequency Trading [17.42] High-Frequency Trading is (1) subject to specific rules under MiFID II designed to deal with market risk. It is also (2) subject to some provisions designed to reduce the risk of market abuse.

1. The Specific Regulation of High-Frequency Trading [17.43] MiFID II subjects (A) investment firms that engage in HFT to specific rules as well as (B) trading venues.

A. Investment Firms that Engage in HFT [17.44] The most important measure in MiFID II is the removal of the exemption for proprietary trading done by HFT traders. Article 2(1)(d)(iii) provides that the Directive does not apply to persons dealing on own account in financial instruments other than commodity derivatives or emission allowances or derivatives thereof and not providing any other investment services or performing any other investment activities in financial instruments other than commodity derivatives or emission allowances or derivatives thereof unless such persons […] apply a high-frequency algorithmic trading technique.

[17.45] Therefore, proprietary traders carrying out HFT in the EU will need to apply for a licence as an investment firm. This requirement is a major change since it triggers the application of capital requirements in line with the Capital Requirements Directive (CRD IV). This impact has probably not been fully realized by the European legislator. Therefore, HFT traders will face a huge increase in their operational costs. Although this does not imply that they will not be able to continue doing HFT, it is likely to make their cost of doing business much higher and potentially significantly reduce their activity. [17.46] HFT traders will be subject to record-keeping obligations. An investment firm that engages in HFT must store in an approved form accurate and time-sequenced records of all its placed orders, including cancellations of orders, executed orders, and quotations on trading venues, and shall make them available to the competent authority upon request.36 [17.47] Article 29 of the ESMA Draft RTS 6 provides for a recordkeeping period of five years.

B. Regulated Markets [17.48] Regulated markets (RMs) will also be subject to specific requirements targeted at HFT. First, Member States must allow a RM to adjust its fees for cancelled orders according to the length of time for which the order was maintained and to calibrate the fees to each financial instrument to which they apply.37 [17.49] Second, Member States may allow a RM to impose a higher fee for placing an order that is subsequently cancelled than an order which is executed, and to impose a higher fee on participants placing a high ratio of cancelled orders to executed orders and on those operating a high-frequency algorithmic trading technique in order to reflect the additional burden on system capacity.38 [17.50] Also, fee structures must be transparent, fair, and nondiscriminatory and must not create incentives to place, modify, or cancel orders or to execute transactions in a way which contributes to disorderly trading conditions or market abuse.39 Finally, there should be a flagging of orders generated by algorithmic trading.40 Flagging of orders allows regulators to audit the order trail more easily across markets.

2. The Specific Regulation to Prevent Market Abuses [17.51] Some regulators consider that some HFT techniques can lead to market abuses. However, proving that HFT can be used to abuse the market is difficult. Therefore, MiFID II includes requirements designed to (A) prevent and identify market abuses. A 2015 decision by the French securities regulator provides clues as to what could constitute market abuse for all other regulators, but also suggests the considerable difficulties encountered in identifying it (B).

A. Requirements Designed to Prevent and Identify Market Abuse [17.52] The ESMA 2012 Guidelines included requirements designed to help supervisors identify suspicious activities by HFT traders. The Guidelines were based on provisions of MiFID I and of the Market Abuse Directive (MAD). Although they apply to all algorithmic trading, they are designed more, by the nature of the suspicious trades, for HFT traders. Some of those provisions apply to trading venues while others apply to investment firms. [17.53] ESMA 2012 Guideline 5 required regulated markets and multilateral trading facilities to have effective arrangements and procedures, taking account of the specific supervisory arrangements/regulation in their Member State, which enable them to identify conduct by their members/participants and users that may involve market abuse (in particular market manipulation) in an automated trading environment.41 Guideline 5 identified potential cases of market manipulation that could be of particular concern in an automated trading environment and which are to be found with HFT traders rather than simply through algorithmic trading. [17.54] For instance, ESMA mentioned ping orders, which consist of entering small orders in order to ascertain the level of hidden orders and are used to assess what is resting on a dark platform. Quote stuffing consists of entering large numbers of orders and/or cancellations/updates to orders so as to create uncertainty for other participants, slowing down their process and to camouflage an HFT trader’s own strategy. Momentum ignition consists of entry of orders intended to start or exacerbate a trend, and to encourage other participants to extend the trend in order to create an opportunity to unwind/open a position at a favourable price. Layering and spoofing consist of submitting multiple orders on one side of the order book with the intention of executing a trade on the other side. Once that trade has taken place, the manipulative orders are removed. [17.55] ESMA 2012 Guideline 6 required investment firms to have policies and procedures in place to minimize the risk that their automated trading activity gives rise to market abuse (in particular market

manipulation). The sorts of market manipulation that might be of particular concern in a highly automated trading environment were described in Guideline 5.42 [17.56] MiFID II provides that investment firms that engage in algorithmic trading provide effective systems and risk controls to ensure the trading systems cannot be used contrary to the 2014 Market Abuse Regulation.43 Article 13 of the ESMA draft RTS 6 requires an automated (not manual) monitoring of algorithmic trading activities, but subject to a proportionality principle. Investment firms that provide direct electronic access (DEA) also have an obligation to have appropriate risk controls to prevent trading that could be contrary to the 2014 Market Abuse Regulation.44 Trading venues are also subject to requirements to prevent and identify market abuses. The fee structure should not create incentives which would contribute to market abuses.45 However, it is difficult to understand what type of fee structure could contribute to market abuses.

B. Sanctions [17.57] Sanctions against HFT traders and trading venues for market abuse are rare. However, there has been a recent example, in France in December 2015, with the sanction by the Sanction Commission of the Autorité des Marchés Financiers (AMF) of Virtu and Euronext.46 This decision is rare and significant. In practice, such investigations require cross-border cooperation which makes market abuses by HFT difficult to prove. There are several reasons for this situation. Competent authorities are reluctant to invest in information technology and staff to monitor order books at a national level in order to identify violations by HFT traders. It took the AMF several months to get order book data from foreign regulators in the context of the formal investigation. [17.58] In this case, Euronext was sanctioned for providing Virtu with an exemption to its fee above OTR without public disclosure. Euronext had acted with a lack of neutrality and impartiality. Virtu was sanctioned for market manipulation (layering). The decision holds that it created a

disorderly market through the high number of cancelled trades (on 27 liquid stocks on the CAC 40), that it created a false appearance of the order book (orders removed before the other faster players could operate on them), and that it occupied a dominant position (80 per cent of the order book on those stocks). The Sanction Commission of the AMF decided on a €5 million sanction each for Euronext and Virtu. This was more than had been requested by the AMF and six times the profit made by Virtu on these trades. Euronext filed an appeal with the Paris court of Appeal but Virtu, which has exited France, did not. [17.59] This decision is equivalent to a de facto ban on HFT in France, as the Sanction Commission held that the usual behaviour of high-frequency traders constituted a violation of French law (a high number of cancelled trades and dominating position). [17.60] Because the sanction describes in detail the alleged behaviour, it could be used by other supervisors as a template if they should also wish to prosecute HFT traders.

IV. Conclusion [17.61] Algorithmic trading raises serious issues of volatility and systemic risk while HFT raises issues of systematic ‘front-running’ of investors. However, there is serious divergence within the EU as to the benefits and risks of those two trading techniques, and especially of HFT. Therefore, MiFID II adopts a technical approach mostly focused on prevention of a repeat of the 2010 Flash Crash and some provisions on market abuse. However, as to HFT, the requirement that proprietary traders carrying out this technique in the EU will need to apply for a licence as an investment firm should have a considerable impact. [17.62] As the MiFID II will not be implemented before January 2018, the ESMA 2012 Guidelines will remain the most effective regulation to frame the development of HFT. This reveals the usefulness of having ESMA adopting Guidelines in order to tackle market developments with relative speed. However, since the EU legislator decided not to tackle HFT

strongly, and since implementation of the Directive is still far away, it is probable that some European legislators and supervisors will do it themselves. The successful case of enforcement in France is a sign of this possible trend. Therefore, regulation by prosecution of market abuse by HFT traders could lead to a de facto ban of HFT in some Member States. However, this would be at a huge cost for supervisors, who would need to allocate scare resources to this topic, which means that only the most motivated supervisors will supervise it thoroughly. On the issue of HFT, Europe stands to stay divided for a long time.

1

Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU. 2 Article 4(1)(39) of MiFID II ‘excludes any system that is only used for the purpose of routing orders to one or more trading venues or for the processing of orders involving no determination of any trading parameters or for the confirmation of orders or the post-trade processing of executed transactions’. 3 ESMA, Economic Report, High-frequency trading activity in EU equity markets, Number 1, 2014. 4 Bank for International Settlements, Markets Committee, ‘High-frequency trading in the foreign exchange market: Report submitted by a Study Group established by the Markets Committee’, September 2011 . 5 ‘Findings Regarding the Market Events of 6 May 2010: Report of the Staffs of the CFTC and SEC to the Joint Advisory Committee on Emerging Regulatory Issues’, 30 September 2010 . 6 Michael Lewis, Flash Boys: A Wall Street Revolt (New York/London: W.W. Norton & Company, 2014), p. 274. 7 M. Baron, J. Brogaard, and A. Kirilenko, ‘The Trading Profits of High Frequency Traders’, 2012. The article is available at SSRN. 8 J. Macey and D. Swensen, ‘One Way to Unrig Stock Trading by Dec. 24, 2015’, New York Times . 9 See The Netherlands Authority for the Financial Markets, ‘High Frequency Trading: The Application of Advanced Trading Technology in the European Marketplace’, Amsterdam, November 2010 . 10 Gesetz zur Vermeidung von Gefahren und Missbräuchen im Hochfrequenzhandel (Hochfrequenzhandelsgesetz), BR 156/1/13 Ausschussempfehlung, 2012.

11

ESMA, Guidelines. ‘Systems and controls in an automated trading environment for trading platforms, investment firms and competent authorities’, 24 February 2012, ESMA/2012/122. 12 Position paper, ESMA’s Consultation on systems and controls in a highly automated trading environment for trading platforms, investment firms and competent authorities, 26 October 2011, ESMA/2011/SMSG/12. 13 The G20 Seoul Summit Declaration, 11–12 November 2010, available at . 14 IOSCO, Regulatory Issues Raised by the Impact of Technological Changes on Market Integrity and Efficiency, Final Report, FR09/11, October 2011. 15 ESMA, Guidelines, Systems, and controls in an automated trading environment for trading platforms, investment firms and competent authorities, 24 February 2012, ESMA/2012/122 (EN). 16 ESMA, Final Report, Draft Regulatory and Implementing Technical Standards MiFID II/MiFIR, 28 September 2015, ESMA/2015/1464. The draft RTS are in Annex I. 17 Commission Delegated Regulation (EU) of 25.4.2016 supplementing Directive 2014/65/EU of the European Parliament and of the Council as regards organizational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive. 18 The legal basis is Article 13, paras (2), (4), (5), and (6) of MiFID I, and Articles 5, 6, 7, 9, 13, 14, and 51 of the MiFID I Implementing Directive. 19 Article 17(1) of MiFID II. 20 The legal basis is Article 13, paras (2), (4), (5), and (6) of MiFID I, and Articles 5 to 9, 13, 14, and 51 of the MiFID I Implementing Directive. 21 Article 17(2) of MiFID II. 22 Article 17(3) of MiFID II. 23 Article 17(4) of MiFID II. 24 Article 4(1)(41) of MiFID II and Article 20 of Commission Delegated Regulation of 25.4.2016. 25 Article 2(1)(d)(iii) of MiFID II. 26 Article 17(5) of MiFID II. 27 Article 17(6) of MiFID II. 28 Article 48(1) of MiFID II. 29 The legal basis is Article 39, paras (b) and (c) of MiFID I for regulated markets. Articles 14, para. (1), and 13, paras (2), (4), (5), and (6) of MiFID I and Articles 5 to 9, 13, 14, and 51 of the MiFID Implementing Directive for multilateral trading facilities. 30 Article 48(1) of MiFID II. 31 Article 48(2) and (5) of MiFID II. 32 The legal basis is Article 14, paras (1) and (4), Article 13, paras (2), (5), and (6), Article 42, para. (3), and Article 26 of MiFID I, and Articles 13, 14, and Article 51 of the MiFID I Implementing Directive for multilateral trading facilities.

33

Article 48(6) of MiFID II. Article 48(8) of MiFID II. 35 Article 48(9) of MiFID II. 36 Article 17(2) of MiFID II. 37 Article 48(9) of MiFID II. 38 Article 48(9) of MiFID II. 39 Article 48(9) of MiFID II. 40 Article 48(10) of MiFID II. 41 The legal basis is Article 39, paras (b) and (d), and Article 43 of MiFID I, Article 6, paras (6) and (9), of MAD, and Articles 7 to 10 of the MAD Implementing Directive 2004/72/EC for regulated markets. 42 The legal basis is Article 13, paras (2), (5), and (6), of MiFID I, Articles 5, 6, and 9 of the MiFID I Implementing Directive, Article 6, para. (9), of MAD and Articles 7 to 10 of the MAD Implementing Directive 2004/72/EC. 43 Article 17(1) of MiFID II. 44 Article 17(5) of MiFID II. 45 Article 48(9) of MiFID II. 46 Decision of the Sanction Commission of the AMF against Euronext Paris SA and Virtu Financial Europe Ltd, Paris, 4 December 2015. 34

18 MIFID II AND EQUITY TRADING A US View Merritt B. Fox

I. Basic Building Blocks 1. The Economics of Liquidity Provision 2. The Evaluative Framework II. 1. 2. 3.

High-Frequency Trading Electronic Front-Running Slow Market Arbitrage High-Frequency Trading and Volatility

III. Dark Pools and Internalization 1. Understanding the Function of Dark Pools 2. The Question of Whether Even Honestly Operated ‘Ideal’ Dark Pools Are Socially Desirable 3. Practices Damaging Traders Using Dark Pools 4. Internalization IV. 1. 2. 3.

Conclusions: Reflections on Issues Raised by MiFID II Dark Pools and an Effective Price-Formation Process HFTs: Price Volatility and Market Abuse Trading Venue Competition

[18.01] At the most fundamental level, today’s equity markets in the European Union and in the United States largely resemble each other.1 Equally importantly, each resembles very little what either looked like a few

decades ago. Currently, in the markets of both places, the stock of the typical significant public company is traded in a number of competing venues, including multiple exchanges and a number of dark pools. Almost all of these competing trading venues are electronic limit order books, where a trader can post, as a limit order, its firm commitment, until cancelled, to buy or sell up to a specified number of shares at a quoted price. These are the source of the market’s quotes that used to be provided by a dealer. A computer (the venue’s matching engine) matches these posted limit orders with incoming buy and sell market orders, which are orders from traders willing to trade at whatever is the best available price in the market. Also, on both continents, high-frequency traders (HFTs) post a substantial portion of the limit orders that are matched in this fashion and result in executed trades. An HFT uses high-speed communications to constantly update its information concerning transactions occurring in each stock that it regularly trades, and concerning changes in the buy and sell limit orders posted by others on every major trading venue. This information is automatically fed into a computer that uses algorithms to change the limit prices and quantities associated with the HFT’s own limit orders posted on each of the various trading venues. Most of the remaining trades that do not occur on electronic limit order book venues involve a broker internally matching the buy and sell orders received from its own retail customers. [18.02] Data concerning the speed of trading, its cost, and the apparent amount of liquidity in the system suggest that today’s US stock market is a substantial improvement on what went before it.2 Given its similarity in fundamental structure, there is every reason to believe that this is so for the EU stock market as well, although others writing in this volume suggest that there is doubt about this among some European commentators.3 On both continents, however, it is certainly possible that this improvement, which is, at least in part, clearly related to huge advances in information technology, could be even greater with a different regulatory or institutional structure. Certainly, not everyone is happy with the current state of US affairs. Michael Lewis’s best-selling book Flash Boys: A Wall Street Revolt4 is emblematic of this dissatisfaction. HFTs, for example, are said to use their speed in finding out changes in the market and in altering their own orders to take advantage of other traders in the market.5 Dark pool

operators, which promise customers to keep their orders secret and to restrict the kinds of parties that can end up as counterparties, are said to often break these promises to the customers’ disadvantage,6 and there is fear that even properly run dark pools undermine equity price discovery. [18.03] A look across the Atlantic at the discussions concerning MiFID II and the European equity markets more generally reveals many of the same kinds of concerns. This is not surprising: similar markets have the potential to raise similar issues. My purpose here is to reflect on some of these issues in the context of the US markets—the ones I know best—with the hope that doing so may shed light on the choices facing Europe as well. Specifically, after laying out some basic building blocks for analysis, I will address some issues raised by dark pools, HFTs, and efforts at promoting competition among trading venues.

I. Basic Building Blocks [18.04] Much of the criticism of today’s US stock market simply consists of taking a representative single transaction exemplifying some controversial practice, showing that the transaction benefits one party at the expense of another, and labelling the resulting transfer as ‘larcenous’, ‘extractive’, ‘predatory’, or simply ‘unfair’. Similarly, in Europe, some commentators criticize some changes—for example the effects of HFTs on liquidity, or dark pools on effective price formation—simply based on plausible-sounding intuitions rather than rigorous theory or sound empirical evidence. [18.05] Serious analysis requires digging deeper. There needs to be a consideration of the effects of each criticized practice as something that occurs on a repeated basis among competing actors, taking into account the reaction of the various other participants in the market to their knowledge that the practice is going on. Additionally, there needs to be an evaluation of these effects in terms of their ultimate impact on the multiple social goals that equity trading markets are expected to serve and that form the justificatory basis for regulation where the markets fall short.

1. The Economics of Liquidity Provision [18.06] Markets benefit enormously from competing businesses that post on trading venues limit orders against which marketable orders can transact. The resulting availability of these limit orders substantially increases liquidity. These businesses are referred to as ‘liquidity providers’ or ‘market-makers’. A professional supplier of liquidity for an issuer’s shares —typically, today, an HFT posting buy and sell limit orders—engages in both the frequent purchase and frequent sale of these shares. In doing so, it stands ready to buy and sell shares up to the limit order’s stated amounts at their stated bid or ask prices. [18.07] The liquidity supplier makes money if on average it sells the shares that it buys for more than the price paid. It might appear that making money this way is easy, even in markets where the offer is only a penny above the bid: buy at the bid and sell at the offer and make a half cent per share on every transaction. Do this for a billion shares and pretty soon you are talking about real money. In fact, however, it is not so easy to make money. This is because the persons with whom a liquidity supplier trades generally do not reveal their identities and there is always the possibility that the person (the ‘trader’) who places a marketable order that executes against the liquidity supplier’s quote is doing so because the trader has private information not known to most of the market or to the liquidity supplier.7 [18.08] An informed trader of this kind will buy from the liquidity supplier when her private information suggests that the stock’s value is above the liquidity provider’s offer. And she will sell to the liquidity supplier when her private information suggests the value is below the liquidity provider’s bid. In such transactions, the liquidity supplier sells at prices on average below the value of the stock and buys at prices on average above the value of the stock, not a formula for success. Despite this, the liquidity supplier, if skilful, can still make money on a net basis, because the remaining traders with whom it transacts do not possess private information and the liquidity supplier can profit on these transactions.8

A. Kinds of Private Information and Their Sources [18.09] Information concerning a stock is ‘private’ if it allows a more accurate appraisal of the stock’s value than the assessment of its value implied by the current market price. There are three primary kinds of private information: inside information, announcement information, and fundamental value information. i. Inside Information [18.10] Inside information’s ultimate origins come from within some institutional source. Frequently, this institution is the issuing company of the stock itself. This is information that the institution seeks to prevent from becoming public or from being the basis of trading by others. In the US, trading on such information is, under many circumstances, illegal under Section 10(b) and Rule 10b-5 of the Exchange Act, as it is in EU countries under national legislation implementing the Market Abuse Directive (MAD). The existence of cases of successful prosecutions under these provisions shows that such information is the basis of at least some of the informed trading that occurs in the market. ii. Announcement Information [18.11] Announcement information is information that has only just been publicly revealed, for example a government statistic about the economy or a company earnings announcement. A trader who acts on this information extremely quickly, before other traders and the liquidity suppliers themselves can react, is also an informed trader. iii. Fundamental Value Information [18.12] Fundamental value information is a superior estimate of an issuer’s future cash flows based on a person gathering bits of publicly available information and analysing what this person has gathered in a sophisticated way. The traders whose trades are informed due to this kind of information include hedge funds, actively managed mutual and pension funds, non-profit institutions, and very wealthy individuals with actively managed portfolios. Liquidity suppliers are vulnerable to trades based on this kind of private information too, because liquidity suppliers tend to specialize in the business of supplying liquidity. Thus, they generally do not engage in their own fundamental analysis.

B. Transacting with Informed versus Uninformed Traders [18.13] An informed trader will buy from the liquidity supplier only when her superior assessment of the stock’s value suggests that the value is above the liquidity supplier’s offer. And she will sell to the liquidity supplier only when her superior assessment suggests that the value is below the liquidity supplier’s bid. Thus, in transactions with an informed trader, the liquidity supplier sells at prices that the informed trader’s information suggests is below the value of the stock, and buys at prices that the informed trader’s information suggests is above the value of the stock. These on average will be losing transactions for the liquidity supplier. In essence, the liquidity supplier faces a classic adverse selection situation.9 [18.14] Fortunately for the liquidity supplier, the rest of its transactions are with uninformed traders. On average, these transactions should be profitable. As we will see, the assessment of value of the stock implied by current market prices is the mid-point between the national best offer (NBO) and national best bid (NBB).10 Because the uninformed trader has no private information, there is no reason to think that on average this market assessment is wrong. So when a liquidity supplier purchases from an uninformed trader at the NBB and sells to an uninformed trader at the NBO, each of these transactions on average yields an expected profit equal to half the spread between the two quotes, with the liquidity supplier on average buying for a little less than value and selling for a little more than value. [18.15] In sum, whatever the source of an informed trader’s private information, the liquidity provider will be subject to adverse selection and on average lose money when it buys at the bid from informed sellers or sells at the offer to informed buyers. As long as there are enough uninformed traders willing to suffer, in order to accomplish their reasons for trading, the inevitable expected trading losses of buying at the offer and selling at the bid, the liquidity provider can still break even, however.11 Thus the resulting bid–ask spread will be the one such that what the liquidity supplier makes from trading with uninformed traders just equals what it loses from trading with informed traders. There simply needs to be a

large enough spread between the bid and offer that the losses accrued by transacting with informed traders are offset by the profits accrued from transacting with uninformed investors.12

C. How Liquidity Suppliers Set Their Bids and Offers [18.16] There are two ways to think about the calculations that liquidity providers need to perform in setting particular bids and offers, sometimes referred to as the ‘accounting perspective’ and the ‘information perspective.’ One of the fundamental theorems of microstructure economics is that each of these two perspectives leads a liquidity supplier to quote the same bid–ask spread at any given point in time.13 i. The Accounting Perspective [18.17] The accounting perspective is based on the proposition that a liquidity supplier operates in a competitive market and that, to survive, it must set its quotes aggressively enough to attract business, but not so aggressively that the money it makes by buying from, and selling to, uninformed traders is less than what it loses by engaging in such transactions with informed traders. Thus the resulting bid– ask spread will be the one such that what the liquidity supplier makes from trading with uninformed traders just equals what it loses from trading with informed traders.14 ii. The Information Perspective [18.18] The information perspective starts by noting that a liquidity supplier knows that there is a certain possibility that the next marketable order that arrives to execute against one of its quotes will be from an informed trader. The liquidity supplier knows that if the next marketable order to arrive is a buy, there is a certain chance that it is motivated by positive private information and there is no chance it is motivated by negative private information. Similarly, if the next order to arrive is a sell, there is a certain chance that it is motivated by negative private information and there is no chance it is motivated by positive private information. Thus, the liquidity supplier knows that whichever kind of order arrives next, it will alter the liquidity supplier’s estimate of the stock’s value: up if the order is a buy and down if it is a sell order.

[18.19] The offer and the bid are set in advance of knowing which it will be, but with the offer being contingent on the next arriving order being a buy and the bid on it being a sell. Thus, when a liquidity supplier is deciding on its offer price, it knows that an informed trader will only transact against this price if the information possessed by the informed trader is positive and thus that the arrival of a buy order will cause the liquidity supplier to revise its estimate upwards. So, for a transaction with a buy order to be regret free, the liquidity supplier must, in advance of the arrival of the order, set its offer quote to reflect this upward revision of estimated value that will inevitably accompany the buy order’s arrival. The same logic applies for setting the bid: to be regret free it must reflect the downward revision that would inevitably accompany the arrival of a sell order.15

D. The Pattern of Transaction Prices in the Presence of Informed Trading [18.20] This second approach highlights an important characterization of rational liquidity provision in a market with informed traders. Liquidity suppliers will be constantly updating their bids and offers in response to transactions. With a sufficient number of trades, the market price will come to reflect the informed traders’ information. The behaviour of rational liquidity providers thus reflects a kind of ‘invisible hand’: simply as a result of their efforts to minimize losses to informed traders, liquidity providers are repeatedly revising their quotes so that, with time, they come to fully reflect informed traders’ information. [18.21] To see this in an example, suppose that there were one or more informed traders possessing a particular piece of positive information. During their period of trading, there would of course also be buying and selling by uninformed traders. So both marketable buy and marketable sell orders will arrive at trading venues, but there will be more buys than sells. As a result, although there will be ups and downs in the offers and bids as the liquidity-supplier estimates of value move up and down with the arrival of each buy and sell order, the ups will predominate and the mid-point between the bid and offer will trend upwards until the offer gets high

enough that it equals the informed traders’ estimate of the share’s value.16 Empirical evidence strongly supports the results from these adverse selection models. Analyses of intra-day changes in quotes and in the prices of executed transactions consistently show that they respond to the pattern of buy and sell orders at the time and that the adjustment in price described here often completes itself quite quickly.17

2. The Evaluative Framework [18.22] With this explanation of the economics of the liquidity supply in hand, we must consider the evaluative framework that I will use to analyse stock market public policy issues. The parties participating in today’s equity markets do so in a competitive market on a repeated basis. Each actor in the system, in determining its own actions, generally takes into account the fact that each of the other actors in the system will act in accordance with its individual advantage, subject to whatever constraints are imposed by effective regulation, contractual obligations, and reputational concerns. The question for good regulation is how the existence of any given practice in the market affects the system as a whole in terms of its ultimate impact on the multiple social goals that equity trading markets are expected to serve and that form the justificatory basis for regulation when the markets fall short.

A. Goals [18.23] A number of social goals animate discussion of secondary equity markets18 and their regulation: (i) promoting the efficient allocation of capital so that it goes to the most promising new investment projects in the economy; (ii) promoting the efficient operation of the economy’s existing productive capacity; (iii) promoting the efficient allocation of resources between current and future periods so as to best satisfy the needs of firms seeking funds for real investments (trading the promise of future dollars to obtain current dollars) and the needs of savers seeking to forgo current consumption in order to enjoy future consumption (trading current dollars to obtain the promise of future dollars); (iv) promoting the efficient

allocation among investors of the risks associated with holding securities so that the volatility in the cash flows generated by productive enterprises is borne by risk-averse investors in a way that generates the least disutility; (v) fostering an overall sense of fairness; (vi) economizing on the real resources that society devotes to the operation of the trading markets and to the enforcement and compliance costs associated with their regulation; and (vii) fostering innovation that over time can improve the capacity of the system to serve these preceding goals. [18.24] Any particular practice in the market may, of course, have a positive impact in terms of some of these seven goals and a negative impact in terms of others. It is nevertheless desirable to structure the market for the secondary trading of equities so that there are no unnecessary trade-offs— that is, so that it satisfies each goal to the fullest extent possible without compromising one or more others—and to identify the nature of the remaining unavoidable trade-offs so that intelligent choices can be made.

B. Market Characteristics That Impact on These Goals [18.25] The stock market’s operations relate to these social goals in complex ways that are the product of its interacting characteristics. The two most important characteristics are share price accuracy and liquidity. A third characteristic, anonymity, importantly affects these two and is also discussed. The impact of any given practice on the goals above is most easily evaluated through a two-step process; first assessing the effect of the practice on each of these three market characteristics, and then identifying the effect of the characteristic on the goals. i. Price Accuracy [18.26] Price accuracy relates to the accuracy with which the market price of an issuer’s shares predicts the issuer’s future cash flows. Because the price of any new share offering by a publicly traded issuer will be determined largely by the price of its already outstanding shares in the stock market, more accurate stock market prices will lead to capital being more likely to go to the issuers with the most promising new real investment projects. Share price also influences the availability of new project funding from other outside sources and the willingness of managers

to use internal funds for investment, and so greater price accuracy assists the efficient allocation of capital in these other ways as well.19 [18.27] More generally, more accurate share prices help reveal managers who are performing poorly, both in terms of their deployment of internal funds for new investment projects (again assisting the efficient allocation of capital) and in terms of their management of the issuer’s current assets (assisting the efficient operation of the economy’s existing productive capacity).20 [18.28] Over time, more accurate share prices today also likely lead to a greater sense of fairness on the part of investors because they will experience fewer negative surprises.21 ii. Liquidity [18.29] A second characteristic is how liquid the market is. Liquidity is a multi-dimensional concept that relates to the size of a trade, the price at which it is accomplished, and the time it takes to accomplish the trade. Generally, the larger the size of the purchase or sale and the faster one wishes to accomplish it, the less desirable will be the price. The more liquid the market is, however, the less severe are these trade-offs. For a small retail purchase or sale of stock, the spread between the best offer and best bid available in the market (referred to in the United States as the NBO and NBB) is a good measure of liquidity because the trader can effect the buy or sell transaction immediately at those respective prices and, in essence, will be paying half the spread to do so. For larger orders, how much is available at prices not too inferior to the best offer and best bid (the ‘depth of the book’) will become relevant as well. [18.30] Liquidity also has an impact on a number of social goals. More Efficient Allocation of Resources over Time [18.31] To start, the prospect of greater liquidity promotes more efficient allocation of resources over time. Consider this first in terms of enterprises seeking new capital to devote to real investment projects. In essence, they are purchasers of current dollars in return for the promise of future dollars. The more liquid an issuer’s shares are, the more valuable their shares are to hold for any given level of expected future dollar cash flow.22 Thus, when an issuer offers

shares in the primary market, the more liquid that investors anticipate the shares will be in the future, the higher the price, all else equal, at which the issuer can sell its shares. Hence, the lower the issuer’s cost of capital.23 [18.32] In welfare economics terms, just like a tax, illiquidity results in a ‘wedge’ between the value of what the savers (the purchasers of future dollars) expect to receive in the future and what the entrepreneurs or issuers (the suppliers of future dollars in the form of future dividend streams) expect to give up in the future. As a result, illiquidity results in resources being allocated less efficiently over time. This wedge prevents certain transactions from occurring that would have occurred if the shares were expected to be more liquid. The fact that, absent this wedge, the issuer and savers would have willingly entered into these transactions means that the transactions prevented by illiquidity are ones that would have made both parties better off on an expected basis. These lost transactions are projects with expected returns that are lower than the marginal project that gets funded in a world with a given level of illiquidity, but that nevertheless are high enough to make some people feel that, if the market were more liquid, sacrificing their current dollars for the projects’ promises of future ones would be worthwhile.24 Greater Share Price Accuracy [18.33] More liquidity also lowers the transaction costs associated with speculative trading based on acquiring a variety of bits of publicly available information and analysing them to make more accurate predictions of an issuer’s cash flows, that is, creating fundamental value information. Thus, it stimulates such activity and in the process increases share price accuracy, with the attendant benefits in terms of more efficient capital allocation and utilization of existing productive capacity discussed above. More Efficient Allocation of Risk [18.34] Greater liquidity also promotes the more efficient allocation of risk. Constant change in the world means that what constitutes an optimal portfolio, in terms of diversification and of each investor’s relative degree of risk aversion, is always shifting. By making both the purchase and sale of securities less expensive, greater liquidity allows the individual investor to cost-effectively adjust her

portfolio over time to keep it closer at each moment to what is optimal for her. iii. Anonymity [18.35] A third relevant characteristic of a trading market is the extent of anonymity that it offers. Price Accuracy [18.36] Anonymity’s direct effect on price accuracy can be positive. Consider persons engaging in speculative trading based on acquiring various pieces of publicly available information and analysing them to make more accurate predictions regarding an issuer’s cash flows, that is, those who are in the business of developing and trading upon fundamental value information. Suppose, for example, that a trader acquires and analyses such information and concludes that an issuer’s future cash flow will be better than the issuer’s current share price suggests. The trader purchases a significant number of the issuer’s shares based on this conclusion. These trades will start a process in which this information will ultimately get reflected in the issuer’s share price, making the price more accurate. Greater anonymity, however, means that the trader is likelier to have more or all of its orders executed before others detect that they are based on superior information. In other words, with greater anonymity, his orders will have less impact on the average price that the trader will pay for the shares. Thus, his trades will be more rewarding. The prospect of more remunerative trading encourages further fundamental information gathering and trading. This enhances price accuracy, with the social benefits discussed above. [18.37] On the other hand, as will be discussed in Section II, when traders with private information can more effectively hide the fact that their orders are informed, liquidity suppliers will provide less aggressive quotes, thereby lessening liquidity. This makes the socially useful activities of fundamental information traders more costly. [18.38] How the benefits and costs of anonymity balance out in terms of the profitability of fundamental information trading—and hence anonymity’s effect on share price accuracy—is complicated. As will be discussed below in connection with the practice of electronic front-running, as a group, traders with private information—fundamental value information, inside information, and announcement information—get all of

the gain from a higher level of anonymity, but bear only a portion of the additional costs, the rest of which are borne by uninformed traders. Whether or not the gains of fundamental value information traders are greater than their costs, however, depends on how much the higher level of anonymity helps them hide the informed nature of their trades relative to how much it helps those trading on the basis of announcement information or inside information.25 Efficient Allocation of Resources over Time and Risk [18.39] In any event, regardless of the net effect of greater anonymity on price accuracy, it will diminish liquidity because it makes it harder for liquidity suppliers to identify when informed trading is occurring. Thus liquidity suppliers will need to protect themselves with wider bid–ask spreads and less depth of book. This has unambiguously deleterious social consequences, as we have just seen, in terms of less efficient allocation of risk and of resources over time.

II. High-Frequency Trading [18.40] A number of practices associated with HFTs have been criticized in the United States. Two such practices are discussed here: electronic frontrunning and slow market arbitrage. Also discussed here are accusations that US HFTs contribute to market instability, a concern also reflected in the formulation of MiFID II. [18.41] It should be noted at the outset that electronic front-running and slow market arbitrage each involve an HFT benefiting itself by taking advantage of having a ‘co-location’ facility at each exchange. Co-location involves the HFT having a computer located right next to an exchange’s matching engine. This arrangement allows the HFT to find out about transactions occurring on the exchange, and changes in quoted prices, sooner than other traders. It also allows the HFT to cancel old limit orders posted on the exchange, and submit new ones, very quickly. The HFT’s colocation facility at each exchange is connected to its co-location facility at every other exchange by specialized fibre-optic cables, which permit extremely rapid communication among the HFT’s co-location facilities at

the different exchanges, all of which, in the United States, have their matching engines in northern New Jersey.

1. Electronic Front-Running [18.42] So-called electronic front-running involves a situation where an HFT, before others in the market, learns of a transaction that has occurred at one exchange and infers from this that similar orders may still be in transit heading towards other exchanges. The HFT races ahead of these orders still on their way to the other exchanges and, before they arrive at their destinations, the HFT changes its quotes on these other exchanges. [18.43] Electronic front-running has come in for harsh criticism. For example, Charlie Munger, vice chairman of Berkshire Hathaway, has objected that high-frequency trading is ‘legalized front-running […] and it should never have been able to reach the size that it did’.26 Similarly, New York Attorney General Eric Schneiderman has complained that ‘[w]hen blinding speed is coupled with early access to data, it gives small groups of traders the power to manipulate market movements in their own favor before anyone else knows what’s happening’.27 Flash Boys, published after these comments, makes electronic front-running its principal focus.28

A. An Example [18.44] We will examine the practice of electronic front-running through use of an example. For simplicity of exposition, just one HFT, Lightning, and two exchanges, BATS Y and the NYSE, are involved. Lightning has co-location facilities at the respective locations of the BATS Y and NYSE matching engines. These co-location facilities are connected with each other by a high-speed fibre-optic cable. [18.45] An actively managed institutional investor, Smartmoney, decides that Amgen’s future cash flows are going to be greater than its current price suggests. The NBO is $48.00, with 10,000 shares being offered at this price on BATS Y and 35,000 shares at this price on NYSE. Smartmoney decides

to buy a substantial block of Amgen stock and sends a 10,000 share market buy order to BATS Y and a 35,000 share market buy order to NYSE.29 The 35,000 shares offered at $48.00 on NYSE are all from sell limit orders posted by Lightning. [18.46] The order sent to BATS Y arrives at its destination first and executes. Lightning’s co-location facility there learns of the transaction very quickly. An algorithm infers from this information that an informed trader might be looking to buy a large number of Amgen shares and thus may have sent buy orders to other exchanges as well. Because of Lightning’s ultra-high-speed connection, it has the ability to send a message from its BATS Y co-location facility to its co-location facility at NYSE, which in turn has the ability to cancel Lightning’s 35,000 share $48.00 limit sell order posted on NYSE. All this can happen so fast that the cancellation would occur before the arrival there of Smartmoney’s market buy order. If Lightning does cancel in this fashion, it has engaged in ‘electronic frontrunning’. [18.47] Why might Lightning wish to cancel its sell limit order on NYSE? One possibility is that, given its inference that a large market buy order is likely soon to arrive at NYSE, Lightning wishes to submit, in place of its cancelled order, a new sell limit order for the same number of shares at a higher price, say at $48.02. If Lightning does so and Smartmoney’s buy order executes against this new higher quote, the HFT will be better off, and Smartmoney worse off, by $.02 per share. [18.48] Note, though, that the HFT will be able to improve its position in this way only if there is room in the NYSE limit order book so that the $48.02 offer price is still more attractive to potential buyers than any other offers with respect to Amgen already posted on NYSE. This may well not be the case. Suppose, for example, that prior to Lightning’s cancellation, the next best offer on the NYSE was 15,000 shares at $48.01 and the best offer after that was 20,000 shares at $48.02. The price and time priority rules would mean that Smartmoney’s buy order would execute against these other two standing offers, not against any new $42.02 offer by Lightning.

[18.49] This cautionary note, though, hides a more critical point: Lightning may wish to cancel its $48.00 sell limit order even if in fact there is no room in the book to improve its position by selling to Smartmoney at a higher price. Recall that to survive in a competitive market, a market-maker like Lightning must set its quotes aggressively enough to attract business, but not so aggressively that the money it makes by buying from, and selling to, uninformed traders is less than what it loses by engaging in such transactions with informed traders. $48.00 was what Lightning calculated at the time it posted its sell limit order to be the optimal price for an offer of 35,000 shares, based on what it knew then about the likelihood of the existence of positive private information. Now, however, Lightning knows something more: a large buy order has transacted on BATS Y. This will cause Lightning to revise upward its assessment of the likelihood that private information suggests that the value of a security is higher than the market previously thought. The upward revision is very possibly great enough that $48.00 is no longer the optimal price at which to offer to sell shares. In that case, Lightning will be better off cancelling its $48.00 limit offer on NYSE.

B. Wealth Transfer Considerations [18.50] To see the distributive effects of electronic front-running, we need to consider how the world would differ if the practice were eliminated. As a first cut for this discussion of the practice’s wealth effects, and for the discussion below of its efficiency effects, we will make the assumption, later relaxed, that there are only three kinds of market participants: HFTs, informed traders who trade on the basis of fundamental value information, and uninformed traders.30 i. Electronic Front-Running Narrows Spreads [18.51] As the analysis of the example makes clear, the practice of electronic front-running by HFTs makes orders by large purchasers and sellers somewhat less anonymous in the sense that the practice allows HFTs to better detect the possibility that informed market orders are headed for their limit orders. If HFTs did not have the ability to learn these things and alter their standing limit orders accordingly, they would know that a larger percentage of the

trades that will execute against their limit orders will come from informed traders. The primary cost of being a liquidity supplier—the losses incurred from dealing with informed traders—would therefore go up. Accordingly, HFTs would increase their initially posted spreads to compensate. [18.52] Going back to our example, if Lightning were not able to electronically front-run, it might have initially posted its limit sell order for 35,000 shares at $48.01 instead of $48.00. For the same reasons, it would also have a lower bid. So if, with electronic front-running being allowed, its bid would have been $47.96, without the practice its bid might instead have been $47.95, with the bid–ask spread, an inverse measure of liquidity, thus rising from four cents to six cents. ii. Electronic Front-Running Helps Uninformed Investors and Hurts Informed Investors [18.53] If electronic front-running were eliminated, uninformed traders and informed traders would each suffer from the resulting larger spreads—the higher offers and lower bids—because for both it would be more expensive to trade. For uninformed traders, that is the end of the story. Informed traders, however, would get a benefit that more than compensates. [18.54] To see why, the starting point again is the fact that the elimination of electronic front-running would make it more difficult for HFTs to detect indications of possible informed trading, and so more informed trades would execute against their quotes. Trading is a zero-sum game. Thus, if HFTs did not increase their spreads in response to the end of the practice, the gains enjoyed by informed investors would just equal the increased losses suffered by HFTs. In fact, however, if electronic front-running is eliminated then HFTs will increase their spreads, and will do so by an amount just sufficient to cover what these losses would otherwise be.31 This is because, as we learned in Section I, the economic pressures on HFTs operating in a competitive market require them to set their spreads at a level such that they just break even. [18.55] The increased spreads will be borne by all traders, informed and uninformed alike, because the HFTs cannot condition their exchange-posted limit orders on the identity of the person who sends the market orders

against which their limit orders execute.32 This means that informed traders come out ahead: the gains they would have enjoyed without the increase in spreads are not fully dissipated by the extra they must pay because the spreads in fact are increased. The rest of what HFTs need to break even comes from uninformed traders, who must pay the increased spread too. [18.56] In sum, without electronic front-running, HFTs would find it harder to detect indications of possible trading on private information and as a result would increase their spreads. Informed traders would get all of the gains from being better able to hide the informed nature of their trades. But they pay, through the increased spreads, only part of the added costs incurred by HFTs as a result of entering into more losing transactions. The rest of these added costs are borne by uninformed investors, who receive no such benefit. So, electronic front-running benefits uninformed investors and harms informed ones. iii. The Ultimate Incidence of Electronic Front-Running [18.57]  Electronic front-running has been regularly attacked as harming ‘ordinary investors’.33 Our analysis, however, suggests that this attack is unmerited. To start, consider retail investors, the paradigmatic ordinary investors. Retail investors generally lack any significant private information and hence are properly assumed to be uninformed. Uninformed investors, as we have just seen, are helped, not hurt, by electronic front-running. [18.58] Most of the persons whose money is invested in index-based mutual funds and pension funds would also presumably count as ordinary investors. These entities too, by definition, are uninformed traders: their purchases and sales are not prompted by any kind of private information; rather they purchase all the stocks in the index when they receive a net inflow of investor funds and sell all stocks in the index when the volume of investor redemptions is sufficient to result in a net outflow of funds. Again, electronic front-running, by narrowing spreads and reducing the cost of trading, helps, not hurts these funds and derivatively their ordinary investors. [18.59] What though about people who invest in managed mutual or pension funds? They too are presumably mostly ordinary persons.34 These

entities do fundamental value research and thus have the potential of being informed investors. The analysis above suggests that electronic frontrunning hurts informed investors. While these funds can be expected to enjoy gains from the elimination of electronic front-running, these gains might well not be passed on to the ordinary people who invest in them. The investment industry and those that work in it each appear to operate in fairly competitive markets. To the extent that these markets are in fact competitive, much of whatever above-market returns are generated by these institutions’ informed trading will be captured in the form of higher fees or salaries for the professionals who make the actual investment decisions.35 This suggests that any gains in these entities’ trading returns that might result from the elimination of electronic front-running are likely to go primarily to increase the fees and salaries of the professionals who make the actual investment decisions, not to the ordinary persons on whose behalf they trade. So even these ordinary investors are not hurt by electronic frontrunning. [18.60] The beneficiaries of electronic front-running, according to the critics of the practice, are the exchanges and the HFTs themselves.36 Here, the critics are closer to the mark. An exchange charges HFTs fees for permitting co-location; namely, the right to place the HFT’s server very near the exchange’s matching engine. If electronic front-running were eliminated tomorrow, HFT co-location facilities would be worth less to the HFTs and they might consequently not be willing to pay as much in fees. This might reduce the rents collected by the exchanges. Any such reduction in rents certainly would hurt the exchanges, at least in the short run. The exchange business, however, has become much more competitive than in the past, making the exchanges’ longer run ability to collect rents questionable. In the longer run, the revenues of firms in a competitive industry can be expected to just equal their costs, including an ordinary market return on capital. Thus, to the extent that the exchange business has in fact become competitive, eventually any revenues lost from co-location fees would need to be made up through higher charges to investors who trade on the exchange. [18.61] Similarly, the lower volume of HFT business that would result from the elimination of electronic front-running would reduce the profits of

firms now in the HFT business and thus lower the value of their existing assets. But in the longer-run future, investments in the industry can be expected to earn a competitive return, with or without the practice. Persons with abilities and skills that are uniquely valuable to the business of HFTs would, however, suffer both a short- and longer-term diminution in their wealth positions from its elimination as rents for these abilities and skills decline.

C. Efficiency Considerations [18.62] Elimination of electronic front-running would have three effects in terms of the efficient operation of the economy, two of which would appear to be efficiency enhancing and one efficiency diminishing. i. Improved Share Price Accuracy [18.63] We have just seen that informed traders would be net gainers from the elimination of electronic front-running. Their cost of trading would go up from the increase in spreads, but this would be more than compensated for by the more advantageous trades they could make with HFTs because of the reduced ability of HFTs to detect indications of possible informed trading. In the simplified world that we are analysing in this first cut at the problem, the only informed traders are persons who trade on fundamental value information. These are the speculative investors that make money by searching out bits of publicly available information, analysing what has been gathered in a sophisticated way, and coming up with a superior estimate of a share issuer’s future cash flows than is implied by the current market price of its shares. Hedge funds and actively managed mutual funds, pension funds and endowments of non-profits are examples of such informed traders. [18.64] Because these informed traders buy when their superior estimate of share value suggests that a stock is underpriced and sell when it indicates a stock is overpriced, their activities make share prices more accurate. The elimination of electronic front-running would make it more profitable for these traders to engage in their activity and so they would do more of it. As a result, prices would be more accurate. As we have seen, more accurate

prices benefit the economy by helping to allocate the economy’s scarce capital to the most promising potential real investment projects and by improving the utilization of the economy’s existing productive capacity through optimizing the signals provided to management about investment decisions and the signals given to boards and shareholders about the quality of management decisions.37 ii. Reduced Resources Going to HFT Activities [18.65] The second positive effect to be gained from eliminating electronic front-running relates to the productive resources that are currently being devoted to undertaking the practice, including the skills and abilities of highly sophisticated technical personnel, advanced computers, and fibre-optic networks. With this infrastructure and human capital no longer needed to support electronic front-running, they would be freed up to increase other productive activities in the economy. iii. Allocation of Resources over Time and Allocation of Risk [18.66]  The elimination of electronic front-running, by widening spreads, would make the market for equities less liquid. This has an unambiguously negative effect on the efficient allocation of resources over time. As we have seen, the prospect that an issuer’s shares will have less liquidity in the secondary trading market increases the issuer’s cost of capital. Just like a tax, illiquidity results in a ‘wedge’ between the value of what the savers (the purchasers of future dollars) expect to receive in the future and what the entrepreneurs or issuers (the suppliers of future dollars in the form of future dividend streams) expect to give up in the future. This blocks transactions that both parties would otherwise have found advantageous if the market for the stock was expected to be more liquid, and hence diminishes economic welfare.38 [18.67] Less liquidity would similarly have an unambiguously negative effect on the efficient allocation of risk. The greater transaction costs deter each investor from adjusting as finely her portfolio when circumstances alter what would be optimal in terms of diversification and suitability to her risk preferences.39

D. Taking Other Kinds of Informed Traders into Account [18.68] The preceding discussion assumed that the only informed traders are ones trading on the basis of fundamental value information. In fact, we know that there are two other types of private information that can give a trader a significant advantage: announcement information and inside information. Taking account of these additional kinds of private information does not change the conclusions above that electronic front-running has positive effects on uninformed investors, as well as on the efficiency with which risk is allocated and resources are allocated over time, nor does it alter the fact that electronic front-running has negative social effects in terms of the real resources that it consumes. It also does not change the conclusion that it has a negative impact on informed traders as a group. But, depending on one’s assessment of the parameters involved, taking account of these additional kinds of private information may well change the conclusion above concerning the impact of electronic front-running on fundamental value information traders and hence of the impact of electronic front-running on price accuracy. [18.69] The issue is as follows. Suppose that electronic front-running is much more helpful at enabling market-makers to respond to trading based on announcement information and inside information than on trading based on fundamental value information. Suppose as well that trading on the basis of these other two kinds of information would, at least absent electronic front-running, constitute a large portion of all informed trading. Then most of the increased trading gains from the elimination of the practice will be enjoyed by traders informed by these other two types of information. If electronic front-running were eliminated, HFTs would need to increase spreads sufficiently to cover their correspondingly increased trading losses, most of which would be due to inside information and announcement information traders. With the elimination of electronic front-running, fundamental value information traders will thus have to pay as much extra per trade from the increased spread as traders on the other two kinds of private information, but will only get a small portion of the additional trading gains. It is thus quite possible that fundamental value information traders will gain less than they pay in increased spread and thus will be hurt by the elimination of the practice.

[18.70] A key factor in determining the likelihood of this possibility is the susceptibility of fundamental value information trading to detection by electronic front-running relative to that of trading on the basis of the other two kinds of private information. Announcement information traders are clearly particularly susceptible because they need to do all of their trading in a very short period of time. They therefore need to engage in larger transactions, which are easier for HFTs to detect and react to. Fundamental value traders, in contrast, may often have days to complete their planned purchases or sales and can break the total amount they wish to transact into small packets that look more like the trades of uninformed traders. But we would need to know much more to resolve the question definitively. Existing empirical research is not very enlightening concerning several other important factors: the proportion of informed trades based on each of the three kinds of private information, the average value of the information associated with each, and the exact sensitivity of the trading patterns associated with each of the three kinds of informed traders to detection by electronic front-running. [18.71] If further empirical research ultimately suggests that electronic front-running actually helps, not hurts, fundamental value information trading, it would suggest that the practice, contrary to our earlier analysis, actually helps share price accuracy by making the business of fundamental value information trading more rewarding. In contrast, announcement information trading is not important in terms of the social benefits that are derived from share price accuracy because the information will be reflected in price very quickly even without the trading. For similar reasons, inside information trading is likely not socially useful either.40 In sum, if electronic front-running helps fundamental value-informed trading, the practice enhances both price accuracy and liquidity and is thus an unambiguously socially positive activity, assuming that the value of the resulting price accuracy and liquidity gains exceeds the cost of the real resources consumed by the activity.

2. Slow Market Arbitrage

[18.72] In the United States, slow market arbitrage can occur when an HFT has posted a quote representing the NBO or NBB on one exchange, and subsequently someone else posts on a second exchange an even better quote, which the HFT learns of before it is reported by the national system. If, in the short time before the national report updates, a marketable order arrives at the first exchange, the order will transact against the HFT’s now stale quote. The HFT, using its speed, can then make a riskless profit by turning around and transacting against the better quote on the second exchange. [18.73] Slow market arbitrage was a target of criticism in Flash Boys,41 which in turn reflected growing discontent among commentators in the years preceding the book’s publication.42

A. An Example [18.74] To understand the practice in more detail, let us return to our HFT, Lightning. Suppose that Lightning has a limit sell order for 1,000 shares of IBM at $161.15 posted on NYSE. This quote represents the NBO at that moment. Mr. Lowprice then posts a new 1,000 share sell limit order for IBM on EDGE for $161.13. [18.75] The national reporting system is a bit slow, and so a short period of time elapses before it reports Lowprice’s new, better offer. Lightning’s co-location facility at EDGE very quickly learns of the new $161.13 offer, however, and an algorithm sends an ultra-fast message to Lightning’s colocation facility at NYSE informing it of the new offer. During the reporting gap, Lightning keeps posted its $161.15 offer. Next, Ms. Stumble sends a marketable buy order to NYSE for 1,000 IBM shares. Lightning’s $161.15 offer remains the official NBO, and so Stumble’s order transacts against it. Lightning’s co-location facility at NYSE then sends an ultra-fast message to the one at EDGE instructing it to submit a 1,000-share marketable buy order there. This buy order transacts against Lowprice’s $161.13 offer. Thus, within the short period before the new $161.13 offer is publicly reported, Lightning has been able to sell 1,000 IBM shares at $161.15 and purchase them at $161.13, for what appears to be a $20 profit.

[18.76] It is worth noting that the first step in this story—Lowprice’s posting of the $161.13 offer on EDGE—does not guarantee that Lightning can make this profit. No marketable buy order may arrive at NYSE during the reporting gap. Also, even if one does, by the time Lightning is able to submit its marketable buy order at EDGE, some other person may already have submitted a marketable buy order to EDGE that picks off the $161.13 offer. This becomes particularly likely if, as is the case in the real world, there are a number of HFTs besides Lightning with co-location facilities at EDGE and at the other exchanges. Depending on the nature of their own respective offers posted on various exchanges, one or more of these other HFTs may be competing with Lightning to pick off the one $161.13 offer.

B. Wealth Transfer Effects [18.77] Who is helped and who is hurt in the example above and what are the larger distributive consequences with slow market arbitrage as an ongoing practice? In the example, the first thing to note is that Ms. Stumble, the person who, during the reporting gap, submits the marketable order that transacts against Lightning’s stale $161.15 offer, is not harmed by Lightning’s slow market arbitrage activities. Stumble would have suffered the same fate if Lightning had not engaged in slow market arbitrage because that course of action would have also left the $161.15 offer posted on NYSE and so Stumble’s buy order would still have transacted against it. [18.78] Still, someone must be worse off: Lightning is better off than if it had not engaged in the slow market arbitrage, and trading is a zero-sum game. To see who this worse-off person might be, consider first why Lightning is better off. Lightning is in the business of buying and selling shares, not holding on to long or short positions for any significant period of time. So it needs to reverse quickly each transaction it enters. Here, it sold shares when Stumble’s order transacted against Lightning’s $161.15 offer on NYSE. To reverse this transaction, Lightning needed to buy shares. By engaging in slow market arbitrage, it did so by seizing the best offer in the market—Lowprice’s $161.13 offer on EDGE—before others in the market even knew the offer was available. If Lightning had not detected this new offer ahead of others and seized it, Lightning’s reversal of the situation

would occur through posting a bid that a marketable order transacts against. We know from Section I that the sale of the shares at $161.15 and their repurchase at this newly posted bid would each, on an expected basis, be a break-even transaction. By successfully engaging in slow market arbitrage, Lightning instead made a certain $.02 profit per share sold and purchased. [18.79] To figure out who is hurt from Lightning engaging in slow market arbitrage—that is, detecting the $161.13 offer and seizing it—consider who would have been better off if Lightning had posted a new buy limit order instead of seizing Lowprice’s $161.13 offer. The person or persons helped would come from one of two groups of potential liquidity takers. One group is potential sellers who submit marketable sell orders; the posted bid that Lightning would need filled would improve the terms for the marginal seller. The other group is potential buyers who submit marketable buy orders; the opportunity for members of this group to seize Lowprice’s $161.13 offer, which was better than anything else available in the market at the time, would improve terms for the marginal buyer. [18.80] The results from this example can be generalized. The persons who are hurt by HFTs engaging in the practice of slow market arbitrage on an ongoing basis are regular traders, both informed and uninformed.43 In contrast to electronic front-running, where the practice decreases the effective cost of trading for uninformed traders but increases it for informed traders, slow market arbitrage increases the effective cost of trading for all regular traders.

C. Efficiency Considerations [18.81] In most situations, arbitrage activities, at least if they do not consume any real resources, have positive economic welfare effects. The actions of arbitrageurs equilibrate prices across two markets, each of which has its own group of potential participants, and as a result, presumptively welfare enhancing transactions are entered into that otherwise would not have occurred. However, as the example shows, slow market arbitrage has little in common with ordinary arbitrage. Slow market arbitrage adds a third party, the liquidity supplier, whose only socially valuable purpose is to

facilitate trades between regular traders, but who instead is the only gainer from the so-called arbitrage activity. Regular traders, both informed and uninformed, are in fact losers because their cost of trading goes up. So the normal presumption in favour of activities carrying the label ‘arbitrage’ does not apply here. [18.82] In fact, even if slow market arbitrage consumed no real resources, it would have an unambiguously negative impact on welfare. Consider first the effect of the increased effective cost of trading for fundamental valueinformed traders. Slow market arbitrage, by raising the effective cost of trading for such informed traders, makes it less rewarding to seek out bits of publicly available information and analyse their implications in a sophisticated way. This reduces share price accuracy, which, as we have seen, would in turn have negative effects on the allocation of capital for new real investment projects and the efficient utilization of existing productive capacity. As for the increased effective cost of trading of uninformed traders, it has the now familiar negative effects on the efficient allocation of resources over time and on the efficient allocation of risk.44 [18.83] Slow market arbitrage does in fact consume real resources, which is another efficiency consideration. If it were the only HFT practice dependent on co-location facilities and ultra-fast connections, it would use substantial amounts of real resources that could otherwise be usefully employed increasing the production of other goods and services. If HFTs were to continue the practice of electronic front-running, however, the marginal cost in real resources of engaging in slow market arbitrage as well would probably be fairly low.

3. High-Frequency Trading and Volatility [18.84] A different, but also important, strain of criticism of HFTs has alleged a causal connection between HFT activity and greater volatility in equity markets.45

A. General Increase in Volatility [18.85] One criticism is that HFTs have made the US markets more volatile on an ongoing day-to-day basis. Michael Lewis, in Flash Boys, for example, asserts that the intra-day price volatility of the stock market was 40 per cent greater between 2010 and 2013 than it was between 2004 and 2006, and associates this change with the enactment of Reg. NMS and the rise of HFTs.46 [18.86] Lewis, however, is comparing an unusually volatile three-year period with an unusually stable two-year period. A better comparison sample would be 2012 to the present, which shows market volatility that is generally lower than the 1990s and early 2000s, despite the greatly increased role of HFTs in the latter period.47 As far as one can tell so far, there is no serious evidence showing a causal link between the rise of HFTs and day-to-day increased volatility, nor any well-articulated theoretical basis for expecting such a link. Instead, the majority of academic evidence on the subject suggests that HFTs reduce volatility.48

B. The Flash Crash [18.87] More interesting is a second claim: that HFTs occasionally exacerbate volatility in a very extreme manner when there has been some kind of disruption in the market, such as the infamous 6 May 2010 ‘Flash Crash.’ The Flash Crash occurred within a window of less than thirty minutes, during which the Dow Jones Industrial Average (‘DJIA’) dropped about 1,000 points, losing 9 per cent of its value, and then recovered almost its entire loss.49 Accenture fell from trading at $39.98 at 2:46 to one cent at 2:49, only to return to $39.51 by 2:50.50 Apple, on the other hand, at one moment traded for almost $100,000 per share.51 [18.88] The Flash Crash was widely taken to ‘highlight […] the risks of electronic trading’ as NYSE’s then head of operations suggested.52 However, the report eventually issued by federal regulators explained the Flash Crash not as the result of HFT predation, but as the result of a liquidity crisis caused by a large sell order that triggered a flight of liquidity

from the market. This flight involved HFTs, but only in the sense that many market-making HFTs left the market in response to the large sell order. This temporary disappearance of the HFTs removed substantial liquidity.53 [18.89] The crucial question is: Why would a large market sell order trigger a flight by HFTs, when the business of HFTs is to provide liquidity to persons submitting marketable orders? The answer to this question returns us to the overarching theme of this chapter: the role of adverse selection in shaping the provision of liquidity.54 A large, aggressive sell (or buy) order suggests to liquidity providers that the order submitter may have important private information. HFTs know that if this apparent private information in fact turns out to exist, then they will lose money from trading with that order and so they will widen their spreads.55 If the threat of being adversely selected by the order becomes extreme enough, many or all liquidity providers will, to seek safety, temporarily exit from the market altogether and prices will, as a result, fluctuate widely in the absence of quotes reflecting any plausible estimate of a security’s fundamental value.56 This, in essence, is what happened on a large scale during the Flash Crash. [18.90] The behaviour of HFTs during the Flash Crash was not predatory; it was simply unheroic. Perceiving the large sell order to have a higher probability of being motivated by private information, given its size and aggressiveness,57 HFTs removed their quotes to minimize their trading losses, and liquidated the long positions they had accumulated, exacerbating pressures on price declines.58 Because HFTs provide a large share of liquidity, in their absence, a security’s only remaining quotes lay far from its true value.59

C. Wealth-Transfer Considerations [18.91] Assessing the wealth transfers resulting from gyrations, such as in the Flash Crash, is equivalent to asking who wins and loses when HFTs stop providing liquidity. Despite suggestions by critics of predatory behaviour, HFTs cannot make money if they do not trade. Among traders, the losers are persons who put in market sell orders for stocks that temporarily went a long way down and market buy orders for stocks that

temporarily went a long way up. The winners were the persons who posted previously way-out-of-the-money limit orders against which these market orders transacted.

D. Efficiency Considerations [18.92] Events such as the Flash Crash seem bound to occur from time to time with an HFT-dominated system for providing liquidity. The old NYSE specialist system, where the specialist was supposed to ‘lean against the wind’ to provide liquidity may have been less prone to such problems. So perhaps was the dealer system more generally, where human beings made the trading decisions. [18.93] These occasional brief moments of total collapse of liquidity do not really seem very important in terms of our touchstones for efficiency, however. Very brief sharp deviations of share prices from fundamental values do not seriously undermine the role of share prices in aiding the efficiency with which capital is allocated to new real investment projects and with which existing productive capacity is utilized. It is accuracy most of the time that matters. And investors can protect themselves from extreme results by refraining from submitting market orders, and instead using limit orders that would appear to make them marketable but that protect against the rare moment where the best quote is well away from the most recent posted transaction. They can also stay briefly out of the market without seriously undermining the efficient allocation of resources over time or the efficient allocation of risk. The modern stock market’s overall performance in terms of liquidity provision and operational costs is far better than the market of the past, which matters for the ultimate social goals promoted by a well-functioning equity market much more than avoiding the occasional Flash Crash-type disruption.

III. Dark Pools and Internalization [18.94] Dark pools have been the subject of considerable negative commentary in the United States, just as HFTs have. Some of this arises

from simple suspicion arising from their lack of pre-trade transparency. Some of the negative commentary arises from a concern that dark pools undermine equity market price discovery; a concern that has been expressed through the whole MiFID II process as well. Finally, some of the negative commentary relates to allegations that the operators of dark pool break their promises to customers to the operators’ advantage and the customer’s disadvantage. [18.95] Internalization involves a retail broker that either (i) matches its incoming buy orders with nearly simultaneous incoming sell orders, buying from the sellers and selling to the buyers at prices (usually very modestly) better than the NBB and NBO, or (ii) sells its order flow (i.e. receives payment for order flow) to a freestanding internalizer that does the same kind of matching. Internalization raises some parallel issues to dark pools and these will also be discussed briefly.

1. Understanding the Function of Dark Pools [18.96] A dark pool, like an exchange, is typically an electronic limit order book, but, unlike an exchange, it does not publicly reveal the limit orders that are posted on it.60 Moreover, it has the ability to restrict who can post limit orders and submit marketable orders.61 In the traditional dark pool arrangement, a trader will submit to a dark pool a ‘mid-point’ limit buy or sell order, the terms of which are that it will execute against the next marketable order with the opposite interest to arrive at the pool and will do so at a price equal to the mid-point between the best publicly reported bid and offer at the time of execution. [18.97] Dark pools, despite their nefarious-sounding moniker, can, if honestly operated, provide services that are very useful to their customers. In the United States, they arose because of the more liberal regulatory environment established by the NMS Amendments to the Exchange Act and the information technology revolution. MiFID I established a similarly more liberal regulatory environment in the EU. The key force driving their rise—as with so many other institutions and practices within the new stock market—was concern to mitigate adverse selection. A dark pool’s most

valuable characteristic, from this perspective, is to provide a venue where uninformed buyers and sellers, seeking to trade substantial amounts of stock, can minimize the movement of prices against them and transact at prices potentially much better than the NBO and NBB.62 These advantages arise because the mid-point limit orders posted on dark pools are not disclosed and do not generate public displayed quotes and because dark pool operators have the ability to exclude traders. In terms of serving these functions, the ideal dark pool would be one where both the parties posting mid-point limit orders and parties sending in marketable orders are completely uninformed. The mid-point is a substantially better price for the buyer than the NBO, and it is the same for the seller relative to the NBB. The system begins to break down to the extent that the parties posting limit orders, or those submitting marketable orders, are in fact informed. [18.98] When the limit order poster is informed, the breakdown is because its counterparties—the parties submitting marketable orders— would be disadvantaged by being in a dark pool since they would not be able to see from the size of the posted limit orders that there might be an informed party on the other side. With regard to the informed marketableorder submitter, the breakdown is because its order will only transact against the limit orders in the dark pool when the submitter’s information suggests that the mid-point is a price that makes the transaction advantageous to it, which means it is a price that makes the transaction disadvantageous to the person posting the limit offer against which the marketable order transacts. Thus, the dark pool operator provides a service to the extent that it can effectively monitor both the parties posting the midpoint limit orders and the parties sending in marketable orders. This assures each side that the other side is relatively unlikely to be informed.63

2. The Question of Whether Even Honestly Operated ‘Ideal’ Dark Pools Are Socially Desirable [18.99] Suppose that all dark pools were of the ideal type, each with an honest operator that was so competent that it was totally effective at

excluding informed traders from among those who post either mid-point limit orders or marketable orders. Would even these dark pools be socially desirable? They certainly would not raise any apparent fairness concerns. They also would benefit both sides of every executed transaction by getting each side a better price than the best prices available on the exchanges, that is, by lowering the cost of trading. Still, whether the presence of these dark pools would enhance or diminish efficiency in the larger economy is an open question. [18.100] On one hand, lowering the cost of trading for the uninformed traders using the dark pools will, for the reasons discussed in Section I, increase the efficiency of both the allocation of resources over time and the allocation of risk in the economy. On the other hand, the availability of such dark pools would, with their lower costs of trading, draw uninformed investors away from trading on the exchanges. When fewer uninformed investors trade on exchanges, the spreads are wider.64 The uninformed investors trading on exchanges subsidize informed investors and this subsidy encourages those in the business of generating and trading on fundamental value information, which increases price accuracy that in turn enhances efficiency in the production of goods and services in the real economy.65

3. Practices Damaging Traders Using Dark Pools [18.101] If dark pools are permitted, they are not useful to uninformed traders in the way that an ideal dark pool is supposed to be to the extent that their counterparties turn out to be informed traders. In the United States, large investment banks, which are both important providers of brokerage services and operators of most of the largest dark pools,66 are accused of routing their brokerage customers’ orders to the banks’ own dark pools, even when the orders will receive inferior execution there.67 Related to this first practice is the claim that it is common for a dark pool operator to misrepresent the nature of other parties trading in its pool in order to induce brokerage customers to agree to have their orders sent to this pool.

[18.102] To the extent that a dark pool does not function in accordance with the ideal described above, an order that is sent there may execute at less desirable terms than if it were sent to another venue. An investment bank that operates a dark pool has intimate knowledge of the extent to which it in fact falls short of this ideal. If a brokerage unit of an investment bank sends a trader’s order to the bank’s own dark pool when the broker knows, or should know, that the order would receive superior execution elsewhere, the bank gains from the extra volume of trade in its dark pool and in other possible ways, and the customer loses from the inferior terms of execution. The same result is also likely if the bank operating the dark pool misrepresents to customers the nature of the parties allowed to trade on the bank’s dark pool, in order to create the impression that there is less danger of informed counterparties there than is in fact the case. Such a misrepresentation is likely to attract orders that could execute on better terms elsewhere. All of these results generalize if these failures are common practices: they make investment banks richer and traders poorer. Ironically, if the calculation is that ideal dark pools would do more damage to efficiency through their negative effects on price accuracy than enhance efficiency through lower-cost trading-induced improvements in the efficiency of the allocation of resources over time and allocation of risk, these practices might actually enhance the overall efficiency of the economy. This is because a growing awareness of these improper practices would scare uninformed investors away from the dark pools, who would instead trade on exchanges, thereby lowering spreads and increasing fundamental value-informed trading. However, even if this were to be the case, it would not be desirable to abandon the traditional rules of upright commercial practice that are breached by these practices. There are other ways to achieve the same thing, for example by prohibiting or limiting dark pools.68

4. Internalization [18.103] Internalization raises similar issues. Retail order flow is properly regarded as uninformed. Thus, in an ‘ideal’ internalization arrangement, there is no reason for any adverse-selection-induced spread between what the seller receives and what the buyer pays, the way there is between the bid

and the ask on the exchange. Either the seller should receive approximately what the buyer pays, or, if there is a significant difference between what is received and what is paid, the difference should be split between the two in the form of reduction in brokerage commissions. Again, the uninformed traders’ lower cost of trading with an ideal internalization arrangement will increase the efficiency of both the allocation of resources over time and the allocation of risk in the economy. However, internalization again draws uninformed investors away from trading on the exchanges, thereby widening spreads which in turn discourages fundamental value informed trading with the consequent reduction in share price accuracy and efficiency losses in the real economy. If the efficiency gains from the one effect are less than the efficiency losses from the other, ideal internalization is again undesirable. [18.104] Internalization as it is actually practiced in the United States may not come close to this ideal in any event. The traders probably truly are uninformed. But we know that what the seller receives is typically only slightly more than the NBB and what the buyer pays is only slightly less than the NBO. It is not clear that brokers pass on most of the difference between what is received and what is paid in the form of reduced commissions.

IV. Conclusions: Reflections on Issues Raised by MiFID II [18.105] A comparison of the EU and US market structure rules, and the concerns that generated them, suggests three key differences. Relative to the United States, the EU shows (i) more concern with having an effective price-formation process, (ii) more concern with the possibility that HFTs contribute to price instability and engage in market abuse, (iii) less concern with promoting competition among trading venues. These differences characterized the MiFID I era and are reflected in MiFID II as well, although MiFID II does evince somewhat greater concern about competition among trading venues than was true before.

1. Dark Pools and an Effective Price-Formation Process [18.106] Discussion leading up to MiFID II and the rules themselves evince great concern with maintaining what officials term ‘effective price formation processes’.69 This concern is reflected in a limit placed on waivers to the general rules requiring pre-trade transparency of bids and offers. This limit is the main tool in an effort to steer more trading from dark pools to ‘lit’ venues that are regulated markets.70 Specifically, MiFIR Article 5 provides that a venue cannot obtain a waiver from the requirement that it display the mid-point limit orders posted on it in a given stock if either the trading on the venue represents more than 4 per cent of total EUwide trading in the stock or unlit trading across the EU represents more than 8 per cent of the total EU-wide trading in the stock. There is no equivalent provision in the US regime. [18.107] Dark pool mid-point trading poses two potential challenges for price formation: it ‘free rides’ on prices established in a lit, non-mid-point market, and it hides the existence of interest by the traders posting orders in such venues in buying or selling the stock. Consider these two challenges first in connection with what I have referred to as an ‘ideal’ dark pool, where the parties posting the mid-point limit orders, and the parties sending in marketable orders, are both completely uninformed.71 Suppose such dark pools were completely eliminated, so that participants who otherwise would have sent orders to dark pools, to the extent they were still interested in trading, would need to send their orders to lit, non-mid-point venues. The prices established in these lit venues would gain nothing in informedness from orders sent in by these participants because they are uninformed. Indeed, without any ideal dark pools, if, in a given period, there is some kind of order imbalance because an uninformed investor wanted to make a large purchase or sale, the price would deviate temporarily from what the information in the market would suggest. This is because professional liquidity suppliers would calculate that the imbalance suggests a certain percentage chance that there is informed trading going on when in fact there is not.72

[18.108] Now consider less than ‘ideal’ dark pools, where the policing by their operators, purposefully or unintentionally, is not fully effective at excluding informed traders from posting orders. Suppose these less than ideal dark pools were completely eliminated, so that participants who otherwise would have sent orders to them, to the extent that the participants still wanted to trade, would need to send their orders to lit, non-mid-point venues. This might well cause prices to fully reflect the information possessed by informed traders more quickly. This is primarily because the banned, less than ideal dark pools would have been hiding informed buying or selling interest in the form of the unreported posted mid-point buy or sell limit orders. The improvement in the time by which the information possessed by informed traders is reflected in price would be very small, however, because dark pools are as transparent as lit ones in publicly reporting executed transactions. So evidence of the buying or selling interest is not hidden for long. [18.109] In addition, to the extent that banning the less than ideal dark pools leads to more concentrated trading on each of the lit venues (as opposed to trading over more lit venues), there might be a brief improvement in price accuracy because there would be fewer situations where transactions occur at about the same time in different venues at different prices. Again, the improvement in the time by which the information possessed by informed traders is reflected in price would be very small, however, because these differences would vanish quickly due to arbitrage and information flow across markets. [18.110] Whether the MiFIR Article 5 caps, once they are imposed, really do improve the speed with which informed traders’ information is reflected in price should be testable empirically. When information takes time to seep into prices, they become autocorrelated. If the caps reduce this amount of time, prices should become less autocorrelated. Any such reduction can be measured as follows. If there is no autocorrelation, the variances of return for time periods of different lengths should be proportional to these lengths of time. The extent to which they are not proportional is thus a measure of the extent of autocorrelation and hence a measure of short-term price inefficiency.73 If the ratios of various intra-day and other short-term period return variance ratios come closer to being proportional after the caps are

imposed, it will suggest that they resulted in an improvement in the rate at which prices reflect information possessed by informed traders.74 [18.111] Even if tests show there is an improvement, however, it is not immediately obvious that it will have led to any significant social welfare gain from the gain in price accuracy. A very brief delay—minutes, hours, or at most a few days—in the full absorption of new information into price is not the kind of improvement in price accuracy that matters for the real economy. The important causal links between price accuracy and efficiency in the real economy relate to the efficiency with which the market allocates capital for real investment and to better reductions in the agency costs of management through better use of firms’ internal cash flows and better operation of their existing productive capacity.75 In sum, the MiFID II emphasis on effective price formation of the type being discussed here may be misplaced because very short-term price efficiency is arguably not very important. [18.112] Interestingly, however, to the extent that it is in fact uninformed traders that use dark pools—a phenomenon that has no negative effect on short-term price efficiency—the caps are likely to improve the longer-term type of price accuracy that does matter to the real economy. The caps will drive more of this uninformed trading into non-mid-point lit venues. An increase in uninformed trading in such venues will, as explored in the discussion above concerning dark pools, narrow the bid–ask spreads demanded by professional liquidity suppliers, and, by lowering the cost of doing fundamental value-informed trading, lead to the generation of more such information.76 The social welfare question, again, would be whether the welfare gains from the resulting improvement in price accuracy would dominate the welfare losses from the reduction in liquidity as a result of uninformed investors needing to buy at the offer and sell at the bid rather than being able to trade at the mid-point.

2. HFTs: Price Volatility and Market Abuse [18.113] Professor Pierre-Henri Conac reports in his chapter in this volume that ‘there is strong disagreement among national regulators on

HFT’, with regulators in the Netherlands having a positive attitude and those in France and Germany taking the position that HFT’s ‘efficiency gains are marginal while the risks of market abuse are high’.77 Professor Conac also suggests, in contrast, a consensus among regulators exists that algorithmic trading generally, of which HFT is a subcategory, can create systemic risk.78 Against this background, we see that MiFID II, as finally implemented, will require HFTs to be licensed, to keep comprehensive records of all orders that they place, to have effective systems of risk controls to avoid disorderly markets, and, if acting as market makers, to do so on a continuous basis with two-sided quotes for a specified portion of trading hours. It also allows Member States to require disclosure to authorities of algorithmic trading strategies and does not prohibit Member State regulators from interpreting the concept of market abuse broadly. There are thus two themes relating to the EU reactions to HFTs: volatility and market abuse.

A. Price Volatility [18.114] The discussion above concerning the US experience with HFTs suggests that there is no serious evidence showing a causal link between HFTs and increased day-to-day volatility: markets are in fact less volatile today than in the late 1990s and early 2000s, the period before the rise of HFTs.79 As for occasional extreme volatility events such as the 2010 Flash Crash, the evidence is that HFTs did not precipitate the Flash Crash; rather a very large sell order did. HFTs temporarily pulled out of the market in response to this order, however, thereby removing much of the market’s liquidity, with a resulting severe swing in prices for a brief time. [18.115] The MiFID II record-keeping provisions seem sensible and unobjectionable, as does the requirement of an effective system of risk control, at least as long as what is required is not so burdensome that it prevents HFTs from efficiently providing their fundamental liquidity supply function (described in Sections I and II). Requirements that HFTs disclose their algorithms to authorities may be unobjectionable as well, but they raise two concerns. One is whether the disclosure of the algorithms is the first step in regulating their use, in which case there is again the worry that,

because of regulator misunderstanding or the pressure of vested interests, such regulation will interfere with HFTs efficiently providing liquidity. The other is whether the disclosures are kept sufficiently confidential to avoid the disincentives to their development that would come from copying by competitors. [18.116] Requirements that HFTs that generally play a market-making role by providing two-sided quotes for a certain percentage of trading hours also seem unobjectionable. It should be realized, though, that if these requirements are such that an HFT will still be in compliance if it immediately changes its two-sided quotes radically when a large uncertainty arises, the requirement will do little to buffer Flash Crash-type price swings. If, instead, HFTs are expected to maintain two-sided quotes that are restricted in how rapidly they can change in the face of major uncertainties, HFTs, to stay in business, will need to quote wider spreads during normal times to make up for the expected losses from being constrained in changing their quotes when large uncertainties occur. Moreover, the October 1987 market break experience with specialists on the NYSE, who did have the opportunity to make above supernormal profits during ordinary times, suggests that an obligation not to change quotes too quickly is a difficult one to enforce: they only ‘leaned against the wind’ so much. [18.117] The bottom line is that it would be wise not to be aggressive in implementing HFT regulations aimed at avoiding price volatility. HFTs do not appear to contribute to day-to-day volatility and apparently only contribute to Flash Crash-type brief price spikes in the sense that they rationally withdraw from providing liquidity in the face of major uncertainties. The requirement to maintain two-way quotes is likely to be either ineffective at preventing such spikes, or to be somewhat effective but at the cost of reduced liquidity during ordinary times in the form of wider spreads. The other requirements may be unobjectionable and may possibly help avoid such brief price spikes, but, if pursued aggressively, could also impede HFTs in efficiently providing their fundamental liquidity supply function described in Sections I and II without any counterbalancing significant social gain.

B. Market Abuse [18.118] As discussed by Professor Conac in Chapter 17 in this volume, there are a variety of manipulative trading strategies that are made easier to implement by a trader with the technological capabilities of an HFT. These include quote stuffing, momentum ignition, and layering and spoofing.80 Because trading is a zero-sum game, these strategies, to the extent that they are profitably utilized by an HFT, add to the cost of trading by all other participants and they serve no socially useful purpose. Thus they are socially undesirable. These practices are the subject of rules intended to deter them pursuant to MiFID I, the Market Abuse Directive (‘MAD’), and the ESMA 2012 Guidelines 5 and 6, which in turn are supplemented by MiFID II rules, under Article 17(1), requiring investment firms using algorithmic trading to have controls on their systems to prevent their use to undertake such strategies.81 [18.119] There are dangers, however, that the term ‘market abuse’ will be so broadly construed that it reaches strategies that are an integral part of the methods, as described in Sections I and II, by which HFTs efficiently supply liquidity. A case in point is the December 2015 decision by the Enforcement Committee of the French Autorité des Marchés Financiers (AMF), which became the basis of sanctions being imposed on a major HFT, Virtu, and on Euronext.82 Among other things, the Committee found, for a group of twenty-seven CAC 40 stocks (stocks constituting a representative index of French large cap stocks), that Virtu was responsible for an extremely high proportion of all messages placing or cancelling orders on Euronext Paris relative to the number of trades executed there. Specifically, the Committee found that Virtu was responsible for 62.7 per cent of all order placement and cancellations messages sent to the venue but only 2 per cent of the venue’s executed trades. The Committee concluded that the large volume of Virtu limit orders that were cancelled before they could be executed against distorted the representation of the order book, particularly because, by the time that other traders saw the orders, they had often been cancelled. [18.120] The AMF decision in the Virtu case appears to represent a fundamental misunderstanding of the way HFTs work to provide liquidity

to the market. As explained in Sections I and II, HFTs, in order to minimize their exposure to adverse selection, adjust their orders (by cancelling old orders and submitting substitute new ones) constantly as they receive information about executed transactions and changes in the quotes posted by others. The reduction in losing transactions for HFTs resulting from these constant adjustments reduces their cost of doing business, which, in a competitive market for liquidity supply, is passed on to traders in the form of narrower spreads. Narrower spreads reduce the costs of trading, quite the opposite of the effects of successful quote stuffing, momentum ignition, and layering and spoofing, which increase the cost of trading for others. [18.121] Professor Conac characterizes the decision as ‘equivalent to a de facto ban on HFT in France’.83 If this is really true, or even if HFT activities in France are, for the reasons discussed here, at least significantly hampered, the situation presents an interesting opportunity for an empirical test of whether, as argued here, HFTs do in fact contribute to liquidity in Europe. The issue for investigation would be whether spreads widened and depth of book declined on the Euronext Paris after the decision and did so in a way that suggests a causal connection between the two.

3. Trading Venue Competition [18.122] When a given stock trades in multiple venues, the resulting competition is likely to generate the upsides of greater cost efficiency and a higher rate of innovation. Multiple venue trading, however, has the downside that orders from potential traders are fragmented among these venues, which means that it is less likely that a buyer willing to pay a given price will find a seller willing to sell at that price. The vast improvements in information technology over the last couple of decades, however, have offered the potential for a significant reduction in this downside, by making it easier for traders to see what is going on in each of these venues. [18.123] In the United States, Congress, in its adoption in 1975 of the National Market System (‘NMS’) amendments to the Exchange Act, foresaw that improving information technology could significantly reduce this downside. The NMS amendments pushed the system to develop in the

direction of having multiple trading venues for each stock; a push that has since been consistently supported by the SEC, culminating in the adoption of the Regulation NMS Rules in 2005.84 The effect has been dramatic. For example, twenty years ago, 83 per cent of all the trading volume in NYSE stocks occurred on the NYSE, whereas today the figure on NYSE venues is about 24 per cent.85 [18.124] MiFID I was also intended to promote competition in trading venues.86 It has not been accompanied by as dramatic a deconcentrating effect as the reforms to the US trading regulation regime, however. For example, in May 2016, the London Exchange still had 49 per cent of the trading volume of FTSE 100 stocks. Similarly, Euronext Paris had 59 per cent of the CAC 40; the Bolsa de Madrid had 67 per cent of the IBEX 35; the Frankfurt Exchange had 57 per cent of the DAX.87 [18.125] Professor Giovanni Petrella suggests that the difference in the extent of deconcentration between the United States and the EU is due to contrasting regulations in three areas.88 First, MiFID I does not have an equivalent to NMS Rule 611, which requires an exchange that does not have the nationally best quote for such stock and that receives a marketable order for a stock where it does not have the nationally best quote to forward the order to the venue that does have the NBO or NBB. Second, MiFID I has no equivalent to NMS Rules 602 and 603, which require, respectively, the consolidation of the best quotes available on each exchange into a single tape and the consolidation of executed transactions on each exchange into a single tape. Finally, MiFID I has no equivalent to two NMS rules that permit monitoring of how well trading venues and brokers are doing their job: NMS Rule 605, which requires trading venues to periodically disclose certain statistics indicative of the average quality of execution on the venue, and NMS Rule 606, which requires brokers to show which venues they send customer orders to. [18.126] The lower level of regulation in the EU with respect to each of these three regulatory areas increases the likelihood that a trader will decide to send her order to the largest exchange. For a marketable order, the largest exchange is where the most non-marketable limit orders have been posted and so, in the absence of readily available information about other venues,

the fact that a venue is the largest means that there is the largest chance that it is the one whose posted limit orders include the one with the best available terms. For a non-marketable limit order offering the best terms available in the system, in the absence of a mechanism to steer marketable orders in its direction, the largest exchange is where it is most likely to be found by a marketable order and execute. [18.127] MiFID II moves in the US direction with respect to providing for a consolidated tape, but does not provide equivalents to US rules in the two other regards. Perhaps the decision not to undertake the other two reforms represents a judgment that the lack of a consolidated tape is the most important impediment to deconcentration and that the other two regulatory reforms would have costs greater than the benefits from whatever further deconcentration they might bring. Or perhaps the decision is simply the result of there being less of a political consensus in the EU as regards genuinely favouring deconcentration in the first place. In any event, the technology in the United States and the EU is the same and so going forward we will have an opportunity to see how much the remaining regulatory differences matter.

1

Parts of this chapter are adapted from an earlier article concerning US equity markets, ‘The New Stock Market: Sense and Nonsense’ (2015) Duke L J 65, 191, which I coauthored with Lawrence R. Glosten and Gabriel V. Rauterberg. 2 See James J. Angel, Lawrence E. Harris, and Chester S. Spatt, Equity Trading in the 21st Century: An Update (2013) 11–12,

(showing significant increases in the speed of execution, decreases in the bid–ask spread, decreases in commissions, and increases in the number of quotes per minute); see also James J. Angel, Lawrence E. Harris, and Chester S. Spatt, Equity Trading in the 21st Century (2010), . 3 See Chapter 3 in this volume. 4 Michael Lewis, Flash Boys: A Wall Street Revolt (New York: W. W. Norton, 2014). 5 Charles Schwab, founder of the well-known brokerage firm bearing his name, recently suggested, e.g., that ‘[h]igh-frequency traders are gaming the system, reaping billions in the process and undermining investor confidence in the fairness of the markets. […] It’s a growing cancer and needs to be addressed’. Steven Russolillo, ‘Schwab on HFT: “Growing

Cancer” That Must Be Addressed’, Wall S J, 3 April 2014, . 6 Sam Mamudi, ‘UBS Hit With Record Dark Pool Fine for Breaking US Rules’, Bloomberg, 15 January 2015, (imposing on UBS the largest fine ever imposed on a dark pool operator); Sam Mamudi, ‘Dark Pools Opening Up Amid Increased Scrutiny’, Bloomberg Business Week, 20 May 2014, (reporting on industry unease with dark pools). 7 See Lawrence E. Harris, Trading and Exchanges (Oxford: Oxford University Press, 2002) 158 (discussing analyses indicating that in most markets adverse selection accounts for the majority of the bid–ask spread). 8 The seminal microstructure economics models of this process are Lawrence R. Glosten and Paul R. Milgrom, ‘Bid, Ask and Transaction Prices in a Specialist Market with Heterogeneously Informed Traders’ (1985) J. Fin. Econ. 14, 71; Albert S. Kyle ‘Continuous Auctions and Insider Trading (1985) Econometrica 53, 1315. 9 See generally George A. Akerlof, ‘The Market for “Lemons”: Quality Uncertainty and the Market Mechanism’ (1970) Q.J. Econ. 84, 488 (analysing how informational asymmetries can drive declines in the quality of goods traded in a market until only ‘lemons’ are left). Liquidity suppliers face the constant threat that they are trading under conditions of information asymmetry and are thus only transacting when the trade is adverse to their interests. 10 See infra Section I.1.D. 11 For expositional simplicity, I am assuming that the liquidity supply business involves no costs of operations or utility decreasing risks and requires no capital. A real world liquidity supplier would need to take account of such factors in setting its bids and asks, but doing so does not change the basic analysis. See Lawrence R. Glosten and Lawrence E. Harris, ‘Estimating the Components of the Bid–Ask Spread’ (1988) J. Fin. Econ. 21, 123 (testing a model in which the bid–ask spread is divided into an adverse selection component and a transitory component due to inventory costs, clearing costs, and other factors). In large, well-functioning equity markets, adverse selection accounts for the majority of the bid–ask spread. See Harris, supra n. 7, 158. 12 It is possible, however, for the market to break down entirely so that there is no trade. The smaller the portion of trading attributable to uninformed traders, the bigger the spread needs to be to compensate for the losses from the informed traders. But the bigger the spread, the fewer are the uninformed investors willing to tolerate the associated trading losses. 13 See Glosten and Milgrom, supra n. 8 (providing model of trading behaviour under information asymmetries in securities markets). 14 For expositional simplicity, I am again assuming that the liquidity supply business involves no costs of operations or utility decreasing risks and requires no capital. See supra

n. 11. 15 Applying Bayes’ Rule leads to exactly the same calculation of the bid and ask as was generated by the accounting perspective. See Glosten and Milgrom, supra n. 8. 16 One qualification should be noted, however. At the completion of this process, the market still does not fully reflect the information possessed by the informed traders. Full reflection of the information would require that the bid and offer move up enough that the mid-point between the two equals the value of the stock taking account of the new information, which is what would happen with its public announcement. The process described above only assures that the bid and offer move to the point that the offer price equals this value. For many US stocks today, given the current structure of the market and its regulation, spreads are down to just one or a few cents. For these stocks, the qualification raised here is of little significance. Some US stocks have much wider spreads, however, as is true of many stocks in some other countries. In these cases, this qualification strengthens the argument in favour of public disclosure of any given type of information, whatever are the strengths of the arguments, if any, against. 17 See Glosten and Harris, supra n. 11 (testing a model in which the bid–ask spread is divided into an adverse selection component and a transitory component due to inventory costs, clearing costs, and other factors); Kalok Chan, Y. Peter Chung, and Herb Johnson, ‘The Intraday Behavior of Bid–Ask Spreads for NYSE Stocks and CBOE Options’ (1995) J. Fin. & Quant. Anal. 30, 329 (suggesting that adverse selection is an important determinant of the intra-day behaviour of bid–ask spreads). 18 In the primary market, stocks are purchased from the company issuing those stocks, while in the secondary market, traders buy and sell stocks from each other. Stock exchanges are fundamentally secondary markets. 19 Merritt B. Fox, ‘Civil Liability and Mandatory Disclosure’ (2009) Colum. L. Rev. 109, 237, 260–4.. 20 ibid. at 258–60. 21 In an efficient market, the market price, whether it is relatively accurate or inaccurate, is an unbiased predictor of an issuer’s future cash flows. If it is inaccurate, it is just more likely to be far off, one way or the other, from how things ultimately turn out. Thus an efficient, but relatively inaccurate, price would result in as many positive surprises as negative ones. To many investors, the negative surprise is likely to be more salient, however. So when a negative surprise materializes, it generates a sense of grievance even though, ex ante, a positive surprise was equally likely. 22 For a purchaser of the shares in the primary market—the sellers of current dollars in return for the promise of receiving future dollars—more liquidity means it is less costly to sell her shares in the future to provide for future consumption because the bid will be less below the mid-point between the bid and the offer. In addition, more liquidity means that buyers in the market at the time of this sale would value the shares more highly so that this mid-point will be higher. This is because it is less expensive to buy the shares in the sense that the offer will be less above the mid-point and again it will be less expensive for these

buyers to sell at yet some further point in the future because the bid then will be less below the mid-point. 23 The cost of capital is lower because the prospect of a smaller bid–ask spread results in the same issuer’s expected future cash flow being discounted to present value at a lower discount rate. See also Yakov Amihud and Haim Mendelson, ‘Asset Pricing and the Bid– Ask Spread’ (1986) J. Fin. Econ. 17, 223; Yakov Amihud and Haim Mendelson, ‘Liquidity and Asset Prices: Financial Management Implications’ (1988) Fin. Mgmt. 17, 5. 24 Harris, supra n. 7, at 214–15. 25 See II.1.D infra. 26 Sam Mamudi, ‘Charlie Munger: HFT is Legalized Front-Running’, Barron’s, 3 May 2013, . 27 Linette Lopez, ‘New York’s Attorney General Has Declared War On Cheating HighFrequency Traders’, Bus. Insider, 24 September 2013, . 28 See Lewis (n. 4), 108, 126, 172. 29 This example fleshes out the story by Michael Lewis of how electronic front-running could occur with Amgen stock in such a situation. Lewis (n. 4), 33–4. Lewis asserts that the HFT could profit at the expense of others by cancelling its quotes on another exchange, but he does not discuss exactly why it would be profitable for the HFT to do so. Nor does he analyse how the quotes initially available might be different if the practice of electronic front-running were eliminated. The discussion that follows fills in these holes. 30 See I.1.A supra for a discussion of the different types of private information. We will revisit this discussion later with a more nuanced analysis that focuses on the fortunes of each of the three kinds of informed traders: fundamental value information traders, announcement traders, and inside information traders. See II.1.D infra. 31 For reasons of expository simplicity, this statement assumes that the increase in spreads would not decrease the volume of trading. In fact it would, but my simplification does not alter the basic logic of the analysis in the text. 32 Reg. NMS precludes exchanges from restricting access to trading on their facilities. See Regulation National Market System Rule 610(a), 17 CFR § 242.610(a) (2005) (prohibiting ‘national securities exchange[s] [from] … prevent[ing] or inhibit[ing] any person from obtaining efficient access’ to trading against the buy and sell quotes posted on exchanges); Securities and Exchange Act Section 6, 15 U.S.C. § 78f (1934) (providing that ‘the rules of [a registered] exchange [must] provide that any registered broker or dealer … may become a member of such exchange’). 33 See, e.g., Lewis (n. 4), 104. 34 Indeed, it appears to be these particular ordinary investors that Michael Lewis has in mind when arguing that electronic front-running takes money from ordinary folks on Main Street and gives it to HFTs. See Lewis (n. 4), 81, 102, 108, 172.

35

Jonathan B. Berk and Richard C. Green, ‘Mutual Fund Flows and Performance in Rational Markets’ (2004) J. Pol. Econ. 112, 1269. 36 See, e.g., Lewis (n. 4), 126, 176. 37 There is ample empirical evidence to suggest that accurate price signals do in fact have efficiency-enhancing effects on managerial decisions—see Thierry Foucault, Marco Pagano, and Ailsa Röell, Market Liquidity: Theory, Evidence, and Policy 361–68 (2013) (collecting relevant empirical studies). Theory also suggests that accurate financial information will often be under-produced due to its status as a public good—see, e.g., Philip Bond, Alex Edmans, and Itay Goldstein, ‘The Real Effects of Financial Markets’ (2012) Ann. Rev. Fin. Econ. 4, 339. 38 See I.1.B.ii supra. 39 See I.1.B.ii supra. 40 See Michael J. Fishman and Kathleen M. Hagerty, ‘Insider Trading and the Efficiency of Stock Prices’ (1992) Rand J. Econ. 23, 106, 110; Zohar Goshen and Gideon Parchomovsky, ‘On Insider Trading, Markets, and “Negative” Property Rights in Information’ (2001) Va. L. Rev. 87, 1229, 1238–43. 41 Lewis (n. 4), 172. 42 See, e.g., Nanex, Latency On Demand?, 23 August 2010,

(discussing discrepancies between NYSE quotes in the public quotation system and its private feeds and the potentially manipulative gaming of those feeds by HFTs). 43 In the example, if Lightning did not engage in slow market arbitrage, it is possible that it would be another HFT engaging in slow market arbitrage, not an ordinary trader, who would transact against the $161.13 offer. The ultimate question we are asking, however, is: what would happen if no HFT engaged in the practice? 44 See II.1.B supra. 45 See, e.g., Kit R. Roane, ‘How NYSE Plans to Use “Flash Crash” to Reclaim Its Glory’, CNN Money,

(reporting attempts by NYSE executives to connect HFTs with the Flash Crash); Bob Dannhauser, ‘Debating Michael Lewis’ “Flash Boys”: High-Frequency Trading Not All Bad’, Seeking Alpha, 7 April 2014, . 46 Lewis, p. 112. 47 See Angel et al., supra note 2, p. 2. 48 See, e.g., Joel Hasbrouck and Gideon Saar, ‘Low-Latency Trading’ (2013) J. Fin. Markets 16, 646 (finding that HFT activity reduces volatility); Jonathan Brogaard, Thibaut Moyaert, and Ryan Riordan, High-Frequency Trading and Market Stability, (May 2014), available at ). But see Sandrine Jacob Leal, Mauro Napoletano, Andrea Roventini, and Giorgio Fagiolo, Rock around the Clock: An Agent-

Based Model of Low- and High-Frequency Trading, (January 2014) (available at ). 49 Report of the Staffs of the CFTC and SEC to the Joint Advisory Committee on Emerging Regulatory Issues, Findings Regarding the Market Events of 6 May 2010, (hereinafter, the ‘Flash Crash Report’); Alexandra Twin, ‘Glitches Send Dow on Wild Ride’, CNN Money, 6 May 2010, . 50 Tom Lauricella and Peter A. McKay, ‘Dow Takes a Harrowing 1,010.14-Point Trip’, Wall St. J., 7 May 2010, 12:01 am, . 51 Tom Lauricella and Scott Patterson, ‘Legacy of the “Flash Crash”: Enduring Worries of Repeat’, Wall S J, 6 August 2010, . Many of the most outlandish transactions executed during the Flash Crash were later cancelled or ‘broken’ by regulators. See Deborah L. Jacobs, ‘Why We Could Easily Have Another Flash Crash’, Forbes, 9 August 2013, . 52 ibid. 53 Flash Crash Report (n. 49), 6. 54 This article focuses on HFTs as liquidity providers, and there is ample evidence they play this role. See, e.g., Albert J. Menkveld, ‘High-Frequency Trading and the New-Market Makers’ (2013) J. Fin. Markets16, 712. 55 See I.1.B and I.1.C supra. 56 Flash Crash Report (n. 49), 2–3. 57 David Easley, Marcos López de Prado, and Maureen O’Hara, ‘The Microstructure of the “Flash Crash”: Flow Toxicity, Liquidity Crashes and the Probability of Informed Trading’ (2011) J. Portfolio Mgmt. 37, 118, 120–6 (suggesting that order flow was especially informed and hence toxic for market-makers in the period preceding the Flash Crash). 58 Flash Crash Report (n. 49), 29; see also Charles M. Jones, What Do We Know about High-Frequency Trading, Columbia Business School Research Paper No. 33-36 (2013). 59 Flash Crash Report (n. 49), 45–57. 60 See 17 CFR § 242.602(b)(1) (defining scope of reporting requirements); 17 CFR § 600(b)(65); 17 CFR § § 242.600(b)(73)(ii)(A). 61 See Regulation of Exchanges and Alternative Trading System Rule 301(b)(5), 17 CFR § 242.301(b)(5) (1997); Concept Release on Equity Market Structure, Securities Exchange Act Release No. 34-61358, 17 C.F.R. § 242, at 72 (‘As [trading systems] that are exempt from exchange registration, [off-exchange platforms] are not required to provide fair access [to all traders] unless they reach a 5% trading volume threshold in a stock, which none

currently do[es]’ and that ‘[a]s a result, access to … [these platforms] … is determined primarily by private negotiation’). 62 See, e.g., Rhodi Preece, Dark Pools, Internalization, and Equity Market Quality, CFA Institute Codes, Standards, and Position Papers, 12–13 (2012). 63 The operator provides a similar service to the extent that it keeps out HFTs that engage in mid-point order exploitation. 64 This is because, as discussed in Section I, liquidity suppliers make money transacting with uninformed investors and lose money transacting with informed investors. To break even, they need to recoup enough from their transaction gains with uninformed investors to cover their losses to the informed ones. If there are fewer share transactions from uninformed investors, the spread will need to be wider: more money needs to be made off each share traded by an uninformed investor to recoup what has been lost to the informed investors. Indeed, if such ‘ideal’ dark pools really existed and what they had to offer was recognized by all uninformed investors, only informed traders would be attempting to trade on the exchanges, at which point all the liquidity suppliers would withdraw and the equity markets would collapse. 65 See Kevin S. Haeberle, ‘Stock-Market Law and the Accuracy of Public Companies’ Stock Prices’ (2015) Col. Bus. L. Rev. 121 (arguing that off-exchange trading venues should be banned because they increase the probability of adverse selection on exchanges). 66 An underlying premise of these criticisms is that the largest investment banks are also among the most prominent brokers and dark pool operators. For instance, Lewis often discusses dark pools as being operated by Wall Street banks, which is accurate—six of the ten largest dark pools are run by major investment banks, see Preece (n. 62), 14–15. All of the ten largest brokers on NYSE are also global investment banks. See NYSE Market Data, NYSE Broker Volume, (accessed 28 May 2016). 67 Michael Lewis, for example, claims that dark pool operators sell access to their trading venues to HFTs—without disclosing this practice to other users—and that these HFTs then exploit other traders. Lewis (n. 4), 123. Inferior execution could also occur on a dark pool if the counterparties trading there are especially informed or were given information about the existence of the customer limit orders posted there. 68 See Haeberle, supra n. 65. 69 European Commission, Market in Financial Instruments Directive (MiFID II): Frequently Asked Questions, 4, 7 (15 April 2014). 70 ibid. at 2; Chapter 14 of this volume, paragraph 14.07 (citing these caps as a ‘main instrument’ in an effort ‘to shift significant parts of dark turnover to regulated lit markets by strengthening the transparency framework’). 71 See III.1 supra. 72 See III.1.C and III.1.D supra. 73 Andrew Lo and C. MacKinlay, ‘Stock Market Prices Do Not Follow Random Walks: Evidence from a Simple Specification Test’ (1988) Review of Financial Studies 1, 41.

74

For an example of a similar study to determine whether the increased market fragmentation resulting from the imposition of MiFID I led to a decrease in short-term pricing efficiency, see Carole Gresse, Market Fragmentation in Europe: Assessment and Prospects for Market Quality (2012) (available at ). 75 See I.2.B.i supra. 76 See III.2 supra. 77 See Chapter 17 of this volume, paragraph 17.17. 78 ibid., paragraph 17.16. 79 See II.1.a supra. 80 See Chapter 17, paragraph 17.54. 81 See Chapter 17, paragraph 17.56. 82 An AMF news release describing the decision of the Committee can be seen at . 83 See Chapter 17, paragraph 17.59. 84 Congress, when the NMS amendments were adopted, expected that there would be a proliferation of competing venues. It self-consciously rejected a proposal for an electronic limit order book where all order flow was directed to a single trading venue, known as a central limit order book (‘CLOB’). See, e.g., S. Rep. No. 75, 94th Cong., 1st Sess., 12 (1975), 1975 USCCAN 179, 190 (‘Senate Report’) (rejecting role for ‘the SEC … as an “economic czar” for the development of a national market system’ and noting that ‘a fundamental premise of the bill is that … a national market system … will depend upon the vigor of competition within the securities industry’); Div. of Mkt. Regulation, US Sec. & Exch. Comm’n, Market 2000: An Examination of Current Equity Market Developments, at III-6 (January 1994) (discussing vigorous industry opposition to the SEC’s proposal of a CLOB in the 1970s. 85 This is a figure for 23 May 2016 and combines NYSE (13 per cent) and NYSE Arca (11%). It comes from the BATS Exchange website. See . 86 Giovanni Petrella, ‘MiFID, Reg NMS and Competition Across Trading Venues in Europe and the United States’ (2010) J Fin. Reg.& Compliance 18, 257(available at ). 87 See . 88 Petrella, supra n. 86, 258.

PART IV

SUPERVISION AND ENFORCEMENT

19 Public Enforcement of MiFID II Christos V. Gortsos

I. Introductory Remarks 1. The ‘Twin Legal Acts’: MiFID II and MiFIR II. Structure of the Study III. Competent Authorities: General Aspects, Powers, and Redress Procedures 1. General Aspects 2. Supervisory Powers 3. Power to Impose Administrative Sanctions and Measures as well as Criminal Sanctions 4. Redress Procedures IV. 1. 2. 3.

Cooperation Arrangements Cooperation between the Member States’ Competent Authorities Powers of Host Member States’ Competent Authorities Cooperation with Third Countries

V. Concluding Remarks and Assessment

I. Introductory Remarks 1. The ‘Twin Legal Acts’: MiFID II and MiFIR

[19.01] Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments . . . (MiFID I),1 as currently in force, is a key source of European Union (EU) capital markets law.2 It will be repealed, as of 3 January 2018, by two legal acts (the ‘twin legal acts’) of the same institutions of 15 May 2014, adopted in accordance with the ordinary legislative procedure laid down in Article 289(1) of the Treaty on the Functioning of the European Union (TFEU):3 • Directive 2014/65/EU on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU4 (MiFID II), which contains ninety-seven Articles; and • Regulation (EU) No 600/2014 on markets in financial instruments and amending Regulation (EU) No 648/20125 (MiFIR), which contains fiftyfive Articles. [19.02] These twin legal acts are addressed to EU Member States6 and are of relevance to the other Member States of the European Economic Area (EEA), that is, Norway, Liechtenstein, and Iceland. They both entered into force on 2 July 2014.7 The transposition of the Directive’s provisions into national legislation must be completed by 3 July 2017 and their vast majority will apply from 3 January 2018.8 [19.03] The legal basis of MiFID II is Article 53(1) TFEU on the coordination of the provisions laid down by law, regulation, or administrative action in Member States concerning the taking up and pursuit of activities as self-employed persons.9 The general provision of Article 114 TFEU on the approximation of laws for the establishment and functioning of the internal market (in accordance with Article 26 TFEU)10 was selected as the legal basis for MiFIR.11 [19.04] MiFID II applies to investment firms,12 market operators,13 datareporting services providers14 (i.e. approved publication arrangements (APAs), consolidated tape providers (CTPs), and approved reporting mechanisms (ARMs15)), and third-country firms16 providing investment services or performing investment activities17 through the establishment of a branch18 in the EU.19 It establishes requirements in relation to:

• the authorization and operating conditions for investment firms (not only for those exercising the right of establishment by branches, but also the freedom to provide services); • the provision of investment services or activities by branches of thirdcountry firms; • the authorization and operation of regulated markets;20 • the authorization and operation of data-reporting services providers; and • the supervision, cooperation, and enforcement by competent authorities, which is the subject of this study.21 [19.05] Certain provisions also apply to credit institutions,22 authorized under Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and prudential supervision of credit institutions and investment firms23 (CRD IV), when providing one or more investment services (and in some cases ancillary services24) and/or performing investment activities.25 Finally, some provisions apply to investment firms and to credit institutions when selling or advising clients in relation to structured deposits.26 [19.06] MiFID II is based on the principle of minimum harmonization. Accordingly, Member States and their competent authorities have been given the discretion to maintain or adopt additional, and in specific cases more stringent or even deviating rules.27 Nevertheless, the power conferred on the Commission to adopt delegated and implementing acts in accordance with Articles 290–1 TFEU, respectively,28 on the basis of draft regulatory and implementing technical standards developed by the European Securities and Markets Authority (ESMA), established by virtue of Regulation (EU) No 1095/2010 of the European Parliament and of the Council of 24 November 2010,29 is a factor accelerating progress towards the creation of a single rulebook.30

II. Structure of the Study

[19.07] The subject of this study is supervision, enforcement, and cooperation by competent authorities under MiFID II (Title VI, Articles 67– 88). The discussion is structured into two main sections. [19.08] Section III, divided into four subsections, deals with the role of Member States’ competent authorities within the system of MiFID II: (i) The general aspects pertaining to the designation of competent authorities, the cooperation between authorities in the same Member State, the professional secrecy regime, and the relations with auditors are covered in subsection 1 (A–D, respectively). (ii) Subsection 2 presents systematically the detailed provisions on the competent authorities’ supervisory powers. The general provisions and those on the exercise of supervisory powers are dealt with in turn (A and B, respectively). (iii) The similarly detailed provisions pertaining to the competent authorities’ power to impose administrative sanctions and measures as well as criminal sanctions are the subject matter of subsection 3. This covers the general provisions (A), those on the exercising of the power to impose administrative sanctions and measures (B), and those on the publication of decisions imposing administrative sanctions and measures (C). (iv) This section concludes, in subsection 4, with the rules on redress procedures, that is, the reporting of infringements (A), the right of appeal (B), and the extra-judicial mechanism for consumer complaints (C). [19.09] The cooperation arrangements among competent authorities are the subject matter of Section IV, structured in three subsections: (i) Subsection 1 presents the detailed provisions on the cooperation between the Member States’ competent authorities, including those on the obligation to cooperate with each other and with ESMA (A), the cooperation in supervisory activities, for on-site verifications or for investigations (B), the exchange of information (C), the binding mediation (D), and the consultation prior to authorization (E). (ii) The specific powers of host Member States’ competent authorities are presented in subsection 2.

(iii) Finally, subsection 3 deals with the provisions on cooperation with third countries.31 [19.10] The chapter finishes with Section V, which is comprised of a concise assessment and conclusion.

III. Competent Authorities: General Aspects, Powers, and Redress Procedures 1. General Aspects A. Designation of Competent Authorities [19.11] Each Member State must designate the competent authorities to carry out each of the duties provided for under the twin legal acts. It must inform the Commission, ESMA, and the other Member States’ competent authorities of the identity of the competent authorities responsible for enforcement of each of those duties, and of any division thereof.32 [19.12] These competent authorities must be public authorities. This is without prejudice to the possibility of delegating tasks to other entities if that is expressly provided for in Article 29(4) on tied agents.33 Any delegation of tasks to entities (other than those competent authorities) is subject to strict conditions. In particular: (i) Prior to delegation, competent authorities must take all reasonable steps to ensure that the entity to which tasks are to be delegated has the capacity and resources to execute all tasks effectively and that the delegation takes place only if a clearly defined and documented framework for the exercise of any delegated tasks has been established, stating the tasks to be undertaken and the conditions under which they are to be carried out.34 (ii) The entity to which tasks are delegated may not involve either the exercise of public authority or the use of discretionary powers of judgement and the final responsibility for supervising compliance with

MiFID II and its implementing measures lies with the competent authorities.35 [19.13] Member States must inform the Commission, ESMA, and the competent authorities of other Member States of any arrangements entered into with regard to delegation of tasks, including the precise conditions regulating such delegation.36 [19.14] A list of all competent authorities and entities mentioned above must be published and kept updated on ESMA’s website.37 [19.15] With reservation to the range of cases in which supervisory tasks may be delegated and the enhanced role of ESMA, these provisions were carried over from MiFID I.38

B. Cooperation between Authorities in the Same Member State [19.16] Member States are given the discretion to designate more than one competent authority to enforce the provisions of the twin legal acts. In such a case, their respective roles must be clearly defined and they must cooperate closely. [19.17] Such a close cooperation must be established between the competent authorities for the purposes of the twin legal acts and the national competent authorities for the supervision of the following entities: credit and financial institutions,39 pension funds, undertakings for collective investment in transferable securities (UCITS) in the meaning of Directive 2009/65/EC of the European Parliament and of the Council of 13 July 2009 on the coordination of laws, regulations and administrative provisions relating to UCITS (UCITS IV),40 as well as insurance and reinsurance intermediaries and insurance undertakings. These competent authorities must exchange any information which is essential or relevant to the exercise of their functions and duties.41

[19.18] In principle, these provisions were carried over, after the necessary updating, from MiFID I.42

C. Professional Secrecy [19.19] Competent authorities, all persons working or having worked for them, entities to which tasks are delegated pursuant to Article 67(2), and auditors and experts instructed by the competent authorities are bound by the obligation of professional secrecy. They are not allowed to divulge any confidential information they may receive in the course of their duties, unless it is in such a summary or aggregate form that individual investment firms, market operators, regulated markets, or any other person cannot be identified. This is without prejudice to the requirements of national criminal or taxation law or of other provisions of the twin legal acts.43 [19.20] Competent authorities, or other bodies or (natural or legal) persons receiving confidential information pursuant to the twin legal acts, may use it solely for two purposes: first, in the performance of their duties and for the exercise of their functions (in the case of the competent authorities) within the scope of these legal acts, or (in the case of other authorities, bodies, or persons) for the purpose for which it was provided to them; and, second, in the context of administrative or judicial proceedings specifically relating to the exercise of those functions. However, with the consent of the competent authority or the other authority, body, or person communicating information, the authority receiving it may use it for other purposes as well. This is also without prejudice to requirements of national criminal or taxation law.44 [19.21] If an investment firm, market operator, or regulated market has been declared bankrupt or is being compulsorily wound up, confidential information not concerning third parties may be divulged in civil or commercial proceedings only if necessary for carrying out the relevant proceeding.45 [19.22] The above-mentioned conditions of professional secrecy do not prevent competent authorities from the following:

(i) They may exchange or transmit confidential information in accordance with the twin legal acts or with any other EU legal acts applicable to investment firms, credit institutions, pension funds, UCITS, alternative investment funds (AIFs),46 insurance and reinsurance intermediaries and insurance undertakings, regulated markets, market operators, central counterparties (CCPs),47 and central securities depositaries (CSDs),48 with the consent of the competent authority or another authority body or person that communicated the information. (ii) They may also exchange or transmit, in accordance with national law, confidential information that has not been received from the competent authority of another Member State.49 [19.23] In principle, these provisions were carried over from MiFID I.50

D. Relations with Auditors [19.24] Specific provisions apply to statutory auditors (natural persons) and audit firms authorized in accordance with Articles 3–14 of Directive 2006/43/EC of the European Parliament and of the Council of 17 May 2006 on statutory audits of annual accounts and consolidated accounts ….51 Any such statutory auditor or audit firm, performing in an investment firm, a regulated market, or a data-reporting services provider either the task described in Article 34 of the Directive 2013/34/EU of the European Parliament and of the Council of 26 June 2013 on the annual financial statements, consolidated financial statements and related reports of certain types of undertakings …,52 or in Article 73 of Directive 2009/65/EC (UCITS IV),53 or any other task prescribed by law, has the following duties: (i) First, they must report promptly to the competent authorities any fact or decision concerning the undertaking of which they have become aware while carrying out that task and which is liable to the following: either to constitute a material infringement of the laws, regulations, or administrative provisions which lay down the conditions governing authorization or which specifically govern pursuit of the activities of investment firms, or to affect the continuous functioning of an

investment, or, finally, to lead to refusal to certify the accounts or to the expression of reservations. (ii) A duty is also imposed to report any facts and decisions of which they become aware in the course of carrying out the above tasks in an undertaking having close links with the investment firm within which these tasks are being carried out.54 [19.25] The disclosure, in good faith, by statutory auditors or audit firms to the competent authorities of any fact or decision referred to above does neither constitute a breach of any contractual or legal restriction on disclosure of information nor does it involve them in liability of any kind.55 [19.26] These provisions were carried over verbatim from MiFID I.56

2. Supervisory Powers A. General Provisions i. Rationale and Member States’ Notification Obligation [19.27] The objective of enhancing competent authorities’ supervisory powers under the MiFID II regime57 is given in Recital 137, which reads as follows: It is necessary to enhance convergence of powers at the disposal of competent authorities so as to pave the way towards an equivalent intensity of enforcement across the integrated financial market. This Directive should therefore provide for a minimum set of supervisory and investigative powers competent authorities of Member States should be entrusted with in accordance with national law. […] When exercising their powers under this Directive, competent authorities should act objectively and impartially and remain autonomous in their decision making.

[19.28] In light of the above, national legislation must endow competent authorities with all supervisory powers, including investigatory powers and powers to impose remedies, necessary to fulfil their duties under the twin legal acts.58 [19.29] By 3 July 2017, Member States must notify the Commission and ESMA of the laws, regulations, and administrative provisions transposing

the above provisions into national legislation, while any subsequent amendments thereto must also be notified without undue delay. In addition, they must ensure that mechanisms are in place to ensure that compensation may be paid or other remedial action may be taken, in accordance with national law, for any financial loss or damage suffered as a result of an infringement of the twin legal acts.59 ii. The Specific Supervisory Powers under Article 69 MiFID II [19.30] The extensive supervisory powers of competent authorities must include, de minimis, the following.60 [19.31] Competent authorities must have access to any document or other data in any form, which they consider as relevant for the performance of their duties and receive or take a copy of it, as well as the power to require or even demand the provision of information from any person (including the auditors of investment firms, regulated markets, and data-reporting services providers) and, if necessary, to summon and question a person with a view to obtaining information. They may also require or demand the provision of information (including all relevant documentation) from any person regarding the size and purpose of a position or exposure in a commodity derivative,61 and any assets or liabilities in the underlying market. [19.32] In order to detect potential infringements, they must have the power to carry out on-site inspections or investigations or allow auditors or experts to carry them out. In this respect, Recital 143 states that they must have, in accordance with national law and with the Charter of Fundamental Rights of the European Union,62 the ability to access the premises of (natural and legal) persons, when there is reasonable suspicion that documents and other data relating to the subject matter of an investigation exist and may be relevant to prove an infringement of the twin legal acts, if the person to whom a demand for information has already been made fails to comply, or if there are reasonable grounds for believing that if a demand were to be made, it would not be complied with, or that the documents or

information to which the information requirement relates would be removed, tampered with, or destroyed. [19.33] Also of importance are the powers to require existing recordings of telephone conversations or electronic communications or other data traffic records held by an investment firm, a credit institution, or any other entity regulated by the twin legal acts executing and documenting the executions of transactions, as well as to require existing data traffic records held by telecommunication operators, provided that this is permitted under national law, if there is reasonable suspicion of an infringement and if such records are relevant to an investigation into infringements of the twin legal acts. In this respect, Recital 144 states the following: • Access to such data and recordings constitute crucial, and sometimes the only, evidence to detect and prove the existence of market abuse in accordance with the relevant EU legal acts63 and to verify compliance by firms with investor protection and other requirements set out in the twin legal acts. • In order to introduce a level playing field in the EU in relation to this access competent authorities must, in accordance with national law, be able to require these data and recordings if a reasonable suspicion exists that such records relating to the subject matter of an inspection or investigation may be relevant to prove behaviour that is prohibited under the MAR or infringements of the twin legal acts. • Access to telephone and data traffic records held by a telecommunications operator should not encompass the content of voice communications by telephone. [19.34] Competent authorities may also require the freezing and/or sequestration of assets, the temporary prohibition of professional activity, the temporary or permanent cessation of any practice or conduct that they consider to be contrary to the provisions of the MiFIR and the provisions adopted in the implementation of MiFID II and prevent repetition of that practice or conduct, the suspension of trading in a financial instrument,64 the removal of a financial instrument from trading, whether on a regulated market or under other trading venues, and the removal of a natural person from the management board of an investment firm or market operator.

[19.35] In addition, they have the power to request any person to take steps to reduce the size of a position or exposure and limit its ability from entering into a commodity derivative (including by introducing limits on the size of a position any person can hold at all times in accordance with Article 57).65 They are also empowered to suspend the marketing or sale of financial instruments or structured deposits if the conditions of Articles 40– 2 MiFIR are met, as well as the marketing or sale of financial instruments or structured deposits if the investment firm has not developed or applied an effective product-approval process or otherwise failed to comply with Article 16(3) governing organizational requirements. [19.36] Finally, competent authorities are empowered to: • adopt any type of measure to ensure that investment firms, regulated markets, and other persons to whom the twin legal acts apply, continue to be compliant with legal requirements; • issue public notices; and • refer matters for criminal prosecution.

B. Exercise of Supervisory Powers [19.37] The extensive supervisory powers (including the investigatory powers and the powers to impose remedies) given to competent authorities under MiFID II may seriously interfere with fundamental rights, such as the right to respect for private and family life, home, and communications. Accordingly, competent authorities must exercise these powers within a comprehensive national legal framework and with due respect to the ‘principle of proportionality’. Recital 138, last sentence states on this: ‘Member States should allow the possibility for competent authorities to exercise such intrusive powers to the extent necessary for the proper investigation of serious cases where there are no equivalent means for effectively achieving the same result.’ [19.38] These powers may be exercised directly, in collaboration with other authorities, or by delegation to entities to which tasks have been delegated pursuant to Article 67(2), under their responsibility.66 As a

safeguard against potential abuses, they may also need to be exercised by application to competent judicial authorities for prior authorization, in accordance with national law.67 [19.39] According to the new Article 78, the processing of personal data collected in or for the exercise of the supervisory powers must be carried out in accordance with national law implementing Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data68 and with Regulation (EC) No 45/2001 of the same institutions of 18 December 2000 on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data,69 if applicable.

3. Power to Impose Administrative Sanctions and Measures as well as Criminal Sanctions A. General Provisions i. The Obligations Imposed on Member States and Their Exemptions [19.40] One of the main innovative elements of MiFID II (compared to MiFID I), if not the most innovative, is the detailed and intrusive character of the provisions of its Articles 70–1 on the competent authorities’ power to impose ‘administrative sanctions and measures’.70 Recital 148 states that the Directive should make reference to the terms ‘sanctions and measures’ as overarching terms covering all actions applied after an infringement, irrespective of their qualification as a sanction or a measure under national law, which are intended to prevent further infringements. [19.41] There are two main obligations imposed on Member States: 1. They must provide in their national legislation and ensure that their competent authorities are empowered to impose administrative

sanctions and measures applicable to infringements of the twin legal acts and of the national provisions adopted in implementation of MiFID II. 2. In addition, they must take all the necessary measures to ensure that such sanctions and measures are implemented.71 [19.42] Such sanctions and measures must be ‘effective, proportionate and dissuasive’ and apply to any kind of infringement, including (but not confined to) the ones specifically referred to in Article 70(3)–(5).72 This is without prejudice to the above-mentioned supervisory powers of competent authorities and their right to provide for and impose criminal sanctions.73 [19.43] Ratione personae and, in order to ensure similar treatment across the EU, administrative sanctions and measures must apply to investment firms, market operators authorized to operate an MTF or OTF, regulated markets, data-reporting services providers, credit institutions (in relation to investment and ancillary services or investment activities), and branches of third-country firms.74 They may also be applied, subject to the conditions laid down in national law in areas not harmonized by MiFID II, to those who effectively control the business of those entities and to the members of the investment firms’ and market operators’ management bodies,75 as well as to any other (natural or legal) persons which, under national law, are responsible for the infringement.76 [19.44] Exceptionally, in a manifest application of the principle of minimum harmonization, the Directive allows Member States not to lay down rules for administrative sanctions in the case of infringements which are subject to criminal sanctions under national law. In such cases, the Member State must communicate to the Commission the relevant criminal law provisions.77 [19.45] The justification for this is laid down in Recital 150: Even though nothing prevents Member States from laying down rules for administrative and criminal sanctions for the same infringements, Member States should not be required to lay down rules for administrative sanctions for the infringements [of the twin legal acts] which are subject to national criminal law. In accordance with national law, Member States are not obliged to impose both

administrative and criminal sanctions for the same offence, but they should be able to do so if their national law so permits.

The last sentence of this Recital delineates, however, this national discretion, by stating that the maintenance of criminal sanctions instead of administrative sanctions for infringements should not reduce or otherwise affect the ability of competent authorities to cooperate, and access and exchange information, in a timely way with competent authorities in other Member States, including after any referral of the relevant infringements to the competent judicial authorities for criminal prosecution.78 [19.46] By 3 July 2017, Member States must notify to the Commission and to ESMA the laws, regulations, and administrative provisions transposing (this) Article 70 into national legislation, together with any relevant criminal law provisions. Any subsequent amendments thereto must also be notified without undue delay.79 ii. Infringements of the Twin Legal Acts—Administrative Sanctions and Measures under Article 70 MiFID II [19.47] Infringements of the twin legal acts are considered to be, de minimis, the following:80 • an infringement of the provisions of the twin legal acts listed exhaustively in Article 70(3), • the provision of investment services or the performance of investment activities without the required authorization or approval,81 and • failure to cooperate or comply in an investigation or with an inspection or request under Article 69.82 [19.48] In the cases of these infringements, competent authorities must be empowered to take and impose, de minimis, the following administrative sanctions and measures, which, according to Recital 141, second sentence, must satisfy certain essential requirements in relation to: addressees, criteria to be taken into account when applying a sanction or measure, publication, key powers to impose sanctions, and levels of administrative fines.83 [19.49] With regard to (natural or legal) persons, competent authorities may issue a public statement, in accordance with Article 71,84 indicating

the infringing person and the nature of the infringement, and/or an order requiring that person to cease the conduct and desist from a repetition thereof. [19.50] Their powers also include the imposition of the withdrawal or suspension of the authorization of an investment firm (in accordance with Article 8), a third-country firm (in accordance with Article 43),85 a regulated market (in accordance with Article 44(5)), another market operator authorized to operate an MTF or an OTF, or, finally, a datareporting services provider (in accordance with Article 62).86 The withdrawal of credit institutions’ authorizations by their competent authorities is governed by Article 18 CRD IV, with reference to Article 67(1), which does not cover the infringements set out in Article 70 MiFID, since the CRD IV was adopted prior to MiFID II. In the author’s opinion, Article 67(1) CRD IV should be amended accordingly by MiFID II. [19.51] In addition, they may impose a temporary or, for repeated serious infringements, a permanent ban against any member of the investment firm’s management body or any other natural person responsible exercising management functions in investment firms, and a temporary ban on any investment firm being a member of or participant in regulated markets or MTFs or any client of OTFs. [19.52] Finally, they are empowered to impose the following fines, which, according to Recital 142, should be sufficiently high to offset the expected benefits and dissuasive even for larger institutions and their managers: 1. For legal persons, maximum administrative fines of either at least €5 million (or in the Member States with a derogation, the corresponding value in the national currency on 2 July 2014, that is, the date of the Directive’s entry into force) or of up to 10 per cent of its total annual turnover according to the last available accounts approved by the management body.87 2. For natural persons, maximum administrative fines of at least €5 million (or in the Member States with a derogation, the corresponding value in the national currency on 2 July 2014).

3. Maximum administrative fines of at least twice the amount of the benefit derived from the infringement, if that benefit is determinable, even in excess of the above maximum amounts. [19.53] At national discretion, competent authorities may be given the power to impose additional types of sanctions or to impose fines exceeding the amounts referred to above.88

B. Exercise of the Power to Impose Administrative Sanctions and Measures [19.54] Similarly to the provisions on the exercise of supervisory powers,89 competent authorities must exercise the power to impose administrative sanctions and measures in accordance with their national legal framework, either directly, in collaboration with other authorities, or by delegation to entities to which tasks have been delegated in accordance with Article 67(2), under their responsibility, or, finally, by application to the competent judicial authorities.90 [19.55] For the sake of consistent application across the EU, when determining the type and level of an administrative sanction or measure to be imposed, competent authorities must take into account all relevant circumstances and, inter alia: • the gravity and duration of the infringement, as well as the degree of responsibility of the (natural or legal) person responsible for the infringement; • the financial strength of the responsible person, as indicated in particular by the total turnover of the responsible legal person or the annual income and net assets of the responsible natural person; • if determinable, the importance of profits gained or losses avoided by the responsible person and the losses for third parties caused by the infringement; • the level of cooperation of the responsible person with the competent authority, without prejudice to the need to ensure disgorgement of profits gained or losses avoided by that person; and

• previous infringements by the responsible person.91

C. Publication of Decisions Imposing Administrative Sanctions and Measures as well as Criminal Sanctions i. General Provisions [19.56] The justification for the publication of decisions imposing administrative sanctions and measures is laid down in Recital 146 (first to fifth sentences), which reads as follows: In order to ensure that decisions made by competent authorities have a dissuasive effect on the public at large, they should normally be published. The publication of decisions is also an important tool for competent authorities to inform market participants of what behaviour is considered to infringe this Directive and to promote wider good behaviour amongst market participants.

If such publication causes disproportionate damage to the persons involved, jeopardises the stability of financial markets or an ongoing investigation the competent authority should publish the sanctions and measures on an anonymous basis in a manner which complies with national law or delay the publication. Competent authorities should have the option not to publish sanctions where anonymous or delayed publication is considered to be insufficient to ensure that the stability of financial markets will not be jeopardised. Competent authorities should not be required to publish measures which are deemed to be of a minor nature where publication would be disproportionate. [19.57] On the basis of these considerations, the following rules have been adopted. [19.58] Competent authorities must in principle publish any decision imposing an administrative sanction or measure for infringements of the MiFIR or of the national provisions adopted in the implementation of MiFID II on their official websites, without undue delay, after the person on whom the sanction was imposed has been informed of that decision. The publication must include at least information on the type and nature of the infringement and the identity of the persons responsible. That obligation does not apply to decisions imposing measures of an investigatory nature.92

[19.59] Nevertheless, if the competent authority considers, at its own discretion, that the publication of the legal persons’ identity or the natural persons’ personal data is ‘disproportionate’, following a case-by-case assessment, or if publication might jeopardize the stability of financial markets or an ongoing investigation, it may at its discretion take any of the following three courses of action:93 1. Defer the publication of the decision to impose the administrative sanction or measure until the reasons for non-publication cease to exist. 2. Publish that decision on an anonymous basis in a manner compliant with national law, if such a publication ensures an effective protection of the personal data concerned. In this case, the publication of the relevant data may be postponed for a reasonable period of time if it is envisaged that within that period of time the reasons for anonymous publication will cease to exist. 3. Not publish that decision at all, if the above options are considered to be insufficient to ensure the proportionality of the publication of such decisions, with regard to measures which are deemed to be of a minor nature, and that the stability of financial markets would not be put in jeopardy. [19.60] If the decision to impose a sanction or measure is subject to appeal before judicial or other authorities,94 the competent authorities must also publish, immediately, on their official website such information and any subsequent information on the outcome of such appeal. Moreover, any decision annulling a previous decision to impose a sanction or a measure must also be published.95 [19.61] Competent authorities must ensure that any publication in accordance with the above-mentioned will remain on their official website for a period of at least five years after its publication. Personal data contained in the publication may be kept on this website only for the period necessary under the applicable data-protection rules.96 [19.62] MiFID II neither requires nor prevents the publication of criminal sanctions imposed for infringements of the twin legal acts.97 Accordingly, this is left to national discretion.

ii. Notification and Reporting Requirements Introductory Remarks [19.63] Article 71(3)–(6) contains specific provisions with regard to the submission of information on sanctions and measures to ESMA by the Member States (see paragraph 19.65) as well as by competent authorities (see paragraphs 19.66–19.73). ESMA’s courses of action in this respect are also detailed therein. [19.64] Acting in accordance with Article 71(7), ESMA developed draft implementing technical standards (ITS) concerning the procedures and forms for the submission of such information and submitted them to the Commission on 11 December 2015.98 Article 2 of these ITS stipulates that competent authorities must designate contact points for communications on any issue relating to the submission of information and notify ESMA accordingly. The latter must also designate a contact point for communication of information and for any issue relating to the reception thereof, publishing information about its contact point on its website. [19.65] Power is conferred on the Commission to adopt these draft ITS in accordance with Article 15 ESMA Regulation (by an Implementing Regulation).99 The General Obligation [19.66] Member States must provide annually ESMA with aggregated information regarding all administrative sanctions and measures imposed (published and not published), with the exception of measures of an investigatory nature.100 In discharging this obligation, their competent authorities must provide ESMA with this information by filling in the form set out in Annex II of the Implementing Regulation (which must include the information on all sanctions and measures imposed by the competent authority during the previous calendar year).101 ESMA must publish this information in its Annual Report.102 Requirements Pertaining to Administrative Sanctions and Measures Imposed and Published [19.67] If a competent authority has disclosed an administrative sanction or measure to the public, it must concurrently report this fact to ESMA as well.103 It is worth mentioning that ESMA already

compiles information on published sanctions on the basis of notifications made to it by competent authorities.104 [19.68] On the other hand, if a published administrative sanction relates to an investment firm, market operator, data-reporting services provider, credit institution (in relation to investment services or activities and ancillary services), or a branch of a third-country firm, ESMA must add a reference to the published sanction in the relevant register.105 Requirements Pertaining to Administrative Sanctions Imposed but Not Published [19.69] Competent authorities must inform ESMA of all administrative sanctions imposed but not published,106 including any appeal in relation thereto and the outcome thereof, in order to enable competent authorities to take them into account in their ongoing supervision.107 This notification must be made by using the existing interfaces provided by ESMA’s IT system, in a report file in the format set out in Annex I of the Implementing Regulation, and by sending this file within ten working days, at the latest, of the decision not to publish the sanction being taken.108 [19.70] ESMA must maintain a central database of these sanctions only for the purposes of exchanging information between competent authorities, accessible only to them and updated on the basis of the information provided by the competent authorities.109 Requirements Pertaining to Criminal Sanctions Imposed [19.71]  Competent authorities must receive information and the final judgement pertaining to any criminal sanction imposed, and submit it to ESMA within ten working days at the latest of it having received the information or the final judgement.110 The just-mentioned central database of sanctions maintained by ESMA must also contain the criminal sanctions imposed.111 [19.72] If a Member State has opted to enact criminal sanctions for infringements of Article 70, its competent authority must provide ESMA annually with anonymized and aggregated data regarding all criminal investigations undertaken and the criminal sanctions imposed.112 They must do so by:

• using the existing interfaces provided by ESMA’s information technology system, in a report file in the format set out in Annex I of the Implementing Regulation, and • filling in the form set out in Annex III thereto (which must include data on all criminal investigations undertaken and criminal sanctions imposed by the competent authority during the previous calendar year).113 ESMA must publish in its Annual Report the anonymized and aggregated data regarding all criminal investigations undertaken and the criminal sanctions imposed.114 [19.73] If a competent authority has disclosed an administrative measure or sanction or a criminal sanction to the public, it must concurrently report this fact to ESMA as well.115 [19.74] Finally, as in the case of administrative sanctions, if a published criminal sanction relates to an investment firm, market operator, datareporting services provider, credit institution (in relation to investment services or activities and ancillary services), or a branch of a third-country firm, ESMA must add a reference to the published sanction in the relevant register.116

4. Redress Procedures A. Reporting of Infringements [19.75] With regard to the reporting of potential or actual infringements of the provisions of the MiFIR and of the national provisions adopted in the implementation of MiFID II, the Directive lays down two (new in MiFID II) rules. [19.76] First, the competent authorities must establish effective and reliable mechanisms to enable and encourage the reporting to them of such infringements, without prejudice to adequate safeguards for accused persons. These mechanisms must include, de minimis, the following: specific procedures for the receipt of reports on infringements and their follow-up (including the establishment of secure channels for the

communication of such reports), appropriate protection for employees reporting infringements committed within their own institution (at least against retaliation, discrimination, or other types of unfair treatment), and protection of the identity of both the person reporting the infringements and the natural person allegedly responsible for these at all stages of the relevant procedures (unless such disclosure is required by national law in the context of further investigation or subsequent administrative or judicial proceedings).117 [19.77] In addition, investment firms, market operators, data-reporting services providers, credit institutions (in relation to investment services or activities and ancillary services), and branches of third-country firms must have in place appropriate procedures for their employees to report such infringements internally, through a specific, independent, and autonomous channel.118

B. Right of Appeal [19.78] According to Recital 147, in finem, procedures must be in place to ensure the right of the accused person of defence and to be heard before the adoption of a decision concerning him/her, as well as the right to seek effective remedy before a court against a decision concerning him/her.119 Hence, any decision taken under the MiFIR or the national provisions adopted in the implementation of MiFID II must be ‘properly’ reasoned and be subject to the right of appeal before a tribunal. This right of appeal also applies if, in respect of a completed application for authorization, no decision is taken within six months of its submission.120 [19.79] Public bodies or their representatives, consumer organizations having a legitimate interest in protecting consumers, and professional organizations having a legitimate interest in acting to protect their members may also, in the interests of consumers and in accordance with national law, take action before the courts or the competent administrative bodies in order to ensure that MiFIR or the national provisions adopted in the implementation of MiFID II are applied.121 All these provisions were carried over almost verbatim from MiFID I.122

C. Extra-Judicial Mechanism for Consumers’ Complaints [19.80] With a view to ensuring the protection of clients and without prejudice to their right to bring an action before the courts, MiFID II requires the setting-up of efficient and effective complaints and redress procedures for the out-of-court settlement of consumer disputes concerning the provision of investment and ancillary services provided by investment firms, using existing private and public bodies as appropriate, and taking into account the Commission Recommendation 98/257/EC of 30 March 1998 on the principles applicable to the bodies responsible for out-of-court settlement of consumer disputes123 and the Commission Recommendation 2001/310/EC of 4 April 2001 on the principles for out-of-court bodies involved in the consensual resolution of consumer disputes.124 [19.81] All investment firms must adhere to one or more such bodies implementing such complaint and redress procedures. Those bodies must actively cooperate with their counterparts in other Member States in the resolution of cross-border disputes, using existing cross-border cooperation mechanisms, and in particular the Financial Services Complaints Network (FIN-Net), launched by the Commission in 2001.125 [19.82] The competent authorities must notify ESMA of the complaint and redress procedures available in their jurisdictions. A list of all extrajudicial mechanisms must be published and kept updated on ESMA’s website.126 These provisions were carried over almost verbatim from MiFID I as well.127

IV. Cooperation Arrangements 1. Cooperation between the Member States’ Competent Authorities A. Obligation to Cooperate

i. General Provisions [19.83] MiFID II establishes a comprehensive general framework of cooperation between Member States’ competent authorities for the implementation of its provisions, which is based on that of MiFID I, but is definitely more detailed. Within this framework, Member States’ competent authorities are under the obligation to cooperate if necessary for the purpose of carrying out their duties under the twin legal acts, making use of their powers whether set out therein or in national law.128 They must also cooperate with ESMA for the purposes of MiFID II in accordance with the ESMA Regulation. [19.84] In this respect, they must provide ESMA, without undue delay, with all information necessary to carry out its duties under the twin legal acts in accordance with Articles 35–6 ESMA Regulation on the collection of information and the relationship with the ESRB.129 This general cooperation obligation is further specified as follows. [19.85] If a Member State has laid down criminal sanctions for infringements of this provision in accordance with Article 70,130 its competent authority must have the necessary powers to liaise with judicial authorities within its jurisdiction in order to receive specific information relating to criminal investigations or proceedings commenced for potential infringements of the twin legal acts and provide the same to other competent authorities and to ESMA.131 [19.86] Member States’ competent authorities must render assistance to each other and, in particular, exchange information and cooperate in any investigation or supervisory activities and with respect to facilitating the recovery of fines.132 In order to facilitate and accelerate cooperation, and more particularly information exchange, Member States must designate a single competent authority as a ‘contact point’ for the purposes of the twin legal acts and communicate to the Commission, ESMA, and the other Member States the names of the authorities designated to receive requests for exchange of information or cooperation.133 ESMA must publish and keep updated on its website a list of those authorities.134 [19.87] For the sake of facilitation of this assistance and for the purpose of this cooperation, the necessary administrative and organizational

measures must be taken and competent authorities may use their powers even if the conduct under investigation does not constitute an infringement of any regulation in force under national law.135 [19.88] If a competent authority reasonably suspects that acts contrary to the provisions of the twin legal acts are being or have been carried out by entities not subject to its supervision on the territory of another Member State, it must notify specifically the other Member State’s competent authority and ESMA. Without prejudice to the notifying competent authority’s competences, the notified competent authority must take appropriate action and inform the notifying authority and ESMA of the outcome of the action and, if possible, of significant interim developments.136 [19.89] Without prejudice to the above provisions, competent authorities must notify ESMA and other competent authorities of the details of any requests to reduce the size of a position or exposure pursuant to Article 69(2), point (o) and/or any limits on the ability of persons to enter into a commodity derivative pursuant to Article 69(2), point (p).137 [19.90] This notification must include, if relevant, specific features and be made in principle not less than twenty-four hours before the actions or measures are intended to take effect. A competent authority receiving such a notification may take measures in accordance with these Articles if it is satisfied that such measures are necessary in order to achieve the other competent authority’s objective (with due notification).138 ii. Substantial Importance of a Trading Venue in a Host Member State [19.91] Several trading venues (i.e. regulated markets, MTFs, and OTFs) are established in host Member States,139 the competent authorities of which do not have direct supervisory powers over their operation. Within this context, Article 79(2) provides that, if the operation of a trading venue in a host Member State has become of ‘substantial importance’ for the functioning of securities markets and for investor protection therein, and taking into account the situation of these securities markets, the home140 and host Member State’s competent authorities of the trading venue must establish ‘proportionate’ cooperation arrangements.

[19.92] Similar were the provisions of Article 56(2) MiFID I, which nevertheless applied only to regulated markets. Recital 140 MiFID II clarifies that its provisions should apply also to MTFs and OTFs, in view of the significant impact and market share they have already acquired or are expected to acquire, respectively. [19.93] In order to achieve the highest possible degree of flexibility due to the fact that such cooperation arrangements may be prolonged and intense, these must take the appropriate form amongst possible cooperation modalities, such as ad hoc or periodic information sharing, consultation, and assistance, proportionate to the needs for cross-border supervisory cooperation (resulting, in particular, from the nature and scale of the impact on securities markets and on investor protection in the host Member State).141 [19.94] In this respect, the Commission is empowered to adopt delegated acts in accordance with Article 89 to establish the criteria under which the operations of such a trading venue could be considered to be of ‘substantial importance’ for the functioning of securities markets and for investor protection.142 [19.95] According to ESMA’s advice to the Commission in 2014,143 this should be the case if either the host Member State used to be the home Member State of the trading venue in question, or the trading venue in question has acquired through a merger, takeover, or any other form of transfer the business of a trading venue which was previously operated by a market operator or investment firm registered in the host Member State.144 [19.96] In addition, ESMA, acting in accordance with Article 79(9), developed a draft ITS to establish standard forms, templates, and procedures for these cooperation arrangements and submitted them to the Commission on 11 December 2015.145 The main points of this draft ITS can be summarized as follows: • ESMA considers that the development of these templates, forms, and procedures should be consistent with the existing practice set out in the ‘Multilateral Memorandum of Understanding on Cooperation

Arrangements and Exchange of Information’ of 5 June 2014 (ESMA/2014/608).146 • Under the flexible approach proposed, home and host Member States’ competent authorities should use the standard template in Annex I as a basis for their agreement, adapting or complementing it in order to ensure that its provisions are proportionate to particular circumstances of each specific case and the type and level of cooperation sought.147 A non-exhaustive list of events where competent authorities may take action is also laid down, reflecting that the need for cross-border cooperation depends on the nature and scale of the operations or structure of trading venues.148 iii. Specific Provisions on Greenhouse Gas Emission Allowance Markets and Agricultural Commodity Derivatives [19.97] In order to ensure the acquisition of a consolidated overview of greenhouse gas emission allowance markets, competent authorities must cooperate with public bodies competent for the oversight of spot and auction markets and competent authorities, registry administrators, and other public bodies charged with the supervision of compliance under Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community (…).149 [19.98] In addition, they must report to and cooperate with public bodies competent for the oversight, administration, and regulation of physical agricultural markets under Regulation (EU) No 1308/2013 of the European Parliament and of the Council of 17 December 2013 establishing a common organization of the markets in agricultural products (…)150 in relation to agricultural commodity derivatives.151

B. In Particular: Cooperation for Supervisory Activities, On-Site Verifications, or Investigations [19.99] A Member State’s competent authority may request the cooperation of another Member State’s competent authority in a supervisory

activity, for an on-site (or on-the-spot) verification or for an investigation.152 In the case of investment firms that are remote members or participants of a regulated market, this market’s competent authority may choose to address them directly, informing accordingly the competent authority of the home Member State of the remote member or participant. A competent authority receiving a request with respect to an on-site verification or an investigation must, within the framework of its powers, either carry them out itself, or allow the requesting authority, auditors, or experts to carry them out.153 These provisions were carried over verbatim from MiFID I.154 [19.100] The significant role of ESMA in this respect is threefold. In particular: 1. With the objective of converging supervisory practices, ESMA may participate in the activities of colleges of supervisors, including on-site verifications or investigations, carried out jointly by two or more competent authorities in accordance with Article 21 ESMA Regulation.155 These provisions were also carried over verbatim from MiFID I.156 2. For the specification of the information to be exchanged between competent authorities when cooperating in supervisory activities, onsite verifications, and investigations and acting in accordance with Article 80(3), ESMA developed draft regulatory technical standards (RTS) and submitted them to the Commission on 29 June 2015.157 The detailed provisions of its Articles 2–4 refer to the information to be exchanged between cooperating competent authorities concerning investment firms, market operators, data-reporting services providers, credit institutions (in relation to investment services or activities and ancillary services), or any other (natural or legal) person or unincorporated entity or association.158 They aim at ensuring that the information to be exchanged is of sufficient scope and nature to allow competent authorities to discharge their supervisory duties and functions effectively. The power is conferred on the Commission to adopt these RTS in accordance with Articles 10–14 ESMA Regulation (by a Delegated Regulation).159

3. Acting in accordance with Article 80(4), ESMA also drafted ITS establishing standard forms, templates, and procedures for competent authorities with regard to cooperation in such cases and submitted them to the Commission on 11 December 2015.160 They stipulate that competent authorities must designate a specific contact point to facilitate communication of request and cooperation.161 They also detail the procedures to be laid down with regard to cooperation requests, acknowledgement of receipt, replies to requests and sending and processing requests for cooperation, as well as requests for taking a statement from a person and for competent authorities to open an investigation or carry out an on-site verification or investigation.162 Finally, specific provisions govern the unsolicited exchanges of information and requirements to notify competent authorities.163

C. Exchange of Information [19.101] The rationale for the exchange of information between competent authorities is laid down in Recital 153, which reads as follows: It is necessary to reinforce provisions on exchange of information between national competent authorities and to strengthen the duties of assistance and cooperation which they owe to each other. Due to increasing cross-border activity, competent authorities should provide each other with the relevant information for the exercise of their functions, so as to ensure the effective enforcement of this Directive, including in situations where infringements or suspected infringements may be of concern to authorities in two or more Member States. In the exchange of information, strict professional secrecy is needed to ensure the smooth transmission of that information and the protection of particular rights.

[19.102] On the basis of these considerations, the following rules have been adopted. [19.103] The competent authorities designated as ‘contact points’ in accordance with Article 79(1)164 must immediately supply one another with the information required for the purposes of carrying out the duties of the competent authorities designated in accordance with Article 67(1),165 which are set out in the provisions adopted pursuant to the twin legal acts. They

may indicate at the time of communication that the information exchanged may not be disclosed without their express agreement; that is, it can only be exchanged for the purposes for which their agreement was given.166 Any confidential information received by a Member State’s contact point through the contact point of another should not be regarded as ‘purely domestic’.167 [19.104] The competent authority designated as contact point may transmit the above received information (as well as the information received under Article 77 on relations with auditors and Article 88 on cooperation with third countries)168 to the competent authorities designated in accordance with Article 67(1). The latter may transmit it further to other bodies or persons only with the express agreement of the competent authorities which disclosed it and only for the purposes for which the agreement was given, except in duly justified circumstances. In the latter case, the contact point which transmitted the information must be immediately informed.169 [19.105] Authorities, as referred to in Article 76 on professional secrecy,170 and other bodies or persons receiving confidential information, may use it only in the course of their duties, and in particular: • to check that the conditions governing the taking-up of the investment firms’ business are met and to facilitate the monitoring of its conduct, especially with regard to the capital adequacy requirements imposed by the CRD IV, administrative, and accounting procedures and internalcontrol mechanisms; • to monitor the proper functioning of trading venues; • to impose sanctions; • in administrative appeals against competent authorities’ decisions; as well as • in court proceedings initiated under Article 74 and in the extra-judicial mechanism for investors’ complaints provided for in Article 75.171 [19.106] Neither the above provisions nor those of Article 76 on professional secrecy and Article 88 on cooperation with third countries172

can prevent a competent authority from transmitting confidential information intended for the performance of their tasks to: • ESMA; • the European Systemic Risk Board (ESRB), established under Regulation (EU) No 1092/2010 of the European Parliament and of the Council of 24 November 2010 on European Union macro-prudential oversight of the financial system (…);173 • central banks—the European System of Central Banks (ESCB) (more specifically for the euro area, the Eurosystem) and the European Central Bank (ECB), in their capacity as monetary authorities; and, • if appropriate, other authorities responsible for overseeing payment and settlement systems.174 [19.107] Likewise, these authorities or bodies may communicate to the competent authorities any information needed for the purpose of performing their functions provided for in the twin legal acts.175 These provisions were carried over verbatim from MiFID I.176 [19.108] Acting in accordance with Article 81(4), ESMA drafted ITS establishing standard forms, templates, and procedures for the exchange of information and submitted them to the Commission on 11 December 2015.177 These ITS are contained in the same document pertaining to the exchange of information between competent authorities when cooperating in supervisory activities, on-site verifications, and investigations, and Articles 1–7 apply also in this case.178 In addition, specific provisions govern unsolicited exchanges of information.179

D. Binding Mediation and Its Exemptions [19.109] If a request relating either to carrying out a supervisory activity, an on-the-spot verification, an investigation, or to exchanging information180 has been rejected or has not been acted upon within a reasonable time, the competent authorities may refer the case to ESMA.181 The latter may act in accordance with Article 19 of its founding Regulation

on the settlement of disagreements between competent authorities in crossborder situations (usually referred to as ‘binding mediation’).182 [19.110] This is without prejudice to two possibilities.183 First, if a mediation involves a case of a Member State not having correctly implemented EU law, the mediation powers under Article 19 ESMA Regulation usually apply without prejudice to ESMA’s powers under Article 17 of that Regulation on the breach of EU law.184 The justification is that ESMA may opt not to wait for the parties involved to bring their disagreement to the mediation panel under Article 19, but may prefer to take action under Article 17.185 This is also provided for in Article 82(2) MiFID II. These provisions were carried over verbatim from MiFID I.186 [19.111] Second, ESMA may also act under Article 19 ESMA Regulation without prejudice to the possibility of a competent authority refusing to act on a request for information. A competent authority may refuse either to act on a request for cooperation in carrying out an investigation, on-the-spot verification, or supervisory activity or to exchange information, only if judicial proceedings have already been initiated in respect of the same actions and the same persons before the authorities of the Member State addressed, or final judgment has already been delivered in the Member State addressed also in respect of the same persons and the same actions. In case of such a refusal, the competent authority must notify the requesting competent authority and ESMA, providing detailed information.187

E. Consultation Prior to Authorization [19.112] Competent authorities must consult each other prior to granting authorization to an investment firm which is either a subsidiary of an investment firm or market operator or credit institution authorized in another Member State, or a subsidiary of the parent undertaking of such an investment firm or credit institution, or controlled188 by the same persons who control such an investment firm or credit institution. In addition, the competent authority of the Member State responsible for the supervision of credit institutions or insurance undertakings must also be consulted prior to granting an authorization to an investment firm or market operator which is

either a subsidiary of a credit institution or insurance undertaking authorized in the EU, or a subsidiary of the parent undertaking of such a credit institution or insurance undertaking, or controlled by the same person which controls such a credit institution or insurance undertaking.189 [19.113] In both cases, these relevant competent authorities must, in particular: • enter into consultations when assessing the suitability of the shareholders or members and the reputation and experience of the persons effectively directing the business and involved in the management of another entity of the same group;190 and • exchange all information necessary for the granting of an authorization and for the ongoing assessment of compliance with operating conditions.191 [19.114] These provisions were carried over verbatim from MiFID I.192 [19.115] ESMA, acting in accordance with Article 84(4), developed a draft ITS to establish standard forms, templates, and procedures for the consultation of other competent authorities prior to granting an authorization, and submitted them to the Commission on 11 December 2015.193 Just like the ITS on information to be exchanged between competent authorities when cooperating in supervisory activities, on-site verifications, and investigations and on the exchange of information,194 they stipulate that competent authorities must designate a specific contact point to facilitate communication between them prior to granting authorization.195 They also detail the procedures to be laid down with regard to requests for consultation, acknowledgement of receipt and replies to request for consultation, the procedures for the consultation itself, and the conditions for the use of information.196

2. Powers of Host Member States’ Competent Authorities

A. Powers with Regard to Branches Operating Under Freedom of Establishment [19.116] A host Member State’s competent authority may, for statistical purposes, require investment firms with branches within its territory to report to it periodically on the activities of those branches. It may also require these branches to provide directly the information necessary for the monitoring of their compliance with the standards set by the host Member State and applying to them for the cases provided for in Article 35(8), with due respect to the ‘principle of national treatment’.197 [19.117] This Article provides that the host Member State’s competent authority: • must ensure that the services provided by the branch within its territory comply with the obligations laid down in Articles 24–25 and 27–28 MiFID II and Articles 14–26 MiFIR and the measures adopted pursuant thereto by the host Member State if allowed under Article 24(12) MiFID II; and • has the right to examine branch arrangements and to request changes strictly needed in order to be able to enforce the above obligations with respect to the services and/or activities provided by the branch within its territory. These provisions were carried over verbatim from MiFID I.198

B. Precautionary Measures by Host Member States’ Competent Authorities [19.118] If a host Member State’s competent authority has clear and demonstrable grounds for believing that an investment firm acting within its territory under the freedom to provide services infringes the obligations arising from Article 34 MiFID II or that an investment firm that has a branch within its territory infringes obligations not conferring powers on it, it must refer those findings to the home Member State’s competent authority. If, despite the measures taken by the latter or because such

measures prove inadequate, the investment firm persists in acting in a manner that is clearly prejudicial to the interests of host Member State investors or the orderly functioning of markets, the host Member State’s competent authority must, after informing the home Member State’s competent authority, take all appropriate measures, including the possibility of preventing offending investment firms from initiating any further transactions within their territories.199 [19.119] On the other hand, if a host Member State’s competent authority ascertains that an investment firm that has a branch within its territory infringes the legal or regulatory provisions adopted in its territory pursuant to the MiFID II provisions conferring powers on it, it may require the investment firm concerned to put an end to its irregular situation. If the investment firm concerned fails to take the necessary steps, it can take all appropriate measures to ensure that the investment firm concerned puts an end to its irregular situation, communicating the nature of those measures to the home Member State’s competent authorities. [19.120] If, despite the measures taken by the host Member State’s competent authority, the investment firm persists with the infringement, the former must, after informing the home Member State’s competent authority, take all the appropriate measures needed to protect investors and the proper functioning of the markets.200 [19.121] If the host Member State’s competent authority of a trading venue has ‘clear and demonstrable grounds’ to believe that such a trading venue infringes the obligations arising from MiFID II, it must refer those findings to the home Member State’s competent authority. If, despite the measures taken by the latter or because such measures prove inadequate, that trading venue persists in acting in a manner clearly prejudicial to the interests of host Member State investors or the orderly functioning of markets, the host Member State’s competent authority must, after informing the home Member State’s competent authority, take all the appropriate measures needed, including the possibility of preventing that trading venue from making its arrangements available to remote members or participants established in its territory.201

[19.122] In all three cases, any measure adopted involving sanctions or restrictions on the activities of an investment firm or of a regulated market must be properly justified and communicated to the investment firm or the regulated market concerned.202 In addition, the Commission and ESMA must be informed of such measures without undue delay, while the host Member State’s competent authority may refer the matter to ESMA, which may act in accordance with the powers conferred on it under Article 19 ESMA Regulation on binding mediation.203 [19.123] These provisions were carried over verbatim from MiFID I.204

3. Cooperation with Third Countries [19.124] Member States and ESMA, in accordance with Article 33 ESMA Regulation on its international relations,205 may conclude cooperation agreements providing for the exchange of information with the competent authorities of third countries206 under two conditions: the information disclosed is subject to guarantees of professional secrecy, at least equivalent to those required under Article 76;207 and the exchange of information is intended for the performance of the tasks of those competent authorities. They may also conclude, under the same conditions, cooperation agreements providing for the exchange of information with third-country authorities, bodies, and natural or legal persons responsible for: • the supervision of credit institutions, other financial institutions, insurance undertakings, and financial markets; • the liquidation and bankruptcy of investment firms or any similar procedures, as well as the oversight of the bodies involved in these procedures; • the carrying out of statutory audits of the accounts of investment firms, credit institutions, financial institutions, and insurance undertakings, in the performance of their supervisory functions, or which administer compensation schemes, in the performance of their functions; as well as • the oversight of persons charged with carrying out such statutory audits and the statutory audits of the accounts of persons active on emission

allowance markets and on agricultural commodity derivatives markets for the purpose of ensuring a consolidated overview of financial and spot markets. In both cases, any transfer of personal data to a third country by a Member State must be in accordance with Articles 25–26 of Directive 95/46/EC, while transfer of personal data to a third country by ESMA in accordance with Article 9 of Regulation (EU) No 45/2001.208 [19.125] If the information originates in another Member State, it may be disclosed only with the express agreement of the competent authorities which have transmitted it and, if appropriate, solely for the purposes for which the agreement was given. This applies also to information provided by third-country competent authorities.209 [19.126] With the exception of the (newly introduced) reference to Regulation (EU) No 45/2001, these provisions were carried over verbatim from MiFID I.210

V. Concluding Remarks and Assessment [19.127] In the author’s view, in order to ensure the stability of capital markets and in general the financial system (in terms of prudence and not crisis management) and the attainment of the other goals underlying (public) capital markets law, of essential importance are the quality and targeting (and not the quantity) of regulations, as well as the following five elements pertaining to financial supervision: (i) micro-prudential supervisory effectiveness, including the appropriate staffing of supervisory authorities in order to ensure the quality of the supervision exercised;211 (ii) the efficient and unobstructed exercise of supervisory authorities’ sanctioning powers; (iii) the substantial safeguarding of supervisory authorities’ institutional, personal, financial, and functional independence (following the model

of central banks as bodies responsible for the definition and implementation of monetary policy212), and concurrently the safeguarding of proper accountability; (iv) the establishment of an appropriate framework on supervisory authority responsibility vis-à-vis investors, supervised entities, and their shareholders; and (v) the effectiveness of cooperation arrangements between national supervisory authorities, both in the same country and internationally, in relation to cross-border issues. [19.128] The financial system of several states was not exposed (at least primarily), or was less significantly exposed, to the recent (2007–9) international financial crisis, not only because it was operating within a strong institutional and regulatory framework, but also because microprudential supervision was, admittedly, adequate. The most appropriate reference in this regard is to the Report drawn up by the de Larosière HighLevel Group,213 tasked with identifying the causes of that crisis. One of its main conclusions, stated in para. 151 of the Report, was that poor supervisory organization or unduly intrusive supervisory rules and practices translate into costs for the financial sector and, in turn, for customers, taxpayers, and the wider economy, and that hence supervision should be carried out as effectively as possible and at the lowest possible cost.214 [19.129] The provisions of MiFID II on public enforcement, laid down in Articles 67–88 and presented above, address three of these elements. The safeguarding of competent authorities’ institutional, personal, and functional independence and of proper accountability, as well as competent authorities’ responsibility, are outside its scope. An assessment of these provisions (some of which contain, to a significant extent, several key elements of MiFID I, which is to be repealed) results in the following remarks. [19.130] As stated in Recital 137, second sentence: ‘a common minimum set of powers coupled with adequate resources should guarantee supervisory effectiveness’. Under MiFID II, guaranteeing supervision effectiveness is sought by means of enhancing the competent authorities’

supervisory powers,215 while the established redress procedures act as a safeguard.216 On the second aspect, however, which also partly covers the aspect of financial independence, MiFID II is silent, since the adequacy of resourcing still remains a national discretion, with Member States’ competent authorities (still) being on an unequal footing. [19.131] The second element, that of efficiency in the exercise of sanctioning powers, is a cornerstone of MiFID II and, in the author’s view, its most important innovative element, with the establishment of the extensive new framework on competent authorities’ power to impose administrative sanctions and measures, its exercise, and the publication of decisions to impose such sanctions and measures.217 The established redress procedures act as a safeguard in this case as well. Given the current position of significant divergence of national rules, and despite the multitude of national discretions, the Directive paves the way for higher degrees of efficiency. [19.132] Finally, the effectiveness of cooperation arrangements both between national supervisory authorities in the same Member State and in relation to cross-border issues is addressed in Article 68218 and Articles 79– 88,219 respectively. In addition, by partly addressing the potential for regulatory arbitrage, Recital 139 states that no action taken by any competent authority or ESMA in the performance of their duties should directly or indirectly discriminate against any Member State or group thereof as a venue for the provision of investment services and activities in any currency. [19.133] There is no doubt, however, that in this respect the catalytic role for efficiency and effectiveness remains that of ESMA. [19.134] A final remark and point of general concern: as in the case of any new regulatory and/or supervisory framework, complacency as a result of the establishment of new institutions and/or the adoption of new rules in capital markets may lead to a short-sighted approach and exacerbate the issue of designing effective long-term policies. Equally problematic is the eventuality of the framework shortly proving inadequate, and thus necessitating readjustment. This usually results in heightened uncertainty in

terms of the adequacy of interventions, both undermining trust on the part of investors and increasing the operational cost of supervised entities (which is rolled over to either the consumers or the shareholders and leads to a squandering of funds), without, on the other hand, (sufficiently) achieving the goals of public intervention.

1

[2004] OJ L 145/1. See on this, by mere indication, the various contributions in Guido Ferrarini and Eddy Wymeersch (eds) Investor Protection in Europe: Corporate Law Making, the MiFID and Beyond (Oxford/New York: Oxford University Press, 2006); Emilios Avgouleas (ed.) The Regulation of Investment Services in Europe under MiFID: Implementation and Practice (West Sussex: Tottel Publishing, 2008). 2 The author defines EU capital markets law as the branch of EU financial law containing rules aimed at ensuring the liberalization within the EU of trade in investment services; the stability of capital markets; the protection of investors and of capital markets’ integrity, efficiency, and transparency; and the operation of adequate investor compensation schemes. 3 [2012] OJ C 326/47. 4 [2014] OJ L 173/349. 5 [2014] OJ L 173/84. 6 MiFID II, Article 97, and MiFIR, Article 55, last subpara. 7 MiFID II, Article 96, and MiFIR, Article 55, first subpara. 8 MiFID II, Article 93(1), first subpara, first sentence, and second subpara, respectively. 9 On this TFEU Article see, by mere indication Martin Schlag, ‘Artikel 53 des AEUV’ in EU-Kommentar, 3, Herausgeber Jürgen Schwarze, Ulrich Becker, Armin Hatje, and Johann Schoo (Baden-Baden: Auflage, Nomos Verlagsgesellshaft, 2012), pp. 809–18. 10 On this TFEU Article see, by mere indication Hans-Holger Herrnfeld, ‘Artikel 114 des AEUV’ in Schwarze, Becker, Hatje, and Schoo (n. 9), 1420–76, as well as Paul Craig and Graine de Búrca, EU Law: Texts, Cases, and Materials (6th edn, Oxford/New York: Oxford University Press, 2015), pp. 93–4 and 614–20. 11 On various aspects of the twin legal acts see Nis Jul Clausen, ‘Reforming the Regulation of Trading Venues in the EU under the Proposed (MiFID II): Leveling the Playing Field and Overcoming Market Fragmentation?’ (2012) Nordic & European Company Law LSN Research Paper Series, no. 10-23; Guido Ferrarini and Paolo Saguato, ‘Reforming Securities and Derivatives Trading in the EU: from EMIR to MiFIR’ (2013) Journal of Corporate Law Studies 13(2), 319–59; Rik Mellenbergh, ‘MiFID II: New Governance Rules in Relation to Investment Firms’ (2014) European Company Law 11, 172–7; Athanasios Panagopoulos, Thomas Chatzigagios, and Ioannis Dokas, ‘The Main Effects of MiFID on European Capital Markets and European Integration’ (2015) International Journal of Business Administration 6(5), 52–62; Jürgen Vandenbroucke,

‘(Non-) Complexity Through the Eyes of MiFID’ (2014) European Journal of Law and Economics 37(3), 477–88; Gaetane Willemaers, ‘Client Protection on European Financial Markets—From Inform Your Client to Know Your Product and Beyond: An Assessment of the PRIIPs Regulation, MiFID II/MiFIR and IMD 2’ (Autumn 2014) Revue Trimestrielle de Droit Financier and the other contributions in this volume. 12 The term ‘investment firm’ is defined, in detail, in Article 4(1), point (1) MiFID II. 13 The term ‘market operator’ is defined (in Article 4(1), point (18)) as meaning a person or persons managing and/or operating the business of a regulated market and may be the regulated market itself. 14 MiFID II, Article 4(1), point (63). 15 These providers are defined in Article 4(1), points (52)–(54). 16 The term ‘third-country firm’ is defined (in Article 4(1), point (57)) as meaning a firm which, if its head office or registered office were located within the EU, would be an investment firm or a credit institution providing investment services or performing investment activities. 17 The term ‘investment services and activities’ is defined (Article 4(1), point (2)) as meaning any of the services and activities listed in section A of Annex I relating to any of the instruments listed in section C of Annex I. 18 The term ‘branch’ is defined (in Article 4(1), point (30)) as meaning a place of business other than the head office which is a part of an investment firm, which has no legal personality, and which provides investment services and/or activities, and which may also perform ancillary services for which the investment firm has been authorized; all the places of business set up in the same Member State by an investment firm with headquarters in another Member State shall be regarded as a single branch. 19 MiFID II, Article 1(1). 20 The term ‘regulated market’ is defined (in Article 4(1), point (21)) as meaning a multilateral system operated and/or managed by a market operator, which brings together or facilitates the bringing together of multiple third-party buying and selling interests in financial instruments—in the system and in accordance with its non-discretionary rules—in a way that results in a contract, in respect of the financial instruments admitted to trading under its rules and/or systems, and which is authorized and functions regularly in accordance with Articles 44–56. Other ‘trading venues’ (Article 4(1), point (24)) are:

• the ‘multilateral trading facilities’ (MTFs), defined (Article 4(1), point (22)) as meaning multilateral systems, operated by an investment firm or a market operator, which bring together multiple third-party buying and selling interests in financial instruments—in the system and in accordance with non-discretionary rules—in a way that results in a contract in accordance with the provisions of Articles 5–43; and

• the ‘organized trading facilities’ (OTFs), defined (Article (1), point (23)) as meaning multilateral systems which are not a regulated market or an MTF and in which multiple third-party buying and selling interests in bonds, structured finance products, emission allowances, or derivatives are able to interact in a way that results in a contract also in accordance with the same above-mentioned provisions. 21

MiFID II, Article 1(2). The term ‘credit institution’ is defined (Article 4(1), point (27)) with reference to Article 4(1), point (1), of Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012 (CRR) [2013] OJ L 176/1. 23 [2013] OJ L 176/338. 24 The term ‘ancillary services’ is defined (Article 4(1), point (3)) as meaning any of the services listed in Section B of Annex I. 25 MiFID II, Article 1(3). 26 MiFID II, Article 1(4). The term ‘structured deposit’ is defined (Article 4(1), point (43)) as meaning a deposit as defined in point (c) of Article 2(1) of Dir 2014/49/EU of the European Parliament and of the Council of 16 April 2014 on deposit guarantee schemes (recast) [2014] OJ L 173/149, which is fully repayable at maturity on terms under which interest or a premium will be paid or is at risk, according to a formula involving factors such as those mentioned in that point. 27 On this, see in particular, paragraphs 19.43–19.45. 28 On these acts see Paul Craig, The Lisbon Treaty: Law, Politics and Treaty Reform (Oxford/New York: Oxford University Press, 2010), pp. 57–66 and 260–82, and Johann Schoo, ‘Artikel 289–291 des AEUV’ in Schwarze, Becker, Hatje, and Schoo (n. 9), 2337– 44. 29 [2010] OJ L 331/84. 30 The term ‘single rulebook’ is commonly used to refer to the total harmonization of rules pertaining to the micro- and macro-prudential regulation and the micro-prudential supervision of financial firms. The term was first introduced in June 2009, when the European Council called for the establishment of a ‘European single rulebook applicable to all financial institutions in the Single Market’, that is, a single set of harmonized prudential rules (European Council Conclusions, 18/19 June 2009, 11225/2/09 REV 2, para. 20, first sentence: ). 31 The asymmetry in the length of these three subsections is due to the relative size of the provisions addressed therein. 32 MiFID II, Article 67(1). 33 MiFID II, Article 67(2), first subpara (equivalent to Article 48(2) MiFID I). It is noteworthy that, under Article 48(2) MiFID I, tasks can also be delegated to other entities if that is expressly provided for in Articles 5(5), 16(3), and 17(2) on investment firms 22

providing only investment advice or (in the first case) the service of reception and transmission of orders. 34 Those conditions must include a clause obliging the entity in question to act and be organized in such a manner as to avoid conflict of interest and so that information obtained from carrying out the delegated tasks is not used unfairly or to prevent competition. 35 MiFID II, Article 67(2), second subpara. 36 MiFID II, Article 67(2), third subpara. 37 MiFID II, Article 67(3). 38 MiFID I, Article 48. 39 The phrase ‘credit and other financial institutions’ used in this article is not compatible with EU banking law, whereby financial institutions are defined as precluding credit institutions. The beginning of Article 4(1), point (26) of the CRR reads as follows: ‘“Financial institution” means an undertaking other than an institution (i.e. a credit institution or an investment firm according to point (3)) ….’ 40 [2009] OJ L 302/32. The term UCITS is defined in Article 1(2) of that Directive. 41 MiFID II, Article 68. 42 MiFID I, Article 49. 43 MiFID II, Article 76(1). 44 MiFID II, Article 76(3). 45 MiFID II, Article 76(2). 46 AIFs are defined in Article 4(1), point (a) of Directive 2011/61/EU of the European Parliament and of the Council of 8 June 2011 on alternative investment fund managers … [2011] OJ L 174/1, which governs their operation, as meaning collective investment undertakings not requiring authorization pursuant to Article 5 of the UCITS IV Directive. 47 CCPs are defined (Article 4(1), point (51)) with reference to Article 2, point (1) of Regulation (EU) No 648/2012 of the same institutions of 4 July 2012 on over-the-counter (OTC) derivatives, central counterparties and trade repositories [2012] OJ L 201/1, which governs their operation. 48 CSDs are defined in Article 2(1), point (1) of Regulation (EU) No 909/2014 of the same institutions of 23 July 2014 on improving securities settlement in the EU and on central securities depositories … [2014] OJ L 257/1, which governs their operation. 49 MiFID II, Article 76(4)–(5). On the latter aspect, see Section III.1.C. 50 MiFID I, Article 54. 51 [2006] OJ L 157/87. 52 [2013] OJ L 182/19. This Article provides that statutory auditors or audit firms may carry out statutory audits (on the basis of Directive 2006/43/EC) of the financial statements of public-interest entities and medium-sized and large undertakings. They may also express an opinion on whether the management report is consistent with the financial statements for the same financial year, and whether the management report has been prepared in accordance with the applicable legal requirements, state whether they identified material

misstatements in the management report, and give an indication of the nature of any such misstatements. 53 Under this Article, statutory auditors or audit firms may audit the accounting information given in the annual report of the UCITS’ investment company and management companies. 54 MiFID II, Article 77(1). 55 MiFID II, Article 77(2). 56 MiFID I, Article 55. 57 Under MiFID I, this subject-matter is governed by Article 50. 58 MiFID II, Article 69(1). 59 MiFID II, Article 69(2), second and third subparas. 60 MiFID II, Article 69(2), first subpara. The list of these powers is definitely more extensive than the one in Article 50(2) MiFID I. 61 The term ‘commodity derivative’ is defined (Article 4(1), point (50)) with reference to Article 2(1), point (30) MiFIR. 62 [2012] OJ C 326/391. 63 This aspect is governed by Regulation (EU) No 596/2014 on market abuse (MAR) [2014] OJ L 173/1, and Directive 2014/57/EU on criminal sanctions for market abuse (MAD II) of the European Parliament and of the Council of 16 April 2014 [2014] OJ L 173/179 (also ‘twin legal acts’ in their field of application). It is worth mentioning that, unlike MiFID II and MiFIR, the MAD II introduces, for the first time, EU rules on criminal sanctions for market abuse, albeit on the basis of minimum harmonization. See on this Michael Faure and Claire Leger, ‘Towards a Harmonization of Insider Trading Criminal Law at EU Level?’ (2014) available at: , and Willemaers (n. 11). 64 The term ‘financial instrument’ is defined (Article 4(1), point (15)) as meaning the instruments specified in S C of Annex I, MiFID II. 65 On these two aspects, see also paragraph 19.88. 66 See paragraph 19.12. 67 MiFID II, Article 72(1) and Recital 138. 68 [1995] OJ L 281/31. 69 [2001] OJ L 8/1. 70 It is noteworthy that under MiFID I this subject matter is governed only by the two (short) paragraphs of Article 51, while under MiFID II (as discussed below in this subsection) it is covered by the lengthy Articles 70–71, as well as by Article 72 (the first paragraph of which applies also to supervisory powers, as already mentioned). 71 MiFID II, Article 70(1), first subpara, first sentence. 72 MiFID II, Article 70(1), first subpara, second sentence, and Recital 141, first sentence. On Article 70(3)–(5) see paragraph 19.46. 73 MiFID II, Article 70(1), first subpara, first sentence, and Recital 149. 74 MiFID II, Recital 141, first sentence.

75

The term ‘management body’ is defined (Article 4(1), point (36)) as meaning the body or bodies of an investment firm, market operator, or data-reporting services provider, which are appointed in accordance with national law, which are empowered to set the entity’s strategy, objectives, and overall direction, and which oversee and monitor management decision-making, including persons who effectively direct the entity’s business. 76 MiFID II, Article 70(2). 77 MiFID II, Article 70(1), second subpara. 78 On this aspect, see Section IV.1. 79 MiFID II, Article 70(1), third subpara. 80 MiFID II, Article 70(3)–(5). 81 This aspect is governed by Articles 5 or 6(2), or Articles 34–5, 39, 44, and 59 MiFID II, and by Article 7(1), third sentence, or 11(1) MiFIR. 82 On this aspect of Article 69, see paragraph 19.31. 83 MiFID II, Article 70(6). 84 See paragraphs 19.55–19.61. 85 By mistake, Article 70(6) makes reference to Article 43 without stating that this applies to third-country firms. 86 By mistake again, Article 70(6) makes reference to Article 65 instead of Article 62. 87 Specific rules apply if the legal person is a parent undertaking or a subsidiary of the parent undertaking preparing consolidated financial accounts in accordance with Directive 2013/34/EU (Article 70(6), point (f) in finem). The terms ‘parent undertaking’ and ‘subsidiary’ are defined (Article 4(1), points (32) and (33)) with reference to Articles 2(9)– (10) and 22 of that Directive. 88 MiFID II, Article 70(7). 89 See paragraphs 19.37–19.39. 90 MiFID II, Article 72(1). 91 MiFID II, Article 72(2) and Recital 145. 92 MiFID II, Article 71(1), first subpara. 93 MiFID II, Article 71(1), second and third subparas. 94 See paragraphs 19.78–19.79. 95 MiFID II, Article 71(2). 96 MiFID II, Article 71(3), first subpara. 97 MiFID II, Recital 146, seventh sentence. 98 ESMA/2015/1858, ‘Draft implementing technical standards under MiFID II: Final Report’, ESMA (11 December 2015), , pp. 149– 61. 99 MiFID II, Article 71(7), third subpara. Under Article 15(1) ESMA Regulation upon submission of the ITS by ESMA, the Commission has three months (a period extendable by one additional month) to decide whether to endorse them. On this Article see Eddy

Wymeersch, ‘The European Financial Supervisory Authorities or ESAs’, in Eddy Wymeersch, Klaus Hopt, and Guido Ferrarini (eds) Financial Regulation and Supervision: A Post-Crisis Analysis (Oxford: Oxford University Press, 2012), Ch. 9, pp. 254–5. 100 MiFID II, Article 71(4), first subpara, first and second sentences. 101 Draft ITS, Article 5 and Recital 4. 102 MiFID II, Article 71(4), first subpara, second sentence. 103 MiFID II, Article 71(5). 104 See . 105 MiFID II, Article 71(6). 106 See paragraph 19.58. It is worth noting that both Article 71(3) and the draft ITS apply only to sanctions and not to measures imposed but not published. 107 MiFID II, Article 71(3), second subpara, first sentence, and Recital 146, sixth sentence. 108 Draft ITS, Article 2 and 4(1). 109 MiFID II, Article 71(3), second subpara, third and fourth sentences. 110 See paragraph 19.61. 111 MiFID II, Article 71(3), second subpara, second to fourth sentences, and draft ITS, Article 4(2). 112 MiFID II, Article 71(4), second subpara, first sentence. 113 Draft ITS, Articles 2 and 6. 114 MiFID II, Article 71(4), second subpara, second sentence. 115 MiFID II, Article 71(5). 116 MiFID II, Article 71(6). 117 MiFID II, Article 73(1) and Recital 147. 118 MiFID II, Article 73(2). 119 MiFID II, Recital 147, in finem. 120 MiFID II, Article 74(1). On this see also Articles 6 and 7(2)–(3). 121 MiFID II, Article 74(2). 122 MiFID I, Article 52. 123 [1998] OJ L 115/31. 124 [2001] OJ L 109/56. 125 MiFID II, Article 75(1)–(2) and Recital 151. On the FIN-Net see: . 126 MiFID II, Article 75(3). 127 MiFID I, Article 53. 128 MiFID II, Article 79(1), first subpara. With the exceptions mentioned hereinafter, Article 79 is based on Article 56 MiFID I. 129 MiFID II, Article 87 (almost identical to Article 62a MiFID I). On Articles 35–6 ESMA Regulation see Wymeersch (n. 99), 278–9 and 285 (at 9.195), respectively. 130 See paragraphs 19.43–19.45.

131

MiFID II, Article 79(1), second subpara. This is a new rule in MiFID II. On this aspect see paragraphs 19.99–19.100. 133 On this aspect see paragraphs 19.101–19.107. 134 MiFID II, Article 79(1), third to fifth subparas. 135 MiFID II, Article 79(3). 136 MiFID II, Article 79(4). 137 See paragraph 19.33. 138 MiFID II, Article 79(5) (new provision). When an action relates to wholesale energy products, the competent authority must also notify the Agency for the Cooperation of Energy Regulators (ACER) established under Regulation (EC) No 713/2009 of the European Parliament and of the Council of 13 July 2009 [2009] OJ L 211/1 (last subpara). 139 In Article 4(1), point (56), the term ‘host Member State’ is defined as meaning the Member State, other than the home Member State, in which an investment firm has a branch or provides investment services and/or activities, or the Member State in which a regulated market provides appropriate arrangements in order to facilitate access to trading on its system by remote members or participants established in that same Member State. 140 The term ‘home Member State’ is defined (Article 4(1), point (55)) differently for investment firms (point (a)), regulated markets (point (b)), and data-reporting services providers (point (c)). 141 MiFID II, Recital 154. 142 MiFID II, Article 79(8). 143 ESMA/2014/1569, ‘Final Report: ESMA’s Technical Advice to the Commission on MiFID II and MiFIR’, ESMA (19 December 2014): ), S 6.3, pp. 377–80. 144 This is consistent with Article 16 of Commission Regulation (EC) 1287/2006 [2006] OJ L 241/1, adopted as an implementing measure by virtue of Article 56(5) MiFID I. 145 ESMA (n. 98), 34–56. The power is conferred on the Commission to adopt these ITSs in accordance with Article 15 ESMA Regulation (Article 79(9), third subpara). 146 Draft ITS, Recital 5. This MoU is available at: . 147 Draft ITS, Article 1(2)–(3). 148 Draft ITS, Recital 6. 149 [2003] OJ L 275/32. 150 [2013] OJ L 347/671. 151 MiFID II, Article 79(6)–(7). Both these rules are new in MiFID II. 152 See paragraph 19.31. 153 MiFID II, Article 80(1). 132

154

MiFID I, Article 57(1). MiFID II, Article 80(2). On Article 21 ESMA Regulation see Wymeersch (n. 99), 282–3. 156 MiFID I, Article 57(2). 157 ESMA/2015/1006, ‘MiFID II/MiFIR draft Technical Standards on authorization, passporting, registration of third-country firms and cooperation between competent authorities: Final Report’, ESMA (29 June 2015): ), pp. 102–9. 158 Draft RTS, Article 1. 159 Draft RTS, Article 80(3), third subpara. Just like in the case of ITSs under Article 15(1) ESMA Regulation, upon submission of the RTS by ESMA, the Commission has three months (a period extendable by one additional month) to decide whether to endorse them (ESMA Regulation, Article 10(1), fifth subpara). On Arts 10–14 ESMA Regulation see Wymeersch (n. 99), 249–54. 160 ESMA (n. 98), 114–35. The power is conferred on the Commission to adopt such ITS in accordance with Article 15 ESMA Regulation (Art 80(4), third subpara). 161 Draft ITS, Article 1. 162 Draft ITS, Articles 2–7. 163 Draft ITS, Articles 8 and 9, respectively. 164 See paragraph 19.85. 165 See paragraphs 19.11–19.15. 166 MiFID II, Article 81(1). 167 MiFID II, Recital 136. 168 See paragraphs 19.24–19.26, and 19.124–19.126, respectively. 169 MiFID II, Article 81(2). 170 See paragraphs 19.19–19.23. By mistake, Article 81(3) makes reference to Article 71 instead of Article 76. 171 MiFID II, Article 81(3). On Articles 74–75 see paragraphs 19.78–19.79 and 19.80– 19.82, respectively. 172 See paragraphs 19.19–19.23, and 19.124–19.126, respectively. 173 [2010] OJ L 331/1. 174 Under Article 127(2), last indent TFEU, the oversight of payment and settlement systems in the euro area is a basic task of the Eurosystem. Nevertheless, it must be clarified that Article 81(5) MiFID II does not apply only to euro area countries (to which Article 127(2) TFEU is applicable), but also to Member States with a derogation. 175 MiFID II, Article 81(5). 176 MiFID I, Article 58. 177 The power is conferred on the Commission to adopt these ITS in accordance with Article 15 ESMA Regulation (Article 81(4), third subpara). 155

178

ESMA (n. 98), 114–35 (see paragraph 19.99). Draft ITS, Article 8. 180 See paragraphs 19.99–19.100 and 19.101–19.108, respectively. 181 MiFID II, Article 82(1). 182 Under this Article, in case of disagreement between competent authorities in crossborder situations, ESMA assists the competent authorities concerned in finding a settlement by mutual agreement. In case of failure, ESMA can adopt a binding decision to settle the matter. See on this Wymeersch (n. 99), 266–71. This is set apart from the nonbinding mediation provided for in Article 31, point (c) ESMA Regulation. 183 MiFID II, Article 82(2). 184 Under this Article, ESMA has the power to investigate and take further action concerning the failure of competent authorities to comply with their obligations under the legislation referred to in Article 1(2) ESMA Regulation. See on this Wymeersch (n. 99), 255–63. 185 See Wymeersch (n. 99), 266 (at 9.120). 186 MiFID I, Articles 58a–59. 187 MiFID II, Article 83. 188 The term ‘control’ is defined (Article 4(1), point (35(b))) as meaning the relationship between a parent undertaking and a subsidiary in the cases referred to in Article 22(1)–(2) of Directive 2013/34/EU. 189 MiFID II, Article 84(1)–(2). 190 The term ‘group’ is defined (Article 4(1), point (32)) with reference to Article 2(11) of Directive 2013/34/EU. 191 MiFID II, Article 84(3). 192 MiFID I, Article 60. 193 ESMA (n. 98), 136–48. The power is conferred on the Commission to adopt these ITSs in accordance with Article 15 ESMA Regulation (Article 84(4), third subpara). 194 See paragraphs 19.100(3) and 19.108. 195 Draft ITS, Article 1. 196 Draft ITS, Articles 2–6. 197 MiFID II, Article 85. 198 MiFID I, Article 61. 199 MiFID II, Article 86(1). 200 MiFID II, Article 86(2). 201 MiFID II, Article 86(3). 202 MiFID II, Article 86(4). 203 MiFID II, Article 86(1)–(3), in finem. On Article 19 ESMA Regulation see paragraphs 19.109–19.111. 204 MiFID I, Article 62. 179

205

On ESMA’s current role in the field of international relations and cooperation see: . 206 It is worth recalling (see introductory remarks, Section I.1) that Norway, Lichtenstein, and Iceland, Member States of the EEA, are not considered third countries. 207 See paragraphs 19.19–19.23. 208 MiFID II, Article 88(1). 209 MiFID II, Article 88(2). 210 MiFID I, Article 63. 211 In this context, the ‘theory of public choice’ addresses the important aspect of supervisory ‘capture’, which may result in weak public responses. 212 For more details on the concept and extent of central bank independence, see, by mere indication, Fabian Amtenbrink, The Democratic Accountability of Central Banks—A Comparative Study of the European Central Bank (Oxford: Hart Publishing, 1999); Central Bank Governance Group, ‘Issues in the Governance of Central Banks’, Bank for International Settlements Report (May 2009) chs 5–6. 213 The High-Level Group on Financial Supervision in the EU, Chaired by Jacques de Larosière, Report, Brussels, 25 February 2009: . This Report is analysed in Jean-Victor Louis, ‘The Implementation of the Larosière Report: A Progress Report’ in Mario Giovanoli and Diego Devos (eds) International Monetary and Financial Law: The Global Crisis (Oxford/New York: Oxford University Press, 2010), Ch 7, pp. 146–76. 214 These arguments are fleshed out mainly with regard to banking supervision, but are still of general application to the financial system, in Christos Gortsos, The Single Supervisory Mechanism (SSM): Legal Aspects of the First Pillar of the European Banking Union (Athens: Nomiki Bibliothiki/European Public Law Organization (EPLO), 2015), pp. 314–21. 215 See Section III.2. 216 See Section III.4. 217 See Section III.3. 218 See paragraphs 19.16–19.18. 219 See Section IV.

20 THE PRIVATE LAW EFFECT OF MIFID I AND MIFID II The Genil Case and Beyond Danny Busch

I. General II. May Civil Courts Be Less Strict Than MiFID I and MiFID II? III. 1. 2. 3.

May Civil Courts Be Stricter Than MiFID I and MiFID II? General Nationale-Nederlanden v Van Leeuwen May Civil Courts Thus Be Stricter Than MiFID?

IV. May Contracting Parties Be Less Strict Than MiFID I and MiFID II? V. May Contracting Parties Be Stricter Than MiFID I and MiFID II? VI. Influence of MIFID I and MiFID II on the Principle of Relativity VII. Influence of MiFID I and MiFID II on Proof of Causation VIII. Influence of MiFID I and MiFID II on a Contractual Limitation or Exclusion of Liability IX. MiFID I and MiFID II Assessments by the Courts of Their Own Motion in Relation to Private Investors?

X. Conclusion

I. General [20.01] As regulatory provisions are in most jurisdictions classified as public law, any failure by an investment firm to comply with one or more regulatory provisions applicable to it will primarily affect its relationship with the competent financial supervisor.1 In other words, the relevant financial supervisor can enforce these provisions under administrative law in the event of an infringement, for example by imposing an administrative fine on the firm.2 However, the regulatory provisions, in particular the conduct-of-business rules under MiFID I and MiFID II,3 also have a major influence on relations between the investment firm and its clients under private law.4 It is now commonly accepted in most European jurisdictions that the regulatory rules help to define the pre-contractual and contractual duty of care of investment firms (and other financial undertakings as well) under private law. Moreover, in many jurisdictions, an infringement of national implementing provisions can constitute not only a breach of the civil duty of care but also a tort (unlawful act) for contravention of a statutory duty. It should also be noted that in the context of institutional portfolio management (for pensions funds, insurers, and so forth) duties of care under public law and other regulatory provisions are regularly explicitly incorporated into the contract, with all the contractual consequences that this entails. Institutional portfolio management contracts routinely include a provision in which the portfolio manager declares that s/he has an authorization from the competent financial supervisor and will at all times comply with the applicable regulatory law.5 [20.02] This chapter examines to what extent the civil courts are bound by MiFID I/MiFID II under European law. In what follows, MiFID I and MiFID II will be jointly referred to as ‘MiFID’.

II. May Civil Courts Be Less Strict Than MiFID I and MiFID II? [20.03] It seems to follow from the Genil case that the European principle of effectiveness (effet utile) prevents the civil courts from imposing private law duties on investment firms that are less strict than that to which they are subject under the MiFID rules.6 In Genil the Court of Justice of the European Union (henceforth ‘the Court of Justice’) held that in the absence of EU legislation it is for the Member States themselves to determine the contractual consequences of non-compliance with the know-your-customer (KYC) rules under MiFID I, but that the principles of equivalence and effectiveness must be observed (para. 57).7 The Court of Justice referred in this connection to para. 27 of a judgment of 19 July 2012 concerning a tax matter (Littlewoods Retail and Others, Case C-591/10) and the case law cited there. This paragraph reads as follows: In the absence of EU legislation, it is for the internal legal order of each Member State to lay down the conditions in which such interest must be paid, particularly the rate of that interest and its method of calculation (simple or ‘compound’ interest). Those conditions must comply with the principles of equivalence and effectiveness; that is to say that they must not be less favourable than those concerning similar claims based on provisions of national law or arranged in such a way as to make the exercise of rights conferred by the EU legal order practically impossible [emphasis added] (see, to that effect, San Giorgio, paragraph 12; Weber’s Wine World, paragraph 103; and Case C-291/03 MyTravel [2005] ECR I-8477, paragraph 17).

[20.04] In the MiFID I context, the principle of effectiveness therefore means that the conditions which an investor must fulfil in order to bring a civil action against an investment firm may not be such that success is practically impossible. The judgment appears to mean, among other things, that civil courts may not be less strict than MiFID I. Where, according to MiFID I, there is non-compliance with KYC rules in a specific case and the aggrieved investor brings a civil action for damages, the civil courts are not permitted to dismiss this claim by arguing that in the particular circumstances it was not necessary to comply with the KYC rules. This would, after all, be at odds with the principle of effectiveness.8 This approach can be extended to claims for damages for non-compliance with other MiFID I provisions, particularly infringements of other conduct-of-

business rules. And the approach can also be extended to MiFID II, especially as under MiFID II the operation of the principle of effectiveness has been explicitly codified in Article 69(2), last paragraph, MiFID II: Member States shall ensure that mechanisms are in place to ensure that compensation may be paid or other remedial action be taken in accordance with national law for any financial loss or damage suffered as a result of an infringement of this Directive or of [MiFIR].9

III. May Civil Courts Be Stricter Than MiFID I and MiFID II? 1. General [20.05] Genil does not seem to provide a definitive answer to the vexed question of whether civil courts may impose stricter duties of care under private law than those resulting from MiFID.10 If a civil court holds, for example, that although an investment firm is admittedly not obliged to comply with KYC rules under MiFID (or indeed with other MiFID rules), it is nonetheless obliged to do so in the particular circumstances of the case because of its civil duty of care, the aggrieved client is not denied a claim on account of non-compliance with MiFID rules. If a civil court is stricter than MiFID, there would not appear to be any conflict with the principle of effectiveness as formulated by the Court of Justice in Genil. It should be noted, however, that the question of whether civil courts may be stricter than MiFID was not at issue in Genil and was therefore not explicitly addressed. Genil dealt, after all, only with the question of the private law consequences of non-compliance with MiFID rules.11 However, this does not exclude the possibility that an argument could be made on the basis of other principles of European law that civil courts may not be stricter than MiFID. The recent judgment of the Court of Justice in the case of Nationale-Nederlanden v Van Leeuwen12 concerning the sale of insurance policies (woekerpolissen) with exorbitant management charges provides some leads in this respect. So this is sufficient reason to pause and consider

this judgment at rather greater length, although it should be noted that it relates to the Third Life Assurance Directive and not to MiFID.

2. Nationale-Nederlanden v Van Leeuwen A. Facts [20.06] In 1999 Mr Van Leeuwen concluded a life assurance contract with Nationale-Nederlanden Assurance forming part of an investment known as ‘flexibly insured investing’. It is evident from the policy dated 29 February 2000 that Nationale-Nederlanden insures a benefit of 255,000 Dutch guilders, or the value of participations in investment funds taken out for Van Leeuwen (plus 10% thereof). Under this contract Mr Van Leeuwen was both the policyholder and the insured. [20.07] If Mr Van Leeuwen dies before 1 December 2033 the contract offers two options. Benefit A is a guaranteed and fixed amount of 255,000 guilders. Benefit B is the (variable) sum of the value of his participations in investment funds (based on the value of those participations) as of the date of his death, plus 10 per cent thereof. If, at the time of his death, benefit B is greater than benefit A, then the higher sum is to be paid to the beneficiaries of his life assurance. Thus, benefit A sets a minimum level for the benefit to be paid out in the case of death prior to 1 December 2033.13 [20.08] The ‘gross premium’ consists of a single payment of 8,800 guilders at the start of the contract and then monthly payments of 200 guilders from the inception date of 1 May 1999. This gross premium is invested in investment funds chosen by the policyholder. Costs such as premiums for the death cover are periodically deducted from the value accrued in this way. These premiums are therefore not charged separately, but—like these costs—form an integral part of the gross premium. [20.09] Before Mr Van Leeuwen concluded this insurance contract with Nationale-Nederlanden, he was supplied with a ‘Proposal for flexibly insured investing’. This proposal contained three scenarios based on different returns and management costs of 0.3 per cent. The text under the heading ‘product return’ contained the following sentence: ‘The difference

between the fund return and the product yield is dependent on the risks insured, the costs payable as well as any additional coverage.’

B. Legal Framework [20.10] Article 31 of the Third Life Assurance Directive14 (which has now been repealed and replaced by a more recent version15) plays a crucial role in this respect and reads as follows: 1. Before the assurance contract is concluded, at least the information listed in Annex II(A) shall be communicated to the policyholder. 2. The policyholder shall be kept informed throughout the term of the contract of any change concerning the information listed in Annex II(B). 3. The Member State of the commitment may require assurance undertakings to furnish information in addition to that listed in Annex II only if it is necessary for a proper understanding by the policyholder of the essential elements of the commitment. 4. The detailed rules for implementing this Article and Annex II shall be laid down by the Member State of the commitment. [20.11] The obligation to furnish the information specified in Annex II to the Third Life Assurance Directive was transposed into Dutch law at that time in Article 2 of the 1998 Regulation regarding the provision of information to policyholders (Regeling informatieverstrekking aan verzekeringnemers 1998). In view of the text of the 1998 Regulation, the Netherlands did not at that time make use of the possibility of imposing a duty to furnish additional information under Article 31(3) of the Third Life Assurance Directive. [20.12] It has been established that Nationale-Nederlanden, in compliance with Article 2(2)(q) and (r) of the 1998 Regulation, furnished the policyholder with information about the effect of of the costs and the risk premiums on the return. However, the policyholder did not receive a summary or full overview of the actual and/or absolute costs and their composition. Nor was this obligatory under the 1998 Regulation. In short, it

has been established that Nationale-Nederlanden furnished the policyholder with all information which it was bound to supply under the 1998 Regulation. [20.13] Nonetheless, in its interim judgment Rotterdam District Court held as follows about the fact that Nationale-Nederlanden had not sent the policyholder a summary or full overview of the actual and/or absolute costs and their composition: Although Nationale-Nederlanden fulfilled the requirements referred to in Article 2(2) (q) and (r) of the 1998 Regulation regarding the provision of information to policyholders, it nonetheless infringed the open rules (including, in this legal action, the general and/or special duty of care owed by Nationale-Nederlanden to Van Leeuwen in the context of their contractual relations, pre-contractual good faith and/or requirements of reasonableness and fairness) by confining the information it furnished to information about the effect of costs and risk premiums on the return.16

Nationale-Nederlanden argued that it could not be required to furnish additional information on the basis of open and/or unwritten rules.

C. Questions Referred for a Preliminary Ruling [20.14] The District Court referred the following two questions to the Court of Justice for a preliminary ruling: 1. Does EU law, and in particular Article 31(3) of the Third Life Assurance Directive, preclude an obligation on the part of a life assurance provider on the basis of the open and/or unwritten rules of Dutch law—such as the reasonableness and fairness17 which govern the contractual and pre-contractual relationship between a life assurance provider and a prospective policyholder, and/or a general and/or specific duty of care—to provide policyholders with more information on costs and risk premiums of the insurance than was prescribed in 1999 by the provisions of Dutch law by which the Third Life Assurance Directive was implemented (in particular, Article 2(2) (q) and (r) of the 1998 Regulation)? 2. Are the consequences, or possible consequences, under Dutch law of a failure to provide that information relevant for the purposes of

answering question 1?

D. Are there Duties to Furnish Additional Information on the Basis of Reasonableness and Fairness? [20.15] The first question referred for preliminary ruling is answered in the affirmative. In short, the civil courts may, by reference to the dictates of reasonableness and fairness under Article 6:2 of the Dutch Civil Code (Burgerlijk Wetboek, DCC) and Article 6:248 DCC,18 impose duties to furnish information additional to that required under the 1998 Regulation, provided that three cumulative conditions are fulfilled (this is a matter for the referring court to decide): (i) the information required must be clear and accurate; (ii) the information required must be necessary to enable the policyholder to understand the essential elements of the commitment; (iii) legal certainty for the insurer is sufficiently safeguarded (paras 21, 29–31, and 33). [20.16] The first two conditions follow from the express wording of Article 31(3) of the Third Life Assurance Directive, Annex II and Recital 23 in the preamble to the Third Life Assurance Directive (para. 21). The third condition expresses the principle of legal certainty under EU law. The Court of Justice holds that the legal basis for the use by the Member State concerned of the possibility provided for in Article 31(3) of the Third Life Assurance Directive must be such that, in accordance with the principle of legal certainty, it enables insurance companies to identify with sufficient foreseeability what additional information they must provide and which the policyholder may expect (para. 29). An additional duty to provide information based on the requirements of reasonableness and fairness under Article 6:2 DCC or Article 6:248 DCC would not seem at first sight to fulfil this requirement since this rule is extremely vague and has little if any predictive value. So that seems to be good news for Nationale-Nederlanden. [20.17] But the Court of Justice then goes on to formulate two arguments that are favourable to the policyholder and unfavourable to Nationale-

Nederlanden. It holds that when deciding whether the legal certainty principle has been fulfilled the national court may (not ‘must’) take into consideration the fact that it is for the insurer to determine the type and characteristics of the insurance products which it offers, so that, in principle, it should be able to identify the characteristics which its products offer and which are likely to justify a need to provide additional information to policyholders (para. 30). In short, the ball is played back into the insurer’s court. It knows best what information it should furnish to its clients in order to ensure that they understand the insurance product. What perhaps played a role in this connection is that, according to the Court of Justice, the fact that the policyholder should receive a summary or full overview of the actual and/or absolute costs and their composition to be able to understand the operation of the product is so apparent that the insurer itself should have realized it was necessary to furnish this information to the policyholder. The Court of Justice adds in this connection that, in accordance with the description of the grounds of the 1998 Regulation, its application is governed, in particular, by the national private law in force, ‘including the requirements of reasonableness and fairness’ set out in Articles 6:2 DCC and 6:248 DCC (para. 31). In short, the Court of Justice clearly considers that Nationale-Nederlanden could and should have known that its responsibility did not begin and end with literal compliance with the 1998 Regulation.

3. May Civil Courts Thus Be Stricter Than MiFID? [20.18] It seems to follow from the Nationale-Nederlanden judgment that EU law is blind to the distinction between public and private law when it comes to implementing rules of EU law (para. 28). After all, the Court of Justice has no problem with the fact that Directives are transposed into national law by a combination of public and private law. Annex II to the Third Life Assurance Directive has been transposed into Dutch law by the 1998 Regulation (public law), whereas the Member State option to furnish additional information may be implemented by means of the requirement of reasonableness and fairness under Article 6:2 DCC (private law), provided that three conditions are fulfilled (see paragraphs 20.15–20.17 above).

[20.19] If it is indeed true that EU law is blind to the distinction between public and private law, this also has an important bearing on whether civil courts may impose stricter standards than the rules under MiFID. For the most part, MiFID provides for maximum harmonization. If EU law is truly blind to the distinction between public and private law when it comes to the transposition of EU legal rules, the maximum harmonization standard also applies to the civil courts. They may not therefore impose stricter duties of care than those that apply under the rules resulting from MiFID. In the above-mentioned Genil judgment about the private law impact of MiFID, the Court of Justice admittedly notes that in the absence of EU legislation it is for the Member States themselves to determine what effect noncompliance with MiFID has under private law (provided that it is not practically impossible to recover compensation for the loss or damage suffered), but this refers to the sanction and not to the legal rule itself.19 If this line of reasoning is rejected because it is considered that the civil courts may be stricter than MiFID, the present judgment in any event shows that legal certainty is an important factor that the civil courts must take into consideration in deciding whether they may impose stricter duties of care than apply under MiFID (see paragraphs 20.15–20.17 above). [20.20] What has been said above can be qualified as follows. MiFID itself also contains open rules. One important rule of this kind is that investment firms must act honestly, fairly, and professionally in accordance with the best interests of their clients (below: duty of honesty).20 This obligation is admittedly translated into more specific rules in MiFID (including KYC rules and duties to furnish information), but the general rule does not coincide with the more detailed provisions. The general duty of honesty therefore leaves some scope for additional duties of care. This scope could be used by the civil courts. By doing so, they would not, strictly speaking, be applying stricter standards than MiFID since they would be using the space provided by MiFID itself. The only question is how much space exactly is left by the open rule, bearing in mind the EU principle of legal certainty. Let us take an example. Under MiFID, warnings may be provided in a standardized format.21 An approach in which the civil courts hold that the special duty of care means that investment firms are obliged to provide express investment risk warnings in terms that are not misleading, and that the investment firm must subsequently check to ensure

that the private investor is actually aware of these risks seems to go further than a standard warning,22 although a standard warning too must naturally be sufficiently clear. Would a civil court then be justified in adopting the following reasoning? The investment firm has discharged its duty to provide a warning in standardized format of the risks of the product and has thus complied with its specific duty to provide information under MiFID. But in view of the general duty of honesty, the investment firm should nonetheless have given an express warning in not misleading terms, and should have subsequently checked to ensure that the private investor was actually aware of these risks. Consequently, the investment firm has breached the general duty of honesty under MiFID and must pay damages to the investor. Reservations based on the EU principle of legal certainty could be expressed about this argument. Nonetheless, the NationaleNederlanden judgment shows that the Court of Justice is prepared to adopt a flexible approach to the principle of legal certainty and does not shun acrobatic reasoning in its efforts to achieve a just result. It remains to be seen, therefore, whether the Court of Justice will actually bar civil courts of the Member States from using the argument that investment firms have a general duty of honesty under MiFID as a ground for requiring them to issue personalized rather than standardized warnings on the risk.

IV. May Contracting Parties Be Less Strict Than MiFID I and MiFID II? [20.21] Do contractual provisions that are less strict than MiFID actually have an effect?23 In Genil it was held that although in the absence of European legislation it is admittedly for the Member States themselves to determine the contractual consequences of non-compliance with the MiFID rules, one of the principles that must be observed is the principle of effectiveness. As noted above, the principle of effectiveness has been explicitly codified in Article 69(2), last paragraph, MiFID II. The principle of effectiveness means in this connection that the conditions on which an investor can bring a civil claim against an investment firm may not be such that successful legal actions are practically impossible. Naturally, however, the argument is less strong in cases where the civil courts, regardless of the

contractual provisions, wish to be less strict than MiFID (see Section II above)—the investor has, after all, himself agreed to the contract. On the other hand, private investors in particular often have little influence over the contractual conditions. The effectiveness principle could therefore be cited in support of the argument that the civil courts are obliged to hold that the relevant contractual provision is unacceptable, for example (depending on the applicable private law) according to the criteria of reasonableness and fairness or, if included in general terms and conditions, it constitutes an unreasonably onerous provision. This goes further, by the way, than an assessment by the courts of their own motion since in the above approach the result of the assessment is also predetermined. The subject of assessments by the court of their own motion is dealt with in Section IX below.

V. May Contracting Parties Be Stricter Than MiFID I and MiFID II? [20.22] Do contractual provisions that are stricter than MiFID actually produce an effect?24 At first sight, it would seem that there can be little objection to such provisions since they can only benefit investor protection. Moreover, unlike the situation where civil courts, regardless of the contract, impose stricter duties of care than apply under the MiFID rules (see Section III), legal certainty is not at issue here. After all, the investment firm voluntarily submits to stricter duties of care. Nonetheless, if investment firms in a particular Member State were to voluntarily submit on a large scale to stricter duties of care, for example pursuant to local market usage, this might jeopardize the European level playing field. I should add, however, that in my view this is a rather theoretical argument. [20.23] Just as in connection with the question of whether civil courts may be stricter than MiFID, Genil does not seem to provide a definitive answer to whether contractual provisions that are stricter than MiFID actually produce an effect. In such a case, a client’s claim is in any event not rejected on the grounds of non-compliance with MiFID rules. If contracting parties themselves are stricter than MiFID, there would not

seem to be any conflict with the principle of effectiveness, as formulated by the Court of Justice in Genil. [20.24] Could it perhaps be reasoned on the basis of the NationaleNederlanden case that the civil courts are bound to hold that where a contractual provision is stricter than MiFID it is to this extent unacceptable according to, for example (depending on the applicable private law) the criteria of reasonableness and fairness or, if included in general terms and conditions, constitutes an unreasonably onerous provision? Although it may be possible to draw such a conclusion from a strictly logical approach, I must confess to having some difficulty with it. [20.25] To start with, one of the key objectives of MiFID is to offer investors a high level of protection.25 If an investment firm voluntarily submits to stricter contractual rules than apply under MiFID, there could surely be little objection to this. [20.26] Moreover, offering contractual conditions that go further than MiFID is one of the ways in which an investment firm can compete with its rivals. To this extent the question goes to the root of free enterprise. If an entrepreneur wishes to do more than he is obliged to do by law, this must be possible. Another factor here is that the freedom to conduct a business is included in the Charter of Fundamental Rights of the European Union and is therefore a principle that forms part of the European legal order.26 [20.27] Finally, a client may have valid reasons for requesting that an investment firm submit contractually to rules that are stricter than those applying under MiFID. For example, under the Dutch supervision rules contained in the Pensions Act (Pensioenwet) and the Occupational Pension Scheme (Obligatory Membership) Act (Wet verplichte beroepspensioenregeling), pension funds are permitted to outsource their portfolio management to one or more external asset managers (investment firms), but in doing so are required to ensure that the external portfolio manager complies with the rules applicable to them.27 Insofar as is relevant here, these rules mean that outsourcing to an external portfolio manager is permitted only if the contract regulating the outsourcing or portfolio management meets certain requirements, for example that the external

portfolio manager must enable the pension fund at all times to comply with the provisions laid down by or pursuant to the Pensions Act or the Occupational Pension Scheme (Obligatory Membership) Act.28 Naturally, any such contractual obligation to which the external portfolio manager concerned is subject does not result from MiFID and may to this extent be stricter than the obligations to which it is subject under MiFID.29

VI. Influence of MiFID I and MiFID II on the Principle of Relativity [20.28] In some European jurisdictions (such as Germany and the Netherlands) a tort claim based on breach of statutory duty cannot succeed in the absence of ‘proximity’ or ‘relativity’, which means that the relevant duty must not only serve the general interest, but also the claimant’s patrimonial interests. In the jurisdictions imposing a relativity requirement the question therefore arises of whether the relativity requirement is met in the case of a breach of MiFID duties.30 [20.29] It is apparent from Genil that in the absence of EU legislation it is admittedly for the Member States themselves to determine the contractual consequences of non-compliance with MiFID rules, but one of the principles that must be observed is the principle of effectiveness. According to this principle, the conditions to be fulfilled by an investor in bringing a civil action against an investment firm may not be such as to virtually exclude the possibility of success. As noted above, the principle of effectiveness has been explicitly codified in Article 69(2), last paragraph, MIFID II. In my view, Genil and Article 69(2), last paragraph, MIFID II mean that in view of the principle of effectiveness a claim for damages on account of an infringement of MiFID rules, in particular the conduct-ofbusiness rules, may not fail by virtue of the requirement of relativity.

VII. Influence of MiFID I and MiFID II on Proof of Causation

[20.30] In Europe (and beyond), it is a universal requirement that a causal connection must be established between an investment firm’s breach of duty (be it in tort, contract, or otherwise) and the loss suffered by the client. As a rule, the client claiming damages has the burden of proof with respect to this requirement. However, especially in the case of duties to furnish information or duties to warn, which may or may not be MiFID-derived, proof of this requirement is often problematic. After all, an investment firm may argue that there is no causal connection between the breach and the loss suffered because the client would have made the same investment decision had the manager complied with its duties to provide information and its duties to warn.31 In view of this, another interesting question about the effect of MiFID in private law concerns the influence of MiFID on proving causal link. [20.31] To answer this question, it is worthwhile to consider a wellknown judgment of the Dutch Supreme Court concerning the internet company World Online.32 That case concerned loss allegedly suffered by investors in World Online as a consequence, among other things, of a misleading prospectus published on the occasion of the company’s flotation. In brief, the Supreme Court held as follows. As it is often hard to prove a condition sine qua non link (a ‘but for’ link) in relation to liability for a prospectus, the investor protection which the EU Prospectus Directive is intended to provide may prove illusory in practice. Although this Directive admittedly contains detailed provisions about what information must be included in the prospectus, it does not regulate civil liability if the prospectus is misleading because it is incomplete or incorrect. It does, however, require the Member States to ensure that their laws, regulations, and administrative provisions on civil liability apply to those persons responsible for the information given in a prospectus (Article 6(2), first paragraph). According to the Supreme Court, this means that effective legal protection must be provided in accordance with the rules of national law.33 The Supreme Court also held as follows: With a view to that effective legal protection and having regard to the protection which the prospectus rules are intended to provide for investors and potential investors against misleading information in the prospectus, the basic principle should be that a conditio sine qua non link (‘but for’ link) must exist between the misleading information and the decision to invest. This means that it must be assumed, in

principle, that if there had been no misleading information the investor would not have proceeded with the purchase of the securities or, in the event of purchase on the secondary market, would not have done so on the same conditions. However, the court may conclude—from the submissions of the parties (either party being free to prove the correctness of the alleged facts) and any further information available—that the basic principle referred to above does not apply in the specific circumstances. This will be the case, for example, if it is shown that the investment decision was made before disclosure of the misleading information. It should be noted by the way that as professional investors have knowledge and experience of the relevant market and of analysing the available information, there will generally be more reason than in the case of retail investors to conclude that, despite the misleading information in the prospectus, they were not actually influenced by it in making their investment decision.34

[20.32] In short, the basic assumption of the Supreme Court is that a causal link exists between the misleading information and the investment decision. In the case of a professional investor, however, the court may be justified in concluding, on the basis of his knowledge and experience, that he was not actually influenced by the misleading information. In such case, it is possible to revert to the basic rule that the investor bears the burden of proving the causal link. [20.33] A question that arises in connection with the issue of how MiFID influences the proof of causal link is whether the reasoning of the Supreme Court in its World Online judgment, as set out above, could also be applied if investment firms neglect to comply adequately with one or more duties under MiFID to provide information or warnings. My own view is that, subject to a few adjustments, this is indeed a legitimate argument. As already explained, one of the key objectives of MiFID is investor protection.35 Although MiFID does not contain any provision comparable to Article 6(2) of the Prospectus Directive, we may assume that the EU legislator intends Member States to offer effective legal protection in relation to the MiFID rules as well. The principle of effectiveness is, after all, a fundamental principle of EU law.36 Genil and Article 69(2), last paragraph, MiFID II provide support for this notion. It is apparent from the judgment, after all, that in the absence of EU legislation it is admittedly for the Member States themselves to determine the contractual consequences of non-compliance with MiFID obligations, but that one of the principles to be observed is the principle of effectiveness (para. 57). As noted previously,

the principle of effectiveness has been explicitly codified in Article 69(2), last paragraph, MiFID II. The principle of effectiveness means in this connection that the conditions on which an investor can bring a civil claim against an investment firm may not be such as to virtually exclude the possibility of bringing a successful legal action. [20.34] To provide effective legal protection (to use the terminology of the World Online judgment), it is legitimate, in my view, to argue that where the duty under MiFID to provide information and warnings is infringed, the basic rule should be that a causal (‘but for’) link exists between the infringement of the rule and the loss as the investor protection intended by MiFID may otherwise in practice prove to be illusory.37 [20.35] In keeping with the World Online judgment, an exception could be made in the case of professional investors since it could be concluded on the basis of their knowledge and experience that they are not actually misled by the incorrect information into making their investment decision. However, this exception may be less appropriate in the event of noncompliance with duties to provide information and warnings under MiFID.38 After all, the provisions of MiFID on investment firms make a clear distinction between duties to provide information and warnings to retail clients on one hand and professional clients on the other. The duties under MiFID to provide information and warnings to professional investors are geared to their specific information needs. In the event of noncompliance with one or more of these duties, it is reasonable to suppose that the investment decision of the professional client may have been influenced by this. It therefore seems legitimate to argue that even where an investment firm infringes its duty under MiFID to provide information or warnings to professional clients, the basic principle must be that a causal connection exists between the infringement and the loss. However, whether this approach would be followed by the Supreme Court or other civil courts is at present unclear. Naturally, other approaches which help the client to prove a causal connection may also be in keeping with the principle of effectiveness.39

VIII. Influence of MiFID I and MiFID II on a Contractual Limitation or Exclusion of Liability [20.36] The principle of effectiveness as formulated in Genil and in Article 69(2), last paragraph, MiFID II could be used to argue that in relation to consumers and small businesses the civil courts are obliged to hold that a contractual clause excluding or limiting liability for an infringement of MiFID rules constitutes an unreasonably onerous provision if included in the general terms and conditions, and that the contractual clause does not therefore prevent a claim for damages on account of noncompliance with the MiFID rules. Likewise, it could be argued that the civil courts are obliged to hold that a contractual clause that seeks to exclude or limit liability for infringement of the MiFID rules is unacceptable according to (depending on the applicable private law) the requirements of reasonableness and fairness and that the contractual clause does not therefore prevent a claim for damages on account of non-compliance with the MiFID rules (even in relation to clients other than consumers and small businesses). [20.37] The principle of effectiveness as formulated in Genil and in Article 69(2), last paragraph, MiFID II means, after all, that the national conditions which an investor must fulfil in order to bring a civil action against an investment firm for infringement of MiFID obligations may not be such that success is practically impossible. It can be argued that this also means that contractual clauses that seek to exclude or limit liability for infringement of MiFID rules are contrary to the principle of effectiveness. Naturally, however, the argument is less strong in cases where the civil courts, regardless of the contractual provisions, are less strict than MiFID. After all, the client has himself agreed to the contractual clause. On the other hand, retail clients in particular often have little influence over the contractual conditions. Arguments that also carry weight are, naturally, that clauses of this kind jeopardize the high level of investor protection which MiFID intends to provide, and also detract from the level playing field envisaged by MiFID. An example may help to clarify this. Article 14(1) of the MiFID I Implementing Directive provides that

Member States shall ensure that, when investment firms outsource critical or important operational functions or any investment services or activities, the firms remain fully responsible for discharging all of their obligations under [MiFID I].40

[20.38] It follows, for example, that where a portfolio manager outsources part of the management to a third party (e.g. a more specialized portfolio manager), it remains fully responsible (despite the outsourcing) for observance of the regulatory provisions applicable to the outsourced activities under MiFID. In short, if the third party infringes conduct-ofbusiness rules under MiFID during these activities and the portfolio manager’s client suffers loss as a result, it can be argued that, according to the principle of effectiveness, the civil courts are obliged in relation to consumers and small businesses to hold that a contractual provision limiting the liability of the portfolio manager to carefully selecting third parties (including independent agents to whom activities have been outsourced) and excluding his liability for infringements of MiFID rules by a third party to whom aspects of the portfolio management have been outsourced constitutes an unreasonably onerous condition if included in general terms and conditions. Likewise (depending on the applicable private law), it can be argued that the civil courts are obliged here to hold that the contractual clause is unacceptable in light of the requirements of reasonableness and fairness and is not therefore a bar to a claim for damages for infringement of the MiFID rules. This goes further, by the way, than an assessment by the courts of their own motion since in the above approach the result of the assessment is also predetermined. The subject of assessments by the courts of their own motion is dealt with in the following section.

IX. MiFID I and MiFID II Assessments by the Courts of Their Own Motion in Relation to Private Investors? [20.39] This brings me, to what I regard as an intriguing question that has a bearing on the intensity with which MiFID impacts on private law. At present, the parties to a legal action are often unaware that they could invoke an infringement of MiFID (conduct-of-business) rules. Are the civil courts obliged in such cases to determine of their own motion whether the MiFID (conduct-of-business) rules have been infringed? I would certainly

not exclude this possibility. It is apparent from the settled case law of the Court of Justice that the national courts must determine of their own motion whether, on the basis of the European principle of effectiveness, unreasonably onerous clauses in contracts between businesses and consumers are ‘unfair’ within the meaning of Directive 93/13/EEC. The Court of Justice may also direct the civil courts to determine of their own motion whether the legislation is applicable.41 Indeed, it would seem to be extending the protection to the entire field of consumer protection Directives. Recently, the Court of Justice gave such a direction in the case of the Consumer Purchases Directive.42 In any event, the MiFID conductof-business rules can, in my view, be treated as consumer protection provisions insofar as they must be observed in relation to private investors.43 National civil courts should in that case determine of their own motion whether there has been an infringement of MiFID conduct-ofbusiness rules in disputes between investment firms and private investors.

X. Conclusion [20.40] The last word has not been spoken about the effect of MiFID on private law. MiFID II is as unclear about this as MiFID I. Although the possible contours are somewhat clearer as a result of the judgments in the Genil and Nationale-Nederlanden cases, the Court of Justice has not yet explicitly answered the main questions. For more definitive answers, it will be necessary to await the further judgments of the Court of Justice.

1

Note that in Germany a minority view in the legal literature qualifies the MiFID conduct-of-business rules (which most clearly pursue investor protection) as norms with a dual legal nature (Doppelnatur), with the effect that they qualify not only as regulatory rules, but also as private law norms. See inter alia M. Casper and C. Altgen, ‘Chapter 4: Germany’ in D. Busch and D. A. DeMott (eds) Liability of Asset Managers (Oxford: OUP 2012), § 4.23; R. Veil, ‘Anlageberatung im Zeitalter der MiFID: Inhalt und Konzeption der Pflichten und Grundlagen einer zivielrechlichen Haftung’ (2007) Wertpapier-Mitteilungen (WM) 1821, 1825–26; T. Weichert and T. Wenninger, ‘Die Neuregelung der Erkundigungsund Aufklärungspflichten von Wertpapierdienstleistungsunternehmen gem. Art. 19 RiL

2004/39/EG (MiFID) und Finanzmarkt-Richtlinie-Umsetzungsgesetz’ (2007) WM 627, 635. One German author even advances the view that the MiFID conduct-of-business rules qualify solely as private law norms because they place an obligation on a private firm towards its clients. See D. Einsele, ‘Anlegerschutz durch Information und Beratung: Verhaltens- und Schadenserzatzpflichten der Wertpapierdienstleistungsunternehmen nach Umsetzung der Finanzmarktrichtlinie (MiFID)’ (2008) Juristen Zeitung (JZ) 477, 482. In Italy, the MiFID duties have a dual nature because they are considered both public and private law duties that an asset manager owes its clients. See P. Giudici and M. Bet, ‘Chapter 5: Italy’ in Busch and DeMott (eds) ibid., § 5.42. 2 The same applies if a regulated market infringes MiFID rules. 3 Directive 2004/39/EC, OJ L 145, 30 April 2004, pp. 1–47 (MiFID I); Commission Directive 2006/73/EC, OJ L 241, 2 September 2006, pp. 26–58 (MiFID I Implementing Directive); Commission Regulation (EC) no. 1287/2006, OJ L 241, 2 September 2006, pp. 1–25 (MiFID I Implementing Regulation). MiFID I and the MiFID I Implementing Directive have been transposed into national legislation in the various Member States of the European Union (EU) and the European Economic Area (EEA). The MiFID II regime consists of (1) Directive 2014/65/EU, OJ L 173, 15 May 2014, pp. 349–496 (MiFID II); (2) Regulation (EU) No. 600/2014, OJ L 173, 15 May 2014, pp. 84–148 (MiFIR); and (3) an impressive number of implementing measures. The relevant Directives pertaining to MiFID II will in a similar fashion as MiFID I be transposed into national law. 4 The same applies to the relationship between regulated markets and their participants. 5 See for a comparative overview of the civil law effect of MiFID in several European jurisdictions: Busch and DeMott (n. 1); D. Busch, ‘Why MiFID Matters to Private Law: The Example of MiFID’s Impact on an Asset Managers’ Civil Liability’ (2012) CMLJ 386–413. 6 At least prior to the Genil case, this question had hardly been addressed in the legal literature across Europe, let alone in case law. Nevertheless, there has been some discussion of this question in Germany, where some authors advance the view that the civil courts are allowed to be less demanding in the circumstances of a specific case. See A. Fuchs, Wertpapiergesetz (2009) Vor ss 31ff para. 61. Other German authors submit that the civil courts are not so permitted, because in their view MiFID provides minimum standards in civil law. See E. Schwark, in E. Schwark and S. Zimmer (eds) Kapitalmarktrechtskommentar (4th edn, 2010), Vor ss 31 ff WpHG para. 16. Two Luxembourg authors have explicitly addressed this question as well. At first sight the strict separation between public law duties and civil law duties under Luxembourg law suggests that the civil courts are a priori free to be less demanding. However, it is difficult to conceive that the duties under public and civil law could be totally inconsistent. It is the view of these authors that a court might not decide that there is neither a duty of best execution nor a duty to assess the client (suitability test) in civil law. Given that MiFID is largely a maximum harmonization regime and that according to Luxembourg law the MiFID regime probably has a

public order character, these authors hold the view that the civil courts may not subject asset managers to private law duties that are less strict than the relevant regulatory duties. See I. Riassetto and J.-F. Richard, ‘Chapter 6: Luxembourg’ in Busch and DeMott (n. 1), § 6.63. 7 EU CoJ 30 May 2013, no. C-604/11, AA (2013) 663, with note by Busch; JOR 2013/274, with note by Busch (Genil 48 SL and Other v Bankinter SA and Others). 9 For a different view, see O. Eloot and H. Tilley, ‘Beleggersbescherming in MiFID II en MiFIR’ (2014) Droit Bancaire et Financier, 179–201, 200. 8 As regards the question of how the principle of effectiveness affects the impact of EU law on private law in a general sense, see, e.g., T. Tridimas, The General Principles of EU Law (Oxford: Oxford University Press, 2006), pp. 418–76; W. van Gerven, ‘Of Rights, Remedies and Procedures’ (2000) CMLR pp. 501–36. 10 The answer to the question whether the civil courts may be stricter than MiFID differs across Europe. In addition, in many jurisdictions the answer is simply not clear. See for a comparative overview Busch (n. 5), 394–8 (with further references). 11 In the same sense, see C. Herresthal, ‘Zu den Auswirkungen der MiFID auf das nationale Vertragsrecht’ (2013) ZIP, 1420–2; J. Lieder, ‘EuGH: Anlageberatung bei Zinsswaps’ (2013) LMK 349–404. 12 EU CoJ 29 April 2015, no. C-51/13, AA (2015) 696, with note by Busch and Arons (Nationale-Nederlanden Levensverzekering Mij NV/Hubertus Wilhelminus van Leeuwen). 13 See Opinion of Advocate-General Sharpston, 12 June 2014, Case C-51/13, ECLI:EU:C:2014:1921, para. 15. 14 Directive 92/96/EEC, OJ L 360, 9 December 1992, pp. 1–27. 15 See the present judgment, para. 3. 16 Rotterdam District Court 28 November 2012, ECLI:NL:RBROT:2012:BY5159, para. 2.9. 17 In Dutch: redelijkheid en billijkheid. 18 Article 6:2 DCC reads as follows: (1) A creditor and debtor must, as between themselves, act in accordance with the requirements of reasonableness and fairness. (2) A rule binding upon them by virtue of law, usage or legal act does not apply to the extent that in the given circumstances, this would be unacceptable according to criteria of reasonableness and fairness. See also Article 6:248 DCC: A contract has not only the legal effects agreed to by the parties, but also those which, according to the nature of the contract, result from the law, usage or the requirements of reasonableness and fairness. (2) A rule binding upon the parties as a result of the contract does not apply to the extent that, in the given circumstances, this would be unacceptable according tot he criteria of reasonableness and fairness.

19

EU CoJ 30 May 2013, no. C-604/11, AA (2013) 663, with note by Busch; JOR 2013/274, with note by Busch (Genil 48 SL and Others v Bankinter SA and Others). 20 Article 19(1) MiFID I; Article 24(1) MiFID II. 21 Article 19(3), last sentence, MiFID I. Note that this becomes a Member State option under MiFID II: the Member States may allow the information to be provided in a standardized format (see Article 24(5), last sentence, MIFID II). In short, if a Member State does not allow this, it seems as though the information must always be provided in a personalized format. It seems from the MiFID II Consultation Document that this Member State option is exercised (implicitly) in, for example, the Netherlands. The relevant Dutch implementing provision (Article 4:20(6) Wft) is not altered in the MiFID II Consultation Document, and the Explanatory Memorandum to the Bill to implement MiFID II is also silent on this point. Both documents may be downloaded at . It will therefore remain possible in the Netherlands to provide information in standardized format. The situation will undoubtedly be different in at least a few other Member States. If the Member States had unanimously considered that information could be provided in standardized format, a compromise in the form of a Member State option would have been unnecessary. 22 For this approach in relation to private investors, see Dutch Supreme Court (Hoge Raad, HR) case law: HR 3 February 2012, NJ 2012/95; AA (2012) 752, with note by Busch; JOR 2012/116, with note by Van Baalen (Coöperatieve Rabobank Vaart en Vecht UA v X) (duty of care in relation to the provision of investment advice), para. 3.6.2. It should be noted that these (and other) judgments of the Dutch Supreme Court about the duty to provide warnings relate, without exception, to the pre-MiFID era. Whether the Supreme Court will continue this line of reasoning under MiFID remains to be seen. 23 Most jurisdictions tend towards ineffectiveness in one way or another of contractual clauses setting lower standards than those following from MiFID. Nevertheless, in at least Ireland, Spain, Luxembourg, the Netherlands, and Germany, the answer is open to doubt. See for a comparative overview Busch (n. 5), 399–403 (with further references). 24 The question of whether the contracting parties may be stricter than MiFID has not been much addressed in the legal literature across Europe, let alone in case law. Nevertheless, in Germany, Poland, and Luxembourg there are some authors who have addressed this question explicitly. In Germany, some authors have advanced the view that it follows from the principle of freedom of contract that contractual clauses setting higher standards than those following from MiFID are as a general rule effective. See I. Koller, in H.D. Assmann and U.H. Schneider (eds) Wertpapierhandelsgesetz (5th edn, 2009), Vor s 31 para. 5; Einsele (n. 1), 481. In Poland, one author is similarly of the view that the courts will likely uphold contractual clauses that set higher standards than MiFID. This author submits that a reasonable legislator should not prevent market participants from voluntarily raising the bar, although he realizes that this approach may in theory have a negative impact on the European level playing field. See A. W. Kawecki, ‘Chapter 8: Poland’ in Busch and DeMott (eds) (n. 1), § 8.50–8.52. According to two Luxembourg authors, at first sight it may be argued under Luxembourg law that provisions setting higher contractual

standards than MiFID should be valid because of the strict separation between public law and civil law duties based on case law denying the right of the client to claim damages in the case of a violation of a public law duty. However, according to these authors it is difficult to conceive that the duties under public law and civil law could be totally inconsistent. In contradistinction to the German and Polish authors addressing this question, they are therefore of the view that they can reasonably assert that the public order character of public law rules of conduct, combined with the logic of maximum harmonization of MiFID rules, should impose limits on the possibility of setting higher contractual standards than MiFID. Article 6 Luxembourg Civil Code, which refers to public policy, might be the ground of these limits. See I. Riassetto and J.-F. Richard, ‘Chapter 6: Luxembourg’ in Busch and DeMott (eds) (n. 1), § 6.74. 25 See Recital 2 MiFID and Recital 70 MiFID II. 26 Article 16 EU Charter: ‘The freedom to conduct a business in accordance with Union law and national laws and practices is recognized.’ As regards the significance of the EU Charter for financial supervision law, see Eva Dieben, ‘Vijf jaar bindend EU-Handvest en het financieel toezichtrecht’ in Janneke Gerards, Henri de Waele, and Karin Zwaan (eds) Vijf jaar bindend EU Grondrechtenhandvest (Deventer: Wolters Kluwer, 2015), pp. 277– 350. 27 See Section 34(1) Pensions Act and Section 43(1) Occupational Pension Scheme (Obligatory Membership) Act. These provisions are elaborated in Chapter 4 of the Decree implementing the Pensions Act and the Occupational Pension Scheme (Obligatory Membership) Act. 28 Article 13(2)(e) Decree implementing the Pensions Act and the Occupational Pension Scheme (Obligatory Membership) Act. 29 As regards outsourcing by pension funds under Dutch law, see, e.g., J. A. M. I. Hoens, ‘Uitbesteding: een achilleshiel in de Pensioenwet?’ (2009) P&P, 16–22; R. H. Maatman and J. W. van Miltenburg, ‘Pensioenfondsen’ in D. Busch, D. R. Doorenbos, C. M. Grundmann-van de Krol, R. H. Maatman, M. P. Nieuwe Weme, and W. A. K. Rank (eds) Onderneming en financieel toezicht (Onderneming en Recht no. 57), (2nd edn, Deventer: Kluwer, 2010), 323–59, 339–42; P. Laaper, Uitbesteding in de financiële sector – in het bijzonder van vermogensbeheer door pensioenfondsen (Onderneming en Recht no. 88) (PhD Nijmegen), (Deventer: Kluwer 2015). 30 See for a comparative overview Busch (n. 5), 404–6 (with further references). 31 See for a comparative overview Busch (n. 5), 406–12 (with further references). 34 See HR 27 November 2009, NJ 2014/201, with notes by Du Perron, AA (2010) Raaijmakers, 336; JOR 2010/43, and Frielink (Vereniging van Effectenbezitters an Others v World Online International NV), para. 4.11.2. As regards this aspect of the judgment, see A. C. W. Pijls and W. H. van Boom ‘Handhaving prospectusaansprakelijkheid niet illusoir: vermoeden van causaal verband bij prospectusaansprakelijkheid’ (2010) WPNR 6834, 194– 200; B. J. de Jong, ‘Liability for Misrepresentation: European Lessons on Causation from the Netherlands’ (2011) ECFR 352–75, 364–6.

32

HR 27 November 2009, NJ 2014/201, with notes by Du Perron, AA (2010) 336, Raaijmakers, JOR 2010/43, and Frielink (Vereniging van Effectenbezitters and Others v World Online International NV). 33 See HR 27 November 2009, NJ 2014/201, with notes by Du Perron, AA (2010) 336, Raaijmakers, JOR 2010/43, and Frielink (Vereniging van Effectenbezitters and Others v World Online International NV), para. 4.11.1. 35 See Recital 2 MiFID I and Recital 70 MiFID II. 36 As regards the question of how the principle of effectiveness affects the impact of EU law on private law in a general sense, see e.g. Tridimas (n. 8), 418–76; van Gerven (n. 8), 501–36. 37 See D. Busch. ‘Drie fundamentele vragen voor de Europese Commissie’ (2010) Ondernemingsrecht, 294–5, 295. The following authors also consider that the Supreme Court’s approach in World Online could conceivably be applied to information duties of European origin other than the duties under the Prospectus Directive: Pijls and van Boom (n. 34), at 199 (referring to the duties mentioned in Annex II to the Unfair Commercial Practices Directive; de Jong (n. 34), 271–2; G. T. J. Hoff, ‘Civielrechtelijke aansprakelijkheid van uitgevende instellingen bij niet-naleving van de openbaarmakingsplicht van koersgevoelige informatie’ in D. Busch, C. J. M. Klaassen, and T. M. C. Arons (eds) Aansprakelijkheid in de financiële sector (Onderneming en Recht no. 78), (Deventer: Kluwer, 2013), pp. 711–72, 760–70 (both refer to the Transparency Directive and the obligation to provide immediate disclosure of price-sensitive information under the Market Abuse Directive. For similar reasoning in relation to MiFID, see: Den Bosch Court of Appeal 15 April 2014, JOR 2014/168, with note by Van der Wiel and Wijnberg; Ondernemingsrecht 2014/92, with note by Arons (Holding Westkant B.V., in liquidation v ABN AMRO Bank N.V.), para. 4.11.15. 38 In D. Busch, ‘Het “civiel effect” van MiFID: Europese invloed op aansprakelijkheid van vermogensbeheerders’ (2012) Ondernemingsrecht 67–78, 77, I was still inclined to argue for such an exception. However, I qualified this in the main body of the text. See also Busch (n. 5), 408–409; D. Busch, ‘De invloed van MiFID op aansprakelijkheid in de Europese financiële sector’ in Busch, Klaassen, and Arons (eds) (n. 37), pp. 11–70, 51–2. 39 On this last point, see C. J. M. Klaassen, ‘Bewijs van causaal verband tussen beweerdelijk geleden beleggingsschade en schending van een informatie- of waarschuwingsplicht’ in Busch, Klaassen, and Arons (eds) (n. 37), 127–74, 151. 40 Emphasis added. See also Article 31(1), first sentence, Draft Commission Delegated Regulation, C(2016) 2398 final, 25 April 2016. 41 See EC CoJ 26 October 2006, no. C-168/05, NJ 2007/201, with note by Mok (Mostaza Claro); EC CoJ 4 June 2009, no. C-243/08, NJ 2009/395, with note by Mok (Pannon); EC CoJ 6 October 2009, no. C-40/08, NJ 2010/11 (Asturcom); EU CoJ 30 May 2013, NJ 2013/487, with note by Mok (Asbeek Brusse and De Man Garabito). 42 On this point see A. S. Hartkamp, ‘Ambtshalve toepassing van Europees consumentenrecht. Een nieuw hoofdstuk: de richtlijn consumentenkoop’ (2015) AA, 222.

43

One of the key objectives of MIFID is to offer a high level of investor protection. See Recital 2 MiFID I and Recital 70 MiFID II.

PART V

THE BROADER VIEW AND THE FUTURE OF MIFID

21 MIFID II IN RELATION TO OTHER INVESTOR PROTECTION REGULATION Picking Up The Crumbs of a Piecemeal Approach Veerle Colaert1

I. Introduction II. 1. 2. 3.

MiFID II versus IDD Background The Structure of the IDD Conduct of Business Rules Main Differences Between IDD and MiFID II Conduct of Business Rules 4. Conclusion on the MiFID II–IDD Nexus

III. MiFID II and IDD versus UCITS KII and PRIIPs KID 1. MiFID II and IDD Information Obligations on Costs versus UCITS KII and PRIIPs KID 2. Product Risk Assessment 3. Conclusion with Respect to the MiFID II/IDD–PRIIPs/UCITS Nexus IV. Cross-Selling Practices V. Conclusion

I. Introduction

[21.01] The overall EU legislative approach is—necessarily—piecemeal: the EU has only those competences that are conferred by the Treaties and may, even then, only intervene if it is able to act more effectively than EU countries at their respective national or local levels.2 [21.02] This has resulted in a body of EU legislation lacking common definitions and principles. As the case may be, national legislators need to integrate the different pieces of EU legislation dealing with closely related topics into their relevant national legislation. This, however, only works if they fit together properly. Because of the EU piecemeal approach, this is not always the case. The study of the interaction between different pieces of EU legislation therefore becomes increasingly important—and difficult. [21.03] MiFID II3 having been dissected and analysed in all its aspects in the other chapters of this book, this contribution seeks to take a step back and consider the relationship of the MiFID II conduct of business rules with other relevant pieces of financial legislation. In Section II the relationship with the very similar conduct of business rules of the Insurance Distribution Directive4 is examined. The interaction between information obligations in the UCITS Directive5 and the PRIIPs Regulation6 on the one hand, and the MiFID II and IDD conduct of business rules on the other hand, is the subject of Section III. In Section IV the different rules for cross-selling practices in the financial sector are scrutinized. This overview allows the author to draw conclusions on ways to improve consistency between different pieces of EU financial regulation (Section V).

II. MiFID II versus IDD 1. Background [21.04] Both MiFID II and the Insurance Distribution Directive (IDD) provide for conduct of business rules, which—although highly similar—are not identical. The reason for this half-hearted duplication of conduct of business rules can be found in the EU piecemeal approach.

MiFID I applied to the provision of ‘investment services’7 or ‘ancillary services’8 relating to ‘financial instruments’. Financial instruments were defined by an exhaustive list, including transferable securities,9 moneymarket instruments,10 units in collective investment undertakings, and different kinds of derivative contracts.11 Not covered by MiFID I were, among other things, deposits and insurance products. [21.05] There is evidence that, after the entry into force of MiFID I, mutual funds were often repackaged as life insurance products or as structured deposits in order to avoid the MiFID regime.12 This tendency to structure a product, not to meet economic needs or investor preferences, but merely in order to circumvent certain legislation, is an instance of ‘regulatory arbitrage’. Many Member States therefore soon deemed the scope of application of MiFID I too limited and provided for national extensions of the MiFID conduct of business rules to certain insurance products and/or structured deposits, or even duplicated MiFID-like rules into their insurance laws.13 [21.06] MiFID II has consolidated this Member State evolution. The scope of application of certain MiFID II provisions, including the conduct of business rules, has been expanded to also cover structured deposits.14 The EU legislature recognized that also for insurance-based investment products the same level of investor protection was needed in order to maintain a level playing field.15 Those products have not been included into the scope of MiFID II since they were already covered by Insurance Mediation Directive 2002/92/EC (IMD). The IMD however merely provided for some basic information and know-your-customer (KYC) obligations.16 In order to avoid regulatory arbitrage between products subject to MiFID II conduct of business rules and similar products which only needed to comply with the much lighter IMD regime, MiFID II amended the IMD by introducing a separate chapter with ‘MiFID-like’ conflict of interest and conduct of business rules for ‘insurance-based investment products’ (so-called IMD 1.5).17 Meanwhile the IMD has been fully revised and renamed IDD (‘Insurance Distribution Directive’).18 The conduct of business rules in IDD have been aligned to their MiFID II counterpart to a much greater extent.19

This, however, does not mean that the MiFID II conduct of business rules have been copy-pasted into IDD. There are substantial differences in the order, terminology, and even content of the conduct of business rules in MiFID II and IDD.20

2. The Structure of the IDD Conduct of Business Rules [21.07] The pendant of the MiFID II conduct of business rules can be found in two different IDD chapters. IDD indeed distinguishes between ‘information requirements and conduct of business rules’, applicable to all insurance products and services, on the one hand (Chapter V) and ‘additional requirements in relation to insurance-based investment products’ on the other hand (Chapter VI).

A. IDD General Conduct of Business Rules [21.08] Three of the IDD conduct of business rules that apply to all insurance products and services closely mimic their MiFID II counterpart. This is the case for, first, the general duty to act honestly, fairly, and professionally in accordance with the best interests of the customer.21 Second, the rules on product oversight and governance requirements of Article 25 IDD are clearly equivalent to the MiFID II organizational rules of Article 16(3). However, IDD has no separate conduct of business rule on product governance, equivalent to Article 24(2) MiFID II. Only one additional sentence in Article 25 IDD is inspired by Article 24(2) MiFID II, stating that investment firms22 have to ‘take reasonable steps to ensure that the insurance product is distributed to the identified target market’. Third, the IDD requirement to only provide fair, clear, and not misleading information (Article 1 (2)) clearly corresponds to Article 24(2) MiFID II. The IDD provision, however, explicitly refers to the Unfair Commercial Practices Directive,23 whereas the MiFID II provision does not.24 Further ‘information conditions’ are provided in Article 23 IDD, dealing with elements such as the media through which information can be provided.25 MiFID II does not feature such ‘information conditions’.26 The MiFID II

Delegated Regulation, however, further details the information requirements27 and also features certain information conditions, though these are quite different from the IDD conditions.28 [21.09] The IDD approach to cross-selling practices29 has, in the last stage of the legislative process, been adapted in line with the MiFID II approach.30 Both directives thus provide for extra information obligations and KYC requirements for products or services offered together with another service or product.31 The exact rules, terminology, and definitions, however, still differ quite a lot (see paragraphs 21.52 et seq). [21.10] Two other general IDD conduct of business rules differ significantly from their MiFID II counterparts. This is the case for information obligations and conflict of interest procedures. The general information obligations of Articles 18 and 20–22 IDD, including the obligation for insurance firms to produce a PRIIPs-like standardized insurance product information document for non-life insurance products, are quite different from the MiFID II information obligations of Article 24(4). They moreover explicitly aim at minimum harmonization only,32 whereas the MiFID II conduct of business rules are generally considered to aim at a very high degree of harmonization, leaving only very limited room for Member States to impose additional requirements.33The rules on conflicts of interests and transparency of Article 19 IDD are only a far cry from the MiFID II rules with respect to inducements and independent advice. IDD, however, also features additional information and conflicts of interests provisions for insurance-based investment products only, which are closer to the MiFID II equivalent rule (see paragraph 21.13). In addition, more detailed IDD rules regarding inducements and independent advice apply to insurance-based investment products only (see paragraphs 21.18–21.19). [21.11] The general IDD conduct of business rules finally also feature the —to insurance intermediaries well-known—need analysis: the requirement for insurance distributors to specify, on the basis of information obtained from the customer¸ ‘the demands and the needs’ of that customer.34 The need analysis is insurance-specific and has no MiFID II equivalent. For

insurance-based investment products, one may wonder, however, whether and how the need analysis interacts with the suitability test, which features an obligation to analyse ‘the objectives’ of the client. The question was considered by the Belgian Financial Services and Markets Authority (FSMA), since the Belgian implementation of the MiFID I KYC obligations had already been extended to the insurance sector. The FSMA ruled that an insurance services provider which complies with the MiFID I suitability test can reasonably assume that the product or service in question also meets ‘the demands and needs’ of that client and therefore complies with the need analysis. According to the FSMA, no separate need analysis should therefore be performed in such circumstances.35 This position has however been criticized by insurance practitioners, who claim that the need analysis typically aims at avoiding a client being over- or under- or inadequately insured. The need analysis would thus gear the insurance product or service to the insured risk, whereas the suitability test gears the product or service to the qualities and objectives of the client. In the context of insurancebased investment products, however, the two positions do not seem to differ that much. Insurance-based investment products are investment products rather than insurance products. Analysing whether the client would be overor under-insured when buying such a product therefore seems pointless. Analysing whether such a product would be adequate for the client would come very close to an analysis of the investment objectives of the client, though. If the FSMA interpretation is correct, and both analyses (the need analysis and the analysis of the objectives of the client under the suitability test) would indeed overlap in this manner, the explicit mentioning in Article 30 §1 IDD that the suitability test for insurance-based investment products is without prejudice to the need analysis of Article 20 §1, would have little meaning. It would be useful to receive further guidance from ESMA and EIOPA on this matter.

B. Additional IDD Conduct of Business Requirements for Insurance-Based Investment Products Only [21.12] Next to the general conduct of business rules applicable to all insurance products and services, IDD also features additional requirements in relation to insurance-based investment products only.

[21.13] Five sets of IDD additional conduct of business rules for insurance-based investment products are similar to their MiFID II counterparts. The KYC rules,36 record-keeping, and reporting obligations37 closely mimic their MiFID II counterparts.38 Also the additional IDD provisions dealing with conflicts of interest39 are largely the same as the MiFID II organizational rules in this regard.40 IDD has, however, not copied the explicit MiFID II reference to inducements as a source of conflicts of interests, which fits the different IDD approach regarding inducements (see paragraph 21.18). Finally, the additional information obligations for insurance-based investment products41 are to a large extent based on the MiFID II information obligations.42 There are two differences though: the MiFID II information obligations regarding ‘independent advice’ have not been copied—which fits the different IDD approach to independent advice (see paragraph 21.19)—and there is no IDD equivalent to the MiFID II exception for investment services offered as part of a financial product which is already subject to other provisions of Union law relating to credit institutions and consumer credits.43 [21.14] In contrast, the additional conduct of business rules on inducements44 and regarding (independent) advice45 are very different from their MiFID II counterparts (see paragraphs 21.18–21.19).

3. Main Differences Between IDD and MiFID II Conduct of Business Rules [21.15] Although the above overview shows a considerable alignment between the IDD and MiFID II conduct of business rules, many differences persist. [21.16] Sometimes these differences are minor and seem the result of translation inaccuracies rather than substantiated choices.46 Other differences consist of small additions or deletions. Article 17(2) IDD for instance states that this article is ‘without prejudice to Directive 2005/29/EC’, whereas no such confirmation features in the equivalent article 24(3) MiFID II. It can be assumed, nevertheless, that Article 24(3)

MiFID II also applies without prejudice to Directive 2005/29/EC. Likewise Article 25(2) IDD states that ‘the product approval process shall be proportionate and appropriate to the nature of the insurance product’. The equivalent Article 16(3) of MiFID II does not feature this sentence. It seems plausible though that the same proportionality and appropriateness requirement should apply to the MiFID II product-approval process. Although these small terminological differences and additions do not seem to have much impact on the meaning of the provision, they are regrettable, as they raise unnecessary interpretation questions. [21.17] Some rules do, however, differ substantially. Those differences seem to be the result of lobbying or lack of agreement to apply rules as strict as the MiFID II rules to insurance products, manufacturers, and distributors. [21.18] The rules on inducements in IDD offer a startling example. MiFID II as a principle prohibits all inducements, considering investment firms receiving inducements as ‘not fulfilling their obligations’ to act honestly, fairly, and professionally in accordance with the best interests of the client.47 MiFID II only provides some limited exceptions to this principal ban on inducements (the exceptions being even more limited when independent investment advice and portfolio management are provided).48 IDD starts from exactly the opposite premise: insurance intermediaries or insurance undertakings receiving inducements are regarded as ‘fulfilling their obligations’, if these inducements do not have a detrimental impact on the quality of the service and do not impair compliance with the duty to act honestly, fairly, and professionally in accordance with the best interests of the customer.49 IDD does add that its provisions on inducements only aim at minimum harmonization: Member States may impose stricter requirements on distributors and in particular additionally prohibit or further restrict the offer or acceptance of fees, commissions, or non-monetary benefits from third parties in relation to the provision of insurance advice.50 [21.19] IDD and MiFID II show a similar difference in approach with respect to the provision of ‘independent advice’. MiFID II provides for strict rules for investment firms who inform their clients that they provide

‘independent advice’: they have to assess a sufficiently wide variety of products and need to comply with the enhanced inducements regime.51 The IDD variant is again much less compelling: Member States do not need to introduce rules regarding ‘independent’ advice; they ‘may’ however require that where an insurance intermediary informs the client that advice is given independently, it should assess a sufficiently wide variety of insurance products.52 A stricter inducements regime is not required. [21.20] The latter differences in approach are substantial and contrary to the spirit of creating a level playing field between investment products which the European legislature considers to be substitutes from an economic perspective.53 Those differences may indeed still provide a fertile breeding ground for regulatory arbitrage: in Member States that do not adopt stricter rules than IDD, investor protection rules for insurance-based investment products will be much less burdensome than for MiFID II financial instruments (such as investment funds). [21.21] It is further remarkable that if a Member State introduces stricter requirements with regard to inducements and/or independent advice, these requirements have to be complied with by all insurance intermediaries or insurance undertakings, including those operating under the freedom to provide services or the freedom of establishment, when concluding insurance contracts with customers having their habitual residence or establishment in that Member State.54 This means that IDD provides for an important exception to the principle of home state control in this area, and thereby seriously hampers the creation of an internal market. [21.22] Finally, beyond these differences resulting from a rule-by-rule comparison, there are also more general differences between the MiFID II and the IDD approach to investor protection. [21.23] First, as is clear from the above discussion, the IDD conduct of business rules aim at minimum harmonization, whereas the MiFID II conduct of business rules in principle aim at maximum harmonization. National implementation rules of the IDD conduct of business rules may therefore deviate even more from each other, from IDD, and from the MiFID II framework. Member States may, on the other hand, also use this

freedom to align their national conduct of business regime for (certain) insurance products and services to the MiFID II regime. [21.24] Second, MiFID makes an explicit and well-defined distinction between retail and professional investors, which the IDD does not make. IDD only refers to a Member State’s option to soften certain information requirements in relations with professional clients, and empowers the Commission to adopt delegated acts with respect to the KYC requirements, taking into account, among other things, ‘the retail or professional nature of the customer or potential customer’.55 If insurance-based investment products are indeed substitutes for many MiFID financial instruments, this different approach seems hard to justify.56

4. Conclusion on the MiFID II–IDD Nexus [21.25] Even though the European legislature is convinced of the risks of regulatory arbitrage, it attempts to deal with them within the pre-existing sectoral framework as far as conduct of business rules are concerned. We believe that this approach testifies to short-termism, as it hampers the creation of a true level playing field for all investment products in the broad sense.57 First, substantial differences remain between IDD and MiFID II conduct of business rules. Second, even for rules that are are fully equivalent today, future developments in MiFID II would need to be meticulously transposed into IDD and vice versa. Third, EU financial legislation is interpreted and regulated by sectorally divided European Supervisory Authorities (ESAs): MiFID II and IDD are the working field of ESMA and EIOPA respectively, advising the Commission on Level 2 legislation and creating important Level 3 guidance.58 Even though an attempt has been made to level the legislative playing field with respect to conduct of business rules in the MiFID II and the IDD Level 1 directives, the danger of divergent interpretations and application at Levels 2 and 3 is real.59 [21.26] A true level playing field would require that the rules are the same, unless there is an objectively justifiable reason to introduce different rules or terminology. It would therefore have been more effective if MiFID

II had been converted into a cross-sectoral piece of legislation dealing with conduct of business rules for all investment products in the broad sense, including insurance-based investment products. IDD could in addition still apply to these insurance-based investment products, but only to cover specific insurance aspects if needed.

III. MiFID II and IDD versus UCITS KII and PRIIPs KID [21.27] A second example of the EU piecemeal legislative approach giving rise to cross-legislation interpretation issues relates to the interplay between the PRIIPs and UCITS key information documents and the MiFID II and IDD conduct of business rules. Two such issues are developed below. First the interaction between the cost information provided in the PRIIPs KID and the UCITS KII on the one hand and the cost information requirements under MiFID II and IDD will be scrutinized. Then the question will be raised as to what extent the KID and KII product risk assessment method could or should be used for purposes of the MiFID II and IDD product governance process and suitability test.

1. MiFID II and IDD Information Obligations on Costs versus UCITS KII and PRIIPs KID [21.28] The PRIIPs Regulation, adopted in November 2014, introduces a ‘Key Investor Document’ (KID) for all ‘Packaged Retail Investment and Insurance-Based Products’, similar to the ‘Key Investor Information’ which had been introduced for UCITS funds three years earlier.60 The KID and KII are short information documents,61 which should increase product transparency and investor understanding, and allow investors easy comparison. They are to be produced by the product manufacturer, but any person advising on or selling a UCITS fund or a PRIIP should provide investors with the KII or KID.62

[21.29] A key section of both the PRIIPs KID and the UCITS KII deals with the costs of the product.63 MiFID II—and for insurance-based investment products, IDD—adds that the client should receive, at least annually, aggregated information about all costs and charges, including costs and charges in connection with the service and the product, which are not caused by the occurrence of underlying market risk. Where the client so requests, an itemized breakdown should be provided.64 The MiFID II Delegated Regulation further details these information requirements,65 whereas IDD has not empowered the Commission to adopt delegated acts in this respect. [21.30] Even apart from such differences between MiFID II and IDD with respect to information on costs, several problems arise in respect of the interplay between the UCITS Directive and the PRIIPs Regulation on the one hand and the MiFID II and IDD information obligations on the other. When services providers need to disclose product information to clients,66 they can in principle rely on the PRIIPs KID or the UCITS KII for information on product costs.67 To the extent there is no KID or KII, or the required MiFID II or IDD information does not feature in the KID or the KII, the investment firm will, however, need to bridge this gap. [21.31] A first example relates to information on transaction costs. A PRIIPs manufacturer should disclose information on transactions costs in the PRIIPs KID. UCITS providers, however, are not obliged to disclose precise quantitative information on transaction costs in the KII. ESMA believes that this problem should be solved in the UCITS implementing rules,68 but is of the opinion that in the meantime investment firms should liaise with UCITS management companies to obtain the relevant information.69 The sector has obviously reacted strongly against this point of view.70 It has nevertheless been inserted in the Proposal for a Commission Delegated Regulation, which, more generally, requires investment firms to calculate and disclose product costs and charges that are not included in the UCTIS KIID, for example, by liaising with UCITS management companies to obtain the relevant information.71

[21.32] Second, the MiFID II measures on cost disclosure apply to all categories of clients.72 This however means that in situations where MiFID II requires product distributors to provide information on product costs,73 but those products are not targeted at retail investors—and therefore no PRIIPs KID is available—the product distributor still needs to obtain the information on product costs from the product manufacturer in order to comply with the MiFID II requirements.74 The same problem may arise under IDD, which does not even make a distinction between customer categories. [21.33] Finally, MiFID II information requirements not only apply in respect of the structured products in scope of the PRIIPs Regulation or the UCITS Directive, but also in respect of all ‘simple’ financial instruments. This means that if a MiFID II financial product is subject to the PRIIPs Regulation (e.g. derivatives, non-UCITS funds, structured securities), the investment firm should provide its clients with the PRIIPs KID, and will by doing so comply with the MiFID II ex ante information requirements with respect to costs of the product. The product distributor should only add information with respect to the costs of the service.75 If no PRIIPs KID is available, however, (e.g. for simple securities76) and the product distributor needs to provide information regarding the costs of the product, it will again have to obtain this information from the product provider. [21.34] Each of the above-mentioned issues necessitates increased cooperation between product providers and distributors in order to ensure compliance with the MiFID II and IDD requirements on cost disclosure. Although increased interaction between product providers and distributors should in principle be regarded as a positive evolution, it may be problematic that MiFID II and IDD place the responsibility of gathering missing product information upon the distributor. Obtaining this information may not be a big issue for large distributors with sufficient bargaining power, but for smaller distributors this may prove quite a challenge. If they fail to meet this challenge, their only option may be to cease distribution of the product in question.

2. Product Risk Assessment [21.35] A second question relating to the UCITS/PRIIPS—MiFID II/IDD interplay concerns the product risk assessment in the UCITS KII and the PRIIPs KID. Product risk is also highly relevant under MiFID II and IDD, more particularly in the context of the product governance process and the suitability test. The question arises as to what extent the UCITS/PRIIPs risk assessment methodology could or should be used for MiFID II/IDD purposes.77

A. The Product Risk Indicator [21.36] A key section of both the UCITS KII and the PRIIPs KID relates to the risk and return of the product.78 The most prominent part of this section is without doubt the KID summary risk indicator79 / KII ‘synthetic risk and reward indicator’ (SRRI),80 which should give an overall view of the risk of the product. It should be supplemented by, among other things, a narrative explanation of that indicator, its main limitations, and a narrative explanation of the risks which are materially relevant to the PRIIP/UCITS fund and which are not adequately captured by the risk indicator.81 [21.37] The UCITS risk indicator is well established; the design and underlying methodology of the PRIIPs risk indicator is still under development.82 Nevertheless, this research into the question of whether such product risk assessment methodology can or even should be used for the purposes of the MiFID II/IDD product governance and suitability processes will only refer to the PRIIPs risk indicator and underlying methodology, for the following reasons. First, both the PRIIPs and the UCITS risk indicators are composed of a scale with seven discrete buckets. The methodologies underlying the classification of products, however, differ between UCITS and PRIIPs. The risk assessment methodology underlying the PRIIPs risk indicator is the more refined and includes a broader range of risk categories. Whereas the UCITS risk indicator only caters for market risk, the PRIIPs risk indicator will also cover credit risk and liquidity risk.83 Second, a UCITS fund categorized in the highest risk bucket, 7, of the UCITS SRRI is arguably less risky than certain PRIIPs

classified in the highest PRIIP risk bucket (e.g. certain derivatives). For investors this may be quite confusing. It should, however, only be a matter of time before UCITS funds become subject to the PRIIPs Regulation, which features a transitional period of only five years, during which UCITS will not be subject to the PRIIPs Regulation (this period can, however, be extended).

B. Risk Assessment and Product Governance [21.38] As explained in more detail in Chapter 5 of this volume, MiFID II and IDD require investment firms which produce financial instruments for sale to clients to ensure that those financial instruments are designed to meet the needs of an identified target market of end clients.84 These product governance rules are, however, not limited to the product design phase. An intense interaction with the distribution side is necessary. After the product has been designed with a target group in mind, the product distributor should ensure that the product is sold to the right target group of clients. Therefore, product manufacturers are expected to provide adequate information to distributors, regularly review investment products offered or marketed, and check that products function as intended.85 Product distributors, on the other hand, have to ensure that products and services are compatible with the characteristics, objectives, and needs of the target market and have to take into account how the products and services relate to other applicable MiFID conduct of business and organizational requirements.86 [21.39] The questions should be raised (i) whether the risk assessment methodology which has been developed for purposes of the PRIIPs risk indicator could or even should be used by product manufacturers as one of the tools to define the target market for the product in the product governance process; and (ii) whether—vice versa—feedback from the distributor of the product, which may suggest an adaptation of the target market, should also induce the product manufacturer to review the PRIIPs KID in order to check whether any update is necessary of, for example, the risk profile or risk warnings for that product.

[21.40] As product risk is indeed an important element in defining the target market for a product, it would at first sight seem plausible to use the PRIIPs risk assessment methodology to assess this aspect in the product governance process. In reality, though, the risk scales may be insufficiently refined to be useful in this respect. Derivatives, such as options, futures, and swaps, will typically receive a high risk categorization. In the product governance process, however, a difference could be made between whether the use of such products is for speculation or for hedging purposes. In the second scenario, the product—which may have a high PRIIPs risk score— may also be considered useful for risk-averse investors (see also n. 86). [21.41] It therefore seems useful that investment firms use the PRIIPs risk assessment procedure in defining the product target group, but do so without strictly linking a certain product score to a certain target group. [21.42] It would also be a good idea to explicitly require product manufacturers to check whether an update of the KID or KII is necessary (e.g. the risk profile or risk warnings), upon product governance feedback from the distributor suggesting an adaptation of the target market.

C. Risk Assessment and Suitability [21.43] As explained in more detail in Chapter 6 of this volume, MiFID II requires investment firms providing investment advice or portfolio management services to perform a suitability test.87 One of the main aspects of this test is the assessment of the objectives, including the risk tolerance of the investor. In order to match suitable products to the investor, most financial institutions use an internal product risk classification model to attribute a certain risk level to each product. The 2012 ESMA guidelines on suitability even require investment firms and credit institutions to define a priori the level of risk and illiquidity of all financial instruments included in their offer to investors.88 Such attribution of a risk and liquidity level usually only serves internal use in facilitating the product–client matching process for suitability purposes. The investment firm will define for each investor profile what product risk classes are suitable for the client and in what proportion of the client’s portfolio. There are no EU requirements on

how financial institutions should attribute this risk/liquidity level, what risk aspects should be considered, or what scale they should use. [21.44] Again the question can be raised whether a link should be made with the PRIIPs risk assessment methodology, which could or should be used internally by investment firms for purposes of the MiFID II suitability test. At first sight it seems efficient to use the same risk assessment methodology both for categorizing PRIIPs and for the MiFID suitability test. If on the contrary a risk assessment methodology is used in the suitability process which differs from the PRIIPs methodology, this could result in strange outcomes, especially if a product that would be categorized as ‘high risk’ on the PRIIPs risk indicator would in the suitability process be considered sufficiently safe for a risk-averse investor.89 [21.45] On the other hand there are good arguments for allowing investment firms to use a risk assessment procedure for suitability purposes which does not perfectly match the PRIIPs methodology. The most important argument is that investment firms are fully responsible to their clients for the products they recommend as being suitable. If they are convinced that a product—even when it has a ‘low risk’ score on the PRIIPs indicator—does involve high risks for a certain category of investors, investment firms should have the liberty to use a different internal risk score. Even though the PRIIPs indicator takes into account a wide variety of risk factors, certain aspects which an investment firm may deem important for purposes of its suitability process may not be included. [21.46] The former argument already indicates another reason why investment firms should not be forced to use the PRIIPs risk assessment methodology as part of their suitability test. If there were a mistake or an inaccuracy in the PRIIPs risk assessment methodology, the mandatory use of this same methodology for suitability purposes could lead to a sectorwide mis-selling problem. It would in such circumstances moreover be hard to attribute responsibility to the investment firm. We have argued before that the use of one common risk assessment system for all PRIIPs creates an inherent systemic risk.90 This risk would be exacerbated if investment firms used the same risk assessment procedure for suitability purposes. If the investment firm on the other hand remains responsible for the risk

assessment for suitability purposes, (i) there is a double check on the acceptability of product risk for a certain client; and (ii) mistakes or inaccuracies in the PRIIPS risk assessment methodology will be more easily detected. This moreover encourages investment firms to remain vigilant and to constantly update their risk assessment methodologies.

3. Conclusion with Respect to the MiFID II/IDD– PRIIPs/UCITS Nexus [21.47] The Joint Committee of the ESA’s has already indicated that: There is a clear interaction between disclosures in the KID of costs and disclosure by the intermediary of costs and that it will be important to consider how these two kinds of disclosure work together for consumers so as to avoid inconsistencies, potential for misunderstanding, or ‘too much information’ undermining the consumer’s use of the information overall.91

This section has shown that the danger of inconsistencies is real, not only with respect to cost information, but also with respect to product risk assessment. [21.48] We believe that the following steps should be taken in order to improve consistency in this area. [21.49] First, the PRIIPS Regulation has granted a transitional period of five years during which UCITS would not be subject to the PRIIPs Regulation. Following that period UCITS funds will become subject to the Regulation in the absence of any extension of this transitional period.92 The separate rulebook for UCITS KII should in our opinion indeed be abolished sooner rather than later and UCITS funds should be covered by the PRIIPs Regulation instead. The current unsubstantiated differences in approach between both information documents not only undermine the idea that the same rules should apply to all investment products that are substitutes from an economic perspective, but also complicate the relationship with the MiFID II conduct of business rules.

[21.50] Second, product manufacturers and investment firms should be encouraged to align their internal product risk assessment methodology, used in the context of the product governance process as well as for suitability purposes, to the PRIIPs methodology. This should, however, not result in a hard law requirement to use exactly the same methodology. Investment firms should be allowed to factor in other risks or to weigh certain elements in a different manner, since it is the investment firm which is eventually responsible for matching suitable products with individual clients. Moreover, only such an approach allows for continued experimenting and improvement of risk assessment methodologies, which is the only way to avoid large-scale mis-selling problems which could result from the use of one harmonized risk assessment methodology for both PRIIPs/UCITS and MiFID II/IDD suitability purposes. [21.51] Feedback from product distributors to product manufacturers suggesting an adaptation of the target market or some other problem should in our opinion be used by product manufacturers to check not only whether an adaptation to the target group is necessary but also whether an update of the KID or KII is necessary (e.g. of the risk profile or risk warnings).

IV. Cross-Selling Practices [21.52] MiFID II has introduced specific rules with respect to crossselling practices.93 When an investment service is offered together with another service or product as part of a package or as a condition for the same agreement or package, the investment firm should inform the client whether it is possible to buy the different components separately and provide for separate evidence of the costs and charges of each component. Where the risks resulting from such an agreement or package offered to a retail client are likely to be different from the risks associated with the components taken separately, the investment firm should moreover provide an adequate description of the different components of the agreement or package and the way in which their interaction modifies the risks.94

[21.53] MiFID II is, however, not the only financial law directive dealing with cross-selling practices. After an extensive study for the European Commission had measured the impact of cross-selling practices and charted why they are used,95 the Mortgage Credit Directive96 (MCD), the Payment Accounts Directive97 (PAD), MiFID II, and IDD98 have introduced provisions dealing with cross-selling practices. [21.54] A comparison of those provisions reveals that the measures introduced by the different directives are quite divergent.99 First, whereas MiFID II, IDD, and PAD introduce a rather soft approach towards crossselling practices (there are no specific prohibitions,100 but extra information and risk assessment obligations apply), the MCD takes a much stricter approach, banning tying,101 but allowing bundling102 in certain circumstances.103 Second, the scope of the provisions is different. Whereas MiFID II, IDD, and PAD seem to apply to cross-selling practices irrespective of whether the second element of the package is a financial or a non-financial product or service, the MCD only applies to cross-selling practices involving two financial services or products. Finally, the inconsistency in the use of terminology and the lack of definitions in most directives104 is disheartening. [21.55] Since cross-selling will often involve two products or services falling in the scope of different directives, EBA, ESMA, and EIOPA decided to work on common guidelines on cross-selling practices for the financial sector. They therefore launched a joint consultation paper with respect to ‘cross-selling practices’ on 22 December 2014.105 It was ESMA alone, however, that finally issued guidelines on cross-selling practices on 22 December 2015 ‘due to concerns regarding the legal basis provided in current insurance and banking directives, and considering the legal obligation for ESMA to adopt guidelines’.106 Indeed MiFID II mandated ESMA to develop, in cooperation with EBA and EIOPA, guidelines for the assessment and the supervision of cross-selling practices by 3 January 2016, indicating, in particular, situations in which cross-selling practices are not compliant with the MiFID II obligations.107 IDD gives a similar mandate to EIOPA, but the other two Directives do not explicitly mandate EBA to issue guidelines. The bigger problem seems to be that the four directives set out

different rules with respect to cross-selling practices: they use different terminology, and take different stances on whether and on what conditions different cross-selling practices are allowed. This obviously makes it extremely difficult to develop common guidelines.108 [21.56] In a previous contribution, the author has made the case for a cross-sectoral approach of conduct of business rules.109 Especially with respect to cross-selling practices, which inherently often involve two products or services, a common approach is indispensable. The regulation of cross-selling practices clearly demonstrates the difficulty of levelling the playing field between the different sectors of the financial industry (banking, investment, and insurance), which are still predominantly regulated by sector-specific legislation and European Supervisory Authorities.

V. Conclusion [21.57] MiFID—since the adoption of MiFID II renamed ‘MiFID I’—has often been labelled ‘the cornerstone of EC securities regulation’.110 However, the house that is EU financial legislation today would not strike the average passer-by as being constructed by experienced stonemasons in accordance with a thoroughly considered blueprint. Rather, on account of the EU legislature’s ‘piecemeal approach’, any combined application of different pieces of financial legislation relating to particular sectors, products, and/or services is bound to result in serious interpretation and implementation issues. [21.58] Indeed, subsequent pieces of financial legislation have been lacking common concepts and terminology, resulting in a rather baffling Babylonian confusion. Properly defining the scope of application of rules—thus covering the right set of economically substitutable products or services—would be an important first step to take in order to improve the consistency of the legislative framework.

[21.59] We have thus made the case for one cross-sectoral piece of legislation dealing with conduct of business rules for any investment product in the broad sense. To that end, the scope of application of the MiFID II conduct of business rules should have been broadened to also cover insurance-based investment products. After all, insurance-based investment products are generally considered to be substitutes to certain financial instruments and structured deposits, which are subject to the MiFID II conduct of business rules. The Insurance Distribution Directive has introduced conduct of business rules which, especially for insurancebased investment products, are to a certain degree aligned with the MiFID II conduct of business rules. Substantial differences remain, however, leaving room for regulatory arbitrage, particularly in view of the Insurance Distribution Directive’s inducements regime, which is much laxer than its MiFID II counterpart. [21.60] Similarly, with respect to cross-selling practices, one set of rules with an even wider scope of application—the entire financial sector— would have been much more effective. One could even argue that such rules could have been fitted into the Unfair Commercial Practices Directive 2005/29/EC. [21.61] While defining a proper scope of application for new pieces of legislation is one thing, it remains equally important to ensure consistency between different pieces of financial legislation such as the UCITS Directive, the PRIIPs Regulation and MiFID II. The legislature and the Supervisory Authorities have been observed to point to each other as being the ones having to solve inconsistencies. It appears rather cynical to expect financial institutions to succeed in fully complying with the investor protection rules, if the European legislature itself has a hard time combining different pieces of legislation into one consistent whole. [21.62] It all goes to show that, on account of the EU’s piecemeal approach to financial legislation, MiFID II has had to pick up some of the crumbs of other pieces of legislation, and vice versa. Yet, as the various regulators soldier on, trying to come up with novel ways to fix the design and construction flaws in the financial legislation house, the stakeholders have been footing the bill. In EU financial legislation, we might see quite a

few more bills coming before an alternative, cross-sectoral approach to crafting legislation and organizing supervision takes hold.

1

I would like to thank Dr Maarten Peeters for his most helpful comments on an earlier draft of this contribution. 2 These are the ‘principle of conferral’ and the ‘subsidiarity principle’, both enshrined in Article 5 TEU. 3 Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU, OJ L 173/349. 4 Directive (EU) 2016/97 of the European Parliament and of the Council of 20 January 2016 on insurance distribution (recast), OJ L 26/19. 5 Directive 2009/65/EC of the European Parliament and of the Council of 13 July 2009 on the coordination of laws, regulations and administrative provisions relating to undertakings for collective investment in transferable securities (UCITS) (recast) [2009] OJ L 302/32. 6 Regulation (EU) No 1286/2014 of the European Parliament and of the Council of 26 November 2014 on key information documents for packaged retail and insurance-based investment products (PRIIPs) [2014] OJ L 352/1. 7 Such as portfolio management, investment advice, and order execution. See the full list of investment services in Annex I.A of MiFID I. 8 Such as safekeeping and administration of financial instruments for the account of clients. See the full list of ancillary services in Annex I.B of MiFID I. 9 See definition in Article 4(1)18 MiFID I. 10 See definition in Article 4(1)19 MiFID I. 11 See Annex I.C MiFID I. 12 See for concrete examples: Niamh Moloney, EU Securities and Financial Markets Regulation (Oxford: OUP, 2014), p. 780, footnote 71; European Commission, ‘Open Hearing on Retail Investment Products’ (2008) at 11, indicating that in France, sales of unit-linked life insurance have increased following the implementation of MiFID; see also pp. 16–17 for several examples of regulatory arbitrage in the Netherlands, and p. 18, for a quote by Eddy Wymeersch, chairman of CESR at the time, arguing that regulatory arbitrage has been seen on a massive scale through the growth of the certificate market. Also European Commission, Impact Assessment accompanying the Proposal for a Regulation of the European Parliament and of the Council on key information documents for investment products (SWD(2012)187, 3 July 2012) at 16. 13 See Veerle Colaert, ‘Building blocks of investor protection: all-embracing regulation tightens its grip’ (KU Leuven working paper, available via ssrn).

14

Article 1, § 4 of MiFID II. Structured deposits are defined in Article 4(1)43 as ‘deposits’ in the meaning of Article 2(1)3 of Deposit Guarantee Directive 2014/49/EU (i.e. ‘a credit balance which results from funds left in an account or from temporary situations deriving from normal banking transactions and which a credit institution is required to repay under the legal and contractual conditions applicable, including a fixed-term deposit and a savings deposit’), which are ‘fully repayable at maturity on terms under which interest or a premium will be paid or is at risk, according to a formula’ involving certain underlying financial instruments or indexes. 15 Recital 87 to MiFID II and Recital 56 to IDD. 16 Articles 12–13 IMD. 17 Article 91 MiFID II; the Explanatory Memorandum to the IMD II Proposal explicitly mentions that the conduct of business rules for insurance-based investment products are based on the MiFID II conduct of business rules (Proposal COM (2012) 360 for a Directive of the European Parliament and of the Council on insurance mediation (recast) (3 July 2012) at 11). 18 This new name expresses the fact that the IDD rules no longer only apply to insurance brokers or intermediaries (‘mediation’), but also to insurance companies that engage in direct selling, and thus to anyone distributing insurance products. See European Commission, ‘Press Release: Commission welcomes deal to improve consumer protection for insurance products’ (IP/15/5293, July 2015). 19 See the Explanatory Memorandum to the IMD II proposal (COM 2012 (360)) (n. 17), 2: In order to ensure cross-sectoral consistency, the European Parliament requested that the revision of IMD1 would take into account the ongoing revision of the Markets in Financial Instruments Directive (MiFID II). This means that, whenever the regulation of selling practices of life insurance products with investment elements is concerned, the proposal for a revised Directive (IMD2) should meet the same consumer protection standards as MiFID II. 20

It should be noted that some of the rules which IDD categorizes as ‘conduct of business’, such as the conflicts of interest regime (Article 19) and the product governance process (Article 25), are traditionally (under MiFID I and II) regarded as organizational rules. This is relevant for MiFID purposes since the organizational rules apply irrespective of client categories, whereas the conduct of business rules differentiate between retail clients, professional clients, and eligible counterparties. IDD does not distinguish between customer categories. 21 Article 17(1) IDD, which mimics Article 24(1) MiFID II. 22 This seems to be the omitted subject of the sentence. 23 Unfair Commercial Practices Directive 2005/29/EC. 24 On the interaction between MiFID, the UCPD, and other consumer law directives, see (in Dutch): V. Colaert, De rechtsverhouding financiële dienstverlener: belegger (Brugge:

Die Keure, 2011). 25 Paper or another ‘durable medium’ (Article 23 IDD). 26 MiFID II only requires a durable medium for specific information obligations, such as with respect to conflicts of interests (Article 23). 27 Articles 47–50 of the Proposal for a Commission Delegated Regulation (C (2016) 2398 final) supplementing Directive 2014/65/EU of the European Parliament and of the Council as regards organisational requirements and operating conditions for investment firms and defined terms for the purposes of that Directive (‘MiFID II Delegated Regulation’). 28 Article 46 MiFID II Delegated Regulation. 29 Article 24 IDD. 30 Article 24(11) MiFID II. 31 The original IMD II proposal distinguished between tying and bundling and prohibited, in principle, tying but not bundling (Article 21(1) IMD II Proposal). See in this regard, Veerle Colaert and Maarten Peeters, ‘Combined Offers’, in Evelyne Terryn, Gert Straetmans, and Veerle Colaert (eds), Landmark Cases of EU Consumer Law in Honour of Jules Stuyck (Cambridge: Intersentia, 2013), pp. 427−46. Even the terminology used and the terms defined differed. 32 Article 22(2). 33 Article 24(12) only allows Member States to impose additional requirements ‘in exceptional cases’ and after notification of the European Commission. 34 Article 20 IDD. 35 Circular FSMA_2014_02 (16 April 2014) p. 39. 36 Article 30 IDD. 37 Article 30(4) and (5) IDD. 38 KYC rules in Article 25(2)–(4); record-keeping obligations in Article 25(5); and reporting obligations in Article 25(6) of MiFID II. 39 Articles 27–28 IDD. 40 Articles 16(3) and 23 MiFID II. 41 Article 29(1) IDD. 42 Article 24(4) MiFID II. 43 Article 24(6) MiFID II. 44 Article 29(2) IDD. 45 Article 29(3) IDD. 46 The IDD rules seem to have been based on one language version of MiFID II and to have been translated into other languages without always taking into account the official MiFID II translations. See for example the general duty of care in Article 17(1) IDD: ‘Member States shall ensure that …’ versus Article 24(1) MiFID II: ‘Member States shall require that …’. 47 Article 24(9) MiFID II.

48 49

Article 24(7)–(9) MiFID II. See on this issue, Chapter 8. See Article 29(2) IDD:

insurance intermediaries or insurance undertakings are regarded as fulfilling their obligations … where they pay or are paid any fee or commission, or provide or are provided with any non-monetary benefit in connection with the distribution of an insurance-based investment product or an ancillary services … where the payment or benefit (a) does not have a detrimental impact on the quality of the relevant services to the customer and (b) does not impair compliance with the insurance intermediary’s or insurance undertaking’s duty to act honestly fairly and professionally in accordance with the best interest of its consumers. The Commission is empowered to adopt delegated acts in this respect (Article 29(4) IDD). 50 Article 29(3) IDD. 51 Article 24(7) MiFID II. See on this issue Chapter 6. 52 Article 29(3), fourth paragraph. 53 On the economic substitutability of life insurance products and other typical investment products, see European Commission, ‘Need for a Coherent Approach to Product Transparency and Distribution Requirements for “Substitute” Retail Investment Products?’ (Call for evidence) (2007); Communication of the Commission to the European Parliament and the Council COM (2009) 204 final, ‘Packaged Retail Investment Products’ (30 April 2009). 54 Article 29(3), last paragraph, IDD. 55 Recital 51 and Articles 22 (1) and 30(6) IDD. With respect to conflicts of interest, Article 28(3) requires that the disclosure should include ‘sufficient detail taking into account the nature of the customer’. It is not clear whether this refers to a professional or retail classification, or to any other aspects which define the nature of the customer. Interesting in this regard is that, for purposes of the PRIIPs regulation, a ‘retail client’ is defined as (a) a retail client as defined in Article 4(1)11 MiFID II, or (b) a customer within the meaning of IDD, where that customer would not qualify as a professional client as defined in MiFID II (Article 6(4) PRIIPs Regulation). 56 In a recent case dealing with the same issue under current Belgian law, the Belgian Constitutional Court held that the fact that insurance services providers, contrary to investment services providers, had no possibility to distinguish between retail and professional clients (and therefore had to comply with the demanding conduct of business regime for retail clients in respect of all customers) was contrary to the constitutional principle of equal treatment (Belgian Constitutional Court, Judgment no. 89/2016 (9 June 2016). ). 57 See also: Veerle Colaert, ‘European Banking, Securities and Insurance Law: Cutting through Sectoral Lines’ (2015) CML Rev 52, 1603.

58

‘Level 2’ and ‘Level 3’ refer to the Lamfalussy legislative method, a legislative technique used in European financial law to speed up the legislative process. It is based on the idea that only the principles should be agreed upon in the ordinary legislative procedure, involving a proposal by the European Commission and co-decision by the Council and the European Parliament (‘Level 1 legislation’). The technical details are then delegated to the European Commission which can adopt ‘Level 2 legislation’. At ‘Level 3’ the European Supervisory Authorities (ESAs), composed of representatives of supervisors of the Member States, develop guidelines and standards for a common interpretation of the Level 1 and Level 2 legislation. Finally, ‘Level 4’ consists of a compliance check by the European Commission. See Alexandre Lamfalussy and others, ‘Final Report of the Committee of Wise Men on the Regulation of European Securities Markets’ (15 February 2001). 59 The Level 1 Directives make divergent measures at Level 2 almost unavoidable, as IDD has not empowered the European Commission to adopt delegated acts in all instances whereas MiFID II has done so (see e.g. Article 24(13) and (14), which has no equivalent in IDD). Moreover the Commission has already made a Proposal for a Commission Delegated Regulation (C (2016) 2398 final) (n. 27) for MiFID II upon the advice of ESMA. The bulk of the Level 2 and Level 3 work for IDD still needs to be done. Ensuring a level playing field would require EIOPA to closely follow the ESMA advice and only deviate where necessary in function of the specificity of the insurance market. 60 Article 78 UCITS Directive 2009/65/EC and Commission Regulation (EU) No. 583/2010 of 1 July 2010 implementing Directive 2009/65/EC of the European Parliament and of the Council as regards key investor information and conditions to be met when providing key investor information or the prospectus in a durable medium other than paper or by means of a website [2010] OJ L 176/1. 61 Maximum three A4 pages for PRIIPs KIDs (Article 6(4) PRIIPs Regulation); maximum two A4 pages for UCITS KIIs (Article 6 of Commission Regulation (EU) No. 583/2010). 62 Article 13 PRIIPs Regulation; Article 80 UCITS Directive 2009/65/EC. 63 For PRIIPs: Article 8(3)(f) PRIIPs Regulation and Article 5 of the Final Draft Regulatory Technical Standards (JC 2016 21); for UCITS funds: Articles 4(9) and 10–13 Commission Regulation (EU) No. 583/2010. 64 Article 24(4) last paragraph MiFID II; Article 29(4) second para IDD (for insurancebased investment products). 65 Article 24(13) and (14) MiFID II; Articles 46–50 of the Proposal for a Commission Delegated Regulation (C (2016) 2398 final) (n. 27). See also ESMA, Final Report: ESMA’s Technical Advice to the Commission on MiFID II and MiFIR (ESMA/2014/1569, 19 December 2014) 121–7. 66 See Article 50(5) and (6) of the Proposal for a MiFID II Commission Delegated Regulation (C (2016) 2398 final) (n. 27). 67 Recital 78 MiFID II and ESMA (n. 65), 120, para. 30:

Recital 78 of MiFID II states that where sufficient information in relation to the costs and associated charges in respect of a financial instrument is provided in accordance with Union law that information should be regarded as appropriate for the purposes of providing information to clients under this Directive. It is ESMA’s interpretation that this is primarily referring to the UCITS KIID and PRIIPS KID. 68

It ‘believes that as transactions costs are linked to the product itself, rather than to the service provided, the issue of their disclosure could be better addressed, following the necessary consultations, in implementing measures related to PRIIPs and UCITS, rather than MiFID II’. See ESMA (n. 65), 115, para. 23. 69 ESMA (n. 65), 118 no. 23 and at 123 para. 14: In relation to UCITS, the Commission should consider the possibility to require the disclosure of product costs and charges that are not included in the UCITS KIID. In line with Recital 78 of MiFID II, where transaction costs have not been provided by a UCITS management company, the investment firms should calculate and disclose these costs (for example, by liaising with UCITS management companies to obtain the relevant information). 70

See ESMA (n. 65), 114 para. 11:

ESMA received several specific comments on the draft technical advice as respondents noted that firms should not be required to provide any additional information about the financial instrument that is not already provided by the product manufacturer and/or disclosed in the UCITS KIID or will be required in the PRIIPs KID. See also at 117, para. 21: With regards to the list of costs and charges to be disclosed to clients, that where presented in Annex 2.14.1 of the CP, respondents raised the following comments: (i) with regards to consistency with other Directives and Regulations, respondents highlighted that it is essential that requirements on cost disclosure in the MiFID implementing measures are consistent with and directly based on the requirements set out for the UCITS KIID and PRIIPS KID. 71

Article 50(4) of the Proposal for a Commission Delegated Regulation (C (2016) 2398 final) (n. 27). 72 With the possibility for professional clients and eligible counterparties to agree, under certain conditions, to a limited application of the detailed requirements of the delegated regulation. See Article 50(1) of the Proposal for a Commission Delegated Regulation (C (2016) 2398 final) (n. 27).

73

Article 50(5) of the Proposal for a Commission Delegated Regulation (C (2016) 2398 final) (n. 27). 74 ESMA (n. 65), 113 para 6. 75 See nn. 67–9. 76 It should be noted however that the proposal for a new Prospectus Regulation intends to model the summary prospectus on the PRIIPs KID. See Recital 25 and Article 7 of the Proposal for a Regulation of the European Parliament and of the Council on the prospectus to be published when securities are offered to the public or admitted to trading (30 November 2015, COM (2015) 583 final). 77 With respect to costs a very similar issue was explicitly raised. See ESMA (n. 65), 120 para. 33: ‘In relation to the methodology for the calculation of ex-ante figures, the majority of respondents encouraged ESMA to ensure consistency between the methodologies to be used for MiFID, accounting standards (GAAP/IFRS) and PRIIPS/UCITS regulatory frameworks.’ 78 For UCITS funds: Article 4(8) of Commission Regulation (EU) No. 583/2010; For PRIIPs: the section: ‘What are the risks and what could I get in return?’ (Article 8(3)(d) of the PRIIPs Regulation). 79 Article 8(3)(d) PRIIPs Regulation; Article 3(2)(a) and Annex III of the PRIIPs Final Draft Regulatory Technical Standards (JC 2016 21 – see also footnote ). 80 583/2010. 81 Article 8(3)(d) PRIIPs Regulation and Annex III of the PRIIPs Final Draft Regulatory Technical Standards (JC 2016 21 – see also footnote); Article 8(1) UCITS Implementing Regulation (EU) No. 583/2010. 82 On the basis of the Draft Regulatory Technical Standards of the Joint Committee of the European Supervisory Authorities (JC 2016 21), the Commission on 30 July 2016 proposed a Delegated Regulation with regulatory technical standards with regard to the presentation, content, review and revision of key information documents and the conditions for fulfilling the requirement to provide such documents (C(2016) 3999 final). The methodology and presentation of the PRIIPs risk indicator were developed in Annexes II and III. On 14 September 2016, however, the European Parliament voted to object to the draft Regulatory Technical Standards on account of a number of shortcomings and inconsistencies. The European Commission should propose a new draft in due course. 83 PRIIPs Final Draft Regulatory Technical Standards (JC 2016 21 – see footnote ) at 6 and RTS Impact Assessment at 84. 84 Article 24(2) MiFID II 2014/65/EU; Article 25 IDD. 85 ESMA (n. 65), 59–61. Those requirements do not seem to have been taken over in the Proposal for a Commission Delegated Regulation (C (2016) 2398 final) (n. 27). 86 According to ESMA, this would, for instance, mean that: (i) if the product is difficult to explain, the sales process should be adapted; (ii) the product governance arrangements should be periodically reviewed; (iii) sales information should be provided to manufacturers to assist them in meeting post-sale responsibilities; and (iv) copies of

promotional material should be supplied to support product reviews by manufacturers (ESMA (n. 65), 59–61). 87 Article 25(2) MiFID II. 88 Final report of ESMA 2012/387, ‘Guidelines on certain aspects of the MiFID suitability requirements’ (25 June 2012) 30 and 31, footnotes 16 and 18. Article 48(2)(c) of the Proposal for a Commission Delegated Regulation requires in this respect that a description of risks should be provided to clients, including, where relevant, to the specific type of instrument and the status and level of knowledge of the client, concerning impediments or restrictions for disinvestments, for example as may be the case for illiquid financial instruments. 89 This could be the case because an investment firm uses a ‘portfolio approach’ to suitability, meaning that not every product marketed to a risk-averse investor should correspond to a low-risk level, but that a few more risky products can be suitable as part of a largely defensive portfolio. A second reason why a product distributor may deem a high-risk classified product suitable for a risk-averse investor relates to the limits of PRIIPs risk classification methodology. For instance, the market risk measure (MRM) for ‘PRIIPs where investors could lose more than the amount they invested’—as is the case for many derivatives—is 7 (Annex II, part I, paras 4 and 8 of the PRIIPs Final Draft Regulatory Technical Standards). If, however, a risk-averse investor uses a derivative for hedging purposes, having also invested in the underlying product, such a derivative product may well be suitable for a risk-averse investor. The PRIIPs risk classification methodology indeed does not allow one to differentiate risk classification depending on the use made of such product (speculation or hedging). 90 V. Colaert, ‘The Regulation of PRIIPs: Great Ambitions, Insurmountable Challenges’ (2016) Journal of Financial Regulation 2 (2), 205. 91 Joint Committee of the European Supervisory Authorities (JC/DP/2014/02) (n. 81), 48. 92 Recital 35 and Articles 32(1) and 33(1) PRIIPs Regulation. 93 Defined in Article 4(1 42 as ‘the offering of an investment service together with another service or product as part of a package or as a condition for the same agreement or package’. 94 Article 24(11). 95 CEPS, ‘Tying and other potentially unfair commercial practices in the retail financial services sector’. Final report submitted to the European Commission, 24 November 2009. 96 Article 12 Mortgage Credit Directive 2014/17/EU. 97 Articles 8, 4(3), and 5(2) Payment Accounts Directive 2014/92/EU. 98 Article 21 IDD. 99 See for an in depth analysis: Veerle Colaert, ‘Cross-selling Practices: Whose Cross to Bear?’ (KU Leuven working paper, available via SSRN).

100

The general rules of the Unfair Commercial Practices Directive 2005/29/EC should obviously be complied with. 101 Defined as ‘the offering or the selling of a credit agreement in a package with other distinct financial products or services where the credit agreement is not made available to the consumer separately’ (Article 4(26) Mortgage Credit Directive 2014/17/EU). 102 Defined as ‘the offering or the selling of a credit agreement in a package with other distinct financial products or services where the credit agreement is also made available to the consumer separately but not necessarily on the same terms or conditions as when offered bundled with the ancillary services’ (Article 4(27) Mortgage Credit Directive 2014/17/EU). 103 Article 12 Mortgage Credit Directive 2014/17/EU. 104 The Mortgage Credit Directive only defines ‘bundling’ and ‘tying’. It does not use the term ‘cross-selling practices’. MiFID II defines the term ‘cross-selling practices’, but does not use the terms ‘bundling’ and ‘tying’. IDD uses the term ‘cross-selling practices’ without defining it. The PAD does not define any of those terms and uses the (undefined) term ‘package’. 105 Joint Committee of the European Supervisory Authorities, ‘Consultation Paper on guidelines for cross-selling practices’ (JC/CP/2014/05, 22 December 2014). 106 ESMA, ‘Final Report: Guidelines on cross-selling practices’ (ESMA/2015/1861, 22 December 2015) at 3. 107 Article 24(11) MiFID II 2014/65/EU. 108 See in this respect the most remarkable letter of 26 January 2016 from the chairs of the three European Supervisory Authorities to Commissioner Jonathan Hill (ESAs 2016/07), with an urgent request to assess the differences in the existing legislation and to consider any necessary steps in order to ensure that the ESAs can regulate cross-selling practices in a consistent way across the three sectors. 109 Colaert (n. 57) at 1611. 110 See among others European Commission, ‘Financial Services: Implementing the framework for financial markets. Action Plan’ (COM 1999 (232), 11 May 1999) at 5; Eilis Ferran, Building an EU Securities Market (Cambridge: Cambridge University Press, 2004) at 3; European Commission, ‘Markets in Financial Instruments Directive (MiFID II): Frequently Asked Questions’ (15 April 2014) at 1; Niamh Moloney, How to Protect Investors: Lessons from the EC and the UK (Cambridge: Cambridge University Press, 2010) at 52.

22 SHADOW BANKING AND THE FUNCTIONING OF FINANCIAL MARKETS Eddy Wymeersch

I. Introduction II. Shadow Banking: Concept and Terminology III. Non-Regulated Activity According to the FSB Approach IV. What Financial Activity is not Subject to Banking Supervision? V. The European Response to Shadow Banking 1. Measures Addressing the Banks 2. Measures Relating to Mandatory Clearing for OTC Derivatives (EMIR) 3. The Short-Selling Regulation 4. Money Market Funds (MMFs) 5. CSDR 6. Securities Financing Transactions 7. The Changes to the Prospectus Regime 8. New Forms of Direct Financing VI. Conclusion

I. Introduction

[22.01] Nearing the end of 2016, the main building blocks of the new financial regulatory measures that were adopted after the financial crisis are now in place. A considerable amount of work has been undertaken with respect to the substantive rules applicable to the fields of banking and insurance, while the securities markets will also be subject to a significantly updated set of rules. One could state that the three main fields of financial activity—banking, insurance, and securities—have now been subjected to new or updated regulations that will contribute to protecting the national economies against further upheaval, while market participants will enjoy better protection. Although many of the details still have to be filled in, one could consider that in these fields most of the work has now been done. It would be repetitive to list the numerous changes that have been introduced —especially in the securities field—as these will be dealt with in other chapters of this volume. Importantly, beyond the detailed regulatory provisions, the organizational framework that is now in place for each of these three sectors reflects very substantial changes in the way Europe will be dealing with its financial sector. Centralization of supervision, in banking and to a certain extent in securities as well, will over time considerably modify the landscape and should lead to a more uniform, more resistant financial system. But it is clear that more efforts have to be made with respect to the supervisory and administrative framework. [22.02] The financial activity that is taking place outside the traditional and often unregulated financial sphere of the securities markets has become quite substantial and especially diverse. This segment is often designated as ‘shadow banking’—a misnomer. The activities to be classified under this characterization have been mapped by the Financial Stability Board (FSB), and include a wide variety of entities which specialize in specific financial activities, or offer financial services as part of their overall product offer. The risks they create are essentially of a macro or ‘systemic’ nature, leading to major financial disruption and contagion. In the securities markets, this activity is often indirectly regulated as part of the wider market regulation. This is the case for the market in derivatives, or for the use of securitized instruments; for derivatives, certain trading venues have been mandated, not always with comprehensive results. The post-trade sector may be the source of considerable risk, and therefore OTC derivatives are subject to mandatory clearing in the CCP, while title security for most traditional

securities will be organized within the Central Securities Depositories. The recent Securities Financing Regulation, dealing with repos and similar instruments, clarifies the position of investors that see their securities ‘reused’ in a second collateral transaction, for which in future their express consent will be necessary. [22.03] Although thought to be on the border of the traditional securities systems, these matters and the related regulations have a considerable impact on the framing of certain classes of securities and may determine the safety of the overall system, including that of the final investor. The objective of the present contribution is to look beyond these core regulatory schemes—what, not without imprecision, could be called the frontiers of the financial regulatory field. Here not only the most recently adopted measures, but also the likely future developments will be mentioned. Attention will be drawn to some interesting shifts in policy objectives over that period. These numerous regulations illustrate a clear intention of the legislature to cover the entire financial activity by one or more layers of regulation, so that in future no part of the financial system would escape the piercing eyes of the regulators. Much of it goes back to the 2009 conclusions of the Pittsburgh G20 meeting.1 [22.04] What is outside the traditionally regulated area of banking, insurance, and securities regulation can be defined in different ways, whether in macro or micro terms. From the macro angle, these are the fields where systemic risk may develop, putting in danger a large part of the financial or economic system; developments that were feared a couple of times in recent years. This subject is now mainly targeted by the FSB, a worldwide body.2 Part of the concern is dealt with in the domestic regulation, in this case in the directives or regulations adopted by the European institutions, often related to regular banking, insurance, or securities regulation. Part of it escapes the powers of the legislators and their agents: these are the monetary imbalances, the sovereign debt concerns, and more recently the effects on the commodities and foreign exchange markets or of the negative interest rate, not to mention the interconnection of the big national economies, the political threats, and the destabilizing terrorist attacks. It is also against this background that some of this new regulation can be viewed.

[22.05] But there also a micro side which is often lost from sight: the new market mechanisms have a profound influence on the way investors are protected, in most cases as a ‘by-product’ of the more solid overall mechanisms. Protections may be legal but are often included in the operational structure or models: interestingly, traditional investor protection techniques prevail, often pursued by disclosure, or by organizational means, or otherwise. The idea is that there is a clear link between the macro and micro aspects, as investors will evidently suffer from systemic crises, but will also be protected or affected by restrictive measures adopted for macro reasons. [22.06] In the first years after the crisis, attention was drawn to a number of financial activities which remained outside the realm of traditional supervision, or were supervised from a different angle than the one applicable to the main field; that is, in terms of their relevance for the financial system.3 Some of these activities had strikingly contributed to propagating the financial crisis: the money market funds (MMFs) especially in the US had unleashed market reactions that might have resulted in a meltdown, while the credit default swaps (CDS) were at the basis of a major confidence crisis affecting not only the banks, but also some sovereign debtors. The question therefore arose whether the existing regulatory and supervisory regime should be augmented by including objectives of financial stability, so avoiding major market turmoil and contagion, and ultimately avoiding systemic risk. Indeed, if investor protection should first be addressed at the micro level, in the relationship between a service provider and the investors, investors should also be protected against risk that would affect the investor community at large, or at least a segment of it. This has been the contribution of the measures that were adopted after the financial crisis: originally most regulations aimed at individual protection—MiFID II being the example—the shift took place in CRR and CRD IV and in later measures: financial stability became the driving force behind this new layer of regulation. By way of a superficial touchstone, MiFID II does not refer, and MiFIR4 only in a few instances expressly refers, to ‘financial stability’ or ‘systemic risk’. The likely explanation is that MiFID II, although dated 2014, was conceived before 2008, while MiFIR work was started later, once the crisis had erupted.

[22.07] There is a relatively clear evolution of regulatory objectives: in the pre-crisis phase, first harmonization of regulation,5 later the integration of the markets, along with the level playing field and protection of investors were the favourite considerations addressed to those already regulated. In a second phase, the central banks and associated bodies drew attention to the potentially disturbing effects of these entities, and to some of the instruments or practices used to disturb the overall stability of the financial markets, whether on their own (see e.g. the SIFIs) or through contagion (e.g. the CDS or derivative markets) and stated that additional measures needed to be considered to avoid a major upheaval in international or even domestic financial markets. The third wave seems to have started very recently: under the pressure of low growth and negative evolutions in the real economy, attention focuses on the role of the financial system in financing the economy and in supporting growth, leading to its adopting a facilitating approach, without abandoning instruments aimed at protecting against the financial stability risks. At the same time one sees a renewed focus on protection for investors and consumers that make use of these new instruments.6 The Capital Markets Union proposal (CMU) opens the door to measures that will help to achieve these goals. [22.08] Until the financial crisis, financial stability issues were essentially dealt with by the central banks, and were mostly considered from the monetary angle. They were generally not addressed—or only obliquely—in financial regulation. The crisis dramatically revealed the relationship between the regulated banking sector and the other parts of the financial system, and led to the awareness that the financial system may be gravely distorted by evolutions that had not been included in the then existing regulations. New financial products like Asset-Backed Securities (ABS), CDS, or MMFs were identified as drivers behind systemic developments, while organizational deficiencies in existing market segments—especially in the field of derivatives—revealed major pockets of potentially global destruction. [22.09] Starting from the analysis of these unregulated, or weakly regulated, activities, the awareness grew that financial activity is not restricted to the banking, insurance, and securities sectors, but that much is

taking place outside these regulated areas. This leads to the notion of ‘shadow banking’.

II. Shadow Banking: Concept and Terminology [22.10] In a first wave, these concerns were regrouped under the very misleading heading of ‘shadow banking’. The term is misleading as it seems to point to a somewhat irregular activity of financial players that would take place in shadowy places—back rooms where no regulation applies or even worse where crime and illegal profit call the tune.7 Some found a more generous explanation according to which shadow banking would refer to activities in the ‘shadow of the bank’, in parallel but not necessarily outside the overall coverage of the banking licence. Over time, however, several of the activities that were labelled ‘shadow banking’ have become the subject of explicit EU regulations, therefore coming out of the shadow in which they had been supposed to dwell. Moreover, shadow banking has been described as useful in the overall functioning of the financial system. To quote Mark Carney, chairman of the FSB: Properly structured, shadow banking can increase efficiency, provide diversification, and spur competition and innovation. It has the potential to make the system more robust, provided it does not rely on the regulated sector for liquidity or pretend to provide it with liquidity in times of stress.8

[22.11] This terminology, and the content behind it, is misleading, as most of this activity takes place between entities that are subject to general prudential regulation (banks, insurance companies) or to more specific forms of regulation and supervision (undertakings for collective investment in transferable securities (UCITS), alternative investment funds (AIFs), MMFs). Also, it often is not ‘banking’, as it is taking place between entities that are not banks, but exercise some of the banking functions, such as credit intermediation in the wide sense. The transactions that are the subject of this activity are increasingly regulated and even specifically organized (in the field of derivatives, for example) and are subject to specific disclosure requirements, allowing the authorities to adopt whatever necessary measures if the overall financial situation takes a negative turn, as

happened in 2008. In most cases the approach is an institutional one, prescribing rules that are applicable to the entity engaged in the transactions viewed, although in other fields the approach is transaction oriented, prescribing rules that address the transaction itself, irrespective of the identity of the party engaging in it. [22.12] By way of example, transaction-based regulation is found as applicable to many of the credit intermediaries which are not regulated or supervised as legal entities; their transactions are regulated (e.g. for mortgage or consumer credit) but from the angle of customer protection, and only incidentally from a prudential (capital requirements) or systemic (loan to value limitations) angle. In the absence of a clear conceptual definition of the shadow banking area, the best approach to obtain an overview is to identify the lists of entities, or activities that are considered to be included in the term ‘shadow banking’. [22.13] As to terminology, the term ‘shadow banking’ is now widely used and accepted. It is used by official bodies such as the FSB, the BCBS (Basel Committee for Banking Supervision), the European Commission, and the ECB (European Central Bank). Many national supervisors also employ the same terminology. In their reports they have several times and with regrets mentioned the imprecision of the terminology, but have nevertheless bowed to its general acceptance.9 We reluctantly join this approach. [22.14] There have been numerous definitions and translations of the expression ‘shadow banking’: ‘market-based financing’, ‘less regulated financial sector’, ‘unregulated financial sector’, ‘market-led institutions’, and ‘alternative financial institutions’ are found in different types of documents and are all equally imprecise. In French, the usual term is ‘finance parallèle’, while in German reference is often made to ‘Schattenbanken’ and in Dutch to ‘schaduwbankwezen’, both literal translations.10 [22.15] The concept of shadow banking can be defined from several angles. The most authoritative one has been developed by the FSB and refers to ‘credit intermediation involving entities and activities (fully or partially) outside the regular banking system’. As non-bank credit

intermediation, shadow banking activity may relate to maturity or liquidity transformation and lead to a build-up of leverage. Referring to activity outside the regular banking system, the definition directly addresses the gaps in the supervisory system where considerable risks may accumulate that could reverberate around or even destabilize the banking system. The approach of the FSB is mainly driven by financial stability and systemic risk concerns, and not on an overarching concept of shadow banking. This approach reflects reality as in today’s analysis it is based on a series of activities that are mainly interconnected from the angle of financial stability.11 [22.16] The analysis of shadow banking should, however, not be restricted to financial stability or ‘macro’ concerns: on one hand, macro provisions often have a micro effect, leading to more transparency and better protection of the market participants, while on the other, some regulatory provisions apply to specific aspects irrespective of the commercial context in which they occur. Some of these provisions play a role in improving protection of market participants or indirectly create the conditions in which this protection can be achieved, for example in terms of market organization. Therefore, the subject of shadow banking, although primarily developed in the context of macro risks, often contributes to solving of or improving on micro issues. In EU regulatory instruments one regularly finds reference to this double objective, where the recitals refer to both financial stability and investor protection aspects. In the following paragraphs, attention will be drawn to this two-sided effect, which is sometimes explicitly pursued, sometimes a mere side effect.

III. Non-Regulated Activity According to the FSB Approach [22.17] How to define the activities that should be included in the shadow banking field is a difficult subject. The FSB has made considerable efforts to better describe this field by formulating criteria that would identify fields that might be of concern especially from the financial stability angle.

[22.18] The FSB defines shadow banking according to economic functions that could trigger financial stability risks in the non-banking area. The following overview lists the main approaches followed by the FSB to define shadow banking in terms of entities exposed to the defined types of risks: • A certain type of entity is exposed to runs: this is mainly the case of investment funds of different types, the MMFs being the most prominent example. • Another type of risk flows from the term transformation, whereby shortterm deposits are used for longer-term lending; these are the finance companies—leasing, factoring, consumer credit. • Market intermediation by broker-dealers,12 dependent on short-term funding. • Credit creation, facilitated by credit insurance, guarantors, monolines. • Securitization-related activities and intermediation [22.19] Identifying these alternative ‘credit intermediators’ and assessing the volumes of their activity has been the significant contribution of the FSB and for the euro area of the European Central Bank. The FSB engages in a worldwide observation and identifies the major providers of credit outside the banking fields—these are the so-called Other Financial Institutions (OFIs)—measuring their activity and relative importance in the financial system of the leading economies. [22.20] According to the FSB, these OFIs include: • • • • • • •

SPV for securitization purposes ABCP conduits, finance companies,13 structured finance vehicles, credit hedge funds, broker dealers, real estate investment trusts, and trust companies.14

[22.21] In order to determine the importance of this phenomenon in the overall financial system and in the respective economies, the FSB analyses

on a yearly basis the development of these alternative credit intermediaries, their growth, and their relative importance. The FSB distinguishes between a ‘broad’ and ‘narrow’ measure; in the latter case, especially addressing the potential impact on systemic risks. According to its 2015 report, the narrowly defined OFIs’ part of the financial sector—which may thus pose a systemic risk threat—corresponds in a worldwide analysis relating to the year 2014, to $36 trillion, equivalent to 59 per cent of the GDP of the twenty-six jurisdictions concerned, or 12 per cent of their total financial system assets. In this total volume, 60 per cent is represented by investment funds of all types that are susceptible to runs. [22.22] Under the broad measure, the analysis reveals the following. Including pension funds and insurance companies, the total volume stands at $137 trillion, or 40 per cent of the global financial system.15 In these jurisdictions the non-bank financial intermediaries stood for $80 trillion in 2014, or 124 per cent of GDP. [22.23] Within the European Union too, much attention is being paid to the shadow banking issue, mostly along the lines of the data analyses developed by the FSB. [22.24] Data about shadow banking activity have especially been compounded by the ECB, relating only to the euro area.16 In its 2015 report on financial structures, the ECB indicates that the shadow banking entities stand for €23 trillion, against €37 trillion for the banking sector proper, increasing their relative share since 2009 by 5 per cent, while the banking sector’s share lost 7 per cent. For a large part of the unregulated sector, no detailed data are available. Of this total volume for the shadow banking segment, 8 per cent represents so-called financial vehicle corporations (FVCs)17 and 4 per cent MMFs. As to non-MMF funds, their growth has been very rapid, resulting in €10.5 trillion of assets being held by all types of funds. These are managed mainly (90%) out of a handful of Member states18 and they mainly finance financial institutions and governments. MMFs are very sensitive to systemic developments: they stood for €1 trillion at the beginning of 2015, mainly located in Ireland, France, and Luxembourg19 their funding being largely collected from non-euro-area investors. On the asset side, euro-area MMFs hold more than 40 per cent of

their assets in euro-area financial institutions, in the form of loans or securities.20 [22.25] The data for the FVCs, mainly securitization vehicles (65%), indicate a net decline since 2010, with the exception of 2012 (a decline of 23% since 2009) now standing for €1.8 trillion. Most of these are ‘retained deals’, used by banks for collateral in central bank operations. 51 per cent of these loans have been generated by banks, with only 15 per cent by nonbank institutions. New securitization issuance in the euro area is still a fraction of the comparable figure in the US, although the issuance of mortgage-backed securities has been picking up, alongside the evolution of the real estate market.21 [22.26] These figures indicate that shadow banking activity is very considerable, but also strongly connected to the traditional financial world, and therefore could endanger stability in the euro area and beyond. Some of these risks are directly related to the activity of the supervised banks, and others might affect them as a consequence of contagion and different types of interconnectedness. No volume-wise significant financial activity should leave the overseers indifferent.

IV. What Financial Activity is not Subject to Banking Supervision? [22.27] According to the definition followed in most of the EU measures, a ‘credit institution’ is defined as ‘an undertaking the business of which is to take deposits or repayable funds from the public and to grant credits for its own account’.22 One of the decisive elements of this definition consists in the credit institution or ‘bank’ receiving funds from the ‘public’. The public may be the general, mainly retail public, or a more limited number of larger investors. The definition does not extend to the funds received from shareholders, or from a limited number of large stable investors. This constitutive element is directly related to the supervisory system that aims at protecting the deposits from the public at large, including the individual depositor,23 but also refers to the systemic dimension the subject of which is

the protection of the overall financial and economic system, indirectly providing a society-wide protection. Indeed, both the well-known case of the ‘run on the bank’, and the equally destructive ‘fire sale’ of assets, often induced by depositors reclaiming their savings, come under this ‘systemic’ category. [22.28] In most European jurisdictions the definition of a bank relates to both the asset and the liability side of the balance sheet: hence entities only engaging in credit transactions, such as factoring and leasing with the public, or other specialized credit activities (conduits, SPV) are not included. The number and business volumes that come under this ‘unregulated credit activity’ are impressive, as evidenced by ECB data.24 Apart from the mentioned activities—leasing, factoring, mortgage credit, and so on—these are the numerous lending subsidiaries of commercial firms, such as car manufacturers and general distribution firms, but also the new forms of provision of credit such as business-to-business firms, crowdfunding, etc. The prevailing definition of ‘credit institution’ therefore leaves a significant population of financial service providers outside the systemic analysis and guidance of the financial supervisory system. This does not mean, however, that these entities are unregulated: in many cases these firms are part of a banking group, exercising specialized activities within the group, and frequently funded by the banking entity, hence indirectly funded by ‘deposits from the public’. As their risks ultimately relate to the banking entity, they will be included in the consolidated position of the group, at least when they are controlled entities. In resolution schemes they would qualify to contribute to the resolution of the bank. Information on the nature, volume, and risks flowing from their activities would be available, although indirectly, through the consolidating bank. In practice, however, these entities have often been held outside the consolidation perimeter, as being ‘non-controlled’ at law. For entities which are not bank-controlled, these are often also partially funded by the banking sector, and hence their demise would percolate into the regulated field. Information on these activities and on the entities involved has generally been lacking, and therefore collecting detailed information on these entities and on their business activity is the first step to undertaking corrective action, including rescue operations. The effect of their insolvency on the financing banks or entities or on the wider banking system has proved to be

equally devastating. As was revealed by a prominent US case,25 this issue was not limited to banking, it also occurred in the insurance field. [22.29] In terms of applicable regulation and oversight, the situation is very diverse. Some of these entities are indirectly regulated and overseen through their parent or group entities, although this might be more limited if the entity falls outside consolidation. In other cases, they are subject to specific regulation—for example, on investment funds—that is not directly geared to financial stability issues. Some are not regulated as financial entities but from the angle of their operations; see, for example, the regulation that protects the users of their services under the headings of consumer credit, mortgage credit, or similar forms. It would be misleading to qualify them as ‘unregulated’. Recently, complaints have been tabled about some of these non-financial credit providers overextending credit, which may be a consequence of the commercial policy of manufacturing or distribution groups. This may lead to the concentration of considerable risks in their portfolios, undermining their solvency and eventually even putting the industrial group to which they belong in danger. Here again, the financing banks may be at risk. [22.30] Being outside the regular financial system, these entities are not connected to the crisis refinancing mechanisms that are available to banks and insurance or securities firms, for example under the form of central resolution support, ELA liquidity provision, or DGS. It is well known that some of these players avail themselves of the lighter forms of regulation which they enjoy in certain jurisdictions or certain market segments, a premium difficult to justify in times when banking regulation is considerably tightened. Regulatory arbitrage risks are rightly related by the EU Commission to ‘market integrity and confidence of savers and consumers’.26 Once more, the macro and the micro issues go hand in hand. [22.31] From a wider perspective, this evolution is part of the wider trend of ‘debancarization’ which motivates regular financial institutions to transfer part of their portfolios to third parties, as their lending activity is increasingly restricted due to prudential or regulatory measures and reduced due to negative interest rates. Other players—such as hedge funds or private investor groups investing on a grand scale in non-performing exposures—

will take advantage of this gap, buying large credit portfolios, which then disappear from the regulators’ radar screens. A similar development can be noticed in the securitization field, at least as it was practised before the crisis: opacity characterized some of the asset-backed securities that, although they mainly originated in the regular financial system, presented unidentified risks which were fully transferred to final investors. Once they left the originator, little information was available, leading to massive liabilities for the banks that sold these securities to the unsuspecting investor. Even in the field of government bonds, the lack of market data allowed market participants to speculate against a government by shorting its bonds, or by buying CDS. Therefore, the first priority for dealing with these issues is transparency and reporting, to make sure that the activity on these markets is well registered and well documented, so that the authorities in charge of the public interest can observe, analyse, and where needed monitor this activity. However, today and except for some specific sectors, many jurisdictions still do not dispose of an overall system of recording transactions in the different types of lending and financing. One could add with some cynicism that states are better informed about their defaulting citizens. [22.32] The population of OFIs is, however, not stable: while existing types may disappear, mainly due to their absorption into banking groups, new players, such as hedge funds and even pension funds or new forms of alternative financing, come on the market at regular intervals: internet credit and mobile banking operators, some also offering credit services, so-called ‘peer-to-peer lending’, or agent firms offering direct access to crowdfunding pools have all recently sprung up. Up to now, their significance has been very limited and therefore they have not yet attracted much attention from the international regulatory bodies. Time will tell how they have to be dealt with. The FSB has drawn attention to the risks from new and innovative shadow banking activities and entities.27 [22.33] The concerns raised by the FSB mainly relate to the development of systemic risk and the risk of regulatory arbitrage.28 The FSB provides a broad definition, focusing on credit intermediation outside the regular banking system. The risks may be generated by operations on the asset side of securitization as well as on the liability side (MMFs).

[22.34] The FSB29 has made a certain number of recommendations on how to bring the shadow banking world into that of resilient market-based financing; the concrete measures recommended are: • Guidance on the scope of consolidation of bank prudential regulation: – to be developed by the BCBS to ensure all-encompassing scope of banking regulations; the step-in risk30 is addressed; – monitoring large exposures, to be limited to 25 per cent of Tier 1 capital; – risk-sensitive capital requirements for investment in equity funds. • Re-hypothecation: harmonization of regulatory approaches: – securities financing transactions: haircuts;31 – harmonization and limiting interconnectedness. • Monitoring of MMFs, their definition, and the limiting of their exposure to runs (CNAV v VNAV). • Monitoring of securitization: alignment of the incentives and STS securitization.32 • Monitoring instruments: peer review. • Monitoring of global trends and risk; data collection.

V. The European Response to Shadow Banking [22.35] In its 2013, report to the Council and to the European Parliament, the Commission gave an overview of the actions it has undertaken or was planning to put on the agenda to curb systemic risk flowing from the shadow banking sector. In so doing the Commission largely followed up on the recommendations of the FSB. The topics mentioned cover a wide range of provisions in directives and regulations which contribute to reducing systemic risk, partly by imposing certain requirements on the regulated firms, partly by introducing new regulations that would extend to another segment of financial activity or address specific risks in the shadow banking sphere. Dealing with shadow banking has therefore not resulted in an overarching measure under the label of ‘shadow banking’, but consists of a series of measures addressing aspects of activities or decisions that otherwise would contribute to the risk of financial instability. This lack of

specificity will sometimes make it difficult to classify a specific action as related to shadow banking, as the measure will be embedded in overall banking or insurance regulation. [22.36] In its 2012 green paper,33 the Commission originally listed as ‘alternative financing entities’ to which it was addressing its attention as belonging to shadow banking: • special purpose entities (SPVs), especially in the context of securitization; • MMFs; • investment funds, including ETFs; • finance companies; and • insurance and reinsurance with respect to guarantees. [22.37] The EBA has published its advice to the Commission on how to define the shadow banking sector for regulatory purposes.34 This list— which generally follows the FSB list—was later expanded also to allow inclusion of other entities that present specific systemic or contagion risks.

1. Measures Addressing the Banks [22.38] In the field of banking regulation, as was put in place since the crisis, numerous provisions serve to curb excessive risk-taking, be it by imposing capital requirements, prescribing governance provisions, or declaring recovery or resolution procedures as applicable. These measures address risk-taking in relation to the financial structure of the bank, its business model, or its relation to other entities in the financial system. Systemic risk issues appear on the agenda when risks exceed certain thresholds in comparison to the overall financial system, or when risk in one sector may spill over into other sectors.35 It is not the purpose of this analysis to deal with these general risk matters, but rather to focus on risks that originate from the ‘unregulated’ part of the financial system, or are proper to that segment.

A. Credit Risk Concentration on Shadow Banking Entities [22.39] Credit activity of the banking sector with shadow banking entities presents a specific risk. Apart from risk concentration, there is a concern with firms arbitraging away some of the more heavily regulated bank risks, to entities—such as MMFs—that are exposed to runs. The same phenomenon is well known from risks being located at SPVs.36 In December 2015, the EBA published its guidelines defining the entities that should be considered as ‘shadow banking’, taking into account the risks that these represent and this with a view of setting limits on the individual exposures to shadow banking entities.37 So, for example, UCITS and AIFs are excluded,38 but MMFs are considered part of shadow banking. Firms controlled on an individual basis, or included in the consolidation of a banking or insurance group, would not be included.39 Under the guidelines, banks will have to consider their individual exposure and their group exposure. The guidelines40 provide that, with respect to the CRR concentration risk, exposures to shadow banking entities, as defined, which exceed 25 per cent of eligible capital—whether per institution or in aggregate—should adopt appropriate measures. These guidelines are based on a thorough fact-finding report on the volume of these exposures in the different member states. The total amount of these exposures exceeds €1 trillion, and stands for 45 per cent of eligible capital, justifying additional supervisory attention.

B. Securitization [22.40] The FSB also found that the interconnectedness between the banking and the non-banking financial sector decreased, more markedly for the securitization activity. [22.41] Securitization has been a concern starting from the first years of the crisis: different measures were introduced in a wide range of EU legislation, mainly to the effect of restricting securitization and reducing its feared risk profile. The successive capital requirement directives have strengthened their treatment in the banks. In a later directive, the capital

requirement position with respect to re-securitizations was strengthened whether by imposing a deduction from own funds, or by applying a strict capital requirement.41 Prudent valuation rules and additional disclosures were mandated. The large exposure rules were also tightened.42 [22.42] To deal with the position of banks with interests in separate vehicles including securitization or structured entities as SPVs—these did not have to be included in the consolidated statement as there was no legal control, although often operational control subsisted—the accounting rules on consolidation have been adapted to oblige banks to disclose their interest, in derogation to the principle of non-inclusion in the absence of legal control. In particular, IFRS 12,43 dealing with ‘Disclosures of Interests on Other Entities’ requires disclosures of interests in, inter alia, joint arrangements, associates, but especially in ‘unconsolidated structured entities’. The information to be disclosed is necessary to understand the nature and the extent of a bank’s interests in unconsolidated structured entities and to evaluate the nature of the risks associated with its interests in these entities. [22.43] Although the experiences with securitization in the EU were less negative than in the US, the European measures have had a considerable effect on this part of the market, as is illustrated in the following graphs (Figures 22.1, 22.2 and 22.3).44 These graphs illustrate that there is a real concern with securitization weakness in the EU, especially for small- and medium-sized enterprise (SME) financing—certainly in comparison to similar figures about the US SME securitization market.

Figure 22.1: Securitization issuance.

Figure 22.2: Issuance of SME securities.

Figure 22.3:

C. EU Measures Relating to Securitization [22.44] There are numerous provisions in EU legislation dealing with securitization, most of the time going back to the time of the financial crisis, and these are generally very, if not excessively, restrictive on securitization transactions, as the experiences were rather bad with previous, especially the imported, securitization species. Among these pre-existing measures applicable to securitizations, one should refer to a large number of existing or proposed provisions in applicable legislative bodies such as CRR, Solvency II, AIFMD, UCITS, CRA III, Prospectus D, BSR, and MMF and their respective delegated acts.45 [22.45] With all this restricting legislation in place, and in addition the considerable complexity of the resulting legal regime, it should not astonish that securitization in Europe is not prospering and therefore has not delivered on its potential in terms of contribution to growth.

[22.46] For these reasons, it was time for a change and this is the objective of the Commission’s recent proposal. [22.47] The proposal, dated 30 September 2015, intends to introduce common rules on securitization in order to reactivate this activity, especially in favour of SMEs. This proposal should be seen as the cornerstone of the common future securitization plans. Voluntary market-led securitization proposals of a non-regulatory nature are welcomed and some have already been developed. The concept consists essentially in providing a blueprint for reliable securitization, offering sufficient safety and reliability to investors, but allowing the financial sector to access the existing lending sector to mobilize its lending portfolios to long-term investors—mainly insurance companies, pension funds, long-term investment funds, and similar. [22.48] At the same time, it should be borne in mind that the interest for securitization is part of the CMU proposal, according to which increased lending by banks or other entities is expected to contribute to economic growth and employment. Although systemic considerations are mentioned in the context of the ‘originate and distribute model’, the proposed regulation mainly pursues a policy of reinvigorating growth by facilitating the lending of assets from the balance sheet of banks and other lending entities to a wider investor community. [22.49] The proposed regulation contains a general blueprint for all securitizations based on requirements relating to disclosure, due diligence, and risk retention, while adding specific requirements for ‘simple, transparent and standardized’ (STS) securitizations,46 the latter to be considered of higher technical quality. Worldwide-applicable models for STS securitization have been developed by IOSCO and the Basel Committee.47 At the same time, the regulation looks at the creation of a Europe-wide securitization market by introducing similar criteria applicable throughout the Union, thereby avoiding the danger of regulatory arbitrage, and contributing to market efficiency. Securitized portfolios also qualify for ECB interventions under its Asset-Backed Securities Purchase Programme (ABSPP) and should therefore meet common standards.48 The treatment of transactions in accordance with different national provisions would be made

more difficult, as the basic rules are partly laid down in the CRR. This was one of the reasons for proposing a Europe-wide standard. [22.50] The regulation proposes criteria for standardization of all securitization transactions, based on due diligence essentially at origination, reflecting sound and well-defined criteria and established processes. These conditions do not apply to originations by banks or investment firms where the conditions will apply on the basis of other regulation.49 Risk retention meeting the criteria of the regulation—5 per cent as a rule—and disclosures in accordance with the requirements of the regulation are core principles. Interesting to note is that the regulation imposes due diligence requirements not only on the offerer but also on the institutional investors acquiring these instruments such as procedures to establish the risk profile of their overall securitization position, stress testing, and regular internal reporting.50 Special provisions apply to Asset Backed Commercial Paper (ABCP) securitizations, as the latter instruments have a more short-term maturity. [22.51] For STS securitizations, the risk retention requirement may be replaced by a reference to an index provided the underlying reference entities are identical to those included in the index.51 This is likely to facilitate standardized securitizations. [22.52] The criteria ‘simplicity, standardization, and transparency’ are further detailed in several sub-requirements.52 They should be based on ‘true sale’ transfers, without application of claw-back provisions in the case of the debtor’s insolvency. This would exclude synthetic securitizations from the STS label. Clear eligibility criteria should be followed based on homogeneous pools of underlying assets, thereby avoiding adverse selection. Also, the portfolio should be stable, and not actively nor discretionarily managed. Re-securitizations or defaulted exposures should be excluded from these portfolios. These features should also avoid the application of the ‘originate or distribute model’. Risk retention is confirmed, with exceptions.53 Effective monitoring mechanisms measuring and managing risk should be in place.

[22.53] Ample pre-contractual information is dealt with in detail, with a transaction summary containing the main features of the securitization.54 [22.54] Interesting is the idea that institutional investors are expected to play a ‘guardian’ role: before entering into a securitization transaction, they are expected to undertake a certain number of diligences relating to the securitization process; but this only with respect to securitizations originating from non-bank entities. [22.55] Institutional investors55 should establish procedures to monitor their risks for which general criteria should be established, some of which apply specifically to STS transactions: • Sound and well-defined lending criteria and processes as the basis of a pre-established process.56 • Material net economic interest is retained;57 not less than 5 per cent.58 • Regular stress tests on cash flow and collateral. • Regular internal reporting information is provided. • With respect to all securitizations, institutional investors will have to develop, inter alia, risk procedures, and stress tests on cash flow and on collateral. • For STS transactions whether the general criteria for STS securitizations are applied; but investors may also rely on a previous STS notification affirming compliance with STS requirements. [22.56] It is open to analysis what the effect of this role of the institutional investors will be for other investors, who will be able to consult the list of STS securitizations on the ESMA website.59 Mentioning a certain securitization portfolio on the website will give investors the impression that these are of higher quality and fit for investment even by retail investors. Will this allow investors to rely on this impression, but also hold the parties having delivered the STS status liable in the case of deficiency? One can expect even more explicit disclaimers than in regular prospectuses. [22.57] The above-mentioned criteria apply to STS securitizations with some additional, stricter information including on defaults, with an external opinion on the accuracy of the data.60

[22.58] Originators or sponsors initiating an STS transaction should notify it to ESMA, mentioning that the securitization meets the STS criteria, or, as the case may be, that due to intervening changes these are not further met: this information is then published on the ESMA website (but only for information purposes).61 The recitals make it clear that the information is released under the responsibility of originators or sponsors and does not bind ESMA as to whether the STS criteria have been met. ESMA should inform national supervisors about transactions in their national spheres. [22.59] As to supervision of these criteria, the regulation refers to the competent authorities for each of the classes of financial intermediaries involved. One can assume that this reference is to both the supervisor in charge of the originating or sponsoring entity and the supervisor dealing with the institutional or other investors with respect to the securities in its portfolio. Securitization for which no competent authority has been referred to on the basis of the existing legislation will be designated, on a default basis, by the Member States. [22.60] By making securitization transactions more reliable, the regulation will not only protect the institutional investors which are the parties mainly investing in these securities, but all investors, opening new avenues to retail investors. In terms of investor protection, the proposed regulation shifts the focus from the general regime based on disclosures— for example, in the prospectus directive a previously applied—to a detailed regime based on substantive requirements, subject to a form of supervision that comes close to that of the investment funds. Interesting is the idea that investors should also make efforts to protect themselves: institutional investors should be especially aware of their duties to protect their stakeholders by implementing specific provisions in that respect. [22.61] The Commission proposal for STS securitization has received support from the main trading associations, both from the sell and the buy sides.62 In the European Parliament the proposal has met with delay.63 [22.62] With respect to the secondary trading of securitized instruments, MiFID II provides that the trading of structured finance products,64 including securitization instruments, is subject to the requirements

formulated in MiFID II for ‘structured finance products’ that are traded on a regulated market, an MTF, or an OTF, a provision also applicable to the systemic internalizers. These requirements relate to both pre- and post-trade transparency. The information will have to be stored at trade repositories (TRs) for which the TRs under EMIR might qualify as Approved Reporting Mechanisms.65 These data-storage tools should be accessible to all on a non-discriminatory basis. Due to the specific nature of these instruments, they will most of the time be traded OTC.66

2. Measures Relating to Mandatory Clearing for OTC Derivatives (EMIR) [22.63] MiFIR contains detailed obligations with respect to trading of derivatives: the general principle is that derivatives should be traded on regulated markets, at least on organized trading facilities (OFTs) and are to be cleared on CCPs.67 These are the exchange-traded derivatives, usually cleared through a CCP belonging to or designated by the trading venue. The decision as to which derivatives will be traded is left to the market itself, that is, to the trading venues.68 A condition is that CCP links have been provided for and that the CCP has received the authorization to clear that specific class of derivatives.69 For classes of derivatives that have not been picked up by a trading platform or by a CCP, ESMA will define the criteria for deciding whether a derivative is tradable, opening the possibility for a regular trading,70 and will make publicly available the register for derivatives subject to trading obligation.71 [22.64] A large part of the derivatives market is not traded on a trading venue, but over the counter (OTC); that is, trades are carried out via direct negotiation between buyer and seller. EMIR contains the obligation to also clear these OTC derivatives through a Central Counterparty (CCP), being a central body that will act as buyer for the seller of the derivatives and seller for the buyer. This would reduce the outstanding amount of derivatives, as these will be set off internally in the CCP. While derivatives entered into by financial counterparties have to be cleared through a CCP, this obligation

only applies to non-financial counterparties provided they exceed a threshold fixed by the EMIR.72 [22.65] The purpose of EMIR73 is to impose clearing of derivatives through a CCP and ensure adequate reporting and disclosure of information on derivatives that are traded OTC. The introduction of a mandatory clearing regime for OTC derivatives is a landmark step in reducing risk, especially systemic risk, in this important segment of the financial market. Before the adoption of this regime, derivatives were cleared on a bilateral basis, leaving considerable positions unmatched and unsettled even for longer periods of time. Some preliminary work was undertaken by firms specializing in compression of positions, leading to a considerable reduction of risk.74 But many positions could not be compressed. The financial crisis and the cases of Lehman and AIG had illustrated the need to proceed in a more structured way. The G20 meeting in Pittsburgh75 had stated that OTC derivatives—at that time, the vast majority of all derivatives—had to be centrally cleared while stricter capital requirements would apply to the other derivatives. The outstanding liabilities on the global OTC derivatives market amount today to $552 trillion, of which $434 trillion relate to interest rate swaps, $74 trillion to Forex contracts,76 and $14 trillion CDS. [22.66] To increase the safety of transactions it was ordered that OTC derivatives—that is, the contracts that are not traded on regulated markets77 but are negotiated directly between banks and other financial intermediaries or commercial enterprises—should be mandatorily cleared through a CCP. These contracts lead to a considerable counterparty risk, especially credit risks. In order to mitigate risks, standardized derivatives will have to be cleared through a central body (the CCP). The role of the CCP is often defined as the party that acts as the buyer for all sellers and the seller for all buyers,78 thereby assuming final risks for the net positions. The risk incurred by the CCP is covered, apart from regulatory capital,79 mainly by the margin requirements that clearing members—and their clients—must deliver to the CCP. These margins stand for 99 per cent of the overall potential exposure of the CCP and will be fully collateralized by its clearing

members.80 A pre-funded default fund will protect the CCP against the default of one or more of its clearing members, while in addition pre-funded resources of the CCP will protect against the default of the two largest clearing members. In the case of the default of a member, the default funds of the non-defaulting members can be used, in a sequence designated as the ‘waterfall’.81 These different safeguards should lead to the conclusion that although CCPs concentrate a large part of the risks in the OTC derivatives markets, their financial position is solidly protected. [22.67] Clearing through a CCP is mandatory for all financial counterparties—as defined82—and for most non-financial counterparties83 that cross a pre-established threshold. The classes of derivatives that are subject to the clearing obligation will be defined in an RTS to be proposed by ESMA for Commission approval.84 The individual derivatives will be identified by the CCPs in their application for authorization. For derivatives that have not been so proposed for clearing, ESMA may define the class of derivatives, with notification to the Commission. In so acting, ESMA will take into account the objective of reducing systemic risk, which will lead in concreto to evaluating the degree of standardization; the volume and liquidity of the class; and the availability of fair, reliable, and accepted pricing information. [22.68] The purpose of the EMIR set-up consisted essentially to reduce risk for all market participants by ordering central clearing and so to secure their position as it would then be linked to adequate financial safeguards at the CCP level. At the same time, transparency in the market had to be increased considerably, allowing market participants to adapt their trading strategies. The volume of cleared OTC derivatives has increased over time, and this already before EMIR has entered into force (Figure 22.4). As this market takes place on a worldwide basis, specific—but for a long time controversial85—provisions take account of this dimension.

Figure 22.4: Percentage and volume of cleared OTC derivatives. *Interest rate swaps ** Forward rate arrangements Source: LCH.Clearnet, published in Phillip Stafford, ‘Centralised risk raises systemic worries over derivatives, Clarity needed on liquidity and clearing models’, Financial Times, 9 June 2016.

[22.69] As the OTC derivative market is vast, transparency was a primary objective of the EMIR allowing supervisors to monitor more precisely the evolutions in the market and the development of specific risks. As a consequence of the clearing requirement, standard derivatives will be more frequently used, contributing to fluidity in the clearing process and offering better protection for indirect users such as buyers of certain investment products or industrial or commercial firms that also use frequently traded derivatives.

[22.70] However, this remains largely an OTC market, where price formation is a function of individual transactions without central price discovery. It is unclear to what extent the MIFIR rules will change this mode of trading. It does not seem unlikely that the stricter requirements applicable in the case of OTC trading as opposed to trading on a regulated market will over time make the latter options more attractive. [22.71] From the angle of the extension of the supervisory regime, this is another example where a complex and detailed regime has been introduced that covers not only the regulated financial institutions, but extends to wider sections of the users of financial services including the shadow banking sphere. [22.72] Supervision of the derivatives market is in the hands of national regulators, generally the securities regulators. The EU Regulation86 provides for the traditional mechanism of decision-making, including the colleges of regulators. ESMA plays its role of supporting the coordination of national regulators, and where necessary mediating between divergent opinions of national supervisors. [22.73] Traded derivatives—whether traded on regulated markets or OTC87—will have to be centrally reported to TRs. ESMA has been put in charge of their supervision, starting with their registration, their supervision including the collection of all useful information, possibly by on-site inspections, while adopting supervisory measures or exercising disciplinary powers, allowing ESMA to impose fines for specifically described breaches. The EMIR Regulation contains no express statement as to the reasons why ESMA has been put directly in charge of the supervision of the TRs. But the need to collect a full overview, including on the positions of the shadow banking entities, can be mentioned.88 The Trade Repository does not confer specific rights to the investor, it is essentially a mechanism to collect information. [22.74] EMIR also contains some provisions that, apart from the advantages of the overall solid organization of the market, are relevant from an investor point of view. Article 39 of EMIR obliges clearing members of a CCP to offer their clients the choice between omnibus and individual

accounts. The legal position would be different and the clearing member has to inform its clients about the level of protection, including ‘the main legal implications of the respective levels of segregation including information on the insolvency law applicable’.89

3. The Short-Selling Regulation [22.75] In 2008, the financial crisis led to strong speculation against the shares of banks and against the sovereign debts of a few of the weaker euro member states. In the case of the banks, the danger was essentially that deposits would be massively withdrawn as a consequence of a fall in the share price. The sovereign debt was mainly attacked through credit default swaps related to these sovereign debt instruments, as their prices went up inversely to the confidence in the sovereign debtor, hence leading to considerably increased interest rates. Voluntary measures to stop at least short selling were considered by the competent national regulators, but these appeared unwilling to suspend the practice (or more generally any trading). The Commission proposed a regulation, in which short selling in equity instruments of listed companies would be reported, while in exceptional circumstances, a short-term ban could be imposed whether by the national regulator, or in the case of multistate listings and if disagreements between national regulators persisted, by ESMA.90 This delegation of power, at that time a rather unique provision based on Article 18(3) of the ESMA regulation, has been challenged before the European Court of Justice (ECJ). The court recognized that, taking into account the specific limitation on the use of the power and the strict conditions under which it can be exercised, there is no reason to consider it contrary to the Treaty.91 This decision is in line with the more recent interpretations of the Meroni doctrine.92 [22.76] In this matter too, the regulation first introduces an elaborate reporting regime, with disclosure to the public for positions beyond the 5 per cent threshold.93 With respect to the CDS, only those on sovereign debt instruments are in scope, as these financially express a short position on the underlying risk. CDS on sovereign rights could only be acquired by parties

that have a—neutralizing—risk position on the underlying bonds; in other words where the CDS is a protection, not a mere speculation on the demise of the sovereign risk. In fact, this amounts to a ban on single CDS on sovereign risks, thereby limiting a much-feared upward trend on the risk premium and hence on the interest rate. To this requirement in principle, some flexibility can be granted by the supervisors, thereby allowing some uncovered CDS activity: this could be the case if the market in the underlying debt instruments is not working properly leading to excessive margins, whereby a decision to allow the CDS bond market to function more correctly would reduce the interest rates. A similar flexibility could be granted on the basis of similar criteria mentioned in the regulation.94 [22.77] This regulation was first and foremost inspired by systemic and financial stability considerations, although its scope is broader, extending to all listed equity and CDSs. But at the same time it mentions the need to organize and maintain fair and orderly markets; objectives that more directly serve the interests of the investors. The broader context of the functioning of the internal market also refers to the legal basis being Article 114 of the Treaty on the Functioning of the European Union (TFEU). [22.78] The discussion that a short-selling ban may be detrimental to investors’ interests, seems to have subsided. The economic effectiveness of this regulation should be assessed over time.

4. Money Market Funds (MMFs) [22.79] MMFs are another example of the interrelationship between the macro and the micro level. Legally, MMFs are investment funds, often taking the form of UCITS in accordance with the applicable European directives. Under a common definition, these funds are classified on the basis of the liquidity of their assets as either ‘short-term MMFs’ or simply ‘MMFs’.95 The funds are generally highly liquid, investing in short-term assets of mostly good to excellent quality, some even investing only in government bonds. Interests may be calculated on a daily basis and then so passed on to the investor. The fund distributes its return, or may be based on capitalization. Most MMFs qualify as UCITS and have a well-established

legal structure, applying some of the less restrictive provisions addressing their specific portfolio composition.96 [22.80] Some of these funds are managed in such a way that the value of their shares remains constant—for example one share is worth $1—and this value remains constant over the entire life of the fund. Investors often assimilate these funds to deposit accounts, on which—as is the case in the US—investors can withdraw at nominal value, order payments, or write cheques. The constant net asset value (CNAV) funds have been very popular as equivalents to bank accounts, which, however, offer a higher return. In fact, the asset value of these funds is not a constant one, but depends on changes in the valuation of the underlying assets. The constant par value is ensured by offering the manager of the fund the right to apply certain pricing and valuation conventions, by sponsor support, or by setting aside certain revenues that can be used if the value falls below par. Often they enjoy an AAA rating. A mismatch between asset value and nominal value would not alarm investors and normally it would not lead to a run on the fund. [22.81] There have been a few cases in the US where the CNAV of the MMFs fell below par. The most recent challenge came with the collapse of Lehman brothers, where a big shock to Lehman instruments lead to ‘breaking the buck’; that is to say the value of Lehman securities fell dramatically, leading the funds’ $1 NAV to fall below the par value, resulting in a significant confidence shock and a run for the exit, in turn provoking fire sales on the market. The SEC therefore proposed a series of changes aimed at dealing with the liquidity, interest rate, and credit risk in their portfolios. It also proposed that asset calculation should be based on Variable Net Asset Value (VNAV) for the prime institutional funds, and on liquidity fees and redemption gates for other funds. This move was very strongly opposed by the Investment Management Association. In July 2014 the SEC adopted rules reflecting the proposals mentioned above.97 [22.82] In the European Union, the Commission tabled a comparable proposal addressing the MMFs. Their systemic importance did not need to be underlined, as they stand for more than €1.1 trillion of assets (compared with $2.3 trillion for the US).98 In Europe, they are mainly owned by banks,

institutional investors, and governments, but less frequently by individual investors. The vast majority qualify as traditional UCITS while about 40 per cent have adopted the form of an AIF, subject to the AIFMD.99 It is important to mention that most funds are located in a limited number of Member States, among which France, Ireland, and Luxemburg stand for 90 per cent of the market. The MMFs are also supervised by the supervisors competent for investment funds in general, and the basic principles applicable to UCITS or AIFs apply, with some adaptations due to their specific purpose. [22.83] As a consequence of the financial crisis and to follow up on the recommendations of the FSB, the European Commission tabled a proposal for a regulation in 2013100 which has not yet been adopted by the Union legislature. In Europe, too, the controversy centres on whether to use CNAV or VNAV. [22.84] The main highlights of the Regulation provide for requirements in terms of diversification, quality of the investments, clear and harmonized valuation rules, and minimum liquidity requirements. The Commission proposal clearly opted for VNAV as the standard model, but allowed CNAV provided a number of additional safeguards were fulfilled, the most visible being the requirements to dispose of a buffer that would allow it to bridge the difference between market value and ‘constant NAV’ or par value.101 The buffer was fixed at 3 per cent of assets, the variance margin also observed in the US. The cost of the buffer leading to an increase in the management fee was to be borne partly by the asset manager and partly by the investors. This idea of mandating a buffer was strongly opposed by the industry and was not endorsed by the Parliament’s Econ Committee which adopted an alternative measure such as requiring 90 per cent of assets to be invested in short-term assets of up to ninety days. [22.85] After the parliamentary election of 2014, a new report by Neena Gill MEP contained several other changes to the original proposal. Central is retail CNAV being permitted to be open for distribution only to certain non-profit organizations, charities, and public authorities or foundations. The public debt CNAV would be publicly available and have to invest 99.5

per cent of its assets in public debt. The buffer idea was abandoned and replaced by liquidity fees, and redemption gates. [22.86] It is still unclear how this debate will be concluded.

5. CSDR [22.87] MIFID II lays down the basic settlement rules relating to the connection between the trading market, its participants, and the settlement systems or CCPs. [22.88] Regulated markets should adopt the necessary provisions in their internal regulation to provide for clearing and settlement of their transactions. They should designate the settlement systems for these transactions102 provided that the system has the necessary links with other settlement systems to allow for efficient and economic settlement. But settlement in another system is allowed provided the market supervisor has determined that this will allow for smooth and efficient settlement. 103 [22.89] Investment firms and market operators on MTFs and OTFs have the responsibility to establish and adopt arrangements allowing for the settlement of their transactions.104 This may take the form of links with a CSD or a CCP, or with banks offering private settlement services. [22.90] Freedom of access to CCPs and settlement arrangements on a cross-border basis are important building blocks in the functioning of a Europe-wide financial market. Therefore, Member States are admonished not to prevent regulated markets from entering into ‘appropriate arrangements’ with entities established in other Member States for the settlement or clearing of trades realized in that Member State.105 Normally this would include direct access. For investment firms from other Member States, the rule granting direct access is formulated and this on a nondiscriminatory, transparent, and objective basis.106 [22.91] The European Regulation on ‘improving securities settlement in the EU and on central depositories’, of 23 July 2014 (known as ‘CSDR’)107

is another ground-breaking measure in the sense that a field—securities settlement—which had for a long time remained a national competence remaining outside the ambit of specific EU regulation will henceforth be subject to quite demanding European rules. Here again the systemic argument has been the leading force for redefining the boundaries of regulation and supervision. In addition, the new regulation contains a certain number of micro provisions, aiming at better protecting securities investors, which before were only formulated in voluntary statements or international reports.108 The new regulation is therefore an important case of the transformation of soft law instruments into directly applicable European hard law. [22.92] The adoption of these new measures should not, however, give the impression that a considerable loophole has been closed, or that during the crisis considerable issues or deficiencies at the level of securities settlement were discovered. Rather, to the contrary, the European securities settlement system shows a very high degree of resilience and no significant concerns or dysfunctions have been mentioned. The CSDR mainly organizes the legal framework within which in the future this systemically highly relevant activity will be organized, and the necessary safeguards such as supervision—national or cross-border— introduced. On the other hand, it also aims at creating a more level playing field by ensuring that all market operators and CSDs are subject to identical applicable regulations, standards, and rules.109 [22.93] The CSDR contains different bodies of rules, all applicable to Central Securities Depositories (CSDs) and securities settlement and related activities. CSDs were already subject to national regimes of supervision, but according to the CSDR a regime comparable to the prudential regime applicable to banks will be applied. Importantly, the regulation organizes several aspects of the process of settlement of securities’ transfers, and organizes the private law transaction of executing a transfer of securities, but with very important structural aspects. The latter point will have considerable consequences for protecting the private law position of the parties active in the securities markets, whether directly or indirectly.

[22.94] Central securities depositories or CSDs are mostly organized as separate legal entities, offering an independent service. In some jurisdictions they are organized in the margin of the stock exchanges and are responsible for the settlement of the securities traded on the exchange. In other states, they are independent entities offering a central facility to the intermediaries trading on these markets. As a core market infrastructure they are subject to authorization and supervision, similar to that applicable to banks, but with a twist towards their public interest function. This applies to risk management and resulting own fund requirements, to be established in function of the different types of risks, among which operational risk is predominant. Governance is stricter than for banks as one third of the board should be composed of independent directors, and their remuneration should not be related to the business’ performance. The regulation contains detailed provisions dealing with the organization and risk management of CSDs. CSDs are subject to the oversight of the central banks,110 who oversee the SSS, the systems of settlement of securities, whether at CSD level or in the banking system in general. [22.95] Open and fair access should be guaranteed to all market participants. Cross-border transfers of securities between the EU CSDs should in the future be realized within the framework of a pan-European securities transfer system known as T2S, Target 2 Securities.111 The ‘fair and open access’ principle also applies to issuers who can choose to apply to any European CSD for having their securities included in its settlement system, except if local company law or other regulations would prevent this.112 This provision may contribute to integrate securities trading, at least from the angle of settlement. [22.96] A second part of CSDR relates to organizational matters, which are essential to secure the safe and secure transfer of securities, taking into account the volumes and the speed at which these transactions have to be executed. Therefore, all securities have to be held in book-entry form, based on immobilization of the physical securities or on securities held in dematerialized form, allowing transfer by simple accounting, mostly electronic booking. The regulation does not, however, contain a rule that title is transferred upon booking, although a recommendation to that effect was already formulated in the 1990s.113 Settlement will take place within

two days of the trading date (T + 2), and late settlement—settlement ‘fails’, eventually putting involved parties at risk—will be tracked and sanctioned by the CSD, by imposing fines or applying buy-ins at the risk and expense of the failing party, and this within four days maximum, or ultimately by disclosing the name of a repeatedly failing participant. These provisions should avoid delays and ensure fast and faultless execution, as delays are disruptive and if sufficiently prolonged may lead to blockage in the markets and hence constitute a systemic factor. Similarly, settlement finality— already supported in Directive 98/26—will insure that once a transfer order is introduced into the system, execution will take place making the order final. Similarly, the ‘delivery v payment’ (DVP) rule avoids securities being transferred without corresponding payment. All transactions should be settled in central bank money, and for foreign currencies through cash correspondents. [22.97] CSDs also carry considerable amounts of securities on their books. Therefore, several provisions of the CSDR aim at protecting the investors. The ‘integrity’ of the issue must therefore be secured by imposing daily reconciliation throughout the holding chain and with all parties involved, to verify that the number of outstanding securities is not greater than the one created. Securities overdrafts or securities creation would put the reputation of the entire system gravely at stake, but would be very damaging to investors, whose confidence would be severely shaken. In the CSD’s books, the securities of the CSD and those of the clients have to be segregated, but the latter can be held in omnibus accounts. There is a tendency to move to individual client segregation, avoiding any overspill between individual holdings of securities. The settlement procedures to be followed have to be fully secure, for example by imposing DVP or settlement in central bank money. [22.98] Special mention can be made of the cross-border aspects: within the EU, freedom of services allows a regime akin to the banking regime to be applied—no authorization needed for starting an activity in another Member State. For foreign CSDs, a recognition process is put in place, leading, for third-country CSDs, to an equivalence decision by the Commission and effectively decided upon by ESMA.114 CSDs offering

banking services—directly or through a separate entity—are subject to stricter requirements.115 [22.99] This new regulation—and its implementing regulations116—have raised the legal status of this core activity and its participants from a regime that was largely regulated at the national level and in non-regulatory statements117 to one that will in the future be regulated at EU level and strictly supervised. Driven originally mainly by systemic considerations, it has introduced quite an elaborate system of rights and duties of the different parties, strengthening the legal position of the market participant. In particular, the provisions on settlement finality, DVP, segregation, and strict execution within two business days are instrumental in providing, by guaranteeing efficient execution mechanisms, adequate protection to the participants in the securities markets.

6. Securities Financing Transactions [22.100] Securities financing transactions (SFT) have been identified as one of the major vectors of systemic risk in the financial system. The subject refers to a certain number of frequently used systems for creating liquidity or collateral mainly in the interbank market. In the official documents Securities Refinancing Transactions have been defined as ‘repurchase transactions’, better known as ‘repos’ or ‘commodities or securities lending’. The call for taking a regulatory initiative has first been formulated by the FSB,118 where it was mainly approached from the angle of procyclicality.119 The FSB also published recommendations on haircuts for not centrally cleared SFT.120 Moreover it paid attention to the need for reliable data.121 [22.101] Data could be collected from different sources, especially also from the CCP, CSDs, and trade repositories for relevant market data. Reporting on securities lending should extend to fund managers and endinvestors, important users of these techniques.

The European Commission is taking advantage of the FSB’s recommendations in the latter’s 2013 Policy Framework: ‘Strengthening Oversight and Regulation of Shadow Banking’.122 The Commission first published a Green Paper on 19 March 2012 followed by a Commission Communication to the Council on shadow banking.123 In the proposal for structural reform (‘Liikanen report’124) it was considered that the structural measures proposed might lead to a shifting of these activities to a less regulated space, outside the supervisory view. This might lead to systemic concerns in the shadow banking area, due to excessive transformation of maturity and liquidity provision, procyclicality, and especially the risk of contagion to other financial sectors, particularly the regulated one. [22.102] SFTs play a central role in the creation of—sometimes excessive —liquidity in most types of interbank or market operations, not only at the interbank level, but also between banks and non-banks, or between nonbanks. This is thus clearly a shadow banking issue. [22.103] The Commission’s proposal on ‘Regulation on Reporting and Transparency of Securities Financing Transactions’125 covers only part of the subject: it essentially requires information of SFT transactions in which different types of counterparties may engage. In addition, it imposes a consent condition with respect to rehypothecation, the latter probably also covering the original idea to deal with asset managers’ use of of their clients’ portfolios. There is no mention of a requirement relating to haircuts for banks financing non-banks against collateral, other than government securities, as was developed by the FSB.126 The proposal to interpose CCPs in the interdealer repo market also remains unmentioned. [22.104] The SFT regulation would cover all counterparties—both financial and non-financial—in the EU and for branches also outside the EU, including UCITS management companies and AIFM. With respect to the rehypothecation provision, it would also apply to the counterparties established in the Union, or outside the Union, to an EU branch or relating to the provision of collateral outside the Union by a counterparty in the Union.127

[22.105] As proposed by the FSB, the first step should consist of collecting information on SFTs. All counterparties are therefore obliged to report their transactions to a Trade repository registered with ESMA, and ESMA will act as the default data collector in the same sense as under EMIR. Here too, ESMA will be the supervisory agency for the TRs. The trade repository will publish aggregate data, but data allowing a party to be identified will remain confidential. Further disclosure is mandated for the investment companies of UCITS or AIFs about the use they make of SFTs; investment funds will inform investors in their prospectuses about the SFTs, including in total return swaps, which they use in their asset management. [22.106] Rehypothecation or ‘reuse’ of collateral has been an oftendiscussed item: it is widely considered a regular market practice, as it leads to an obligation to replace the securities by identical ones. But at the same time it exposes the owner of the securities to a loss of his security interest. This aspect has not prevented rehypothecation from being used on a massive scale, as most of the time the owner or holder of the securities recovers equivalent assets. In the Lehman case this feature was a considerable disappointment for many inadequately warned investors. It would be worth considering how this protection can be harmonized. [22.107] The regulation requires that the party providing the collateral will expressly consent to the rehypothecation, whether in a written agreement or by ‘an equivalent alternative mechanism’.128 Special guarantees should be provided for cash re-use,129 and the assets so used will be booked in a separate account in the name of the receiving party. Asset managers will also inform their clients beforehand about their intended use of rehypothecation. [22.108] The future regulation of SFTs is an example of extending regulation to the shadow banking field, while introducing some essential investor protection provisions.

7. The Changes to the Prospectus Regime

[22.109] A recent Commission proposal intends to adapt the requirement to publish a listing or a public offer prospectus to the changed circumstances after the financial crisis. This proposal is another example whereby the strict provisions of the EU regime are loosened in order to allow for more flexible financing, especially of the SMEs. The proposal is part of the growth agenda, and reduces the safeguards that are normally offered to investors in shares or bonds. More generally this is a field in which systemic concerns would rarely come to the fore, although one might find cases of regulatory arbitrage leading to a concentration of risks in certain jurisdictions. [22.110] Prospectus requirements have been on the table for a very long time, originally only for admission to a stock exchange, later for the public offer of securities.130 The regime was quite stringent, with very elaborate disclosure obligations to which considerable liabilities are attached, disclosures being based on detailed lists of information to be included in the prospectus and submitted to the securities markets supervisors. The regime has become so impracticable and so much geared to avoiding liability for the parties involved that, while scaring issuers away from the market, traditional prospectuses have become almost unreadable. The effectiveness of this regime in terms of investor protection has to be put in doubt. In order to alleviate these concerns somewhat, a more user-friendly document had to be developed. [22.111] A Commission proposal dated 30 November 2015131 intends to adapt the existing regime to make the prospectus a more relevant tool for investors, while reducing the administrative burden for secondary issues or for frequent issuers, and opening access to the market to meet the financing needs especially of small and medium-sized firms. The proposal has also to be situated in the efforts of the Commission to create a Capital Markets Union, with the aim of increasing market funding and correspondingly reducing bank financing, an undeniable macrofinancial objective. [22.112] It is striking that the proposed regime on one hand maintains the prospectus requirement in full force for those issues that aim a first listing on the main securities markets. Indeed, these markets are essential vectors

for public confidence and have to be kept intact from a financial stability perspective. [22.113] But some simplification can be considered even for those largest, listed entities: admission to trading of securities that are identical (‘fungible’) to already listed ones, or are issued in conversion of other securities—exchange of shares for others of the same class, issues of shares for free, for example dividend shares—should not call for a new prospectus. Also a listing on an additional regulated market would be exempted under certain conditions. The existing exemptions for private placements (qualified investors), small issues to less than 150 persons solicited or for less than €500,000 in total or at least for €100,000 per investor, dividend shares are examples where no prospectus is needed in the case of an offer to ‘the public’. These provisions are maintained or slightly adapted. A new exemption applies to offers of more than €10 million, made only in one Member State, provided that the state has opted in to this regime. [22.114] The prospectus itself will contain a summary, the content of which is defined in the regulation in a detailed but limiting way, while the warnings will be restricted to four items. It is unclear whether the full prospectus will still contain the full list of disclaimers and whether these can be invoked against the investors. In the case where a PRIIP has been published,132 the latter may replace the summary. [22.115] Further simplifications are proposed for the admission of nonequity securities (base prospectus and securities note), and for repeated issues of listed securities: a ‘universal registration document’, valid for three years can be drawn up, and will stand for the prospectus. It will be accompanied with a securities note and the summary. [22.116] The most innovative regime is the one applicable to SME issues. They may opt for a minimum regime consisting of a registration document and a securities note, both adapted to the SME case. The prospectus will be composed of a questionnaire with standardized texts, the content of which will be determined in a Commission delegated act. The regime is applicable to the offer of securities by SMEs provided it is not planned to have the securities admitted to a regulated market, though they might be traded on

other trading venues such as an SME growth market under the MTF regime (Article 33 MiFID II). [22.117] The registration document and the securities note will follow the same format. The overview of the risks relating to the issuer will be limited to three distinct categories, defined along criteria of materiality, probability, and magnitude. For sensitive information—especially commercial information—equivalent information will be allowed. This regime is not accessible for SMEs with securities listed on a regulated market. [22.118] This prospectus will be approved following the same procedure as applicable to all other prospectuses, and will further be made available on the website of the home supervisor, and of ESMA. [22.119] From the point of view of extending the regulatory ambit, this proposal will—among other changes—offer a more liberal regime to securities issues of small import, while opening the collection of capital to SMEs, where previously this additional funding could only be delivered by the banking world. Where previously the public issue of securities by smaller firms was only possible by submitting to the very heavy regime of presenting a full-scale prospectus, the new regime withdraws from it, reducing considerably the level of requirements applicable. At the same time, in order to obtain access to financing on the public market, SMEs will have to draw up an information document that one hopes will be sufficiently simple. The link with crowdfunding should be mentioned here. [22.120] A philosophical note may state that in the field of investor protection fewer requirements are admissible, in contrast to the systemic risk of the financial stability field, where more is the rule.

8. New Forms of Direct Financing A. Marketplace Lending, Unregulated Financing, and Other New Forms of Financing [22.121] Linked to the restrictions imposed on the banking industry and taking advantage of the communication facilities offered on the internet, a

new kind of financial activity has developed and has proved to be quite successful. It essentially consists of direct financing, those in need of the funds addressing themselves to the ultimate owners of the funds. The existing conduits are bypassed, and although no new financial products have been developed, the ways the activities are organized are quite innovative. These funding techniques present themselves in many different forms: a main dividing line would be whether the funding is debt financing —including the issuance of securities—or leads to participation in a business venture, rights being represented by shares or other return-related instruments. Another important dividing line distinguishes funding serving economic purposes, and therefore pursuing some profit, from non-profit contributions where returns are mainly non-economic. The dividing line is not always very clear, at least if one analyses some of the leading funding platforms. A further division relates to whether banking institutions are involved, supporting the activity by either intervening themselves or not. Also the terminology under which these new forms of financial activity are carried on is not very well established; some refer to marketplace lending,133 others to unregulated lending, or frequently to ‘crowdfunding’—the latter seems to be the term used by the official institutions.134 From a comparative point of view, the equivalent French terminology is ‘financement participatif’, and the translated Italian is ‘crowdfunding related to innovative start-ups’. Italy was the first country to adopt a specific regulation especially addressing equity crowdfunding, but Germany has also decided to apply regulation to this activity. The types of services that are being offered this way are not limited to funding: automated investment advice135 and insurance services are also being offered this way. [22.122] Direct lending or financing by ultimate investors has always existed and was the original method for financing economic activity, and governments as well. Historically, funding of firms was usually provided by founders, the family, or by private investors known to the founders. This way of proceeding lies at the basis of many of today’s business empires and their related concentrated ownership structure, with the founding families still playing a considerable role. Many of the larger SMEs even today are mainly directly financed this way.

[22.123] Even before the financial crisis, some of the more specialized lending activities emigrated to different types of financial intermediaries: this is the case for credit offered through credit cards, under the form of leasing or factoring contracts, or of different types of credit in which the OFIs are specialized. Most of these financial intermediaries are still financed in the traditional way, through bank credit, by long-term group credit, by publicly issued debt securities, or by concentrated, mostly stable deposits. The new lending models differ from these as they organize direct funding between different types of market participants, the owners of the funds—depositors, investors—and their ultimate users. But they also take the form of direct lending by institutional investors to business firms or infrastructure projects.136

B. Crowdfunding [22.124] Crowdfunding is another example where the widest range of projects, activities, or services are offered to the public. Much of this activity relates to funding, or other financial services, but the range of items on offer is unlimited. Especially in the US, the most unexpected offers can be met. [22.125] The crowdfunding models usually differ as the amounts involved first on the funding side, but also often on the user side, remain, in view of the needs of the users, quite modest—although in some cases they may be very large. The undefined nature of this new development makes it somewhat difficult to describe it more precisely, to measure its significance, or to outline its regulatory status. [22.126] According to IOSCO: Crowd-funding is an umbrella term describing the use of small amounts of money, obtained from a large number of individuals or organizations, in order to raise funds for a project, business/personal loan or other financing needs through online web based platforms. Peer-to-peer lending is a form of crowd-funding used to fund loans, which are paid back with interest. Equity crowd-funding is the raising of capital through the issuance of stock to a number of individual investors using the same method as crowd-funding.137

[22.127] The success of these alternative techniques for SME funding is still relatively limited, as can be derived from the following table relating to 2013 figures as published by AFME, based on 2013 figures (see Figure 22.5).138 While in the US the non-bank sector distributes the majority of the lent funds, the relationship in Europe is more complex, as in some schemes the banks play an active role. The majority of this alternative funding in the EU comes from regulated financial institutions such as banks, insurers, and pension funds (55 per cent compared to 41 per cent in the US). European SMEs strongly prefer bank lending to personal or alternative sources of financing such as family and friends, equity crowdfunding, marketplace lending, loans from non-banks, business angels, and venture capital or private equity funds. For instance, €26 billion was invested by venture capital firms in SMEs in the US in 2013, compared to only €5 billion in Europe. At the same time, €20 billion was invested by business angels in US SMEs, compared to €6 billion in Europe. In Europe, equity crowdfunding investments are still marginal, with €83 million in 2014, but are growing considerably (from €47 million in 2013).139

Figure 22.5: Sizes and sources of financing for SMEs in Europe and the US. * Estimates used. Source: AFME and BCG, ‘Bridging the Growth Gap: Investor Views on European and US Capital Markets and How They Drive Investment and Economic Growth’, February 2015.

[22.128] Crowdfunding is another development that raises numerous issues at the borders of the existing practices and regulations, although up to now at least, no clear financial stability issues have been identified. Central bankers140 have expressed their interest in and support of the new phenomenon. In the meantime, the issues raised generally relate to investor protection, and implementation of the existing regulatory system. [22.129] This provisional conclusion does not mean that there are not a series of items to be further discussed. The excellent and very detailed overview of the legal position of crowdfunding by Guido Ferrarini and Eugenia Macchiavello (Chapter 23) has indicated the present balance of advantages and drawbacks, the latter prevailing from a perspective of protection of market participants, especially retail investors. It leads to the conclusion that it would be useful to impose some order on this matter:

there is a fear that, with the present divergent national interpretations, in some jurisdictions crowdfunding will be very much restricted, while in others widely admitted and considered the perfect case of regulatory competition. The growth agenda which is now strongly defended by the European Commission in the context of the Capital Market Union might suffer from this rather chaotic regulatory environment. It is important that the activities undertaken as part of crowdfunding are adequately regulated, with a view also to protect the parties involved in these schemes— especially the investors—while a clear regulatory support for this financing technique should be pursued. Without being overly restrictive, the necessary minimum safeguards should be provided. Also, placing this financing technique outside the traditional banking world would help to realize one of the objectives of the CMU, that is, diversification of financing techniques, and to that extent would contribute to a shift from a bank-dominated financial system. [22.130] However, crowdfunding is not only about attracting investors to subscribe to securities, mostly of SME, but often relates to other financing transactions than securities or other financial rights, lending being the most important. The need for diversification is equally important in the segment of medium-sized companies, which are often too much geared to bank financing. Many other initiatives, for example in the non-profit sector, are financed this way, sometimes leading to amazing applications,141 where the risk of abuse is as great as in the financial form of crowdfunding. Therefore, an independent overarching framework, that formulates high-level principles while allowing for the different activities that are financed by crowdfunding, seems the right objective. This framework should not consist of a patchwork of changes to existing directives or regulations—see the above-mentioned proposal for amending the prospectus directive—as this would make the scheme overly complex, non-transparent, and unmanageable without expert legal support.142 It would rather consist of an independent regime, containing uniform definitions and well-defined concepts that would be binding in the different EU Member States.143 On the other hand, new regulation should not prevent new developments taking place, and should accommodate many potential needs that today are unknown or undefined. The cross-border aspects have been mentioned by many writers on the subject: this presupposes harmonization of the

applicable regulation and of the definition used in the Member States. It should allow the granting of a European passport. [22.131] The supervision of the existing crowdfunding schemes should be discussed: at present, in most Member States that have published their opinion on the matter the position was negative, tracing the limits where no supervision is applicable, and indicating the lines which should not be crossed in that case. A more constructive attitude could be developed based on a common template which laid down in an EU regulation, would lead to a registration once the different conditions have been met. This approach would also reduce the danger of regulatory arbitrage that might lead to unhealthy developments in some places.

VI. Conclusion [22.132] There can be little doubt that in many segments of financial regulation the financial crisis has been a game changer. Although some of the changes were already on the table before the crisis broke out, the rules have been strengthened and the structures tightened along with developments on the crisis scene. Changes have been numerous: little has been mentioned here about the new supervisory bodies, especially in the banking field. But in the field of securities too, ESMA has received powers to directly supervise CRA and trade repositories, decide on short selling, and register third-country CSDs.144 And many substantive changes will have lasting effects on Europe’s financial markets. [22.133] One can distinguish four policy drivers that have been active during this phase stretching from the beginning of the financial crisis to now. These drivers have regularly been active simultaneously, but their respective impact has changed depending on the phase of the crisis. Nor does the legislative work follow this sequence: even today, several instruments are still pending before the EU legislature, although they mainly relate to issues that were encountered in the earlier years of the crisis.

[22.134] In the first phase, stabilizing the banking system was the highest priority: prudential instruments were introduced by several EU measures, leading to considerable capital increases, new supervisory instruments, strong attention to solvency and liquidity measures, and stronger risk management even with a proposal for structural measures. The work is not finished, but the end is in sight. [22.135] The second lead is safeguarding financial stability and avoiding systemic risk; this has been the dominant consideration in most of the recent regulation. Apart from direct ‘macroprudential tools’ such as interest rate policies, loans to value limitations, and capital buffers, several of the new structures or requirements are directly inspired by macroprudential considerations. [22.136] The third driver is investor or consumer protection, a concern that has not received the most attention during the crisis, but is now high on the priority list again: the recent trend seems to go for some flexibility, at least if that can help in contributing to confidence and stability. [22.137] The fourth motivation is creating growth and reviving financial activity, as expressed in the Growth Agenda and in the CMU: a balance is struck between this objective and the previous ones. The banking world insists on more flexibility to allow for more lending, thereby invoking the fourth objective while the firms look out for alternatives. [22.138] The evolution of the applicable regulation has followed the development of the crisis: In a first stage, in which many banks were directly threatened, the position of the banking sector was strengthened, mainly by increasing capital requirements. Banking supervision in the euro area has been centralised with new institutions in charge of supervision and resolution. Side effects became visible e.g. in the securitisation area, while transactions emigrated to the shadow banking field.

[22.139] Another wave—second not in time, but in impact—flows from overall financial stability and systemic concerns. A considerable number of new legislative instruments are directly inspired by systemic concerns: EMIR is a good example, but MMF future regulation, or SFT, or even short selling are also largely based on this concern. The securities trading

mechanism is being tightened in MiFIR, while CSDR constitutes a considerable step in securing effective settlement of securities after their transfer. Even if systemic risks have not been eliminated, they should be easier to govern. [22.140] Striking for this part of the new regulation is the attention paid to transparency: indeed, the authorities did not have information, or sufficient useful information on large parts of the markets, preventing them from taking effective action. This status has been remedied with respect to the more volatile market segments, such as in the derivatives sphere, in the short-selling trading segment, and for other complex instruments (e.g. repos). These reporting mechanisms do not necessarily go along with drastic supervisory instruments: detailed knowledge of market developments allows supervisors to make use of their existing powers to deal with worrisome developments. However, some segments remain under the radar: parts of shadow banking activity are only indirectly known, through the financing by the banking sector. But other parts remain in the dark: this might be the case for some of the OFIs, but is certainly the case for the much smaller, unofficial lending activity. The concerns in this segment are mainly related to consumer or investor protection, on which additional research should be undertaken. [22.141] Finally, there is the issue of regulatory abundance and complexity: more than 1,500 new official journal pages have been published since 2012, not to mention many of the implementing regulations and national additions. Many more are still to come. Coordination between these numerous provisions is becoming a real efficiency concern. Even access to all available regulations is a challenge, and this in Internet times. Action should be undertaken to integrate and restate large parts of the new regulatory body, with the promise not to change existing texts unless absolutely necessary.145 A final wish: the Commission should strive to create one single integrated database, where all applicable regulations are available and continuously updated. This source of information should be considered authoritative.

1

See Pittsburgh meeting () and the 2010 Seoul meeting: IMF (2011); see Macroprudential Policy Tools and Frameworks Progress Report to G20 27 Oct 2011, Seoul 2011 for the declarations . 2 The link between micro and macro fields is often associated with Sir Andrew Crockett, managing director of the BIS (Bank for International Settlements) who first drew attention to the role of central banks and that of banking supervisors. 3 See among the early positions, P. Tucker, ‘Shadow Banking, Financing Markets and Financial Stability’, Speech 21 January 2010 (Bank of England Speeches, available at ). 4 MiFIR contains a couple of express references to financial stability but this within the context of a central bank activities, or of derivatives trading. 5 Several of the regulations or directives analysed below refer to Article 114(1) TFEU as the legal basis. This provision allows the adoption of measures for the approximation of national provisions aiming at the establishment and functioning of the internal market. 6 See Commission Green Paper on Retail Financial Services, December 2015: Consultation. 7 The European Commission mentioned in its Communication on Shadow Banking that it was continuing to use the term as it had not found a better one, and that the term was in common use in the financial world; see Communication from the Commission to the Council and the European Parliament: ‘Shadow Banking—Addressing New Sources of Risk in the Financial Sector’, COM/2013/0614 final. 8 Quoted from: Some Issues in Financial Reform, Institute of International Finance (25 September 2011). See also: Shadow Banking: A Forward Looking Framework for Effective Policy, IIIF, June 2012, with case studies. 9 See for example FSB Global Shadow Banking Monitoring Report 2015, 12 November 2015, . 10 In Italian, reference is made to ‘banche ombra non regolamentate’. 11 See for other attempts to define shadow banking: E. D. Murphy, ‘Shadow Banking: Background and Policy Issues’, Congressional Research Service, 13 December 2013 , identifying mainly repos, non-bank intermediaries, ABCP, securitization, and MMFs. See also the Deloitte shadow banking index. 12 See for this case: Tucker (n. 3), 7. 13 See by way of an example: General Electric Capital and the list of financial services offered, now GER; see also: . 14 A type of business that is particularly widespread in China.

15

Relating to twenty jurisdictions and the Euro Area. See also the data published by the ECB in its report on Financial Structures 2015, October 2015: . 17 Financial vehicle corporations, e.g. issuers of securitized products: see . 18 Ireland and Luxembourg, but also Germany, France, and the Netherlands. 19 Holding respectively 43 per cent, 30 per cent, and 24 per cent of total MMF assets. 20 This stands for €327 billion of euro-area bank debt and €337 billion non-euro-area bank debt: ECB, Report on financial structures (n. 16), 4.2.2. 21 See ECB, ‘Shadow Banking in the Euro Area: An Overview’, Occasional paper no. 133, April 2012, Charts 2 and 3 by different authors, . 22 Article 4(1)(1) CRR. 23 Reference can be made here to the Deposit Guarantee systems: Directive 2014/49/EU of 16 April 2014 on deposit guarantee schemes, OJEU 12 June 2014, L 173/149; Proposal for amending Regulation (EU) 806/2014 in order to establish a European Deposit Insurance Scheme (2015), but also to the Resolution provisions, where the ‘public interest’ is a condition for applying the resolution procedures (see Article 18(1)(c) of the Single Resolution Mechanism, Regulation 806/2014 of 15 July 2014). 24 See ECB, Report on financial structures (n.16), 46 et seq. 25 See AIG as insurance company which had developed a very substantial CDS business that was not supervised. 26 See Commission Communication: ‘Shadow Banking: Addressing New Sources of Risk in the Financial Sector’ COM(2013) 614 final . 27 FSB, ‘Transforming Shadow Banking into Resilient Market-based Finance: An Overview of Progress’, 12 November 2015, , p. 12 et seq. 28 See FSB, ‘Shadow Banking: Scoping the Issues’, 11 April 2011, . 29 FSB (n. 27). 30 See BIS, ‘Identification and Measurement of Step-In Risk: Consultative Document’ December 2015, , where this risk is defined as follows: ‘Step-in risk refers to the risk that a bank will provide financial support to an entity beyond, or in the absence of, its contractual obligations should the entity experience financial stress.’ 31 See FSB, ‘Regulatory Framework for Haircuts on Non-centrally Cleared Securities Financing Transactions’ 12 November 2015, . 16

32

BCBS and IOSCO, ‘Criteria for Identifying Simple, Transparent and Comparable Securitisations’ July 2015, . 33 Commission Green paper, ‘Shadow Banking’ 19 March 2012, . 34 EBA, ‘Report to the European Commission on the Perimeter of Credit Institutions Established in the Member States’ 27 November 2014, . 35 See Commission Delegated Regulation No 1187/2014 of 2 October 2014 supplementing Regulation (EU) No 575/2013 of the European Parliament and of the Council as regards regulatory technical standards for determining the overall exposure to a client or a group of connected clients in respect of transactions with underlying assets, L 324/1 van 7 11 2014. 36 See among many, the Enron case: C. Williams Thomas, ‘The Rise and Fall of Enron’ (2002) Journal of Accountancy, April, . 37 EBA Guidelines: Limits on exposures to shadow banking entities which carry out banking activities outside a regulated framework under Article 395(2) of Regulation (EU) No 575/2013 EBA/GL/2015/20 14 December 2015; see also EBA Report on Institutions’ Exposures to ‘Shadow Banking Entities’, 2015 data collection. 38 This does not mean that the activity of these funds cannot raise shadow banking issues, such as contributing to unregulated lending activity, or create runs. So, for example, real estate investment funds would not be included. The AIFs are subject to certain financial stability provisions, according to Directive 2011/61 of 8 June 2011, for example Article 25(7) or 53. This exclusion in the report is probably due to institutional reasons, funds being normally under the supervision of the securities supervisors. 39 ‘Excluded undertakings’, which are subject to an appropriate and sufficiently robust prudential framework, are not to be considered as shadow banking entities. This underlines the ‘objective-based’ definition of the notion of shadow banking. Excluded undertakings include bank-like activities involving traditional maturity transformation, liquidity transformation, leverage, credit risk transfer, or similar activities. These are defined in EBA/GL/2015/20 03/06/2016, nt. 36. 40 Based on Article 395(2) of CRR. 41 CRD III 2010/76, of 20 November 2010 ‘regarding capital requirements for the trading book and for re-securitisations, and the supervisory review of remuneration policies’ amending Annex V to VI (standardized approach of Directive 2006/48). 42 See Directive 2009/111/EC of the European Parliament and of the Council of 16 September 2009 amending Directives 2006/48/EC, 2006/49/EC, and 2007/64/EC as regards banks affiliated to central institutions, certain own funds items, large exposures, supervisory arrangements, and crisis management. 43 Also IFRS 7 on disclosures relating to off-balance-sheet exposures in the case of transfers of financial assets; IFRS 10 and 11.

44

Source: AFME. An overview of these can be found at the Commission’s website, . 46 Proposal for a regulation laying down common rules on securitization and creating a European framework for simple, transparent, and standardized securitization and amending Directives 2009/65/EC, 2009/138/EC, 2011/61/EU and Regulations (EC) No 1060/2009 and (EU) No 648/2012, COM(2015) 472 final . 47 See Basel Committee on Banking Supervision and IOSCO, ‘Criteria for Identifying Simple, Transparent and Comparable Securitisations’, July 2015, (). 48 See ‘Press Release: ECB Announces Operational Details of Asset-backed Securities and Covered Bond Purchase Programmes’, European Central Bank, 2 October 2014 . Thomas Hale, ‘ECB Drops Deutsche and SSgA from ABS Buying Programme’, Financial Times, 23 September 2015, mentions that the asset-backed purchased up to mid-September 2015 did not exceed €12 billion while €116 billion covered bonds were bought, and €314 billion sovereign debt. 49 See Commission Proposal for Regulation laying down common rules on securitization and creating a European framework for simple, transparent, and standardized securitization and amending Directives 2009/65/EC, 2009/138/EC, 2011/61/EU and Regulations (EC) No 1060/2009 and (EU) No 648/2012, 30 September 2015 COM(2015) 472 final, referring to the CRR. 50 Article 3(3) Commission proposal. 51 Article 4(5) Commission proposal. This exception only applies for the retention requirement. 52 Articles 8 to 10 Commission proposal. 53 See for the specific, Article 4(2) Commission proposal. 54 Article 5 Commission proposal. 55 As defined in Article 2(12) as being insurance, reinsurance, pension funds, AIFs, and self-managed UCITs. 56 See Article 3(1) and (3) Commission proposal. 57 Article 122(a) of CRD II, Directive 2009/111 of 16 September 2009. 58 Not applicable to index-based securitization; see Article 4(4) referring to a clear, transparent, and accessible index. 59 See in that sense Recital 11; but investors should also exercise ‘appropriate due diligence’. 60 Article 10(2) Commission proposal. 61 Article 14(4) Commission proposal. 45

62

See AFME, ‘Press Release: Buy and Sell Side Join Forces in Support for SDS Securitization’, 3 March 2016; and the Joint position paper, March 2016. 63 See ‘European Parliament Puts a Brake on Plan to Boost Capital Markets’, Financial Times, 3 March 2015, obviously linking the subject to the creation of a Europe-wide deposit guarantee scheme. 64 MiFIR, Article 2(28) defines ‘structured finance products’ as ‘those securities created to securitize and transfer credit risk associated with a pool of financial assets entitling the security holder to receive regular payments that depend on the cash flow from the underlying assets’. 65 Article 26 MiFIR. 66 See Article 20 MiFIR. 67 The admission of a derivative to trading is subject to a condition or orderly pricing and effective settlement, MiFIR Article 1(1)(f); Article 51(2) MiFID II. 68 Article 32 MiFIR. 69 Article 14 MiFIR. 70 See Article 32(4) MiFIR. 71 Article 34 EMIR. 72 See Article 10 EMIR, applying the rule to the ‘financial derivatives’ except those that are directly related to commercial activity of treasury management. 73 Regulation No. 648/2012 of 4 July 2012
on OTC derivatives, central counterparties, and trade repositories, OJEU 27 February 2012, L201/1.
 74 Portfolio compression consists of compensating trades between a number of market participants on the basis of pre-established criteria. The compensated trades are then considered to be terminated. This process would allow the reduction of the flow of transactions that have to pass through the CCP and in that sense reduce the overall risk. Compression is regulated under Article 31 of MiFIR; as an alternative to market execution, similar provisions apply with respect to transparency, reporting to an APA, and recordkeeping Article 31. It may be practised by specialized, not regulated firms (MiFIR, Recital 8), by CSDs or CCPs, trade repositories, and investment firms. 75 See Pittsburg conclusions, nt. 1; see MiFIR, Recital 25. 76 See BIS Statistics, Table D5, H1 2015. 77 The derivatives traded on regulated markets are cleared in the markets’ clearing systems (e.g. Eurex Clearing). Starting from 2016, these are eligible as CCPs under the EMIR regime. 78 See Article 2(1) of EMIR. 79 According to Article 16 EMIR, for an initial amount of €7.4 million to which retained earning and reserves have to be added. But the real creditor protection is to be found in the margins to be posted by clearing members, in the default fund. 80 Article 41 EMIR referring to losses resulting from at least 99 per cent of exposures over an appropriate time horizon. Along with the default fund to be created under Article 42, additional financial resources will enable the CCP to ‘withstand the default of at least

the two clearing members to which it has the largest exposures under extreme but plausible market conditions’ (Article 43). Under the waterfall, after having used the margins of the defaulting member and its contribution to the default fund, the contribution of nondefaulting members can also be used. 81 See Article 45 EMIR. 82 Applicable to CCPs and their clearing members, to financial counterparties and to trade repositories. State financing and financing by State guaranteed entities are not subject, but local communities are, see Article 2(8) of EMIR; a special regime applies to pension funds. 83 Are exempted: intragroup derivatives and derivatives relating to risks from commercial or treasury activity: Article 10(3) 84 See Draft technical standards under the Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012 on OTC Derivatives, CCPs and Trade Repositories, 27 September 2012, 2012/600; the numerous delegated regulations applicable to derivatives can be found under: European Commission Derivatives. 85 See Recital 23 for requirements applicable to third-country derivative contracts. A recent press release indicates that an agreement on a US–EU coordination agreement about cross-country clearing has been reached. The United States Commodity Futures Trading Commission and the European Commission: ‘Common approach for transatlantic CCPs’, 10 February 2016, . 86 Regulation No. 648/2012 of 4 July 2012. 87 These will be subject to stronger risk-management and collateral requirements. 88 See also Article 25 for non-EU CCPs: Commission Equivalence Procedure: Practical implementation of the EMIR framework to non-EU central counterparties (CCPs), 13 May 2013. 89 See Article 39(7) Regulation No. 648/2012 of 4 July 2012 and Recital 64. 90 Regulation (EU) No 236/2012 of 14 March 2012 on short selling and certain aspects of credit default swaps, OJEU 24.3.2012; Commission delegated Regulation No. 918/2012 of 5 July 2012 supplementing Regulation (EU) No 236/2012 of the European Parliament and of the Council on short selling and certain aspects of credit default swaps with regard to definitions, the calculation of net short positions, covered sovereign credit default swaps, notification thresholds, liquidity thresholds for suspending restrictions, significant falls in the value of financial instruments and adverse events, 9 October 2012. L. 2714/1. 91 ECJ C-9/56 of 13 June 1958 (Meroni): ECJ, Case C-270/12, United Kingdom of Great Britain and Northern Ireland. 92 See for a detailed analysis: M. Chamon, ‘Transforming the EU Administration: Legal and Political Limits to Agencification’, Doctoral Thesis (2015, University of Gent); EU Agencies OUP, 2016; M. Chamon, ‘EU Agencies between Meroni and Romano or the Devil and the Deep Blue Sea’ (2011) Common Market Law Review 1055–75.

93

The reporting regime under the short selling regulation differs in certain respects from the more general reporting under MiFIR (Recital 34 MiFIR). 94 See Article 14(2) Short Selling Regulation. 95 CESR, ‘Guidelines on a Common Definition of Money Market Funds’, 19 May 2010, 10-049; ESMA, ‘Follow Up Peer Review: Money Market Funds Guidelines’, 16 February 2016, 2016/297; see also: IOSCO, ‘Peer Review of Regulation of Money Market Funds’, Final Report, September 2015, FR 19/2015. 96 See Article 52 on the derogations from the diversification requirements under the UCITS Directive 2009/65 of 13 July 2009, OJ L 302/32, of 17 November 2009. 97 See and regulation 33-9616. The US preference for CNAV is related to preferential accounting and tax treatment. 98 See EFAMA, ‘International Statistical Release’, Q2, 2015. 99 According to the Commission, 60 per cent of the funds, standing for 80 per cent of the assets: Commission Staff Working Document. 100 Proposal for a regulation on Money Market Funds, COM(2013) 615 final— 2013/0306 (COD). 101 See also the position of the Bank of England in Tucker (n. 3), 4, stating they should otherwise become banks. 102 Article 53(2) for the general obligation as to internal regulation by the regulated market, along with requirement for its members. 103 Article 37(2). 104 Article 18 MiFID. 105 Article 55 MiFID. This provision is an application of the Treaty freedom, but was sometimes refused in the past. 106 Article 37(1) MiFID; Article 38 also allows for MTF arrangement with settlement systems in other Member States. 107 Regulation (EU) No 909/2014 of 23 July 2014 on improving securities settlement in the European Union and on central securities depositories and amending Directives 98/26/EC and 2014/65/EU and Regulation (EU) No 236/2012, 28 August 2014, L 257/1 (CSDR). 108 See CPMI-IOSCO, ‘Principles for Financial Market Infrastructures’, April 2012, . 109 Recital 5 of the CSDR. 110 Article 12(1) CSDR, referring to the central bank issuing the most relevant settlement currency and the central bank where the cash leg is settled. 111 See about T2S: ECB, T2S, General principles T2S, 201, elaborating on the 19 principles. 112 Article 49 CSDR. 113 See for the Giovanni barriers, ‘Cross-Border Clearing and Settlement Arrangements in the European Union’, November 2001; ‘Second Report on EU Clearing and Settlement

Arrangements’, Brussels, April 2003. 114 Article 25 CSDR. 115 Articles 54 and 59–60 of CSDR. 116 See ESMA, ‘Technical Advice under the CSD Regulation’, 4 August 2015, ESMA 2015/1219 . 117 See CPSS–IOSCO Principles for Financial Market Infrastructures April 2012; see also in general: BIS, ‘Committee on Payments and Market Infrastructure; Recovery of Financial Market Infrastructures’, October 2014. 118 FSB, ‘Strengthening Oversight and Regulation of Shadow Banking: Policy Framework for Addressing Shadow Banking Risks in Securities Lending and Repos’, 29 August 2013. See also FSB Global Shadow Banking Monitoring report 2012, . 119 See FSB, ‘Shadow Banking: Overview of Progress’ (n. 27). 120 FSB, ‘Transforming Shadow Banking into Resilient Market-based Finance: Regulatory Framework for Haircuts on Non-centrally Cleared Securities Financing Transactions, Including Non-bank to Non-bank Transactions’, 12 November 2015. 121 FSB, ‘Transforming Shadow Banking into Resilient Market-based Finance: Standards and Processes for Global Securities Financing Data Collection and Aggregation’, 18 November 2015, . 122 FSB, ‘Strengthening Oversight and Regulation of Shadow Banking’, 29 August 2013, . 123 Proposal for a Regulation on reporting and transparency of securities financing transactions, COM(2014) 40 final, 4 September 2013, . 124 See Liikanen report ‘Report of the High-level Expert Group on Bank Structural Reform’, 2 October 2012, ; Commission Proposal for a Regulation on structural measures improving the resilience of EU credit institutions COM(2014) 43 final. 125 Of 29 January 2014, COM(2014) 40 final. 126 See FSB, ‘Transforming Shadow Banking into Resilient Market-based Finance Regulatory Framework for Haircuts on Non-centrally Cleared Securities Financing Transactions’, 12 November 2015, . 127 Article 2(1). The scope is mentioned as being the same as under of EMIR, covering CCPs and their clearing members, the financial counterparties, and the trade repositories. It shall apply to non-financial counterparties and trading venues where so provided. 128 One can presume that this refers to ‘express knowledge and consent’. 129 FSB, ‘Transforming Shadow Banking into Resilient Market-based Finance: Possible Measures of Non-Cash Collateral Re-Use’, 23 February 2016, . 130 See for the restated version: Directive 2010/73/EU of 24 November 2010 amending Directives 2003/71/EC on the prospectus to be published when securities are offered to the public admitted to trading and 2004/109/EC on the harmonization of transparency requirements in relation to information about issuers whose securities are admitted to trading on a regulated market, OJ Directive 2010/73/EU of 24 November 2010. 131 Proposal for a Regulation on the prospectus to be published when securities are offered to the public or admitted to trading Brussels, 30.11.2015 COM(2015) 583 final. 132 On the basis of Regulation (EU) No 1286/2014 of 26 November 2014 on key information documents for packaged retail and insurance-based investment products OJ L 352, 9.12.2014, p. 1. 133 See ‘Market Based Lending and Financial Stability Aspects’, FSB Letter of the Chairman to G20 Finance Ministers and Central Bank Governors, 22 February 2016, . 134 Crowdfunding is analysed in more detail in this volume in Chapter 23. 135 So-called ‘robo advisers’ may compare fees and management costs for investment funds. See also: ETFs (e.g. Vanguard Personal Advisor Services); or organize venture capital investment, act as business angels, assist in ETF selection, etc. Sometimes insurance benefits are offered to cover losses due to an abandoned crowdfunded project. 136 See the Dutch formula of ‘regiepartijen’, where Dutch pension funds and insurers lend directly through a private fund, allowing for longer duration and bypassing securitization restrictions, but including less liquidity: Thomas Hale, ‘Dutch Pension Funds Turn Mortgage Lenders’, FT, 15 October 2015. Often, direct lending relates to leveraged buyouts or M&A deals. Some lenders buy debt in the secondary market instead. 137 IOSCO Statement on Addressing Regulation of Crowdfunding December 2015, ; Crowdfunding 2015 Survey Responses Report, FR29/2015, December 2015, . 138 See AFME BCG, ‘Bridging the Growth Gap’ February 2015, . 139 AFME, ‘Raising Finance for Europe’s Small & Medium-Sized Businesses’, October 2015. 140 See Keynote speech by Mario Draghi, President of the ECB, Eurofi Financial Forum, Milan, 11 September 2014. 141 See, for example, the Australian ‘GoFundMe’ platform, and the ‘Send Ballarat Survivors to Rome’ Campaign. 142 See Chapter 23 in this volume. 143 See ‘25 Recommendations pour un Union des marches de capitaux axee sur l’investissemennt et le financement’, Rapport pour le ministre des Finances et des Comptes publics, May 2015 (Report under the leadership of Fabrice Demarigny).

144

See about these institutional developments, E. Wymeersch, ‘How to Make Europe’s Financial Supervisory System More Efficient’ in Festschrift für Professor Theo Baums (Mohr Verlag, 2017). 145 See on this topic, E. Wymeersch, ‘How to Make Europe’s Financial Regulation System More Efficient’ (2016) Revue trimestrielle de droit financier, 2, 18 .

23 INVESTMENT-BASED CROWDFUNDING Is MiFID II Enough? Guido Ferrarini and Eugenia Macchiavello

I. Introduction 1. Expected Benefits 2. Main Risks 3. Key Challenges II. Benefits, Risks, and Challenges III. MiFID I 1. Financial Instrument 2. Investment Service IV. 1. 2. 3. V. 1. 2. 3. 4. 5.

National Approaches to Crowdfunding Is MiFID Too Much? Common Focus on Disclosure Obligations Special Requirements MiFID II A Response to the Great Financial Crisis Definitions and Exemptions Tied Agents and OTFs EU Passport and Applicable Law Policy Considerations

VI. Concluding Remarks

I. Introduction [23.01] In this chapter, we explore the core issues generated by investment-based crowdfunding from a policy and regulatory perspective with special reference to the most recent developments in EU financial regulation.1 We refer, in particular, to the use of crowdfunding platforms for the offering of equity and debt securities issued by start-ups and SMEs to investors with expectation of financial returns (‘investment-based crowdfunding’ as a subcategory of ‘financial return (FR) crowdfunding’ which also includes peer-to-peer lending).2 These offers are addressed to the public on the Internet through digital platforms, which enable investors to subscribe the investment securities directly on the basis of information published through them by the issuer, concerning the latter and the offered securities.3 A variant of this model is represented by the ‘private’ offer of investment securities issued by start-ups to either accredited or professional investors through digital platforms similar to those employed for crowdfunding in general.4 The concept of ‘marketplace investing’ is sometimes used to cover both public and private offers on digital platforms.5 [23.02] The business models found in practice vary significantly, not only amongst countries but also between platforms. In some cases, equity is issued through the platform to investors who become direct shareholders in the firm.6 In other cases, the investors either appoint the platform as a nominee shareholder in the firm7 or invest in a holding company that underwrites the equity of the firm.8 This is done both to simplify the relationship between issuer and investors and to solve collective action problems, given investors’ rational apathy and low interest in taking part in the corporate governance of the issuer. In still other cases, investors—rather than purchasing shares—enter into contracts attributing them a share in the profits of the firm.9

[23.03] From a policy perspective, crowdfunding entails finding a balance between, on one hand, the benefits of an atomistic distribution of financial instruments and wider access to finance, as allowed by the Internet, and, on the other, the costs of investor protection in a system that by lowering the distribution costs makes fraud more likely.10 Moreover, insufficient control by investors will likely aggravate the moral hazard of issuers, for the low amounts invested by the individuals in the crowd may not justify the time and effort required for selecting and monitoring their investments. This raises the question whether and to what extent crowdfunding should be incentivized under capital markets law through carve-outs in prospectus and investment services regulation.11

II. Benefits, Risks, and Challenges 1. Expected Benefits [23.04] In a 2014 Communication the European Commission explored the issues and perspectives of crowdfunding in the EU, arguing that there is great potential in it to complement traditional sources of finance and contribute to the financing of the real economy.12 In the Commission’s opinion, crowdfunding can offer various benefits to a large spectrum of users.13 Indeed, access to finance is one of the most pressing problems for SMEs, due to deterioration in public financial support, access to loans, trade credit, and willingness of investors to invest in equity: ‘Many projects’ demand for financing is not met by any existing source of finance, which is referred to generally as the financing gap. Crowdfunding matches small— or even bigger—contributors and investors directly with the projects in need of funds, mainly in the early stages’.14 Compared to other types of finance, it can also reduce transaction costs for SMEs.15 In addition, crowdfunding can foster entrepreneurship by offering ‘an additional market testing and marketing tool, which can help entrepreneurs acquiring relevant knowledge of customers and media exposure’.16 Crowdfunding can in fact represent for entrepreneurs an important pre-sale validation of their product,

as well as a feedback and publicity tool before bringing it to the market, also taking advantage of the intense use of social networks and other communication channels by the crowd.17 Similar non-financial aspects of crowdfunding have been identified as key to its success.18 However, although many equity offerings also include other forms of rewards,19 nonfinancial motivations appear less relevant in the investment-based context where financial motivations prevail.20 [23.05] From the investors’ perspective, crowdfunding offers not only diversification opportunities and potentially higher returns, but also ‘direct choice over where to put one’s money and a sense of involvement with the project’.21 Surveys attest that investors also find the challenge appealing, and the consequent experience and the opportunity to participate in innovative projects and to expand their network.22 Finally, it can also represent an investment resilient to local economy fluctuations when targeting projects located in countries with different employment and economic sectors characteristics.23 [23.06] From a more general point of view, the existence of crowdfunding platforms improves competition in the financial markets, increasing incentives for traditional operators to innovate and serving all parts of the community for lower costs.24 Furthermore, it can stifle innovation by ‘financing innovative projects that do not have the level of maturity that traditional financial market sources require’.25

2. Main Risks [23.07] There are also risks to crowdfunding activities. An IOSCO working paper of February 201426 analyses the potential systemic risks of FR crowdfunding27 and the main investor protection concerns arising from these activities.28 No doubt, the size of the equity crowdfunding market is also small in the UK, which is the country where marketplace investing is more successful.29 Globally, ‘there are very few equity crowdfunding platforms, with the majority focusing on angel investors or sophisticated

investors due to regulatory requirements’.30 This reduces concerns as to the relevance of equity crowdfunding to systemic stability. Nonetheless, the interconnectedness between this sector and the mainstream might exponentially increase in the near future and consequently entail systemic risk.31 In the lending compartment, already between 80 and 90 per cent of the total capital invested on the two major US platforms comes from professional investors (including banks),32 while in the UK institutions account for one third of Funding Circle loans.33 Banks are also considering creating their own platforms.34 A similar trend is noticeable in the equity segment: for example, banks and institutional investors hold 30 per cent of the capital of Triodos Renewable.35 [23.08] Other risks arise to investors participating in equity crowdfunding. First, ‘there is little to no secondary market for the equity of start-up companies’.36 At present there are only a few investment-based platforms providing their investors with some sort of secondary market.37 The only realistic chance that investors have of liquidating their holding in a start-up is in the case of a public float.38 Liquidity events, however, may also occur if the company is sold to another firm or to a private equity or venture capital investor. Second, the crowdfunding market is highly concentrated as there are few established players with a limited number of campaigns reaching the target amount of investors;39 furthermore, there is the risk of investors’ exposure to a single asset, as diversification is not widespread amongst European investors.40 In the case of equity crowdfunding, ‘both the practices and regulations are designed to protect investors from concentration risk and prevent them from incurring losses’.41 In fact, several platforms self-impose investment limits with reference to the same issuer or even same platform and recent crowdfunding reforms have introduced similar limits (see Section IV below on German and Spanish legislation).42 In many platforms the average number of investors per project is around 200.43 Third, the crowdfunding market is relatively transparent, but there are risks to transparency in the absence of regulation, particularly if the equity offer does not fall under the prospectus obligation. The lack of sufficient disclosure could lead retail investors to losses, for they ‘may not adequately understand the risks involved, invest in a product

not aligned with their risk appetite and ultimately incur a loss which they may not have the resources to absorb’.44 As an example, the FCA, analysing the UK crowdfunding market, has reported [a] lack of balance, where many benefits are emphasized without a prominent indication of risks; [i]nsufficient, omitted or the cherry-picking of information, leading to a potentially misleading or unrealistically optimistic impression of the investment; [t]he downplaying of important information. For example, risk warnings being diminished by claims that no capital had been lost or the relevant risk warnings being less prominent than performance information.45

[23.09] The above-cited IOSCO paper also highlights concerns relating to investor experience: ‘investors can and do make decisions based on personal biases and persuasive narrative, rather than on financial experience, due to the social networking aspect of peer-to peer lending platforms’46 and of equity crowdfunding. As an example, a recent survey about investment-based crowdfunding in the UK reports that only ‘38% of investors surveyed were classified as professional or high net worth individuals’, while ‘62% described themselves as retail investors with no previous investment experience of early stage or venture capital investment’.47 Investment-based crowdfunding primarily entails investments in seed companies, a market characterized by a high level of asymmetric information, not benefiting from book-building systems or the aggregation of public information of secondary markets and so typically attracting specialist and sophisticated investors such as venture capitalists.48 Furthermore, according to some studies based on empirical evidence, online investors are even more subject than the average traditional investor to herding behaviour, optimism, confirmation bias, and information overload.49 While there is a rich and growing literature about crowdfunders’ decisions in general and, in particular, in the peer-to-peer lending context (providing, however, mixed results),50 data about investment decisions specifically of equity crowd-investors are still limited (the size of the sample is also limited due to the novelty of the sector and the limited number of projects per platform). Preliminary results attest that crowd-investors tend to rely for their investment decisions on the information provided through the portal and on various forms of feedback by other users. In addition, they are inclined to follow large investments by

sophisticated investors with public profiles (signalling the quality of the venture) and to fund projects receiving a large number of early investments (presumably coming from family and friends or others with inside knowledge) and public attention on social networks and media as well as where funders present high retention rates.51 Evidence about the rationality or efficiency of such behaviour is still limited,52 but the sector has shown innovative attempts to reduce asymmetric information. As an example, syndicate deals seem to be attracting a growing interest in the market (see AngelList in the US): angel investors on an equity crowdfunding platform can show their interest in leading a syndicate and so in selecting projects through a due diligence process in which they will directly invest backed by a number of other crowd-investors who agree to participate with a certain size of investment and to pay to the leader a 5–20 per cent carry per deal.53 Furthermore, the sector has recently witnessed a massive influx of venture capitalists (VCs) as users of such platforms and this could lead to a situation in which retail investors are left with only ‘lemons’. However, some platforms (e.g. MyMicroInvest in Belgium) have allowed retail investors to co-invest with experienced business angels, taking advantage of the aboveobserved ‘signalling effect’.54 Finally, there are risks of fraud to investors (which are heightened by the lack of disclosure) and of money laundering through crowdfunding platforms. [23.10] As regards the risks for the firms requiring financing, the disclosure imposed by platforms might imply the dissemination of relevant corporate information and innovative processes or products among the public, including competitors.55 Furthermore, firms resorting to crowdfunding might accidentally violate clauses of previous contracts with business angels which recognize special rights to the latter or limit firms’ freedom in funding in order to avoid dilution. Alternatively, crowdfunding contracts drafted by platforms might include clauses, possibly required by supervisors, limiting the recognition of special rights to future financers (see Section IV.1.C below about the Italian regime).56 Similarly, complex ownership structures might be an obstacle to the smooth exercising of shareholders’ rights and present free-rider or hold-up problems.57 To solve these problems, some platforms ask investors to provide them with a mandate to act on their behalf with the investees.

3. Key Challenges [23.11] Financial return crowdfunding is subject to several rules at both EU and national level, which are intended to reduce some if not all the highlighted risks (see Section III below).58 Moreover, some Member States have already taken specific regulatory action to facilitate crowdfunding, while also aiming to adequately protect investors (as shown in Section IV). However, ‘the danger is that too burdensome and premature regulatory action could stymie the development of crowdfunding, while too lax policies could lead to losses to investors, harming consumer confidence and trust in crowdfunding’.59 Moreover, the different approaches adopted in Member States, some of which rely on mere guidelines, may create legal uncertainty at the European level, where different regimes are in place for investment-based crowdfunding, possibly raising the need for harmonization (see Sections IV–VI).

III. MiFID I [23.12] Investment-based crowdfunding implies some form of intermediation between issuers and investors. However, whether it also gives rise to an investment service under MiFID and what kind of service is possibly involved are questions which are not always easy to answer. First, the investment should refer to a financial instrument. Second, the digital platform should not limit its activity to the listing of investment opportunities, but should offer a facility for the execution of transactions between issuers and investors. Note that in this chapter the use of ‘MiFID’ on its own denotes the MiFID regime as a whole, i.e. MiFIDs I and II, and MiFIR, or refers to Articles that have remained unchanged in the passage from MiFID I to MiFID II.

1. Financial Instrument [23.13] MiFID includes a definition of financial instrument and sets out a list of products that ought to be considered financial instruments at Section

C of Annex 1. The financial instruments most likely to be issued in crowdfunding are ‘transferable securities’ such as shares or ‘mini-bonds’, though others would be available (e.g. units in collective investment undertakings).60 However, some Member States, including Austria, Belgium, Germany, and Sweden, have experience of investment-based crowdfunding involving forms of equity participation that are not considered financial instruments under national interpretations of MiFID.61 Consequently, the relevant platforms are not authorized as investment intermediaries. As an example, profit-participating loans were not qualified as investment products in Germany before the 2015 reform, while silent partnerships and profit-participating rights are considered financial instruments but not transferable securities (see Section IV.1.D below). As a result, most platforms did not even need to apply for a commercial licence, while others were only licensed as brokers under section 34c of the German Trade, Commerce, and Industry Regulation Act (Gewerbeordnung— GewO).62 [23.14] The absence of a financial instrument in principle bars loan-based crowdfunding from qualifying as an investment service. Nonetheless, transferable securities could in theory also be issued for loan-based crowdfunding, which is actually the practice in the US, where loans granted to individuals (P2P) or firms (P2B) are first securitized and then sold to clients of crowdfunding platforms.63 Such a practice makes the two types of crowdfunding very similar and both subject to SEC jurisdiction. However, the analogy between investment-based and loan-based crowdfunding is strong even when the latter does not foresee the issuance of transferable securities, but the investors get slices of the loans collectively extended by them through the platform. Building on this analogy, UK law treats the two types of crowdfunding similarly, broadly applying the same rules to them but restricting, in the case of investment-based crowdfunding, the types of investors allowed given the riskiness and illiquidity of the instruments offered to them (Section IV.1.A below).

2. Investment Service

[23.15] When the platform actively facilitates the execution of transactions concerning financial instruments, the need arises to identify the type of investment service performed.64 As argued by ESMA, the activity most likely to be carried out by investment-based crowdfunding platforms, in the absence of regulatory constraints, is the reception and transmission of orders: ‘the platform receives orders from investors and transmits them to the issuer or another third party intermediary’,65 generally also arranging the deal through the provision of standard contracts and information material. However, the subscription of financial instruments through the platform might even account as execution of orders when the platform acts on behalf of clients to simplify procedures and investor relations management.66 The reception and transmission of orders qualification is implicit in the Italian regime of equity crowdfunding (Section IV.1.C below), which requires platforms to transmit their clients’ orders to either banks or investment firms and is based on the optional exemption foreseen under Article 3 MiFID for undertakings providing only the service of reception and transmission of orders (or investment advice) in transferable securities and not holding clients’ money or instruments. [23.16] ESMA further argues that the service/activity of investment advice is generally not part of the crowdfunding model. However, depending on how platforms present projects, they might in fact make recommendations constituting investment advice,67 which is defined by Article 4(1)(4) MiFID as ‘the provision of personal recommendations to a client, either upon its request or at the initiative of the investment firm, in respect of one or more transactions relating to financial instruments’. Furthermore, ESMA acknowledges that reliance on the platform’s due diligence concerning the proposed investments could lead investors to consider that they are receiving advice even in the absence of personal communications.68 Indeed, French financial authorities (AMF and ACPR) stated that crowdfunding services constitute investment advice in the case of platforms expressing opinions or value judgements on the projects concerned, through the selection and rating of the latter and presentation of the same to clients as suitable for them (the ‘best for you’).69 The recent French reform on crowdfunding pushes this view to the point of considering investment advice as a characterizing feature of crowdfunding and is based

on the optional exemption foreseen by Article 3 MiFID with respect to firms that only provide investment advice to clients (see Section IV.1.C below). [23.17] In theory, the service/activity of placing could also define crowdfunding from MiFID’s perspective. Indeed, ESMA examined the question ‘whether platforms that undertake market offers for project owners are thereby carrying out the MiFID service/activity of placing without a firm commitment basis in regard to project owners’.70 MiFID does not define this service, while traditional public and private placements are in many respects different in practice from crowdfunding through digital platforms. However, one cannot exclude a priori the case of a platform offering the service/activity of placing without a firm commitment basis to its clients.71 Some national financial authorities (including the French and Italian ones) require that a promotion agreement is entered into between the underwriter and the issuer for this service to occur. The French AMF and ACPR, in particular, ask platforms not to actively engage in soliciting new subscribers for individual offers, even by posting a detailed description of projects on a publicly accessible page of their website, in order for their service not to qualify as placement.72 In general, the agreements entered into between the platform and its clients and/or other circumstances will show whether the relevant activity should be defined as either receiving orders from investors or placing securities on behalf of issuers. The impact of the relevant choice depends on the applicable law and the circumstances of the case. In principle, the characterization of the service as placement could entail higher capital requirements and the exclusion of Article 3 exemption, and would create a risk of liability for the platform, to the extent that the latter may appear to perform due diligence with respect to the offered securities and their issuer.

IV. National Approaches to Crowdfunding [23.18] In this section, we analyse the national legislation on crowdfunding in Europe concerning issues like authorization, type and number of investors allowed, type and size of issuers, etc.73 We examine, in

particular, the UK crowdfunding provisions, the French law on financement participatif, the Italian law on equity crowdfunding, and the recent crowdfunding reforms in Spain and Germany, all of which represent, in different ways, an enabling approach to crowdfunding, as discussed below. Interestingly, the countries concerned offer different levels of market development. In 2014, money invested in equity crowdfunding reached the total amount of €111 million in the UK (with a 420% average growth rate (AGR) per annum),74 €29.8 million in Germany (AGR 174%), €18.9 million in France (AGR 94%), and €10.5 million in Spain (AGR 234%).75 Italian equity crowdfunding is underdeveloped:76 there are nineteen platforms in Italy, but only thirty-one offerings have been made, of which only twelve were successfully completed.77

1. Is MiFID Too Much? [23.19] The comparative study of crowdfunding regulation in the Member States raises the question whether MiFID might be too burdensome for the relevant sector and preclude its full development. Indeed, in most countries, with the as an UK exception, there is either a flight or a deviation from MiFID as to equity crowdfunding, as we explain in this subsection.

A. The UK Model [23.20] UK legislation regulates crowdfunding along the MiFID’s model, with some exceptions. UK crowdfunding platforms are subject to the regime of FCA-regulated firms (including disclosure, risk management, and conflict-of-interest rules), which however allows for exemptions.78 The platforms’ activities constitute one of the following: (i) ‘financial promotion’, which is an FCA-regulated activity (not a MiFID investment service), save when addressed to sophisticated investors;79 (ii) ‘arranging deals in investments’, which is a regulated activity substantially including MiFID’s reception and transmission of orders;80 and (iii) ‘execution of orders’, where the platform acts as an agent of the investor.81 However, platforms could also limit their activities to simply bringing the issuer

together with potential investors (which does not amount to reception and transmission of orders in FCA’s opinion)82 or to take advantage of the ‘tied agents’ exemption.83 In the latter case, non-regulated firms offer crowdfunding services as representatives of authorized ones. FCA also supervises loan-based crowdfunding, which is considered a type of investment,84 however it is less risky than investment-based crowdfunding (where illiquid securities and riskier firms are involved).85

B. Spanish Rejection of MiFID [23.21] Spain avoids defining FR crowdfunding as an investment service, while dedicating a special regime to it. The new Ley de fomento de la financiación empresarial (of 28 April 2015, No. 5)86 defines crowdfunding as the ‘putting in contact’, through the Web or other electronic means, a plurality of either individuals or legal entities who offer financing in exchange for a financial return (investors) with either individuals or legal entities who solicit financing in their own name for a project of ‘financiación participativa’ (promoters) (Article 46 Ley de fomento). Crowdfunding platforms are primarily supposed to facilitate financial dealings between promoters and investors (Article 51(1) Ley de fomento) in addition to assessing the admission of investment projects and their compliance with due diligence requirements (Article 66 Ley de fomento). The crowdfunding activity is reserved to a new registered intermediary (Article 48 Ley de fomento) which must register with the Securities Commission (CNVM) under the condition of meeting certain requirements, such as a minimum capital of €60,000 (or, alternatively, a professional insurance or guarantee), ‘fit and proper’ requirements for managers, a business plan, adequate resources, etc. (Articles 55–56). The financial authority can request information, carry out inspections, and impose sanctions (Articles 89–93). Platforms can offer, in addition to the reserved crowdfunding activity, other services, such as advising promoters on marketing and publicity; analysing the projects and assessing the relevant risks, provided that the information published does not constitute a recommendation to investors; creating communication channels amongst users, investors and promoters; drafting standard contracts for the execution

of transactions; informing investors on the development of individual projects; and taking care of debt collection on behalf of investors (Article 51 Ley de fomento). Nonetheless, platforms cannot hold clients’ money, nor perform investment services, such as reception and transmission of orders and portfolio management (Article 52).

C. French and Italian Middle Course [23.22] Italy and France steer the middle course. While assuming that crowdfunding is an investment service—either reception and transmission of orders (Italy) or investment advice (France)—both countries exploit the exemption foreseen by Article 3 MiFID for the relevant services and formulate a lighter regime for crowdfunding platforms. In France, the 2014 law on ‘financement participatif’87 features two new providers of crowdfunding services: the ‘Conseiller en Investissement Participatif—CIP (investment adviser) (Articles 547-1 et seq. Code monétaire et financier) for investment-based crowdfunding, and the intermédiaire en financement participatif—IFP for crowd-lending (prêt participative),88 both supervised by a private organization (still to be created) responding to the AMF and whose registration is only subject to a few requirements (adequate resources, ‘fit and proper’ managers and owners, professional insurance, membership of a professional association). In Italy, firms managing portals must be registered with Consob (the Securities Commission) through a simple and relatively fast procedure basically presupposing the corporate form and ‘fit and proper’ requirements for executives and owners (Article 50-quinquies(3) TUF). However, portal managers are required to transmit their clients’ orders for execution to either banks or investment firms, which must also perform the appropriateness test with respect to clients unless the amount invested is below €500 per investment and €1,000 per year or, in the case of legal entities, €5,000 and €10,000 respectively (Articles 17(2)–(3) and former 17(4) of Consob Crowdfunding Regulation).89 However, under new rules recently approved by Consob,90 platforms can directly perform the appropriateness test (even when below the above-mentioned thresholds: new Article 17(4) of Consob Regulation) provided that they adopt adequate

organization arrangements (Article 13(5-bis) of Consob Regulation). Instead, if they decide not to perform the test directly, the new Regulation, to reduce their regulatory burden, no longer requires them to make investors fill in an online questionnaire (Articles 13(5-ter) and 17(3) of Consob Regulation). In addition, professional investors—now including business angels—should underwrite at least 5 per cent of the total amount of each issue (Article 24(2) and new Article 24(2-bis) of Consob Regulation). [23.23] Differently from Spain, France and Italy also allow MiFID investment undertakings to offer crowdfunding services. In France, investment firms (Prestataire en Services d’investissement—PSI) authorized for the service of investment advice can perform equity crowdfunding services (Articles L547-3 and L547-1.IV Code monétaire et financier).91 In Italy, banks and investment firms (Società di investimento mobiliare—SIM) can also provide equity crowdfunding, subject both to MiFID’s business conduct obligations and to some of the additional disclosure requirements specifically foreseen for crowdfunding platforms (see Section IV.2).

D. The Fragmented German Approach [23.24] Germany follows a case-by-case approach to crowdfunding, depending on the model used and the products offered.92 In fact, the typical German crowdfunding products—profit-participation loans and subordinated debt93—were not considered as investment products until 2015 and therefore, although necessarily originated (according to German banking law) by a bank, they could be split into multiple parts and distributed to the crowd through unregulated platforms.94 Also, after the Small Investor Protection Act (Kleinanlegerschutzgesetz) of 23 April 2015,95 platforms dealing with such products and generally with investment products other than transferable securities only need to apply for a special investment broker licence outside MiFID’s scope (section 34f GewO) (see about the previous regime Section III.1 above). In fact, § 32.1 of the German Banking Act (Kreditwesengesetz—KWG) requires BaFin authorization for firms conducting banking and investment services, but exempts from registration (§ 2.6) investment brokerage, provided that the

broker does not hold clients’ money and acts only as an intermediary between customers and issuers/offerors, and the placement business (which is a special type of investment brokerage).96 Such exempted broker status is however subject, since 2015, to special conditions as to professional insurance; competence; and duties of information, consultation, and documentation (similar to those found in the securities trading law— Wertpapierhandelsgesetz).97 [23.25] However, platforms that either do not meet the conditions set in § 2.6 KWG or offer transferable securities must apply to BaFin for a licence as investment firms subject to MiFID’s implementing provisions. Loanbased crowdfunding remains unregulated (except for the need of a bank originating and splitting the loans) and a commercial licence is only required for the platform under section 34c GeWo as long as investment products are not involved.98 As a consequence, Germany allows unregulated brokers to offer loans and exempted brokers to offer investment products (including financial instruments) other than transferable securities, while requiring an investment firm authorization only in the case of transferable securities.

2. Common Focus on Disclosure Obligations [23.26] The national crowdfunding regimes primarily focus on disclosure obligations, but also consider business conduct rules and asset-separation mechanisms. Even when keeping crowdfunding outside the scope of MiFID, countries regulate it through disclosure provisions and rules of conduct, which are often modelled along MiFID’s provisions and satisfy similar needs in terms of investor protection. For instance, platforms operating in France must perform investment advice in the best interest of clients, including the selection of offers, due diligence checks, disclosure of risks, and other information, and the adequacy test concerning the investment (Article L547-9 Code monétaire et financier). Moreover, they must provide, in a clear, balanced, and non-misleading way, public information concerning the platform and its functioning (e.g. charter, professional association) and the risks of the specific kind of investment (Article 325-35 of the AMF general regulation). In addition, the platform

must offer only to registered investors, who have passed the adequacy test, all the information, in non-technical language, needed to assess their investment, including the issuer’s business plan and financial accounts, the securities offered and the relevant rights, shareholder agreements and relevant clauses of the articles of association, as well as the costs and risks of illiquidity and capital loss (Article 325-35 of the AMF general regulation).99 [23.27] The Italian regime also includes significant disclosure obligations and other requirements meant to contain the risk of fraud by the issuers. As anticipated, some of them, especially the disclosure obligations and the duty to grant certain rights to investors, also apply to banks and investment firms managing portals, which are in any case primarily subject to investment services regulation.100 First of all, the managers of online portals must describe their activities, and disclose the identity of their controlling shareholders and directors and the measures adopted for the prevention of fraud and conflicts of interests. They also have to provide investors with detailed, clear, and not-misleading information about the financial instruments issued through the crowdfunding portal and explain the relevant risks, including the risk of loss of capital and that of illiquidity of the relevant investment. Furthermore, they must inform on derogations from typical shareholders’ rights (Articles 14 and 15 of Consob Crowdfunding Regulation) and on drag-along and tag-along rights in case of change of control (Article 24(1) and Annex 3 of Consob Crowdfunding Regulation). In addition, shareholder agreements must be disclosed to investors (Article 24(1)b of Consob Crowdfunding Regulation). Platforms should also warn retail investors that their investment in high-risk securities should be commensurate to their financial resources and inform the them that they are entitled to withdraw from their investment within seven days (Articles 13(3), 13(5), and 15(1)g of Consob Crowdfunding Regulation). [23.28] Spanish law also requires platforms to inform investors about their functioning and on the relevant risks (such as the risk of loss of capital, market illiquidity, limited governance powers, etc.) and costs of the investment. Platforms should warn investors that they are not an investment firm nor a bank, and that they are neither subject to public supervision nor to a deposit or investment guarantee scheme. They should also provide all

information required for an informed decision, in a clear, concise, and nontechnical way (Article 61). The platform is responsible only for ensuring the completeness of the information provided (Article 71). Platforms are subject to business conduct obligations similar to those applicable to investment firms. They should perform their services with due diligence, in the best interest of clients, and with adequate disclosure, but also have in place adequate policies (proportionate to their size) on conflict of interests (Articles 60 and 62).

3. Special Requirements [23.29] Furthermore, national regimes foresee quantitative thresholds to retail investing in crowdfunding or the issuer offering’s size, in order to limit individual exposure to similar investments and investment concentration. The underlying assumption is that investing in unlisted securities through crowdfunding platforms is a risky business, which should be restricted particularly with regard to retail investors. The UK, in particular, allows platforms to offer their products to professional investors and to certain types of retail investors, such as those who take regulated advice, those who qualify as high net worth or sophisticated investors, and those who confirm they will invest less than 10 per cent of their net assets in this type of security (when no investment advice has been provided, firms must perform an appropriateness test).101 [23.30] Spanish law also provides for quantitative thresholds: promoters can present their project only on one platform at a time and can request no more than €2,000,000 (€5,000,000 when the offer is restricted to accredited investors) (Article 68 Ley de fomento). Retail investors can only invest €3,000 per project and €10,000 per year (taking account of all projects on all platforms), but these limits do not apply to accredited investors, including professional investors (institutional investors and financial entities); entrepreneurs satisfying certain requirements; individuals asking to be considered as accredited investors, provided that either their annual income is more than €50,000 or their financial assets are in excess of €100,000; and individuals receiving investment advice from a regulated entity (Article 81 Ley de fomento).

[23.31] German investors in profit-participating loans, subordinated debts, and similar investments, when they are individuals or entities other than companies, encounter special thresholds to their investment, the amount of which varies depending on their annual income, but only applies to the amount invested in an individual issuer. In particular, when the investor declares the full availability of €100,000, such investment limit is set at €10,000; when the investor denies such availability, the limit corresponds to the double of monthly net income, in any case capped at €10,000. In all other cases (including when the investor refuses to make a statement about the funds availability) the investment limit is €1,000. No distinction is made between retail and professional investors. [23.32] The Italian regime does not foresee any limitation to the overall amount of financial instruments that can be sold either to an individual investor or by the same issuer through crowdfunding. Rather it sets certain thresholds from the duty to perform the appropriateness test. In any case, retail investors can only be granted access to the details of the offers after having read the investor education material drafted by Consob, having passed a test proving their understanding of the risks and characteristics of the investment, and having declared an ability to afford the possible loss of the money that they intend to invest (Article 15(2) Consob Crowdfunding Regulation). Similarly, in France investors do not encounter investment limits but are permitted to access to more detailed information about the projects once they have passed the suitability test and confirmed acceptance of the specific risks. Limits also exist with reference to the total amount of the offer from the same issuer. In Italy each issuer has to limit its offer to €5,000,000. In France and Germany the threshold (respectively of €1,000,000 and €2,500,000) is referred to the offer of each issuer on a certain platform, therefore allowing them to use multiple platforms. Instead, the Spanish €2,000,000 limit (or €5,000,000 in the case of audited financial statements) is set per offering as well as per issuer on any platform. [23.33] Some countries entail additional protections for retail investors. The Italian regime takes care of the illiquidity problem affecting the financial instruments issued through crowdfunding by requiring investors to

be granted either a withdrawal right (seven days) and/or drag-along obligation in the start-up company’s charter for the case that the controlling shareholders sell their shares to third parties (see Article 24(1)a of Consob Crowdfunding Regulation). German and English crowd-investors also have a fourteen-day withdrawal right (and, in the UK, access to redress mechanisms).

V. MiFID II 1. A Response to the Great Financial Crisis [23.34] The MiFID review was clearly influenced by the 2008 financial crisis.102 Indeed, the review has reduced Member States’ discretion by resorting to a maximum harmonization Directive, a Regulation, and detailed Level 2 measures.103 Moreover, investor protection has been boosted in several ways. First, MiFID II extends core conduct-of-business obligations —such as the duty to act fairly, honestly, and professionally; and the duty to disclose information in a comprehensive, fair, clear, and not misleading way —to all categories of investors, including professional ones and eligible counterparties.104 This has implications for our main topic, to the extent that professional investors also get access to crowdfunding platforms. Second, MiFID II enhances the pre-contractual information duties with regard to the investment firm and its services, the financial instruments and proposed investment strategies, execution venues, and all costs and related charges.105 [23.35] Third, MiFIR assigns wide product-intervention powers to ESMA and national authorities,106 while MiFID II also provides for productdevelopment and distribution-related obligations.107 No doubt, productintervention powers could also apply to crowdfunding activities and/or to the instruments that are distributed through them. However, this also depends on the meaning and scope assigned to the term ‘manufacturing’, since the products are usually issued by a client-firm which offers them on the platform.108 More likely, a crowdfunding intermediary could be seen as

‘marketing’ the financial instruments which are offered by its client firms on the platform.109 [23.36] Fourth, MiFID II restricts the scope of the execution-only regime, which does not entail the appropriateness test with respect to clients, to certain ‘non-complex’ instruments, such as shares in companies (provided that they are admitted to trading on a regulated market or MTF), shares in non-structured UCITS, or bonds and securitized debt traded on regulated markets or MTFs, excluding those that embed a derivative or incorporate a structure which makes it difficult for the client to understand the risk involved.110 This provision could be of interest to crowdfunding to the extent that either bonds or securitized debt are involved, given that shares in crowdfunded firms are generally not traded in regulated markets/MTFs.

2. Definitions and Exemptions [23.37] The new framework, by reducing Member States’ discretion, restricts the scope for national crowdfunding regimes. In particular, the new formulation of MiFID’s exemptions will require reconsideration of them. Article 3(1) MiFID II allows Member States not to apply the Directive to any persons for which they are the home Member State, provided that the activities of those persons are authorized and regulated at national level and those persons are neither allowed to hold client funds or client securities and/or money nor to provide any investment service except the reception and transmission of orders in transferable securities and units in collective investment undertakings and/or the provision of investment advice in relation to such financial instruments.111 Moreover, the same persons, while providing the relevant service, are allowed to transmit orders only to investment firms, credit institutions, collective investment undertakings, and investment companies, under the conditions foreseen by Article 3(1)(c). In addition, national regimes shall submit the persons at issue to requirements that are at least analogous to the following MiFID requirements: (a) conditions and procedures for authorization and ongoing supervision; (b) conduct-of-business obligations; and (c) organizational requirements.112 Furthermore, Member States shall require the persons

exempt from the Directive to be covered by an investor-compensation scheme.113 [23.38] On the whole, the MiFID II exemptions regime has become tighter than the previous one, as it requires the exempt persons to be authorized, regulated, and supervised at national level and to be subject to conduct-of-business obligations and organizational requirements similar to those carried by MiFID. The national crowdfunding regimes will, therefore, have to be reviewed in light of these additional requirements and may need amendment to the extent that they do not comply with the same. As a result, the benefits of an exemption in terms of lower regulatory burdens for the platforms will be smaller, while investor protection will be enhanced. At the same time, pressure will mount in practice to place crowdfunding activities under the MiFID regime, with ensuing benefits in terms of uniformity at EU level and integration of European crowdfunding markets. [23.39] Moreover, MiFID II has failed to either clarify or adapt some of the concepts that have an impact on investment-based crowdfunding, such as the definition of financial instrument and the different categories of investment services subject to MiFID. Investment services and activities are still defined as ‘any of the services and activities listed in Section A of Annex I relating to any of the instruments listed in Section C of Annex I’.114 Crowdfunding activities generally relate to transferable securities, which are indicated in Section C(1) and are defined by Article 4(1)(44) as ‘those classes of securities which are negotiable on the capital market’.115 However, some crowdfunding platforms make recourse to investment contracts or products that do not necessarily qualify as financial instruments or transferable securities (such as, in Germany, respectively, profitparticipation loans on one hand, and silent-partnerships and profitparticipation rights on the other) under MiFID. This determines the Directive’s non-applicability to the relevant activities on grounds that appear rather formalistic and would possibly suggest the need for either a new and more inclusive definition of financial instrument or a broader interpretation of the existing one by ESMA and national authorities. [23.40] Section A of Annex I specifies the different services and activities subject to the Directive, including the three which are particularly relevant

for crowdfunding, which are: reception and transmission of orders in relation to one or more financial instruments, placing of financial instruments without a firm commitment basis, and investment advice. However, a definition is only offered for investment advice.116 As a result, the problems dealt with in ESMA’s Opinion on crowdfunding will remain under MiFID II, while the answers given by ESMA should still be valid considering that the relevant provisions are almost unchanged.117 [23.41] Moreover, Article 4(2) MiFID II empowers the Commission to adopt delegated acts to specify some technical elements of the definitions laid down in para. (1) of the same article (including investment services and activities), to adjust them to market and technological developments and to ensure an uniform application of the Directive. This power could be exercised by ESMA also with respect to crowdfunding activities, in order to overcome differences in MiFID’s interpretation across Member States and to ensure the uniform application of the Directive. Points for clarification could include the following: (i) whether and to what extent crowdfunding platforms can be considered as offering only services other than investment services (an issue particularly raised by the Spanish law considered in Section IV.1.B above); (ii) the criteria for classifying crowdfunding activities as either reception and transmission of orders or placement of financial instruments; and (iii) the extent to which investment advice can be held to characterize crowdfunding activities (an issue particularly raised by the French law considered in Sections III.2 and IV.1.C above).118

3. Tied Agents and OTFs [23.42] As argued already, UK practice sees crowdfunding activities as often performed by tied agents of regulated firms (Section IV.1.A). This has been possible under former Article 23(1) MiFID I which states: Member States may decide to allow an investment firm to appoint tied agents for the purpose of promoting the services of the investment firm, soliciting business or receiving orders from clients or potential clients and transmitting them, placing financial instruments and providing advice in respect of such financial instruments and services offered by that investment firm.

No doubt the recourse to tied agents allows, in practice, firms other than investment firms to perform crowdfunding services in ways which are not too different from those seen, for instance, with respect to Italy, where crowdfunding portals that are not managed by an investment firm or a bank are bound to transmit their clients’ orders of equity securities to either a bank or an investment firm. In essence, the recourse to tied agents can be regarded as a means for reducing the costs of crowdfunding and responds to a rationale similar to that supporting the national laws that regulate crowdfunding outside MiFID’s scope. [23.43] Given that, as explained in the previous section, benefiting from exemptions has become more difficult under MiFID II, one could speculate that the UK approach to crowdfunding, which is based on tied agents, will become more popular in the rest of Europe. This hypothesis is supported by the fact that MiFID II now requires (rather than only permitting) Member States to allow investment firms to appoint tied agents (Article 29(1) MiFID II).119 Member States retain discretion only in deciding whether to allow tied agents to hold clients’ money and instruments. However, the investment firm appointing tied agents is fully and unconditionally responsible for them.120 [23.44] Also, the organized trading facility (OTF)—a new category that was introduced by MiFID II to cover trading venues which are neither investment intermediaries stricto sensu nor regulated markets/MTFs— could in principle be instrumental to further crowdfunding activities. However, the definition of an OTF clarifies that only bonds, structured finance products, emission allowances, and derivatives can be traded on this type of venue.121 As a result, shares and similar equity instruments, which are the securities typically offered on crowdfunding platforms, are excluded from the OTF category. Moreover, OTFs are multilateral systems in which multiple third-party buying and selling interests interact in a way that results in a contract. As argued by ESMA, in crowdfunding there are only one seller and multiple buyers; in other words, crowdfunding gives rise to a primary market, while MTFs and now OTFs are multilateral systems giving rise to secondary markets.122

4. EU Passport and Applicable Law [23.45] The diversity of national regimes applicable to crowdfunding is problematic for platforms trying to offer their services in more than one EU country. Indeed, cross-border crowdfunding is still limited. According to the 2013 public consultation on crowdfunding held by the European Commission, only 38 per cent of European platforms (either equity- or lending-based) operated across borders because of insufficient clarity about applicable rules and excessive diversity of national laws, but 48 per cent declared an interest in extending their reach to other countries.123 More recently, the Finnish Invesdor (both a lending and investment platform) has obtained a MiFID licence as an investment firm authorized to offer investment advice, reception and transmission of orders, and placement of financial instruments, and raised €1,000,000 on its platform to expand in other EU countries.124 [23.46] Clearly, platforms that do not establish a branch in another Member State nor promote their services across the borders (through mailing, cold calls, etc.) are free to accept clients from other countries. However, transnational crowdfunding may require either a branch to be established or a tied agent to be appointed in another Member State, especially if the platform wants to solicit start-ups to make issues and intends to be closer to them. Similarly, investors may need to be solicited in their own countries through direct marketing actions, which give rise to cross-border offer of services by the platform.

A. Crowdfunding as an Investment Service [23.47] In jurisdictions that consider crowdfunding an investment service, platforms registered as investment firms in another Member State can offer their services under the single licence, either by establishing a branch or through a tied agent based in the host country or by providing cross-border services.125 In the first case (right of establishment), the competent authority of the host Member State is responsible for ensuring that the services provided by the branch in its territory comply with its core rules of

conduct (Article 35(8) MiFID II) and with the measures adopted in accordance with Article 24(12),126 which states: Member States may, in exceptional cases, impose additional requirements on investment firms in respect of the matters covered by this Article. Such requirements must be objectively justified and proportionate so as to address specific risks to investor protection or to market integrity which are of particular importance in the circumstances of the market structure of that Member State.127

This provision reflects the ‘general good’ doctrine, which allows host States to impose additional requirements (such as special business conduct rules, but not necessarily limited to them) restricting EU firms’ freedoms when needed to protect the ‘public interest’ of the host state. Under the ‘general good’ test, the requirements must not be discriminatory nor duplicative of the home state’s rules; in addition, they must be objectively necessary to protect a certain ‘imperative general good’ (such as consumer protection), proportionate and not covering a matter already harmonized at EU level.128 [23.48] In the case of crowdfunding, the national laws examined above in this chapter offer several examples of rules that might fall under the notion of special requirements that are justified by the ‘general good’. For instance, the Italian regulation includes special requirements for the disclosure of equity crowdfunding projects to investors, which might qualify as either specification of conduct rules in consideration of the type of business or special requirements under Article 24(12) MiFID II that are objectively justified from an investor protection perspective and should therefore also be applied to crowdfunding platforms operating from other Member States. Similarly, UK regulation restricts the access to investmentbased crowdfunding to retail investors who either get regulated advice or invest less than 10 per cent of their net assets in the securities concerned. This could be seen as a specification of the appropriateness test, which must be made by the platform with respect to its clients, but is more likely an additional requirement under Article 24(12) MiFID II that is justified by the fact the securities issued through investment-based crowdfunding are by definition illiquid and therefore more risky from an investor perspective or a measure of product intervention. Other requirements—such as one limiting the total amount that a firm can issue in one year on a crowdfunding platform to a given threshold—could even represent a case of

product intervention by national authorities, which may need to follow the special criteria and procedure foreseen under Article 42 MiFIR in order to restrict crowdfunding activities with respect to issuers (see Section V.1 above).

B. Crowdfunding as a Different Type of Service [23.49] Jurisdictions which do not qualify crowdfunding as an investment service, but regulate it by requiring ad hoc authorization and some form of supervision, might refuse access to platforms from other Member States under MiFID’s single licence. This is the case of Spain (Section IV.1.B above), which defines crowdfunding as a sui generis activity and requires platforms to be specifically authorized (Article 46 Ley de fomento). The new law applies to all platforms providing services in the Spanish territory and to the participation of Spanish investors and firms on these platforms, unless an investor or firm participates on its own initiative (Article 47 Ley de fomento). As a result, no entity other than platforms authorized under the special regime can offer crowdfunding services in Spain (Article 48 Ley de fomento). The same provision clarifies that an investor or firm cannot be considered as acting on its own initiative when the platform either announces or promotes itself to, or attracts clients or potential clients in Spain, or directs its services or products specifically to either investors or firms in Spain.129 [23.50] Spain could therefore deny access to an investment firm licensed in another Member State and willing to offer crowdfunding services in Spain. A similar claim, however, would raise three objections. The first is based on the qualification of crowdfunding from a MiFID’s perspective. Assuming that crowdfunding activities give rise to either reception and transmission of orders or placement, as argued above (Section III.2), the Spanish position would appear to be in breach of MiFID. The second objection is that, even assuming that crowdfunding could be confined to activities that do not involve either reception and transmission of orders or placement, these activities should also be lawfully performed by investment firms from other countries under the single licence based on Treaty fundamental freedoms. Italy and France would support this contention. In

both countries crowdfunding activities can be performed either by special firms (platforms) authorized ad hoc or by investment undertakings authorized to the relevant investment service (Section IV.1.C above). The third objection is that the Spanish position may not pass the ‘general good’ test, since MiFID is highly protective of the general good and of consumers in the financial sector.130 [23.51] More complex issues could arise if a crowdfunding platform licensed in a country exploiting the Article 3(1) exemption tried to offer its services in Spain, without benefiting from the single licence as the Directive would not apply. Assuming that Spain also keeps its present regime for crowdfunding under MiFID II, this would not necessarily preclude the possibility of the foreign firm offering its services in Spain. The issue would have to be assessed under the general good test, mainly asking whether the other Member State’s crowdfunding regime is substantially duplicative of the Spanish one. Clearly, the possibility that this will occur in practice is greater under MiFID II, for Article 3(2) foresees that exempted firms should be subject to requirements analogous to the Directive’s requirements in several areas. This will make the national regimes of exempted firms more similar and enhance the chances of a firm licensed under one of them being able to offer its services in other Member States under the general good doctrine.

5. Policy Considerations [23.52] To sum up, MiFID II leaves some of the crowdfunding issues unresolved, but may help to resolve others. To exploit the full potential of the Directive, Level 2 and Level 3 measures will also be important for crowdfunding. First, the definitions of investment services could be specified and improved, so as to clarify which criteria should assist in the qualification of an activity as reception and transmission of orders or placement. Similarly, the definition of financial instruments should be clarified, so as to avoid easy circumvention of MiFID II through the use of instruments the qualification of which is presently uncertain.

[23.53] Second, the rules of conduct for investment intermediaries should be specified with regard to crowdfunding activities and with particular reference to the appropriateness test. From such a perspective, the Commission and ESMA should carefully consider the national regimes examined in the previous section, and others which may emerge, in order to identify alternative approaches to investor protection in crowdfunding. They should analyse, in particular, the case for quantitative thresholds to crowdfunding investments, first with reference to the UK approach limiting investment-based crowdfunding to retail investors who get regulated advice or invest no more than 10 per cent of their portfolios; second, with regard to the provisions that are in force in most countries limiting the total amount of equity that a company can issue through crowdfunding on a given platform and/or in a year. The first type of threshold is aimed at protecting retail investors from excessive exposure to risky investments in illiquid instruments; the second is intended to restrain the moral hazard of issuers and to allow crowd-investors to sufficiently diversify over a greater number of issuers. [23.54] Third, the Commission and ESMA should monitor national regimes exempting crowdfunding activities from MiFID, so as to allow sufficient uniformity about applicable conditions under Article 3(2) and protect the right of establishment of firms operating platforms, in view of the formation of an effective pan-European crowdfunding market.

VI. Concluding Remarks [23.55] This chapter has shown that investment-based crowdfunding, while offering benefits to firms, raises serious investor protection concerns, particularly when directed at retail investors. The small size and relative opacity of issuers, the absence of a business history for start-ups, and the difficulty of assessing the business models of innovative firms aggravate the information asymmetries that are typical of capital market transactions. Moreover, the illiquidity of financial instruments issued on crowdfunding platforms makes an exit from the investment difficult, if not impossible, except for cases in which a liquidity event occurs (such as an acquisition of the issuer by a venture capital fund) and the outside investors are either

allowed or required to sell their equity to the new entrants. Diversification is therefore important for managing the risks of investments on crowdfunding platforms, but only institutional investors or affluent individuals have the portfolios required for efficient diversification, also taking advantage of innovative information tools such as big data.131 In addition, the information asymmetries affecting investment in SMEs’ securities generally require the knowledge and competence of professional investors and venture capitalists, while retail investors find it difficult or impossible to assess the available information, unless assisted by specialized advisers. However, on some platforms retail investors might overcome similar problems, for example by investing alongside professional investors or venture capitalists, possibly within a syndicate in which the lead investor also performs due diligence over the issuer on behalf of the other participants and in exchange for a fee (see Section II.2). [23.56] As argued throughout this chapter, a trade-off exists between investor protection and economic growth, which legal systems try to solve either by adopting special requirements and/or restrictions for equity crowdfunding or by exempting the same from existing provisions of securities regulation. Indeed, investor protection is a prerequisite for the development of markets, but regulation has a cost which could chill the recourse to capital markets by small firms. As a result, the choice for economic growth and technological development may, to some extent, require loosening the regulatory requirements which would otherwise apply to issuers and investment firms for the protection of investors. [23.57] MiFID includes exemptions that have been exploited by some Member States to lighten the regulatory burden for crowdfunding platforms. In principle, MiFID offers the natural regulatory framework for crowdfunding intermediaries and activities, as particularly shown by ESMA’s opinion on investment-based crowdfunding, which defines the relevant services as either reception and transmission of orders or placement. In addition, MiFID II will enhance investor protection and also the protection of crowd-investors by setting several conditions to the Member States’ recourse to exemptions from the Directive in the case of services like reception and transmission of orders and investment advice. This will reduce Member States’ incentives to adopt special regimes for

crowdfunding, which will fall in any case within the core MiFID II provisions, while incentives may be reinforced to the use of tied agents for the offer of crowdfunding services by investment firms. [23.58] The national regimes analysed in this chapter show different ways to solve the trade-off between economic growth and investor protection, and also different views of the practice of crowdfunding. While the UK considers crowdfunding as a regulated activity that is subject to provisions implementing MiFID (or in any case reflecting MiFID’s model), other countries either reject the definition of crowdfunding as an investment service (Spain) or exempt the relevant activities from MiFID in an attempt to lighten the regulatory burdens for crowdfunding platforms (France, Italy, and Germany). As a result, platforms are subject to a variety of national regimes, which break-up MiFID’s harmonization and make it difficult to create a pan-European market for crowdfunding services. Indeed, nonMiFID platforms cannot operate cross-border under a single licence, while platforms managed by investment firms and willing to exercise their right of establishment might be asked by host Member States to comply with local provisions specifically regarding crowdfunding. [23.59] MiFID II, while enhancing investor protection, does not consider crowdfunding explicitly, nor does it solve the interpretative questions that were answered by ESMA’s opinion on investment-based crowdfunding. As a result, regulatory fragmentation will likely persist in the EU, creating hindrances to the formation of a cross-border crowdfunding market, unless clarification occurs through Level 2 measures. Additional measures—either at Level 2 or 3—could enhance retail investors’ protection in cross-border investment-based crowdfunding, particularly by fixing or recommending monetary thresholds to their participation in individual transactions and by limiting the issuers’ funding requests on crowdfunding platforms. At the same time, national carve-outs from the MiFID regime could still be helpful for the growth of local crowdfunding platforms, which might otherwise find the Directive’s regulatory burden too heavy, particularly in early stages of their development.

1

Although this chapter is the result of joint work, the following sections should be attributed to Eugenia Macchiavello: II.2, III.2, IV, V.1–2, V.4.B. 2 For an up-to-date overview of crowdfunding in general, see OECD, ‘New Approaches to SME and Entrepreneurship Financing: Broadening the Range of Instruments’ (2015) 53 et seq., available at . Financial crowdfunding or financial return crowdfunding includes two types of transactions: lending, whereby investors/lenders expect to receive the interest and the principal at the end of the lending period; and equity, where a privately held company offers securities to the general public through the medium of an online platform. The distinction is therefore made between loan-based crowdfunding—commonly referred to as peer-to-peer (P2P) lending—and equity or, more generally, investment-based crowdfunding, which is analysed in this chapter. See also Eleanor Kirby and Shane Worner, ‘Crowd-funding: An Infant Industry Growing Fast’, Staff Working Paper of the IOSCO Research Department, (22 February 2014), 8–9, available at . 3 A concise description of investment-based crowdfunding is offered by ESMA, ‘Opinion: Investment-based crowdfunding’, (18 December 2014), ESMA/2014/1378, 6–7, stating that: it involves at least three parties: the project owner seeking finance, the platform which acts as intermediary between the project and the investor, and the investor who forms part of the ‘crowd’ funding the project. At least one party must also issue an instrument, often but not always a security. Often this will be the project owner seeking the finance. In this case, investors invest directly in the project, either by buying the security or by acquiring the beneficial rights in the security which is held by the platform in a nominee account. In other cases a company, special purpose vehicle (SPV) or collective investment scheme (CIS) established by the platform or a third party will issue a security which is bought by the investor, such that the investor is indirectly exposed to the project. 4

See e.g. Yannis Pierrakis and Liam Collins, ‘Crowdfunding: a New Innovative Model of Providing Funding to Projects and Businesses’, (5 May 2013), 4, available at , arguing that ‘these platforms are a welcome innovation to the field of business angel investing bringing transparency to what is an opaque process and greater visibility of potential investors for start-ups’. 5 The concepts of marketplace and of platform are used interchangeably in alternative finance parlance to designate the web portals where either lending or securities transactions take place. The notion of marketplace lending refers in particular to loan-based crowdfunding, while that of marketplace investing refers to both investment-based and loan-based crowdfunding (and similar activities). See, amongst many, Ryan Caldbeck, ‘Why an Equity Crowdfunding Site Could Become the Largest Marketplace in the World’,

Forbes, 11 November 2013, . 6 See e.g. Crowdcube in the UK, . 7 This is the case of Seedrs in the UK, . 8 This is the case of WiSEED in France, . 9 This is the case of AngelMe in Belgium, . See European Commission, ‘Crowdfunding Innovative Ventures in Europe. The Financial Ecosystem and Regulatory Landscape’ (2014), 6–7 and 35–9, . 10 On the comparative law, economics and policy aspects of crowdfunding, see Ross S. Weinstein, ‘Crowdfunding in the US and Abroad: What to Expect When You’re Expecting’ (2013) Cornell International Law Journal 46, 427; Eugenia Macchiavello, ‘Peer-to-Peer Lending and the “Democratization” of Credit Markets: Another Financial Innovation Puzzling Regulators’ (2015) Columbia Journal of European Law 21(3), 521; Guido Ferrarini, ‘Investment-based Crowdfunding: Policy Issues and Regulatory Responses’ in Festschrift für Johannes Köndgen (De Gruyter, 2016), p. 183. 11 C. Steven Bradford, ‘The New Federal Crowdfunding Exemption: Promise Unfulfilled’ (2012) Securities Regulation Law Journal 40, 195. 12 See European Commission, ‘Communication from the Commission to the European Parliament, the Council, the European Economic Council and the Committee of the Regions. Unleashing the potential of Crowdfunding in the European Union’, (27 March 2014), Brussels, COM(2014) 172 final, 2. On the general framework of the Commission’s initiatives in this area, see the Green Paper, ‘Long Term Financing of the European Economy’, (25 March 2013), COM(2013) 150 final (dealing with the different factors that enable the European economy to channel funds towards the long-term investments needed to ensure economic growth) and the follow-up ‘Communication on Long Term Financing of the European Economy’, (27 March 2014), Brussels, COM(2014) 168 final (outlining priority areas where the Commission intends to take initiatives to help SMEs attract funding, one of which is crowdfunding). 13 Communication, ‘Unleashing’ (n. 12), 4, making reference to crowdfunding’s ‘flexibility, community engagement, and the variety of financing forms it can offer’. 14 Communication, ‘Unleashing’ (n. 12), 4. 15 Guido Ferrarini and Andrea Ottolia, ‘Corporate Disclosure as a Transaction Cost: The Case of SMEs’ (2013) European Review of Contract Law 9, 363, 375 et seq., analysing crowdfunding as a tool for reducing information and transaction costs in general. 16 Communication, ‘Unleashing’ (n. 12), 4–5. 17 See Armin Schwienbacher and Benjamin Larralde, ‘Crowdfunding of Small Entrepreneurial Ventures’ in Douglas Cumming (ed.) Handbook of Entrepreneurial Finance (Oxford University Press, 2012), pp. 5, 7, and 17, available at ; Ajay K. Agrawal, Christian Catalini, and Avi Goldfarb, ‘Some Simple Economics of Crowdfunding’ (2013) NBER Working Paper No. 19133, 12–13, available at ; Kathryn Judge, ‘The

Future of Direct Finance: The Diverging Paths of Peer-to-Peer Lending and Kickstarter’, (18 September 2015) Wake Forest Law Review 50 (forthcoming), Columbia Law and Economics Working Paper No. 520, 129–30, available at . 18 See Judge (n. 17), 103, 105, 119, 135. 19 Vismara reports that 73 per cent of his sample (which includes 111 offerings listed on CrowdCube) also offered rewards alongside equity participations (Silvio Vismara, ‘Information Cascades Among Investors in Equity Crowdfunding’, 3 April 2015, 31, available at ). This appears to be a widespread and growing model: see Magdalena Cholakova and Bart Clarysse, ‘Does the Possibility to Make Equity Investments in Crowdfunding Projects Crowd Out Reward-Based Investments?’ (2015) Entrepreneurship Theory and Practice 39(1), 145, 160 (analysing the Dutch Symbid and referring also to Sellaband, Buzzbnk, and BankToTheFuture). 20 Vismara (n. 19), 24; Cholakova and Clarysse (n. 19). 21 Communication, ‘Unleashing’ (n. 12), 5. See also Schwienbacher and Larralde (n. 17), 16; Judge (n. 17), 105. 22 See Schwienbacher and Larralde (n. 17), 16, 17; Judge (n. 17), 127, 128. 23 See Adair Morse, ‘Peer-to-Peer Crowdfunding: Information and the Potential for Disruption in Consumer Lending’, (2015), NBER Working Paper 20899, 5, available at . 24 See European Commission, ‘Consultation document: Crowdfunding in the EU— Exploring the added value of potential EU action’, (3 October 2013), 7, available at ; Kirby and Worner (n. 2). 25 Communication, ‘Unleashing’ (n. 12), 5; see also Ferrarini and Ottolia (n. 15), 382, highlighting the use of crowdfunding portals for the voluntary disclosure of e.g. rankings of VCs and angels investing in the issuer, online systems providing independent IPRs and business plan evaluation systems, in order to help investors to analyse the quality of innovation. 26 See Kirby and Worner (n. 2). 27 See (n. 2). 28 Kirby and Worner (n. 2), 33 et seq. 29 Kirby and Worner (n. 2), 36. In the first half of 2014 only 2.2 per cent of the total investment in equity in the UK was attributed to equity crowdfunding but the figure rises to 32 per cent when considering only seed deals: Beauhurst, ‘British Business Bank and Department for Business, Innovation & Skills, Equity Crowdfunding in the UK: Evidence from the Equity Tracker’, (2015), 9–12, available at . Moreover, crowdfunded equity has been growing at a faster rate (at 410% between 2012 and 2014) than traditional equity: Beauhurst, ‘British Business Bank’, ibid., and Peter Baeck, Liam Collins, and Bryan Zhang, ‘Understanding Alternative Finance. The UK

Alternative Finance Industry Report 2014’, (November 2014), 13, available at . 30 Kirby and Worner (n. 2), 36. 31 Kirby and Worner (n. 2), 43. 32 See Nav Athwal, ‘The Disappearance of Peer-To-Peer Lending’, Forbes, 14 October 2014, available at (accessed 8 October 2015); Cortese, ‘Loans that Avoid Banks? Maybe Not’, New York Times, 3 May 2014, available at (accessed 8 October 2015). 33 See John Gapper, ‘The Lenders of the Revolution Look Familiar’, Financial Times, 18 June 2015. 34 Gapper (n. 33), 9; Judge (n. 17), 111. 35 Judith Evans, ‘Equity Crowdfunding Thrives Despite High Risks’, Financial Times, 7 December 2014, . 36 Kirby and Worner (n. 59), 36–7. 37 See Giuliana Borello et al., ‘The Funding Gap and The Role of Financial Return Crowdfunding: Some Evidence From European Platforms’ (2015) Journal of Internet Banking and Commerce 20(1), 1, 16: ‘Our survey reveals that very few platforms (three out of 21) created a secondary market for their investors to sell shares previously bought on that same equity crowdfunding portal’ (in the lending segment the proportion is seven to twenty-four: ibid., 13). However, secondary markets are expected to develop sooner or later: David M. Freedman and Matthew R. Nutting, ‘The Growth of Equity Crowdfunding: Crowdfinance Options for Private Companies, and Secondary Markets for Investors, will Keep Expanding’, The Value Examiner, July/August 2015, available at

(forecasting a growth in secondary markets for crowdfunding shares). 38 Kirby and Worner (n. 59), 36–7, where the authors specify that the liquidity risk will depend on the type of investors involved: if the investors in equity crowd-funding are inexperienced and unaware, they might overreact in times of stressed market conditions or difficult personal circumstances. This raises concerns about investor protection and has led to many jurisdictions placing limits on who can invest in such equity. In some cases platforms are only allowed to market to sophisticated investors, and/or are limited to the number of individuals such investments can be marketed to. 39

Vismara (n. 19), 18) reports that among the observed 111 offerings on the British platform CrowdCube, only 38.7 per cent reached or exceeded the target amount. The average number of investors per project was sixty-nine.

40

See Niamh Moloney, How to Protect Investors: Lessons from the EC and the UK (Cambridge: Cambridge University Press, 2010), p. 71. 41 Kirby and Worner (n. 2), 41. 42 With reference to the UK equity-based crowdfunding market, see Baeck et al. (n. 29), 52: ‘From the investor side, the average investment portfolio size is £5,414 with an average diversification rate of 2.48 (i.e. on average, one investor has invested in 2.48 equity– crowdfunding deals).’ 43 Seedrs in 2014 reported an average number of investors per project of 165: . As regards CrowdCube, as of December 2015, the average number of investors per pitch was 189: see . 44 Kirby and Worner (n. 2), 41. 45 FCA, ‘A Review of the Regulatory Regime for Crowdfunding and the Promotion of Non-readily Realisable Securities by Other Media’, February 2015, 8. 46 FCA (n. 45), 44. 47 FCA (n. 45), 5 (quoting Baeck et al. (n. 29)). See also Karen E. Wilson and Marco Testoni, ‘Improving the Role of Equity Crowdfunding in Europe’s Capital Markets’, (2014), Bruegel policy contribution 2014/09, 6,

(describing crowdinvestors as generally inexperienced). Contra Lars Hornuf and Armin Schwienbacher, ‘Should Securities Regulation Promote Crowdinvesting?’, 11 June 2015, 29, (describing the typical user of equity platforms as generally equipped with experience in investments and business, especially where platforms set high tickets size, de facto excluding small retail investors). 48 See Ronald J. Gilson, ‘Engineering a Venture Capital Market: Lessons from the American Experience’ (2003) Stan. L. Rev. 55, 1067, 1076–7; Ajay K. Agrawal, Cristina Catalini, and Avi Goldfarb, ‘The Geography of Crowdfunding’ (2011) NBER Working Paper No. w16820, available at . With special reference to innovative start-ups, see John Armour and Luca Enriques, ‘Financing Disruption’, Preliminary Draft (June 2015), available at

(accessed December 2015). 49 See Brad M. Barber and Terrance Odean, ‘The Internet and the Investor’ (2001) The Journal of Economic Perspectives 15(1), 41; Barber and Odean, ‘Online Investors: Do the Slow Die First?’ (2002) Review of Financial Studies 15(2), 455; Barber and Odean, ‘Does Online Trading Change Investor Behavior?’ (2002) European Business Organization Law Review 3, 83. 50 The results range from the evidence of crowd-lenders’ ability to analyse soft information and other funders’ decisions investing wisely (deducting from soft information and other elements when early investments signal high quality of borrowers) to signs of influence of irrational factors/prejudices and herding bias as well as prevalent use of automatic-investment mechanisms with little time spent on investment research: about such

topics and for other references, see Macchiavello (n. 10). An interesting stream of studies attests the relevance of social networks backed by financial commitments (in the form of early financing) in determining a campaign success as well as in reducing information asymmetries: see Sven C. Berger and Fabian Gleisner, ‘Emergence of Financial Intermediaries in Electronic Markets: The Case of Online P2P Lending’ (2009) BuR Business Research Journal 2(1), available at ; Martina E. Greiner and Hui Wang, ‘The Role of Social Capital in People-to-People Marketplaces’, ICIS 2009 Proceedings Paper 29, available at ; Andrea Ordanini et al., ‘Crowd-Funding: Transforming Customers into Investors through Innovative Service Platforms’ (2011) Journal of Service Management 22(4), 443; Yong Lu et al., ‘Social Influence and Defaults in Peer-to-Peer Lending Networks’ (2012) Conference on Information Systems and Technology (CIST 2012) White Paper, available at ; Mingfeng Lin et al., ‘Judging Borrowers by the Company They Keep: Friendship Networks and Information Asymmetry in Online Peer-to-Peer Lending’ (2013) Management Science 59(1), 17 (2013); Ethan Mollick, ‘The Dynamics of Crowdfunding: An Exploratory Study’ (2014) Journal of Business Venturing 29(1), 1; Massimo G. Colombo, Chiara Franzoni, and Cristina Rossi-Lamastra, ‘Internal Social Capital and the Attraction of Early Contributions in Crowdfunding’ (2015) Entrepreneurship Theory and Practice 39(1) (Special Issue: Seeding Entrepreneurship with Microfinance), 75. 51 See Agrawal et al. (n. 48); Lars Hornuf and Armin Schwienbacher, ‘Funding Dynamics in Crowdinvesting’, (September 2015), available at ; Gerrit K.C. Ahlers et al. ‘Signaling in Equity Crowdfunding’ (2015) Entrepreneurship: Theory and Practice 39(4), 955 (studying the success of signalling in 104 equity crowdfunding offerings on the largest Australian platform, identifying as effective signals intellectual capital, retaining equity, and disclosure of risks); Vismara (n. 19) (studying the factors—such as social capital and early investments, investments by public profile investors—influencing the success of 111 equity offerings listed on the British platform CrowdCube, measured based on the number of investors, funding amount, and reaching or even exceeding the target amount). 52 An interesting study about mobile apps shows that the herding behaviour of crowdinvestors on such platforms resulted as rational since sophisticated investors invested in firms that turned out to be successful: see Keongtae Kim and Siva Viswanathan, ‘The Experts in the Crowd: The Role of Reputable Investors in a Crowdfunding Market’ (2016) TPRC 41: The 41st Research Conference on Communication, Information and Internet Policy, available at . 53 See Ajay Agrawal, Christian Catalini, and Avi Goldfarb, ‘Are Syndicates the Killer App of Equity Crowdfunding?’, (25 February 2015) MIT Sloan Research Paper No. 512615 and Rotman School of Management Working Paper No. 2569988, available at . 54 Wilson and Testoni (n. 47), 9. In response to such diversification in the nature of investors, some platforms have provided distinct types of contracts available to different

types of users (boilerplate in case of retails and customized in case of VCs): see Lars Hornuf and Armin Schwienbacher, ‘Crowdinvesting: Angel Investing for the Masses? (8 October 2014) Handbook of Research on Venture Capital: Volume 3. Business Angels, (forthcoming), 15, available at . 55 See European Commission (n. 24), 6–8; Hornuf and Schwienbacher (n. 54); Robert Steinhoff, ‘The Next British Invasion is Securities Crowdfunding: How Issuing NonRegistered Securities Through the Crowd Can Succeed in the United States’ (30 November 2014) University of Colorado Law Review (forthcoming), 705, available at ; Agrawal et al. (n. 17), 16; Ferrarini and Ottolia (n. 15), 368. 56 About the risks of such clauses in the venture capital market, see Jeffrey M. Leavitt, ‘Burned Angels: The Coming Wave of Minority Shareholder Oppression Claims in Venture Capital Start-up Companies’ (2005) North Carolina Journal of Law & Technology 6(2), 223. For a brief reference to the crowdfunding sector, see Hornuf and Schwienbacher (n. 54), 17. 57 Hornuf and Schwienbacher (n. 54), 17; European Commission (n. 24), 8; Steinhoff (n. 55), 707; Agrawal et al. (n. 17), 17–18; Wilson and Testoni (n. 47), 8. 58 Communication, ‘Unleashing’ (n. 12), 6–7, citing the various EU Directives in the areas of prospectuses, payment services, markets in financial instruments, capital requirements, alternative investments, consumer credit, distance marketing, etc., and emphasizing that at national level different additional rules may apply. See also Macchiavello (n. 10). 59 Communication, ‘Unleashing’ (n. 12), 7. 60 ESMA (n. 3), 14. See also Macchiavello, ‘Peer-to-Peer, (n. 10). 61 ESMA (n. 3), 14. 62 See Jörg Begner, ‘Crowdfunding and Regulatory Laws’ (March 2012) BaFin Quartely, 10; V. Müller-Schmale, ‘Crowdfunding: Supervisory Requirements and Investor Responsibility’ (16 June 2014) BaFin Annual Report—Federal Financial Supervisory Authority (Bundesanstalt für Finanzdienstleistungsaufsicht—BaFin), 65–6, available at ; Tanja Aschenbeck-Florange and Thomas Nagel, ‘Germany’ in Oliver Gajda et al. (eds) Review of Crowdfunding Regulation. Interpretations of Existing Crowdfunding Regulations in Europe, North America and Israel (European Crowdfunding Network, 2014), pp. 106 et seq., available at ; Lars Klöhn, Lars Hornuf, and Tobias Schilling, ‘The Regulation of Crowdfunding in the German Small Investor Protection Act: Content, Consequences, Critique, Suggestions’, (forthcoming 2015) Diritto della Banca e del Mercato Finanziario, 5 et seq., available at . 63 Kirby and Worner (n. 2). 64 Some early stage investment platforms restrain their activity to the listing of investment opportunities concerning start-ups: see e.g. Gust () and

AngelList (). To the extent that these platforms offer mere information services, US broker–dealer regulation does not seem to apply. Some US crowdfunding platforms do not offer brokerage services directly to investors, but partner with broker– dealers: e.g. EquityNet (), and CircleUp () which employs a wholly owned subsidiary registered as broker– dealer. 65 ESMA (n. 3), 16, mentioning that Article 4(1)5 of MiFID clarifies that the concept of order is relevant both in primary and secondary markets. In ESMA’s opinion, the service/activity of execution of order on behalf of clients is generally excluded from the crowdfunding business model. 66 ESMA (n. 3), 16: ‘In many cases the agreement could be effectively concluded by the platform on behalf of the investor, and in this case the platform would be carrying out execution of orders on behalf of clients. However, there could also potentially be models where the deal was concluded elsewhere without involvement from the platform’. 67 ESMA (n. 3), 17. 68 ESMA (n. 3), 17. 69 See AMF and ACP, ‘Crowdfunding: a guide for funding platforms and project owners’, (14 May 2013), 9, available at . 70 ESMA (n. 3), 17. 71 ESMA (n. 3), 17. 72 See ACPR, ‘Position de l’ACPR relative au placement non garanti et au financement participative’, 2014-P-08, (30 September 2014), available at ; AMF, ‘Position AMF: Placement non garanti et financement participative’, DOC-2014-10, (30 September 2014), available at . As regards Italy, see Assoreti, ‘Linee guida per la relazione di servizio con il cliente’ (2009) [‘Guidelines on investment services approved by Consob’], available at ; Consob Communication No. DAL/97006042 of 9 July 1997. 73 See Kirby and Worner (n. 2), 29 et seq. For a recent overview of regulatory approaches to crowdfunding, see IOSCO, ‘Crowdfunding 2015: Survey Responses Report’, (December 2015), . During the publication process of this chapter, the European Commission released a Staff Working Paper about crowdfunding in the EU, supporting the regulatory trends here highlighted: European Commission, ‘Crowdfunding in the EU Capital Markets Union’ (3 May 2016) Commission

Staff Working Document SWD(2016) 154 final, available at . For an example of light-touch regulation based on existing general securities law, see the Netherlands: Ronald Kleverlaan, ‘A View from the Field: The Netherlands’ in Robert Wardrop et al. (eds) Moving Mainstream: The European Alternative Finance Benchmarking Report, University of Cambridge & EY, (30 February 2015), available at : There is no specific crowdfunding regulation in the Netherlands. At the moment, 30 companies have a licence or exemption to offer financial products through online platforms based on existing financial regulations. For investors, it is not permitted to either invest in more than 100 projects, invest more than €20,000 in equity through an online platform, or invest more than € 0,000 in debt. For projects raising in excess of €2.5m a prospectus is required. 74

See Baeck et al. (n. 29), 52. See Wardrop et al. (n. 73), 27, 28, and 33. 76 The alternative finance sector in Italy amounted to a total of €8.2 million, ranking tenth in Europe, but seventeenth when considering per capita data. The corresponding figures for the other countries considered with reference to the entire alternative finance market are: €2,337 million in the UK, €140 million in Germany, €154 million in France, and €62 million in Spain. (Wardrop et al. (n. 73), 16.) 77 See data provided by the ‘Osservatorio sul crowdfunding del Politecnico di Milano’: (accessed 11 January 2016). 78 See FCA, ‘The FCA’s regulatory approach to crowdfunding and similar activities’, CP13/13, (October 2013); FCA, ‘The FCA’s regulatory approach to crowdfunding over the internet, and the promotion of non-readily realisable securities by other media. Feedback to CP13/13 and final rules’, PS14/4, (March 2014). 79 See David Blair and Aisling Pringleton, ‘United Kingdom’ in Gajda et al. (eds) (n. 62), 224–5; Tanja Aschenbeck-Florange et al., ‘Regulation of Crowdfunding in Germany, the UK, Spain and Italy and the Impact of the European Single Market’, European Crowdfunding Network (June 2013), 18, available at . 80 See FCA, ‘The Perimeter Guidance Manual, Chapter 13: Guidance on the scope of MiFID and CRD IV’ (November 2015), available at . In particular, in table 1, A1: ‘The activity of arranging (bringing about) deals in investments is wider than A1, so a firm carrying on this regulated activity will not always be receiving and transmitting orders.’ 75

81

See FCA Questions and answers about investment services and activities, Q13 about PERG 13.3 , ‘If you are party to a transaction as agent for your client or commit your client to it, you may be doing more than receiving and transmitting orders and will need to consider whether you are providing the investment service of executing orders on behalf of clients.’ And Q15: You will be providing this investment service if you participate in the execution of an order on behalf of a client, as opposed simply to arranging the relevant deal. In our view, you can execute orders on behalf of clients either when dealing in investments as agent (by entering into an agreement in the name of your client or in your own name, but on behalf of your client) or, in some cases, by dealing in investments as principal (for example by back-to-back or riskless principal trading). 82

See FCA (n. 81) Q 13 about PERG 13.3:

The extended meaning of the service only applies if the firm brings together two or more investors and a person issuing new securities, including a collective investment undertaking, should not be considered to be an ‘investor’ for this purpose. This limitation does not apply though to the general definition of the service. Accordingly whilst an arrangement whereby a person, on behalf of a client, receives and transmits an order to an issuer will, in our view, amount to reception and transmission, one in which it simply brings together an issuer with a potential source of funding for investment in a company, will not. 83

See Blair and Pringleton (n. 79), 225. At least with reference to Client money rules (CASS), the FCA states (FCA, ‘Feedback to CP13/13 and final rules’, PS14/4 (n. 78)): ‘Loan-based crowdfunding is a type of investment activity, so should be subject to the rules for investment business’. Anyway, because of different risks involved, FCA justifies (with particular reference to promotion and marketing rules but also under other aspects) a different approach towards investment-based crowdfunding, on one hand, and loan-based crowdfunding, on the other (ibid., 39): 84

we see sufficient similarities between the equity and debt securities issued by companies, and sufficient distinctions between debt securities and article 36H loan agreements, to justify the different approaches proposed. […] So at present, in the P2P loan market, we consider it reasonable to assume a lower risk of significant capital losses, and less need for consumer protection measures. 85

See FCA, ‘The FCA’s regulatory approach’, CP13/13, (n. 78), 6; ‘FCA Handbook’, COBS 2.2.-1R, 14.3.1R and 14.3.7A.G, glossary terms; SYSC from 4.1.8A to 4.1.8E; IPRU(INV) 12.1–12.3.

86

On the new Law, see Fernando Zunzunegui, ‘Régimen jurídico de las plataformas de financiación participativa (crowdfunding)’ (2015) Revista de Derecho del Mercado Financiero, Working Paper 3/2015, available at . 87 Ordinance no. 2014-559 of 30 May 2014 and implementing decree no. 2014-1053 of 16 September 2014. 88 IFPs are registered in the national register of financial intermediaries and lack a European passport. Their activity is limited in scope to crowd-lending, but can also be exercised by other financial intermediaries (such as banks, investment firms, insurance companies, etc.) and by CIPs. 89 Delibera Consob no. 18592/2013, ‘Regolamento sulla raccolta di capitali di rischio da parte di start-up innovative tramite portali on-line’, (26 June 2013). 90 Delibera Consob no. 19520/2016, ‘Modifiche al “Regolamento sulla raccolta di capitali di rischio da parte di start-up innovative tramite portali on-line”, adottato con delibera n. 18592 del 26 giugno 2013’, (24 February 2016); Consob, ‘Relazione illustrativa delle conseguenze sull’attività delle imprese e degli operatori e sugli interessi degli investitori e dei risparmiatori, derivanti dalle modifiche al regolamento sulla raccolta di capitali di rischio tramite portali on-line’, (25 February 2016). 91 The differences between a CIP and a PSI are substantial. A CIP can only deal in ordinary shares and fixed-rate bonds and cannot offer its services outside France in the EEA. In addition, it cannot hold instruments or money of its clients and is not subject to special capital requirements. A PSI can deal in all kinds of instruments and offer its services under a passport in all EEA countries. It is allowed to hold money and instruments of clients and is subject to capital requirements and other prudential requirements. Both CIPs and PSIs are subject to organization requirements and rules of conduct stated in the general Regulation of the Autorité de Marchés Financiers (AMF, the French securities supervisor). 92 See Müller-Schmale (n. 62); J.P. Bußalb, ‘Subordinated loans and loans with profit participation: BaFin urges caution in connection with capital raising’ (16 March 2015), BaFin, ; Begner (n. 62), 8 et seq. 93 See Bußalb (n. 92): Under a profit participation loan (partiarisches Darlehen), investors transfer capital to the offeror (borrowing entity) for a specific purpose and receive an interest in profits in return. This is the profit-based component of the loan, which depends on the amount of profits generated. Many of these loan agreements also offer interest on the loan amount. 94

Aschenbeck-Florange and Nagel (n. 62), 104. For a comment, see Klöhn et al. (n. 62). 96 Begner (n. 62), 10 defines such services as follows: 95

Investment broking is defined in the Banking Act (KWG) as ‘the brokering of business involving the purchase and sale of financial instruments’. Investment broking is provided by anyone who as agent of the investor transmits the investor’s declaration of intent to buy or sell financial instruments to the entity with whom the investor wishes to conclude the transaction. The brokerage can also be carried out electronically: anyone making an IT system available by which declarations of intent may be transmitted to potential contractual partners is also providing investment broking. This also applies to operators of crowdfunding platforms. If the platform operator acts not only as agent but also as authorised representative, this may constitute contract broking. This is defined by statute as the purchase and sale of financial instruments in the name of and for the account of others (section 1 (1a) sentence 2 no. 2 of the KWG). The platform operator is an authorized representative if it is authorised to accept the investor’s declaration of intent to buy a capital investment. One special form of contract broking is placement business. This is provided by anyone who sells financial instruments upon emission—when they are first issued—on the capital markets or to a limited circle of investors on behalf of and for the account of third parties. The issuer must then have appointed the underwriter to place the financial instruments on the capital markets (placement agreement) without the underwriter being committed to buy the financial instruments. 97

See Peter Mayer and Robert Michels, ‘Changes for Crowdfinancing resulting from the German Small Investor Protection Act (Kleinanlegerschutzgesetz)’, (September 2015), (accessed November 2015). 98 See Mayer and Michels (n. 97); Müller-Schmale (n. 62), 3. 99 The duty to inform investors implies that there is also a duty of the platform to get informed about the offered investment: see Régis Vabres, ‘Les statuts et les obligations des plates-formes de crowdfunding equity’ (2015) in Anne-Valérie Le Fur (ed.) Le cadre juridique du crowdfunding. Analyses prospectives (Paris: Société de législation comparée, 2015), pp. 51–168, 163. 100 A Consob communication clarified that banks and investment firms managing an equity crowdfunding portal are subject to general investment services regulation but need to adapt to the specific crowdfunding disclosure obligations as well as to conditions set in Article 100-ter TUF and Articles 24–25 of Consob crowdfunding regulation (concerning emission limits for prospectus exemption, presence of certain clauses in the issuer’s articles of incorporation, and respect of certain disclosure obligations, professional investors’ mandatory participation, investors withdrawal right): see Consob, ‘Comunicazione n. 0066128 dell’1.08.2013 riguardante lo svolgimento da parte di banche ed imprese di investimento dell’attività di gestione di portali on line per la raccolta di capitali per le start up innovative’.

101

See FCA, ‘Feedback to CP13/13 and final rules’, PS14/4 (n. 78), 7; ‘FCA Handbook’, from COBS 4.7.7R to COBS 4.7.10R. 102 About MiFID II/MiFIR, see Niamh Moloney, EU Securities and Financial Markets Regulation (3rd edn, Oxford: Oxford University Press, 2014), pp. 336 et seq. 103 Moloney (n. 102), 339–40. 104 See Recital 89, which states: One of the objectives of this Directive is to protect investors. Measures to protect investors should be adapted to the particularities of each category of investors (retail, professional and counterparties). However, in order to enhance the regulatory framework applicable to the provision of services irrespective of the categories of clients concerned, it is appropriate to make it clear that principles to act honestly, fairly and professionally and the obligation to be fair, clear and not misleading apply to the relationship with any clients. See also Article 30(1) second indent, which states: ‘Member States shall ensure that, in their relationship with eligible counterparties, investment firms act honestly, fairly and professionally and communicate in a way which is fair, clear and not misleading, taking into account the nature of the eligible counterparty and of its business.’ 105 See Article 24(4) and (5) MiFID II. 106 See Article 39 on market monitoring by ESMA; Article 40 on ESMA temporary intervention powers; and Article 42 on product intervention by competent authorities. Product intervention means that either ESMA or a competent authority may, under certain conditions, prohibit or restrict the marketing, distribution, or sale of certain financial instruments or a type of financial activity or practice. 107 See Article 16(3) MiFID II. 108 Therefore, the following obligations provided for by Article 16(3), second and third indents should presumably not apply to crowdfunding firms: An investment firm which manufactures financial instruments for sale to clients shall maintain, operate and review a process for the approval of each financial instrument and significant adaptations of existing financial instruments before it is marketed or distributed to clients. The product approval process shall specify an identified target market of end clients within the relevant category of clients for each financial instrument and shall ensure that all relevant risks to such identified target market are assessed and that the intended distribution strategy is consistent with the identified target market. 109

Therefore, the following obligations provided for by Article 16(3), fourth indent, could apply to crowdfunding firms: An investment firm shall also regularly review financial instruments it offers or markets, taking into account any event that could materially affect the potential risk to

the identified target market, to assess at least whether the financial instrument remains consistent with the needs of the identified target market and whether the intended distribution strategy remains appropriate. 110

See Article 25(4) MiFID II. The current text of Article 3 MiFID does not require that the exempted persons are authorized and regulated at national level, while including the other conditions mentioned above. 112 See Article 3(2) MiFID II, the provisions of which are not found in Article 3 MiFID. 113 See Article 3(2) second indent MiFID II, which is also not found in Article 3 MiFID. 114 Article 4(1)(2) MiFID II. 115 Article 4(1)(44) MiFID II further specifies that the following are transferable securities: 111

(a) shares in companies and other securities equivalent to shares in companies, partnerships or other entities, and depositary receipt in respect of shares; (b) bonds and other forms of securitised debt, including depositary receipt in respect of such securities; (c) any other securities giving the right to acquire or sell any such transferable securities or giving rise to a cash settlement determined by reference to transferable securities, currencies, interest rates or yield, commodities or other indices or measures. 116

See Article 4(1)(4) MiFID II defining it as ‘the provision of personal recommendations to a client, either upon its request or at the initiative of the investment firm, in respect of one or more transactions relating to financial instruments’. 117 See Section III.2. 118 The answer to the third question is no doubt influenced by the position taken by ESMA in its ‘Technical Advice to the Commission on MiFID II and MiFIR’, ESMA/2014/1569, (19 December 2014), 14. 119 See Article 29(1) MiFID II: ‘Member States shall allow an investment firm to appoint tied agents […]’ (emphasis added; the rest of the provision is identical to Article 23(1) MiFID I quoted above in the text). Under Article 4(1)(29) MiFID II: ‘tied agent’ means a natural or legal person who, under the full and unconditional responsibility of only one investment firm on whose behalf it acts, promotes investment and/or ancillary services to clients or prospective clients, receives and transmits instructions or orders from the client in respect of investment services or financial instruments, places financial instruments or provides advice to clients or prospective clients in respect of those financial instruments or services. 120

See Article 29(2) MiFID II.

121

Article 4(1) No. 23 MiFID II: ‘“organised trading facility” or “OTF” means a multilateral system which is not a regulated market or an MTF and in which multiple thirdparty buying and selling interests in bonds, structured finance products, emission allowances or derivatives are able to interact in the system in a way that results in a contract in accordance with Title II of this Directive’. 122 See ESMA (n. 3), 18. Contra Zunzunegui (n. 86), 15 (proposing in case the qualification as SME growth market under Article 33 MiFID II). 123 European Commission, ‘Unleashing’, (n. 12), 2. 124 See

and . 127 See Moloney (n. 102), 396 et seq. 125 See Article 34 MiFID II on the freedom to provide investment services and activities and Article 35 on the establishment of a branch. 126 Article 35(8) MiFID II makes reference to Articles 24 (general principles for investor protection and information to clients), 25 (assessment of suitability and appropriateness and reporting to clients), 27 (obligation to execute orders on terms most favourable to the client), and 28 (client order handling rules); and to MiFIR’s Articles 14 (obligation for systematic internalizers to make public firm quotes) and 26 (obligation to report transactions). Of these provisions, mainly the first two (general principles and assessment of appropriateness) are relevant for crowdfunding. 128 See Cassis de Dijon—Rewe Zentral AG v Bundesmonopolverwaltung Fuer Branntwein [1979] case 120/78 [1979] ECR 649; Reinhard Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano [1995] ECR 1465; Commission of the European Communities v Federal Republic of Germany—case 205/84, [1986] ECR 3775. Specifically in the financial sector, Alpine Investments v Minister Van Financier [1995] ECR I-1141; Commission of the European Union v Italy [1996] ECR I-2691; CaixaBank France v Ministère de l’Économie, des Finances et de l’Industrie—Case C-442/02, 1 CMLR 39. See also Moloney (n. 102), 8–10; George Walker and Robert Purves, Financial Services Law (3rd edn, Oxford: Oxford University Press, 2014), pp. 87–9. 129 Article 47: Ámbito de aplicación territorial: (1) Estarán sujetas a lo previsto en este título las plataformas de financiación participativa que ejerzan la actividad prevista en el artículo anterior en territorio nacional, así como la participación en ellas de los inversores y promotores. (2) A los efectos de lo previsto en este título, no se considerará que un servicio ha tenido lugar en territorio nacional cuando un residente en España participe por iniciativa propia, como inversor o promotor, en una plataforma con domicilio social en el extranjero que preste los servicios previstos en el artículo 46.1 de esta Ley. (3) A los efectos previstos en el apartado anterior, no se considerará que la actividad se pone en marcha a iniciativa propia: (a) Cuando la empresa anuncie, promocione o capte clientes o posibles clientes en España. (b)

Cuando la empresa dirija sus servicios específicamente a inversores y promotores residentes en territorio español. 130

See Moloney (n. 102), 397–8 (note). An author denies the availability of the ‘general good test’ in the whole financial sector because of the protective nature of MiFID: see Michel Tison, ‘The General Good Exception: The Case of Financial Services’ in Mads Andenas and Wulf-Henning Roth (eds) Services and Free Movement in EU law (The Hague: Kluwer, 2002), p. 305. 131 See Agrawal et al. (n. 17), 38.

INDEX

administrative sanctions power to impose 19.40–19.74 Article 70 MiFID II 19.47–19.53 exercise 19.54–19.55 general provisions 19.40–19.53 publication requirements 19.67–19.68 unpublished decisions 19.69–19.70 see also competent authorities advice, see independent financial advice agency and principal dealing 9.01–9.55 acting as agent or principal 9.23–9.54 arbitrariness 9.26–9.29 civil duty of care 9.51 dealing on own account 9.51 general 9.23 Grant Estates Ltd v Royal Bank of Scotland 9.30–9.45 investors’ reasonable expectations 9.24–9.25 Kay Review 9.52–9.54 manipulation 9.26–9.29 MiFID, lack of practicable distinctions 9.46–9.50 tenuousness 9.26–9.29 context 9.01–9.03, 9.55 investor protection 9.19–9.22 transaction types 9.04–9.18 dealing solely on own account 9.08 execution of client orders 9.05–9.18 general 9.04 see also execution of client orders agricultural commodity derivatives 19.97–19.98 see also cooperation arrangements algorithmic trading 17.22–17.41 context 17.01–17.21, 17.61–17.62

investment firms 17.23–17.32 direct electronic access (DEA) 17.29–17.32 general clearing 17.29–17.32 requirements 17.23–17.28 trading venues 17.33–17.40 organizational requirements 17.34–17.38 trading requirements 17.39–17.40 see also high frequency trading (HFT) Alternative Investment Fund Managers Directive (AIFMD) third-country investment firms 1.24 ancillary activity exemption 16.56–16.62 see also commodity derivatives anonymity, see equity trading assets position protection 15.44–15.45 see also clearing obligation at-arms-length services 4.43–4.49 see also best interest duty auditors 19.24–19.26 see also competent authorities authorization cooperation arrangements 19.112–19.115 banking, see shadow banking BATS Group 11.55–11.57 see also financial market infrastructure (FMI) groups Belgium trading venue competition 1.06 best interest duty 4.01–4.94 antecedents 4.04–4.06 assessment 4.83–4.94 basic provisions 4.01–4.03 functions and contents 4.16–4.32 individual services and activities 4.33–4.73 at-arms-length services 4.43–4.49 dealing on own account 4.71–4.73 investment advice 4.35–4.42 manufacturing 4.50–4.52 self-placement 4.53–4.70 MiFID II 4.07–4.15 US comparative policy 4.74–4.82 binding mediation exemptions 19.109–19.111

see also cooperation arrangements Board governance, see governance Bolsas y Mercados Españoles Group 11.49–11.52 see also financial market infrastructure (FMI) groups broker consortia dark pools 14.23 see also dark trading broker/dealer crossing networks (BDCN) 14.19, 14.22 see also dark trading capital requirements directive (CRD) II 3.14–3.19 capital requirements regulation (CRR) investment firms 2.16–2.25 caps 7.85–7.86 see also conflicts of interest central clearing, see clearing obligation central counterparty (CCP) cleared derivatives: acceptance timeframe 15.61, 15.68–15.69 submission timeframe 15.60, 15.64–15.65 trades not accepted 15.71–15.73 non-discriminatory access to 11.30–11.31 protective measures 15.39 third-country investment firms 1.24 see also clearing obligation central limit order book model 15.11 see also trading obligation central securities depositories regulation (CSDR) shadow banking 22.87–22.99 civil courts less strict application than MiFID 20.03–20.04 stricter application than MiFID 20.18–20.20 see also Nationale-Nederlanden v. Van Leeuwen; private law effect cleared derivatives, see straight-through-processing (STP) obligation clearing broker (CB) cleared derivatives: acceptance timeframe 15.66–15.67 clearing obligation 15.43 clearing obligation 15.32–15.48 central clearing: CCP protective measures 15.39 definition 15.35–15.39 multilateral netting 15.38

novation 15.37 general obligation 15.32–15.48 miscellaneous 15.41–15.48 assets and position protection 15.44–15.45 CB position 15.43 central clearing access 15.41 default management 15.47–15.48 direct and indirect clearing 15.42 leapfrog payment 15.47–15.48 netting 15.46 post-default porting 15.47–15.48 segregation at (direct) client clearing 15.43, 15.45 scope 15.40 see also derivatives cognitively optimized disclosure 7.79–7.80 see also conflicts of interest collective investment undertakings 2.55–2.56 commodity derivatives 16.01–16.93 context 16.01–16.07, 16.90–16.93 definition 16.44–16.50 EU provisions 16.17–16.41 intervention powers: authorities and trading venues 16.83–16.89 market features 16.08–16.16 MiFID II/ MiFIR regulation 16.42–16.62 ancillary activity exemption 16.56–16.62 exemption regime changes 16.51–16.55 issues 16.42–16.43 position limits regime 16.63–16.82 calculation and methodology 16.74–16.76 EU v USA approaches 16.77–16.82 financial v non-financial entities 16.66–16.69 hedging exemption 16.70–16.73 overview 16.63–16.65 company law 3.63 see also governance competent authorities 19.11–19.82 administrative sanctions and measures 19.40–19.74 criminal sanctions 19.40–19.74 Article 70 MiFID II 19.47–19.53 exercise of the power 19.54–19.55 general provisions 19.40–19.53 Member State obligations and exemptions 19.40–19.46

general aspects 19.11–19.26 auditors 19.24–19.26 designation of competent authorities 19.11–19.15 Member State cooperation 19.16–19.18 professional secrecy 19.19–19.23 publication of decisions 19.56–19.74 administrative sanctions imposed and published 19.67–19.68 administrative sanctions imposed and unpublished 19.69–19.70 context 19.63–19.64 criminal sanctions imposed 19.71–19.74 general obligation 19.66 general provisions 19.56–19.62 notification and reporting requirements 19.63–19.74 redress procedures 19.75–19.82 consumer complaints 19.80–19.82 extra-judicial mechanisms 19.80–19.82 reporting of infringements 19.75–19.77 right of appeal 19.78–19.79 supervisory powers 19.27–19.39 Article 69 MiFID II 19.30, 19.36 exercise of 19.37–19.39 general provisions 19.27–19.36 Member States’ notification obligation 19.27–19.29 rationale 19.27–19.29 see also cooperation arrangements; supervision and enforcement conduct-of-business (COB) rules governance 3.04–3.20 investment protection and IDD: general conduct 21.08–21.11 insurance-based investment products 21.12–21.14 MiFID II 21.15–21.24 structure 21.07–21.14 conflicts of interest 7.01–7.88 fee regime 7.55–7.60 market structure approach 7.56 MiFID I to MiFID II transition 7.57 MiFID II regime parameters 7.58–7.60 financial market infrastructure (FMI) groups 11.76 organizational requirements 7.32–7.42 company structure solution 7.33–7.35 MiFID I to MiFID II transition 7.36–7.37 MiFID II regime parameters 7.38–7.42 overview 7.87–7.88

regime examples 7.31–7.86 regime foundations 7.01–7.30 goals and consistency 7.08–7.13 topic scope 7.02–7.13 scope of conflicts 7.03–7.07 regulatory approaches 7.25–7.30 contract/disclosure rules 7.26–7.29 heterogeneous approach 7.30 organizational rules 7.26–7.29 strategic regulatory development 7.78–7.86 caps 7.85–7.86 cognitively optimized disclosure 7.79–7.80 debiasing 7.81–7.84 theoretical approaches 7.14–7.24 cognitive limitations 7.21–7.24 efficiency 7.18–7.20 fairness 7.15–7.17 fundamental lack of transparency 7.18–7.20 see also Investment Services Contracts; linked contracts contracting parties less strict application than MiFID 20.21 stricter application than MiFID 20.22–20.27 see also private law effect consumer complaints 19.80–19.82 see also redress procedures contractual limitation 20.36–20.38 see also private law effect contractual obligations independent financial advice 6.37–6.43 see also suitability cooperation arrangements 19.83–19.126 host member states’ competent authorities 19.116–19.126 freedom of establishment 19.116–19.117 precautionary measures 19.118–19.123 third countries 19.124–19.126 Member States’ competent authorities 19.83–19.115 binding mediation and exemptions 19.109–19.111 consultation prior to authorization 19.112–19.115 exchange of information 19.101–19.108 investigations 19.99–19.100 supervisory activities 19.99–19.100 obligation to cooperate 19.83–19.98 agricultural commodity derivatives 19.97–19.98

general provisions 19.83–19.90 greenhouse gas emission allowance markets 19.97–19.98 on-site verification 19.99–19.100 trading venues 19.91–19.96 see also competent authorities; supervision and enforcement counterparty performance portfolio compression obligation 15.115–15.116 credit institutions 2.15 credit rating agencies (CRA) regulation third-country investment firms 1.24 credit risk shadow banking 22.39 criminal sanctions 19.71–19.74 see also competent authorities cross-selling 21.52–21.56 see also investment protection crowdfunding as a different type of service 23.49–23.51 as an investment service 23.47–23.48 investment-based 23.01–23.59 benefits 23.04–23.11 challenges 23.04–23.11 definitions 23.37–23.41 disclosure obligations 23.26–23.28 EU passport and applicable law 23.45–23.51 exemptions 23.37–23.41 financial instrument 23.13–23.14 Great Financial Crisis 23.34–23.36 investment service 23.15–23.17 issues 23.01–23.03, 23.55–23.59 MiFID and scope of application 23.19–23.25 MiFID I 23.12–23.17 MiFID II 23.34–23.54 OTFs 23.42–23.44 policy considerations 23.52–23.54 risks 23.04–23.11 special requirements 23.29–23.33 tied agents 23.42–23.44 national approaches to 23.18–23.33 French approach 23.22–23.23 German approach 23.24–23.25 Italian approach 23.22–23.23 Spanish rejection of MiFID 23.21

UK model 23.20 shadow banking 22.121–22.131 dark pools, see dark trading dark trading 14.01–14.106 academic literature 14.24–14.38 empirical papers 14.32–14.38 theoretical papers 14.25–14.31 dark pool classification 14.16–14.23 broker consortia dark pools 14.23 broker/dealer crossing networks (BDCN) 14.19 direct market access (DMA) 14.20 full service broker dark pool (BDCN) 14.22 non-directed orders 14.21 order flows 14.20 order management systems (OMS) 14.21 OTCs 14.16, 14.19 RMs and MTFs 14.16–14.18 dark pool internalization 18.94–18.105 function 18.96–18.98 ‘ideal’ dark pools and social desirability 18.99–18.100 damaging practices for traders 18.101–18.102 internalization 18.103–18.104 dark pools (MiFID I) 14.39–14.58 venue classification 14.40–14.51 pre-trade transparency 14.40–14.58 waivers 14.52–14.58 dark pool regulation (MiFID II/MiFIR) 14.59–14.100 double volume cap 14.86–14.91 equity instruments 14.67–14.76 investment firms’ trading obligations 14.92–14.93 MiFID-ready solutions 14.94–14.100 regulatory technical standards 14.86–14.91 volume cap mechanism 14.77–14.85 waivers 14.67–14.76 equity trading: price formation 18.106–18.112 history and context 14.01–14.10 future and outlook 14.101–14.106 motivation 14.11–14.15 data reporting service providers 2.57–2.59 services 10.12–10.14

dealing on own account best interest duty 4.71–4.73 Dutch Supreme Court 9.51 solely on own account 9.08 debiasing 7.81–7.84 see also conflicts of interest Denmark trading venue competition 1.06 derivatives 15.01–15.132 clearing obligation 15.32–15.48 definition 15.03–15.05 history and impact 15.01–15.02, 15.128–15.132 indirect clearing obligations 15.76–15.104 portfolio compression obligation 15.105–15.127 in scope, see trading obligation straight-through-processing (STP) obligation 15.49–15.75 trading obligation 15.06–15.31 see also clearing obligation; indirect clearing obligation; portfolio compression obligation; straight-through-processing (STP) obligations; trading obligation Deutsche Börse Group 11.41–11.45 see also financial market infrastructure (FMI) groups direct electronic access (DEA) algorithmic trading and HFT 17.29–17.32 direct market access (DMA) dark pools 14.20 see also dark trading disclosure obligations crowdfunding 23.26–23.28 SME growth markets 13.34–13.46 current event reports 13.42–13.46 key operating milestones 13.35–13.41 distributors product governance 5.09–5.10 investor protection 5.16 monitoring obligation 5.38–5.39 product governance procedures 5.25–5.29 staff 5.31 double volume cap regulatory technical standards 14.86–14.91 duty to act in the best interests of the client, see best interest duty electronic front-running 18.42–18.67 efficiency 18.62–18.67

HFT activities and resource reduction 18.65 resource and risk allocation 18.66–18.67 share price accuracy 18.63–18.64 examples of 18.44–18.49 wealth transfer 18.50–18.61 informed investors 18.53–18.56 narrowing of spread 18.51–18.52 uninformed investors 18.53–18.56 ultimate incidence 18.57–18.61 see also equity trading; high frequency trading (HFT) eligible counterparties 10.18–10.45 equivalence decision by the Commission 10.19–10.29 absence of 10.42–10.45 Article 47 MiFIR 10.28–10.29 general 10.18 equivalence test 10.22–10.23 reciprocity test 10.21 transitional regime 10.37 ESMA registration 10.30–10.34 cross-border basis 10.34 withdrawal of 10.38–10.41 third-country firms’ obligations 10.35–10.36 see also retail investors emission allowances 2.42 enforcement, see public enforcement; supervision and enforcement entities in scope, see trading obligation equity trading 18.01–18.127 dark pools and internalization 18.94–18.105 function 18.96–18.98 ‘ideal’ dark pools and social desirability 18.99–18.100 damaging practices for traders 18.101–18.102 internalization 18.103–18.104 evaluative framework 18.22–18.39 anonymity 18.35–18.39 goals 18.23–18.24 liquidity 18.29–18.30 market characteristics 18.25–18.39 price accuracy 18.26–18.28, 18.36–18.38 resource allocation 18.31–18.32, 18.39 risk allocation 18.34, 18.39 share price accuracy 18.33 fundamentals 18.04–18.39 high-frequency trading (HFT) 18.40–18.93

efficiency 18.81–18.93 electronic front-running 18.42–18.67 examples 18.74–18.76 Flash Crash 18.87–18.90 slow market arbitrage 18.72–18.93 volatility 18.84–18.93 wealth transfer 18.77–18.80, 18.91 informed v uninformed traders 18.13–18.15 informed trading and transaction prices 18.20–18.21 liquidity provision 18.06–18.21 liquidity suppliers 18.16–18.19 accounting perspective 18.17 information perspective 18.18–18.19 MiFID II issues 18.105–18.127 dark pools and effective price formation process 18.106–18.112 high frequency trading (HFT) 18.113–18.117 market abuse 18.118–18.121 price volatility 18.114–18.117 trading venue competition 18.122–18.127 private information: types and sources 18.09–18.12 announcement information 18.11 fundamental value information 18.12 inside information 18.10 United States perspective 18.01–18.03 see also dark trading; electronic front-running; high frequency trading (HFT); liquidity equivalence 10.24–10.27 Article 47 MiFIR 10.28–10.29 Commission decision 10.19–10.29 absence of decision 10.42–10.45 equivalence test 10.22–10.23 no decision in effect 10.64 post-equivalence decision 10.59–10.63 retail investors 10.64 withdrawal of registration 10.38–10.41 see also reciprocity; third-country firms Euronext Group 11.46–11.48 see also financial market infrastructure (FMI) groups European Banking Authority (EBA) product intervention, temporary powers 5.63–5.73 emergency situations 5.73 general 5.63–5.64 powers under MiFIR 5.65–5.72

European Economic Area (EEA) third-country investment firms 1.24 national legislation, MiFID 1.01 European Market Infrastructure Regulation (EMIR) mandatory clearing 22.63–22.74 MiFID II 1.29 OTC derivatives 22.63–22.74 third-country investment firms 1.24 trading venues 1.29 European Securities and Markets Authority (ESMA) inducements 8.08–8.17 exceptions 8.11–8.17 direct fees 8.11 fees enhancing quality of service 8.12–8.14 intra-group payments 8.10 proper fees 8.15–8.17 scope of inducement ban 8.09–8.10 standard fees 8.09 product intervention, temporary powers 5.63–5.73 emergency situations 5.73 general 5.63–5.64 powers under MiFIR 5.65–5.72 third-country firms: cross-border basis 10.34 registration 10.30–10.34 withdrawal of registration 10.38–10.41 European Union (EU) commodity derivatives 16.17–16.41 position limits regime 16.77–16.82 competent authorities: Member State cooperation 19.16–19.18 Member States’ notification obligation 19.27–19.29 Member State obligations and exemptions 19.40–19.46 equivalence test 10.24–10.27 member state option 10.59–10.63 EU passport and applicable law 23.45–23.51 crowdfunding as a different type of service 23.49–23.51 crowdfunding as an investment service 23.47–23.48 no European passport 10.53–10.55 non-exercise of member state option 10.51–10.52, 10.65 financial governance 12.01–12.33 approach 12.14–12.17 bilateral trading requirements 12.26–12.33

MiFIR transparency regime 12.14–12.33 purpose of transparency regulation 12.10–12.13 significance 12.01–12.09 trading venue requirements 12.18–12.25 investment protection 21.01–21.03, 21.57–21.62 national legislation, MiFID 1.01 reciprocity test 10.24–10.27 retail and opt up professional clients 10.47–10.50, 10.51–10.52 securitization measures 22.44–22.62 shadow banking 22.44–22.62 single rulebooks 1.22 STP obligation 15.75 third-country investment firms 1.24 trading venues 11.32–11.57 evolution of market 11.33–11.35 FMI Groups 11.36–11.57 regulatory framework 11.15–11.16 see also competent authorities; financial market infrastructure (FMI) groups; transparency regulation exchange of information 19.101–19.108 see also cooperation arrangements exchange-traded derivatives (ETDs) STP obligation 15.53 trading obligation 15.31 exchange trading, see trading obligation execution of client orders agency crosses 9.15–9.18 internalization 9.09–9.14 MTF 9.05–9.07 RMs 9.05–9.07 systematic internalization 9.09–9.14 see also agency and principal dealing exclusion of liability 20.36–20.38 see also private law effect fairness conflicts of interest 7.15–7.17 private law effect 20.15–20.17 fees, see conflicts of interest financial entities non-financial entities v 16.66–16.69 see also commodity derivatives; position limits regime financial instruments 2.43–2.50

definition 1.05, 1.11 financial market infrastructure (FMI) groups 11.57–11.85 case studies 11.36–11.57 BATS Group 11.55–11.57 Bolsas y Mercados Españoles Group 11.49–11.52 Deutsche Börse Group 11.41–11.45 Euronext Group 11.46–11.48 London Stock Exchange Group 11.37–11.40 NASDAQ OMX Nordic Exchange 11.53–11.54 challenges to regulators 11.69–11.72 consolidated supervision 11.80–11.85 MiFID II approach 11.73–11.79 conflicts of interest 11.76 rules on access 11.79 suitability of shareholders 11.77–11.78 transparency of ownership 11.77–11.78 OTF in FMI Groups 11.65–11.68 pan-European group formation 11.57–11.64 see also trading facilities financial markets, see shadow banking Financial Stability Board (FSB) 22.17–22.26 see also shadow banking Flash Crash 18.87–18.90 France high-frequency trading (HFT) 1.43 investment-based crowdfunding 1.51, 23.22–23.23 trading venue competition 1.06 freedom of establishment 19.116–19.117 see also cooperation arrangements frontloading 15.74 see also straight-through-processing (STP) obligation fundamental lack of transparency conflicts of interest 7.18–7.20 Genil case, see private law effect Germany investment-based crowdfunding 1.51, 23.24–23.25 trading venue competition 1.06 governance 3.01–3.65 Board structure 3.25–3.46 committees 3.35 duties 3.28–3.35 holistic structure 3.28–3.34

organizational structure 3.25–3.27 personal requirements 3.36–3.44 remuneration of members 3.45–3.46 senior management 3.45–3.46 corporate 5.12–5.13 regulated markets 11.18–11.20 European history 3.01–3.03, 3.20 general principles and problems 3.21–3.24 governance reporting 3.58 organization and risk management 3.47–3.55 CRD IV regime requirements 3.47–3.49 MiFID II requirements 3.50–3.55 owners with qualifying holdings 3.56–3.57 policy foundations 3.02 product, see product governance rationale 3.02–3.03, 3.20 regulation, prudential and conduct-of-business 3.04–3.20 case for 3.04–3.06 CRD II package 3.14–3.19 ISD (1993) 3.07–3.13 MiFID I 3.14–3.19 shareholders 3.56–3.57 technical framework (MiFID II) 3.21–3.65 assessment 3.59–3.65 company and partnership law 3.63 effective enforcement 3.64 policy rationales 3.59–3.61, 3.65 public-private dichotomy 3.64 regulatory strategy 3.62 technical inconsistencies 3.63 see also product governance; risk Grant Estates Ltd v Royal Bank of Scotland agency and principal dealing 9.30–9.45 Great Financial Crisis 23.34–23.36 Greece trading venue competition 1.06 greenhouse gas emissions 19.97–19.98 see also cooperation arrangements hedging 16.70–16.73 see also commodity derivatives; position limits regime high frequency trading (HFT) 17.42–17.60 context and overview 17.01–17.21, 17.61–17.62

equity trading 18.40–18.93 electronic front-running 18.65 issues raised by MiFID II 18.113–18.117 price volatility 18.114–18.117 reduction of HFT resources 18.65 slow market arbitrage 18.84–18.93 volatility 18.84–18.93 market abuse regulation 17.51–17.60 high frequency trading (HFT) 18.118–18.121 prevention and identification 17.52–17.56 sanctions 17.57–17.60 specific regulation 17.43–17.50 investment firms 17.44–17.47 regulated markets (RMs) 17.48–17.50 see also algorithmic trading; electronic front-running; equity trading; slow market arbitrage households bias 6.04–6.06 financial market participation 6.03 see also independent financial advice independent financial advice 6.01–6.47 advice 6.07–6.12 conflicted 6.09–6.10 demand side 6.11–6.12 supply side 6.07–6.08 contractual obligations (MiFID II) 6.37–6.43 investment firms’ personnel 6.37–6.38 product governance 6.39–6.42 suitability assessment 6.37–6.43 written statement of suitability 6.43 economic background 6.03–6.06 households 6.03–6.06 bias 6.04–6.06 financial market participation 6.03 information 6.20–6.29 broad analysis 6.25–6.27 independent basis of advice 6.21–6.24 other information 6.29 periodic suitability assessment 6.28 restricted analysis 6.25–6.27 investment services (MiFID II) 6.30–6.36 independent advice and distribution 6.33–6.34

independent vs non-independent advice 6.30–6.32 tied agents and advice 6.35–6.36 MiFID I 6.15–6.19 advice 6.15 issues with regime 6.16–6.19 MiFID II 6.20–6.44 portfolio management 6.44, 8.40–8.46 missing research 6.13–6.14 scope and overview 6.01–6.02, 6.45–6.47 see also contractual obligations; inducements; suitability indirect clearing obligations 15.76–15.104 default management 15.89–15.91 definition 15.80–15.81 extended chains 15.94–15.97 general obligations 15.76–15.79 information flow 15.92–15.93 concerns about 15.104 leapfrog payment 15.89–15.91 miscellaneous 15.97–15.101 default management concerns 15.98–15.99 extent of protection 15.98–15.99 obligations of means 15.100–15.101 post-default porting 15.89–15.91 scope 15.82 segregation 15.84–15.88 structure of arrangements 15.83 territoriality 15.103–15.104 see also derivatives inducements 8.01–8.68 current legislation 8.04–8.17 ESMA guidance 8.08–8.17 direct fees 8.11 exceptions 8.11–8.17 fees enhancing quality of service 8.12–8.14 intra-group payments 8.10 proper fees 8.15–8.17 scope of the inducement ban 8.09–8.10 standard fees 8.09 MiFID 8.04–8.05 implementing Directive 8.06–8.07 Dutch inducement ban 8.19–8.29 ‘gold plating’, legal basis and necessity 8.25–8.28 investment services outside scope of ban, rules for 8.29

territorial scope 8.24 MiFID II 8.37–8.59 ancillary services 8.47–8.57 deviating requirements 8.58–8.59 independent investment advice 8.40–8.46 investment services 8.47–8.57 minor non-monetary benefits 8.43–8.46 overview 8.37–8.39 portfolio management 8.40–8.46 overview 8.01–8.03 research as an inducement 8.60–8.67 definition 8.61–8.63 inducement 8.64–8.65 minor non-monetary benefit 8.66–8.67 overview 8.60 UK inducement rules 8.30–8.36 MiFID implementation 8.31 overview 8.30 retail distribution review 8.32–8.36 inducement rules 8.33–8.36 informed trading, see equity trading initiative test 10.66–10.76 own exclusive initiative 10.73–10.76 under MiFID II 10.70–10.72 under MiFIR 10.66–10.69 see also third-country firms insurance companies 2.51–2.54 Insurance Distribution Directive (IDD) conduct of business rules: general conduct 21.08–21.11 insurance-based investment products 21.12–21.14 MiFID II, differences with 21.15–21.24 structure 21.07–21.14 investment protection 21.04–21.26 context 21.04–21.06 MiFID II-IDD nexus 21.25–21.26 UCITS KII and PRIIPs KID 21.27–21.51 Italy investment-based crowdfunding 23.22–23.23 internalization dark pools 18.94–18.105 function 18.96–18.98 ‘ideal’ dark pools and social desirability 18.99–18.100

damaging practices for traders 18.101–18.102 internalization 18.103–18.104 intervention powers commodity derivatives 16.83–16.89 investigations 19.99–19.100 see also cooperation arrangements investment activities 10.05–10.08 see also third-country firms investment advice best interest duty 4.35–4.42 portfolio management 6.44 investment firms algorithmic trading 17.23–17.32 direct electronic access (DEA) 17.29–17.32 general clearing 17.29–17.32 requirements 17.23–17.28 authorization 1.12 contractual obligations 6.37–6.38 CRD IV 1.15 credit institutions 1.13 CRR 2.16–2.25 dark trading: trading obligations 14.92–14.93 dealing on own account 1.09 definition 1.08–1.13 execution of orders on behalf of clients 1.09, 1.10 financial instruments 1.09 governance of, see governance high frequency trading (HFT) 17.44–17.47 investment activities, 1.09 investment advice 1.09, 1.10 ‘investment fund’ 1.08 investment services 1.09 MiFID scope 2.02–2.14 multilateral trading facility (MTF) 1.09 organized trading facility (OTF) 1.09 personnel and suitability 6.37–6.38 portfolio compression obligation 15.112 market operators 15.112 notional value reduction 15.113–15.114 portfolio management 1.09, 1.10 reception and transmission of orders 1.09 underwriting 1.09

investment protection 21.01–21.62 EU legislative approach 21.01–21.03, 21.57–21.62 MiFID II: conduct of business rules 21.07–21.24 context 21.04–21.06 cross-selling practices 21.52–21.56 evaluation 21.47–21.51 IDD vs 21.04–21.26 information obligations on costs 21.28–21.34 MiFID II-IDD nexus 21.25–21.26 PRIIPs KID 21.27–21.51 product risk assessment 21.35–21.46 UCITS KII 21.27–21.51 see also risk investment services investment activities vs 10.05–10.08 independent advice 6.30–6.36 distribution 6.33–6.34 non-independent advice 6.30–6.32 tied agents 6.35–6.36 see also independent financial advice; third-country firms Investment Services Contracts execution 7.61–7.70 MiFID I to MiFID II transition 7.63–7.64 MiFID II regime parameters 7.65–7.70 strict investors’ interest approach 7.61–7.62 formation 7.43–7.54 contract standards solution 7.44 MiFID I to MiFID II transition 7.45–7.46 MiFID II regime parameters 7.47–7.54 see also conflicts of interest Investment Services Directive (ISD) 3.07–3.13 adoption 1.04 alternative trading venues 1.06 economic functions 1.06 origins, Europe and US 1.06 pros and cons 1.06 commodity derivatives 1.05 criticism 1.04 harmonization 1.04–1.05 investment advice 1.05 investor protection 1.05 optional concentration rule 1.06

risks 1.05 stability 1.05 stock markets 1.06 see also MiFID I investment-based crowdfunding, see crowdfunding investor protection agency and principal dealing 9.19–9.22 product governance 5.14–5.16 distributors 5.16 manufacturers 5.15 Ireland trading venue competition 1.06 Italy investment-based crowdfunding 1.51 trading venue competition 1.06 key information documents (KID), see PRIIPs KID key investor information document (KII), see UCITS KII Know Your Customer (KYC) rules product governance 5.40 leapfrog payment clearing obligation 15.47–15.48 indirect clearing obligation 15.89–15.91 legal documentation portfolio compression obligation 15.123 linked contracts 7.71–7.77 cross-selling investment services 7.75–7.77 loan contracts and investment contracts 7.73–7.74 Unfair Trade Practices/Competition Law approach 7.72 see also conflicts of interest liquidity economics 18.06–18.21 accounting perspective 18.17 information perspective 18.18–18.19 private information 18.09–18.12 informed vs uninformed traders 18.13–18.15 informed trading and transaction prices 18.20–18.21 liquidity suppliers 18.16–18.19 market characteristics 18.29–18.30 resource allocation 18.31–18.32 risk allocation 18.34 share price accuracy 18.33

in scope, see trading obligation London Stock Exchange Group 11.37–11.40 see also financial market infrastructure (FMI) groups Luxembourg trading venue competition 1.06 mandatory clearing OTC derivatives (EMIR) 22.63–22.74 manufacturers product governance 5.07–5.08 investor protection 5.15 monitoring obligation 5.33–5.37 product approval process 5.18–5.24 staff 5.30 manufacturing best interest duty 4.50–4.52 market abuse high frequency trading (HFT) 17.51–17.60 identification 17.52–17.56 prevention 17.51–17.60 price volatility 18.114–18.117 regulation 17.51–17.60 sanctions 17.57–17.60 Market Abuse Regulation (MAR) SME growth markets 13.24–13.26 market operators 2.60 MiFID I (EU Markets in Financial Instruments Directive) agency 1.23, 9.01–9.55 practicable distinctions 9.46–9.50 commodity derivatives 1.05 conflicts of interest 1.21, 7.08–7.13 organizational requirements 7.36–7.37 core pillar of EU financial market integration 1.07 crowdfunding, investment-based 23.12–23.17 financial instrument 23.13–23.14 investment service 23.15–23.17 limitations of MiFID 23.19–23.25 dark trading 14.39–14.58, 14.94–14.100 dark pools 1.34 pre-trade transparency 14.40–14.58 venue classification 14.40–14.51 financial crisis 1.02 financial instrument, definition 1.05

financial market transparency 1.01, 1.02 governance-related regulation 3.14–3.19 harmonized investor protection 1.01 Implementing Directive: inducements 8.04–8.05 investment firms 1.01 national legislation in Europe 1.01 origins 1.01 regulated markets 1.01 Implementing Regulation: direct effect 1.01 investment firms 1.01 origins 1.01 regulated markets 1.01 independent financial advice 6.15–6.19 on advice 6.15 regime problems 6.16–6.19 inducements 8.04–8.05 implementation 8.31 investment advice 1.05 investment firms 1.01 Investment Services Contracts: contract formation 7.45–7.46 execution 7.63–7.64 maximum harmonization 1.05 national approaches to 1.01, 23.19–23.25 optional concentration rule 1.06 origins 1.01, 1.03, 1.04–1.06 principal dealing 1.23 private law effect 1.46–1.47 civil courts 20.03–20.20 contracting parties 20.21–20.27 contractual limitation 20.36–20.38 private investors 20.39 proof of causation 20.30–20.35 relativity principle 20.28–20.29 strict approach 20.03–20.27 regime on fees 7.57 regulated markets 1.01 rejection of 23.19–23.25 review of regime 1.07 scope of 2.01–2.61 basic provision 2.01

broad impact 2.61 collective investment undertakings 2.55–2.56 credit institutions 2.15 CRR, investment firms under 2.16–2.25 data-reporting service providers 2.57–2.59 dealings in emission allowances 2.42 financial instruments 2.43–2.50 general exemptions to applicability 2.29–2.36 insurance companies and intermediaries 2.51–2.54 investment firms 2.02–2.14, 2.16–2.25 market operators 2.60 structured deposits 2.26–2.28 trading on own account 2.37–2.41 strict application 1.02, 1.05 supervision rules 1.05 trading venue competition 1.01 transition to MiFID II 1.07, 1.21 conflicts of interest 7.36–7.37 Investment Services Contracts 7.45–7.46, 7.63–7.64 regime on fees 7.57 see also investment firm; Investment Services Directive (ISD) MiFID II ((EU Markets in Financial Instruments Directive II) 1.02 administrative sanctions 19.47–19.53 advisory and distribution services 1.14 agency 1.23, 9.01–9.55 algorithmic trading (AT) 1.42–1.43, 17.01–17.62 benefits and risks 1.42 Flash Crash (2010) 1.42 best execution 1.17 best interest duty 1.17–1.18, 4.07–4.15 client’s best interest standard 1.18 ‘better’ regulation 1.55 Brexit 1.54, 10.41 broad perspectives 1.48–1.51 EU piecemeal approach 1.48–1.49 Insurance Distribution Directive 1.48 PRIIPs Directive 1.48 UCITS Directive 1.48 Call for Evidence (Sept 2015) 1.56 commodity derivatives 1.40–1.41, 16.42–16.62 ancillary activities 1.41 ESMA 1.41 ex-ante positions 1.41

exemption for traders 1.14 G20 mandate 1.40 methodology 1.41 NCAs 1.41 product intervention 1.41 types 1.41 conflicts of interest 1.17, 1.21 core parameters 7.38–7.42 organizational requirements 7.36–7.37 cross-selling practices 21.52–21.56 crowdfunding, investment-based 23.34–23.54 definitions 23.37–23.41 EU passport and applicable law 23.45–23.51 exemptions 23.37–23.41 Great Financial Crisis 23.34–23.36 OTFs 23.42–23.44 policy considerations 23.52–23.54 tied agents 23.42–23.44 dark pools 1.34 regulation 14.59–14.100 data reporting service providers 1.14 Directive 1.02 duty to act in client’s best interest 1.16–1.18, 1.23 equity trading 18.01–18.127 dark pools and price formation 18.106–18.112 high-frequency trading (HFT) 18.113–18.117 issues raised 18.105–18.127 market abuse 18.118–18.121 price volatility 18.114–18.117 financial market infrastructure (FMI) groups 11.57–11.85 MiFID approach 11.73–11.79 IDD vs 21.04–21.26 conduct of business rules 21.15–21.24 MiFID II-IDD nexus 21.25–21.26 independent financial advice 6.20–6.44 contractual obligations 6.37–6.43 independent advice 6.30–6.36 information 6.20–6.29 investment services 6.30–6.36 portfolio management 6.44 inducements 8.37–8.59 ancillary services 8.47–8.57 deviating requirements 8.58–8.59

independent investment advice 8.40–8.46 investment services 8.47–8.57 minor non-monetary benefits 8.43–8.46 portfolio management 8.40–8.46 investment firms, governance of 3.01–3.65 organization and risk management 3.50–3.55 technical framework 3.21–3.65 investment protection 21.01–21.62 Investment Services Contracts: contract formation 7.45–7.46 core parameters 7.65–7.70, 7.47–7.54 execution 7.63–7.64 KYC rules 5.40 limitations to inducements 1.17 NCA powers 5.61–5.62 PRIIPs KID vs 21.27–21.51 products 1.18 SME growth markets 13.15–13.33 future regime 13.18–13.23 staff remuneration practices 1.17 emission allowances 1.14 equity trading regulation 1.44 US perspective 1.44 EU Commission 1.54–1.55 European financial sector 1.02 financial instrument, definition 1.14 financial markets, functioning of 1.50 future and impact 1.48–1.57 high-frequency trading (HFT) 1.42–1.43 benefits and risks 1.42–1.43 ESMA adopting guidelines 1.43 market abuse 1.44 market developments 1.43 price instability 1.44 independent advice 1.20 independent EU bodies 1.54 inducements 1.22 investment-based crowdfunding 1.51 investment-based insurance products 1.14 investor protection 1.48–1.49 lobby groups 1.54 TheCityUK 1.54 MiFID III, duty of care 1.23

MiFIR, ‘twin’ relationship 19.01–19.06 non-independent advice 1.20 origins 1.07 OTFs 1.14 principal dealing 1.23 product intervention 1.19, 5.01–5.74 product governance 1.19, 5.01–5.74 costs 1.19 suitability assessment 1.20 prudential supervision 1.14 public enforcement 19.01–19.06 regime on fees 7.57 core parameters 7.58–7.60 regulatory inconsistency 1.20 scope 1.14 shadow banking 1.50 Central Securities Depositories 1.50 derivatives (OTC/CCP) 1.50 Financial Stability Board 1.50 securities markets 1.50 securitized instruments 1.50 SME growth markets 1.33 disclosure obligations 1.33 strict application 1.02 structured deposits 1.14 supervision and enforcement 1.45–1.47 administrative sanctions 1.45 authorities’ sanctioning powers 1.45 cooperation arrangements 1.45 conduct of business rules 1.46 contractual limitation 1.46 criminal sanctions 1.45 EU Court of Justice 1.47 exclusion of liability 1.46 financial supervision 1.45 micro-prudential supervisory effectiveness 1.45 private law effect 1.46–1.47 proof of causation 1.46 proximity principle 1.46 public enforcement 1.45 redress procedures 1.45 relativity principle 1.46 supervisory cooperation arrangements 1.45

supervision of investment firms 1.14 supervisory powers 19.30–19.36 third-country investment firms 1.24–1.28, 10.01–10.79 applicability of requirements 10.63 eligible counterparties 1.28, 10.18–10.45 ESMA 1.26, 10.30–10.34 initiative test 1.28, 10.66–10.76 opting up 1.26, 10.46–10.57 professional investors, EU/EEA 1.26, 10.46–10.57 retail clients 1.28, 10.46–10.57 structured deposits 1.28, 10.09–10.11 third-country regime, clarity of 1.28 trading: transparency regime 1.30–1.32 bilateral/OTC trading 1.30 dark trading 1.30 ESMA 1.32 EU financial governance 1.32 liquidity 1.30 MTFs 1.30 OTFs 1.30 Regulated Markets (RMs) 1.30 waivers 1.30–1.32 trading on own account 1.14 trading venues: CCPs 1.29 EMIR 1.29 Financial Markets Infrastructures (FMIs) 1.29 governance and organization 1.29 MTFs 1.29 prudential supervision 1.29 Regulated Markets (RMs) 1.29 OTFs 1.29 transition from MiFID I 1.07, 1.21 conflicts of interest 7.36–7.37 Investment Services Contracts 7.45–7.46, 7.63–7.64 regime on fees 7.57 UCITS KII vs 21.27–21.51 see also investment firm; MiFID II MiFIR (Markets in Financial Instruments Regulation) 1.02 commodity derivatives 16.42–16.62 dark pool regulation 14.59–14.100 derivatives trading: Dodd-Franks trading mandate 1.36

ETD space 1.39 execution costs 1.36 new mandatory trading obligation 1.35–1.39 portfolio compression 1.38 STP obligation 1.37 veiled impact 1.39 European financial sector 1.02 Level 2 legislation 1.02 KYC rules 5.40 MiFID II, ‘twin’ relationship 19.01–19.06 NCA powers 5.42–5.56 product intervention 1.19, 5.01–5.74 product governance 1.19, 5.01–5.74 strict application 1.02 third-country firms 10.01–10.79 equivalence decision 10.28–10.29 initiative test 10.66–10.69 post-equivalence decision 10.66–10.69 transparency: administrative regulatory governance 12.45–12.55 maximalist approach 1.30 regime 12.14–12.33 supervisory administrative governance 12.34–12.44 see also MiFID II Money Market Funds (MMFs) 22.79–22.86 see also shadow banking monitoring obligation product governance 5.32–5.39 distributor 5.38–5.39 general 5.32 manufacturer 5.33–5.37 multi-dealer request-for-quote model 15.12 see also trading obligation multilateral netting clearing obligation 15.38 multilateral trading facility (MTF) dark pools 14.16–14.18 trading facilities 11.22–11.29 corporate governance 11.25 definitions 11.23–11.24 organization 11.26–11.29 NASDAQ OMX Nordic Exchange 11.53–11.54

see also financial market infrastructure (FMI) groups national competent authorities (NCAs) consultation with other NCAs 5.57 EBA: consultation with 5.57 coordinating rule 5.59–5.60 ESMA: consultation with 5.57 coordinating rule 5.59–5.60 MiFID II powers 5.61–5.62 MiFIR powers 5.42–5.56 general 5.42–5.44 other requirements 5.52–5.56 ‘specific concern’ or ‘threat’ 5.45–5.51 product intervention 5.42–5.62 Nationale-Nederlanden v. Van Leeuwen 20.06–20.17 see also private law effect Netherlands civil courts 20.06–20.17 Dutch Supreme Court: civil duty of care 1.23, 9.51 inducement ban 8.19–8.29 gold plating 1.22, 8.25–8.28 investment services, rules for 8.29 territorial scope 8.24 Nationale-Nederlanden v. Van Leeuwen 20.06–20.17 private law effect 20.06–20.17 trading venue competition 1.06 netting clearing obligation 15.46 novation clearing obligation 15.37 multilateral 15.38 order management systems (OMS) dark pools 14.21 see also dark trading obligation to cooperation, see cooperation arrangements obligations, see clearing obligation; derivatives; indirect clearing obligation; portfolio compression obligation; straight-through-processing (STP) obligation; trading obligation on-site verification 19.99–19.100 see also cooperation arrangements

organized trading facilities (OTFs) financial market infrastructure (FMI) groups 11.65–11.68 investment-based crowdfunding 23.42–23.44 trading facilities 11.22–11.29 corporate governance 11.25 definitions 11.23–11.24 organization 11.26–11.29 over-the-counter (OTC) markets dark pools 14.16, 14.19 mandatory clearing (EMIR) 22.63–22.74 shadow banking 22.63–22.74 STP obligation 15.54–15.56 contractual timeframes 15.55–15.56 see also dark trading packaged retail investment and insurance-based investment products, see PRIIPs KID partnership law 3.63 see also governance portfolio compression obligation 15.105–15.127 definition 15.107–15.110 regulatory developments 15.110 general obligations 15.105–15.106 investment firms 15.112 market operators 15.112 notional value reduction 15.113–15.114 record-keeping 15.126–15.127 requirements 15.117–15.123 legal documentation 15.123 process 15.118–15.122 scope 15.111–15.123 criteria 15.111–15.116 counterparty performance 15.115–15.116 transparency 15.124–15.125 see also derivatives Portugal trading venue competition 1.06 position limits regime 16.63–16.82 calculation and methodology 16.74–16.76 EU v USA approaches 16.77–16.82 financial v non-financial entities 16.66–16.69 hedging exemption 16.70–16.73 overview 16.63–16.65

see also commodity derivatives post-crisis administrative governance, see transparency regulation post-default porting clearing obligation 15.47–15.48 indirect clearing obligation 15.89–15.91 pre-trade transparency venue classification 14.40–14.51 waivers 14.52–14.58 see also dark trading price accuracy equity trading 18.26–18.28 share price accuracy 18.33 price volatility 18.114–18.117 market abuse and 18.113–18.117 see also equity trading; high frequency trading (HFT) PRIIPs KID 21.27–21.51 see also investment protection principal dealing, see agency and principal dealing private information types and sources 18.09–18.12 announcement information 18.11 fundamental value information 18.12 inside information 18.10 see also equity trading private investment 20.39 see also private law effect private law effect 20.01–20.40 assessment 20.40 civil courts 20.03–20.20 duties to furnish additional information 20.15–20.17 facts 20.06–20.09 fairness 20.15–20.17 general 20.05 legal framework 20.10–20.13 less strict approach than MiFID 20.03–20.04 Nationale-Nederlanden v. Van Leeuwen 20.06–20.17 questions referred for a preliminary ruling 20.14 reasonableness 20.15–20.17 stricter approach than MiFID 20.18–20.20 contracting parties 20.21–20.27 less strict approach than MiFID 20.21 stricter approach than MiFID 20.22–20.27 contractual limitation 20.36–20.38

exclusion of liability 20.36–20.38 general 20.01–20.02 Genil case 1.47, 9.43, 9.45, 20.03, 20.05, 20.19, 20.21, 20.23, 20.29, 20.36–20.37, 20.40 private investment 20.39 principle of relativity 20.28–20.29 proof of causation 20.30–20.35 product governance 5.02–5.40 corporate governance 5.12–5.13 investor protection 5.14–5.16 distributors 5.16 manufacturers 5.15 KYC rules 5.40 manufacturers and distributors 5.06–5.10 distributors 5.09–5.10, 5.16 general 5.06 manufacturers 5.07–5.08, 5.15 monitoring obligation 5.32–5.39 distributor 5.38–5.39 general 5.32 manufacturer 5.33–5.37 organizational requirements 5.17–5.39 general 5.17 perspectives 5.11–5.39 product approval process: manufacturers 5.18–5.24 product governance procedures: distributors 5.25–5.29 product governance requirements: staff 5.30–5.31 distributors 5.31 manufacturers 5.30 product intervention and 5.01, 5.74 scope 5.02–5.05 product intervention 5.41–5.73 ESMA and EBA, temporary powers 5.63–5.73 emergency situations 5.73 general 5.63–5.64 powers under MiFIR 5.65–5.72 NCAs 5.42–5.62 consultation with ESMA/EBA 5.57 consultation with other NCAs 5.57 ESMA and EBA, coordinating rule 5.59–5.60 general powers 5.42–5.44 MiFID II powers 5.61–5.62

MiFIR powers 5.42–5.56 other requirements 5.52–5.56 specific concerns 5.45–5.51 threats 5.45–5.51 product governance and 5.01, 5.74 scope 5.41 professional clients, see retail clients professional investors, see retail investors professional secrecy 19.19–19.23 see also competent authorities proof of causation 20.30–20.35 see also private law effect prospectus regime 22.109–22.120 see also shadow banking public enforcement MiFID II 19.01–19.06 twin legal acts (MiFID / MiFIR) 19.01–19.06 see also supervision and enforcement qualifying holdings 3.56–3.57 see also governance reasonable expectations investors 9.24–9.25 reasonableness private law effect 20.15–20.17 reciprocity 10.24–10.27 reciprocity test 10.21 see also equivalence; third-country firms record-keeping portfolio compression obligation 15.126–15.127 redress procedures 19.75–19.82 extra-judicial mechanism for consumer complaints 19.80–19.82 reporting of infringements 19.75–19.77 right of appeal 19.78–19.79 see also competent authorities regime on fees, see conflicts of interest regulated markets (RM) dark pools 14.16–14.18 high frequency trading (HFT) 17.48–17.50 trading facilities 11.17–11.21 corporate governance 11.18–11.20 organization 11.21–11.21

relativity principle 20.28–20.29 see also private law effect research 8.60–8.67 definition 8.61–8.63 as an inducement 8.64–8.65 as a minor non-monetary benefit 8.66–8.67 overview 8.60 see also inducements resource allocation liquidity 18.31–18.32 retail clients branch authorization, withdrawal of 10.50 European passport, absence of 10.53–10.55 exercise of member state option 10.47–10.50 general 10.46 member state option, non-exercise of 10.51–10.52 opt up professional clients 10.46–10.57 third-country firm, branch resolution of 10.56–10.57 retail investors 10.58–10.65 equivalence decision: none in effect 10.64 post-equivalence decision 10.59–10.63 member state option: exercise 10.59–10.63 not exercised 10.65 MiFID II/MiFIR applicability 10.63 see also third-country firms right of appeal 19.78–19.79 see also redress procedures risk allocation: liquidity 18.34 assessment: product governance 21.38–21.42 product risk 21.35–21.46 product risk indicator 21.36–21.37 suitability 21.43–21.46 management 3.47–3.55 CRD IV regime requirements 3.47–3.49 MiFID II requirements 3.50–3.55 see also governance sanctions

high frequency trading (HFT) 17.57–17.60 market abuse 17.57–17.60 see also administrative sanctions; competent authorities; criminal sanctions securities financing transactions (SFT) 22.100–22.108 see also shadow banking securitization EU measures 22.44–22.62 segregation client clearing obligations 15.43, 15.45 indirect clearing 15.89 self-placement best interest duty 4.53–4.70 shadow banking 22.01–22.141 banking supervision 22.27–22.34 concept and terminology 22.10–22.16 European response 22.35–22.131 crowdfunding 22.121–22.131 CSDR 22.87–22.99 financial activity 22.27–22.34 regulation and crisis 22.01–22.09, 22.132–22.141 FSB approach 22.17–22.26 mandatory clearing for OTC derivatives (EMIR) 22.63–22.74 measures addressing the banks 22.38–22.62 credit risk concentration 22.39 securitization 22.40–22.43 EU measures 22.44–22.62 Money Market Funds (MMFs) 22.79–22.86 non-regulated activity 22.17–22.26 prospectus regime developments 22.109–22.120 securities financing transactions (SFT) 22.100–22.108 short-selling regulation 22.75–22.78 share price accuracy liquidity 18.33 shareholders 3.56–3.57 see also governance short-selling regulation 22.75–22.78 see also shadow banking slow market arbitrage 18.72–18.93 efficiency considerations 18.81–18.93 examples 18.74–18.76 high-frequency trading and volatility 18.84–18.93 efficiency considerations 18.92–18.93 Flash Crash 18.87–18.90

general increase in volatility 18.85–18.86 wealth-transfer considerations 18.91 wealth transfer effects 18.77–18.80 see also equity trading small and medium-sized enterprises (SMEs), see SME growth markets SME growth markets 13.01–13.54 alternative disclosure obligations 13.34–13.46 disclosure of current event reports 13.42–13.46 disclosure of key operating milestones 13.35–13.41 alternative European markets 13.04–13.14 admission to trading 13.08–13.09 overview 13.04–13.07 regulatory approaches 13.08–13.14 requirements for being public 13.10–13.14 unity and diversity 13.08–13.14 history and context 13.01–13.03, 13.54 interest swap rates 1.19 labelling, role for 13.47–13.53 MiFID II 13.15–13.33 concept 13.16–13.17 criticism 13.27–13.33 future MiFID II regime 13.18–13.23 MAR requirements 13.24–13.26 regulatory aim 13.15 Spain investment-based crowdfunding 1.51, 23.21 MiFID, rejection of 23.21 trading venue competition 1.06 straight-through-processing (STP) obligation 15.49–15.75 CCP: trades not accepted 15.71–15.73 cleared derivatives: bilateral 15.64–15.70 exemption 15.70 non-electronic means 15.72 timeframe for CB acceptance 15.66–15.67 timeframe for CCP acceptance 15.68–15.69 timeframe for submission by counterparties to CCP 15.64–15.65 cleared derivatives: trading venues 15.58–15.63 electronic means 15.71 exemption 15.63 non-acceptance 15.62 pre-determined information 15.58 pre-trade checks 15.59 timeframe for CCP acceptance 15.61

timeframe for submission by trading venue to CCP 15.60 clerical problems 15.73 definition 15.52–15.56 contractual timeframes 15.55–15.56 ETDs 15.53 OTC Derivatives 15.54–15.56 EU/US comparative approach 15.75 frontloading 15.74 general obligation 15.49–15.51 scope 15.57–15.75 technical problems 15.73 see also derivatives structured deposits 2.26–2.28 selling and advising clients 10.09–10.11 suitability assessment 6.37–6.38 periodic 6.28 risk 21.43–21.46 investment firms’ personnel 6.37–6.38 product governance 6.39–6.42 shareholders 11.77–11.78 written statement 6.43 see also contractual obligations; independent financial advice supervision and enforcement 19.01–19.134 assessment 19.127–19.134 competent authorities 19.11–19.82 cooperation arrangements 19.83–19.126 public enforcement of MiFID II 19.01–19.06 ‘twin legal acts’ 19.01–19.06 structure 19.07–19.10 see also competent authorities; cooperation arrangements supervisory activities 19.99–19.100 see also cooperation arrangements supervisory powers 19.27–19.39 Article 69 MiFID II powers 19.30, 19.36 exercise of 19.37–19.39 general provisions 19.27–19.36 Member States’ notification obligation 19.27–19.29 rationale 19.27–19.29 see also competent authorities territoriality indirect clearing obligation 15.103–15.104

third-country firms 10.01–10.79 AIFMs and AIFMD 1.24 branch resolution 10.56–10.57 continuing obligations 10.35–10.36 depositaries 1.24 eligible counterparties 10.18–10.45, 10.58–10.65 initiative test 10.66–10.76 opt up professional clients 10.46–10.57 overview 10.01–10.03, 10.77–10.79 professional clients 10.18–10.45 professional investors 10.58–10.65 retail clients 10.46–10.57 retail investors 10.58–10.65 scope 10.04–10.17 data-reporting services 10.12–10.14 investment services v investment activities 10.05–10.08 operating trading venues 10.15–10.17 structured deposits, selling and advising clients 10.09–10.11 third-country firms 10.04 see also eligible counterparties; initiative test; professional clients; retail clients; retail investors third-country entities 15.14 see also trading obligation third-party service providers 15.115, 15.118, 15.125 tied agents independent financial advice 6.35–6.36 investment-based crowdfunding 23.42–23.44 trading 11.01–11.86 history and context 11.01–11.02, 11.86 see also dark trading; financial market infrastructure (FMI) groups; trading facilities; trading obligation; trading venues trading facilities 11.03–11.16 CCPs 11.30–11.31 MTFs and OTFs 11.22–11.29 corporate governance 11.25 definitions 11.23–11.24 organization 11.26–11.29 regulated markets (RM) 11.17–11.21 corporate governance 11.18–11.20 organization 11.21–11.21 scenario of 11.03–11.14 trading venues: EU regulatory framework 11.15–11.16

non-discriminatory access to 11.30–11.31 see also financial market infrastructure (FMI) groups trading obligation 15.06–15.31 derivatives in scope 15.17–15.25 alternative process 15.25 liquidity in scope 15.20–15.24 venue test 15.19 entities in scope 15.13 exchange trading 15.08–15.31 exemptions 15.15–15.16 general obligation 15.06–15.07 miscellaneous 15.26–15.31 ETDs 15.31 package transactions 15.30 register 15.29 timing 15.26–15.28 scope 15.13–15.16 third-country entities 15.14 trade life cycle 15.08 trading models 15.09–15.12 central limit order book model 15.11 multi-dealer request-for-quote model 15.12 see also derivatives trading on own account 2.37–2.41 trading venues algorithmic trading and HFT 17.33–17.40 organizational requirements 17.34–17.38 trading requirements 17.39–17.40 cleared derivatives 15.58–15.63 bilateral execution 15.72 CCP 15.60–15.61 electronic execution 15.71 exemption 15.63 non-acceptance 15.62 non-electronic execution 15.72 pre-determined information 15.58 pre-trade checks 15.59 commodity derivatives: intervention powers 16.83–16.89 competition 18.122–18.127 cooperation arrangements: host member states 19.91–19.96 equity trading 18.122–18.127 EU scenario 11.32–11.57

evolution of trading venues market 11.33–11.35 FMI groups 11.36–11.57 market fragmentation 1.06 non-discriminatory access 11.30–11.31 operating 10.15–10.17 transparency regulation 12.18–12.25 transparency ownership 11.77–11.78 portfolio compression obligation 15.124–15.125 transparency regulation 12.01–12.57 administrative regulatory governance 12.45–12.55 bilateral trading requirements 12.26–12.33 EU financial governance 12.01–12.33 MiFIR transparency regime 12.14–12.33 overview 12.56–12.57 purpose 12.10–12.13 significance 12.01–12.09 supervisory administrative governance 12.34–12.44 trading venues requirements 12.18–12.25 UCITS KII 21.27–21.51 see also investment protection undertakings for the collective investment in transferable securities, see UCITS KII United Kingdom EU membership 1.54 inducement rules 8.30–8.36 MiFID implementation 8.31 overview 8.30 retail distribution review 8.32–8.36 specific provisions 8.33–8.36 investment-based crowdfunding 1.51, 23.20 Kay Review: duties of care 1.23 government response 9.52–9.54 lobby groups 1.54 trading venue competition 1.06 United States best interest duty 4.74–4.82 commodity derivatives: position limits regime 16.77–16.82 equity trading 18.01–18.03 STP obligation 15.75

venue test 15.19 see also trading obligation volume cap mechanism dark trading 14.77–14.85 double volume cap 14.86–14.91 waivers dark trading: equity instruments 14.67–14.76 pre-trade transparency 14.52–14.58 wealth transfer effects 18.77–18.80 electronic front-running 18.50–18.61 slow market arbitrage 18.91