International and Comparative Secured Transactions Law: Essays in honour of Roderick A Macdonald 9781849467650, 9781782257844, 9781509901142

The law of secured transactions has seen dramatic changes in the last decade. International organisations, particularly

241 90 3MB

English Pages [369] Year 2017

Table of contents :
Preface
Table of Contents
List of Contributors
Table of Legislation
Introduction
1
Secured Credit Legislation: Functionalism or Transactional Co-Existence
I. Introduction
II. What is the Position in England?
III. The Nature and Limits of Functionalism
IV. Models of Transactional Co-Existence
2
‘Functional Formalism’ in the Treatment of Leases under Secured Transactions Law: Comparative Lessons from the Canadian Experience
I. Introduction
II. The Economic Substance Approach to the Characterisation of Leases under the PPSAs and the UNCITRAL Model Law
III. Extension of the PPSAs to Long-term ‘True’ Leases
IV. The Functionalist Underpinnings of the Formal Characterisation of Leases under the Civil Code of Quebec
V. Characterisation of Leases outside Secured Transactions Law
VI. Conclusion
3
Reflections on Misgivings about a Model Law
I. Rod’s Misgivings
II. My Misgivings
III. Assessment of the UNCITRAL Model Law in Light of Our Misgivings
IV. A Final Appreciation and Concern
4
The UNCITRAL Model Law on Secured Transactions
I. Introduction
II. Key Objectives and Fundamental Policies
III. Scope and General Provisions
IV. Creation of a Security Interest
V. Third-party Effectiveness of a Security Interest
VI. The Registry System
VII. Priority of a Security Interest
VIII. Rights and Obligations of the Parties and Third-party Obligors
IX. Enforcement of a Security Interest
X. Law Applicable to Security Interests
XI. Conclusions
5
Non-assignment Clauses and their Treatment under UNCITRAL’s Secured Transactions Law Instruments
I. Introduction
II. What are Anti-assignment Clauses and Why Are They Used in Contracts?
III. The Legal Effect of Anti-assignment Clauses
IV. Why is it Necessary to Regulate Prohibitions on the Assignment of Receivables?
V. UNCITRAL’s Approach to Anti-assignment Clauses
VI. Conclusions
6
Dealing with Concepts of Property in the UNCITRAL Legislative Guide on Secured Transactions and UNCITRAL Model Law on Secured Transactions (2016)
I. Introduction
II. Rights to Claims that Support Encumbered Assets
III. Rights of Persons with Contractual Rights to Encumbered Assets
IV. Rights of a Licensee of Intellectual Property
V. Conclusion
7
Current Issues in Cross-Border Asset-Based Lending: Lessons from the Field on the Need for Secured Transactions Reform
Author’s Note: Remembering Rod Macdonald
I. Introduction
II. The Growing Cross-border Financing Needs of Middle-market Companies
III. Asset-based Lending
IV. Current Issues in Cross-border Asset-based Lending
V. Conclusion
8
The Rights and Obligations of the Parties to the Security Agreement According to the UNCITRAL Model Law on Secured Transactions
I. Introduction
II. General Remarks
III. The Parties Concerned
IV. The Sources of the Reciprocal Rights and Obligations of the Parties (Article 52)
V. Obligation to Preserve the Encumbered Asset (Article 53)
VI. Obligation of the Secured Creditor to Return the Encumbered Asset (Article 54)
VII. The Secured Creditor’s Right to Use and Inspect the Encumbered Asset (Article 55)
VIII. Right of the Grantor to Obtain Information (Article 56)
IX. Conclusion
9
Sûretés Mobilières et Stocks : ou l’Art et la Manière de Résoudre la Quadrature du Cercle
I. L’Art: Comment Réconcilier l’Inconciliable
II. La Manière: le Traitement Légistique du Gage des Stocks
10
Some Thoughts (and Facts) about the Process of Secured Transactions Law Reform, with Special Emphasis on Registration, the Key to Achievement of Reform’s Goals
I. Introduction
II. Key Features of a Modern Efficient Registry
III. Reform Approaches: Comprehensive or Piecemeal
IV. The Quixotic Search for Brevity
V. Allocation of Functions
VI. Interrelationship between Policy Choices, the Statutory Rules Governing the Registration System and the Registry Design
VII. The Crucial (but Limited) Functions of a Secured Transactions Registry
VIII. Omissions from and Gaps in the UNCITRAL Model Law
IX. Registry Fees
X. The Need for Education
XI. Conclusion
11
Conflict-of-Laws Rules on Security Rights in Non-Intermediaries Securities
I. Concepts and Terminology
II. Role and Scope of Conflict Rules and Policy Considerations
III. Creation, Effectiveness against Third Parties and Priority
IV. Effectiveness against the Issuer and Enforcement
V. The UNCITRAL Rules
12
Comparative Approaches to the Enforcement of Secured Credit in Insolvency
II. The Enforcement of Secured Credit in Insolvency Proceedings Generally
III. The Treatment of Secured Credit in Rehabilitation Proceedings
IV. Conclusion
13
Comparative Study on Indian Secured Transactions Law and the UNCITRAL Legislative Guide on Secured Transactions
I. Introduction
II. Historical Background of New Secured Transactions Law in India
III. Comparative Analysis
IV. Modifications Required in Indian Secured Transactions Law to Conform with the Legislative Guide
V. Conclusion
VI. Recent Reforms in India—an Addendum
14
Security Interests over Intellectual Property Rights in Italy: Critical Analysis and Reform Proposals
I. Introduction
II. Legal Framework for Taking Security over Intellectual Property Rights
III. Taking Security over Intellectual Property Rights
IV. Critical Analysis of the Italian legal Regime for the Taking of Security over Intellectual Property Rights
V. Legal Reform
15
Reverse Engineering the Law: Reforming Secured Transactions Law in Italy
I. Introduction
II. Secured Transactions Law Reforms: Common Roots, Different Routes
III. From Legal Functionalism to Reverse Engineering
IV. From Reverse Engineering to Re-Engineering: Implementing the UNCITRAL Registry System in the Italian Legal System
V. Conclusion
Index

Recommend Papers

Secured Transactions Law Reform in Africa 9781509913077, 9781509913107, 9781509913084

Over the last few decades, many countries have reformed their secured transactions law. One of the main reasons has been

149 82 10MB Read more

The Method and Culture of Comparative Law: Essays in Honour of Mark Van Hoecke 9781474201636, 9781849466233

Awareness of the need to deepen the method and methodology of legal research is only recent. The same is true for compar

166 36 5MB Read more

Reasons and Context in Comparative Law: Essays in Honour of John Bell 1009246372, 9781009246378

In honour of the work and writings of Professor John Bell, leading scholars present essays on factors affecting the cour

159 41 2MB Read more

Crime, Procedure and Evidence in a Comparative and International Context: Essays in Honour of Professor Mirjan Damaška 9781472564528, 9781841136820

This book aims to honour the work of Professor Mirjan Damaška, Sterling Professor of Law at Yale Law School and a promin

145 2 2MB Read more

Secured Transactions Law in Asia: Principles, Perspectives and Reform 9781509926497, 9781509926527, 9781509926503

This collection of essays offers a unique insight and overview of the secured transactions law in many of the most impor

148 13 6MB Read more

Law, Policy, and International Justice: Essays in Honour of Maxwell Cohen 9780773564275

Law, Policy, and International Justice is a collection of essays published in honour of Judge Maxwell Cohen. As a law pr

105 47 27MB Read more

Availability of Credit and Secured Transactions in a Time of Crisis 9781107703704, 9781107027442

In the light of the financial crisis, it has become clear that the globalisation of financial markets has not been match

160 98 3MB Read more

Contemporary Developments in International Law: Essays in Honour of Budislav Vukas 9789004245310, 9004245316

For the Liber Amicorum, dedicated to Professor Budislav Vukas, his colleagues and former students have contributed essay

98 64 5MB Read more

Man's Inhumanity to Man: Essays on International Law in Honour of Antonio Cassese (International Humanitarian Law) 9041119868, 9789041119865

This volume contains a unique collection of essays on various aspects of current interest within the field of public int

119 99 132MB Read more

The Achievements of International Law: Essays in Honour of Robin Churchill 1509917373, 9781509917372

The aim of this collection of essays in Robin Churchills honour to discuss the key examples of the achievements of inter

384 84 7MB Read more

International and Comparative Secured Transactions Law: Essays in honour of Roderick A Macdonald
9781849467650, 9781782257844, 9781509901142

Author / Uploaded
Spyridon V Bazinas
N Orkun Akseli (editors)

0 0 0
Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

File loading please wait...

Citation preview

INTERNATIONAL AND COMPARATIVE SECURED TRANSACTIONS LAW The law of secured transactions has seen dramatic changes in the last decade. International organisations, particularly the United Nations Commission on International Trade Law (UNCITRAL), have been working towards the creation of international legal standards aimed at the modernisation and harmonisation of secured financing laws (eg, the United Nations Convention on the Assignment of Receivables in International Trade, the UNCITRAL Legislative Guide on Secured Transactions and its Intellectual Property Supplement, the UNCITRAL Guide on the Implementation of a Security Rights Registry and the UNCITRAL Model Law on Secured Transactions). The overall theme of this book is international (or cross-border) secured transactions law. It assembles contributions from some of the most authoritative academic voices on secured financing law. This publication will be of interest to those involved in secured transactions around the world, including policy-makers, practitioners, judges, arbitrators and academics.

ii

International and Comparative Secured Transactions Law Essays in honour of Roderick A Macdonald

Edited by

Spyridon V Bazinas and N Orkun Akseli

OXFORD AND PORTLAND, OREGON 2017

Hart Publishing An imprint of Bloomsbury Publishing Plc Hart Publishing Ltd Kemp House Chawley Park Cumnor Hill Oxford OX2 9PH UK

Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

www.hartpub.co.uk www.bloomsbury.com Published in North America (US and Canada) by Hart Publishing c/o International Specialized Book Services 920 NE 58th Avenue, Suite 300 Portland, OR 97213-3786 USA www.isbs.com HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2017 © The editors and contributors severally 2017 The editors and contributors have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as Authors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the authors, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www. nationalarchives.gov.uk/doc/open-government-licence/version/3) except where otherwise stated. All Eur-lex material used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2017. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: HB: 978-1-84946-765-0 ePDF: 978-1-50990-114-2 ePub: 978-1-50990-115-9 Library of Congress Cataloging-in-Publication Data Names: Bazinas, Spyridon V., editor. | Akseli, N. Orkun (Nazim Orkun), editor. Title: International and comparative secured transactions law / Edited by Spyridon V. Bazinas and Nazmi Orkun Akseli. Description: Portland, Oregon : Hart Publishing, 2017. | Includes bibliographical references and index. Identifiers: LCCN 2017024445 (print) | LCCN 2017026449 (ebook) | ISBN 9781509901159 (Epub) | ISBN 9781849467650 (hardback : alk. paper) Subjects: LCSH: Security (Law) Classification: LCC K1100 (ebook) | LCC K1100 .B39 2017 (print) | DDC 346.07/4—dc23 LC record available at https://lccn.loc.gov/2017024445 Typeset by Compuscript Ltd, Shannon To find out more about our authors and books visit www.hartpublishing.co.uk. Here you will find extracts, author information, details of forthcoming events and the option to sign up for our newsletters.

PREFACE

We had hoped to complete this book and present it to Rod. Unfortunately, the ultimate ferryman beat us to it. Yet, we decided to complete the book. We all, editors and contributors, agreed that Rod had earned this post-mortem honour and we all wanted to say farewell to him. Personally, I feel I owe Rod this foreword. I first met Rod in 2002, when he joined the Canadian delegation to the UNCITRAL Working Group VI (Security Interests), of which I was and still am the Secretary. His whole appearance and demeanour was that of an extremely intelligent, kind and modest person. His interventions during our Working Group sessions quickly earned him the position of an opinion leader and a consensus builder. With these characteristics, he also became a key member of a small informal group of experts who used to advise the UNCITRAL secretariat as to how to draft to implement the decisions of the Working Group. The policy discussions at the Working Group sessions were often difficult because we had to deal with complex issues arising in the context of the various legal systems of the world and in the six official languages of the United Nations (Arabic, Chinese, English, French, Russian and Spanish). The drafting discussions at the informal expert group meetings were not less difficult. While we had only to find the appropriate words to implement the decisions of the Working Group, we often discovered new policy issues. So, notes raising new issues became a regular feature of our working papers. And we could always rely on Rod’s assistance for a balanced formulation of those notes, as a result of which the Working Group was able to resolve many important issues. During Working Group sessions, the Chairman had to summarise discussions, draw conclusions and reach decisions. And he or she could always count on Rod’s support to reach consensus and spell it out in a generally acceptable way. During informal expert group meetings, I had to chair the discussion, summarise and reach conclusions. And I could always count on Rod’s support to succeed in performing these most difficult tasks. Here are some concrete examples of Rod’s contribution to our work on secured transactions. The UNCITRAL Legislative Guide on Secured Transactions (2007), of which he was one of the main draftsmen, would have never been completed if Rod had not helped us draft in particular the chapter on acquisition financing. The flexibility of this chapter reflects Rod’s skills as a comparative law lawyer and one who knows, not only how to explain complex points of law, but also how to listen and address the concerns of others. Our exchanges on drafting

vi

Preface

or policy matters during the time I had to implement the final decisions of the Commission with respect to the Guide will remain one of my most cherished memories of Rod and my work in UNCITRAL. He had to leave the Working Group in 2010 initially to focus on his work as a newly-appointed Commissioner in the Construction Corruption Commission and ultimately on the fight for his life. He did so with honour and bravery. And he never missed a chance to respond to my many questions and engage in discussions, which for me was always a rare intellectual pleasure. In 2013, he even participated in one of the conference calls I organised to assist us with a most difficult issue, the effectiveness of amendment or cancellation notices that are not authorised by the secured creditor. I initially challenged part of his text and we had an unforgettable discussion, the result of which became the basis for the Working Group to reach consensus. With very few changes, this text was finally adopted in the UNCITRAL Guide on the Implementation of a Security Rights Registry (2013). The last message I received from Rod was on 15 May 2014 (less than four weeks before the day he passed away) and it was an effort to help me address concerns raised by another delegate about the placement in what later became the UNCITRAL Model Law on Secured Transactions of the chapter on the rights and obligations of third-party obligors and the question whether the relevant rules should be regrouped at a higher level of generality, which we ultimately did. Rod was not only a mentor for many in Working Group VI, including me. He was also a cherished friend and esteemed colleague of all. We all learnt from him and enjoyed his company. While he questioned everything, he also had a deep faith, and at time we had discussions about spiritual matters. Rod may have closed his bodily eyes in this vain world but has opened the eyes of his soul in a better world. While we miss him, with this book in particular, we celebrate his life and the difference he made in ours. We are all grateful to him and to UNCITRAL which brought us together. Spyridon V Bazinas February 2017

TABLE OF CONTENTS

Preface��v List of Contributors�� ix Table of Legislation�� xi Introduction�� xxvii 1. Secured Credit Legislation: Functionalism or Transactional Co-Existence��1 Michael Bridge 2. ‘Functional Formalism’ in the Treatment of Leases under Secured Transactions Law: Comparative Lessons from the Canadian Experience��25 Catherine Walsh 3. Reflections on Misgivings about a Model Law��41 Neil B Cohen 4. The UNCITRAL Model Law on Secured Transactions��55 Spyridon V Bazinas 5. Non-assignment Clauses and their Treatment under UNCITRAL’s Secured Transactions Law Instruments��77 N Orkun Akseli 6. Dealing with Concepts of Property in the UNCITRAL Legislative Guide on Secured Transactions and UNCITRAL Model Law on Secured Transactions (2016)��95 Steven O Weise 7. Current Issues in Cross-Border Asset-Based Lending: Lessons from the Field on the Need for Secured Transactions Reform��101 Richard M Kohn 8. The Rights and Obligations of the Parties to the Security Agreement According to the UNCITRAL Model Law on Secured Transactions��119 Bénédict Foëx 9. Sûretés Mobilières et Stocks : ou l’Art et la Manière de Résoudre la Quadrature du Cercle��133 Jean-Francois Riffard

viii

Table of Contents

10. Some Thoughts (and Facts) about the Process of Secured Transactions Law Reform, with Special Emphasis on Registration, the Key to Achievement of Reform’s Goals��157 Marek Dubovec and Harry C Sigman 11. Conflict-of-Laws Rules on Security Rights in Non-Intermediaries Securities�� 185 Michel Deschamps 12. Comparative Approaches to the Enforcement of Secured Credit in Insolvency��207 José M Garrido and Edwin E Smith 13. Comparative Study on Indian Secured Transactions Law and the UNCITRAL Legislative Guide on Secured Transactions��239 Madhukar R Umarji 14. Security Interests over Intellectual Property Rights in Italy: Critical Analysis and Reform Proposals��259 Andrea Tosato 15. Reverse Engineering the Law: Reforming Secured Transactions Law in Italy��285 Giuliano G Castellano

Index��327

LIST OF CONTRIBUTORS

(in alphabetical order) Dr N Orkun Akseli Associate Professor of Commercial Law, Durham University, Durham, United Kingdom. Mr Spyridon V Bazinas Senior Legal Officer, International Trade Law Division (UNCITRAL Secretariat), Office of Legal Affairs, United Nations, Secretary of the UNCITRAL Working Group VI (Security Interests) and Lecturer, University of Vienna Law School, Vienna, Austria. Professor Michael Bridge Cassel Professor of Commercial Law, London School of Economics, London, United Kingdom and Professor of Law, National University of Singapore. Dr Giuliano Castellano Assistant Professor of Law, University of Warwick, School of Law, Research Associate, i3-CRG, École polytechnique, CNRS (Université Paris-Saclay), and member of the Italian delegation to the UNCITRAL Working Group VI (Security Interests). Professor Neil B Cohen Jeffrey D Forchelli Professor of Law, Brooklyn Law School, New York, NY, USA and member of the USA delegation to the UNCITRAL Working Group VI (Security Interests). Mr Michel Deschamps McCarthy Tetrault, Montreal, member of the Canadian delegation to the UNCITRAL Working Group VI (Security Interests) and Professor of Law, Faculty of Law, University of Montreal, Montreal, Quebec, Canada. Dr Marek Dubovec Senior Research Attorney, National Law Center for Inter-American Free Trade, Representative of the Center to the UNCITRAL Working Group VI (Security Interests) and Part-time Professor of Practice, University of Arizona, James E. Rogers College of Law, Tucson, AZ, USA.

x

List of Contributors

Professor Benedict Foex Professor of Law, University of Geneva Law School, Geneva, Switzerland, member of the Swiss delegation to the UNCITRAL Working Group VI (Security Interests) and of Counsel, Schellenberg Wittmer, Geneva. Mr Jose Maria Garrido Professor of Commercial and Corporate Law, University of Castilla-La Mancha (Spain) and Senior Counsel, International Monetary Fund (IMF), Washington DC. Mr Richard M Kohn Goldberg Kohn Ltd., Chicago, IL, USA; Co-General Counsel, Commercial Finance Association (CFA), New York, NY, USA; representative of CFA to the UNCITRAL Working Group VI (Security Interests). Professor Jean Francois Riffard Professor of Law, Université D’Auvergne—Clermont I, France and member of the French delegation to the UNCITRAL Working Group VI (Security Interests). Mr Harry Sigman Member of California Bar, Los Angeles, CA, USA; Representative of the National Law Center for Inter-American Free Trade to the UNCITRAL Working Group VI (Security Interests); former member of the USA delegation to the UNCITRAL Working Group VI (Security Interests). Mr Edwin E Smith Morgan, Lewis & Bockius LLP, New York. NY, and Boston, MA, USA; former member of the USA delegation to the UNCITRAL Working Group VI (Security Interests), and Lecturer in Law, Morin Center for Banking Law Studies, Boston University Law School, Boston, MA, USA. Dr Andrea Tosato Assistant Professor of Law, University of Nottingham Law School and member of the Italian delegation to the UNCITRAL Working Group VI (Security Interests). Mr Madhukar R Umarji Former Chief Legal Adviser, Indian Banks Association, Mumbai, India. Professor Catherine Walsh Professor of Law, Faculty of Law, McGill University, Montreal, Quebec, Canada and member of the Canadian delegation to the UNCITRAL Working Group VI (Security Interests). Mr Steven O Weise Proskauer Rose LLP, Los Angeles, CA, USA; Representative of the American Bar Association to the UNCITRAL Working Group VI (Security Interests).

TABLE OF LEGISLATION

International Basel Accords��314 Berne Convention for the Protection of Literary and Artistic Works, Art 5(2)��283 Cape Town Convention on International Interests in Mobile Equipment 2001�� 3, 23, 66, 160, 170–1, 297 Art 2(2)��23 Art 2(4)��23 Art 2(5)��23 Art 5(3)��23 Art 8��23 Art 10��23 Art 12��23 Art 14��23 Art 31��66 Art 38(2)��2 Art 54��23 Art 60(1)��170 Protocol on Matters Specific to Aircraft Equipment��23, 160, 170 Convention on the International Sale of Goods (CISG), Art 7��60 Convention Providing a Uniform Law for Bills of Exchange and Promissory Notes 1930��292 Geneva Securities Convention (GSC)��186, 188–90 Art 1��186 Art 1(a)��189 Art 31��190 Hague Securities Convention (HSC)��157, 186, 188–9 Art 1��186 OAS Model Law on Secured Transactions��45, 66, 124 Art 33��124, 126, 129 Art 68��59 OAS Registry Regulations��66 OHADA Model Law on Secured Transactions��45 UN Convention on International Bills of Exchange and International Promissory Notes 1989��292 UN Convention on the Assignment of Receivables�� 2, 75, 78, 90–2, 157, 194, 246 Art 1(a)��90 Art 2��194

xii

Table of Legislation

Art 3��90 Art 7��60 Art 9�� 85, 89, 92, 108 Art 9(1)��90 Art 9(2)��91 Art 9(3)��92 Art 10(3)��80, 91 Art 18(3)��91 Art 40��92 UNCITRAL Model Law on Secured Transactions�� 2, 16, 26–9, 34, 38–9, 41–53, 55–76, 78, 85, 89, 92, 95–6, 99, 102, 106, 108–9, 119–32, 135, 140, 144, 146, 148, 152, 157, 159, 163–7, 169, 172–4, 178–81, 183, 186, 188–93, 198, 201, 203–4, 206, 240, 285, 291, 296–7, 299–300, 307–8, 319, 321, 323 Art (6)��57 Art 1��28, 56 Art 1(5) and (6)��61, 90 Art 1(6)��58 Arts 1–5��120 Art 2�� 120, 297, 299, 308 Art 2(1)��121 Art 2(1)(o)��120 Art 2(11)��121, 188 Art 2(b)��29, 109 Art 2(bb)��62, 141 Art 2(dd)��121 Art 2(e)��69, 190 Art 2(g)(i)��65 Art 2(ii)��130, 204 Art 2(jj)��121, 186 Art 2(kk)��28, 109, 129 Art 2(w)��186 Art 2(z)��121, 124 Art 3�� 28, 120, 123, 191 Art 3(1)�� 59–60, 72, 128, 131 Art 3(3)��59 Art 4��59, 72, 120, 124, 127, 129, 131 Art 4(1)��120 Arts 4–6��28 Art 5��60 Art 5(2)��167 Art 6(1)��128 Art 6(1) and (2)��60 Art 6(3)��60 Art 6(3)(b)��128 Art 6(4)��61

Table of Legislation

xiii

Art 7��61, 128, 131 Art 8��61, 131 Art 8(d)��126 Art 9��62 Art 10��140 Art 10(1)��62 Art 10(c)��143 Art 11��149, 153 Art 11(2)��149 Art 11(3)��149 Art 12��127 Art 13��89, 92, 108 Art 13(1)��62, 85, 90 Art 13(2)��91 Art 13(3)��92 Art 14��63, 96–7 Art 16��63 Art 17��63 Art 18(1)��64–5, 127 Art 18(2)��64 Art 18(ff)��130 Art 19(1)��64, 144 Art 19(2)��145 Art 23��205 Art 25��65 Art 26��65, 321 Art 27��65 Art 28��65 Art 29��69 Art 30��69 Art 33��149, 153 Art 34�� 69, 98–9, 153, 321 Art 34(2)��127 Art 34(4)��153 Art 35��57, 69 Art 36��57, 69, 110 Art 37��70, 180 Art 38�� 29, 57, 70, 146, 153 Art 38(1)(b)��173 Art 39��57 Art 39(1)��70 Art 41��70 Art 43��71 Art 46(1)��70 Art 47��70 Art 48��71 Art 49��71

xiv

Table of Legislation

Art 51��71 Art 52�� 121–2 Arts 52–56��120–1, 131 Art 53��72, 121–6, 128 Art 54��72, 121–2, 126–8 Art 55��121–2, 130 Art 55(1)�� 128–9 Art 55(1)(a)��128 Art 55(1)(b)��129 Art 55(2)��121, 128, 130 Art 56��122, 131 Art 56(1)��131 Art 56(2) and (3)��131 Arts 57–60��120 Art 58��72 Arts 61–71��74 Art 64��72 Art 71��72 Art 72(3)��72 Art 76��59 Art 77��125 Art 77(2)��59, 73 Art 78(4)��59 Art 78(4)–(8)��73 Art 79��73 Art 79(2)(c)��59 Art 80(2) and (3)��73 Art 81(3)��73 Art 81(4)��73 Art 82(1)��73 Art 82(2)��74 Art 82(3)��74 Art 82(4)��74 Art 82(5)��74 Art 83��74 Art 85(1) and (2)��74 Art 85(3)��75 Art 85(4)��75 Art 86��75 Art 88(a)��75 Art 90��75 Art 91��205 Art 95��191, 204 Art 97��75 Art 98��206 Art 99��76 Art 100��202–4, 206 Registry Provisions��76, 164, 166

Table of Legislation

xv

Arts 1–29��119 Art 5��168 Art 8(1)��175 Art 8(b)��67 Art 9��66, 175 Art 11��67 Art 14��182 Art 18��125 Art 21��68 Art 22��67 Art 22(b)��67 Art 23��177 Art 25��177 Art 26�� 68–9 Art 31��68 Art 32��68 Art 33��181 Art 34��69 UNIDROIT Convention on International Financial Leasing 1988��316 UNIDROIT Principles for International Commercial Contracts��4, 83 Art 9.1.9��4, 83 Australia Corporations Act 2001, s 471C��213 Personal Property Securities Act 2009��10, 159, 161, 163, 174, 178, 183 s 12(1)��10 s 44��179 s 153(1)��179 Belgium Law of 11 July 2013, art 26��138 Brazil Bankruptcy Law Art 6��231 Arts 41–46��234 Art 48(3)��212 Art 58�� 234–5 Art 83��227 Art 85��212 Civil Code, Arts 80–83��212 Canada Federal Bankruptcy and Insolvency Act 1985��20, 36 s 2��36 s 65.1(4)(a)��37

xvi

Table of Legislation

Companies’ Creditors Arrangements Act 1985�� 36, 230, 232, 237 s 2��36 s 11.01��37 s 34(4)(a)��37 Constitution Act 1867, s 91(21)��36 Alberta Fair Trading Act 2000 ss 90–93��33 s 94��33 Personal Property Security Act 1990��25 British Columbia Business Practices and Consumer Protection Act 2004 s 77��33 ss 100–101��33 s 102��33 Personal Property Security Act 1996��20, 25 s 1(1)��29 s 3��29 s 41(9)��78 s 67��32 s 67(1)(c)��20 s 71(2)��20 Manitoba Consumer Protection Act 2000 ss 36–39��33 s 40��33 Personal Property Security Act 1978��25 New Brunswick Personal Property Security Act 1995��25, 168 Newfoundland and Labrador Personal Property Security Act 1999��25 Northwest Territories Personal Property Security Act 2001��26 Nova Scotia Personal Property Security Act 1997��25 Nunavut Personal Property Security Act 2001��26 Ontario Personal Property Security Act 1976��25 Personal Property Security Act 1990�� 11, 29, 31–2, 161, 197–8 s 1(1)��28 s 2��29 s 2(a)��28 s 2(a)(ii)��11 s 7.1��194 s 7.1(1)(a)��196 s 7.1(1)(b)��199

Table of Legislation

xvii

s 7(3)��197 s 8(1)��201 s 17(3)��11 s 33��29 s 39��78 Securities Transfer Act 2006��187, 189, 199 s 1(1)��187, 189 s 44(5)��199 ss 44–46��194 s 95(1)��187 s 95(3)��188 Prince Edward Island Personal Property Security Act 1998��25 Quebec Civil Code�� 25–6, 34, 36, 39, 186, 188–9, 191, 195–6 Art 7��36 Art 307��198 Art 1263��34 Art 1375��36 Art 1437��36 Art 1745��34 Art 1749��34 Art 1750��34 Art 1756��34 Art 1842��35 Arts 1842–1850��35 Art 1847��35 Art 1851��35 Art 1852��35 Art 2660��34 Art 2674��140, 142 Art 2954��34 Art 3077��191 Art 3103��199 Art 3108��196 Consumer Protection Act��35 Securities Transfer Act 2008��187 Saskatchewan Personal Property Security Act 1993�� 10, 19, 25, 79 s 2(1)(qq)(ii)(C)��19 s 3(1)��19 s 3(1)(a)��10 s 3(2)��19, 297 s 20(2)��19 Yukon Territory Personal Property Security Act 1982��25

xviii

Table of Legislation

China Bankruptcy Law Art 37��215 Art 75��230, 232, 237 Art 82��234 Art 109��223 Costa Rica Law No 9246��178 European Union Capital Directive, Art 23��113 Capital Requirements Regulation, Art 210(d)��315 Community Trademark Regulation Art 19��277 Art 24��277 Draft Common Frame of Reference�� 2, 27, 45, 163 Art IX—3:301��160 EBRD, Model Law on Secured Transactions��16, 45 Art 6.1��297 Art 6.2.2��297 Art 6.4.2��297 Art 8.4��308 Art 9��297 Financial Collateral Directive��9, 263, 316 Quality Schemes for Agricultural Products and Foodstuffs Regulation��312 Rome I Regulation, Art 3��274, 280 France Civil Code��150, 288 art 2093��303 art 2277��138 art 2337��137 art 2342��137 art 2372��142 Civil Procedure Code, art R162–9��143 Commercial Code art L257–6��154 art L257–7(3)��154 art L524–1 to L524–21��137 art L525–1 to L524–20��137 art L527–1�� 150–1 art L527–2��139 art L622–7–I–3º��227 art L622–23–1��231 art L642–20–1–1º and 2º��227

Table of Legislation

xix

Law of 23 May 1863��137 Law of 2 January 1981��158 Law of 25 January 1985 art 47��214 art 78��227 art 93�� 226–7 art 159��227 art 161��217 Monetary and Financial Code, arts L313–23 to 313–35��158 Ordonnance No 2016–56��150 Rural Code, art L342–1 to 342–17��137 Germany Act on Limited Liability Companies, s 30��106, 111 Civil Code (BGB)��261, 288 Commercial Code, s 354(a)��79 Insolvency Code��109 ss 49–52��212 ss 170–171��109 s 171��225 India Arbitration & Conciliation Act 1996��240 Code of Civil Procedure 1908, s 60��250 Constitution��247, 254 Contract Act 1872�� 240, 245, 247, 249 s 172��250 Copyright Act 1957��240 Depositories Act 1996��245 Designs Act 2000��240 Factoring Regulation Act 2011��240, 245–6 Forward Contracts (Regulation) Act 1952��246 Information Technology Act 2000��240 Merchant Shipping Act 1958��240 s 3(55)��250 Motor Vehicles Act 1988��240 Negotiable Instruments Act 1881��245, 247 Patents Act 1970��240 Sale of Goods Act 1930��249 s 47��250 Securities and Exchange Board of India Act 1992��246 Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest Act 2002�� 241, 245, 248, 258 s 2(1)(zd)�� 255–6 s 31(e)��256 Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest Ordinance 2002��243

xx

Table of Legislation

Trade Marks Act 1999��240 Transfer of Property Act 1882��240, 247 Warehousing (Development and Regulation) Act 2007��245, 247 Italy Civil Code��261–5, 267, 269, 271–3, 275, 277, 288, 303, 309, 311, 314, 317, 319, 324 Art 1153��263, 268, 315 Art 1154��263 Art 1156��315 Art 1186��263, 275 Art 1523��263 Art 1524��268, 313 Art 2023��309 Art 2082��267 Art 2740��262 Art 2741��262, 303, 310 Art 2741.1��262 Art 2741.2��262 Art 2743��263, 266, 275 Art 2744��263, 316 Art 2746��262 Art 2748��310 Art 2748.1��268 Art 2751bis��311 Art 2777��311, 314–15 Art 2777.3��268 Art 2784��264, 316 Art 2784.1��271 Arts 2784–2807��263 Art 2786��264, 276, 309–10 Art 2787��264 Art 2787.3��264 Art 2789��265 Art 2791��264 Art 2792��265 Art 2795��265, 275 Art 2800 et seq��316 Art 2806��265, 272–3 Art 2806.1��264–5, 271, 277 Art 2806.2��264 Art 2808��262 Arts 2808–2899��266 Art 2809��262 Art 2810��266, 271, 311 Art 2811��266 Art 2813��267

Table of Legislation

xxi

Art 2821��266 Art 2823��266 Arts 2838–2839��266 Art 2843��262, 267 Art 2847��266 Art 2851��266 Art 2852��262, 267 Art 2900��263, 265, 275–6 Art 2901��263 Codice della Proprietà Industriale (CPI)��269, 271–3, 277–8 Arts 78–79��269 Art 104.1��269 Arts 137–140��269 Art 138�� 269, 273, 277, 279 Art 138.3��269 Arts 138–140��269, 277–9 Art 139.1��269 Arts 139–140��271 Art 140�� 277–8 Art 140.1��283 Art 140.2��269 Art 140.3��269 Art 195��269 Arts 195–196��269 Art 196��269 Constitution, Art 77��318 D.lgs 1 September 1993 (TUB)��267–8, 314, 316–17 Art 46�� 267, 273, 313–15, 317, 322 Art 46.1��267–8, 271–2, 278, 314 Art 46.1–bis��268 Art 46.2��268 Art 46.3��268 Art 46.4��268, 315 Art 46.5�� 268, 272, 315, 320 Art 46.6��314 D.lgs 24 February 1998��317 D.lgs 24 June 1998��317 D.lgs 4 July 2006 (Bersani Decree)��312 Law of 22 April 1941 (LA)��270, 272–3, 278–9, 283 Art 103��270 Art 104��270, 272 Art 107��270 Art 111�� 270, 272, 278, 283 Art 112��278 Law of 24 July 1985��3, 264, 312 Law of 14 July 1993��316 Law of 31 May 1995 (PILR)��273

xxii

Table of Legislation

Art 51��273 Art 54�� 273–4 Law of 27 March 2001, Art 7��312 Law of 30 June 2016��318 Art 1��319 Maritime and Navigation Code Art 565��309 Art 1027��309 Royal Decree of 29 March 1942��317 Jamaica Security Interests in Personal Property Act 2013��49 Malaysia Bankruptcy Act 1967, Schedule C��222 Mexico Commercial Code, Arts L625–7 and L625–8��224 Ley de Concursos Mercantiles Art 214��216 Art 224��237 Art 224.1��237 Art 225��237 Netherlands Civil Code 1838��288 New Zealand Personal Property Securities Act 1999��162, 165, 178–9 s 17(1)(a)��10 s 17(3)��22 Nigeria Registry Regulations, s 2(1)��178 Russia Bankruptcy Law, Art 131��213 Sierra Leone Borrowers and Lenders Act 2014, s 10��172 Spain Ley Concursal Art 56��218 Art 59��231

Table of Legislation

xxiii

Art 59.1��219 Art 90.3��223 Art 134.2��234 Switzerland Code of Obligations, Art 164��79 Thailand Bankruptcy Act s 90/28�� 230–1 s 90/42��234 s 96��221 s 110��215 Turkey Code of Obligations, Art 162(1)��79 United Kingdom Companies Act 1985��8 Companies Act 1989, Pt IV��8 Companies Act 2006��8 s 859A��8 s 859D(2)(c)��9 s 859H��18 Enterprise Act 2002��7, 224 Financial Collateral Arrangements (No 2) Regulations 2003, SI 2003/3226��9 Fraudulent Conveyance Act 1571��288 Insolvency Act 1986��4 Insolvency Act 1986 (Prescribed Part) Order 2003��224 Insolvency Act 1986 s 175��4, 293 s 176��293 s 176A��4, 224 s 176ZA��4 ss 238–245��4 s 283��17 s 386��4 ss 423–425��4 Sch 6��4 Sch B1 para (3)��7 para 65(2)��4 Law of Property Act 1925, s 136��80 Pension Schemes Act 1993, s 159��80 Pensions Act 1995, s 91��80 Social Security Administration Act 1992, s 187��80 Third Parties (Rights against Insurers) Act 2010��4

xxiv

Table of Legislation

United States Bankruptcy Code�� 111, 213, 219–20, 222, 225, 227, 232–6 § 361��220 § 362(d)��220 § 362(d)(1)��219 § 362(d)(2)��219 § 364(d)��236 § 506(a)(1)��222, 228 § 506(b)��222 § 506(c)��225 § 541��213 § 1122��233 § 1124��233 § 1126(c)��233 § 1126(f)��233 § 1129��233 Exchange Act 1934�� 295–6 Federal Tax Lien Act��180 Securities Act 1933��295 Uniform Commercial Code (UCC)�� 5, 10, 13, 16, 18–19, 28, 31, 44, 78, 86, 91, 95, 138, 145, 164, 172, 186–91, 193, 195–6, 198–200, 202, 204, 289–90 § 1–201(35)��16, 28 § 1–201(b)(35)��20, 108 § 1–203��19–20, 31 § 1–203(a)��31 § 1–203(c)��19 § 1–301(2)��191 § 2–401��16 § 2–401(1)��108 § 2–403��14 § 2–709��5 § 8–102��187 § 8–102(a)(9)��187 § 8–110��194 § 8–501(b)��187 § 8–501(d)��188 § 9–102(a)(20)��21 § 9–102(a)(52)��199 § 9–103��50 § 9–103(b)(2)��146 § 9–103(d)��22 § 9–108(b)��164 § 9–109(a)(1)��21 § 9–109(d)(10)��13 § 9–110��16 § 9–203(b)(2)��14 § 9–301��201

Table of Legislation

xxv

§ 9–301(1)��196 § 9–305��194 § 9–306��142 § 9–307��198 § 9–315 (b)(2)��142 § 9–317(a)(2)��15, 18 § 9–320(a)��16 § 9–336��148 § 9–342(g)(1)��17 § 9–401��79 § 9–401(2)��86 § 9–402��79, 91 § 9–406��86, 90–1 § 9–406(d)�� 79, 86, 90, 107 § 9–406(d)(2)��90 § 9–406(e)��86 § 9–408��86 § 9–408(a)��79 § 9–408(a)(2)��90 § 9–502(a)(2)��115 § 9–504��164 § 11–548��111 § 931–5(a)(2)��142 Art 1��16 Art 8��186, 189 Art 9��1–3, 7–8, 10, 13, 15–19, 21–2, 28, 31, 51, 66, 78–9, 83, 86, 90, 137–8, 142, 148, 157–8, 164–5, 169–72, 186, 281, 289–90, 297–8, 300, 308 Uniform Trust Receipts Act 1933��289, 308

xxvi

INTRODUCTION

SPYRIDON V BAZINAS AND N ORKUN AKSELI

This book focuses on one of the fields of law that Roderick A Macdonald served with great distinction: international and comparative secured transactions law, as developed in the work of UNCITRAL from 2002 to 2010. Throughout the world, secured transactions law is out of date. Markets simply run faster than the law. Legislators and courts have to keep up, but they are always left behind market developments. In addition, as they try to keep up with market developments, legislators and courts increase the multiplicity of rules, and thus decrease the certainty and transparency of the law. This process also creates gaps and inconsistencies in the law. In some countries, lawyers, courts, markets and the other relevant infrastructure reduces the negative impact of these gaps and inconsistencies on the availability and the cost of credit. This does not mean, though, that there is no problem. It simply means that we often cannot assess with accuracy the economic impact of those gaps and inconsistencies. A simple way to verify the accuracy of these assertions is to take one or both of the following steps. First, one may look at the multiplicity of laws and registries applying to interests in personal property performing security functions. It is not easy to have an overview of the relevant rules even in just one and the same jurisdiction. Second, one may collect all the relevant rules in one document. The complexity, the gaps and the inconsistencies among those rules, and at the same time the need for reform, becomes immediately apparent. The purpose of this book is to provide an analysis of the relevant issues and make a case for law reform. It begins with a couple of chapters discussing the functional approach to secured transactions (Bridge and Walsh). Bridge shows that that the functional approach to secured transactions ‘does not expunge difficult proprietary issues’ and does not mean that ‘all transactions should be treated in an identical way just because they fulfil in economic terms a security function’. Walsh focuses on the issue of whether, and to what extent, leases that function to finance the acquisition of ownership of the leased asset by the lessee should be subjected to the same rules that apply to economically equivalent secured transactions, and demonstrates that an apparently formalistic approach may often simply reflect, in the words of Rod Macdonald, ‘not choices about formalism or functionalism, but rather choices about the location and level at which functional analysis should take place.’

xxviii

Introduction

The next couple of chapters discuss the UNCITRAL Model Law on Secured Transactions (the Model Law) (Cohen and Bazinas). Cohen discusses the misgivings that he and Rod had expressed with regard to the preparation of a model law on secured transactions at a 2010 colloquium of UNCITRAL and concludes that their ‘biggest concerns (the risks of relitigation, undermining of the value of the Legislative Guide, and the possible creation of disharmony) did not take hold’. Bazinas confirms this point by presenting the Model Law and comparing its provisions with the recommendations of the UNCITRAL Legislative Guide on Secured Transactions (the Legislative Guide) and other relevant UNCITRAL texts. A number of chapters follow, which discuss specific issues (Akseli, Weise, and Kohn). Akseli examines the approach to anti-assignment clauses followed in the UN Convention on the Assignment of Receivables in International Trade (the ‘Receivables Convention’), the Legislative Guide and the Model Law, arguing that there is a policy need to limit the effectiveness of contractual clauses restricting the assignment of receivables. Weise discusses one of the main challenges faced by UNCITRAL in preparing the Legislative Guide and the Model Law, that is, to develop a set of recommendations and provisions that accommodated both civil law and common approaches, in particular when considering questions of what constitutes ‘property’ for the purposes of applying the scope rules of the Legislative Guide and the Model Law. He concludes that the substance-over-form, functional approach ‘served as the basis for solving this apparent conundrum’. Kohn examines the role of secured transactions reform in promoting crossborder financing transactions, by examining secured transactions reform through the lens of asset-based lending and argues that, because of its intense focus on collateral, cross-border asset-based lending provides an ideal laboratory in which to investigate how the secured transactions regimes currently in effect in many countries negatively affect the cost and availability of credit, thereby making the case for secured transactions reform all the more compelling. These chapters are followed by two contributions on the rights and obligations of parties and on security interests in inventory (Riffard and Foëx), one translated from the French original and the other in French. Riffard discusses the challenge that a security interest in inventory and the proceeds of inventory pose for many jurisdictions and concludes that, by enacting a few specific rules and making some room for party autonomy, it is possible for those States to reconcile the need of the grantor to freely dispose of its inventory and of the secured creditor to repossess and sell the inventory in the event of default. Foëx focuses on the provisions of the Model Law on the rights and obligations of the parties to a security agreement and shows that, while these provisions have been carefully drafted and provide balanced solutions, some of them, such as Article 53 (obligation of the party in possession of the encumbered

Introduction

xxix

asset to exercise reasonable care) and Article 54 (obligation of the secured creditor to return the encumbered asset) should not have been formulated as mandatory law rules. Next, is a chapter discussing registration issues (Dubovec and Sigman). Dubovec and Sigman provide insights on the theory and practice of registration of notices of security interests as a means of achieving successful secured transactions law reform. They lay out the main characteristics of a modern efficient registry, and discuss several issues concerned with the process of secured transactions law reform and the Model Law. Like the Model Law, which contains a conflict-of-laws code dealing with the law applicable to security interests in personal property, this book contains a conflict-of-laws chapter (Deschamps). Deschamps compares in particular the conflict-of-laws rules of the Model Law that apply to security interests in nonintermediated securities with the relevant Canadian and US rules, and discusses their comparative advantages and disadvantages. Recognising the need for coordination of secured transactions law with insolvency law, this book includes a chapter on this matter (Garrido and Smith). Garrido and Smith examine selected issues arising in the case of the enforcement of a security interest in the case of insolvency, comparing the recommendations of the Legislative Guide (on Secured Transactions) and the UNCITRAL Legislative Guide on Insolvency Law with the relevant rules of various insolvency laws. The book closes with three chapters that discuss national secured transaction law issues, in view of the recommendations of the Legislative Guide and the provisions of the Model Law (Umarji, Tosato and Castellano). Umarji compares Indian secured transactions law with the recommendations of the Legislative Guide. He points out the fragmented nature of Indian secured transactions law and makes reform suggestions in the light of the recommendations of the Legislative Guide. Tosato examines the Italian law governing security interests in intellectual property and argues that it is in need of reform. He recommends either wholesale reform that overhauls the whole of Italian secured credit law, or piecemeal reform that introduces specific rules on security interests in intellectual property. Castellano examines the new Italian law on non-possessory pledges in the light of the UNCITRAL texts on secured transactions and suggests that further improvement of that law, in particular the regime applicable to registration of notices of security interests, is both desirable and feasible. With its discussion of the Legislative Guide, the Model Law, other relevant UNCITRAL texts and various national laws, this book is expected to become a valuable resource for those interested in secured transactions law research and reform throughout the world.

xxx

1 Secured Credit Legislation: Functionalism or Transactional Co-Existence MICHAEL BRIDGE*

I. Introduction A. General The purpose of this chapter is principally to highlight a select number of features of modern personal property security legislation in an attempt to show that functional legislation does not expunge difficult proprietary issues. A further purpose is to demonstrate that the inherently reductionist character of functional legislation does not mean that all transactions should be treated in an identical way just because they fulfil, in economic terms, a security function. The legislation that comes in for examination does not itself go quite so far, though it comes close. Its pursuit of identical treatment might yet have been more restrained without prejudice to the core purposes of the legislation. With the enactment in the various states of Article 9 of the Uniform Commercial Code in the 1950s, the United States demonstrated a revolutionary approach to the subject of secured credit. This approach involved embracing under one legislative roof all forms of security and related devices that served the common purpose of facilitating the recovery of debts from a disposal of personal property. It involved, too, an extensive measure of top-down standardisation with a common set of priority rules and a common set of creditor remedies, both personal and proprietary. Particularly striking, since it involved a departure from freedom of contract, was the treatment of all secured creditors as having the remedies of a mortgagee, even those supplying goods under a reservation of title scheme. The key characteristic of Article 9 is the functional thinking embedded in it that captures a wide range of * Michael Bridge FBA, Cassel Professor of Commercial Law, London School of Economics; Professor of Law, National University of Singapore.

2

Michael Bridge

transactions in order to bring them under a single statutory roof.1 Considerations of time and length prevent me from considering the full range of transactions that might be covered by functional legislation of this type. Hence, subjects such as Quistclose trusts,2 sale and leaseback agreements,3 priority (or inter-creditor) agreements4 and ‘repo’ transactions5 fall by the wayside. Article 9 was designed for both corporate and non-corporate debtors, and for commercial and consumer debtors. In its original version, it undoubtedly simplified a great deal of complex law, building on a large uncodified base and bringing together a range of statutes and registration systems dealing with charges, conditional sales and trust receipts, to name a few examples. This was no small achievement in any case, but was especially impressive in the case of a federal country where the law on secured transactions is state-based law. The Article 9 approach has been replicated in other jurisdictions, notably and at different intervals in the case of Canada, New Zealand and Australia. It has been influential in the drawing up of international conventions, such as the UN Convention on the Assignment of Receivables in International Trade 2001. The influence of Article 9 is also manifest in soft law instruments, such as the UNCITRAL Legislative Guide on Secured Transactions 2007,6 the UNCITRAL Draft Model Law on Secured Transactions7 and the European Draft Common Frame of Reference.8 To the above examples, one might add an instrument that demonstrates a debt to Article 9 but that does not follow it all the way. This is

1 See also MG Bridge, ‘The Scope and Limits of Security Interests’ (2008) European Company and Financial Law Review 180–214 (Special Volume: The Future of Secured Credit in Europe). 2 The lending of money dedicated in trust terms to a particular purpose, reverting to the creditor on a resulting trust if the purpose fails, taking its name from the leading case, Barclays Bank Ltd v Quistclose Investments Ltd [1970] AC 567. 3 These are more often found in the case of land where the freehold interest is conveyed to a financier who then grants back a leasehold interest. The financier may have its own motives for acquiring the freehold estate over and above the giving of financial accommodation to the lessee. This feature is absent in short-term transactions dealing with stock-in-trade: see Re Curtain Dream Ltd [1990] BCLC 925. 4 These are accepted in modern personal property security regimes (e.g. the Cape Town Convention 2001, Art 38(2)) but a question mark hangs over the mechanics of the transaction. Are they merely personal undertakings between two creditors (see, eg, Re Portbase Clothing Ltd [1993] Ch 388) or is something transferred from one creditor to the other, so as to attract, for example, the rules against divesting of a party’s assets on its insolvency (see, eg, Belmont Park Investments Ltd v BNY Corporate Trustee Services Ltd [2011] UKSC 38, [2012] 1 AC 383)? 5 The sale of fungible securities from A to B, to accommodate B’s financial needs, with a sale back, at an enhanced price, of equivalent, but not identical, securities: see Re Lehman Brothers International (Europe) [2011] EWCA Civ 1544, [2012] 2 BCLC 151. 6 United Nations Publication Sales No E.09.V.12. The text of the Legislative Guide is available at www.uncitral.org. The text was adopted by consensus on 14 December 2007: Sixty-second Session, Supplement No 17 (A/62/17, Part II) paras 99 and 100. The previous draft Guide was contained in documents A/CN.9/631/Add.1-3 and A/CN.9/637 and Add 1–8. 7 A/CN.9/WG.VI/WP.61. 8 See Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference (Sellier, Outline Edition 2009) Book IX (Proprietary Security in Movable Assets).

Secured Credit Legislation: Functionalism or Transactional Co-Existence 3 the Cape Town Convention on International Interests in Mobile Equipment 2001. It is the extent of its departure from Article 9 that interests me, partly because of a conviction that an attempt to sell the whole of the Article 9 package to a receiving country might fail, where a less ambitious attempt to sell certain of its essential parts might not.

B. Desiderata of a Modern Secured Transactions Regime The sought-after values of a modern secured transactions regime have been set out at some length in the UNCITRAL Legislative Guide on Secured Transactions. In contrast with the UNCITRAL recommendations contained in this text, one might characterise as essentially uncommitted to such a legal regime a State that permits, or indeed requires, the proliferation of different types of security, each with its own arcane rules and each frequently limited to a particular type of collateral,9 with, overall, certain types of collateral eluding the reach of security, the whole amounting to less than might be encompassed by a general enterprise charge.10 The UNCITRAL list is premised upon the inherent desirability of secured credit as an engine that facilitates the grant of credit and, by diminishing the financial risk associated with debtor default, reduces credit costs.11 One stated value is that of fairness to all parties affected by a secured transactions regime, expressed as a balancing of interests.12 It is well known that certain categories of creditor, particularly involuntary creditors such as tort claimants, are adversely affected by modern schemes if these are allowed to operate unchecked. Older schemes, such as the contrasting example in the previous paragraph, might be said systemically, even deliberately, to impose a check on efficiency in order to protect vulnerable creditors. It is not my purpose in this paper, however, to deal in any detail with this theme. That said, vulnerable creditors may, broadly, be protected by a legal system’s rules relating to secured transactions and insolvency in one of two ways: the first is by imposing limitations and restrictions on the taking of security (the concept of over-security in some legal systems13 is an example of this); and the second is by limiting the enforcement of security rights in the event of the debtor’s insolvency. These two ways can be implemented by the following separate methods. One method is by the use of clawback measures dealing with 9 A favourite example is the Italian security right over Parma ham (pegno non possessori sui prosciutti): Law of 24 July 1985, No 401. See the report on Italy (Veneziano) in H Sigman and E-M Kieninger (eds), Cross-Border Security over Tangibles (Sellier, 2007). 10 The process of reform may take the road of retaining existing partial coverage and supplementing it with a range of additional rights of varying dimensions: for the case of France, see L Aynès and P Crocq, Les Sûretés: La Publicité Foncière (4th edn, Defrénois, 2009); the French reports (Leavy) in Sigman and Kieninger (eds) (n 9) and H Sigman and E-M Kieninger (eds), Cross-Border Security over Receivables (Sellier, 2009). 11 Legislative Guide, Introduction, paras 5–6, 46 and 49. 12 Legislative Guide, Introduction, para 58. 13 See, eg, the German report (Rakob) in Sigman and Kieninger (eds) (n 9).

4

Michael Bridge

preferences, undervalue transactions, fraudulent conveyances and related matters in the twilight period when the individual is sliding into personal bankruptcy or the company into insolvent liquidation.14 Another method, adopted in particular by English law,15 even though that law is broadly friendly to the taking of security, goes beyond clawback measures. It does so by expropriating the holders of floating charges in favour of preferential16 and general unsecured creditors (up to a financial limit)17 in the event of distribution via either administration or standard liquidation proceedings.18 Collaterally, tort claimants may be protected in altogether different ways. For the most common tort claims, such as those made against negligent employers and drivers, compulsory insurance can be required by legislation with direct access by claimant to insurer if needed to avoid a windfall benefit to other creditors of the insured.19 The values mentioned by the UNCITRAL Guide that are intrinsic to the type of regime that enhances security and therefore the flow of credit are: the need for a simple and efficient system of security entitlement; access to credit deriving from a debtor’s ability to use all of its assets for this purpose; the existence of a clear set of priority rules; transparency and certainty in the shape of a general registry of security interests; equality of treatment of diverse types of transaction serving a common credit purpose; and the efficient enforcement of creditors’ rights.20 Of these values, my primary focus in this paper will be on the last two: how far should we go to embrace diversity of transactions? And how far does an effective system of creditors’ rights demand uniformity of rights? The latter value prompts the further question whether we should override freedom of contract between debtor and creditor in the cause of uniformity.21 In handling the various questions that arise out of these two latter values, the looming presence of ownership (or title) asserts itself. Is there something special 14 See, eg, Insolvency Act 1986, ss 238–45, 423–25. Unless otherwise stated, all references to legislation are to English or UK legislation. 15 In addition to clawback measures. 16 Insolvency Act 1986, ss 175 and 386, and Schs 6 and B1, para 65(2). 17 See the Insolvency Act 1986, ss 176A (creating a financially limited fund for unsecured creditors out of assets the subject of a floating charge), 176ZA and 386 (rights of preferential creditors to prior payment out of floating charge assets); H Beale and others, The Law of Security and Title-Based Financing (2nd edn, Oxford University Press, 2012) para 20.24ff. 18 In the process, floating charge assets are used also to fund the expenses of liquidation and administration. See Insolvency Act 1986, s 176ZA. 19 See, eg, Third Parties (Rights against Insurers) Act 2010, which entered into force on 1 August 2016. Attempts to achieve the same outcome by contract may fall foul of insolvency principles, as where a payment of moneys in settlement of a disputed claim under a liability insurance policy, conditional upon the moneys being paid out to the third party claimant, was held to infringe the rule that the assets of an insolvent person (here, the insured) should be preserved from its creditors and not divested upon its insolvency: Folgate London Market Ltd v Chaucer Insurance Plc [2011] EWCA Civ 328, [2011] BCC 675. 20 See Legislative Guide, Introduction, paras 46–59. 21 A consistent theme of modern personal property security instruments and uniform law instruments (eg, the Unidroit Principles of International Commercial Contracts 2010, Art 9.1.9) is the way that they override non-assignment clauses in favour of the free alienation of the creditor’s payment rights.

Secured Credit Legislation: Functionalism or Transactional Co-Existence 5 about ownership that prevents its full equation with lesser, or at least different, property interests for the purposes of a modern secured transactions regime? It is notorious that the leading proponent of the Uniform Commercial Code, Karl Llewellyn, was a title sceptic: ‘Nobody ever saw a chattel’s Title.’22 Article 2 of the Code, on sales, contains no rules on the passing of title (or property) between title and seller, instead defining the rights and remedies of buyer and seller without the aid of title.23 Whilst it may be possible to avoid mention of title between buyer and seller, this is far from being the case when third parties intrude. A security agreement concerns rights in collateral that affect the position of competing secured creditors, the trustee-in-bankruptcy or administrator or liquidator representing unsecured creditors, and outright purchasers of the collateral from the debtor who expect to obtain a clear title.

II. What is the Position in England? A. The Law Commission Project As a common law jurisdiction that has resisted the call for an Article 9 scheme remapping the secured transactions landscape, and that seems intent on continuing to do so for the indefinite future, England24 provides an interesting case study. Some years ago, the English Law Commission conducted an extensive inquiry into the subject of company charges in three publications, namely, a consultation paper, a consultation document and a report.25 From the very beginning, the Law Commission was hampered in its work in two principal respects: its terms of reference extended only to companies and, furthermore, they did not include insolvency legislation. The former restriction inhibited a general root and branch reform of the general law of security and the

22 K Llewellyn, ‘Through Title to Contract and a Bit Beyond’ (1938) 15 New York University Law Review 159, 165. At 166ff, he also criticised the use of the ‘lump’ concept of title to resolve diverse issues arising between sellers and buyers without regard to the different practical considerations raised by those issues. 23 See, eg, UCC, § 2-709 (seller’s action for the price dependent upon acceptance of the goods by the buyer). 24 Increasing devolution in the inchoate federal state that is the United Kingdom creates presentational difficulties. English-and-Welsh law would be a more accurate representation of the common law than English law, just as England-and-Wales would be more accurate for present purposes (since the issues dealt with in this paper have not been devolved to the Welsh Legislative Assembly). This is before we turn to Northern Ireland (not to mention the offshore Crown dependencies, such as the Channel Islands and the Isle of Man). 25 Law Commission Consultation Paper No 164, Registration of Security Interests: Company Charges and Property other than Land (2002); Law Commission Consultation Paper No 176, Company Security Interests (2004); Law Commission Report No 296, Company Security Interests (August 2005).

6

Michael Bridge

latter posed a particularly difficult problem in that the floating charge, even if eliminated from the general law of security, would necessarily remain to the extent that it played a key role in insolvency legislation.26 The language and forms of existing secured transactions law being embedded in insolvency law, no reform of the former could fluently proceed independently of the latter. The split between secured transactions law and insolvency law, as awkward as it is in a federal system that assigns them to different legislative competencies, is therefore also capable of causing problems in a unitary system. The amount of ground that the Law Commission covered in the two consultation documents in particular, and the depth of that coverage, were both impressive but, by the time of the Report, it was evident that late-declared opposition to the broader features of the Law Commission’s proposed scheme of radical reform had administered a fatal blow to that reform project. In brief, the consultation documents favoured notice filing and relieving the Registrar of Companies from the irksome task of checking the brief particulars of charge against the instrument of charge. At the time of the Law Commission’s involvement with secured transactions,27 the chargee’s priority position was determined by the contents of the instrument of charge and not by the barebones particulars of charge, even though only the latter was entered on the register by the registrar. The chargee’s priority position was preserved, not created, by delivering the required documents to the registrar. Consequently, if the wrong details were inserted on the particulars of charge and entered on the register, the registrar’s certificate that the requirements of the legislation had been satisfied was conclusive, even though third parties might have been misled by the divergence of the entry on the register and the instrument of charge, the latter not being available to them for inspection.28 Under these consultation proposals of the Law Commission, priority would have derived, not from the charge itself, but from the act of so-called perfection and from the date of that perfection.29 Consequently, where registration was the chosen route to perfection of the security interest, this necessarily entailed the abandonment of the registrar’s conclusive certificate. Permission to register in advance of any security agreement was also recommended by the Law Commission in its consultation documents.30 Departing from the approach then existing for registering only designated charges, the Law Commission recommended at the consultation stage the perfection of all security interests.31 Going further still, the Law Commission at the consultation stage tentatively recommended the bringing of quasi-securities, or at least some of them, into the proposed scheme of filing.32 This involved a recognition that 26

As mentioned above. For the current position, see below. 28 See National Provincial and Union Bank of England v Charnley [1924] 1 KB 431. 29 Consultation Paper No 176, para 3.201. 30 Consultation Paper No 176, para 3.139. 31 Consultation Paper No 176, passim. 32 Consultation Paper No 176, para 3.33. 27

Secured Credit Legislation: Functionalism or Transactional Co-Existence 7 certain quasi-securities might be integrated only to the limited extent of the rules on registration and priority (and not remedies). The sale (or discounting) of receivables33 would be such a case;34 consignment agreements,35 even if not securing payment or performance of an obligation, were less explicitly marked out for it.36 The extension of registration and priority rules to quasi-securities involving title reservation necessitated the recognition of an exception to the first-in-time approach to perfection, namely the so-called purchase-money security interest. Assuming that the value added to the company by collateral subject to title reservation was equal to the depletion of funds needed to acquire the collateral on credit terms, the advancement of the purchase-security interest holder would not come at the expense of a prior general financier of record. Furthermore, as is well-known, the purchase-money super-priority would open up alternative avenues of financing for debtors that would not depend upon there first having to be a priority agreement between earlier and later financier. In sum, these provisional recommendations amounted to something that was close to the Article 9 model. Finally, the second consultation document proposed a uniform scheme of creditor remedies.37 At this point, one might point to two aspects of the law of remedies. The first concerns the steps that a creditor might take to deal with collateral in the event of debtor failure, including resale and the treatment of any deficiency or surplus. This aspect was very much in the Law Commission’s line of sight. The second aspect concerns the intersection of creditor remedies and insolvency law, the subject of major reform in the Enterprise Act 200238 and outside the Law Commission’s terms of reference. My concern in this chapter is only with the former aspect. By the time of the Law Commission’s Report in 2005, the scope of the proposed reform had been substantially reduced. Notice filing remained, with the Registrar exempted from checking the particulars of charge against the instrument of charge.39 Priority, however, would continue to be dated from the creation of the charge and not from the date of perfection, despite the allowing of advance registration.40 All charges would have to be registered, except those for 33 Where the price paid for a receivable due on a future date reflects the current time value of that receivable, measured with the aid of a notional interest charge that is added to the current value of that receivable to bring it up to its face value. 34 Consultation Paper No 176, paras 3.48, 5.11. 35 Where goods are delivered to a future buyer in the expectation that they will be sold to that buyer, and thus generate an obligation to pay the price, only at the point where the buyer enters into a sub-sale contract with a customer. 36 Consultation Paper No 176, paras 3.43 (with which cf para 3.48), 5.11. 37 Consultation Paper No 176, Pt 5. 38 See now the Insolvency Act 1986, Sch B1 (especially para (3)), laying down a moratorium on the enforcement of security and property rights while a company is in administration and requiring that the administrator first try to save the company and then achieve a fair balance of all creditors before making a distribution to secured and preferential creditors. 39 Report No 296, para 3.76. 40 Report No 296, para 3.155.

8

Michael Bridge

which a case could be made for exemption.41 The registration of quasi-securities was abandoned with one exception.42 That exception was the sale of receivables (subject to a few exceptions). Since title reservation was left out of the reduced proposed reform, no provision was made for a purchase-money security interest given the consequential absence of transactions constituting the lion’s share of the case for this super-priority rule.43 The Law Commission’s aspirations were thus severely reduced with the fading away of any prospect of major reform.44

B. The Limited Changes of 2013 The above account of the Law Commission’s work is only part of a story that demonstrates the chronically vexed affair of the reform of the law of company charges in the United Kingdom. Part IV of the Companies Act 1989, which would have introduced significant changes to the law, though it would not have gone so far as was later proposed in the Law Commission’s consultation documents, was never brought into force. A lesson to be drawn from this experience was the need, in this area of law especially, for a preliminary testing of the waters and a sustained process of consultation. It is by no means just a question of the inherent conservatism of the legal profession and the practical exposure of many of its members, operating at varying levels within the profession, to the law on company charges. The rapid pace of change in the financial markets and the evolution of new ‘products’ meant that legal change might have, or be thought to have, unintended consequences in those markets.45 Further to powers conferred on the Secretary of State in the Companies Act 2006, regulations were made and came into force in 2013 modifying primary legislation, namely, the provisions on company charges in the 2006 Act, which had been carried over without substantial changes from the Companies Act 1985. The major changes that may be noted here are that all charges, subject to very limited exceptions, have to be registered if they are not to be void against liquidators and administrators, as well as competing secured creditors.46 It should, nevertheless, be remembered that a very large exemption from registration exists under different legislation for financial collateral arrangements, defined in terms

41

Report No 296, para 3.16. Report No 296, paras 4.18 and 4.25. 43 There are rare instances of transactions of a purchase-money security interest that do not involve the retention of title, eg, Re Connolly Bros Ltd (No 2) [1912] 2 Ch 25. 44 Professor Sir Roy Goode, with some judicial and considerable academic backing, has continued to press for major reform, along the lines of Art 9, via a system of committee and sub-committees, currently under the overall chairmanship of Professor Louise Gullifer. The City of London Law Society is also considering the case for substantial reforms of the security system, but not along the lines of Art 9. 45 Which is why, when reform is in the air, the nostrils of market participants dilate like those of antelopes in the Serengeti. 46 Companies Act 2006, s 859A. 42

Secured Credit Legislation: Functionalism or Transactional Co-Existence 9 broad enough to capture charges in relation to cash and credit claims transactions.47 Interestingly, the move towards the (near) comprehensive registration of charges that are not financial collateral arrangements places the spotlight upon the continuing exemption of title-based transactions. If selective registration of charges gives less than a complete picture, then what about a selective registration scheme that disregards title transactions? It seems impossible to make a case for requiring registration of all, or almost all, charges while resisting the case for registration of title schemes.48 If title transactions were to be registered, it would by no means follow that the contractual and remedial terms applicable between the parties to the transaction should have to be recharacterised so as to fall in line with the law relating to charges and mortgages. Title transactions would remain title transactions as far as the rights and duties of the parties to such transactions are concerned. The company charges register could simply be transformed into a register of charges and reservation of title, or the latter could be deemed for the limited purpose of registration to be company charges. Like a little learning, incomplete knowledge derived from the semblance of a comprehensive register can be a dangerous thing. The 2013 changes do not affect the rule that priority dates from the instrument giving rise to the charge and not from the date on which the taking of security was publicly notified. They bring us no nearer to a rational system of priority rules49 and they do not affect in any way the difference between legal and equitable interests. Since priority dates from the instrument and not from registration, the doctrine of constructive notice remains a relevant consideration. The extent to which registration of a company charge serves as constructive notice of the existence and contents of the charge has never been settled at the highest level.50 In contrast with the pre-2013 scheme, where only barebones particulars of charge were placed on the register, the current provisions now extend the range of registrable particulars of an instrument of charge. For example, they now include negative pledge clauses.51 On the slender back of existing case law, a subsequent financier might therefore be deemed to have notice of a negative pledge clause in a registered floating charge of earlier date. In addition, the registration system now provides for access to the instrument of charge itself. Does this mean that the doctrine of constructive notice applies to the contents of the instrument even if 47 See the Financial Collateral Arrangements (No 2) Regulations 2003, SI 2003/3226, as amended by SI 2010/2993, transposing Directive 2002/47/EC as amended by Directive 2009/44/EC. 48 There would be compelling reasons for not registering short-term reservations of title by trade suppliers, whilst requiring it for, say, a three-year finance lease that, but for tax reasons, might have been structured instead as a charge or mortgage. 49 The lack of which, however, does not seem to have generated significant difficulties in practice. 50 There has long been reliance on a first instance decision that this amounts to notice of the particulars of charge—Wilson v Kelland [1910] 2 Ch 306—but there is no clear view as to the categories of interested persons for whom this amounts to notice. Are purchasers of debts (factors) and of surplus machinery deemed to have notice? 51 Companies Act 2006, s 859D(2)(c) (‘terms of the charge [that] prohibit or restrict the company from creating further security that will rank equally with or ahead of the charge’).

10

Michael Bridge

they are not registrable particulars? The silence of the legislation on this point is profoundly unsatisfactory. Moreover, in this case, as in the case of the registrable particulars themselves, the continuing uncertainty about the size of the community of persons who should be expected to search the register in order to safeguard their interests remains an issue. The availability of the instrument of charge for inspection has at least this beneficial consequence: the registrar’s conclusive certificate has now gone and chargees are protected only to the extent of what is visible on the register. Overall, the Law Commission’s initial grand scheme is far from being implemented by these legislative changes. There are issues remaining, such as the need for an orderly and comprehensible system of priority rules, that may one day have to be addressed. For the present, it may be that a few trees have been cut down and others planted, but the view out of the carriage window remains more or less the same.

III. The Nature and Limits of Functionalism A. The Substance Test In well-known language, Article 9 provides that, with stated exceptions, ‘this Article applies to … a transaction, regardless of its form, that creates a security interest in personal property or fixtures by contract’.52 It then goes on to apply also to, inter alia, sales of accounts and consignments, though it does not seek to classify them as security interests properly so-called.53 Consequently, the location of title as between debtor and secured party may be relevant here when this is not the case for a transaction falling squarely within the above definition.54 A security interest is broadly defined as ‘an interest in personal property or fixtures which secures payment for performance of an obligation’.55 A slightly different approach is to be found in the Saskatchewan Act56 which applies to ‘every transaction that in substance [emphasis added] creates a security interest, without regard to its form and without regard to the person who has title to the collateral’,57 which places added emphasis on the functional character of the definition. The Saskatchewan Act goes on to apply the Act to transfers of

52

UCC, § 9-109(a)(1). UCC, § 9-109(a)(3), (4). 54 UCC, § 9-202. 55 UCC, § 1-201(37). 56 The Personal Property Security Act, 1993, c P-6.2. 57 ibid s 3(1)(a). Very similar language is to be found in the New Zealand Personal Property Securities Act 1999, No 126, s 17(1)(a) and in the Australian Personal Property Securities Act 2009, No 130, s 12(1). 53

Secured Credit Legislation: Functionalism or Transactional Co-Existence 11 accounts receivable and to commercial consignments, even though they cannot be said to secure payment or the performance of an obligation that has not yet arisen and may never arise.58 A similar definition denying the relevance of the location of title is to be found in the Ontario Act,59 the scope of which extends to ‘every transaction without regard to its form and without regard to the person who has title to the collateral that in substance creates a security interest’.60 Of interest is the Ontario Act’s inclusion in that definition of consignments securing performance of an obligation.61 The New Zealand Act retains a significant link to the old law. Like other non-US models, it simply disregards the location of title for the purpose of characterising an interest as a security interest; it does not deny the conditional seller’s reserved title. It recites the functional definition of security and then, for the avoidance of doubt, recites also a series of transactions that fall within it, namely, ‘a fixed charge, floating charge, chattel mortgage, conditional sale agreement (including an agreement to sell subject to reservation of title), hire purchase agreement, pledge, security trust deed, trust receipt, consignment, lease, an assignment, or a flawed asset arrangement,62 that secures payment or performance of an obligation’.63 This is tantamount to preserving old thinking whilst holding on to a functional definition for the rare case where it might be needed. It suggests that a great deal of rationality can be arrived at by clustering a variety of security or quasi-security t ransactions under the same statutory roof without obliterating the differences between them. This in turn means that a modern, rational system need not eliminate the distinction between security and title reservation. Perhaps the first question that ought to be asked is whether a functional definition provides a greater measure of certainty than the existing definition of a charge. I am dwelling on charges instead of mortgages because in practice and for various reasons charges are significantly more important than mortgages in the world of secured lending.64 It is easy enough to define a charge as an encumbrance falling short of a proprietary transfer that applies the collateral towards the satisfaction of an obligation in the event of the obligor’s default. In practice, however, significant difficulties of interpretation arise, especially if the word ‘charge’ is not employed by the parties. The distinction between a charge and a declaration of trust is far from being an easy matter: both involve a dealing with the beneficial

58

ibid s 3(1)(b). Personal Property Security Act, RSO 1990, c P.10. 60 ibid s 2(a). 61 ibid s 2(a)(ii). 62 Which implicitly assumes that a charge-back is a security interest. 63 Ontario Act, 17(3). 64 Freedom of contract in the provision of remedies leads to the elimination in practice of any distinction between mortgages and charges. Even in the highest courts, references may be found to a chargor’s equity of redemption: Re Bank of Credit and Commerce International SA (No 8) [1998] AC 214. Since a charge is merely an encumbrance and not a transfer, there is no legal title in the hands of the chargee that remains to be redeemed. 59

12

Michael Bridge

interest and the terminology of the law relating to charges sometimes mingles with that of the law relating to trusts.65 Furthermore a charge must involve a positive undertaking to apply collateral towards the discharge of an obligation and not a negative undertaking to refrain from drawing upon it or dissipating it.66 There is also some degree of tension in the case of appropriated funds: directing payment out of a particular fund may not necessarily involve the creation of a charge,67 yet payment out of a fund has been authoritatively expressed to be the required means of discharging the obligation.68 This cannot be true of all collateral: if I charge my collection of rare books, I should not be required to sell the collection in order to pay down the loan. I do not want to make too much of the case law, now abated, where unpaid sellers sought unsuccessfully to carve proprietary interests out of new goods produced or manufactured by the buyer, or out of book debts due to buyers from sub-buyers who had bought the original or new goods.69 These cases owed their existence largely to the energy or desperation of unpaid sellers in the face of an intractable English judiciary. That said, especially in the case of book debts, that judiciary if so minded might have taken the kinder line adopted by the Australian High Court when dealing with the claim of an unpaid seller to the buyer’s book debts.70 In sum, if a critic were to say that English law provides less than complete certainty when defining a charge, the criticism would be hard to rebut. That uncertainty, however, lies very much at the margins of the definition and can be avoided by competent drafting. It should not be supposed that a functional definition will be any easier to apply, especially when applied as it will have to be against an evolving commercial background, where new forms of transactions are thrown up, in such a way that prior case law may not be provide an effective guide in the application of the statutory test. There is a certain reductive character attaching to a functional definition that requires the brakes to be applied on the definition if it is not to be pressed too far. Functional thinking, pursued to its limits, encourages ‘over-capture’. To a man with a hammer, everything looks like a nail. Just because a transaction is built upon a foundation of commercial risk avoidance or limitation should not mean that the devices it adopts to protect an obligee should be treated as security interests. The leading architect of Article 9, Professor Grant Gilmore, once spoke in cautionary terms about ‘a no-man’s land, in which strange creatures do strange things’ for which ‘the rules which professionals have developed for professional transactions’ do not provide a measure.71 A striking example of overreach is the 65 See M Bridge and others, The Law of Personal Property (2013) para 7-083, citing Tailby v Official Receiver (1888) 13 App Cas 523, 541. 66 Flightline Ltd v Edwards [2003] EWCA Civ 63, [2003] 1 WLR 1200. 67 Swiss Bank Ltd v Lloyds Bank Ltd [1982] AC 584. 68 Palmer v Carey [1926] AC 703. 69 See MG Bridge, The Sale of Goods (3rd edn, Oxford, Oxford University Press, 2014) para 3.76ff. 70 Associated Alloys v CAN 001 452 106 Pty Ltd (2000) 202 CLR 568. A recent outrider case in England is FG Wilson (Engineering) Ltd v John Holt & Co (Liverpool) Ltd [2013] EWCA Civ 1232, [2014] 1 WLR 2065. 71 G Gilmore, Security Interests in Personal Property (Boston, Little Brown & Co, 1965) 336–37.

Secured Credit Legislation: Functionalism or Transactional Co-Existence 13 decision of the Canadian Supreme Court in Caisse Populaire Desjardins de l’Est de Drummond v Canada,72 which mentions a remark of Gilmore, who only reluctantly acceded to a specific exclusion of set-off from the Article 9 scheme,73 on the ground that one might as well go to the length of excluding fan dancing. In Caisse Populaire, a deemed trust in favour of the federal tax authority, over the assets of a company in insolvent liquidation, for sums deducted from employees’ salaries for income tax and health insurance, included assets of the company the subject of a ‘security’. The company had obtained a loan from a financial institution and had lodged with that institution a substantial sum by way of deposit. The deposit was subject to a negative pledge clause in that the company covenanted not to encumber it in favour of a third party. Moreover, it was not repayable to the company until the company had repaid the amount of the loan to the institution. In the conventional so-called ‘triple cocktail’ of English law,74 the bank advancing funds will have a right to combine accounts, whether by operation of law or reinforced by contract, a right not to repay the deposit until the loan is paid down (a flawed asset arrangement) and a charge over its own indebtedness to the borrower (the so-called charge-back). By a majority, the Canadian Supreme Court, influenced by the terms attaching to the repayment of the deposit, concluded that the institution had a ‘security’ over its indebtedness in the deposit account. Given especially the fact that the institution could not have protected itself by filing in the precautionary way often associated with Article 9 systems, this is an extraordinary outcome. A sum of money not repayable to the company, because the conditioning event in question never occurred, now has to be paid over to the federal tax authority. Yet the definition cannot be applied too narrowly, or else it will lose its evolutionary utility. One of its merits is that it provides a kind of statutory shorthand so that, if the draftsman wishes to bring a wide range of transactions under the same roof, there is no need constantly to mention mortgages, charges, conditional sales, pledges, trust receipts, equipment trusts and so on. Though too much should not be made of this as a measure of statutory achievement, a further virtue of the broader functional label is that it can attach to new forms of transaction that have not yet acquired a commercial existence at the time the statute is enacted. The pressure of regulatory need and regulatory predictability tends over time towards a security interest acquiring a conventional and more or less settled meaning in the way that an English charge is understood. Indeed, a well-known text, referring to a previous version of Article 9-109(1) adverted both to the paucity of litigation to which it gave rise and also to its success ‘even in bringing most deviant secured transactions under the umbrella of Article 9’.75 As for transactions that evade even the broad reach of the functional language in Article 9, notably, sales of book debts

72

Caisse Populaire Desjardins de l’Est de Drummond v Canada [2009] 2 SCC 29. See now UCC, § 9-109(d)(10). 74 See Re Bank of Credit and Commerce International (No 8) [1998] AC 214. 75 JJ White and RS Summers, Uniform Commercial Code (5th edn, 2000) 716. 73

14

Michael Bridge

and consignments, the need for a particular label continues. What is the alternative to this? The following formula captures the mischief of the statute yet is unattractive because of its dangerously subjective character: ‘This statute applies to any other transaction where the rational application of a general scheme of public notice and priority rules requires this to be so.’

B. Rights in the Collateral Any system of security must co-exist in harmony with principles of property law. A functional system of security presents its own challenges. UCC, section 9-203(b)(2) requires that the debtor have ‘rights in the collateral or the power to transfer rights in the collateral to a secured party’ as a condition of the creation of a security right. The reference to power catches the case of a debtor who as a result of the owner’s actions or statements has the appearance of ownership of the collateral or the apparent authority to burden it with a security interest.76 The interesting question, nevertheless, is whether it goes further than that, which requires an a nalysis of the phrase ‘rights in the collateral’. The first and most obvious question to ask is whether the security attaches only to the debtor’s rights or apparent rights in the collateral, or whether the existence of those rights acts as a threshold that must be crossed in order for the debtor to create an effective security interest in the whole of the encumbered asset. In a priority contest between two security or deemed security interests, there would not really be a priority conflict in the first place if the former position were correct. This would be because the security interest coming second in time would attach only to what remained of the asset after the first interest was subtracted. The fundamental purpose of the legislation would thus be subverted. The Official Comment to Article 9-203 expresses the position in the following way: A debtor’s limited rights in collateral, short of full ownership, are sufficient for a security interest to attach. However, in accordance with basic personal property conveyancing principles, the baseline rule is that a security interest attaches only to whatever rights a debtor may have, broad or limited as those rights may be.

The second mention of attachment in this passage is to say the least puzzling. The Official Comment, repeating Article 9-203(b)(2), then goes on to say that in some cases the debtor has power to transfer greater rights than he in fact has. It then makes a cross-reference to the priority rules, including the rule that a registered security interest has priority over an earlier unregistered interest. This leads to a circular conclusion. Article 9 concludes that a later secured creditor with a perfected security interest has priority over an earlier unperfected interest because the rights that the debtor needs to encumber the collateral on the second occasion 76

See, eg, UCC, § 2-403.

Secured Credit Legislation: Functionalism or Transactional Co-Existence 15 are to be found in the priority position accorded to the second creditor, which is itself dependent upon whether the debtor had rights in the collateral in the first place. This circularity expresses itself with particular force in those cases where title-based transactions are recharacterised as security agreements. At the heart of the matter lies the principle that the perfection of security interests creates a priority position; the failure to perfect does not invalidate a security interest. The unperfected first interest is still good against the debtor but the debtor has rights in the collateral that override those of the non-perfecting secured creditor. Someone taking a functionalist view of these provisions of Article 9, however, might conclude that deeming the debtor to have rights in the collateral for the purpose of creating the second security interest comes close to invalidating the first unperfected security interest as against the second security interest. Similarly, with greater force, one might say that the ability of unsecured creditors, via the trustee-in-bankruptcy, to trump the non-perfecting secured creditor amounts to de facto invalidity.77 In this case, there is no overriding second transaction to which the debtor is a party. Contrast the position of this unperfected security interest with that of a perfected security interest that is second in time to another perfected security interest. In this latter case, if the first perfected security interest exhausts the value of the collateral, a second security interest, even if perfected, is an empty vessel except where, as in the case of a purchase-money security interest, it can vault over the first security interest. Even if an empty vessel, the junior security interest might be replenished from time to time as the collateral expands or is upwardly revalued, or the senior secured creditor is paid in full or in part. It is therefore unsurprising that a leading Canadian architect of modern personal property security legislation confesses that the functionalist approach of Article 9 to finance leases and other title-based devices really amounts to a denial that the seller or lessor, as the case may be, is the owner of the collateral: Since a title retention sales contract or a lease falls within a secured financing regime because it functions as security device, it follows that the seller or the lessor is not the owner of the goods sold or leased. What the seller or lessor has is a security interest; the owner of the goods is the buyer or lessee … What is troublesome is that outside this regime, the recharacterization might not be successful with the result that the same transaction is viewed differently depending upon the legal issues being addressed.78

The functional view of security interest taken in Article 9 and in systems based upon it has a particular bearing on title-based schemes.79 The definition of 77

See UCC, § 9-317(a)(2). RC Cuming, ‘Internationalizing Secured Financing Law’ in R Cranston (ed), Making Commercial Law (Oxford, Oxford University Press, 1997) 523. The UNCITRAL Draft Legislative Guide on Secured Transactions reminds States that one version of the unitary approach as applied to acquisition financing is that the buyer would be treated as owner for all purposes, which would necessitate a change to taxation statutes if it were desired to tax the seller as owner: A/CN.9/631/Add.9, para 68. 79 Outside the field of acquisition finance, where it leaves it to individual States whether to take a functional view of security, the UNCITRAL Legislative Guide on Secured Transactions adopts a functional view of security: see Ch IX para 60 ff. 78

16

Michael Bridge

‘security interest’ in Article 1 of the Uniform Commercial Code makes it explicit in stark terms that a seller who delivers goods to a buyer on reservation of title terms is taking a security interest in those goods.80 Article 2-401 also states that ‘[a]ny retention or reservation by the seller of the title (property) in goods shipped or delivered to the buyer is limited in effect to a reservation of a security interest’.81 This absorption of title-based devices requires a major inroad to be made into the order of priority. The general priority rule, that the first in time should prevail, is set aside in the case of so-called purchase-money security interests.82 The Article 9 scheme, furthermore, does more than disregard the relevance of title: it recharacterises the transaction. A more equivocal note is struck in the UNCITRAL Legislative Guide on Secured Transactions, which poses the question whether the most efficient way to achieve equality of outcomes among acquisition financiers requires functionally equivalent outcomes. The Guide notes that in acquisition finance some financiers are lenders and some sellers. It is therefore for States to decide whether the logic of secured transactions should override the logic of sale and lease or vice versa.83 Two matters of considerable importance flow from the above observations. First, UCC Article 9, as facilitative as it is of secured transactions, exhibits a legislative willingness to curtail freedom of contract in the cause of financial rationality. One may observe the querulous note in the words ‘in effect’ on Article 2-401(1), as well as the fact that most of the aims of Article 9 can be accomplished without recharacterising the seller’s interest in this way. Secondly, the scheme laid down in Article 9 is deeply rooted in the remainder of the Uniform Commercial Code and in property law. A free-standing version of Article 9 cannot be transplanted into another legal system without considerable thought being given to all features of the legal terrain, especially the property law of the receiving jurisdiction, into which it is being transplanted.84 For example, the protection given in Article 9 to the buyer of goods in the ordinary course of business,85 prevailing against prior secured creditors of record, is not matched in a straightforward way in English law. English law gives little protection to good faith purchasers against the holders of legal interests in the goods who are not the disponors of those goods. That said, security interests in English law are usually equitable interests, in which case the good faith purchaser without notice of the legal interest will prevail over the secured creditor. Article 9 systems, however, efface the distinction between

80

UCC, § 1-201(35). So too Art 9.1 of the EBRD Model Law on Secured Transactions. This provision is brought into Art 9 via Art 9-110. A similar provision is to be found in the EBRD Model Law on Secured Transactions for vendors’ charges (Art 17.3). 83 Chap IX para 66. 84 See MG Bridge, ‘The Law Commission’s Proposals for the Reform of Corporate Security Interests’ in J Getzler and J Payne, Company Charges: Spectrum and Beyond (Oxford, Oxford University Press, 2006) 267, 288–89. 85 UCC, § 9-320(a). 81

82

Secured Credit Legislation: Functionalism or Transactional Co-Existence 17 legal interests. Consequently, the implications of transplanting such a system into English law would require a careful evaluation of the general provision made for title transfer and good faith purchasers.

C. Functionalism and the Need for Purchase-money Security Interests Article 9 and cognate systems accord a super-priority to a so-called purchasemoney security interest. A purchase-money security interest is one that secured a purchase-money obligation. An obligation of this kind is incurred as part of or all of the price of the collateral, which would involve the recharacterisation of a retained title as giving rise to a purchase-money security interest. The purchasemoney obligation also arises where an advance is given to the debtor for the purpose of acquiring rights in or the use of the collateral from a third party. It is strongly arguable that, apart from a few minor cases where English law explicitly recognises this super-priority,86 its recognition is also implicit in the very recognition of reservation of title as not amounting to security. This is because collateral subject to a reservation of title clause does not become part of the debtor’s estate.87 For present purposes, a reservation of title clause may be seen in an ordinary supply of goods on credit, as well as in the terms of a more structured conditional sale agreement or a finance lease. Consequently, it prevails over even a pre-existing charge granted by the debtor, since in English law this charge never captured collateral subject to title reservation that prevented it from forming part of the debtor’s estate. The charge would only attach in the event of the buyer or hire purchaser of the collateral acquiring an unencumbered title on making the final payment for the collateral.88 In the case of a finance lease, the event would be any ad hoc sale agreement reached between lessor and lessee once the agreement had run to term. The concept of purchase-money security interest, therefore, amounts to a form of statutory corrective to the proposition that the distinction between title and security can be eliminated in a broad functional scheme of security. A further aspect of purchase-money security interests also demonstrates how hard it is to suppress considerations of title. In the event of a priority conflict between two or more such interests, in Article 9 an interest in respect of the price prevails over an interest arising from value given to enable the debtor to acquire an interest in or the use of the goods,89 the latter of which best captures the case where the debtor acquires the goods not from the creditor but from a third party.

86 See Re Connolly Bros Ltd (No 2) [1912] 2 Ch 25; Abbey National Building Society v Cann [1991] AC 56. 87 See, eg, Insolvency Act 1986, s 283. 88 By analogy with the capture of future assets: Tailby v Official Receiver (1888) 13 App Cas 523. 89 UCC, § 9-342(g)(1).

18

Michael Bridge

D. Vesting and the Trustee-in-bankruptcy The UCC Article 9 approach to conditional sales, involving a forced transfer of the property in the goods to the buyer, dictates in a way the response that is made to a seller who fails to perfect its security interest by the time the buyer becomes insolvent. According to Article 9-317(a)(2), a trustee-in-bankruptcy, as a so-called lien creditor, takes priority over the non-perfecting conditional seller. Classically, priority contests erupt between competing interests in the same assets. A conditional buyer, if the reservation of title clause is effective, ought not in principle to have any interest in the subject matter of the contract to encumber in favour of another creditor. But by waving the legislative wand, the UCC has recharacterised the buyer as the owner and the seller as a security interest taker, thus giving rise to a priority contest between the seller and other secured creditors. Further, in the case of a bankrupt buyer, the trustee, as the representative of lien creditors, is allowed to prevail against the unperfected seller without any modification of bankruptcy legislation. The means chosen to achieve this was not to give the buyer a power to grant an interest in the collateral that overrides that of the seller: whilst that would work for voluntary transactions, it would not work for property interests arising by operation of law, like those that arise when assets vest in a trusteein-bankruptcy. Instead, the buyer has a proprietary right in the collateral that can vest in the trustee (or lien creditors), who is in effect the buyer itself carried on by other means. The trustee thus succeeds to the buyer’s rights but, unlike the buyer itself, can repudiate the seller’s security interest. The outcome is thus the same under Article 9, a system that treats perfection as a positive priority point, as it is under English law, where a failure to register or perfect by other means a charge or other true security interest is a negative priority point that removes the secured creditor’s entitlement on a stated event.90 Now, if one takes registration as the mode of perfection in a conditional sale case, it is not self-evident that a failure to register should benefit unsecured creditors, who represent the real ‘clients’ of the trustee. If the point is pressed that a debtor in possession of collateral presents a false appearance of wealth if the creditor’s interest is unperfected, one response is to say that in a credit-driven world possession is a weak indicator of ownership. Moreover, to take England as an example, the doctrine of reputed ownership was expunged from bankruptcy law nearly 30 years ago and was never applied to companies anyway. Further, whereas it might be common for secured creditors to search a register for competing interests, it is questionable how often unsecured creditors do this. And of course involuntary creditors never conduct a search. The New Zealand legislation, it should be noted, did not go down the road favouring unsecured creditors as represented by an insolvency office holder over an unperfected secured creditor. There is no provision that subordinates an unperfected security interest to the claims of unsecured 90

Liquidation, administration and challenges by secured creditors: Companies Act 2006, s 859H.

Secured Credit Legislation: Functionalism or Transactional Co-Existence 19 creditors expressed via the company liquidator or a trustee-in-bankruptcy.91 Given the long tradition of avoiding unregistered company charges as against trustees-in-bankruptcy and company liquidators, this is perhaps surprising. The position in other modern PPSA regimes, for example, Saskatchewan, is different.92 The line taken in Article 9 is there followed but the means chosen involve no small amount of manipulation to get there, as the Canadian case, Re Giffen,93 demonstrates. Re Giffen considers in some detail the bankruptcy effect of an unregistered ownership interest. It concerned British Columbia legislation modelled on the Saskatchewan Act. The ownership interest in this case was not that of a conditional seller but that of a lessor. A preliminary discussion of leases is in order before we turn back to the bankruptcy position. There are essentially two approaches to leases in modern personal property security legislation. One is to consider all of the terms of the transaction and to apply the substance test in order to separate leases which are functionally credit transactions from those that are simply bailments for the use of goods. This approach is taken in the UCC, where the definition of a security lease94 is dauntingly complex and, at times, counterintuitive,95 though it is said often enough that it has given rise to little litigation. One of the advantages of modern personal property security legislation is that the cost of compliance is small so that precautionary filing represents a real option in case of doubt about the susceptibility of a transaction to regulatory coverage by the legislation. The other approach, pioneered by the Saskatchewan Act, is to cut the Gordian knot and require compliance in the case of all leases of more than one year in duration. The Saskatchewan approach, in contrast with Article 9, does not involve treating leases as security agreements for the purpose of remedies between lessor and lessee. Leases of this sort are deemed in other respects to fall within the Act, no attempt being made to capture them in the functional test that defines a security interest.96 In that way, they give rise to security interests.97 In Re Giffen, a lessor had repossessed a car with the consent of the lessee’s trustee-in-bankruptcy but a dispute had arisen between lessor and trustee about the fate of the proceeds of sale of the car. The lease, for a term in excess of one year, had not been registered. If the legislation had deemed title to the car to pass to the bankrupt, it would have been a straightforward affair to rule that a security interest of this unperfected type, rendered subordinate to the trustee-in-bankruptcy, had the consequence of preferring the trustee in a contest with the lessor. While, in the United States, UCC Article 9 provides for title transfer in the case of 91 A company liquidator is not a third party able to assert a competing claim against a creditor with an unregistered security interest: Re King Robb Ltd [2007] 3 NZLR 802. 92 Saskatchewan Act, s 20(2). 93 Re Giffen [1998] 1 SCR 91. 94 UCC, § 1-203. 95 An option given to the lessee to become the owner is not sufficient to give rise to a security interest: UCC, § 1-203(c). 96 Saskatchewan Act, s 3(1), (2). 97 Saskatchewan Act, s 2(1)(qq)(ii)(C).

20

Michael Bridge

conditional sale,98 it is not quite so explicit in the case of security leases.99 It took the Supreme Court of Canada, overruling the British Columbia Court of Appeal, to conclude that the trustee defeated the unperfected lessor. The British Columbia provision specifically stated that the lessor’s interest was ‘not effective’100 against the trustee, which arguably is less explicit than it needs to be for the purpose of integrating the position with that prevailing in bankruptcy legislation,101 while the federal Bankruptcy and Insolvency Act went on to provide that there vested in a liquidator or trustee ‘all property wherever situated of the bankrupt at the date of his bankruptcy’.102 The question was how to fill the conceptual gap so as to confer upon the trustee rights greater than those possessed by the bankrupt lessee. This was found to be impossible in the Court of Appeal below but the Supreme Court ruled that the issue had to be resolved without reference to who had title to (or the property in) the car. Rather, the meaning of ‘property’ in the federal Act captured the lessee’s right to use and possess the car. Pausing there, we do not yet have an answer to the questions how the trustee is supposed to realise that property right and how much is that right worth. The Supreme Court’s response was to apply the British Columbia Act so as first to prevent the lessor from asserting its claim against the trustee, and secondly, to put forward the extraordinary proposition that the provincial Act modified federal bankruptcy legislation concerning the meaning of ‘property’. The conclusion one reaches is that, apart from cases where the contest involves other secured creditors, the subject of title or property does not go away. Either a fictitious transfer is deemed to be made to the debtor (conditional sale) or limited property rights of the debtor are inflated so as to be treated as title (leases). Modern personal property security legislation does not quite wipe the conceptual slate clean.

E. The Treatment of Consignments A consignment may for present purposes be designed as a bailment of goods by a prospective seller to a prospective buyer, coupled with an option given to the latter to purchase the goods at a pre-agreed price. The expectation is that the purchase will be made at the point of resale to a sub-buyer, failing which the goods will be returned to the bailor. The ‘debtor’ is in possession of goods, therefore, but is not yet under an obligation to pay for them. That obligation will arise at the 98 UCC, § 1-201(b)(35): ‘The retention or reservation of title by a seller of goods notwithstanding shipment or delivery to the buyer under Section 2-401 is limited in effect to a reservation of a “security interest”’. 99 UCC, § 1-203. 100 The Supreme Court considered that this provision was stronger than the language of the Saskatchewan Act (‘subordinate’). 101 The British Columbia Act goes on to define the scope of the claim that a lessor might make against the lessee’s estate. 102 British Columbia Act, ss 67(1)(c) and 71(2).

Secured Credit Legislation: Functionalism or Transactional Co-Existence 21 point of resale, namely, when the title to the personal property passes evanescently through the buyer to the sub-buying customer.103 This means that the bailment arrangement in itself cannot when initially executed constitute a security agreement because there is no obligation yet resting on the bailee for it to secure. At first sight, it looks extraordinary that a consignment arrangement should be treated, by anticipation as it were, as a type of secured transaction. The consignor, whilst the goods remain unsold by the consignee, is a bailor and not a creditor. Even deeming the consignor to be a creditor, where are the competing creditors? A creditor who acquires a security interest in the debtor’s inventory has to be treated, in the case of a consignment effected by another entity, as acquiring a property entitlement in something in which the debtor has no proprietary interest (apart from bare possession and usually an option to purchase). UCC Article 9 denies that a consignment gives rise to a security interest104 and yet, whilst not subjecting consignment to the unitary system of default remedies, embraces it for the purpose of rules on perfection and priorities. The extraordinary thing about it is not what Article 9 does to the rights of the consignor but what it does to the rights of the consignee’s creditors. It is well known that Article 9 systems override the classical division between common law and equity. In addition, they demonstrate that statutory fiat does not require a coherent integration of the statute into uncodified property law.

F. The Treatment of Receivables The language of Article 9, as we have already seen, whilst embracing a sale of accounts receivable does not place it in the category of ‘a transaction, regardless of its form, that creates a security interest in personal property or fixtures by contract’.105 It does not deny that it is a sale of accounts. We might now conclude that the sale of accounts does not involve a property interest securing an obligation: the buyer (‘creditor’) has already paid for the accounts, present and future, which have already been transferred to him or will automatically be transferred as soon as they come into existence. It is the very opposite of title reservation. There is nothing left for the seller (‘debtor’) to do except not to interfere with payment (if this is a case of direct collection) or to remit the proceeds of the buyer’s debts when they are received (in the case of indirect collection), apart from the possibility, which a debtor will want to exclude from the agreement, of being called upon to make up a deficiency in recoveries by a buyer invoking a recourse provision. If we are dealing with a case of facultative factoring, moreover, then we are looking at an executory agreement on both sides to offer and accept bundles of accounts on an agreed, selected basis. Even though it might be said that one might raise money 103 See

Kirkham v Attenborough [1897] 1 QB 201. UCC, § 9-102(a)(20). 105 UCC, § 9-109(a)(1). 104

22

Michael Bridge

on the basis of accounts in either of two ways, either by selling the accounts in advance of maturity outright or by using them as security for a loan to be repaid, one retort might be that one can travel to Piccadilly either by tube train or by bus, but that does not mean that a tube train is a bus. This, of course, is an oversimplification. Putting aside administrative costs and certain other charges, the discount rate applied to reduce a book debt to present value places in the hands of the seller an amount equivalent to a loan advance that, with interest, would have produced a repayment equivalent to the amount of the book debt. The calculations start from opposite ends but arrive at the same conclusion and are economically the same.106 The common presence in sales of receivables, moreover, of a recourse provision diminishes yet further the difference between loan and sale. The real difficulty in treating a sale as a secured lending operation seems to be to identify something that can be called collateral,107 which is a problem also in determining whether a repo amounts to a type of secured transaction. A considerable advantage in bringing sales of receivables into modern personal property security legislation is that a rational priority rule can be devised between competing sales and charges over the same receivables.

IV. Models of Transactional Co-Existence Finally, I pose the question whether it is in practical terms possible to have modern legislation whilst retaining old learning. It is a marked feature of Article 9 that it changed the terminology of secured credit. The old language and transactions fell away. Those in the secured credit industry could have clung on to what was familiar but they chose not to. The abandonment of the former legal culture is not an inevitable result of new personal property security legislation.108 Because their treatment under such legislation is less than comprehensive, sales of accounts receivable and consignments retain their existing character, whilst they are subjected to registration requirements that formerly may not have existed or may have been different. This process, in addition, fed them into a general scheme of comparative priority entitlement. For example, a consignor was considered to have the super-priority that went with a purchase-money security interest.109 Had it not been for what is now Part 6 of Article 9, which deals with default by the debtor, it is quite possible that other transactional types, especially those of a

106

This is very clearly demonstrated in Re George Inglefield Ltd [1933] Ch 1. under the reverse or second leg of a repo, the transferor does not have to retransfer the exact securities initially transferred but only their fungible equivalents. 108 As seen above, the New Zealand statute, for example, recites a functional test of a security interest and then proceeds to a substantial list of established transactions that, for the avoidance of doubt, fall within that definition: Personal Property Securities Act 1999, No 126, s 17(3). 109 UCC, § 9-103(d). 107 Since,

Secured Credit Legislation: Functionalism or Transactional Co-Existence 23 title-retaining character, might have retained their separate identities. My argument is that the imposition on secured credit transactions, certainly if they are commercial in nature, of a unitary remedial regime, based on the remedies that formerly applied to mortgages, is an unwarranted interference with the freedom of contract of debtor and secured creditor. Moreover, it is also unnecessary to achieve the major goals of modern personal property security legislation, which are to facilitate the efficient and confident flow of credit in a transparent system of entitlement with predictable and rational rules of priority between competing creditors. Legislation of this sort can tolerate the co-existence of security properly so called and reservation of title. What really matters is to have an effective and accessible registration scheme and a clear ordering of priorities. The UNCITRAL Legislative Guide on Secured Transactions recognises, in the case of acquisition financing, the possibility of two fundamental approaches, the unitary and the non-unitary. Whereas the unitary approach would obliterate the distinction between security and reservation of title, this distinction would be maintained under the non-unitary approach. It is interesting to look at an instrument that goes down the latter road. For the Cape Town Convention on International Interests in Mobile Equipment 2001110 to apply, the interest in the relevant equipment must be that of a chargee under a security agreement, a seller under a conditional sale, or a lessor under a leasing agreement.111 This flexible approach does not force on contracting states a functional approach to security, so as to compel them, for example, to treat a financial lease as the equivalent of a charge for the purpose of remedies.112 The Convention thus accommodates different legal philosophies in the various contracting states. It deals severally with the separate remedies of chargees, conditional sellers and lessors.113 We need not explore the precise extent of those differences here. For those interested in the prospects of reform in England, the question should be asked whether there are lessons to be learned from the Cape Town Convention.114 The approach adopted therein would allow any difference between true security and reservation of title, which has in the past been a real obstacle to progress, to be parked on one side until it comes to the elaboration of default

110 See RM Goode, ‘Official Commentary on the Convention on International Interests in Mobile Equipment and the Protocol Thereto on Matters Specific to Aircraft Equipment’ (2002) (as approved for distribution by the UNIDROIT Governing Council pursuant to Resolution No 5 of the Cape Town Diplomatic Conference). 111 Art 2(2). The interest includes also the proceeds of that proprietary interest in the equipment: art 2(5). 112 So far as it is necessary to distinguish security from conditional sale and leasing, this is a matter for the applicable law selected under the forum’s rules of private international law: Arts 2(4) and 5(3). 113 Arts 8, 10, 12, and 14. Again, it encourages adoption by contracting states with different legal philosophies by allowing for declarations against certain remedies under Art 54. 114 The UK ratified the Convention on 27 July 2015 with the associated regulations (SI 2015/912) coming into force on 1 November 2015.

24

Michael Bridge

and remedies, whilst attention is given to a modernised regime dealing with registration, future advances and priority. I am not very optimistic that there will be progress in the short to medium term. Major company law reform comes along about once in every 20 years. The prospect of a major reform embracing individuals, partnerships and companies being slotted into a crowded legislative timetable does not seem bright.

2 ‘Functional Formalism’1 in the Treatment of Leases under Secured Transactions Law: Comparative Lessons from the Canadian Experience CATHERINE WALSH2 As the C.C.Q. [Civil Code of Quebec] reveals, probably better than any other source we have dealt with, functionalism is not the answer. Functionalism is the question.3

I. Introduction This chapter focuses on the issue of whether and to what extent lease arrangements that function to finance the acquisition of ownership of the leased asset by the lessee should be subjected to the same rules that apply to economically equivalent secured transactions. To that end, it compares the treatment of lease financing under the Civil Code of Quebec,4 the Personal Property Security Acts5 (PPSAs) in

1 The reference to ‘functional formalism’ is taken from Roderick A Macdonald, ‘Article 9 Norm Entrepreneurship’ (2006) 43 Canadian Business Law Journal 240, 282. 2 Professor, Faculty of Law, McGill University. 3 M Bridge and others, ‘Formalism, Functionalism and Understanding the Law of Secured Transactions’ (1999) 44 McGill Law Journal 567, 664. 4 Civil Code of Quebec, CQLR c CCQ-1991, text available in both French and English online at www.canlii.org. 5 In order of implementation, see: ONTARIO, 1976 (SO 1967, c 73, in force 1 April 1976, replaced by SO 1989, c 16, in force 10 October 1989, RSO 1990, c P.10); MANITOBA, 1978 (SM 1973, c 5, in force 1 September 1978, CCSM c P35); SASKATCHEWAN, 1981 (SS 1979–80, c P-6.1, in force 1 May 1981, replaced by SS 1993, c P-6.2, in force 1 April 1995); YUKON TERRITORY, 1982 (OYT 1980, c 20, 2d Sess, in force 1 June 1982, RSY 2002, c 169); ALBERTA, 1990 (SA 1988, c P-4.05, in force 1 October 1990, RSA 2000, c P-7); BRITISH COLUMBIA, 1990 (SBC 1989, c 36, in force 1 October 1990, RSBC 1996, c 359); NEW BRUNSWICK, 1995 (SNB 1993, c P-7.1, in force 18 April 1995); NOVA SCOTIA, 1997 (SNS 1995-96, c 13, in force 3 November 1997); PRINCE EDWARD ISLAND, 1998 (SPEI 1997, c 33, in force 27 April 1998, RSPEI 1988, c P-3.1); NEWFOUNDLAND AND LABRADOR, 1999 (SNL

26

Catherine Walsh

effect in Canada’s common law jurisdictions, and the UNCITRAL Model Law on Secured Transactions.6 The broader goal is to demonstrate that an apparently formalistic approach may often simply reflect, in the words of Rod Macdonald, ‘not choices about formalism or functionalism, but rather choices about the location and level at which functional analysis should take place.’7 Part II reviews the ‘economic substance’ approach to the characterisation of security rights adopted by the PPSAs and the UNCITRAL Model Law. It will be shown that these regimes are aligned in characterising leases as secured transactions if the real object of the lessor’s reservation of ownership is to secure the lessee’s obligation to pay the acquisition price of the leased asset over the term of the lease. The effect of this characterisation is to subject ‘security leases’8 to the general publicity and enforcement rules governing all non-possessory security rights, while also according them the special priority both regimes award to an ‘acquisition security right’ over a prior security right taken in after-acquired assets of the lessee of the same kind as the leased asset. As Part III explains, the PPSAs differ from the UNCITRAL Model Law in extending their scope of application to long term ‘true leases.’9 It will be seen that there are two rationales for this extension. First, long-term true leases and security leases both involve a separation of possession and property rights and thus both engage the publicity rationale underlying the general PPSA requirement for nonpossessory security rights to be publicised by registration in order to be effective against third parties. Second, the determination of whether a particular lease is a true lease or a security lease requires a fact-specific laborious analysis with considerable room for uncertainty in cases at the margin. Bringing all leases within the scope of the PPSA publicity requirements thus also reduces the risk of disputes on the characterisation issue. The risk is not completely eliminated, however, since true leases are still excluded from the PPSA default enforcement regime, on the theory that under the economic substance approach it is the lessor not the lessee who is the residual owner of the leased asset under a true lease. Part IV turns to the treatment of leases under the Civil Code of Quebec. It will be seen that while the Code rejects the economic substance approach to the characterisation of security adopted by the PPSAs and the UNCITRAL Model Law, it 1998, c P-7.1, in force 13 December 1999); NORTHWEST TERRITORIES, 2001 (SNWT 1994, c 8, in force 7 May 2001); NUNAVUT, 2001 (in force 7 May 2001, SNWT (Nu) 1994, c 8). The text of all the Canadian Personal Property Security Acts is available online at www.canlii.org. 6 United Nations Commission on International Trade Law, UNCITRAL Model Law on Secured Transactions (2016), text available online at www uncitral.org. 7 Macdonald (n 1) 274. 8 The term ‘security lease’ is used here to denote a lease that from an economic substance perspective, is designed to enable the lessee to acquire ownership—or its economic equivalent—of the leased goods at the end of the lease term through the payment of instalments of ‘rent’ usually combined with payment of a terminal ‘option to purchase’ amount. A ‘true lease’ (sometimes referred to as an operating lease) is one designed simply to enable the lessee to use and enjoy the leased asset during the lease term. As will be seen, there is no simple sharp dividing line between true leases and security leases. 9 ibid.

‘Functional Formalism’ in the Treatment of Leases

27

indirectly adopts a functionalist approach by extending the publicity and enforcement rules applicable to nominate security in the form of a movable hypothec to other juridical institutions where they are deployed as security. When it comes to leases, however, the Code approach is more complicated. Like the PPSAs, the Code requires all long-term leases—regardless of economic substance—to be publicised by registration. Unlike the PPSAs, the Code excludes all long term leases, again regardless of their economic substance, from the constraints on enforcement imposed on secured creditors. This approach entirely eliminates the potential for inefficient disputes regarding the distinction between true leases and security leases, leaving room for a distinction to be drawn instead between giving all lessors in the business lease context comparatively robust enforcement rights while providing all consumer lessees with the same or superior procedural protection to that accorded only to security lessees by the PPSAs and the UNCITRAL Model Law. Part V reviews the characterisation approach adopted in other areas of C anadian law where the distinction between true leases and security leases is potentially relevant. As will be seen, with a limited exception in the insolvency restructuring context, Canadian law and practice has tended to reject the amorphous and malleable ‘economic substance’ test, preferring the efficiency of ‘bright line’ distinctions in line with the Code approach. Part VI offers concluding observations.

II. The Economic Substance Approach to the Characterisation of Leases under the PPSAs and the UNCITRAL Model Law Secured transactions law exists—as a distinct sub-set of private law—because of the perceived need for public order limits on the facilitative principles of party autonomy and property rights that inform contemporary visions of private law. Consider first the need for security of transactions in property markets. This policy goal is reflected in two widely accepted principles. The first is the principle of publicity under which a security right in movable property generally is not effective against third parties unless it has been publicised by registration or dispossession.10 The second is the need for certainty at the level of priority ordering, requiring priority, as a general rule, to be ordered temporally according to

10 See, for example, the UNCITRAL Legislative Guide on Secured Transactions (2010) at 47: ‘Varied as the legislation providing for non-possessory security rights may be, it often shares one common feature, namely, that some form of publicity of the security right is required or available.’ See also RA Macdonald, ‘Transnational Secured Transactions Reform: Book IX of the Draft Common Frame of Reference in Perspective’ (2009) Zeitschrift fur Europaisches Privatrecht 745, 767–76.

28

Catherine Walsh

the time when publicity was achieved, save where non-temporal priority can be justified by virtue of the character of a particular type of transaction. Consider next the need to protect the grantor from the risk of an unfair expropriation by a secured creditor of the residual value of an encumbered asset in excess of the amount needed to satisfy the obligation secured by that asset. In furtherance of this policy, secured transactions laws typically subject secured creditors to mandatory notification and other procedural constraints in the exercise of their enforcement remedies on the grantor’s default.11 Commercial practice is replete with examples of creative drafting aimed at avoiding the application of mandatory rules. It is the desire to thwart that impulse that underlies the economic substance approach to determining the scope of application of the secured transactions regime in the Canadian PPSAs.12 The Acts apply ‘to every transaction that in substance creates a security interest, without regard to its form and without regard to the person who has title to the collateral.’ In line with Article 9 of the Uniform Commercial Code in the United States,13 the PPSAs then define a ‘security interest’ as any interest in personal property that functions to secure an obligation, regardless of the form of the transaction and regardless of the location of ownership of the encumbered asset as between the creditor and the debtor.14 The UNCITRAL Model Law is aligned with the PPSAs (and UCC Article 9) in likewise embracing an economic substance approach. Subject to limited exceptions,15 the Model Law applies to all ‘security rights in movables.’16 It then defines ‘security right’ in generic and functional terms to mean ‘a property right in a movable asset that is created by an agreement to secure payment or other performance of an obligation, regardless of whether the parties have denominated it as a security right, and regardless of the type of asset, the status of the grantor or secured creditor, or the nature of the secured obligation.’17 The economic substance approach to the concept of security under the PPSAs and the UNCITRAL Model Law casts a very wide net. It extends the scope of application of these regimes beyond the traditional nominate forms of security to also include credit sales in which the buyer’s obligation to pay the purchase price is 11

See, for example, Macdonald, ibid 776–81. for example, the Ontario PPSA, above, s 2(a): ‘… this Act applies to … (a) every transaction without regard to its form and without regard to the person who has title to the collateral that in substance creates a security interest including, without limiting the foregoing, (i) a chattel mortgage, conditional sale, equipment trust, debenture, floating charge, pledge, trust indenture or trust receipt, and (ii) an assignment, lease or consignment that secures payment or performance of an obligation …’. 13 For the definition of ‘security interest’ used to determine the scope of Art 9 on Secured Transactions of the Uniform Commercial Code in the United States: see UCC, §1-201(35). 14 See, for example, the Ontario PPSA, above, s 1(1): ‘“security interest” means an interest in personal property that secures payment or performance of an obligation …’ For a detailed analysis, see RCC Cuming, C Walsh and RJ Wood, Personal Property Security Law (2nd edn, Irwin Law, Toronto, 2012) 116–154. 15 See Art 3 as well as the qualifications on scope in Arts 4–6. 16 Art 1. 17 Art 2(kk). 12 See,

‘Functional Formalism’ in the Treatment of Leases

29

secured by the seller’s reservation of ownership.18 By logical extension from that extension, both regimes also apply to economically equivalent lease arrangements, that is to say, leases in which the rental payments, viewed from an economic substance perspective, are aimed at enabling the lessee to acquire ownership of the leased asset at the end of the lease term.19 The effect of this approach is to require financing sellers and lessors to observe the general publicity requirements and enforcement obligations imposed by these regimes on secured creditors. The PPSAs and the UNCITRAL Model Law are also aligned in recognising a special priority for acquisition security rights (‘purchase money security interests’ in PPSA parlance). Under both regimes, an acquisition secured creditor has priority over a prior secured creditor that took security over after-acquired assets of the same kind as the asset subject to the acquisition security right provided that it satisfies certain specified timely registration and other conditions.20 The principal rationale for this special priority is that the assets would not have entered the debtor’s asset base but for the fresh injection of credit by the acquisition financer. That rationale clearly extends to sellers and financing lessors and the concept of acquisition security right in the UNCITRAL Model Law and the PPSAs has been crafted to clearly include not just loan but also sale and lease secured credit.21

III. Extension of the PPSAs to Long-term ‘True’ Leases The PPSAs and the UNCITRAL Model Law part company on one significant point. The application of the UNCITRAL Model Law is limited to leases that in economic substance are meant to finance the lessee’s acquisition of the leased asset at the end of the lease term. In contrast, the PPSAs, extend their scope of application to long-term true leases, defined as a lease for a term of more than one year that does not secure an obligation.22 18

For the PPSAs, see Cuming, Walsh and Wood (n 14) 122–133.

19 ibid.

20 See, for example, s 33 of the Ontario PPSA. For a detailed analysis of purchase money security interest priority under the PPSAs, see Cuming, Walsh and Wood (n 14) 439–467. For the UNCITRAL Model Law, see Art 38. 21 See, for example, the definition of ‘purchase money security interest’ in s 2 of the Ontario PPSA: ‘“purchase-money security interest” means, (a) a security interest taken or reserved in collateral, other than investment property, to secure payment of all or part of its price, (b) a security interest taken in collateral, other than investment property, by a person who gives value for the purpose of enabling the debtor to acquire rights in or to the collateral, to the extent that the value is applied to acquire the rights.’ Art 2(b) of the UNCITRAL Model Law defines ‘acquisition security right’ in a like manner to mean ‘a security right in a tangible asset … which secures an obligation to pay any unpaid portion of the purchase price of an asset, or other credit extended to enable the grantor to acquire rights in the asset to the extent that the credit is used for that purpose.’ 22 See, for example, s 3 of the British Columbia PPSA: ‘… this Act applies to … (c) a lease for a term of more than one year, that do[es] not secure payment or performance of an obligation.’ See also the related definition in s 1(1) of a ‘lease for a term of more than one year.’ And see Cuming, Walsh and Wood (n 14) at 155, 159–161.

30

Catherine Walsh

Two rationales underlie the PPSA approach. The first is that the separation of possession and property rights (regardless of whether the relevant property right is ownership or security) inherent in both security leases and true leases creates transparency problems for third parties equivalent to those posed by non- possessory security rights. Requiring all long-term leases, regardless of whether in economic substance they are security leases or true leases, to be publicised by registration ensures that third parties can objectively verify whether a person owns or is merely leasing assets in her possession.23 The second rationale has to do with the unique characterisation challenges inherent in attempting to distinguish between true leases and security leases. Indeed it was precisely those challenges that persuaded Ontario in 2006 to finally bring its PPSA in line with the other PPSAs on this point.24 To understand why the application of the economic substance approach to leasing arrangements presents unique characterisation challenges, compare a transaction cast in the form of a sale subject to a reservation of ownership by the seller to secure payment of the purchase price. This form of transaction is self-evidently a secured transaction under the economic substance concept. The security function performed by the creditor’s reservation of ownership is evident in the formal terms of the transaction. No further investigation of its economic substance is needed. In contrast, not every lease formally designated as such is self-evidently a security lease. Rather, the determination of whether a particular lease is a ‘true lease’ or a ‘security lease’ requires a painstaking fact-specific analysis of its specific terms.25 In principle, by analogy to a credit sale subject to a reservation of ownership to secure the price,26 the appropriate characterisation depends on whether the terms of the particular lease reflect the parties’ intention at the end of the lease term to leave the lessor with a ‘meaningful residual ownership interest’ (true lease) or to vest ownership (or its ‘functional and economic equivalent’) in the lessee (security lease).27 23

See Cuming, Walsh and Wood (n 14) 155. See M Burke, ‘Ontario Personal Property Security Act Reform: Significant Policy Changes’ (2009) 48 Canadian Business Law Journal 289, observing at 289 that ‘the issue of whether a lease is a true lease or a security lease has kept more litigators gainfully employed than any other specific OPPSA issue.’ On the origin of the Ontario reform, see Canadian Bar Association—Ontario, Submission to the Minister of Consumer and Commercial Relations concerning the Personal Property Security Act (Toronto: Canadian Bar Association—Ontario, 1998) at 8. 25 On the sometimes conflicting criteria Canadian courts have applied to distinguish between true leases and security leases in the PPSA context, see Cuming, Walsh and Wood (n 14) 125–133; Burke (n 24) 292–95; RCC Cuming, ‘True Leases and Security Leases under Canadian Personal Property Security Acts’ (1983) 7 Canadian Business Law Journal 251; J Ziegel, ‘Security Interests and Continuing Challenges in Characterisation of Equipment Leases: DaimlerChrysler Services Canada Inc. v. Cameron’ (2009) 47 Canadian Business Law Journal 283. 26 See Burke (n 24) 291, observing that ‘Canadian courts have, for the most part, used a typical conditional sale or, in the case of a sale-leaseback transaction, a typical chattel mortgage as a benchmark’ in determining whether a lease should be characterised as a security lease. 27 See SL Harris and CW Mooney, ‘When is a Dog’s Tail Not a Leg?: A Property-Based Methodology for Distinguishing Sales of Receivables from Security Interests That Secure an Obligation’ (2014) 82 University of Cincinnati Law Review 1029 at 1050–1052. 24

‘Functional Formalism’ in the Treatment of Leases

31

Some arrangements present no difficulty. In short-term car or equipment rental agreements, the lessee is clearly paying only for the temporary possession and use of the asset with the lessor retaining a meaningful residual ownership interest. At the other end of the spectrum are long-term leases in which the lessee is contractually committed to make the lease payments for the full term of the lease at the end of which it is either obligated to purchase the leased asset or may exercise an option to purchase it for a nominal sum (security lease).28 In practice, the terms of most long-term lease arrangements do not permit easy characterisation. Rather the ultimate determination depends on a detailed analysis of a diversity of factors and the relative weight accorded to each by the court.29 Moreover, there is room for considerable debate as to which factors are relevant and which are merely red herrings, which constitute primary indicia and which are merely secondary or inconclusive, and whether the default presumption in cases at the margin favours a true lease or a security lease characterisation.30 In the United States, the Uniform Commercial Code attempts to draw a sharper line between true leases and security leases by setting out a list of factors presumed to be indicative of the appropriate characterisation.31 The section, however, begins with the acknowledgement that whether ‘a transaction in the form of a lease creates a lease or security interest is determined by the facts of each case.’32 Not surprisingly, it has not produced any greater certainty in the UCC case law in the United States.33 The omission of a lessor to perfect by registration is a fruitful source of characterisation disputes under UCC Article 9 as it was under the Ontario PPSA before it was amended to bring true leases within its scope. By including both security leases and true leases over a term of more than one year within the scope of the

28 But note Harris and Mooney (n 27) 1053, note 116, observing that ‘a lease with a nominal purchase option may be a true lease nevertheless if the lease is terminable by the lessee’ since in that case ‘there may be no meaningful obligation sufficient to support’ a security characterisation. 29 See Burke (n 24), observing at 291 that ‘the characterisation process can be viewed as a scale, with the ultimate characterisation of the lease being based on the weighing of a number of factors or indicia.’ 30 Burke (n 24) at 202–98 attempts to impose some order on the characterisation issue by drawing a set of principled generalisations from the PPSA case law but concludes his analysis at 298 with the acknowledgment that there are, nonetheless, ‘more than a few cases that cannot be easily explained by [these] generalizations.’ For a similar conclusion in the UCC context, see Ihne (n 33). 31 UCC, § 1-203. 32 UCC, § 1-203(a). 33 See Burke (n 24) 291, note 6. And see Matthew L Warren, ‘Preventing Economic Harm from the Miscalculation of Risk in Equipment Leases’ (2013) 87 American Bankruptcy Law Journal 405, observing that UCC, § 1-203 ‘has led to inconsistent interpretation and application by courts.’ For a somewhat contrary view, see MJ Abatemarco and AM Sabino, ‘“True Lease” Versus Disguised Security Interest: Is the United Trilogy Truly the Last Stand? (2008) 40 UCC Law Journal 445. For an attempt to provide a framework to fill the gaps in the statutory guidance provided by the UCC, see RW Ihne, ‘Seeking a Meaning for “Meaningful Residual Value” and the Reality of “Economic Realities”—An Alternative Roadmap for Distinguishing True Leases from Security Interests’ (2007) 62 The Business Lawyer 1439; the author acknowledges, however, that his ‘proposed roadmap is not determinative in every instance’: ibid at 1469.

32

Catherine Walsh

Ontario PPSA, the legislator has succeeded in eliminating an important historical source of characterisation disputes. However, the potential for characterisation disputes has not been completely eliminated. True leases are excluded from the enforcement provisions of the PPSAs.34 Consequently, the distinction remains relevant for the purposes of determining whether or not a lessor’s remedies on the lessee’s default are constrained by the PPSA enforcement regime. In the particular case of the British Columbia PPSA, the characterisation challenge is further complicated by the imposition, under section 67, of a ‘seize or sue’ limitation on a lessor’s remedies against a defaulting consumer lessee. Where the leased assets constitute consumer goods, the creditor must elect between seizure of the assets and suing the consumer personally for the amounts outstanding under the lease. However, that limitation applies only if the lease is characterised as a security lease so as to come within the PPSA enforcement regime. Arguably because of the drastic impact on a lessor’s enforcement rights, the courts have tended to favour a true lease characterisation in cases where the indicia are ambiguous, thereby diminishing the scope of the consumer protection promised by the provision.35 This exception aside, the remedies available to a lessor subject to the PPSA enforcement regime differ from those available to a lessor under a true lease only in the following respects. First, the PPSAs entitle the lessee to the surplus in the rare event that the proceeds realised on its resale of the leased asset following seizure exceed the amount owed by the lessee under the terms of the lease.36 Second, the PPSAs require the lessor to give advance notice to the lessee of the details of its intended disposition of the leased asset and empower the lessee to cure the default and reinstate the lease prior to disposition by paying only the amount actually in arrears, together with any realisation expenses incurred by the lessor to that point.37 It has been suggested that consideration could be given to extending the advance notice and reinstatement provisions of the PPSAs to true leases.38 This would have the advantage of further reducing the need to distinguish between a true lease and a security lease.39 34 See TM Buckwold, ‘The Law of Commercial Leasing in Canada 1999’ (Uniform Law Conference of Canada: Winnipeg Annual Meeting, 1999) available online at www.ulcc.ca/en/annual-meetings/3591999-winnipeg-mb/civil-section-documents/200-the-law-of-commercial-leasing-in-canada-1999, comparing the enforcement rights of lessors under true leases and security leases at paras 61–70 and 72–78. 35 See Mercedes Benz Financial v Wager 2010 BCSC 1090; TransportAction Lease Systems Inc v Fung 2013 ABQB 215 (CanLII); DaimlerChrysler Services Canada Inc v Cameron 2007 BCCA 144 (CanLII). For a critical analysis of the Court’s reasoning on the characterisation question in the latter case, see J Ziegel, ‘Security Interests and Continuing Challenges in Characterization of Equipment Leases: DaimlerChrysler Services Canada Inc v Cameron’ (2009) 47 Canadian Business Law Journal 283. 36 Buckwold (n 34) paras 72–78. 37 ibid. 38 Buckwold (n 34) paras 77, 118. 39 ibid.

‘Functional Formalism’ in the Treatment of Leases

33

There are, however, two difficulties with the suggested solution. First, it would still be necessary to draw the distinction so as to protect the lessee’s entitlement under a security lease to any surplus proceeds following the lessor’s seizure and disposition of the leased assets by the lessor.40 Second, in the business context, it is arguable that a true lease financing structure is particularly suited to financially stressed lessees owing not just to the off-balance accounting treatment traditionally given to operating leases41 but also the absence of advance notice and other procedural constraints on the lessor’s repossession rights.42 It follows that the extension of the procedural constraints imposed by the PPSA to true leases might well diminish the availability of this form of business financing. That having been said, it is generally acknowledged that a more protective approach is warranted in the consumer context, regardless of the economic substance of the particular lease transaction.43 Consumers lack bargaining power and usually do not have the benefit of legal advice and so may not understand the liabilities to which they are potentially exposed. In particular, the standard terms of contemporary lease agreements typically provide that on the lessee’s default, the lessor is entitled to terminate the lease and to recover liquidated damages consisting of the total of all unpaid instalments of rent to the end of the lease term. So-called open-end leases usually also provide for an early termination payment obligation, requiring the lessee to pay any difference between the net amount received by the lessor on a commercially reasonable resale of the leased assets and the estimated residual value amount agreed to in the contract of lease. The courts generally have upheld the efficacy of these terms.44 Although calls for enhanced protection of consumer lessees in the common law jurisdictions have not received widespread take-up, the consumer protection legislation in several PPSA jurisdictions now includes provisions: (1) imposing meaningful cost of credit disclosure obligations on lessors;45 (2) setting objective limits on the residual obligation of a defaulting consumer for the difference between the residual value specified in the lease and the actual value of the leased asset at the end of term;46 and (3) limiting the operation of acceleration clauses by giving the defaulting consumer the right to advance notice and the right to reinstate the lease on payment of the amount actually in arrears together with enforcement expenses incurred by the lessor to that point.47 40

Buckwold (n 34). But see Part V below on the recent changes to the accounting standards for leases. Giner and F Pardo, ‘Operating lease decision and the impact of capitalization in a bank- oriented country’ (2017) 49 Journal of Applied Economics 1886, DOI: 10.1080/00036846.2016.1229416 at 3–5, available online at dx.doi.org/10.1080/00036846.2016.1229416. 43 Buckwold (n 34) paras 131–205. 44 See, for example, Hav-A-Kar Leasing Ltd v Vekselshtein 2012 ONCA 826, paras 48–50; Mercedes Benz Financial v Wager 2010 BCSC 1090 (CanLII). And see Buckwold, (n 34) paras 61–70. 45 See Alberta Fair Trading Act, RSA 2000, c F-2, ss 90–93; Manitoba Consumer Protection Act, CCSM c C200, ss 36–39; British Columbia Business Practices and Consumer Protection Act, SBC 2004, c 2, ss 100–101. 46 Alberta, s 94; Manitoba, s 40; British Columbia, s 102. 47 British Columbia, s 77. 41

42 B

34

Catherine Walsh

IV. The Functionalist Underpinnings of the Formal Characterisation of Leases under the Civil Code of Quebec The initial drafts of the 1991 Civil Code of Quebec reflected an economic substance approach to the concept of real security under which any transaction that secured an obligation would have been presumed to give rise to a hypothec.48 That approach was ultimately rejected. Although the hypothec is the only recognised nominate form of real security,49 the Code expressly acknowledged that other juridical institutions may also be deployed to secure an obligation; namely: the sale with a right of redemption,50 the trust,51 and the instalment sale whereby the seller reserves ownership of the property until full payment of the sale price.52 In most important respects, however, the Code indirectly reflects a functionalist approach. First, the Code extends the publicity requirements and enforcement limitations applicable to hypothecary creditors to creditors who use other institutions to secure an obligation.53 Second, while the Code limits special priority for acquisition financing to sale as opposed to loan credit, sellers otherwise enjoy the same super-priority as their counterparts under the PPSA and the UNCITRAL Model Law. Where the security takes the form of a movable hypothec granted by the buyer, super priority is achieved by a special priority rule, disrupting the general rule that priority is determined by the order of publicity.54 Where it takes the form of an instalment sale subject to a reservation of ownership to secure payment of the price, there is no need for an explicit priority rule. Provided the seller observes the timely registration requirements imposed by the Code,55 its priority flows naturally from its retained ownership, preventing the purchased asset from being captured by an after-acquired property clause in a security agreement entered into by the buyer with a prior hypothecary creditor. 48 See RA Macdonald, ‘Faut-il s’assurer d’appeler un chat un chat? Observations sur la méthodologie législative à travers l’énumération limitative des sûretés, « la présomption d’hypothèque » et le principe de « l’essence de l’opération » in E Caparros (ed), Mélanges Germain Brière (Montreal, Wilson & Lafleur, 1993) 527; Bridge (n 3) 614–18. 49 Art 2660 defines a hypothec as ‘a real right on a movable or immovable property made liable for the performance of an obligation.’ The definition recognises that a hypothec confers on the hypothecary creditor the full range of rights classically associated with real security, including the right to follow the hypothecated asset into the hands of third parties, to enforce the hypothec against the encumbered asset on the grantor’s default, and to enjoy a preference in payment over other creditors in the proceeds of realisation. 50 Although the drafting is infelicitous, CCQ, Art 1756 recognises that the object of a sale with a right of redemption ‘may be to secure a loan’. 51 Art 1263 recognises that a trust patrimony may be created by a settler for the purposes of securing the performance of an obligation. 52 Art 1745. 53 Arts 1750, 1756 (sale with a right of redemption); Art 1263 (trust); Arts 1745, 1749 (instalment sale subject to a reservation of ownership to secure the purchase price). 54 Art 2954. 55 Art 1749.

‘Functional Formalism’ in the Treatment of Leases

35

The Code’s treatment of long-term lease transactions, however, presents a more complex picture. On the one hand, like the PPSAs, the Code requires all long-term leases, regardless of their economic substance to be publicised by registration in order to be set up against third parties.56 On the other hand, unlike the PPSAs, the Code does not subject lessors under security leases to the procedural constraints on enforcement imposed on hypothecary creditors and extended to other creditors who deploy other devices for security purposes.57 Relative to the PPSAs, the Code approach has the great advantage of completely eliminating the potential for characterisation disputes about whether a particular lease is a security lease so as to engage the enforcement regime applicable to secured creditors. But this does not mean that defaulting lessees are left without protection. For consumer lessees, the Quebec Consumer Protection Act establishes extensive protections in relation to the type of issues summarised in the preceding section that have also been the subject of consumer protection legislation in a number of the common law provinces.58 But most importantly, it also effectively addresses the risk that a lessee under what the PPSAs might characterise as a security lease may be unfairly deprived of her accumulated ‘equity’ in the leased asset. The Act requires a lessor to obtain a court order for repossession where the defaulting consumer already has made half or more of the lease payments due under the lease.59 In determining whether it should grant the order, the court has the discretion to dismiss the lessor’s application and permit the lessee to retain the asset pursuant to whatever changes in the terms of payment it considers reasonable. The requirement for a court hearing provides a powerful incentive to the lessor to come to an acceptable settlement with the lessee in scenarios where the risk of an unfair expropriation is present. The Quebec courts have signalled their intention to treat lessors who fail to take these requirements seriously by limiting their subsequent recourse against the lessee and even awarding punitive damages.60

56 Art 1851 defines a lease (louage) as ‘a contract by which a person, the lessor, undertakes to provide another person, the lessee, in return for a rent, with the enjoyment of a movable or immovable property.’ Under Art 1852, a lease for a term of more than one year must be published within 15 days to be set up against third parties. Arts 1842–1850 deal separately with tripartite leasing (crédit-bail) arrangements in which the role of the lessor is purely that of a financing conduit between the supplier of the leased asset and the lessee. Under Art 1842, this type of leasing arrangement may only be entered into for business purposes and under Art 1847 the rights of ownership of the lessor may be set up against third persons only if they have been published within 15 days of the leasing contract, regardless of the duration of the lease term. 57 The enforcement regime for hypothecs is set out in Chapter V of Book Six of the Civil Code. By its own terms it does not apply to lease transactions and there is nothing in the articles governing leases and leasing contracts that requires lessors to observe the restrictions on enforcement imposed on hypothecary creditors. 58 Consumer Protection Act, CQLR c P-40.1. For detailed references and a detailed summary of the extensive protections afforded by the Act to consumer lessees and references, see Catherine Walsh, ‘Leasing Law in Québec (Canada)’ (2011) 16 Uniform Law Review 181, 208–222. 59 Art 150.32. 60 Failure to give advance notice, or to request a court order before initiating repossession, or failure to do both, has resulted in limiting subsequent lessor recourse against the lessee (see Landmark Vehicle Leasing Corp v Leboedec, 2009 QCCQ 8684 (CanLII)), and may justify an award of punitive damages (see Brodeur v Continental Location C&B Ltée., 1999 CanLII 4732 (QC CQ)).

36

Catherine Walsh

In the non-consumer context, the absence of procedural constraints on the lessor’s repossession and other enforcement rights is unlikely to result in unfairness in practice. Business lessees can be assumed to be sufficiently sophisticated to assess the cost/benefit trade off inherent in this form of financing. As well, equipment leasing tends to be leveraged with the result that a lessee typically will not have a meaningful residual interest in the leased asset at the point of default. Where this is not the case, it is suggested that sufficient ex post protection is afforded under the Code by the general prohibition against the bad faith exercise of contract enforcement rights in an excessive or unreasonable manner and the availability of judicial relief from abusive clauses in contracts of adhesion.61

V. Characterisation of Leases outside Secured Transactions Law Outside the secured transactions context, the distinction between true leases and security leases in Canadian law may be relevant where the lessee becomes insolvent.62 As a general rule lessors are entitled to assert their ownership rights in leased assets in bankruptcy proceedings involving the lessee regardless of the economic substance of the underlying arrangement.63 This is subject to a qualification, however, in the context of reorganisation proceedings where federal insolvency legislation provides for a stay on the enforcement rights of creditors in order to give the insolvent debtor sufficient time to reorganise and renegotiate the terms 61 Art 1375. Art 7 of the Code prohibits not just subjective bad faith (‘No right may be exercised with the intent of injuring another …’) but also the excessive or unreasonable exercise of a right measured as an objective norm of behaviour (‘No right may be exercised … in an excessive or unreasonable manner’). In addition, Art 1437 gives Quebec courts the power to avoid an ‘abusive clause’ in a contract of adhesion or reduce the obligation arising from it, with abusive clause defined as ‘a clause which is excessively and unreasonably detrimental to the consumer or the adhering party and is therefore not in good faith; in particular, a clause which so departs from the fundamental obligations arising from the rules normally governing the contract that it changes the nature of the contract is an abusive clause.’ For an application of this provision in the lease context, see Société générale Beaver inc v Métaux ouvré St-Philippe inc J.E. 94-1295, REJB 1994-28759 (jugement porté en appel: CAM, no 500-09-001322940), where a lessor who seized the leased goods at an early point in the lease term and then re-leased the goods to another lessee, was prevented from enforcing a liquidated damages clause in the contract of lease purporting to require the defaulting lessee to pay the amount of the lease payments for the duration of the lease. 62 In Canada, exclusive legislative authority over bankruptcy and insolvency is allocated to the federal Parliament: see the Constitution Act, 1867, 30 & 31 Vict, c 3, s 91(21). The principal federal statutes are the Bankruptcy and Insolvency Act, RSC 1985, c B-3 (‘BIA’), covering both enterprises and individual debtors, and the Companies’ Creditors Arrangements Act RSC 1985, c C-36 (‘CCAA’) applicable only to the reorganisation of large insolvent corporations that owe their creditors in excess of $5 million Canadian. 63 Section 2 of both the BIA and the CCAA define ‘secured creditor’ in classic terms to mean a person holding ‘a mortgage, hypothec, pledge, charge or lien’ on the property of the debtor as security for the indebtedness of the debtor to that person. Lessors are thus implicitly excluded. And while the BIA definition was amended in 2001 to expressly include a seller who retains ownership under an instalment sale under the Civil Code of Quebec, lessors remain excluded.

‘Functional Formalism’ in the Treatment of Leases

37

of its indebtedness. As an exception to the stay, both Acts provide that a lessor can continue to require ‘immediate payment’ for the post-commencement ‘use’ of leased property.64 The courts have interpreted the word ‘use’ to mean that the exception applies only where the lease is a ‘true’ lease in the sense that the lessee is paying purely for the use of the asset.65 Otherwise, the lessor is treated in the same manner as a secured creditor with the result that it will usually not receive any payment for the lessee’s use of the leased asset during the reorganisation period and is precluded by the stay from exercising its contractual rights of termination and repossession. Although the issue remains controversial, courts have consistently justified this interpretation as necessary to advance general insolvency restructuring policy which seeks to ensure that a debtor corporation can restructure its affairs in a way that will permit it to remain a going concern, without being limited by creditors seeking to enforce their pre-insolvency rights.66 In other areas where the distinction between true lease and security lease may be relevant, however, an ‘economic substance’ analysis has been rejected in favour of a bright line distinction in line with the Civil Code approach reviewed in the preceding section. Consider first the determination of whether a lessor is entitled to claim a depreciation allowance under federal and provincial income tax law. Here, legislators have adopted bright line ‘specified leasing property rules’ pursuant to which lessors of most ‘large-ticket’ tangible assets are automatically placed in a similar tax position to a lender who advances a loan in a principal amount equal to the fair market value of the leased property.67 For assets exempt from these rules and for smaller ticket items, the lessor is still entitled to claim the full depreciation allowance if the transaction is characterised as a lease and not a purchase by the lessee.68 In the past, the tax authorities applied an economic substance characterisation approach similar to that reflected in the PPSAs. However, the Supreme Court of Canada in the late 1990s ruled that the economic realities of a transaction cannot be used to re-characterise a taxpayer’s legal relationships.69 Since then, absent an explicit legislative provision (such as the ‘specified leasing rules’), or a finding that the transaction is a sham aimed at the evasion of tax, the taxpayer’s formal choice of legal relationships generally must be and has been respected.70

64

BIA, s 65.1(4)(a); CCAA, ss 11.01, 34(4)(a). seminal case is Re Smith Brothers Contracting Ltd (1998) 53 BCLR (3d) 264 (BCSC). See also International Wallcoverings Ltd (1999) 28 CBR (4th) 48; Re Philip Services Corp, Re Sharp-Rite Technologies Ltd, Re PSINET Ltd (2001) 26 CBR (4th) 288 (Ont SCJ); Produits Forestiers Labrievile Inc (Faillite), Re, 2004 CanLII 21213 (QC CS); Re Winnipeg Motor Express Inc (2009) 56 CBR (5th) 265 (Man QB); leave to appeal refused (2009) 59 CBR (5th) 36 (Man CA [In Chambers]); Royal Bank of Canada v Cow Harbour Construction Ltd 2012 ABQB 59 (CanLII). 66 See the extensive analysis in Royal Bank of Canada v Cow Harbour Construction, ibid. 67 For a more detailed analysis and references, see Walsh (n 58) 225–27. 68 ibid. 69 Shell Canada Ltd v Canada [1999] 3 SCR 622 at 641. And see Continental Bank Leasing Corp v Canada [1998] 2 SCR 298 at para 21, per Bastarache J. 70 See, for example, the complicated leveraged sale/leaseback arrangement upheld in Canada Trustco Mortgage Co v Canada [2005] 2 SCR 601, 2005 SCC 54 (CanLII). 65 The

38

Catherine Walsh

Consider next the financial accounting and reporting standards applicable to Canadian companies and not-for-profit organisations set by the Accounting Standards Board (AcSB).71 Since 2011, the AcSB standard has been commensurate with the standard issued by the International Accounting Standards Board (IASB). Until recently, the relevant IASB standard (IAS 17) drew a dividing line for accounting purposes between operating leases and finance leases broadly equivalent to the PPSA distinction between true leases and security leases. In practice, structuring a lease as an operating lease was widely recognised—and indeed marketed—as a method of off balance sheet accounting. In an effort to enhance transparency for users of financial statements,72 the IASB revised its lease accounting standards with effect in 2016.73 The new standard requires lessees to report all leases on their balance sheets in a manner similar to the traditional treatment of financing leases (with the exception of short-term leases of 12 months or less and leases of small ticket assets such as laptops and office furniture). The new approach has the merit of avoiding the difficult characterisation issues inherent in determining whether and when a lease agreement involves a deemed transfer of ownership while also responding to the transparency concerns that motivated the reform. Consider finally the criteria adopted by the Canadian Bank Act for the purpose of ensuring that the leasing activities in which banks may engage is limited to transactions in which their principal role is the extension of credit to the lessee to finance the acquisition of the leased asset.74 Rather than adopting an amorphous ‘economic substance’ approach to achieve this goal, the Act sets out conservative bright line quantitative and qualitative criteria, thereby ensuring that a bank’s involvement in lease financing is sufficiently close to traditional bank lending activity to be compatible with Canada’s conservative prudent approach to bank regulation.75

VI. Conclusion The PPSAs and the UNCITRAL Model Law adopt an economic substance approach to the characterisation of leases in order to ensure that secured creditors cannot utilise that form of financing as an artificial escape device from the

71

For details and references, see Walsh (n 58) 229–31. For example, a Canadian study concluded that bankers considered the traditional lease accounting standard to be inadequate and felt that requiring lessees to recognise their future lease liabilities under operating leases on their balance sheets would improve their ability to evaluate credit risk: S Durocher, ‘Canadian Evidence on the Constructive Capitalization of Operating Leases’ (2008) 7 Accounting Perspectives 227–256, doi:10.1506/ap.7.3.2. 73 IFRS 16 (IASB 2016). A detailed summary is available at www.ifrs.org/Current-Projects/IASBProjects/Leases/Documents/IFRS_16_project-summary.pdf. See also Giner and Pardo (n 42). 74 For details and references, see Walsh (n 58) 232–34. 75 ibid. 72

‘Functional Formalism’ in the Treatment of Leases

39

andatory requirements of secured transactions law. However, the distinction m between true leases and security leases is not subject to a bright line determination, giving rise to the potential for complex inefficient characterisation disputes. In view of this risk, jurisdictions contemplating the adoption of the UNCITRAL Model Law might consider modifications along the lines of the approaches taken in the PPSAs or the Civil Code of Quebec. As between the two approaches, it is suggested that the Quebec solution achieves at least as appropriate a balance between efficiency and fairness as the PPSAs. Under both regimes, all leases are required to be publicised by registration, thereby protecting the information interests of third parties dealing with the leased assets. On the other hand, at the level of enforcement, Quebec law obligates lessors to extend to all consumers equivalent or even superior protection to that available only to lessees under a security lease in the PPSA jurisdictions. Whatever increased burden this imposes on Quebec lessors in the consumer context is counterbalanced by the elimination of any potential for characterisation disputes in exercising their enforcement rights and by the preservation of their traditional robust repossession rights in the business lease context. It is also compatible with underlying policies of secured transactions law favouring certainty and predictability whilst still ensuring that particularly vulnerable debtors are accorded significant protection. In basing differences in the enforcement rules applicable to lessors on the relatively bright line distinction between consumer and business lessees as opposed to the more amorphous distinction between security leases and true leases, it is also consistent with the general trend away from an economic substance approach to the treatment of leases in other areas. In offering alternative ‘functionalist’ models to the issue of the extent to which the principles underlying secured transactions obligations should extend to leases, this article has sought to pay homage to the work of Rod Macdonald in the field of secured transactions law. While his prodigious contributions to national and international law reform and legal harmonisation are justly celebrated, he was also a conscientious objector to universal models. His writings invariably championed according equal respect to different solutions predicated on differences in local context and policy preferences and consistently challenged the idea that a functionalist perspective dictated only one legal response.76

76 These themes are particularly prominent in the articles cited above as well as in: RA Macdonald, ‘A Model Law on Secured Transactions. A Representation of Structure? An Object of Idealized Imitation? A Type, Template or Design?’(2010) 15 Revue du droit uniforme 419. For appreciations of Rod Macdonald’s secured transactions scholarship, see C Walsh, ‘The Priority of Distributive Justice: Roderick Macdonald and the Reform of Secured Transactions Law’ and JD McCamus ‘Law Reform and Distributive Justice: The Secured Transactions Scholarship of Rod Macdonald’ in R Janda, R Jukier and D Jutras (eds), The Unbounded Level of the Mind: Rod Macdonald’s Legal Imagination (McGill-Queen’s University Press, 2015) chs 12, 13.

40

 

3 Reflections on Misgivings about a Model Law NEIL B COHEN*

Almost 150 years ago, Anthony Trollope wrote about the path of law reform: Many who before regarded legislation on the subject as chimerical, will now fancy that it is only dangerous, or perhaps not more than difficult. And so in time it will come to be looked on as among the things possible, then among the things probable;—and so at last it will be ranged in the list of those few measures which the country requires as being absolutely needed. That is the way in which public opinion is made.1

While Trollope was certainly not writing about the progress of secured transactions law, his observation fits the topic remarkably well. After all, it was only a generation ago that no less an authority than Professor Ulrich Drobnig despaired of the likelihood of success of secured transactions reform in the international sphere.2 Yet, significantly aided by the work of the United Nations Commission on International Trade Law (UNCITRAL), particularly its Legislative Guide on Secured Transactions (the ‘Legislative Guide’),3 and by the work of such influential scholars and law reformers as Rod Macdonald, modernisation of secured transactions law is a phenomenon that is sweeping the world. Once the province of a small number of States with a common law heritage, the move toward modern registry-based secured transactions law is reaching a fever pitch. At UNCITRAL’s Third International Colloquium on Secured Transactions in 2010,4 one of the panels was devoted to the possibility of using UNCITRAL’s * Jeffrey D Forchelli, Professor of Law, Brooklyn Law School. While the author had the honour to serve as a member of the United States delegation to the Working Group that produced the UNCITRAL Model Law on Secured Transactions, the views expressed in this article are solely those of the author and should not be attributed to the delegation or to the United States. 1 Anthony Trollope, Phineas Finn (1869). 2 See the report of Professor Drobnig appearing at Report of the Secretary-General: Study on Security Interests, (1977) 8 UN Commission on International Trade Law YB 171, 173, UN Doc A/CN.9/ SER.A. 3 The Legislative Guide may be found at www.uncitral.org/pdf/english/texts/security-lg/e/09-82670_ Ebook-Guide_09-04-10English.pdf. 4 See www.uncitral.org/pdf/english/colloquia/3rdSecTrans/General_Program.1-3.March_2010_ Final_FINAL.pdf.

42

Neil B Cohen

Legislative Guide as the basis of a Model Law on Secured Transactions to be prepared by UNCITRAL. A majority of the panelists on that panel, and all of the remarks from the attendees, expressed approval, usually quite enthusiastic, of the idea. While the attendees who spoke were unanimous, the panel was not. Two speakers on the panel dissented from the view that a Model Law on Secured Transactions should be prepared. The first of those dissenting panelists answered the question of whether such a model law should be prepared with a clear ‘no’. The other dissenting panelist, while not expressing a final conclusion, expressed serious concerns about the venture. The first of those two speakers was Rod Macdonald;5 the second was the author of this essay. The two dissents differed in many respects, reflecting our different styles. (They also differed in that only one of us brought Rod’s brilliance to the table.) In addition, Rod’s objections were quite conceptual, while my qualms about the possible endeavour were a bit more pragmatic. This chapter: (i) compares the two sets of misgivings as to whether to prepare a model law; and (ii) reexamines them in light of the now nearly-completed UNCITRAL Model Law.6 Before examining and comparing Rod’s and my 2010 misgivings as to the proposal that UNCITRAL prepare a model law, however, it is important to note that, despite my scepticism, there were, and are, strong reasons in favour of preparing a model law. These reasons include the following: 1. a model law might help complete the Legislative Guide’s mission to serve as a guide to States that are considering secured transactions reform. While the analyses provided in the commentary portion of the Legislative Guide, and the recommendations that flow from them in the Terminology and Recommendations portion of the Guide, are quite well-crafted, the Legislative Guide and its 248 recommendations (including those in the Supplement on Security Rights in Intellectual Property) are daunting even for experts in the field. The time and expertise that would be required to transform those recommendations into a legislative enactment might not be available in all States and, more important, might be unavailable in States with the greatest need for secured transactions reform; 2. second, and somewhat related to the first argument in favour of an UNCITRAL Model Law, a model law produced by UNCITRAL may be more proficient technically and lead to higher quality domestic law than a domestic law drafted by local actors solely from guidance of the Legislative Guide. While the recommendations in the Legislative Guide are in a form 5 An expanded version of Rod’s remarks can be found at RA Macdonald, ‘A Model Law on Secured Transactions: A Representation of Structure? An Object of Idealized Imitation? A Type, Template or Design?’ (2010) Uniform Law Review 419. 6 The text of the Model Law is complete and approved by the Commission. As of this writing, however, UNCITRAL is still in the process of preparing the accompanying Guide to Enactment.

Reflections on Misgivings about a Model Law

43

that is, in many ways, similar to statutory language (in some legal cultures, at least), there is still a considerable difference between the language of the recommendations and the more precise needs of legislation. Moreover, there is always the possibility that the recommendations could be inadvertently transformed into legislation that does not fulfil their intent. The creation of a model law based on those recommendations can lessen the possibility of such inadvertent drafting issues that could lead a statute that does not work as intended; 3. third, in the absence of a model law, it is possible that some States that would otherwise consider secured transactions reform are aware of the pitfalls described in the previous two paragraphs and, accordingly, would conclude that they lack sufficient domestic expertise in modern concepts of secured transactions law to assemble a workable statute from the recommendation of the Legislative Guide. This could certainly serve as a disincentive for such a State to modernise its secured transactions laws along the lines recommended in the Legislative Guide. On the other hand, complementing the Legislative Guide with a model law—the availability of which could instil confidence that a domestic enactment of the recommendations of the Legislative Guide in a coherent and consistent manner is possible—could increase the number of States that modernise their law to be consistent with the principles recommended by the Legislative Guide; 4. a fourth consideration favouring the preparation of a model law based on the Legislative Guide does not relate to the difficulties of States in implementing the recommendations of the Legislative Guide. Rather, it relates to the fact that the Legislative Guide, like all human endeavours, is imperfect. In particular, some legal rules that are necessary for a complete secured transactions regime are not addressed in the Legislative Guide. These lacunae could prove troublesome for States that seek to transform the Legislative Guide into domestic legislation. The process of transforming the Legislative Guide into a model law, however, would probably force UNCITRAL to engage in the sort of detailed analysis that will lead to the discovery of such gaps and to the preparation of statutory language to resolve them. Resolution of those issues by UNCITRAL, rather than by individual States, would probably result in greater consistency and coherence; 5. finally, an UNCITRAL Model Law would be promulgated by an international body with work methods that require consensus in order to obtain approval. The instrument resulting from those work methods might prove to be more acceptable than similar efforts promulgated by organisations that develop their products in a manner that is less participatory or that use compliance with their recommendations as a factor in determining aid or assistance to a State. (There is, of course, a down side to consensus-based approaches. They often lead to lowest-common-denominator approaches that can dilute the effect of the clear policies that they seek to further.)

44

Neil B Cohen

I. Rod’s Misgivings Rod’s analysis of the advisability of preparing a model law was typically thoughtful. It consisted, in major part, of three questions, each of which he answered ‘no,’ leading him to provide a negative answer to the advisability question. First, Rod asked ‘Is a Model Law Desirable?’7 Rod’s answer was simple and direct: ‘No. There has never been a “one size fits all” multijurisdictional model law that has really succeeded—unless you count adoption by multiple legislatures of sub-national units.’8 (The latter reference can be read as referring to such successes as the Uniform Commercial Code in the United States and the Personal Property Security Acts in the Anglophone provinces of Canada.) Rod very carefully distinguished, however, between the preparation of a single model law, embracing a one-size-fits-all assumption, and the preparation of a set of model laws from which an enacting state could choose that which fit its legal system best: ‘Indeed, the answer might be different if the question were posed in the plural. For example, it might be that an architect’s sketch document like a Legislative Guide could be transformed into multiple model laws more in the order of an architect’s blueprints, each reflecting a different, contextually sensitive approach to the legislative art.’9 Even that would be fraught with difficulty, according to Rod, leading to him end this portion of his analysis by wondering ‘whether UNCITRAL should take the lead in developing these diverse model laws, or whether the optimal strategy would be to provide technical assistance to individual States, or States that voluntarily come together in regional groupings (for example, Mercosur, OHADA) to work towards enactment of similar national laws.’10 Second, Rod asked ‘Is a model law feasible?’11 Once again, Rod’s answer was in the negative: No. Experience teaches that a transnational model law is feasible in only three situations: (1) where the idea is to create an international normative regime (as in the Convention on mobile equipment financing); (2) where the model creates a regime that deals with a relatively new field not subject to widespread or detailed regulation, and where the goal is to relieve national legislatures of the burden of statutory development (as would be the case if there were a project meant to structure basic principles relating to electronic commerce); or (3) where the model law is aimed at providing a short, specific, patch on an relatively closely defined existing framework of national legislation (as might have been the case of a short model law on, say, letter of credit financing). None of these situations obtain in respect of the Legislative Guide on Secured Transactions.12

7

See Macdonald (n 5) 421 ff. ibid 421. 9 ibid 422. 10 ibid. 11 ibid. 12 ibid. 8

Reflections on Misgivings about a Model Law

45

Rod continued his observation by predicting that, if UNCITRAL were to go forward despite the absence of conditions suggesting feasibility, another risk would be present: [E]very time one seeks to transform an existing chirographic normative instrument (whether a series of judicial decisions, or a practice manual, or a legislative guide) into another form of chirographic instrument such as a model law, one runs the risk of reopening debate on controversial policy issues.13

As noted below, I, too, expressed concerns about reopening (or, as I put it, relitigating) policy questions. Rod, however, took the concern one step further, suggesting that: Whatever one may think of the specific recommendations of the Legislative Guide, it is important to recall that they reflect the best available compromise among UNCITRAL member States. Many such compromises were reluctantly made and consensus was occasionally fragile. Should States seek to revisit policy choices made in the Legislative Guide, UNCITRAL would be put on the horns of a dilemma: either to accept a revisiting of these policy choices, the consequence of which would be to end up sponsoring two projects (a Legislative Guide and a Model Law) that point in different directions, or to refuse to entertain a reopening of policy questions, in which case support for the new project (the Model Law) would likely dissipate and continuing commitment to the Legislative Guide might also be compromised.14

Third, Rod asked ‘Is the timing right?’15 Yet again, Rod’s answer was negative. Rod wrote that: Any particular law reform project succeeds when there is a happy coincidence of supply and demand. A failure of equilibrium on either side of the balance scale is a recipe for failure. Currently, in the realm of secured transactions law, there is an oversupply of model laws: the model law of the European Bank for Reconstruction and Development, the model law of the Organisation pour l’Harmonisation en Afrique du droit des affaires (OHADA), the von Bar Group’s Draft Common Frame of Reference, and the Inter-American model law of the Organization of American States (OAS).

*** One might also observe that there is an under-demand for model laws.16

Rod argued that, in addition to oversupply and under-demand of model laws, the timing of an UNCITRAL Model Law would meet the needs of few states. Some States, he observed, were on the verge of enacting a reformed law. For States in that situation, Rod argued, ‘it is too late to put a process on hold while waiting for a model law, although a document like the Legislative Guide can provide very

13

ibid 423.

14 ibid. 15 ibid. 16 ibid.

46

Neil B Cohen

valuable insight as to how existing proposals may be improved at their margins.’17 Other States, he observed, such as those of the European Union and Organization of American States, likewise have sponsored unofficial attempts to produce model laws for their Member States. For States in the second situation, he asserted that an UNCITRAL Model Law would not be as helpful as the Legislative Guide unless it actually incorporated local practices into its basic framework.18 A third category of States, wrote Rod, are States that are just beginning to consider whether (and how) to modernise their secured transactions law. For States in the third category, Rod concluded, ‘it is too early in the legislative reform process for them to benefit from a model law.’19

II. My Misgivings The concerns that I expressed at the UNCITRAL Colloquium20 were somewhat different than those expressed by Rod. First, I noted that production of a model law based on the recommendations of UNCITRAL appearing in the Legislative Guide would be a substantial undertaking. While, at the time, many participants sometimes jokingly remarked that all that would be necessary in order to transform the Legislative Guide to a model law would be to remove the phrase ‘The law should provide that’ at the beginning of each recommendation, we all knew (or should have known) that this was far from the truth. Rather, the existence of the recommendations of the Legislative Guide would not mean that the transformation of those recommendations to a model law would be simple and speedy. Rather, the nature of legislative language would require slow, painstaking effort notwithstanding the quality and comprehensiveness of the existing recommendations in the Legislative Guide. I noted that, while the substantial talents of the UNCITRAL Secretariat could be expected to result in the production of an early draft that would provide an excellent starting point for creation of a model law, those talents would not be a substitute for the line-by-line vetting of such a draft by the delegates and observers who bring experience from a wide variety of perspectives to analysis of the draft. Yet, the marginal gain that a model law would provide beyond the significant benefits already accruing from the Legislative Guide might be insufficient to justify such a substantial effort. After all, the Legislative Guide, standing alone, already provided not only extensive analysis of most secured transactions issues that occur with some frequency but also provided detailed recommendations for their resolution. Thus, a State wishing to enact the

17

ibid 424.

18 ibid. 19 ibid.

20 An outline of my remarks can be found at www.uncitral.org/pdf/english/colloquia/3rdSecTrans/ Neil_Cohen_Edited.pdf.

Reflections on Misgivings about a Model Law

47

recommendations of the Legislative Guide was already brought very close to the ‘finish line’ by the existing instrument. While a model law would certainly be a useful guide to legislative enactment of the policy recommendations in the Legislative Guide, the time and effort that would necessarily be expended in creating a model law that would be enactable in many States with varying legal cultures and legislative styles would almost certainly be greater than the time and effort needed to transform the recommendations of the Legislative Guide to statutory form in the legal, cultural, and economic context of a single State. Second, I noted that, while the likely charge to a Working Group developing a model law from the recommendations of the Legislative Guide would be to implement those recommendations and not change them, it might be difficult to stick to that charge. Rather, there would be a substantial risk that, in the process of transforming the recommendations into statutory form, participants would reopen, or ‘relitigate,’ policy decisions made explicitly or implicitly in the Legislative Guide. Many of those policy decisions were made only after great difficulty, after extended vigorous debate, and after a slow process of consensus-building in both the Working Group that prepared the Legislative Guide and in the Commission. But the preparation of a model law even by the same Working Group (and, of course, the same Commission) would inevitably involve different people, some of whom, not having been involved in the process by which the Legislative Guide was created, might not share its conclusions and, thus, might seek to change them. In addition, I might have added, veterans of the Working Group that produced the Legislative Guide who had thoughtfully advocated positions or provisions that were ultimately not included because they did not achieve consensus support might be tempted to advocate those positions or provisions once again in the Working Group preparing the model law. Such relitigation might openly be presented as an attempt to have the UNCITRAL Model Law adopt different policies than those in the Legislative Guide, but it might also take a more subtle form in which supporters of the policies in the Legislative Guide would favour broad codifications of its recommendations while doubters as to some of those policies would work to transform those recommendations to legislative language as narrowly as possible. In that regard, I expressed concern that such relitigation might not only change the policy decisions expressed in the Legislative Guide but could also cast doubt on whether UNCITRAL instruments are truly based on a consensus reached by the Member States or, rather, are more accurately seen as amalgams of the views of delegates who participated most actively in their preparation. I expressed further concern that deviations from the recommendations of the Legislative Guide in the preparation of a model law could lead to effects (often unanticipated or unintended) beyond the context of the particular recommendations from which the UNCITRAL Model Law might deviate. Because the Legislative Guide develops an interconnected set of recommendations that depend on their relationship with each other for their collective success, isolated changes in transforming a single recommendation to a legislative provision may not, in fact, be isolated in their effects. Even a simple change to one provision can have effects

48

Neil B Cohen

on the results of several other provisions. The result can be to change significantly the economic effect of the law envisioned by the Legislative Guide or to undermine its internal coherence. Third, I expressed concern that an effort whose goal was the production of a single Model Law on Secured Transactions based on the Legislative Guide would force UNCITRAL to make difficult choices about drafting techniques to implement the recommendations of the Legislative Guide and that the necessity of making those choices could cause significant disharmony without adding significant value. I noted that, so long as national legislation is consistent with the recommendations of the Legislative Guide, the harmonisation goals of the Legislative Guide did not require enactment of identical legislation in each State. Rather, most recommendations of the Legislative Guide are capable of codification in a large number of ways. Yet, drafting a single model law would necessarily force the Working Group to make choices about a single ‘correct’ form of codification where no such choice need be made to achieve the goals of the Legislative Guide, and this process could lead to disharmony in the preparation of the model law. Moreover, given the interconnected nature of the recommendations of the Legislative Guide, a choice made in the codification method for a single recommendation can reverberate throughout legislation codifying the Legislative Guide, necessitating additional contentious choices for internal harmonisation purposes. Not only might this process of choosing drafting methodology be fraught with conflict, but the resulting model law might alienate as many States as it attracts by choosing methods of codification significantly different than those that a State would select for itself. Fourth, I expressed concern that substantial differences in domestic styles of legislation and codification could easily make the goal of a single model law unattainable. UNCITRAL’s accomplishment in the Legislative Guide of achieving consensus as to secured transactions policy in a wide array of States with widely differing traditions of domestic secured transactions law was not accompanied by a consensus as to how those policies should fit into the framework of an enacting State’s law. States’ legislative styles differ in many ways. For one thing, some States legislate in great detail while others prefer grand generalities. Also, some States would probably choose to codify secured transactions law in a single free-standing statute while others might choose to codify different provisions of secured transactions law by amendment or addition to existing bodies of law. In light of those differences, a single ‘one-size-fits-all’ model law might satisfy the legislative preferences of some States but be antithetical to the preferences of many others. Indeed, it might be necessary to draft two or more alternative versions of a model law in order to have available for enactment a model that fits the legislative concept of a variety of States. This necessity would, of course, greatly increase the workload of a Working Group preparing a model law (and, I might have added, create a risk of inconsistency between and among alternative versions). Fifth, I expressed some scepticism as to the soundness of the basic premise behind the proposal to transform the Legislative Guide into a model law—that it would encourage the adoption of domestic legislation consistent with the

Reflections on Misgivings about a Model Law

49

Legislative Guide. I noted the absence of evidence that the availability of a model statutory text would promote enactment of domestic statutes consistent with the Legislative Guide more effectively than if no such text were prepared and States could turn only to the Legislative Guide for guidance.

III. Assessment of the UNCITRAL Model Law in Light of Our Misgivings The concerns that I expressed at the 2010 UNCITRAL Colloquium led me to doubt whether the benefits that could result from undertaking the preparation of a model law on secured transactions would outweigh those concerns. Rod was less diffident, concluding that a model law should not be undertaken. Now, the UNCITRAL Model Law has been completed and approved (although the Guide to Enactment is still a work in progress). Accordingly, this is a propitious time to revisit Rod’s and my misgivings. With the benefit of hindsight, those misgivings can be assessed in light of the reality that has followed. Essentially, such an assessment seeks to answer a very simple question: Were Rod and I ‘worry-warts,’ elevating hypothetical concerns over tangible benefits, or were we prescient? Of course, the answer is only partly visible now. In some cases, the assessment can be made only in retrospect from a safe distance in the future. With respect to Rod’s timing concerns, it is undoubtedly clear that some States went forward with reforms after the completion of the Legislative Guide without believing that they could or should wait for the drafting and completion of the UNCITRAL Model Law. Consider, for example, the case of Jamaica, which enacted its Security Interests in Personal Property Act in December 2013,21 well before the completion of the UNCITRAL Model Law and without significant input from the early drafts of it that existed at that time. Had the UNCITRAL Model Law been available at the time that Jamaica’s law was being prepared, certain aspects of that law might have been constructed in a more technically proficient manner. But the proof of Rod’s conjecture will be in the pudding. How many States will now embark on secured transactions law reform with draft legislation based on the UNCITRAL Model Law? One example is provided by Saint Lucia which, with the assistance of the International Finance Corporation (part of the World Bank Group), is in the process of preparing a draft secured transactions law that is based in large part on the UNCITRAL Model Law. Rod’s concern about the ‘risk of re-opening debate on controversial policy issues’ was quite similar to my concern about relitigation. We both worried that the good results of the Legislative Guide, often achieved with great difficulty and with only a fragile consensus, would be lost in a sea of reconsideration. I am happy 21

Jamaica, Act 38 of 2013.

50

Neil B Cohen

to say that such relitigation did not occur in any material way. While, of course, there were occasional stray proposals that were inadvertently inconsistent with the recommendations of the Legislative Guide, the general lack of intentional relitigation was a positive development. This happy result may have been generated, in significant part, by a carefully drafted charge to the Working Group. The charge to Working Group from the Commission was to produce a Model Law consistent with the prior UNCITRAL instruments, including most especially the Legislative Guide,22 and the Working Group substantially honoured that charge, especially for core issues in secured transactions law. Interestingly, there was one quite positive development that might be characterised as relitigation, but of a different nature than the feared phenomenon of those who did not prevail in the Legislative Guide seeking to reverse their defeat in Model Law. During the development of the Legislative Guide, the Working Group was unable to reach consensus on the matter of acquisition finance. While there was general agreement that security rights in assets acquired by use of the credit advanced to enable the grantor to acquire them should have a preferential priority provision, there was significant disagreement as to whether a functional approach23 should be applied to so-called ‘retention of title’ transactions or ‘conditional sales’ (sales in which, by agreement between the parties, the seller of goods retains ‘title’ to the goods until the full purchase price has been paid), with the result that the retained title would be viewed as a security right, albeit a security right entitled to special priority treatment.24 An alternative viewpoint was that retention of title transactions do not create security rights but, rather, should be governed in accordance with their form as sales transactions in which the last step—transfer of title—is delayed until final payment. A compromise of sorts was reached, under which those who preferred the second viewpoint did not accept the characterisation of these transactions as creating security rights but agreed that some, but not all, rules applying to security rights should nonetheless be governed by rules that require the same steps to be taken for effectiveness against third parties that are taken with respect to security rights.

22 The Commission’s charge to the Working Group was to ‘prepare a simple, short and concise model law on secured transactions based on the recommendations of the Secured Transactions Guide and consistent with all texts prepared by UNCITRAL on secured transactions.’ United Nations, Official Records of the General Assembly, Sixty-eighth Session, Supplement No 17 (A/68/17, para 194). 23 The functional approach of the Legislative Guide is ‘that, to the maximum extent possible, all transactions that create a right in any type of asset meant to secure the performance of an obligation (that is, to fulfil security functions) should be considered to be secured transactions and regulated by the same rules or, at least, by the same principles’: Legislative Guide, Introduction, para 62. 24 By way of example, in the United States both a seller’s retention of title to goods to secure the buyer’s obligation to pay the remaining purchase price and a security interest granted to a third-party lender to secure credit extended to enable the borrower to obtain the property subject to the security interest are considered ‘purchase-money security interests’ and are governed by the same legal principles. See UCC, § 9-103 for a rigorous, if hyper-technical, definition of ‘purchase-money security interest.’

Reflections on Misgivings about a Model Law

51

The result of the inability to reach consensus is that the Legislative Guide contains two alternative chapters about acquisition finance. One version, referred to as the ‘unitary approach,’ adopted the first viewpoint described above, under which retention of title transactions are governed as ‘acquisition security rights’—that is, security rights that have preferential priority. The second alternative chapter adopted the ‘non-unitary approach’ described above, in which retention of title transactions are not seen as security rights but are subject to some of the same rules with respect to effectiveness against third parties.25 The Legislative Guide works to make the two alternative chapters as parallel as possible, but the effort is somewhat strained. In the preparation of the UNCITRAL Model Law, however, consensus was reached early in the Working Group’s deliberations that only the unitary approach should be reflected in the Model Law. This ‘relitigation’ did not occur as a result on one side of this debate seeking to impose its view on the other side, or to obtain a newfound ‘prevailing’ view from a differently-constituted Working Group. Rather, it occurred as a result of Working Group members from States that had favoured the non-unitary approach during the preparation of the Legislative Guide concluding that a single approach was preferable in the Model Law and (much to the surprise of many in the Working Group) announcing that the single approach should be the unitary approach under which retention of title transactions and conditional sales are treated as security rights. Thus, the ‘relitigation’ on this point consisted of the proponents of one side with respect to a contested, unresolved issue withdrawing their preferred position in favour of a unified consensus. This is certainly not the sort of relitigation about which Rod and I had been worried. Rod’s concerns about the desirability of a single model law are, in many ways, similar to my concern that forcing a single style of codification could lead to disharmony and decreased likelihood of success. It is difficult to say at this juncture whether UNCITRAL’s decision to have a single one-size-fits-all codification of the principles in the Legislative Guide will prove to be an advantageous spur toward international uniformity or a stumbling block in the way of enactment in States whose legislative styles differ in important ways from the style of the Model Law. The UNCITRAL Model Law, in my view, adopts a style of codification that fully replicates neither the precise answer-oriented style of Article 9 of the Uniform Commercial Code in the United States and in many States whose legislation is modeled after or inspired by Article 9, nor the broader conceptual statements, from which detailed answers can be reached by deductive reasoning, more typical of civil law states. Perhaps this will be seen in a positive light as a ‘grand compromise’ of sorts, a sui generis drafting technique that does not cast its lot with the traditions of any particular State or group of States. On the other hand, it might be viewed somewhat negatively by potential enacting States, which

25 For a brief introduction to the unitary and non-unitary approaches, see Legislative Guide, Introduction, para 111.

52

Neil B Cohen

collectively conclude that, rather than fitting all, one size fits none. In other words, the intermediate drafting style of the UNCITRAL Model Law, rather than having the virtues of neutrality, might be unsatisfactory to all inasmuch as it matches the typical drafting style of few, if any, States. If so, the result might be a significant amount of redrafting at the enacting State level, perhaps by actors not as familiar with the context of secured credit as those who developed the Legislative Guide and UNCITRAL Model Law.

IV. A Final Appreciation and Concern The UNCITRAL Model Law resulted from an enormous amount of high-quality technical drafting in a remarkably short time. This laudable achievement probably occurred both more quickly and more proficiently than could have been done at the State level in many States. It is the product of a dedicated and talented Secretariat and of the engaged involvement of many talented members of the Working Group. In many ways, this happy result validates reasons for going forward described above26 and might be seen as outweighing the misgivings expressed by Rod and by me. There is one downside, however, to the remarkable speed with which the UNCITRAL Model Law was achieved—in the last stages of preparation, there was limited time to address concerns raised about clarity and precision of several of the provisions. (It is a natural, if unfortunate, phenomenon that such issues are noticed much more at the end of a legislative drafting process, when more difficult policy and conceptual issues have retreated from prominence.) The limited time that was available to address such issues, combined with the consensus-oriented working methods employed by the Working Group that provided the Secretariat with little room to improve expression of ideas without unanimous acquiescence or the development of a prevailing view after discussion for which little time was available, necessarily prevented full consideration of many of those issues. The result is a Model Law that, for the most part, reflects a high level of technical proficiency in translating the challenging concepts of the Legislative Guide to the more precise wordings necessary in a statute but which, in some cases, could have benefited from either a bit more time to catch drafting concerns and address them without the effort being seen as obstruction of progress or, perhaps, a bit more informal power being given to the Secretariat to address such non-policy matters without time-consuming processes of approval. On balance, however, it is my judgement that the benefits flowing from the generally high technical proficiency characteristic of the UNCITRAL Model Law outweigh the down side of the occasional provision that could benefit from additional

26

See pp 42–43 above.

Reflections on Misgivings about a Model Law

53

work. After all, without that high quality, albeit imperfect, work of UNCITRAL, the Model Law would add little usable value. Moreover, small errors can be corrected by enacting States and those who advise them. So, were Rod and I correct in our misgivings? Certainly our biggest concerns (the risks of relitigation, undermining of the value of the Legislative Guide, and the possible creation of disharmony) did not take hold, and I am glad that events have dispelled those concerns. As for the acceptability and usefulness of the final product in a variety of legal cultures, at this point the question is more empirical than theoretical. As the saying goes, time will tell. In the meantime, I am rooting for the product’s success and will be quite happy if the remaining concerns do not come to fruition.

54

 

4 The UNCITRAL Model Law on Secured Transactions SPYRIDON V BAZINAS*

I. Introduction At its forty-ninth session in 2016, the UN Commission on International Trade Law (UNCITRAL or the Commission)1 adopted the UNCITRAL Model Law on Secured Transactions (the ‘Model Law’).2 At that session, the Commission also asked Working Group VI (Security Interests) to complete its work on a Draft Guide to Enactment of the Model Law.3 The Model Law is based on the recommendations of the UNCITRAL Legislative Guide on Secured Transactions (the ‘Legislative Guide’), including the Supplement on Security Rights in Intellectual Property (the IP Supplement), the UNCITRAL Guide on the Implementation of a Security Rights Registry (the ‘Registry Guide’), and the provisions of the United Nations Convention on the Assignment of Receivables in International Trade (the ‘CARIT’).4 The purpose of this chapter is to briefly discuss the

* Spyridon (Spiros) V Bazinas is Senior Legal Officer at the International Trade Law Division of the Office of Legal Affairs of the United Nations, which functions as the Secretariat of the United Nations Commission on International Trade Law. The views expressed in this article are the personal views of the author and do not necessarily reflect the views of the UN or UNCITRAL. This chapter is based on ch 23 of L Gullifer and O Akseli (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Oxford, Hart Publishing, 2016). In that chapter (see pp 482, 484), I respond to doubts expressed by my good friend Rod Macdonald as to whether a model law on secured transactions would be desirable or feasible. I now believe that Rod smiles approvingly upon us from heaven. 1 For the origin, mandate, composition and methods of work of UNCITRAL, see www.uncitral.org/ uncitral/en/about_us.html. 2 See the report of the Commission at its forty-ninth session (A/71/17, para 119). The UNCITRAL Model Law is available at www.uncitral.org/uncitral/en/uncitral_texts/security/2016Model_secured. html. 3 See the report of the Commission at its forty-ninth session (A/71/17, para 122). The current version of the Draft Guide to Enactment is contained in documents A/CN.9/WG.VI/WP.71 and Addenda 1 to 6, which are available at www.uncitral.org/uncitral/en/commission/working_groups/6Security_ Interests.html. 4 See www.uncitral.org/uncitral/uncitral_texts/security.html.

56

Spyridon V Bazinas

UNCITRAL Model Law and compare its provisions with the recommendations of the Legislative Guide, including the IP Supplement, and the Registry Guide.

II. Key Objectives and Fundamental Policies The overall objective of the Model Law is to promote low-cost credit by enhancing the availability of secured credit and is the same as that of the Legislative Guide.5 The same applies to the other key objectives and fundamental policies set out in the Legislative Guide (eg the unitary, functional and comprehensive approach).6 It is left to enacting States to decide whether to include references to those key objectives and fundamental policies in a preamble to their enactment of the Model Law or other statement that could be used for the purpose of the interpretation of and filling of gap left in their enactment (see III.F below).7

III. Scope and General Provisions A. The Unitary, Functional and Comprehensive Approach Generally, the scope of application of the Model Law is as broad as the scope of the Legislative Guide. Thus, the Model Law: (a) uses a single, unitary concept of ‘security right’ rather than several terms (eg, pledge, hypothec, etc); (b) applies to all types of right in movable property created by agreement to secure payment or other performance of an obligation (eg, transfer of title for security purposes, retention-of-title sale or financial lease); and (c) foresees that security interests may: (i) secure all types of obligation, present or future, determined or determinable; (ii) encumber assets described specifically or generally, or even all of the assets of a grantor, present and future, including a changing pool of assets; and (iii) be created or acquired by any legal or natural person, including a consumer, subject to consumer protection law.8 This unitary, functional and comprehensive approach to secured transactions is not dictated by ideological considerations or preferences for one or the other national legal system. It is a practical response to the main problem of secured transactions laws around the world, that is, the fragmentation of secured transactions law into multiple laws dealing differently with the transactions that fulfil security functions. The result of this fragmentation is the creation of gaps and 5 Legislative Guide, Recommendation 1(a) (www.uncitral.org/uncitral/en/uncitral_texts/security/ Guide_securedtrans.html). 6 Legislative Guide, Recommendation 1 and paras 60–72. 7 Draft Guide to Enactment (A/CN.9/WG.VI/WP.71) paras 29 and 30. 8 Legislative Guide, Recommendation 2 and DML, Art 1.

The UNCITRAL Model Law on Secured Transactions

57

inconsistencies, which cannot but have a negative impact on the availability and the cost of credit (even if judges, practitioners and business people do their best to minimise the negative impact). In addition, the comprehensive, unitary and functional approach is the approach typically followed in modern secured transactions legislation.9 Moreover, it is the approach that facilitates harmonisation of the laws of various States as it deals in a comprehensive and similar way with all types of security interest. However, the unitary, functional and comprehensive approach does not mean that borrowers, such as consumers, that deserve special protection are not to be protected. In line with the Legislative Guide, the Model Law provides that secured transactions law cannot affect the rights of consumers under consumer protection law.10 In addition, the comprehensive, unitary and functional approach does not mean that unsecured creditors are not protected. In line with the L egislative Guide, with the exception of limitations based solely on the ground that an asset is a future asset, the Model Law preserves statutory limitations to the creation or enforceability of a security interest in certain types of asset (eg, household items or employment and retirement benefits, at least up to a certain amount).11 Unsecured creditors may also be protected by way of statutory privileges and, in line with the Legislative Guide, the Model Law only recommends that they should be limited and set out in the law in a transparent way for parties to be able to take them into account when deciding whether to enter into a transaction and at which terms.12 Finally, the unitary, functional and comprehensive approach does not mean that retention-of-title sales and financial leases need to be re-characterised as secured transactions for all purposes (ie, tax, accounting, etc). It is sufficient to subject them to secured transactions law for its limited purposes (ie, creation, third-party effectiveness, priority and enforcement). Nor does coverage of these devices in the Legislative Guide or the Model Law mean that providers of goods on credit cannot be protected. In fact, in line with the Legislative Guide, the Model Law provides that, once they register within a short period of time after delivery of the goods, providers of goods of credit are given a special priority over general financiers.13

B. Differences in the Scope of the Model Law and the Scope of the Legislative Guide As already mentioned, generally, the scope of the Model Law is the same as the scope of the Legislative Guide. However, there are certain exceptions. First, unlike

9 See, for example, the new PPSA of Australia, available at www.comlaw.gov.au/Details/ C2012C00151. 10 Legislative Guide, Recommendation 2, sub-para (b), and Model Law, Art 1, para 5. 11 Legislative Guide, Recommendation 18 and Model Law, Art 1, para 6. 12 Legislative Guide, Recommendations 83 and 239 and Model Law, Arts 35 and 36. 13 Legislative Guide, Recommendations 180 and 182 and Model Law, Arts 38 and 39.

58

Spyridon V Bazinas

the Legislative Guide, the Model Law deals with security interests in non- intermediated securities (eg shares and bonds that are not held in a securities account). The reason is that such securities are regularly used in commercial finance transactions and yet neither the Legislative Guide nor other texts prepared by other organisations deal with such securities.14 Second, unlike the Legislative Guide, the Model Law does not deal with security interests in letters of credit. The reason is that letter of credit financing is a very specialised and narrow area that is subject to specialised regulation and does not need to be addressed in the Model Law. States interested in letter of credit financing can still find some guidance in the Legislative Guide. Third, while both the Legislative Guide and the Model Law apply to all receivables financing transactions whether they involve the creation of a security interest or not, a transfer for security purposes or the outright transfer of receivables, enacting States may exclude certain types of outright assignment of receivables that are not financing transactions (eg, an outright transfer of receivables for collection purposes or as part of the sale of the business out of which they arose).15 Fourth, unlike the Legislative Guide, the Model Law does not deal with security interests in attachments to movable or immovable property. The reason is that, while attachments are important, the recommendations of the Legislative Guide provide sufficient guidance for the benefit of States that would like to deal with attachments in their law.

C. International Obligations of the Enacting State As already mentioned, with the exception of limitations based solely on the ground that an asset is a future asset, the Model Law does not override statutory limitations to the creation or enforceability of a security interest in certain types of asset (eg, household items or employment and retirement benefits, at least up to a certain amount).16 In addition, the Model Law leaves to enacting States the issue whether a treaty or other form of international agreement prevails over domestic law (the Legislative Guide did not deal with this matter either).17

14 See Report of Working Group VI on the work of its twenty-fifth session (A/CN.9/802) paras 72 and 73 (www.uncitral.org/uncitral/en/commission/working_groups/6Security_Interests.html). The Unidroit Convention on Substantive Rules for Intermediated Securities (2013) and the Convention of on the Law Applicable to Certain Rights in Respect of Securities held with an Intermediary (206) do not deal with non-intermediated securities. 15 Draft Guide to Enactment (A/CN.9/WG.VI/WP.71) Add 1, para 2. 16 Legislative Guide, Recommendation 18 and UNCITRAL Model Law, Art 1, para 6. 17 Draft Guide to Enactment (A/CN.9/WG.VI/WP.71) Add 1, para 42.

The UNCITRAL Model Law on Secured Transactions

59

D. Party Autonomy The Model Law generally deals with party autonomy in the same way as the Legislative Guide. Accordingly, the parties may derogate from or vary provisions of the Model Law with the exception of provisions dealing with the general standards of conduct, the requirements for a valid security agreement and the description of the encumbered assets, the priority rules relating to the priority of acquisition security interests, the obligation of a person in possession of an encumbered asset to exercise reasonable care, the obligation of a secured creditor to return an encumbered asset or to register an amendment or cancellation notice, certain provisions relating to enforcement, and the conflict-of-laws rules (with the exception of the provision dealing with the law applicable to the contractual rights and obligations of the parties).18 The Model Law also includes a new provision to ensure that any agreement of the parties to use any alternative dispute resolution mechanism is not affected by the Model Law.19 This provision is not intended to address the question whether disputes arising from security agreements may be referred to arbitration or to address third-party rights where a secured creditor may seek to seize an encumbered asset.20 With respect to third-party rights, it should be noted that in the case of judicial enforcement, the Model Law refers the protection of the rights of third parties to the domestic procedural law; and, in the case of extra-judicial enforcement, it includes provisions to protect the rights of third parties (eg Article 76, under which a secured creditor with priority may take over enforcement from the enforcing secured creditor, Article 77, paragraph 2, under which to obtain possession of an encumbered asset out of court, a secured creditor must notify the grantor and any person in possession of the encumbered asset, and Article 78, paragraph 4, to dispose of an encumbered asset out of court, a secured creditor must notify any person with a right in the encumbered asset, and Article 79, subparagraph 2(c), under which any distribution of proceeds of the sale of an encumbered asset must be in accordance with the rules of the Model Law on priority).21

E. General Standards of Conduct The Model Law obligates any person to exercise its rights and perform its obligations in good faith and in a commercially reasonable manner.22 Any l iability to d amages for breach of this obligation is left to the relevant law of the enacting State. 18

Model Law, Art 3, para 1. Model Law, Art 3, para 3. 20 Art 68 of the OAS Model Law on Secured Transactions provides that such disputes may be submitted to arbitration and leaves other matters to the law of the enacting State. 21 See Draft Guide to Enactment (A/CN.9/WG.VI/WP.71) Add 5, paras 55 and 58. 22 Model Law, Art 4. 19

60

Spyridon V Bazinas

The commercial reasonableness of an act depends on the commercial context and relevant best practices. To avoid abuses, the Model Law prohibits a waiver or variation of these standards of conduct by agreement.23 The Legislative Guide contained a similar recommendation but it was applicable only to the enforcement of a security interest.24

F. Interpretation To ensure uniform interpretation of and filling of gaps in the Model Law, it includes a provision along the lines of Article 7 of the Assignment Convention, which is based on Article 7 CISG.25 According to this provision, in the interpretation of the Model Law, regard is to be had to its international origin and the need to promote uniformity; and gaps with respect to issues covered but not settled explicitly in the Model Law should be filled in line with the principles on which the Model Law is based and, in the absence of such principles, in conformity with the law applicable by virtue of the conflict-of-laws rules of the forum. As mentioned above, if included in a preamble or other statement, the key objectives of a modern and efficient secured transactions law set out in Recommendation 1 of the Legislative Guide may serve as a statement of the principles on which the Model Law is based.26

IV. Creation of a Security Interest A. General Rules One of the key objectives of an effective secured transactions law is ‘to enable parties to obtain security rights in a simple and efficient manner’.27 So, in line with the Legislative Guide, the Model Law provides that a security agreement is sufficient for the creation of a security interest, although in the case of assets in which the grantor acquires rights or the power to encumber after the time of the conclusion of the security agreement (‘future assets’), the security interest is created at that later time.28 The Model Law also provides that the security agreement must indicate the intent of the parties, identify them, and describe the secured obligation and the encumbered assets.29 23

Model Law, Art 3, para 1. Legislative Guide, Recommendations 131 and 132. 25 Model Law, Art 5. 26 See n 7. 27 Legislative Guide, Recommendation 1, sub-para (c). 28 Legislative Guide, Recommendations 13 and 14 and Model Law, Art 6, paras 1 and 2. 29 Model Law, Art 6, para 3. 24

The UNCITRAL Model Law on Secured Transactions

61

The Model Law is similarly flexible with respect to the form of a security agreement. A security agreement need be in writing only if it is not accompanied by delivery of the encumbered assets to the secured creditor.30 This approach combines flexibility (no need for written agreement) with the protection of the legitimate interests of the secured creditor (written agreement only if the secured creditor is not in possession of the assets). The flexibility goes even further. First, it is enough if the agreement is concluded or, at least, evidenced in writing. Second, the writing need only indicate the grantor’s intent to create a security interest, that is, the secured creditor’s signature is not required. Third, the writing may lead to that result by itself or in conjunction with the course of conduct of the parties, that is, an order, shipment, delivery and acceptance of the goods may be enough. And fourth, writing includes an electronic communication and signature includes an electronic signature. To facilitate financing practices, such as revolving credit arrangements, the Model Law provides that it should be possible for a security interest to secure any type of obligation, including future, conditional or fluctuating obligations.31 Also, to reduce risks and ensure better credit terms for borrowers, the Model Law provides that it should be possible to create a security interest in future assets, that is, assets created or acquired by the grantor after the creation of a security interest, including all assets of a grantor, present and future.32 In order to facilitate consideration of the matter by the national legislator, the commentary of the Legislative Guide discusses the various approaches taken in different legal systems in this regard to ensure that grantors are not over-committed.33 In order to simplify the creation of security interests in all assets of an enterprise the Model Law provides that a security interest may be created in all assets of a grantor. Characteristic elements of such transactions are two, a security interest may be created in all assets of a grantor with a single agreement and the g rantor has the right to dispose of certain of its assets (eg, inventory) in the ordinary course of its business.34 The Legislative Guide discusses over-collateralisation, suggesting ways in which it could be addressed, but makes no recommendation as the appropriate response may vary widely from State to State.35 In any case, recognising the need to protect certain parties or exclude certain types of asset from the scope of secured transactions law, as already mentioned, the Model Law defers to consumer protection legislation and to legislation according to which certain types of asset (eg, employment benefits or household goods) may not be transferred or encumbered.36

30

Legislative Guide, Recommendation 15 and Model Law, Art 6, para 4. Legislative Guide, Recommendation 16 and Model Law, Art 7. 32 Legislative Guide, Recommendation 17 and Model Law, Art 8. 33 Legislative Guide, ch II, paras 51–55. 34 Legislative Guide, ch II, para 63, and Art 8, sub-para (d). 35 Legislative Guide, ch II, para 69. 36 Model Law, Art 1, para 5 and 6. 31

62

Spyridon V Bazinas

In view of the importance of the required description of the encumbered assets, the Model Law deals with it in a separate provision that implements the policy of the Legislative Guide.37 The assets must be described ‘in a manner that reasonably allows their identification’. What is a reasonable identification of the assets may differ according to the circumstances and the type of asset involved. For example, in the case of inventory, a generic description, such as ‘all my present and future inventory’ or ‘all my present and future inventory of personal computers’ or ‘all my present and future inventory of personal computers in warehouse X’, may be enough. However, in the case of intellectual property, a specific description, such as ‘my patent X’, may be necessary (the same rule applies to the description of the encumbered assets in the notice). The same standard applies to the description of a secured obligation. To protect the secured creditor from unauthorised transfers (in which a security interest follows the asset but the asset may be removed from the secured c reditor’s reach) or transfers in the ordinary course of business or in good faith (in which the transferee takes the asset free of the security interest) by the grantor, the Model Law provides that the security interest automatically extends to any identifiable proceeds of the encumbered assets.38 Even in cases in which the security interest follows the assets in the hands of the transferee, it makes sense for the security interest to extend to the proceeds (where, for example, the proceeds may be of higher value than used assets). This approach ensures that the secured creditor is sufficiently secured and thus is more likely to offer better credit terms to the borrower. However, this does not mean that the secured creditor will ever obtain more than what it is owed.39

B. Asset-specific Rules The general creation rules apply to all types of asset except to the extent modified by specific asset-specific creation rules. First, in line with the Legislative Guide,40 an agreement limiting the creation of a security interest in a receivable does not invalidate a security interest created despite that contractual limitation.41 This essentially means that the agreement between the creditor and the debtor of the receivable is valid as between the parties thereto, but does not have third-party effects. Thus, if by creating a security interest the creditor of the receivable breaches its contract and becomes liable to damages under the relevant law, the Model Law does not override that law. Where the Model Law limits the effect of that

37

Legislative Guide, Recommendation 14, sub-para (d), and Model Law, Art 9. Guide, Recommendation 19 and Model Law, Art 10, para 1. The term ‘proceeds’ is defined to mean ‘whatever is received in respect of encumbered assets’ and includes proceeds of proceeds, as well as natural and civil fruits (see Model Law, Art 2, sub-para (bb)). 39 Legislative Guide, ch II, paras 72–89. 40 Legislative Guide, Recommendation 24. 41 Model Law, Art 13, para 1. 38 Legislative

The UNCITRAL Model Law on Secured Transactions

63

other law is that the debtor of the receivable may not terminate the contract giving rise to the receivable or claim any payment from the third-party secured creditor. In other words, the debtor of the receivable may raise a claim only against, and thus continues to bear the risk of insolvency of, its contractual partner. Second, in principle, rights securing the payment or other performance of an encumbered receivable follow the right they secure without a new act of transfer. However, if an encumbered receivable is supported by an independent guarantee or stand-by letter of credit, the grantor is obliged to transfer the benefit of that right to the secured creditor.42 Third, also in line with the Legislative Guide,43 the Model Law provides that a security interest in a negotiable document extends to the goods covered by the document, provided that the issuer is in possession of the goods at the time of the creation of the security interest.44 And fourth, also in line with the Legislative Guide,45 the Model Law provides that a security interest in assets containing an intellectual property asset, such as inventory of personal computers containing copyrighted software, does not automatically extend to the intellectual property asset and vice versa. This means that, in the absence of agreement, if the grantor defaults, the secured creditor can re-possess and sell the inventory as is, but not use the copyrighted software in other personal computers.46

V. Third-party Effectiveness of a Security Interest A. General Rules Following the approach of the Legislative Guide, the Model Law d istinguishes between the effectiveness of a security interest as between the grantor and the secured creditor, on the one hand, and its effectiveness as against third parties (such as other secured creditors, buyers of the encumbered assets, judgment creditors and the mass of creditors or the insolvency administrator in the grantor’s insolvency), on the other. In this way, the Model Law achieves two of the key objectives of a modern secured transactions law, namely to simplify the creation of a security interest, while enhancing the transparency with respect to security interests.47 In addition, this approach makes it possible for the Model Law to achieve two more of the key objectives of a modern secured transactions law, that is, to

42

Model Law, Art 14. Legislative Guide, Recommendation 28. 44 Legislative Guide, Recommendation 28 and Model Law, Art 16. 45 Legislative Guide, Recommendation 243. 46 IP Supplement, Recommendation 243 and Model Law, Art 17. 47 Legislative Guide, Recommendation 1, sub-paras (c) and (f). 43

64

Spyridon V Bazinas

allow debtors to use the full value of their assets to obtain credit and provide for an equal treatment of diverse sources of credit to all devices serving security purposes, including form-free types of transaction, such as retention-of-title sales and financial leases.48 In this way, the Model Law establishes a comprehensive and rational regime that addresses the problem of uncertainty and inconsistency created by a multiplicity of regimes depending on the formal name of a transaction or the type of asset involved, that is, by a piecemeal approach to secured transactions. This approach does not harm the rights of parties to these title devices. Once a retention-of-title seller registers a notice within a short period of time after delivery of the goods to the buyer, its priority goes back to the time of delivery of the goods. Such a notice may cover a multiplicity of transactions between the same parties for a long period of time. In a cross-border situation, it ensures that the retention-of-title seller will not lose its rights or its priority. Whether the remedies of the seller or the secured creditor are more efficient depends on the specific situation (eg, whether the residual value of the used goods is higher than the part of the purchase price paid by the buyer minus the value of the use of the goods). In legal systems in which a security interest has by nature effects erga omnes, the Model Law may be implemented in the following way: upon its creation, a security interest may be effective against all of the grantor’s creditors as long as the priority is determined on the basis of the time when registration or another third-party effectiveness step occurs.49 The general method for achieving the third-party effectiveness of a security interest is the registration of a notice of the security interest in a publicly accessible registry.50 However, to avoid undermining well-functioning practices, the Model Law recognises that the third-party effectiveness of a security interest may also be achieved: (a) in a tangible asset by a transfer of the possession of the asset to the secured creditor;51 and (b) in an asset, such as a ship, aircraft, patent or trademark registry, by registration in a specialised registry.52 In addition, a security interest in certain types of asset, such as cash proceeds, may be achieved automatically upon the creation of the security interest.53

B. Asset-specific Rules While the general third-party effectiveness rules apply to all types of asset, the Model Law also includes a set of asset-specific third-party effectiveness rules that are intended to accommodate specific financing practices. Thus, the third-party 48

Legislative Guide, Recommendation 1, sub-paras (b) and (d). Legislative Guide, ch II, para 4 and ch III, para 8. 50 Model Law, Art 18, para 1. 51 Model Law, Art 18, para 2. 52 Draft Guide to Enactment (A/CN.9/WG.VI/WP.71) Add 1, para 85. 53 Legislative Guide, Recommendation 39 and Model Law, Art 19, para 1. 49

The UNCITRAL Model Law on Secured Transactions

65

effectiveness of a security interest may also be achieved: (a) in a right to payment of funds credited to a bank account by the transfer of the bank account to the secured creditor, the creation of a security interest in favour of the depositary bank or the conclusion of a control agreement;54 (b) in a negotiable instrument or a negotiable document by transfer of possession of the instrument or the document to the secured creditor;55 and (c) in a non-intermediated security by notation in the books of the issuer or conclusion of a control agreement.56 The main difference between the Model Law and the Legislative Guide relates to the provision dealing with the third-party effectiveness of a security interest in non-intermediated securities. The general methods of third-party effectiveness (notice registration and transfer of possession) apply also to non-intermediated securities. Two asset-specific methods are added with respect to non-intermediated securities, notation in the books of the issuer and a control agreement (among the issuer, the secured creditor and the grantor according to which the issuer agrees to follow instructions from the secured creditor without further consent from the grantor).57

VI. The Registry System A. Notice Registration The UNCITRAL Model Law contains one rule providing for the establishment of a publicly accessible security interests registry (the Registry).58 The details of the registry system are set out in a series of Registry provisions that are based on the recommendations of the Legislative Guide and the Registry Guide. Depending on the legislative approach of each State, the Registry Provisions may be implemented in the secured transactions law, another law, decree, regulation or other act, or a combination thereof. In line with the Legislative Guide, the Model Law avoids imposing unnecessary formalities on the creation of a security interest. Thus, registration is not necessary for the creation of a security interest, but only for its effectiveness against third parties.59 And, as the legal consequence of registration is to make a security interest effective against third parties (provided that there is a valid security agreement), there is no need to register the security agreement but only a notice thereof. As the commentary to the Legislative Guide points out, ‘registries 54

Legislative Guide, Recommendation 49 and Model Law, Art 25. Legislative Guide, Recommendations 37 and 51–53 and Model Law, Art 26. 56 Model Law, Art 27. 57 Model Law, Arts 2, sub-para (g) (i), and 27. 58 Model Law, Art 28. 59 Legislative Guide, Recommendation 34 and Model Law, Art 18, para 1. 55

66

Spyridon V Bazinas

based on a notice-registration concept exist in an increasing number of States and have also attracted considerable international support’.60 It is worth noting that notice-based registration has been so successful that it has been adopted even in the context of specialised (including asset-based) registration systems, such as the registration systems for the types of high-value mobile equipment covered by the Cape Town Convention on International Interests in Mobile Equipment and its Protocols,61 and patent and trademark registries.62 For the same reason, notice registration is even advocated for immovable property registries.63 In such a notice-based registration system, quick and easy registration is ensured by requiring registration of a notice that contains a limited yet sufficient amount of data, that is, the identifier and address of the grantor and the secured creditor or its representative, a description of the encumbered assets and, if permitted by the law, a selection of the period of effectiveness of the registration and, if a State chooses the option offered, the maximum amount for which the security interest may be enforced.64 First, this information is sufficient for the searcher to determine whether some assets of the grantor may be encumbered by a security interest in the sense that it provides a warning. Second, it does more than that, and points the searcher to the source of information about the transaction to which the notice may relate, that is, the secured creditor identified in the notice. Third and most importantly, it provides an objective method for determining priority. Document registration is also discussed in the commentary of the Legislative Guide, but is not recommended. The reason is that a notice-registration system is inherently more efficient. It simplifies the registration process, minimises the administrative burden, delays and costs, reduces the risk of error and liability and is sufficient in view of the legal consequences of registration, that is, third-party effectiveness.65 With regard to the amount of information required in a notice, the Model Law balances the rights and interests of searchers and secured creditors.66 In an effort to ensure the confidentiality of the transaction and avoid use of the registry by

60

Legislative Guide, ch IV, para 14. These registries are asset-based rather than debtor-based registries. Briefly, asset-based registration has the disadvantage that it is necessary to have a unique identification and cannot accommodate a generic description or after-acquired property. But it has the advantage that it shows all registrable interests, not merely those created by the debtor. See Art 31 of the Convention and R Goode, Aircraft Registry Regulations and Official Commentary (Rome, Unidroit, 2008) s 5, at 122. The Aircraft Registry is a particularly successful example of such a registry (see www.internationalregistry.aero/irWeb/ Controller.jpf). 62 IP Supplement, paras 132–134. 63 J Simpson and F Dahan, Mortgages in Transition Economies, Secured Transactions Reform and Access to Credit (EBRD, 2008) 195. 64 Legislative Guide, Recommendation 57 and registry-related provisions Art 9. 65 Legislative Guide, ch IV, paras 10–14. 66 For a comparison of the registry systems under UCC Art 9, the Canadian PPSAs, the OAS Model Law on Secured Transactions and the OAS Registry Regulations, on the one hand, and the Legislative Guide, on the other, see M Dubovec, ‘UCC Article 9 Registration System for Latin America’ (2011) 28(1) Arizona Journal of International & Comparative Law 117. 61

The UNCITRAL Model Law on Secured Transactions

67

lenders to obtain information about the clients of competitors, the Model Law provides that a search be made possible only by the identifier (ie the name) of the grantor (or the registration number), and not by the identifier of the secured creditor.67 In a further effort to ensure confidentiality, the Model Law provides that the secured creditor may choose not to identify itself on the notice but give the identifier and address of a representative.68 The possibility to identify a representative also facilitates secured transactions that involve multiple lenders that appoint one of them to be their representative for the purpose of registration. The description of encumbered assets in a notice may be generic or specific, depending on what is a reasonable identification in each case.69 Whether the notice should describe the encumbered assets by stating their serial numbers, if any, is left to each practice. For example, description of the encumbered assets by serial number should be possible in the case of high-value mobile equipment, such as ships or aircraft, and intellectual property rights, such as patents or trademarks, while such description would be impractical in the case of inventory. With respect to indexing and searching, the Model Law provides that reference should be made mainly to the grantor identifier.70 The reason is that grantor indexing and searching greatly simplifies the registration process to the extent that a single registration can cover a changing pool of assets and future assets.

B. Registry Provisions in the Model Law and the Legislative Guide/Registry Guide Compared Generally, the Registry Provisions follow the policy of the Legislative Guide and the Registry Guide with respect to registration of notices of security interests. Most of the differences are of a drafting nature. However, there are some new rules. First, the Registry Provisions address the impact of a transfer of an encumbered asset on the effectiveness of a registration, that is, whether, in the case of a transfer of an encumbered asset, a new registration is necessary for the security interest to continue being effective against third parties. This issue had been discussed

67 Legislative Guide, Recommendations 54, sub-para (h), and 58–60 and Registry Provisions, Art 22. The Legislative Guide explains that this approach is intended to prevent lenders from using the Registry in order to identify the clients of their competitors. The Legislative Guide also explains that this approach does not prevent a State from designing a Registry so as to permit search by secured creditor identifier for internal purposes of the Registry, such as, for example, for making a global amendment of notices at the request of the secured creditor to whom the notices relate (see Legislative Guide, ch IV, paras 29 and 30). 68 Legislative Guide, Recommendation 57, sub-para (a), and Registry Provisions, Art 8, sub-para (b). 69 Legislative Guide, Recommendation 63 and Registry Provisions, Art 11. 70 The registration number assigned to each registered notice is also a search criterion (see Registry Provisions Art 22, sub-para (b)). The Legislative Guide discusses in the commentary but does not recommend indexing and searching by serial number (see ch IV, paras 31–36). Thus, a legislator may determine whether to adopt grantor indexing, asset indexing or both.

68

Spyridon V Bazinas

in the commentary but not addressed in the recommendations of the Legislative Guide.71 The Model Law offers three options for the enacting State to choose from, one requiring registration of a new notice, another requiring registration of a new notice within a certain period after the secured creditor acquires knowledge of the transfer and another that does not require registration of a new notice, which is the approach followed where the encumbered asset transferred is intellectual property.72 Second, the Registry Provisions address the question whether the secured creditor’s authorisation is required for the registration of an amendment or cancellation notice.73 This question had been discussed in the commentary of the Registry Guide but not addressed in a recommendation as it is a question for the law (not the registry regulations).74 Four different options are offered for discussion, requiring the secured creditor’s authorisation or not with different conditions. Third, the Registry Provisions address two more related issues, correction of errors and limitation of liability of the Registry. Those issues too had been discussed in the Registry Guide but not addressed in a recommendation.75 With respect to correction of errors, options are offered to enacting States, dealing with the following questions: (a) whether the Registry should be able to correct an error itself or inform the registrant to enable the registrant to correct the error; and (b) the time of effectiveness of a notice correcting an error.76 Also with respect to the limitation of liability of the Registry, options are offered to enacting States ranging from limitation of the basis or the amount of the liability to excluding liability altogether.77 Finally, while there is no policy change with regard to the registry fees, the Registry Provisions emphasise the two options foreseen in the Legislative Guide and the Registry Guide.78 The first option foresees fees at cost-recovery level to be specified by the enacting State. It is based on the assumption that one of the key objectives of a modern secured transactions regimes, that is, transparency with respect to secured transactions, cannot be achieved if the Registry is used as an opportunity to generate revenue. The second option foresees no fees. It is based on the assumption that the cost of establishing and operating an electronic registry should be minimal and the cost should be borne by the enacting State. The rationale for this approach is that the Registry is a key component of a modern secured transactions regime, which would enhance the availability of credit at a lower cost,

71 Legislative Guide, ch IV, paras 78–80. Recommendation 62 simply states that the matter should be addressed. 72 Registry Provisions, Art 26. 73 Registry Provisions, Art 21. 74 Registry Guide, paras 249–259. 75 Registry Guide, paras 135–144. 76 Registry Provisions, Art 31. 77 Registry Provisions, Art 32. 78 Legislative Guide, Recommendation 54, sub-para (i), and ch IV, para 37; Registry Guide, Recommendation 36 and paras 274–280.

The UNCITRAL Model Law on Secured Transactions

69

and thus should be treated as a public service and not a benefit to the parties to secured transactions.79 The Draft Guide to Enactment discusses several variants of these two options, focusing on the Registry charging a fee for some services but not for others.80

VII. Priority of a Security Interest A. General Rules To achieve another key objective of the Legislative Guide,81 the Model Law includes a comprehensive set of rules dealing with priority conflicts between a secured creditor and every possible competing claimant (another creditor with a right in the encumbered asset).82 In a conflict between two security interests in the same asset, the first that was made effective against third parties (by registration or otherwise) has priority.83 Other than changes of a drafting nature, a few new provisions have been included in the Model Law. One of them is a proposed rule dealing with priority conflicts between security interests granted by different persons (ie the initial grantor and a transferee of an encumbered asset). In the case of such a priority conflict, priority is determined according to the time when third-party effectiveness was achieved, provided that the initial secured creditor registers an amendment notice adding the name of the transferee within a short period of time after the transfer or after the secured creditor acquires knowledge of the transfer.84 A transferee of an encumbered asset takes the asset subject to the security interest, with the exception of situations that have to do with the good faith acquisition of an encumbered asset or its acquisition in the transferor’s normal course of business.85 Subject to insolvency rules, in principle, a security interest preserves its third-party effectiveness and priority.86 In line with the approach taken in the Legislative Guide,87 the Model Law takes no position as to whether there should be any statutory preferential claims. It simply includes an article for enacting States to list the type and maximum amount of preferential claims.88 The right of an unsecured creditor that has

79

Registry Provisions, Art 34. See Draft Guide to Enactment (A/CN.9/WG.VI/WP.71) Add 3, paras 74–81. 81 Legislative Guide, Recommendation 1, sub-para (g). 82 Model Law, Art 2, sub-para (e). 83 Model Law, Art 29. 84 Model Law, Art 30, and Registry Provisions Art 26, option A or B. 85 Model Law, Art 34. 86 Model Law, Art 35. 87 Legislative Guide, ch V, paras 90–93. 88 Model Law, Art 36. 80

70

Spyridon V Bazinas

obtained a judgment and taken the steps necessary to have it enforced before a security interest became effective against third parties has priority over that security interest.89 Generally, an acquisition secured creditor that has registered a notice within a short period of time after delivery of the goods has priority as of the time of delivery of the goods, not the time of registration.90 Options are offered to enacting States based on whether a distinction is drawn among equipment, inventory and consumer goods. Priority among competing acquisition security interests is determined on the basis of the general rules.91 However, an acquisition security interest of seller, lessor or licensor has priority over a competing acquisition security interest of a creditor other than a seller, lessor or licensor. Finally, options are offered to enacting States with respect to whether the priority of an acquisition security interest (in goods or intellectual property) extends to its proceeds.92

B. Asset-specific Rules The Model Law includes detailed priority rules with respect to security interests in certain types of asset. First, to avoid interfering with the negotiability of negotiable instruments, the Model Law provides that a security interest in a negotiable instrument made effective against third parties by possession of the instrument has priority over a security interest in the same instrument made effective against third parties by registration.93 Second, with respect to rights to payment of funds credited to a bank account, the Model Law sets out the priority of a security interest depending on the method by which it was made effective against third parties. The order is as follows: the transferee of the account, the depositary bank, a secured creditor with a control agreement, in the case of several control agreements, the secured creditor with the earlier control agreement, the secured creditor that registered a notice.94 The depositary bank’s rights of set-off have priority over any security interest, except one made effective against third parties by the secured creditor becoming the account holder. Under this rule, in the case of non-notification factoring or undisclosed invoice discounting, the factor or invoice discounter need not register a notice in the security interest registry. To have priority over all competing claimants, the factor will need to become the holder of its client’s account to which the relevant receivables are being paid. To have priority against all except the depositary bank and a

89

Model Law, Art 37. Model Law, Art 38. 91 Model Law, Art 39, para 1. 92 Model Law, Art 41. 93 Model Law, Art 46, para 1. 94 Model Law, Art 47. 90

The UNCITRAL Model Law on Secured Transactions

71

secured creditor that has become the account holder, the factor will need to take a control agreement. To have priority over the depositary bank (including for its set-off rights), the factor will need to obtain to obtain a subordination agreement.95 Fourth, in the case of a transfer of funds from an encumbered bank that is initiated or authorised by the grantor, the transferee acquires its right free of the security interest, unless it had knowledge that the transfer violates the rights of the secured creditor under the security agreement.96 Fifth, to avoid undermining the negotiability of negotiable documents, a security interest in tangible assets covered by a negotiable document made effective against third parties by possession of the document has priority over a competing security interest made effective against third parties by registration or possession of the assets covered by the document.97 Finally, another new provision is a rule dealing with priority conflicts between security interests in non-intermediated securities. According to that rule, a security interest in certificated securities made effective against third parties by possession of the certificate has priority over a security interest made effective against third parties by registration of a notice in the security interest registry. With respect to uncertificated securities, there is a cascade of rules according to which notation in the books of the issuer beats any other method and control agreement beats any other method except notation in the books of the issuer. The rule also deals with the question whether a transferee of encumbered non-intermediated securities acquires its rights free or subject to the security interest. Two options are offered. The first provides that defers to the law governing transfers of securities. The second provides that resolves the conflict in favour of the transferee of the securities.98

VIII. Rights and Obligations of the Parties and Third-party Obligors In line with the Legislative Guide, the Model Law contains a chapter dealing with the rights of the parties to the security agreement (ie the grantor and the secured creditor) and of third-party obligors (ie the debtor of an encumbered receivable, the obligor under a negotiable instrument, the depositary bank and the issuer of a negotiable document or non-intermediated security). The purpose of the provisions on the rights of the parties is twofold; first, to set out certain default rules (eg the obligation in possession of an encumbered

95

Model Law, Art 43. Model Law, Art 48. 97 Model Law, Art 49. 98 Model Law, Art 51. 96

72

Spyridon V Bazinas

asset to exercise reasonable care)99 and mandatory rules (eg the obligation of the secured creditor to return an encumbered asset);100 and second to set out the rights of the parties (eg to notify the debtor of the receivable).101 The purpose of the provisions on the rights and obligations of third-party obligors is to protect third-party obligors (eg set out the defences and rights of set off of the debtor of the receivable).102 The Model Law generally follows the approach of the Legislative Guide in this regard as well. The main difference relates to the rights and obligations of the issuer of non-intermediated securities, which are referred to the relevant securities law.103

IX. Enforcement of a Security Interest A. General Rules Following the Legislative Guide, the Model Law recognises that an enforcement regime that results in excessive delays or costs is likely to affect the availability and the cost of credit.104 At the same time, the Model Law recognises that the judicial enforcement regime does not lend itself to harmonisation at the international level. Thus, the Model Law leaves judicial enforcement to other law, making only a recommendation for the introduction of expedited proceedings. At the same time, it introduces extra-judicial enforcement, introducing safeguards for the grantor and other parties with interests in an encumbered asset. The first such safeguard is that the enforcing secured creditor must act in good faith and in a commercially reasonable manner (eg, avoid selling the encumbered assets in a remote location or outside the relevant market).105 To protect the grantor from undue pressure on the part of the secured creditor, the Model Law provides that the grantor may not waive the general standard of conduct.106 Other rights existing under the enforcement provisions of the Model Law may be waived, but only after default, not at the time of the negotiation of the security agreement, to avoid putting the grantor in the position of having to give up its rights to obtain a concession from the secured creditor.107 The second safeguard is that the enforcing secured creditor cannot obtain possession of the encumbered assets out of court, unless: (a) the grantor has

99

Model Law, Art 53. Model Law, Art 54. Model Law, Art 58. 102 Model Law, Art 64. 103 Model Law, Art 71. 104 Legislative Guide, ch VIII, para 6. 105 Model Law, Art 4. 106 Model Law, Art 3, para 1. 107 Model Law, Art 72, para 3. 100 101

The UNCITRAL Model Law on Secured Transactions

73

c onsented in the security agreement to the secured creditor obtaining possession of the encumbered assets out of court; (b) the secured creditor gives the grantor and any person in possession of the encumbered assets notice of default and extra-judicial repossession; and (c) at the time of repossession, the grantor and any person in possession of the encumbered asset does not object.108 The third safeguard is an elaborate notice system according to which notice of the secured creditor’s intention to dispose of the encumbered assets out of court must be given well in advance to all affected parties.109 Similar safeguards for the grantor and other parties with interests in the encumbered assets are foreseen in the case of a proposal by the secured creditor to acquire the encumbered assets in satisfaction of the secured obligation.110 The fourth set of safeguards relates to the distribution of proceeds from the extra-judicial disposition of encumbered assets out of court. The Model Law provides that the enforcing secured creditor must apply the net proceeds (after deducting the reasonable costs of enforcement) to the secured obligation; if there is a shortfall, the grantor remains liable, but the secured creditor then has the position of an unsecured creditor; and any surplus remaining must be turned over to the grantor or to other creditors announced during the enforcement proceedings, or, in the case of doubt, be deposited with a competent judicial or other authority.111 To ensure finality of the rights acquired pursuant to an out-of-court disposition, the Model Law recommends that the transferee acquire the encumbered assets free of any security interests that are subordinate to the security interest of the enforcing secured creditor, but subject to any security interests with priority over the security interest of the enforcing secured creditor;112 and a lessee or licensee is entitled to the benefit of the lease or licence during its term except as against creditors with rights that have priority over the right of the enforcing secured creditor.113

B. Asset-specific Rules The Model Law contains only two asset-specific rules. The first deals with the right of the secured creditor to collect payment under a receivable, negotiable instrument, right to payment of funds credited to a bank account or non-intermediated security.114 The addition of collection of payment under a non-intermediated

108

Model Law, Art 77, para 2. Model Law, Art 78, paras 4–8. 110 Model Law, Art 80, paras 2 and 3. 111 Model Law, Art 79. 112 Model Law, Art 81, para 3. 113 Model Law, Art 81, para 4. 114 Model Law, Art 82, para 1. 109

74

Spyridon V Bazinas

security constitutes the main difference between the L egislative Guide and the Model Law enforcement provisions. Collection may take place not only after default, but also before default provided that the grantor has consented.115 A secured creditor that exercises the right to collect an encumbered receivable or other type of asset has also the right to enforce any right that secures or supports payment of that receivable or other asset.116 Collection of funds from a bank account may take place out of court only if the security interest has been made effective against third parties by the secured creditor becoming the account holder or by a control agreement, that is, with the consent of the depositary bank.117 Finally, the right of collection has to be exercised in line with the provisions of the Model Law dealing with the rights of third-party obligors (ie Articles 61–71).118 The other assets-specific rule refers to collection in the case of an outright assignment of a receivable, according to which the transferee may collect at any time after the receivable becomes due. The transferee may also enforce against rights securing or supporting payment of the transferred receivable. And the right of the transferee to collect is subject to the provisions of the Model Law on the rights of third-party obligors (ie Articles 61–71).119

X. Law Applicable to Security Interests A. General Rules The Model Law contains a complete set of conflict-of-laws rules, which are intended to enhance certainty as to the law applicable to security interests in movable property and facilitate the movement and the financing of movable property across national borders. These rules are intended to apply generally to all transactions, whether a particular transaction is linked with more than one State. Of course, if a transaction has connections only with one State, the law of that State will apply. But that determination is based on a conflict-of-laws analysis. The law applicable to the creation, third-party effectiveness and priority of a security interest in a tangible asset is the law of the State in which the asset is located (except if the asset is covered by a negotiable document, where the applicable law is the law of the location of the document).120 There are a few exceptions to this lex

115

Model Law, Art 82, para 2. Model Law, Art 82, para 3. 117 Model Law, Art 82, para 4. 118 Model Law, Art 82, para 5. 119 Model Law, Art 83. 120 Model Law, Art 85, paras 1 and 2. 116

The UNCITRAL Model Law on Secured Transactions

75

situs rule. The law applicable to a security interest in mobile goods that typically cross national borders is the law of the State in which the grantor is located.121 Finally, the law applicable to a security interest in goods in transit or export goods is the law of the State of destination, provided that the goods reach their destination within a short period of time.122 Following generally applicable rules, enforcement of a security interest in a tangible asset is referred to the law of the State in which the asset is located at the time of commencement of enforcement.123 The law applicable to security interests in intangible assets (with the e xception of receivables related to immovable property, bank accounts, proceeds and intellectual property that are subject to special rules), the Model Law follows the approach of the Receivables Convention, providing for the application of the law of the grantor’s location.124 Location of the grantor is defined by reference to its place of business and, in the case of places of business in more than one State, by reference to the place where the grantor has its central administration.125 The rationale of this rule is that it is the rule that: (a) is most likely to lead to the application of the law of one State; (b) can reasonably be determined in most cases; and (c) is the law of the State in which the main insolvency proceeding with respect to the grantor will most likely be commenced. The relevant time for determining location of the asset or the grantor is, with respect to creation issues, the location at the time of the putative creation of a security interest, and, with respect to thirdparty effectiveness and priority issues, the location at the time the issue arises.

B. Asset-specific Rules Following the approach of the Legislative Guide, the Model Law contains a number of asset-specific conflict-of-laws rules, including rules dealing with the law applicable to security interests in rights to payment of funds credited to a bank account, intellectual property and non-intermediated securities. With respect to the law applicable to a security interest in a right to payment of funds credited to a bank account, the Model Law provides two options for States to choose from. Under the first option, the law applicable is the law of the State in which the bank with which the account is maintained has its place of b usiness. Under the second option, the law applicable is the law stated in the account agreement.126 With respect to the law applicable to a security interest in intellectual property, the Model Law follows the approach of the IP Supplement, which contains a hybrid rule.127 Under this rule, the law applicable to the creation, effectiveness 121

Model Law, Art 85, para 3. Model Law, Art 85, para 4. 123 Model Law, Art 88, sub-para (a). 124 Model Law, Art 86. 125 Model Law, Art 90. 126 Model Law, Art 97. 127 IP Supplement, Recommendation 248. 122

76

Spyridon V Bazinas

against third parties and priority of a security interest in intellectual property is the law of the State in which the intellectual property is protected; however, a security interest created and made effective against third parties other than a secured creditor, transferee or licensee (eg an insolvency administrator) under the law of the State of the grantor’s location is also effective; and the law applicable to the enforcement of such a security interest is the law of the State of the grantor’s location.128 The main difference between the Model Law and the Legislative Guide in this regard is the new rule on the law applicable to security interests in non- intermediated securities. Under this rule, the law applicable to security interests in equity securities (ie shares) is the law of the State under which the issuer is constituted; and the law applicable to security interests in debt securities (ie bonds) is the law governing the securities. The main advantages of this rule is that it refers all issues (the creation, third-party effectiveness, priority enforcement of a security interest, as well as the effectiveness of a security interest as against the issuer) to one and the same law, and distinguishes between debt and equity securities.

XI. Conclusions The Legislative Guide, including the IP Supplement, and the Registry Guide, have become the main reference tools for the modernisation and harmonisation of secured transactions law. However, the two guides are long and complex texts and thus not always easy to implement. The Model Law is intended to assist States in this regard. This chapter briefly introduced the Model Law and its main differences from the Legislative Guide and the Registry Guide. These differences may be summarised as follows: (a) several recommendations have been revised to be reflected in legislative language; (b) a number of recommendations that do not belong in a legislative text have been left out; (c) a number of provisions have been added to the Model Law to address security interests in non-intermediated securities; and (d) the recommendations of the Legislative Guide and the Registry Guide have been reflected in a set of Registry Provisions that may be implemented in the secured transactions law or a separate act, or partly in the former and partly in the latter.

128

Model Law, Art 99.

5 Non-assignment Clauses and their Treatment under UNCITRAL’s Secured Transactions Law Instruments N ORKUN AKSELI*

I. Introduction Receivables financing is an important method of raising finance. Raising finance through assignment of receivables ‘is simply bigger business than the financing of mobile goods.’1 Receivables financing has seen considerable growth as ‘receivables are self-liquidating and … an excellent short-term source of cash.’2 Particularly, small businesses routinely use their receivables to finance their operations. Unlike large businesses, which can utilise their wide ranging assets by diversifying their sources of finance, receivables for small businesses are significant assets, which they can use as collateral by either selling them outright (as in factoring and invoice discounting) or by assigning them by way security to a financier. However, contracts that contain restrictions on the assignment of receivables present tension between concepts of freedom of contract and the need for free alienability of property. These two opposite positions involve a number of theoretical questions. One of these questions relate to the doctrine of privity according to which assignment of rights, where the creditor/assignor unilaterally assigns the rights to a third party without the consent of the other contracting party (customer/debtor), can be regarded as a method to avoid the privity doctrine. In a sense, prohibition is

* Associate

Professor of Commercial Law, Durham University, Durham, United Kingdom. NB Cohen, ‘Harmonizing the Law Governing Secured Credit: The Next Frontier’ (1998) 33 Texas International Law Journal 173, 185. 2 S Schwarcz, ‘Towards a Centralized Perfection System for Cross-Border Receivables Financing’ (1999) 20 University of Pennsylvania Journal of International Economic Law 455, 456. For the significance of receivables financing see also F Oditah, Legal Aspects of Receivables Financing (London, Sweet and Maxwell, 1991) 2. 1

78

N Orkun Akseli

said to be a strategy to preserve privity of contract. Another question that antiassignment clauses and their regulation pose is whether contractual rights can be classified within the framework of assignment as property rights.3 However, one thing seems to be certain. Assignability of contracts is regarded as the origin of modern capitalism because it enables small businesses to borrow on a low rate of interest and provide a rapid turnover of capital.4 Small businesses will be able to utilise their property and capital (ie receivables), which may otherwise be prevented.5 Contractual restrictions on the assignment of receivables prevent small businesses accessing certain financing frameworks, including factoring, invoice discounting, supply chain financing and discounting of individual finances over an online platform. This is a particularly pressing matter in the relative absence of bank lending in the post-global financial crisis era.6 To the extent that the assignment relates to receivables or book debts, anti-assignment clauses under English law will begin to lose their significance due to recent legislative intervention. In the UK, with the enactment of Small Businesses, Employment and Enterprise Act 2015, section 1 and related regulations arising from this legislation, bans on assignment will be nullified.7 UNCITRAL’s secured transactions law instruments recognise the effectiveness of assignments made notwithstanding an anti-assignment clause. These include the UN Convention on the Assignment of Receivables in International Trade, UNCITRAL Legislative Guide on Secured Transactions (the Legislative Guide) and the UNCITRAL Model Law on Secured Transactions (the UNCITRAL Model Law). The approach adopted on anti-assignment clauses by these international instruments shows similarity to that in the Canadian Personal Property Security Acts (PPSAs)8 and to a certain extent in the Uniform Commercial Code (UCC),

3 On these points see generally M Bridge, ‘The Nature of Assignment and Non-assignment Clauses’ (2016) LQR 47; GJ Tolhurst and JW Carter, ‘Prohibitions on Assignment: A choice to be made’ (2014) Cambridge Law Journal 405. 4 JR Commons, Legal Foundations of Capitalism (New Brunswick,Transaction Publishers, 1995) 253. 5 M Safavian, H Fleisig and J Steinbuks, ‘Unlocking Dead Capital’ Viewpoint, Note Number 307 (March 2006). 6 S Fraser, ‘The Impact of Financial Crisis on Bank lending to SMEs’ Econometric Analysis from the UK Survey of SME Finances’, (July 2012), available at www.gov.uk/government/uploads/system/ uploads/attachment_data/file/34739/12-949-impact-financial-crisis-on-bank-lending-to-smes.pdf (accessed October 2016). 7 See generally H Beale, L Gullifer and S Paterson, ‘A Case for Interfering with Freedom of Contract? An Empirically Informed Study of Bans on Assignment’ (2016) Journal of Business Law 203; L Gullifer, ‘Should Clauses Prohibiting Assignment be Overridden by Statute?’ (2015) 4 Penn State Journal of Law and International Affairs 47. At the time of writing, the draft Business Contract Terms (Restrictions on Assignment of Receivables) Regulations 2015 that will refine s 1 of the SBEEA had not been finalised, and were still in draft form. 8 Eg Ontario PPSA, s 39 reads as follows: ‘The rights of a debtor in collateral may be transferred voluntarily or involuntarily despite a provision in the security agreement prohibiting transfer or declaring a transfer to be a default, but no transfer prejudices the rights of the secured party under the security agreement or otherwise.’ British Columbia PPSA, s 41(9) reads as follows: ‘A term in a contract between an account debtor and an assignor that prohibits or restricts assignment of the whole of the account or chattel paper for money due or to become due is binding on the assignor, but only to the extent of

Non-assignment Clauses

79

Article 9.9 The basis of these provisions holds the assignor liable in damages for breach of the underlying contract and not the assignee or the secured party. However, the debtor of the underlying contract cannot terminate the contract as a result of that breach. It can be argued that norms that nullify bans on assignment have become international customary law. These can be found under both civil and common law jurisdictions.10 This chapter will examine UNCITRAL’s approach to anti-assignment clauses in its secured transactions law instruments. The chapter will first explain what anti-assignment clauses are and why they are used commercial transactions. The chapter will then explore why anti-assignment clauses need to be regulated. Finally, the chapter will analyse the approach of UNCITRAL’s secured transactions law instruments to anti-assignment clauses.

II. What are Anti-assignment Clauses and Why Are They Used in Contracts? Anti-assignment clauses are contractual clauses that prohibit one party (generally a service provider, supplier or seller in a contract) from assigning or transferring the benefit arising from this contract to a third party. These types of clauses are generally found in standard terms and conditions11 where the negotiation of the contract is not often possible. Debtors or buyers with strong bargaining powers, generally insist on inclusion of an anti-assignment clause in order to protect their interests. These clauses are generally problematic for assignees (financiers/lenders), particularly in the assignment of bulk receivables. This is because the existence of an anti-assignment clause requires assignees to check each document, in order to ascertain whether the assignment is banned between the assignor and the debtor. The administrative cost involved in due diligence increases the cost of credit.

making the assignor liable in damages for breach of contract, and is unenforceable against third parties’ Saskatchewan PPSA, s 41(9) also provides a similar regulation of anti-assignment clauses whereby the anti-assignment clause cannot be enforceable against third parties and breach of anti-assignment clause makes the assignor liable against the debtor thus making it binding on the assignor. 9 UCC, §§ 9-401, 9-406(d) Cmt.5. Particularly, UCC, § 9-402, which protects the secured party/ assignee from any liability based on contract or tort arising from debtor’s acts or omissions. 10 In Latin, free assignability is called ‘Pactum de non cedendo’. Free assignability is generally recognised by some of the Roman law influenced civil law systems, for instance, under Swiss Code of Obligations (Art 164) and Turkish Code of Obligations (Art 162(1)), and under some Common Law systems such as by the UCC Art 9 regime, see eg UCC, § 9-406(d) and UCC, § 9-408(a); s 354(a) of the German Commercial Code (Handelsgesetzbuch, HGB). 11 ‘Large organizations such as universities and local authorities commonly incorporate [anti]assignment clauses in their standard terms of trading’: M Bridge, Personal Property Law (3rd edn, Clarendon Law Series, 2002) 159.

80

N Orkun Akseli

Rendering the restriction on assignments ineffective may have the potential to increase small businesses’ access to finance by allowing them to use different financing techniques. A prohibition on the assignment of right to payment may be considered as ‘contrary to public policy as an unacceptable restraint on alienation.’12 Free assignability is necessary to protect the bona fide purchaser.13 In principle, all receivables are assignable unless the assignment is prohibited by agreement, statute or due to public policy considerations.14 Free assignability is vital for ‘both businesses and consumer affairs, since credit is commonly obtained on the security of insurance policies, hire-purchase contracts and traders’ and builders’ book debts.’15 The debtor has a legitimate interest in insisting on an anti-assignment clause. The debtor prefers predictability and protection against a change in the creditor. The debtor also prefers, for the sake of legal certainty, that there will be good discharge upon making payment to the original creditor, without having to deal with someone else with whom she has no previous business dealing.16 However, it can be argued that to a certain extent, debtors may be protected by debtor protection rules.17 Anti-assignment clauses may take a number of forms.18 These include a simple condition that restricts the assignment of the benefit of the contract; conditional restriction on the assignment of the contract where the condition requires approval of the debtor or satisfaction of certain requirements; the restriction may

12 L Gullifer (ed), Goode on Legal Problems of Credit and Security (4th edn, London, Sweet and Maxwell, 2009) 107–108; Tom Shaw & Co v Moss Empires Ltd (1908) 25 TLR 190. Examples of public policy may include, among others, maintenance and champerty and public servants’ remuneration and tort claims. 13 G Gilmore, ‘The Commercial Doctrine of Good Faith Purchase’ (1954) 63 Yale Law Journal 1057, 1121. 14 A/CN.9/397, para 20. Mulkerrins v PricewaterhouseCoopers (a firm) [2003] UKHL 41, [2003] 1 WLR 1937, para 13 (benefit of a contract may be assigned to a third party without the consent of the debtor and unless there is contractual prohibitions, however, an assignment under the Law of Property Act 1925, s. 136 cannot be objected as there is no debtor’s consent requirement). In the UK, for example, social security benefit and child benefits are not assignable: see Social Security Administration Act 1992, s 187; Pension Schemes Act 1993, s 159; Pensions Act 1995, s 91. See Re Mirams [1891] 1 QB 594, where an assignment of the salary of a public officer has been held to be void; Methwold v Walbank (1750) 2 Ves Sen 238 and Liverpool Corpn v Wright (1859) 28 LJ Ch 868 (where a public officer cannot assign his salary); see also Arbuthnot v Norton (1846) 5 Moo PCC 219 (where it was held that a judge could not assign his salary); for a similar decision in the US see eg Byers v Comer 68 P.2d 671 (Ariz), modified, 70 P.2d 330 (Ariz 1937); S Worthington, ‘The Disappearing Divide between Property and Obligation: The Impact of Aligning Legal Analysis and Commercial Expectation’ (2007) 42 Texas International Law Journal 917, 927–928; A Bell, Modern Law or Personal Property in England and Ireland, (Butterworths, 1989) 380; In Re Freeman 232 B.R. 497, 501 (Bankr. M.D. Fla. 1999); Restatement (Second) of Contracts § 317(2)(b). 15 B Allcock, ‘Restrictions on the Assignment of Contractual Rights’ (1983) 42 Cambridge Law Journal 328. 16 F Mestre, ‘Explanatory Report on the Draft Convention on International Factoring’ (1987) Uniform Law Review 85, 117. 17 Eg the Receivables Convention’s debtor protection principle and protection of debtor against assignor under Art 10(3). 18 See generally Tolhurst and Carter (n 3) 405–406.

Non-assignment Clauses

81

be a promise not to assign; or the contract may be personal to the parties,19 hence the impossibility of its transfer. There are certain reasons why anti-assignment clauses are used in business transactions. The main reason is the protection of the debtor’s interests. Debtors may often, despite receiving notice of assignment, overlook the notice and pay by mistake to the assignor and therefore do not get good discharge. Losing the right to get a good discharge may mean that the debtor may be compelled to pay a second time.20 Debtors may want to protect their defences and rights of set-off arising from their dealings with the assignor. While the debtor’s defences against the assignor that have arisen before the receipt of the notice of assignment may be set up against the assignee, the same arguments cannot be made in relation to the rights of set-off. With the receipt of notice of an assignment, the debtor loses the availability of rights of set-off against the assignee in relation to present and future claims.21 The reason is that the assignee has no present or potentially, a future business relationship.22 However, from a debtor protection perspective, the debtor must not be worse off as a result of the assignment; this reflects the concept that the assignee takes assignment ‘subject to equities’.23 In Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd,24 Lord BrowneWilkinson noted that the significant concern of a debtor, in its contractual relationships and in a possible assignment, was the impossibility of relying upon new equities against the assignor after the receipt of notification of assignment.25 A further reason is that the debtor, for commercial reasons and in order to protect its interests, may not wish to deal with a new creditor that it has not dealt with before.26 Similar points were made in Linden Gardens: Lord Browne-Wilkinson explained the reasons that led to the use of anti-assignment clauses as follows: The reason for including the contractual prohibition viewed from the contractor’s point of view must be that the contractor wishes to ensure that he deals, and deals only, with the particular employer with whom he has chosen to deal. Building contracts are 19

Nokes v Doncaster Amalgamated Collieries Ltd [1940] AC 1014 (HL). Brice v Bannister (1878) 3 QBD 569, CA. 21 Business Computers Ltd v Anglo-African Leasing Ltd [1977] 1 WLR 578 (Ch D) (where, owing to assignment, claims which have accrued due prior to the receipt of the debtor of a notice of the assignment, debtor’s right of set-off have been limited). 22 Christie v Taunton, Delmard Lane & Co [1893] 2 Ch 175 (Ch D); G McCormack, Secured Credit Under English and American Law (Cambridge, Cambridge University Press, 2004) 229. 23 However, it can be argued that as the assignee takes subject to equities the debtor should not have any concerns. For this view see Mulkerrins v PriceWaterhouseCoopers [2003] UKHL 41, [2003] 1 WLR 1937 at [15], where Lord Millett indicates that ‘the reason that the debtor’s consent is not required to an assignment of a debt is that the assignment cannot prejudice him. The assignment is subject to equities, which means that any set-off which the debtor may have against the assignor can be asserted against the assignee’. See also G Gilmore, ‘The Assignee of Contract Rights and His Precarious Security’ (1964) 74 Yale Law Journal 217, 227 where he discusses the position of the assignee and the defences of the debtor. 24 Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85, [1993] 3 All ER 417. 25 [1993] 3 All ER 417 at 429, [1994] 1 AC 85 at 105. 26 Don King Productions Inc v Warren [2000] Ch 291, 319 where Lightman J, sitting in the Court of Appeal, states that ‘the purpose of [a non]-assignment clause is the genuine commercial interest of a party of ensuring that contractual relations are only with the person he has selected as the other party to the contract and no one else’. 20

82

N Orkun Akseli

regnant with disputes: some employers are more reasonable than others in dealing with p such disputes.27

In Linden Gardens Lord Browne-Wilkinson clarified that a contractual prohibition could prevent an assignment as between the assignee and the customer/debtor; but the assignment is still effective as between assignor and assignee.28 This reasoning considers the fact that ‘the personal quality of the creditor may be of importance to the debtor, who wishes to avoid the risk of having to deal with an assignee that is more demanding and less flexible.’29 By the same token, transaction costs may increase with the change of creditor. This is ‘due to the effect of certain provisions in loan agreements such as increased costs and tax clauses’.30 An anti-assignment clause may also increase the cost of credit, as an assignee would prefer transactional certainty, and due diligence actions may contribute to increased transaction costs. From the assignee’s standpoint the debtor should make the payment to the assignee. If there is a risk of non-payment, this may be reflected in the risk premium.

III. The Legal Effect of Anti-assignment Clauses The use of anti-assignment clauses generally reflects the tension between free alienability and property views. While the free alienability versus freedom of contract position seems to be the legal lens through which US legal scholarship analyses anti-assignment clauses,31 English and Australian scholarship generally looks at the issue from a property versus freedom of contract point of view.32 In the former set of scholarship, free alienability denominates the ability of the assignor to assign right to payment (receivables) to an assignee, notwithstanding any

27 [1994] 1 AC 85 at 105, [1993] 3 All ER 417 at 429; see also Yeandle v Wynn Realisations Ltd (1995) 47 Con LR 1 (CA (Civ Div)) (where the Court of Appeal held that ‘the permission given to the subcontractor without consent to assign was the power to assign any sum which were or might become due and payable to them under the sub-contract [and] an assignee of a debt was entitled to enforce it but not to enforce the other provisions of the contract’). 28 [1994] 1 AC 85, 108; see also LS Sealy and RJA Hooley, Commercial Law Text, Cases and Materials (3rd edn, Butterworths, London, 2003) 943ff. The decision in Helstan Securities Ltd v Hertfordshire County Council [1978] 3 All ER 262 (QBD) was distinguished in Linden Gardens; the anti-assignment clause without a wider prohibition cannot invalidate the assignment between the assignor and the assignee. 29 See R Goode, ‘Assignment Clauses in International Contracts’ (2002) 3–4 Revue de Droit des Affaires Internationales/International Business Law Journal 389, 395. 30 A McKnight, ‘Contractual Prohibitions on a Creditor’s Right to Alienate Debts’ (2003) 18 Journal of International Banking Law 1, 3. 31 See eg N Cohen and WH Henning, ‘Freedom of Contract vs. Free Alienability: An Old Struggle Emerges in a New Context’ (2010/11) 46 Gonzaga Law Review 353. 32 See eg Tolhurst and Carter (n 3); Bridge (n 3).

Non-assignment Clauses

83

contractual restrictions, and therefore to utilise the economic value of the asset; while the freedom of contract denominates the restrictions agreed between the debtor (buyer) and the assignor (seller) that protect the debtor from dealing with a third party, possibly that she would not want to deal with in the first place.33 In the latter scholarship, the contractual freedom view generally suggests that the restriction of assignment should not have an impact on the assignor’s right to assign to a third party, but could only affect the debtor’s obligation. This view suggests that the assignor should be free to assign and—as a result of this assignment—the debtor’s obligation to pay in order to obtain a good discharge changes, and the debtor must make its payment to the new creditor (ie the assignee). The assignee, despite having no right to sue the debtor without the assistance of the assignor, unless there is statutory assignment, may avoid the prohibitions through a declaration of trust according to which the assignee has a direct right to enforce the contract as against the debtor.34 The property view (or the benefit of the contract view) generally suggests that the effect of assignment has proprietary characteristics, and in order to be transferred the right needs to be characterised as property. Debts have been recognised as property by the courts of equity.35 Thus, the restriction on assignment has justifiable grounds. These clauses guarantee that the assignor under the contract performs its contract with the debtor and the contract cannot be altered by way of an assignment. The debtor, thus, can protect its property rights and claim that the assignment in breach of an anti-assignment clause is ineffective.36 There is, however, stronger support to recognise the effectiveness of assignments made, notwithstanding anti-assignment clauses in monetary rights.37 The Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd decision paves the way for the view that the legal effect of assignment depends on how a prohibition clause is understood (construction argument).38 In this decision, the House of Lords recognised the effectiveness of a prohibition on assignment, but also

33

See below for more discussion. For this view see eg R Goode, ‘Contractual Prohibitions against Assignment’ (2009) LMCLQ 300. 35 Ellis v Torrington [1920] 1 KB 399, 410–11 ‘The common law treated debts as personal obligations and assignments of debts merely as assignments of the right to bring an action at law against the debtor and, except in a strictly limited number of cases, did not recognize any such, assignments. Courts of equity always took a different view. They treated debts as property, and the necessity of an action at law to reduce the property into possession they regarded merely as an incident which followed on the assignment of the property.’ 36 Barbados Trust Co v Bank of Zambia [2007] EWCA Civ 148, [2007] 1 Lloyd’s Rep 495 at [88] ‘The ineffectiveness of the assignment in breach of a prohibition on assignment is understandable. It is not merely a matter of contract but of property’, per Rix LJ. 37 See the below UN instruments, as well as UCC Art 9 and similar legislation. The difference between non-monetary and monetary rights has been clearly articulated in the Unidroit Principles for International Commercial Contracts (2004). The Unidroit Principles recognise the effectiveness of an assignment between the assignor and the assignee where the right is a monetary right despite an antiassignment clause in the contract between the assignor and the debtor (Art 9.1.9 para 1). On the other hand, Art 9.1.9 para 2 deals with non-monetary rights. Anti-assignment clauses will be effective against the assignee where the assigned right is a non-monetary right. 38 [1994] 1 AC 85, 105. 34

84

N Orkun Akseli

recognised that the assignment was effective between the assignor and the assignee, and held that: … a prohibition on assignment normally only invalidates the assignment as against the other party to the contract so as to prevent the transfer of the chose in action: in the absence of the clearest words it cannot operate to invalidate the contract as between the assignor and the assignee and even then it may be ineffective on the grounds of public policy … [T]he existing authorities establish that an attempted assignment of contractual rights in breach of a contractual prohibition is ineffective to transfer such contractual rights … If the law were otherwise, it would defeat the legitimate commercial reason for inserting the contractual prohibition, viz to ensure that the original parties to the contract are not brought into direct contractual relations with third parties.39

Lord Browne-Wilkinson’s statement suggests that the legal effect of the prohibition depends upon the construction of the clause. This is because the clause may not automatically prohibit the assignment but rather may subject it to the debtor’s consent.40 The right that arises out of the underlying contract between the assignor and the debtor may be a personal right where the performance is a personal one and cannot be performed by anybody but the assignor. Restriction of assignment in this instance is justifiable. That personal right, not being a property right, cannot be assigned. To the extent that an intangible is the subject of the transfer, it can be claimed that the construction argument does not consider commercial realities, particularly for small businesses. This is because the main intention of the assignor is to access finance by transferring the right to payment, and this leaves no room for construction. It may not always be possible to obtain the debtor’s consent. Furthermore, the debtor may have stronger bargaining power, which results in an inequality of bargaining power between the debtor and the assignor. The debtor may withhold consent or even may not have the authority to consent to the assignment. The right may be a personal property right in which case it can be, without a doubt, subject of a transfer.41 While under English law, land and tangible property may be subject to restrictions of assignment, the possibility of restricting the assignability of receivables on the basis of anti-assignment clauses does not meet commercial expectations. It is argued that restricting the ability of the assignor to assign on the basis of stipulations of the party with the stronger bargaining power is simply against contractual freedom. There is no support for construction argument under the UNCITRAL instruments.42 The statement further suggests that the assignment is effective as between the assignor and the assignee, thus the prohibition cannot invalidate the contract (that is also the transfer) between the assignee and the assignor. This would be against public policy by restricting contractual freedom. The UNCITRAL instruments recognise

39

[1994] 1 AC 85, 108. Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85 (HL). 41 Tolhurst and Carter (n 3) 410. 42 See below. 40

Non-assignment Clauses

85

the effectiveness of assignment made notwithstanding a restriction. The statement also suggests that the debtor should not be brought into a relationship with a party (assignee) with whom, in the first place, it did not enter into a contractual relationship. It can be argued that in the assignment of receivables, it is the assignor’s right to payment (receivables) that is assigned, not the performance. Therefore, it is not the personal obligation but the receivable that is transferred. Professor Goode suggests four clear possibilities on anti-assignment clauses.43 Firstly, the anti-assignment clause may be effective as between the assignor and the debtor, but has no effectiveness as against the assignee (the assignment is effective notwithstanding an anti-assignment clause). The breach of restriction only grants the debtor the right to claim damages from the assignor. This approach is similar to the UN Convention, Article 9, the UNCITRAL Model Law, Article 13(1) and (2),44 and the Legislative Guide, Recommendation 24. Second, interpretation provides that the anti-assignment clause may effectively prevent the assignor from assigning the right to payment to third parties and thus the debtor is not affected and can get a good discharge by paying to the assignor. However, under this interpretation the assignor has the right to enter into a contract to assign the fruits of the contract. Under this interpretation the debtor is effectively protected, as any agreement between the assignor and a third party regarding the assignment of the fruits of the contract, after the debtor’s payment, cannot realistically affect the debtor or its interests. Professor McMeel argues that this interpretation appears to be the path that English law is moving towards.45 The third interpretation suggests that the anti-assignment clause effectively prohibits the assignor from assigning both the right to payment and its fruits, even after the payment is done. This interpretation clearly excludes the assignor from access to immediate financing and may have adverse financial affects. Professor Goode states that an anti-assignment clause under this interpretation will be ‘as a matter of law devoid of effect.’46 Although under this interpretation, the aim is to protect the debtor, the assignor’s prohibition of assigning the fruits of the contract should not be limited. The final interpretation suggests that breach of an anti-assignment clause enables the debtor to claim damages and to avoid the contract with the assignor, and thus reduces or extinguishes the assignor’s right to payment.47 That interpretation has been rejected in the UNCITRAL instruments and a breach of an anti-assignment

43

R Goode, ‘Inalienable Rights’ (1979) 42 MLR 553, 554. UNCITRAL Model Law, A/CN.9/884. 45 For this view see G McMeel, ‘The Modern Law of Assignment: Public Policy and Contractual Restrictions on Transferability’ (2004) LMCLQ 483, 500 (although he qualified that the law is not explicitly moving towards the view that the second interpretation takes). See also Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85, 105. 46 Goode (n 43) 554; see also McMeel (n 45) at 500, where he states that the consequences of this could be drastic unless so far as the type (2) consequence was followed. In other words, the anti-fruits clause should not be struck down in its entirety. Furthermore, an anti-assignment clause is inserted for the benefit of the debtor and not anybody else. See also Goode (n 43) 555. 47 Goode (n 43) 554. 44

86

N Orkun Akseli

clause by the assignor is not in itself a sufficient reason for the avoidance of the original contract by the debtor, thus the debtor is precluded from strengthening his bargaining power.48 Until the introduction of UCC Article 9 in the mid-twentieth century, assignability of receivables and the validity of anti-assignment contracts were subject to the freedom of contract doctrine. Under this approach prohibitions over assignments were upheld and the right to payment could be assigned neither as a true sale nor as a security.49 However, views to in favour of free alienability have gained support by courts and scholars and ways have been found to allow free alienability in the face of contractual restrictions.50 UCC Article 9’s approach favours the free assignability of receivables (UCC, section 9-401(2)). Under UCC Article 9 an anti-assignment clause is ineffective to restrict an assignment or to create a security interest. Also, an assignment notwithstanding an anti-assignment clause is not regarded as a breach of contract. UCC, section 9-406 and UCC, section 9-408 override contractual and legal restrictions on the assignor’s right to assignment. UCC, section 9-406(d) promotes free alienability and renders terms restricting assignment ineffective and applies to accounts other than health care insurance receivables and to chattel paper. This section does not apply to the sale of a payment intangible or promissory note.51 On the other hand, UCC, section 9-408 applies in restricted situations where ‘contractual and legal transfer restrictions are overridden to the extent they would otherwise “impair the creation, attachment, or perfection of a security interest,” but the security interest is not enforceable against the person in whose favour the restriction runs’.52

IV. Why is it Necessary to Regulate Prohibitions on the Assignment of Receivables? The above section provides a sketch of the complex theoretical and practical underpinnings of anti-assignment clauses. Following this, it is argued that there 48 For more information on this point see n 82 and the accompanying text thereto. But cf Lord Browne-Wilkinson’s suggestion in Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd that interpretations one and four are ‘very unlikely to occur’: [1994] 1 AC 85, 104. 49 See generally Cohen and Henning (n 31) 354ff. Some of the significant case law examples of this era on the non-assignability of receivables are Allhusen v Caristo Constr Corp 103 N.E.2d 891, 893 (NY 1952) (‘[W]hile the courts have striven to uphold freedom of assignability, they have not failed to recognize the concept of freedom of contract. In large measure they agree that, where appropriate language is used, assignments of money due under contracts may be prohibited’); Parkinson v Caldwell 272 P.2d 934, 937 (Cal Ct App 1954) (‘A contract right has its origin in the agreement of the parties and if the parties by their free agreement place a limitation on the right [of assignment] at the very time of its creation no good reason occurs to us why they may not do so’). 50 G Gilmore, Security Interests in Personal Property (Boston, Little Brown & Co, 1965) § 7.6, at 211. 51 UCC, § 9-406(e). 52 Cohen and Henning (n 31) 363.

Non-assignment Clauses

87

is a policy-based necessity to limit the effectiveness of clauses that restrict the assignment of receivables. While there could be an argument that nullity of antiassignment clauses interferes with market forces and freedom of contract to pursue regulatory goals, anti-assignment clauses are regarded as barriers to finance, particularly for small businesses. Their removal enables those small businesses that lack the necessary collateral to obtain bank lending, to have access to invoice financing or factoring. Traditionally, small businesses rely on bank lending. The global financial crisis also caused a decrease in bank lending to small businesses.53 Raising finance through factoring is important for small businesses. The significant advantage of factoring is that receivables owed to the small business are assigned outright to the financier (factoring company). The factoring company pays a discounted amount in return to the assignor, rather than collateralising these receivables. Collateralisation enables the financier to take the assets as security to satisfy the claims of creditors. If receivables are collateralised, the title on receivables stays with the assignor. In the case of insolvency, receivables will then become part of the estate of the insolvent small business. In this example, the credit risk stays with the small business. This is a significant point in the decision of credit supplied by the factoring company, which is based on the value of the small business’s receivable rather than the creditworthiness of the small business.54 Thus, it is important to have rules that enable and encourage small businesses to utilise factoring more often as a method to raise finance and utilise their dead capital.55 There is a necessity, so far as financing contracts are concerned, to reduce the disadvantages of inequality of bargaining presents. In terms of anti-assignment clauses, it can be argued that there is an inequality of bargaining power. F reedom of bargaining is regarded as ‘the fundamental and indispensable requisite of progress.’56 Small businesses lack both bargaining power and entitlements.57 This can particularly be experienced in commercial relationships, where the small business acts as the supplier or manufacturer for a larger business with stronger and more developed resources. Private law processes do not necessarily assist the weaker party in the bargaining process. Thus, the main question in the regulation of these clauses is what shape this regulation should take. In England, the Small Businesses, Employment and Enterprise Act 2015, s 1 nullifies bans on the 53 ‘Small Businesses seek alternatives as banks leave them in the lurch’ The Economist (16 August 2014). 54 L Klapper, ‘The Role of Factoring for Financing Small and Medium Enterprises’ (2005) World Bank Policy Research Working Paper 3593; see also O Akseli, Vulnerability and Access to Low Cost Credit’ in M Kenny and J Devenney (eds), Consumer Credit, Debt and Investment in Europe (Cambridge, Cambridge University Press, 2012) 4, 15. 55 Safavian, Fleisig and Steinbuks (n 5). 56 J Gordley, The Philosophical Origins of Modern Contract Law (Oxford, Clarendon Press, 1991) 214. 57 For discussions of poverty and lack of entitlements, see I Christoplos and J Farrington, ‘The Issues’ in I Christoplos and J Farrington (eds) Poverty, Vulnerability, and Agricultural Extension (Oxford, Oxford University Press, 2004) 1, 3.

88

N Orkun Akseli

assignment of receivables.58 It was clear that apart from legislative intervention there was no option to remove contractual barriers to assignment (ie anti-assignment clauses). During the pre-enactment process, the impact assessment document of the, then, Department of Business, Innovation and Skills explained the rationale for intervention as follows: 38. We have considered alternative options other than legislating to nullify ban on assignment of trade receivables. We could consider offering guidance to businesses on the merits of nullifying ban on assignment of trade receivable clauses or request that businesses join a voluntary code. These measures would not compel businesses to nullify these clauses however, and could create a disjointed effort to resolve the problem. 39. We have therefore concluded that the only through legislation will we be able to remove this contractual barrier.59

From a regulatory perspective this seems perfectly clear. Economic organisation relies on the market system and the collectivist system. While the market s ystem identifies with private law, the collectivist system identifies with regulation. Private law processes, and therefore the market system, have deficiencies and sometimes cannot assist parties as there are perceived difficulties with the respective bargaining powers of parties.60 Thus, there is a public interest in regulating these deficiencies with legislation or certain rules. Professor Ogus argues that ‘[a]ccording to the public interest theory, regulation is to be justified as a corrective to perceived deficiencies in the operation of the market’.61 If left alone, markets can work inefficiently due to their fragility. The State needs to understand the barriers and to offer businesses solutions.62 Thus, it may be necessary to have a legislative intervention.63 A legislative intervention may alter private rights for the benefit of the weaker party. The aim of contracts should be the achievement of ‘a fair division of wealth among the members of society’.64 Thus, there is a need for the legislature to create mandatory terms that will prevent the free alienability of rights of payment (receivables). Parties can still agree to insert an anti-assignment clause and agree mutually to adhere to it and their private choice may prevail over the legislative/collective choice. However, in this case, there is at least a legislative

58 www.gov.uk/government/uploads/system/uploads/attachment_data/file/383538/ draft-statutory-instrument-business-contract-terms-restrictions-on-assignment-of-receivables-regulations-2015.pdf and see also www.gov.uk/government/news/restrictions-lifted-on-invoicefinance-to-help-small-firms-grow. 59 ‘Measure to Nullify ban on assignment clauses in a debtors terms of sale’ DBIS Impact Assessment document (2014) paras 38–39. 60 See generally for the context of regulation, markets and private law, A Ogus, Regulation: Legal Form and Economic Theory (Oxford, Oxford University Press, 1994) ch 2. 61 Ogus (n 60) 15. 62 A Schwartz and RE Scott, ‘Contract Theory and the Limits of Contract Law’ (2003) 113 Yale Law Journal 541, 555. 63 See, for a similar view, L Gullifer, ‘Should clauses prohibiting assignment be overridden by Statute?’ (2015) 4 Penn State Journal of Law and International Affairs 47, 65ff. 64 AT Kronman, ‘Contract Law and Distributive Justice’ (1979) 89 Yale Law Journal 472, 472.

Non-assignment Clauses

89

protective measure for the benefit of the weaker party in the contractual process.65 These regulatory arguments suggest that ‘the coercive power of government can be used to give valuable benefits to particular individuals or groups’.66 Regulatory legislative intervention is also necessary to reduce social costs (economic costs plus external costs)67 of restricting assignment. While the credit or the business relationship between the assignor and the debtor may be established on the economic/ private costs calculation, which may be characterised and defined by the strong bargaining position of the supplier, this process may not be concerned about external costs.68 The strong bargaining party may rely on the benefit of the contract, and demand that this should only be performed by the contractual partner and impose a contractual restriction on assignment. The contractual restriction on assignment may be regarded as a condition of extending credit or even entering into a business relationship. This restriction, while perfectly normal under the doctrine of contractual freedom, may not take into account third parties’ or the assignor’s interests. In other words, the strong party in the bargaining process (ie the debtor) may not be concerned with the relationship that the assignor, say, a small business, may have with the assignee or other creditors. External costs, in this framework, relate to the costs that people who have business relations with the assignor, creditors of the assignor or the society at large, have to pay a price as a result of the economic cost. An assignor’s inability to access to finance69 due to restrictions on assignment has an adverse effect on external costs. Potentially, the assignor may default in its payments to third parties or may not pay wages. Thus, the ability to transfer a right to payment to a financier (by factoring or invoice discounting) is crucial. This will enable particularly small businesses who are entering into transactions with large businesses to be able to utilise the value of their receivables and access finance via factoring or invoice discounting methods.

V. UNCITRAL’s Approach to Anti-assignment Clauses The UN Convention on the Assignment of Receivables, Article 9, the Legislative Guide, Recommendation 24 and the UNCITRAL Model Law on Secured Transactions, Article 13 regulate, in a uniform manner, the rules that recognise the effectiveness of assignments made notwithstanding anti-assignment clauses.

65

See generally Ogus (n 60) 256–57. RA Posner, ‘Theories of Economic Regulation’ (1974) 5(2) Bell Journal of Economics and Management Science 335, 344. 67 On social costs see eg RH Coase, ‘The Problem of Social Cost’ (1960) 3 Journal of Law and Economics 1. 68 For a similar view see Schwartz and Scott (n 62) 555ff. 69 It is perfectly possible that the assignor may not be able to obtain bank finance, or other sources of finance and may have to rely on factoring or invoice discounting. 66

90

N Orkun Akseli

Their approach shows similarity to UCC, section 9-406, which overrides prohibitions on the assignment of receivables,70 and the Canadian PPSAs.71 The Receivables Convention, Article 9(1) recognises the effectiveness of an assignment by overriding contractual clauses that ban assignment of receivables arising from trade receivables (including sale or lease of goods, credit card receivables or receivables arising out of the licensing of intellectual property).72 This article does not affect any national or domestic rules or statutes that ban assignment of receivables. This is because the Convention only applies to an assignment of a receivable where either the assignment or the receivable are international.73 This requires the assignor and the assignee to be located in different States (‘international assignment’) or the assignor and the debtor must be located in different States (‘international receivables’).74 The assignment made notwithstanding an anti-assignment clause will be effective as against the debtor and the third parties such as the creditors of the assignor and his trustee in bankruptcy. Effectiveness of an assignment in violation of an anti-assignment clause would not adversely affect small debtors, as ‘they do not have the bargaining power to insert anti-assignment clauses in their contracts and … would continue paying the same bank account or post office box.’75 And it would not affect the large debtors, as they have sufficient bargaining power.76 By a similar token, UNCITRAL Model Law, Article 13(1)77 recognises the effectiveness of a security right created notwithstanding a contractual restriction limiting the grantor’s ability to create a security right. This enables the grantor to be able to utilise its receivables by creating security in return for credit. Article 13(1) does not affect statutory limitations on the creation and enforcement of security rights over consumer and sovereign receivables.78

70 UCC, § 9-406(d) overrides restrictions and provides free alienability of rights to payment and that any agreement between an account debtor and an assignor is ineffective. UCC, §§ 9-406(d)(2) and 9-408(a)(2) eliminate all argument that any assignment made notwithstanding an anti-assignment clause nonetheless constitutes breach as between debtor and assignor: Image Point, Inc v JP Morgan Chase Bank, Nat. Ass’n 2014, 27 F. Supp.3d 494. See above. 71 See above. 72 For a more detailed treatment of anti-assignment clauses under the Receivables Convention see eg O Akseli, ‘Contractual Prohibitions on Assignment of Receivables: An English and UN Perspective’ (2009) JBL 650. In the United States under the UCC Art 9 regime, UCC, § 9-406(d) provides free alienability of rights to payment and that any agreement between an account debtor (debtor) and an assignor is ineffective. 73 UNCITRAL Convention, Art 1(a) ‘Assignment of international receivables’ or ‘international assignments of receivables’. 74 UNCITRAL Convention, Art 3. 75 A/CN.9/WG.II/WP.105, para 83; see also A/CN.9/489, para 103. 76 A/CN.9/WG.II/WP.105 para 83. The Addendum to the Draft Legislative Guide on Secured Transactions para 230 clearly indicates that a debtor such as a consumer may protect itself through statutory prohibitions: A/CN.9/631/Add 1. 77 UNCITRAL Model Law, Art 13(1) reads as follows: ‘A security right in a receivable is effective notwithstanding any agreement between the initial or any subsequent grantor and the debtor of the receivable or any secured creditor limiting in any way the grantor’s right to create a security right.’ 78 UNCITRAL Model Law, Art 1, paras 5 and 6.

Non-assignment Clauses

91

The Receivables Convention protects the assignee, under Article 9(2), by providing that the breach of an anti-assignment clause by the assignor is not in itself sufficient reason for the avoidance of the original contract by the debtor. The liability of the assignor for breach of the anti-assignment clause is preserved under the Receivables Convention; however, the debtor may not terminate the agreement on the grounds of breach of an anti-assignment clause (Articles 9(2) and 10(3)) or raise against the assignee. Same approach is adopted under UNCITRAL Model Law, Article 13(2). This approach prevents the debtor avoiding the contract and strengthening his bargaining power.79 It is argued that the assignee is given some confidence in the outcome of the transaction. The assignor may be held liable for damages for breach of contract if this type of liability is available under existing national law. Therefore, the right to compensatory damages that the debtor may have under the applicable law has been left outside the Receivables Convention.80 Article 9(2) expressly protects a person who is not party to an agreement between the assignor and the debtor on the sole ground that he had knowledge of the agreement. In general, the sole knowledge of the assignee of the anti-assignment clause is irrelevant and he cannot be held liable on the sole ground of his knowledge of it by the debtor.81 There must be additional grounds in order for the assignee to be held liable as the third party. The assignee’s inferred knowledge is not enough: his knowledge must be actual knowledge.82 However, knowledge may be relevant in the case of tortious liability of the assignee, such as for malicious interference with advantageous relations.83 The law attributes tortious liability to the assignee where the assignee, whilst having knowledge of an anti-assignment clause, induces the assignor to breach that clause.84 The UCC does not attribute tortious liability to the assignee.85 Article 18(3) of the Receivables Convention does not allow the debtor to make a claim for breach of an anti-assignment clause against the assignee by way of set-off so as to defeat the assignee’s demand for payment. The Contracting States are not permitted to make a declaration to override the effectiveness of the provision of free assignability. Under Article 40, a Contracting State is permitted to declare whether an assignment of a receivable owed by a governmental debtor in that State will be excluded from the Convention rules that override contractual anti-assignment terms. Although the Convention overrides the

79 S Bazinas, ‘Key Policy Issues of the United Nations Convention on the Assignment of Receivables in International Trade’ (2003) 11 Tulane Journal of International and Comparative Law 275, 287. 80 A/CN.9/489, para 99. 81 UNCITRAL Model Law, Art 13(2), which reads as follows: ‘… A person that is not a party to the agreement referred to in paragraph 1 is not liable for the grantor’s breach of the agreement on the sole ground that it had knowledge of the agreement’. 82 Swiss Bank Corp v Lloyds Bank Ltd [1979] Ch 548 at 575, [1979] 2 All ER 853 (ChD); see also McCormack (n 22) 232. 83 A/CN.9/470, para 102; see also A/CN.9/WG.II/WP.105, para 85. 84 A McKnight, ‘Restrictions on Dealing with Assets in Financing Documents: Their Role, Meaning and Effect’ (2002) 17 Journal of International Banking Law 193, 198 ff. 85 UCC, § 9-406, Cmt 5 and Art § 9-402 Cmt 2; Brandes v Pettibone Corp 1974, 79 Misc. 2d 651; 360 N.Y.S.2d 814.

92

N Orkun Akseli

effectiveness of anti-assignment clauses by virtue of Article 9, this provision will have no effect on a sovereign debtor who is located in a Contracting State if that State makes a declaration by virtue of Article 40, and Article 9 does not apply to restrictions arising by statute or other rule of law. It might have been a better and more consolidated approach had the Receivables Convention treated sovereign debtors and ordinary debtors on an equal basis and granted effectiveness to assignments made notwithstanding an anti-assignment clause between assignors and sovereign debtors.86 UN Convention, Article 9, the UNCITRAL Model Law, Article 13 and the Legislative Guide, Recommendation 24 in a similar way apply only to trade receivables. They do not apply to financial receivables.87 In other words, the scope of these rules is limited to the assignment of receivables arising from an original contract that is a contract for the supply or lease of goods or services other than financial services, a construction contract or a contract for the sale or lease of real property; arising from an original contract for the sale, lease or licence of industrial or other intellectual property or of proprietary information; representing the payment obligation for a credit card transaction; or owed to the assignor upon net settlement of payments due pursuant to a netting agreement involving more than two parties.88

VI. Conclusions Financing small- and medium-sized businesses is vital in the relative absence of bank lending in the aftermath of the global financial crisis. One method of reducing the financial tension on small businesses is to enable them to utilise their receivables. This can be achieved by removing barriers to the assignability of receivables. Recognising the effectiveness of assignments made notwithstanding anti-assignment clauses is part of many modern commercial laws. While nullifying anti-assignment clauses may distort market practices and forces, the law should also support small businesses at times of crisis. The law should enable them to utilise their capital by assigning their receivables in bulk and assigning their future receivables. Therefore, to the extent that these clauses are inserted by

86

cf A/CN.9/466, paras 107–115. The reason for inapplicability has been explained in the Legislative Guide as follows: ‘It does not apply to so-called “financial receivables”, because where the debtor of the receivable is a financial institution, even partial invalidation of an anti-assignment clause could affect obligations undertaken by the financial institution towards third parties. Such a result is likely to have negative effects on important financing practices, such as those involving the assignment of receivables arising from or under securities or financial contracts’ para 108, at 93. See also UNCITRAL Model Law, para 100. 88 UN Convention, Art 9(3); UNCITRAL Model Law, Art 13(3); Legislative Guide, Recommendation 24(3). 87

Non-assignment Clauses

93

the strong bargaining party to the contracts against small businesses, they should be rendered ineffective, which will enable small businesses to use factoring and securitisation financing techniques to access to credit.89 This can only be achieved through a legislative intervention that offers lower transaction costs than negotiating the anti-assignment clauses. UNCITRAL’s text has the necessary detailed and modern approach to achieve this result.

89

See also Legislative Guide, para 110, at 93.

94

 

6 Dealing with Concepts of Property in the UNCITRAL Legislative Guide on Secured Transactions and UNCITRAL Model Law on Secured Transactions (2016) STEVEN O WEISE*

I. Introduction A significant challenge in preparing the UNCITRAL Legislative Guide on Secured Transactions (the Legislative Guide) and the UNCITRAL Model Law on Secured Transactions (2016) (the UNCITRAL Model Law) was to develop a set of recommendations and provisions that accommodated both civil law and common law approaches. This need to accommodate each system of law became particularly acute when considering questions of what constitutes ‘property’ for the purposes of applying the scope rules of the Legislative Guide and the UNCITRAL Model Law.1 These documents provide for a substance-over-form, functional approach. The use of the functional approach served as the basis for solving the apparent conundrum of resolving the differences of how civil law and common law regimes addressed the meaning of ‘property’.

* Steven O Weise is a partner in the Los Angeles, California office of the law firm Proskauer Rose LLP. Since 2002 he has been the representative of the American Bar Association to the U NCITRAL Working Group VI. He has been an advisor to and a member of the Drafting Committees for Article 9 (Secured Transactions) of the Uniform Commercial Code. He is a member of the Permanent Editorial Board for the Uniform Commercial Code and is a member of the Council of the American Law Institute. 1 See generally U Drobnig and O Böger, Proprietary Security in Movable Assets (Sellier European Law Publishers, 2015) Introduction, A 17; ch 2, Art IX–2:112, Comments.

96

Steven O Weise

The Legislative Guide begins in its ‘Purpose’ section by telling us that it applies to security rights in all ‘movable assets.’ Paragraph 17 of the Legislative Guide refers to ‘property’ as follows: References to ‘property right’ are to be interpreted as referring to a right in an asset (right in rem) as opposed to a personal right (right ad personam).

Similarly, the UNCITRAL Model Law also provides ‘This Law applies to security rights in movable assets.’ The UNCITRAL Model Law also may also seem to distinguish between a ‘property right’ and a ‘personal right.’ For example, the title to Article 14 is: Article 14. Personal or property rights securing or supporting payment or other performance of encumbered receivables or other intangible assets, or negotiable instruments

In common law systems, ‘property’ may include both in rem rights to a ‘thing’ (even if intangible) as well as personal rights between two persons with respect to a thing. Thus a person’s rights under a contract may be ‘personal’ with respect to the other party to the contract and yet ‘property’ with respect to the claims of third persons with respect to the party’s rights under the contract. Rod Macdonald (and his colleagues from Canada) were well-situated to address these different approaches to ‘property’ as they were familiar with the legal system in Quebec that embraces both civil law and common law approaches to ‘property.’ One of the many achievements of the Legislative Guide and the UNCITRAL Model Law was to arrive at a practical approach to incorporate the results anticipated by the two different types of legal systems. This tension arose in particular in three areas: (1) rights to claims that supported encumbered assets; (2) rights of persons with contractual rights with respect to encumbered assets; (3) rights of a licensee of intellectual property. In each case, by taking a functional approach, the Working Group was able to achieve the policy goals of the Legislative Guide and the UNCITRAL Model Law while respecting the civil and common law legal regimes.

II. Rights to Claims that Support Encumbered Assets An encumbered asset itself is often supported or secured by a security right or secondary obligation. That underlying security right or claim against a secondary obligor might be considered a personal claim of the person granting a security right in the claim that is secured or guaranteed. As such, if it is not ‘property,’ then it might seem to fall outside the scope of the Legislative Guide and the UNCITRAL Model Law and thus it would not be feasible to grant a security right

Concepts of Property and UNCITRAL Instruments

97

in that contractual, personal claim. However, in common law systems, it might be considered part of the property right in which the security right is being created.2 The Working Group, with able assistance from Rod Macdonald, focused on the functional, and not the formal, effect of the security right. It did so by referring to the ‘benefits’ of a contractual right to an encumbered asset and by referring to the secured creditor having the ‘benefit’ of the underlying security right or obligation of the secondary obligor. The Legislative Guide provided in the following recommendations (emphasis added): 25. The law should provide that: (a) A secured creditor with a security right in a receivable, negotiable instrument or any other intangible asset covered by this law has the benefit of any personal or property right that secures payment or other performance of the receivable, negotiable instrument or intangible asset automatically, without further action by either the grantor or the secured creditor; … (d) A secured creditor with a security right in a receivable, negotiable instrument or any other intangible asset covered by this law has the benefit of any personal or property right that secures payment or other performance of the receivable, negotiable instrument or other intangible asset notwithstanding any agreement between the grantor and the debtor of the receivable or the obligor of the negotiable instrument or other intangible asset limiting in any way the grantor’s right to create a security right in the receivable, negotiable instrument or other intangible asset, or in any personal or property right securing payment or other performance of the receivable, negotiable instrument or other intangible asset.

Similarly, the UNCITRAL Model Law provides (emphasis added): Article 14. A secured creditor with a security right in a receivable or other intangible asset or in a negotiable instrument has the benefit of any personal or property right that secures or supports payment or other performance of the encumbered asset without a new act of transfer.

III. Rights of Persons with Contractual Rights to Encumbered Assets A similar issue arose concerning the rights of persons with a contractual right to use encumbered assets. The Working Group readily agreed that certain buyers of

2

See generally Drobnig and Böger (n 1) ch 2, Art IX–2:401, Comments.

98

Steven O Weise

encumbered assets would take their ‘property’ rights free of the security right in the encumbered asset: 81. The law should provide that: (a) A buyer of a tangible asset (other than a negotiable instrument or negotiable document) sold in the ordinary course of the seller’s business takes free of a security right in the asset, provided that, at the time of the sale, the buyer does not have knowledge that the sale violates the rights of the secured creditor under the security agreement;

The UNCITRAL Model Law has a similar provision: Article 34. … (4) A buyer of a tangible encumbered asset sold in the ordinary course of the seller’s business acquires its rights free of the security right, provided that, at the time of the conclusion of the sale agreement, the buyer does not have knowledge that the sale violates the rights of the secured creditor under the security agreement.

However, the answer was not so readily achieved when considering the rights of a third person that obtains a contractual, but not ownership, right with respect to an encumbered asset.3 As with a security right in an encumbered asset having the ‘benefit’ of a security right that secured the encumbered asset or the ‘benefit’ of a secondary obligation with respect to the encumbered assets, the Working Group again focused on the functional, and not the formal, effect of the security right. The Legislative Guide did so by referring to whether the security right would ‘affect’ the rights of the person holding the contractual right to use the encumbered property (emphasis added): 81. The law should provide that: … (b) The rights of a lessee of a tangible asset (other than a negotiable instrument or document) leased in the ordinary course of the lessor’s business are not affected by a security right in the asset, provided that, at the time of the conclusion of the lease, the lessee does not have knowledge that the lease violates the rights of the secured creditor under the security agreement; and (c) The rights of a non-exclusive licensee of an intangible asset licensed in the ordinary course of the licensor’s business are not affected by a security right in the asset, provided that, at the time of the conclusion of the licence agreement, the licensee does not have knowledge that the licence violates the rights of the secured creditor under the security agreement. 82. The law should provide that, if a transferee acquires a right in an encumbered asset free of a security right, any person that subsequently acquires a right in the asset from the

3 See generally Drobnig and Böger (n 1) ch 3, Art IX–3:330, Comments; ch 5, Art IX–5:303, Comments; ch 6, Art IX–6:102, Comments.

Concepts of Property and UNCITRAL Instruments

99

transferee also takes its right free of the security right. If the rights of a lessee or licensee are not affected by a security right, the rights of a sub-lessee or sub-licensee are also unaffected by the security right.

The UNCITRAL Model Law adopted this approach (emphasis added): Article 34. … 5. The rights of a lessee of a tangible encumbered asset leased in the ordinary course of the lessor’s business are not affected by the security right, provided that, at the time of the conclusion of the lease agreement, the lessee does not have knowledge that the lease violates the rights of the secured creditor under the security agreement.

IV. Rights of a Licensee of Intellectual Property A similar issue arose in the context of the ability of a licensee of intellectual property to grant a security right in its contractual rights. Under the intellectual property law of many States, those rights are ‘personal’ and are not ‘property’.4 Once again, the Working Group came up with a practical approach, avoiding concepts of ‘property’ and instead referring to a security right in a licensee’s rights (emphasis added): 107. Typically, a licensee is authorized to use or exploit the licensed intellectual property in line with the terms and conditions of the licence agreement. Some laws relating to intellectual property provide that the licensee may not create a security right in its authorization to use or exploit the licensed intellectual property without the licensor’s consent (although in many States an exception may arise where the licensee sells its business as a going concern). The reason is that it is important for the licensor to retain control over the licensed intellectual property and who can use it. If such control cannot be exercised, the value of the licensed intellectual property may be materially impaired or lost completely. If, however, the rights of a licensee under a licence agreement are transferable and the licensee creates a security right in them, the secured creditor will take a security right in the licensee’s rights subject to the terms and conditions of the licence agreement. If the licence is transferable and the licensee transfers it, the transferee will take the licence subject to the terms and conditions of the licence agreement. The law recommended in the Guide does not affect these licensing practices.

And, once again, the UNCITRAL Model Law follows this approach (emphasis added): Article 34. … 6. Subject to the rights of a secured creditor with a security right in intellectual p roperty in accordance with article 50, the rights of a non-exclusive licensee of an intangible 4

See generally Drobnig and Böger (n 1) Introduction, A, 17.

100

Steven O Weise

encumbered asset licensed in the ordinary course of the licensor’s business are not affected by the security right, provided that, at the time of the conclusion of the licence agreement, the licensee does not have knowledge that the licence violates the rights of the secured creditor under the security agreement.

V. Conclusion Rod Macdonald made an enormous contribution to the Working Group’s successful achievement of accommodating different legal systems and their approaches to concepts of ‘property’ by taking a functional approach to solving these problems. We will always be grateful to him for that.

7 Current Issues in Cross-Border Asset-Based Lending: Lessons from the Field on the Need for Secured Transactions Reform RICHARD M KOHN*

Author’s Note: Remembering Rod Macdonald Secured transactions reform is a high-stakes game. Low-cost credit for small- and medium-sized enterprises can create vitally needed jobs, raise standards of living and promote social stability within a country. When the credit stimulates commerce among businesspeople in different countries, its effects can be even more profound, generating understanding and mutual trust across national boundaries. In this way, secured transactions reform can facilitate the realisation of human aspirations and at the same time promote the common good. Secured transactions reform was, in other words, a perfect field for Rod Macdonald, and it should come as no surprise that his influence in the area was extraordinary. This influence was felt time and time again at the UNCITRAL meetings, in the way he applied his extensive knowledge and practical wisdoms, infused with his impeccable values, to the issues at hand. Often, when we reached an impasse in our deliberations, Rod came to the rescue, offering a well-reasoned and pragmatic solution that carried the day. I had the great privilege of working with Rod on various UNCITRAL projects for more than 12 years. The highlight came when, after eight years of work on the U NCITRAL Legislative Guide on Secured Transactions, Spiros Bazinas of the U NCITRAL Secretariat asked Rod and me to join him in editing the Guide for publication. This gave me the opportunity to work closely with Rod for a full year, over many long meetings and phone calls. The experience was a highlight of my professional career, and continues to inspire me in many ways. * The author wishes to thank Jonathan M Cooper, William A Starshak and Sophie D Rezki for their valuable help in preparing this chapter.

102

Richard M Kohn

Rod’s illness prevented him from working on the most recent UNCITRAL p roject: the Model Law on Secured Transactions. He was sorely missed, but the lessons he taught remained ever-present in the UNCITRAL meeting rooms, wisely and gently informing our work. With deep respect and gratitude, I dedicate this chapter to him.

I. Introduction Recent years have witnessed a growing appreciation of the way in which secured transactions can stimulate economic growth, as more and more countries have adopted modern secured transactions regimes. The strength of this perception was evident when, at the height of the recent financial crisis, Belgium adopted a new secured transactions law based on the UNCITRAL Legislative Guide on Secured Transactions (2007) (the Legislative Guide). As a contemporaneous report to the Belgian Parliament made clear, the law was expected to play a significant role in Belgium’s economic recovery: The economic crisis that Belgium is facing recalled the importance of the proper functioning of credit facilities to fund economic operators and boost growth and employment. However, the granting of credits is inseparable from granting security interests. By improving the efficiency of security, we reduce the cost of credit and therefore promote economic growth.1

Some countries, while stopping short of enacting completely new secured transactions regimes, have nevertheless modified their existing secured transactions regimes in specific ways that stimulate economic activity. Thus, in recent years the United Kingdom, the Netherlands and Singapore repealed their financial assistance laws insofar as they apply to private companies (thereby stimulating mergers and acquisitions of private companies by permitting them to grant security for loans made to finance the acquisition of their own shares) and the United Kingdom is, at the time of this writing, in the process of blunting the impact of bans on the assignment of receivables.2 Although there is wide agreement that a country’s adoption of a modern secured transactions regime can benefit domestic businesses,3 such a regime can also play 1 ‘Exposé introductif de Mme AnnemieTurtelboom, ministre de la justice, sur les projets de loi DOC 53 463/001 et DOC 53 2464/001’ in Rapport fait au nom de la commission de la justice par Mme Carina Van Cauter, 21 December 2012 (‘La crise économique à laquelle la Belgique est confrontée a rappelé l’importance d’un bon fonctionnement des mécanismes de crédit pour financer les opérateurs économiques et stimuler la croissance et l’emploi. Or, l’octroi de crédits est indissociable de la constitution de droits de sûreté. En améliorant l’efficacité de ces derniers, l’on réduit le coût du crédit et l’on favorise donc la croissance économique.’) 2 See, eg, www.gov.uk/government/news/restrictions-lifted-on-invoice-finance-to-help-smallfirms-grow. Financial assistance laws and anti-assignment clauses are discussed below. 3 See, eg, Legislative Guide, Introduction, paras 2–7.

Current Issues in Cross-Border Asset-Based Lending

103

a significant role in promoting cross-border financing transactions to meet the needs of an increasingly globalised middle market, such as cross-border credit facilities to middle-market companies with subsidiaries in multiple countries. This chapter examines the role of secured transactions reform in promoting cross-border financing transactions. It does so by examining secured transactions reform through the lens of ‘asset-based lending,’ a form of secured finance prevalent in the United States and in some other countries as well, that is increasingly used to accommodate the cross-border financing needs of middle-market companies. The chapter discusses various current issues that confront US asset-based lenders and their borrowers as they seek to bridge the gap between the crossborder financing needs of modern middle-market companies and those legal regimes that inhibit the realisation of those needs. Because of its intense focus on collateral, cross-border asset-based lending provides an ideal laboratory in which to investigate how the secured transactions regimes currently in effect in many countries negatively affect the cost and availability of credit, thereby making the case for secured transactions reform all the more compelling.

II. The Growing Cross-border Financing Needs of Middle-market Companies In many countries, middle-market companies have increasingly become globalised. This is certainly true in the US, where many middle-market companies routinely sell their goods and services to customers in other countries, set up foreign manufacturing and distribution operations and acquire foreign companies. This development has presented a significant challenge for banks, finance companies and other lenders that serve the financing needs of middle-market companies. These lenders are no longer able to confine themselves to making purely domestic loans in which the borrowers, guarantors and collateral all are located in the lender’s country and which are governed by domestic laws with which they are comfortable. Today, lenders seeking to accommodate the increasing cross-border financing needs of their customers must contend with unfamiliar foreign legal regimes in which it is often difficult or impossible to obtain the security rights, guaranties and other legal protections that are available under their home laws.

III. Asset-based Lending In the US, many middle-market businesses finance their cross-border activities using a financing technique known as asset-based lending. Asset-based lending

104

Richard M Kohn

has played a major role in financing middle-market businesses in the US for decades and, during the past 20 years, US asset-based lenders have been pioneers in accommodating the cross-border financing needs of their borrowers. An asset-based loan typically is a senior secured revolving loan facility under which the lender makes advances from time to time at the borrower’s request predicated upon the value of a ‘borrowing base’ comprised of the borrower’s receivables or inventory, or both. Not all receivables and inventory are eligible for inclusion in the borrowing base; only those that meet certain criteria spelled out in the loan agreement. Thus, for example, receivables that remain past due beyond an acceptable grace period and inventory that becomes obsolete drop out of the borrowing base. In addition to the primary collateral consisting of receivables and inventory, the lender in an asset-based loan will often secure the loan with other assets owned by the borrower, such as equipment, real property, intellectual property and shares in subsidiaries, and may also provide additional credit based on the value of those assets. When the borrower is a member of a corporate group, the loan often will be supported by guaranties from other entities in the group, which guaranties may, in turn, be secured by the assets of those entities. Because of its intense focus on the current value of receivables, inventory and other assets, an asset-based lender typically will require detailed periodic reports from the borrower concerning the collateral.4 From the lender’s perspective, the rationale for asset-based lending is that, because the loans are never permitted to exceed the borrowing base, the lender may reasonably expect that there will be sufficient collateral to satisfy the outstanding loans in full even if the borrower fails to repay the loans. Thus, it is often said that the lender’s ‘exit strategy’ (eg the avenue by which the loans are expected to be repaid in the event of default) is the sale or other disposition of the collateral. From the borrower’s perspective, asset-based lending is a highly efficient form of financing in which the borrower only pays interest on funds it needs to borrow, and more financing is available at a lower cost than if the lender had less collateral to mitigate its risk.

IV. Current Issues in Cross-border Asset-based Lending Because of its reliance upon receivables, inventory and other collateral as a way of reducing credit risk, asset-based lending provides an ideal format for making cross-border loans in countries with legal regimes that enable lenders to obtain and enforce security rights in such collateral in an efficient and predictable 4 For a more detailed description of asset-based lending, see Legislative Guide, Introduction, paras 28–30.

Current Issues in Cross-Border Asset-Based Lending

105

way. However, when venturing into other countries, US asset-based lenders have been confronted with a variety of challenges. First and foremost is the challenge of identifying ‘lender-friendly countries’, ie countries with laws that enable such lenders to obtain the functional equivalence of the legal protections they enjoy in the US. Yet even in lender-friendly countries, lenders often encounter challenges that require them to adjust the structure and terms of the loans they are willing to make. The following discussion examines certain of these challenges.

A. Challenge No 1: The Identification of Lender-friendly Countries In recent years, US asset-based lenders have devoted substantial attention to identifying countries where the laws make it possible to obtain, in a reasonably efficient and cost-effective way, the functional equivalence of the essential components of US-style asset-based loans. Of greatest importance is the ability to obtain security rights in collateral that are enforceable against third parties (including an insolvency administrator in the grantor’s insolvency proceeding), coupled with clear rules for determining the priority of such security rights and a reasonably efficient method of enforcing them. Also important, however, is the ability to obtain enforceable cross-corporate guaranties from other entities in the borrower’s corporate group, and the ability to obtain a sufficient level of control over the proceeds of encumbered receivables. Thus, the evaluation as to whether a country’s laws are lender-friendly involves much more than an examination of the country’s secured transactions laws. Other areas of the country’s law, such as corporation and insolvency laws, are also implicated, as is the quality of the courts in terms of their reliability, efficiency and freedom from corruption. Many cross-border lenders and their lawyers agree as to the identity of certain lender-friendly jurisdictions, such as Australia, Belgium, Canada, Germany, the Netherlands, New Zealand, Switzerland (with respect to financing receivables)5 and the United Kingdom (as well as certain jurisdictions with legal regimes heavily influenced by UK law, such as Hong Kong and Singapore). Interestingly, this list is not confined to common law countries. It includes certain civil law countries as well, demonstrating that, in the field of secured transactions law, labels are not dispositive. Thanks to the efforts of UNCITRAL and other international organisations, the list of lender-friendly countries can be expected to expand dramatically in coming years. Although the jurisdictions named above are considered to be lender-friendly, none of them is favourable in all respects from a US asset-based lender’s perspective. For example, as of the time of this writing:

5 Swiss law does not currently recognise a non-possessory security right in inventory (ie, a security right that allows for the borrower to retain possession of the inventory).

106

Richard M Kohn

(i) the UK registry system incorporates a 21-day grace period for registration of security rights, which creates uncertainty for lenders by permitting a laterregistered security right to obtain priority in certain circumstances; (ii) US lenders generally accept that they are unable to rely upon an upstream or sidestream guaranty given by a German limited liability company known as a Gesellschaft mit beschränkter Haftung (GmbH) by reason of Section 30 of the German Act on Limited Liability Companies,6 which has been interpreted to permit a GmbH to refuse to honour a demand for payment under such a guaranty if making the payment would render the company insolvent; (iii) the absence of a ‘searchable’ secured transactions registry in the Netherlands,7 and the absence of any registry at all for security rights in most forms of personal property in Germany; (iv) the general recognition in many lender-friendly countries of retention-oftitle claims of suppliers, and of anti-assignment clauses restricting the ability of an owner of receivables to encumber such receivables.8 The existence of such infirmities in a country typically will not make that country an unfavourable lending jurisdiction if its laws are otherwise favourable. Lenders often are willing to accept the risks posed by such infirmities or are able to find ways of addressing them (some of which are discussed below), though doing so will often increase the cost of, or reduce the borrowing availability under, the credit facility. The list of lender-friendly countries will undoubtedly grow in the coming years, as the work of UNCITRAL and other international organisations continues to influence the adoption of modern secured transactions laws. The Model Law on Secured Transactions adopted by UNCITRAL in July 2016 (the UNCITRAL Model Law), which offers countries an efficient way to adopt a state-of-the-art secured transactions regime, promises to play a particularly important role in that process.

B. Challenge No 2: Centralisation of Receivables An increasingly prevalent theme in US cross-border asset-based lending is the tension between a borrower’s desire for a corporate structure that is efficient from a tax perspective, on the one hand, and the borrower’s desire to maximise its ability to obtain financing, on the other. To accomplish the former goal, the borrower and its tax advisors may have developed a complex corporate structure with subsidiaries in many countries, each of which owns the receivables that it generates by selling goods and services locally. However, such a corporate structure may make

6

Gesetz betreffend die Gesellschaften mit beschränkter Haftung (GmbH-Gesetz), Germany. the Netherlands, security rights in receivables and inventory are registered with the taxing authorities for evidentiary purposes. 8 See discussion of anti-assignment clauses below. 7 In

Current Issues in Cross-Border Asset-Based Lending

107

it difficult for the corporate group to obtain financing based on the receivables owned by subsidiaries domiciled in countries where the legal regime is not favourable to secured lending. In some cases, the borrower’s tax savings generated by its optimal tax structure may be outweighed by its inability to obtain the financing it needs to conduct its operations. One possible solution to this problem is for the company to ‘centralise’ its receivables in a lender-friendly country, such as Ireland or the Netherlands, which has a legal regime that accommodates lending against receivables. This is typically accomplished by creating a subsidiary (or designating an existing subsidiary) domiciled in a lender-friendly country that acts as a financing entity to own the receivables. Receivables are either: (i) sold by the local entities to the financing entity as they arise via a ‘true sale’; or (ii) generated by the financing entity in the first place. In either event, it is essential that the receivables be owned by the financing entity, at the time they are included in the borrowing base, in a way that will survive a claim by the insolvency administrators of the local entities that the local entities remained the actual owners of the receivables. Thus, for example, it is not sufficient that the financing entity act purely in an administrative capacity as the entity that merely invoices the receivables; the financing entity must own the receivables outright. Because centralisation of receivables is a technique developed to address the lack of modern secured transactions laws in the jurisdictions in which the borrower’s subsidiaries are domiciled, the need for centralisation will decrease as more countries adopt modern secured transactions regimes.

C. Challenge No 3: Maximising the Borrowing Base Once an asset-based lender determines that a given jurisdiction is acceptable for lending, the focus shifts to determining the composition of the borrowing base. This, in turn, requires a nuanced examination of the jurisdiction’s secured transactions laws to identify factors that could negatively affect the value of the borrower’s receivables and inventory as collateral.

i. Anti-assignment Clauses One such factor is the treatment afforded by the jurisdiction’s laws to ‘antiassignment clauses’, which are provisions in the documents evidencing a receivable that purport to restrict the assignment or encumbering of the receivable by its owner without the consent of the account debtor. The laws of some countries, such as the US, render anti-assignment clauses unenforceable in order to promote receivables financing,9 while the laws of some other countries reduce

9

See, eg, Uniform Commercial Code, § 9–406(d).

108

Richard M Kohn

their effectiveness. However, such clauses remain fully enforceable in many jurisdictions. An asset-based lender typically will treat receivables containing enforceable anti-assignment clauses as ineligible for borrowing, discarding them from the borrowing base (thereby reducing the amount of credit otherwise available to the borrower) unless the account debtor consents to the assignment or security right. At a minimum, the lender will need to incur the cost of due diligence to ascertain the presence of anti-assignment clauses in the documents evidencing the borrower’s receivables. This cost, which must be incurred even if it is ultimately determined that no anti-assignment clauses exist, is typically passed on the borrower, thereby increasing the borrower’s cost of credit. As noted above, an effort is underway in the UK to blunt the effect of anti-assignment clauses, a step that undoubtedly will go a long ways toward promoting receivables financing in the UK, and the UNCITRAL Model Law contains a provision that significantly limits the impact of anti-assignment clauses.10

ii. Retention-of-title Clauses A second factor is the treatment afforded by the jurisdiction’s laws to retentionof title clauses in favour of the borrower’s inventory suppliers. These clauses permit the suppliers to retain title to the inventory until the borrower pays the purchase price for the inventory in full. Although these clauses are recharacterised as unperfected security rights under the laws of some countries,11 they are not only enforceable, but commonplace, in many other countries. Some countries, such as Germany, recognise ‘extended’ retention-of-title claims, which extend the supplier’s claim to the receivables generated from the sale of the inventory, as well as the work-in-process resulting from the manufacturing process. Although retention-of-title claims must be evidenced by a writing, they are typi cally not subject to any registration requirements, which means that their existence can only be ascertained through due diligence conducted by the lender. An asset-based lender typically will treat inventory subject to enforceable retentionof-title claims as ineligible for borrowing, and either exclude such inventory from the borrowing base or create a reserve against borrowing availability in an amount up to the unpaid obligations owing to the suppliers holding such claims, in either case reducing the amount of credit available to the borrower. As in the case of antiassignment clauses, the borrower typically is required to bear the cost of this due diligence, regardless of whether it is ultimately determined that any retention-of-title claims exist, thereby driving up the cost of the financing. Modern secured transactions regimes typically treat retention-of-title claims as security rights that must be disclosed in the secured transactions registry in order 10 See, UNCITRAL Model Law, Art 13. This is also the approach taken in Art 9 of the UN Convention on the Assignment of Receivables in International Trade (2001). 11 See, eg, UCC, §§ 1–201(b)(35) and 2–401(1).

Current Issues in Cross-Border Asset-Based Lending

109

to be enforceable against third parties, and this is also the approach taken by the UNCITRAL Model Law.12

iii. Carve-outs for Unsecured Creditors and other Preferential Claims A third factor is the various claims of creditors that are accorded priority over a security right under local law. One category of such claims is carve-outs for unsecured creditors. For example, UK law currently recognises a carve-out for unsecured creditors of up to £600,000 of the proceeds of collateral subject to a floating charge (known as the ‘prescribed part’ or ‘ring-fenced pot’). Another example is the Finnish enterprise mortgage, which is subject to a carve-out for unsecured creditors of 50 per cent of the proceeds of the assets subject to the mortgage. Another category of preferential claims is claims for certain unpaid wages owing to employees. Mexico and Australia are examples of countries that recognise such claims. Still another category is claims in favour of insolvency administrators. For example, German insolvency law provides for a priority claim for fees owing to an insolvency administrator at a minimum rate of 9 per cent of the proceeds of assets included in the insolvency estate.13 Because these preferential claims are not allocable to specific receivables or inventory, an asset-based lender often will address the existence of such claims by creating reserves against loan availability in the amount of the potential preferential claims. For example, it has become routine practice in the UK for lenders to protect against the prescribed part by reserving up to £600,000 from borrowing availability. Thus, preferential claims can significantly reduce the amount of credit available to the borrower from its asset-based credit facility, ironically depriving the borrower of funds that could be used to pay the very unsecured creditors, suppliers, employees and other parties whom the priority claims were designed to protect. There is an increasing awareness of how the existence of preferential claims can negatively impact the availability of credit. For example, for a number of years, Sweden’s enterprise mortgage was subject to a 45 per cent carve-out for unsecured creditors. However, in 2010 Sweden repealed this provision of its law because the existence of the carve-out was making it difficult for Swedish SMEs to

12 Art 2, para (b) of the UNCITRAL Model Law defines an ‘acquisition security right’ as ‘a security right in a tangible asset … which secures an obligation to pay any unpaid portion of the purchase price of an asset, or other credit extended to enable the grantor to acquire rights in the asset to the extent that the credit is used for that purpose.’ Art 2, para (kk) of the UNCITRAL Model Law defines a ‘security right’ as ‘a property right in a movable asset that is created by an agreement to secure payment or other performance of an obligation, regardless of whether the parties have denominated it as a security right, and regardless of the type of asset, the status of the grantor or secured creditor, or the nature of the secured obligation.’ These two definitions, taken together, make it clear that a retention-of-title arrangement is considered to be a security right for purposes of the UNCITRAL Model Law. 13 German Insolvency Code, Arts 170–171.

110

Richard M Kohn

obtain credit. Modern secured transactions laws that limit or eliminate carve-outs and other priority claims would go a long way toward streamlining the diligence process, thereby reducing the cost of credit, while at the same time increasing the amount of credit available to the borrower. Consistent with this approach, the Legislative Guide acknowledges the adverse impact that carve-outs for unsecured creditors may have on the availability of credit,14 and recommends that preferential claims be limited both in type and amount.15 This policy is implemented in Article 36 of the UNCITRAL Model Law, which seeks to promote certainty for secured creditors by permitting countries to list, in one place, all statutory claims that may have priority over security rights, and to indicate the maximum amount of such claims.

D. Challenge No 4: Consolidated Borrowing Bases As noted above, in a typical asset-based credit facility, periodic revolving advances are made to the borrower predicated on the value of a borrowing base comprised of receivables and inventory that are deemed eligible for borrowing. Purely US domestic asset-based credit facilities to corporate groups comprised of various US entities often will employ a ‘consolidated’ borrowing base, under which multiple entities are designated as ‘co-borrowers’ that are jointly and severally liable for all of the obligations owing under the credit facility, and eligible receivables and inventory owned by all of the co-borrowers are combined to form a single borrowing base. This approach enables one entity in the corporate group to borrow based on receivables and inventory owned by other members of the group, thereby increasing the amount that each entity may borrow, and creating flexibility to deploy loan proceeds to the entities that need funds the most. Thus, if one entity in the group (Company A) lacks sufficient borrowing availability based on its own receivables and inventory to generate all of the working capital advances it needs to run its business, and another entity in the group (Company B) has excess borrowing availability, Company A may borrow based on the receivables and inventory owned by Company B. A current issue in cross-border asset-based lending is the extent to which this technique may be employed in a cross-border context. From the lender’s perspective, permitting Company A to borrow based on the assets of Company B is prudent only if the laws of Company B’s jurisdiction permit Company B to become liable for the obligations of Company A. The reason is that, if the facility is in default, the lender may seek to enforce its security rights in the assets of Company B to support the loans to Company A, ahead of the claims of the unsecured creditors of Company B. This can only be accomplished if, under the laws of the country in which Company B is domiciled, it is possible for Company 14 15

See Legislative Guide, ch II, para 66. See Legislative Guide, ch V, paras 90–93 and Recommendation 83.

Current Issues in Cross-Border Asset-Based Lending

111

B to become liable, as co-borrower or guarantor, for the obligations of an affiliate company. Because countries typically test the enforceability of joint and several liability of co-borrowers and cross-guaranties by affiliates using the same legal principles, for convenience the following discussion will focus on the laws pertaining to guaranties. In the US, an upstream or sidestream guaranty is tested primarily by US creditors’ rights law: the guaranty will be enforceable under section 548 of the US Bankruptcy Code even if the guarantor company did not receive reasonably equivalent value for the guaranty if, measured as of the time of its execution, the guaranty did not render the guarantor insolvent or unable to pay its debts as they fall due, or leave it with insufficient capital to conduct its business.16 States in the US also have creditors’ rights laws that impose a substantially similar test. Directors of US companies often are willing to authorise the execution of such guaranties in the context of a financing that benefits the corporate group as a whole. In many countries, however, the law pertaining to upstream and sidestream guaranties is quite different.17 Such guaranties are tested as a matter of corporate governance, and directors act at their peril in authorising a company to execute a guaranty that is determined not to be in the guarantor’s best interests. The standard for determining the best interests of the guarantor varies greatly from country to country. Under the laws of some jurisdictions (such as Australia, Canada, the Netherlands and the UK), the guaranty may properly be authorised if the financing guarantied is deemed to be in the best interests of the corporate group taken as a whole. In other countries, a financial test is imposed, either by law or market practice. In Luxembourg, for example, under current market practice, directors of a Luxembourg company will usually authorise an upstream or sidestream guaranty so long as it contains a provision that it is only enforceable up to a specified percentage (often 85–95 per cent) of the net assets of the guarantor, determined either at the time the guaranty is executed or the time it is called (whichever is greater). In other countries, such as France, the directors will require that the guaranty be limited based upon the amount of tangible consideration (usually a portion of the loan proceeds) actually received by the guarantor. In Germany, an upstream guaranty given by a GmbH typically will provide that the guarantor is not required to make a payment under the guaranty to the extent the payment would cause the guarantor’s ‘net assets’ to fall below its ‘stated capital’ (measured at the time of enforcement).18 As a result, under current market practice, consolidated borrowing bases are generally used only where the co-borrowers are domiciled in jurisdictions that permit guaranties where the credit facility is deemed to be in the interests of the corporate group taken as a whole (such as Australia, Canada, the Netherlands and 16

USC, § 11–548. Downstream guaranties, under which a company guaranties the obligations of its direct or indirect wholly-owned subsidiary, are generally enforceable in most countries. 18 This provision is based on GmbH-Gesetz, § 30. 17

112

Richard M Kohn

the UK). A German GmbH, on the other hand, is typically not permitted to join in a consolidated borrowing base, because the lender cannot reasonably be assured, at the inception of the credit facility, that a guaranty given by a German GmbH will be enforceable. However, a German entity may nevertheless be permitted to have the benefit of a consolidated borrowing basis. For example, if the corporate group includes a UK company, a Dutch company and a German company, a lender may permit a consolidated borrowing base for the UK and Dutch companies, but may require that the German company have its own borrowing base. However, the German company may be permitted to borrow against the receivables and inventory owned by the UK and Dutch companies, because the lender can be reasonably assured, at the inception of the credit facility, of the enforceability of their guaranties of loans made under the credit facility to their German affiliate. The relevance of restrictions on the enforceability of affiliate guaranties inherent in a country’s laws is by no means limited to consolidated borrowing bases. Affiliate guaranties play a crucial role in credit transactions of all types, providing essential credit support. Although a country’s laws applicable to affiliate guaranties are often deeply rooted in the country’s legal tradition, the case can be made that these laws have not kept pace with the evolution of the modern multi-entity corporate group and its complex financing needs, where all of the entities in the group may function as a single organism. Legislators in countries wishing to promote financing for domestic business entities may wish to examine these laws in order to expressly permit guaranties in situations where financing provided to one or more members of the group benefits the group as a whole.

E. Challenge No 5: Cash Dominion As noted above, an essential component of asset-based lending is the lender’s ability to obtain control over the proceeds of the borrower’s receivables. The lender typically will require that the borrower notify its customers to make all payments of receivables directly to a bank account under the lender’s control (a ‘collection account’). The proceeds are then either: (i) applied automatically to the outstanding balance of the loans, subject to reborrowing by the borrower based on new receivables generated, and new inventory acquired, by the borrower (an approach known as ‘full dominion’); or (ii) applied to the outstanding balance of the loans only upon the occurrence of a triggering event, such as an event of default by the borrower or a drop in borrowing availability under the credit facility below a specified amount (an approach referred to as ‘springing dominion’). Borrowers will generally prefer springing dominion over full dominion because it gives them greater access to the proceeds of receivables remitted by their customers, since they may access these funds directly from the collection account prior to the occurrence of the triggering event, rather than indirectly as proceeds of future loans. In the US, full dominion is often required by asset-based lenders as a means of exercising control over the proceeds of receivables, though it is increasingly

Current Issues in Cross-Border Asset-Based Lending

113

common for lenders to agree to springing dominion, especially in larger assetbased credit facilities where borrowers may have a stronger negotiating position. In some other countries, however, full dominion can be motivated by other reasons. For example, as noted above, in the UK it is generally recognised that full dominion is necessary to convert a floating charge on the proceeds of receivables (which is subject to a carve-out for unsecured creditors of up to £600,000) to a fixed charge (which is not subject to such carve-out).19 This can be a mixed blessing for the borrower: on the one hand, it provides more credit to the borrower by eliminating the possibility that the lender will reserve up to £600,000 from the borrowing availability; on the other hand, it denies the borrower direct access to the funds in the collection account.

F. Challenge No 6: Financial Assistance Laws One of the major challenges that must be addressed when financing acquisitions of companies in many Member States of the EU, and in some other countries with legal regimes based on UK law, is the existence of financial assistance laws. These statutes, adopted by European countries in response to the Second Council Directive of the European Council of Communities (also called the ‘Capital Directive’),20 make it illegal for a domestic company to provide financial assistance, such as a guaranty or a security right in its assets, to support a loan made to finance the acquisition of that company’s own shares. In other words, financial assistance laws prohibit the typical ‘leveraged buyout’ in which the assets of the company whose shares are being acquired (the ‘target company’) are used to provide credit support for a loan made to finance the acquisition of the target company’s shares. Because financial assistance laws make these transactions illegal, great care is taken by borrowers and lenders alike to make certain that the laws are not being violated. Though emanating from the same Directive, financial assistance laws vary greatly from country to country, both in their terms and in the way they have been interpreted. In some countries, for example, financial assistance laws do not apply to certain types of entity. Examples are the German Gesellschaft mit beschränkter Haftung (GmbH) and the French société à responsabilité limitée (Sàrl). In other countries, the financial assistance law incorporates a procedure (often referred to as ‘white-washing’) designed to permit a target company to provide financial assistance in connection with the acquisition of its own shares if it can be 19

This result is derived from the case of Re Spectrum Plus Ltd [2005] UKHL 41. See Art 23 of the Second Council Directive of 13 December 1976 on coordination of safeguards which, for the protection of the interests of members and others, are required by Member States of companies within the meaning of the second paragraph of Art 58 of the Treaty, in respect of the formation of public limited liability companies and the maintenance and alteration of their capital, with a view to making such safeguards equivalent, (77/91/EEC); eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri=CONSLEG:1977L0091:20070101:EN:PDF. 20

114

Richard M Kohn

demonstrated that doing so would not put the company in a precarious financial condition. Depending on the country, this procedure can range from a detailed analysis to a relatively ministerial act. In some countries, financial assistance laws are interpreted to apply only where the shares being acquired are the shares of the target company, while in other countries the laws are interpreted to prohibit the target company from providing financial assistance even when the shares of its direct or indirect parent company are being acquired. In transactions where a financial assistance law presents a potential impediment, much attention is often given to structuring the transaction in a way that does not violate the law. For example, because financial assistance laws apply to share acquisitions as opposed to asset acquisitions, it may be possible to restructure the acquisition as an asset acquisition. However, this approach is often not practical because it could result in the loss of various tax benefits or require thirdparty consents that would not be required in a share acquisition. It may also be possible to effect a post-acquisition merger in which the target company is merged with its parent company (the borrower of the acquisition loan) shortly after the closing, with security rights being granted by the merged entity. In some transactions, it may be possible to assure that the equity funds contributed by the buyer are used to purchase the shares, while the loan proceeds provided by the lender are used for other purposes, such as refinancing the target company’s existing loans or fulfilling the target company’s working capital needs. Notwithstanding the availability of these ‘work-arounds’ in some circumstances, financial assistance laws impede financing for share acquisitions, and thereby inhibit merger and acquisition activity as well. Some jurisdictions, such as the UK, the Netherlands and Singapore, have repealed their financial assistance laws insofar as they apply to private companies. Countries wishing to stimulate merger and acquisition activity may wish to consider following their lead.

G. Challenge No 7: Syndication Not all of the current issues relating to cross-border asset-based lending pertain primarily to borrowers. Some pertain principally, though not exclusively, to lenders. Examples are various issues that arise in the context of credit facilities in which the loans are made by a syndicate of lenders rather than a single lender.

i. Withholding Taxes Many countries impose a withholding tax on payments of interest by a domestic entity to a foreign lender. In many cases, these taxes are reduced (often to zero), either by the provisions of the country’s domestic withholding tax law or by a double tax treaty in force between that country and the country in which the lender is domiciled. Application of withholding taxes in a syndicated credit facility is

Current Issues in Cross-Border Asset-Based Lending

115

determined separately for each lender in the group, and not for the group as a whole. The economic burden of withholding taxes in a credit facility is often imposed on the borrower because the credit agreement will typically require the borrower to increase (ie, ‘gross up’) payments of interest to include withholding taxes on such interest. Thus, the presence of withholding taxes may make it costly to the borrower for particular lenders to participate in the syndication.

ii. Internal Lending Constraints In some instances, US lenders may be inhibited from participating in a syndicated cross-border credit facility either by their internal policies or by agreements with their own financing sources, which may either prohibit the lender from making a direct loan to a non-US entity or make it economically less desirable for the lender to so do. In addition, as a practical administrative matter, the lender may not have ready access to the various foreign currencies required by the borrower under the credit facility. The marketplace has provided ways of addressing this problem. One response in common use today is known as the ‘collateral allocation mechanism’ (or ‘CAM’). A CAM is typically used in credit facilities in which certain loans are made directly to US entities in a corporate group (‘US loans’) by those lenders that are unable to make loans to foreign entities, while other loans are made directly to the foreign entities in the group (‘foreign loans’) by those lenders that are able to do so. Upon the occurrence of a ‘triggering event’ (such as acceleration of the loans), each lender is required to purchase an undivided interest in the loans made by all of the other lenders, with the result that all of the lenders hold undivided interests in all of the US loans and foreign loans in proportion to their respective loan commitments. Thus, each lender in the syndicate bears its proportionate share of the risk of all of the loans made under the credit facility.

iii. Parallel Debt It is a characteristic of syndicated credit facilities that lenders may join or leave the syndicate from time to time. Depending upon the countries in which the collateral is located, this circumstance can present an issue for lenders in a secured cross-border credit facility: how to assure that the security rights in the collateral continue to secure the obligations owing to all of the lenders, even though new lenders may have joined the syndicate and other lenders may have left it. In some jurisdictions, this issue is easily addressed. In the US, for example, the legal relationship between the syndicate of lenders and the agent acting on their behalf is governed by the law of agency, under which security rights in the collateral may be granted to the agent for the benefit of all of the lenders.21 Thus, the security rights benefit all of the lenders that comprise the syndicate at any given 21

See UCC, § 9–502(a)(2).

116

Richard M Kohn

time, and lenders may move in and out of the syndicate without triggering the need for any additional documentation or other action to continue the security rights in effect. In other jurisdictions, such as the UK and jurisdictions with legal regimes based on UK law, the relationship between the syndicate of lenders and the agent is governed by the law of trusts. Here, the security right in the collateral may be granted to the agent in its role as security trustee for the lenders. Yet, the result is the same as in the US: no additional documentation is necessary to continue the security rights in effect each time a lender joins the syndicate. However, the situation is quite different in some civil law countries, such as Germany, Luxembourg and the Netherlands, which recognise neither the law of agency nor the law of trusts in the context of syndicated credit facilities. Instead, the security documents must be amended each time a lender joins the credit facility in order to assure that the security rights remain in effect for the benefit of all of the lenders. In a large syndicated credit facility, in which lenders move in and out with some frequency, this requirement can present a significant and costly administrative burden. Lawyers have responded to this issue with a highly creative solution known as ‘parallel debt’. In addition to the primary loan obligation owing by the borrower to the lenders in the syndicate, a second, ‘parallel’, debt is created, which is owed only by the borrower to the agent. The primary debt and the parallel debt operate in tandem: provisions in the loan documents make it clear that payments on the primary debt also reduce the parallel debt, and vice versa, and that the original debt and the parallel debt, taken together, may never exceed the primary debt. The security rights in the collateral are granted only to secure the parallel debt. Because the parallel debt is only owing to the agent, changes in the composition of the lender syndicate do not require any amendments to the security documents. The concept of parallel debt has become widespread in syndicated credit facilities. Nevertheless, there is no statutory authority, and very little judicial authority, supporting it. The principal judicial authority is the decision of the French Cour de Cassation in the Belvédère case.22 Although this case has been much heralded as the validation of parallel debt, it is less than ideal as authority, given that the French Supreme Court merely enforced a parallel debt provision in a credit agreement governed by New York law on the ground that the provision did not violate French public policy. At the time of this writing, there is little other significant authority supporting the concept of parallel debt.23 As a result, lenders in syndicated cross-border credit facilities currently are exposed to the risk of a court rejecting the concept

22

Arrêt n° 840 du 13 septembre 2011 (10-25.633; 10-25.731; 10-25.908), Cour de Cassation, France. The other case often cited is the decision of the Supreme Court of Poland of 9 October 2009 (file no IV CSK 145/09), which implied that it was possible to use a parallel debt mechanism under Polish law. However, the case has little value as precedent because the parallel debt feature of the credit facility involved in that case was not challenged. 23

Current Issues in Cross-Border Asset-Based Lending

117

of parallel debt and invalidating the security rights for a credit facility. Countries wishing to accommodate the needs of modern, multi-bank credit facilities as a way of stimulating economic growth for domestic businesses may wish to consider modifying their laws to adopt either the agency or security trust concept for syndicated credit facilities, making resort to legal fictions such as parallel debt, and the attendant legal risks, unnecessary.

V. Conclusion Asset-based lending is merely one form of financing used by middle-market companies. Many other forms of financing have evolved, and will to continue to evolve, to fulfil the increasing financing needs, both domestic and cross-border, of modern businesses throughout the world. However, the challenges revealed by asset-based lending, as it has expanded from the US to other countries, are nevertheless instructive as to how the laws of many countries have not kept pace with the financing needs of today’s increasingly globalised business entities. Countries wishing to stimulate the growth of domestic companies may wish to examine their laws through the lens of asset-based lending and other prevalent financing techniques, and consider the adoption of laws that promote such financing. Modern secured transactions laws are, of course, at the very core of a legal regime that promotes credit, but other laws, such as corporate governance laws, financial assistance laws and insolvency laws, also play a vital role in creating a legal environment in which low-cost financing, and the economic benefits that can flow from it, may flourish.

118

 

8 The Rights and Obligations of the Parties to the Security Agreement According to the UNCITRAL Model Law on Secured Transactions BÉNÉDICT FOËX*

I. Introduction The chrysalis has turned into a butterfly: the draft UNCITRAL Model Law on Secured Transactions1 adopted by the UNCITRAL Working Group VI (Security Interests),2 was reviewed by the United Nations Commission on International Trade Law at its 49th Session, held in New York in June–July 2016:3 analysed, discussed, reworked, improved and polished, the draft Model Law became the UNCITRAL Model Law on Secured Transactions.4 In this context, we thought it appropriate to devote the following lines to the provisions of the UNCITRAL Model Law governing the rights and obligations of the parties to the security agreement. Firstly, because of course it is one of the many themes that Professor Roderick Macdonald was interested in, but also because it is sometimes a neglected aspect of the security rights in movables, straddling contract law and property law. *

Professor at the University of Geneva. draft Model Law; Documents A/CN.9/884, A/CN.9/884/Add.1, Add.2, Add.3 and Add. 4, available on the UNCITRAL website at www.uncitral.org/uncitral/en/commission/sessions/ 49th.html. 2 For a summary of the sessions of Working Group VI on the draft Model Law, see www.uncitral. org/uncitral/en/commission/working_groups/6Security_Interests.html. 3 At its 48th session (29 June to 16 July 2015), however, UNCITRAL approved the substance of the provisions concerning the Registry (Art 26 of the draft Model Law and Arts 1–29 of the draft Model Registry Provisions); see Report of the United Nations Commission on International Trade Law, Fortyeighth session (29 June–16 July 2015), Vienna 2015, p 30–40 (this report is available on the UNCITRAL website at www.uncitral.org/uncitral/en/commission/sessions/48th.html). 4 Hereinafter: the UNCITRAL Model Law, the text of which is available on the UNCITRAL website at www.uncitral.org/uncitral/en/uncitral_texts/security/2016Model_secured.html. 1 Hereafter:

120

Bénédict Foëx

II. General Remarks The UNCITRAL Model Law devotes no fewer than nine articles to the rights and obligations of the parties to the security agreement. The first five are referred to as ‘General rules’ (Articles 52–56), while the other four contain ‘Asset-specific rules’ (Articles 57–60).5 These rules are supplemented by the general provisions underlying the whole UNCITRAL Model Law (Articles 1–5) and, in particular, by Article 3 (which enshrines the principle of ‘party autonomy’) and by Article 4 (which lays down ‘general standards of conduct’ imposed on ‘a person’ (Article 4, paragraph 1), which includes in particular the parties to the security agreement). Finally, of course, one must take into account Article 6 of the UNCITRAL Model Law, which is devoted to the security agreement, as well as the definitions enshrined in Article 2. This contribution will focus on the general rules governing the rights and obligations of the parties to the security agreement (Articles 52–56).

III. The Parties Concerned The UNCITRAL Model Law does not define the notion of ‘party’ (see Article 2). One may, therefore, wonder who are the ‘parties’ to the security agreement, referred to in three of the four headings preceding Article 52.6 At first glance, these are the contracting parties who enter into a security agreement, that is to say, the person who constitutes the security right (and who is generally also the debtor of the claim to be secured) and the creditor in favour of whom the security right is being created. This impression can be supported by Article 52, which refers to ‘the mutual rights and obligations of the grantor and the secured creditor’, and by Articles 53, 54, 55 and 56, which also mention the grantor and the secured creditor. However, account must be taken of the fact that the concepts of ‘grantor’ and ‘secured creditor’ are defined by the UNCITRAL Model Law: 1. the term ‘grantor’ means ‘(i) A person that creates a security right to secure either his own obligation or that of another person; (ii) A buyer or other transferee of an encumbered asset that acquires its rights subject to a security right; and (iii) A transferor under an outright transfer of a receivable by agreement’ (Article 2, paragraph 1 o); 5

See titles preceding Arts 52 and 57. VI: Rights and obligations of parties and third-party obligors’; ‘Section I. Mutual rights and obligations of the parties to a security agreement’; ‘Article 52. Sources of mutual rights and obligations of the parties’. 6 ‘Chapter

The Rights and Obligations of the Parties to the Security Agreement 121 2. the term ‘secured creditor’ means ‘(i) A person that has a security right; and (ii) A transferee under an outright transfer of a receivable by agreement’ (Article 2, paragraph 1 ff). Thus, in the system of the UNCITRAL Model Law, the ‘grantor’ is not only a person creating a security right; it also designates the buyer or beneficiary of another form of transfer of an asset encumbered by a security right, or a person that proceeds as transferor to an outright transfer of receivable by assignment. As to the ‘secured creditor’, it may be not only the person in favour of whom the security right is created (or transferred), but also the person who acquires as transferee a receivable under an outright transfer by agreement. It follows from the foregoing that Articles 52–56 of the UNCITRAL Model Law deal with rights and obligations that concern not only the grantors and the secured creditors in the strict sense, but also the subsequent transferees of the encumbered asset and the transferor and transferee under an outright transfer of a receivable by agreement. This being said, on closer inspection, Articles 53, 54 and 55 cannot be applied in the event of an outright transfer of receivable by agreement. These provisions presuppose that the secured creditor (respectively the secured creditor or the grantor, in the case of Articles 53 and 55, paragraph 2) is in possession of the encumbered asset. Yet, according to Article 2, paragraph z of the UNCITRAL Model Law,7 ‘possession’ means ‘the actual possession of a tangible assets’; and ‘tangible asset’ is defined as ‘any tangible movable asset’ (Article 2, paragraph ll of the UNCITRAL Model Law), which includes ‘money, negotiable instruments, negotiable documents and certificated non-intermediated securities’8 but excludes receivables.9 Articles 53–55 cannot therefore apply to the transferor and the transferee under an outright transfer of a receivable by agreement. As to Article 52, it can hardly, in our view, be applicable to the transferee of an encumbered asset. Indeed, this provision refers to the ‘mutual’ rights and obligations of the parties: it therefore does not seem to govern the rights and obligations of third parties which are not bound by the security agreement, such as the transferee of the encumbered asset. Moreover, the definition of ‘security agreement’10 also suggests that the parties are the grantor and the secured creditor stricto sensu (as well as the transferor and transferee under an outright transfer of

7 The term ‘“possession” means the actual possession of a tangible asset by a person or its representative, or by an independent person that acknowledges holding it for that person’. 8 Art 2 para ll, 2nd sentence. 9 Art 2 para dd of the UNCITRAL Model Law defines ‘receivable’ as ‘the right to payment of a monetary obligation, excluding a right to payment evidenced by a negotiable instrument’. 10 Art 2, para jj of the UNCITRAL Model Law: ‘the term “security agreement” means: (i) An agreement, regardless of whether the parties have denominated it as a security agreement, between a grantor and a secured creditor that provides for the creation of a security right; and (ii) An agreement that provides for the outright transfer of a receivable’.

122

Bénédict Foëx

a receivable): one fails to see how the security agreement could give rise to rights and obligations in favour of, and to be borne by, a third party who is not a party thereto. Finally, Article 56 seems to refer only to the grantor and the secured creditor stricto sensu: the first paragraph of this provision expressly indicates that it is not applicable to the transferee under an outright transfer of a receivable; logically, it cannot be applicable to the transferor under an outright transfer of a receivable either. And it is hard to conceive how this provision could entitle the buyer or other transferee of the encumbered asset to request from the secured creditor information about the secured obligation and the assets currently encumbered. To sum up, Article 52 of the UNCITRAL Model Law applies to the ‘parties’ to the security agreement, namely, the person creating the security right and the person in whose favour the security right is created (as well as the transferor and the transferee under an outright transfer of receivable by agreement). Articles 53, 54 and 55 apply to the grantor and the secured creditor stricto sensu (but not to the transferor and the transferee under an outright transfer of receivable), and to the subsequent transferee of the encumbered asset (that acquires its right subject to a security right). As to Article 56, it seems to us that it is intended to apply to the grantor and the secured creditor stricto sensu only. We cannot help but point out that the above illustrates the disadvantages of polysemy, especially when it comes to such fundamental notions: the reading of the UNCITRAL Model Law and its comprehension are not facilitated by the fact that the word ‘grantor’ has (at least) three different meanings and that the locution ‘secured creditor’ means not only ‘a person that has a security right’ but also the transferee under an outright transfer of receivables.

IV. The Sources of the Reciprocal Rights and Obligations of the Parties (Article 52) Article 52 of the UNCITRAL Model Law is devoted to the sources of the mutual rights and obligations of the parties, which it lists. It is based on Recommendation 110 of the Legislative Guide on Secured Transactions.11 While relying on a slightly different wording, it takes up the substance of Recommendation 110, mentioning, in particular, as sources the ‘terms and conditions’ of the security agreement, the ‘rules or general conditions’ referred to in that contract, the ‘usage to which’ the parties ‘have agreed’ and ‘any practices they have established between themselves’. In an earlier version, Article 52 added two sources that are not mentioned in Recommendation 110 of the Legislative Guide: ‘the provisions of this chapter’ 11 The Legislative Guide can be consulted at www.uncitral.org/uncitral/en/uncitral_texts/security/ Guide_securedtrans.html; Recommendation 110 is found at the end of ch VI, at p 257 of the English version of the Legislative Guide.

The Rights and Obligations of the Parties to the Security Agreement 123 (sub-paragraph [a]) and ‘the provisions of other law relating to the mutual rights and obligations of the parties to an agreement’ (subparagraph [b]).12 These were not happy additions. Indeed, they seemed to state the obvious.13 In addition, the wording of subparagraph (b) was not necessarily appropriate, as it seemed to assume that the legislation of the State adopting the UNCITRAL Model Law (ie the enacting State) would necessarily contain general provisions ‘relating to the mutual rights and obligations of the parties to an agreement’. It was therefore without prejudice that subparagraphs (a) and (b) of Article 47 of the draft Model Law (now Article 52) were discarded, to return to the simplicity of Recommendation 110. This also has the advantage of highlighting what constitutes the primary source of the parties’ rights and obligations (in accordance with the principle of party autonomy referred to in Article 3 of the UNCITRAL Model Law): the will of the parties, as expressed in their agreement.

V. Obligation to Preserve the Encumbered Asset (Article 53) Under Article 53 of the UNCITRAL Model Law, ‘a grantor or secured creditor in possession of an encumbered asset must exercise reasonable care to preserve the asset’. This provision essentially14 covers the content of Recommendation 111 of the Legislative Guide.15 It imposes a preservation obligation on the party in possession of the encumbered asset. The conditions for the application of Article 53 are the following: (i) the encumbered asset must be in the actual possession of the debtor of the obligation (resulting from Article 53), or of the representative of that

12 See Art 47 of the draft Model Law, in the summer 2015 version in document A/CN.9/WG.VI/ WP65/Add.3, available at www.uncitral.org/uncitral/en/commission/working_groups/6Security_ Interests.html. 13 See also Note to the Working Group VI, ad Art 47 (in A/CN.9/WG.VI/WP.65/Add.3, available at www.uncitral.org/uncitral/en/commission/working_groups/6Security_Interests.html): ‘Subparagraphs (a) and (b) […] may […] be self-evident’. 14 Recommendation 111 further provides that the party in possession of the property must preserve the value of the asset. This requirement, which was also raised by Art 48 of the draft Model Law (see document A/CN.9/WG.VI/WP65/Add.3, available at www.uncitral.org/uncitral/en/commission/working_groups/6Security_Interests.html) has fortunately not been included in the UNCITRAL Model Law: depending on the circumstances, it might for instance have resulted in the duty to insure the asset, to institute proceedings to settle disputes concerning the encumbered asset (such as the questioning of the authenticity of the encumbered painting by a third party, for example), or even to move it abroad to preserve it from falling market prices in the country in which it is located, or to take into account the impact on the value of the encumbered securities of the exercise of the voting right inherent in these securities at the shareholders’ meeting of the company concerned. 15 Legislative Guide (n 11), end of ch VI, p 257.

124

Bénédict Foëx

debtor or of an independent person that acknowledges holding it for the debtor;16 (ii) the encumbered asset must be a tangible movable asset: it has been noted above that it follows from the definition of ‘possession’ that Article 53 applies only if the encumbered asset is a tangible asset;17 (iii) the person exercising the possession (or on whose account the possession is exercised) must be a grantor or a secured creditor stricto sensu, or a transferee of an encumbered asset that acquires its right subject to a security right. When these conditions are met, an obligation arises at the expense of the party in possession of the encumbered asset: this party must exercise ‘reasonable care’ in order to preserve the asset. It should be noted in passing that the Model interamerican law on secured transactions of the Organization of American States18 contains in its Article 3319 a solution close to Article 53 of the UNCITRAL Model Law. The extent of this obligation will vary depending on the nature of the asset in question: the preservation of an ingot of precious metal or of an emerald requires less care than that of a racehorse, of a collection of cigars or of an Egyptian papyrus. It is ultimately the duty to perform its obligations ‘in good faith and in a commercially reasonable manner’ (Article 4 of the UNCITRAL Model Law) that will determine the limits of the obligation to preserve in a particular case. In any event, this obligation imposes, in the first place, on its debtor the duty to keep the encumbered asset in an appropriate and safe place (temperature, ventilation, humidity, etc), in order to prevent that the asset be damaged, lost or stolen, for example. This obligation then imposes on the debtor the duty to maintain the encumbered asset to the extent necessary to keep its condition. Depending on the encumbered asset, further care will still need to be provided: for example, the animal encumbered by the security right must be adequately fed and kept in good health (exercise, milking, veterinary care, etc),20 the engine which is liable to seize must be started regularly, the ancient musical instrument must be removed from its case to be ‘played’ by a qualified musician, etc. As can be seen, the outline of the obligation imposed by Article 53 is not easy to pin down. The fact that only ‘reasonable care’ is required in this regard and that in any event this obligation must be performed ‘in good faith and in a commercially reasonable manner’ under Article 4 of the UNCITRAL Model Law is likely to limit

16

See definition of possession at Art 2 z of the UNCITRAL Model Law (n 7). See text to n 7 above, and following. For the text of the OAS Model Law of 8 February 2002, see www.oas.org/en/sla/dil/docs/secured_ transactions_BOOK_Model_Law.pdf. 19 Art 33 of the OAS Model Law: ‘A creditor in possession of the collateral: I. Shall exercise reasonable care in the custody and preservation of the collateral. Unless otherwise agreed, reasonable care implies the obligation to take the necessary steps to preserve the value of the collateral and the rights derived therefrom. II. […]’. 20 Legislative Guide (n 11), ch VI, para 28, p 243. 17 18

The Rights and Obligations of the Parties to the Security Agreement 125 the scope of the obligation in question somewhat, but still leaves considerable uncertainty. This being said, the UNCITRAL Model Law makes two main methods for achieving third-party effectiveness available to the parties: possession of the (tangible) encumbered asset by the secured creditor and registration of a notice in the Registry established in accordance with the Model Registry Provisions of the UNCITRAL Model Law (Article 18, paragraphs 2 and 1 of the UNCITRAL Model Law). The parties can therefore determine which one of them will ultimately have possession of the encumbered asset. As a result, they can determine which party has the obligation to preserve the asset, which also allows them to clarify the scope of this obligation in the particular case. It should be borne in mind, however, that the use of one method of third-party effectiveness rather than the other can be guided by considerations other than who should bear the obligation to preserve the asset. Thus, for example, the secured creditor may wish to be immediately in possession of the encumbered asset in order to avoid the uncertainties and difficulties associated with the exercise of the right to obtain possession of it in the event of the debtor defaulting (Article 77 of the UNCITRAL Model Law); the creditor will then find himself, nolens volens, with the burden of preserving the encumbered asset and the associated uncertainties that come from this. It therefore seems desirable for the parties to be able to specify the outline of the obligation under Article 53 of the UNCITRAL Model Law, or even to exclude any duty of the party in possession in this regard. However, the UNCITRAL Model Law provides that Article 53 constitutes a mandatory provision (Article 3, paragraph 1). It thus conforms to the Legislative Guide, whose Recommendation 111 (on which Article 53 is based, as we have seen)21 was placed under the heading ‘Mandatory rules’, along with Recommendation 112. The Legislative Guide explains in this respect that: ‘the basic rule in most legal systems is that the parties may not, by agreement, derogate from mandatory rules setting out the general pre-default rights and obligations. For example, States typically do not permit parties to contract out of their duty to take reasonable care of the encumbered assets. […] The mandatory pre-default rules recommended in the Guide aim at policy objectives consistent with what have been defined as the core principles of an effective and efficient regime of secured transactions (see recommendation 1). They set out pre-default rights and obligations that (a) encourage parties in possession to preserve the encumbered asset; and (b) ensure that once the secured obligation is paid or otherwise performed, the grantor recovers the full use and enjoyment of the previously encumbered assets’.22

This motivation full of wisdom and moderation does not completely remove all the problematic areas. Are the aims just mentioned so fundamental that it is 21 22

Above (n 15). Legislative Guide (n 11), ch VI, p 265, paras 21 and 22.

126

Bénédict Foëx

necessary to demand that the parties pursue them in all circumstances? Should the grantor truly be protected to this extent on this specific issue, even though it is—for instance—at the same time given the freedom to create a security right in ‘all’ its ‘movable assets’?23 Thus, by way of example, should a grantor who is hospitalised for a long period of time not be allowed to create a security right in the movables in his house or in his living room or in his attic, ‘without warranty, and maintenance-free given the circumstances’? And should not the creditor be free to content himself with such a security right? It seems ultimately that there is no reason for imposing on the parties the requirements provided for in Article 53 of the UNCITRAL Model Law. They should be free to deviate from it by defining the scope of the obligation to preserve the encumbered asset as they wish, or by releasing the relevant party from any obligation of that kind at the outset.24 Such a conclusion is all the more necessary since the UNCITRAL Model Law does not in any way specify the sanction for failure by the party concerned to comply with the obligation enshrined in Article 53. The question is therefore governed by other legal provisions of the enacting State, which may, as the case may be, enable the parties to release the debtor in advance of the liability incurred in the event of a breach of that obligation.

VI. Obligation of the Secured Creditor to Return the Encumbered Asset (Article 54) Article 54 of the UNCITRAL Model Law reads as follows: ‘Upon extinguishment of a security right in an encumbered asset, a secured creditor in possession must return the asset to the grantor or deliver it to the person designated by the grantor’. This provision essentially corresponds to the content of Recommendation 112 of the Legislative Guide.25 The ratio of this rule is easily explained: it is ‘to make certain that the grantor recovers the full use and enjoyment of the previously encumbered assets and is able to deal effectively with them in transactions involving third parties, free of any disability arising from the no-longer-extant security right’.26 Indeed, it could be argued that this obligation of restitution is self-evident: when its security right is extinguished, the secured creditor loses its right encumbering the asset and cannot therefore oppose it to the grantor; logically therefore, 23

Art 8, para d of the UNCITRAL Model Law. See in this regard the second sentence of Art 33 of the OAS Model Law (n 19), which allows the parties to define the reasonable care required on the part of the secured creditor: ‘Unless otherwise agreed, reasonable care implies the obligation to take the necessary steps to preserve the value of the collateral and the rights derived therefrom’. 25 Legislative Guide (n 11), p 257. 26 Legislative Guide (n 11), ch VI, para 39, p 246. 24

The Rights and Obligations of the Parties to the Security Agreement 127 the secured creditor must return the encumbered asset if it is in possession of it and it must file a cancellation notice if the security right has been made effective against third parties by the registration of a notice in the Registry in accordance with Article 18, paragraph 1 of the UNCITRAL Model Law. It seems preferable to enunciate this obligation clearly rather than leaving the legal text silent on the subject, which might suggest that the grantor has to reclaim the encumbered asset, the creditor having no obligation to spontaneously return it.27 The beneficiary thus has a legal claim for restitution, in addition to that which arises, for instance and as the case may be, from the security agreement. It will also be observed that, according to Article 12 of the UNCITRAL Model Law, ‘a security right is extinguished when all secured obligations have been discharged and there are no outstanding commitments to extend credit secured by the security right’. The extinction of a security right (within the meaning of Article 54) therefore implies the full payment of the secured obligation (or the discharge thereof in another way); it follows that the secured creditor has no obligation (unless otherwise agreed) to return all or part of the encumbered assets until it has been fully paid: the UNCITRAL Model Law enshrines the principle of the indivisibility of the security right: pignoris causa indivisa est. That being said, it can be observed that, depending on the circumstances, the ‘general standard of conduct’ enshrined in Article 4 of the UNCITRAL Model Law could lead to a different result: one should be able to admit that in the event of gross disproportion between the amount of the secured obligation and the value of the encumbered asset, the duty to exercise its rights ‘in good faith and in a commercially reasonable manner’ (Article 4) may require the secured creditor to partially release the encumbered assets (or to accept a substitution of encumbered assets). For the rest, one can wonder why Article 12 of the UNCITRAL Model Law only mentions the discharge of the secured obligations as a cause of extinction of the security right: there are many other causes of extinction that may28 or may not29 trigger the restitution obligation governed by Article 54. Finally, it is to be welcomed that—contrary to the draft Model Law30—Article 54 does not provide as its only solution the return to the grantor, and allows the asset to be returned ‘to the person designated by the grantor’. Indeed, if it can be

27

See the developments in the Legislative Guide, ch VI, para 37, p 245. Eg conventional extinction of the security right, expiration of the duration for which the security right has been established, etc. 29 Eg sale of the encumbered asset free of the security right authorised by the secured creditor (Art 34, para 2 of the UNCITRAL Model Law). Recommendation 80, para a of the Legislative Guide (p 231) made it clearer that the security right would be extinguished in such a case: ‘A security right does not continue in an encumbered asset that the grantor sells or otherwise disposes of, if the secured creditor authorizes the sale or other disposition free of the security right’. 30 Art 52 of the draft Model Law (document A/CN.9 /884/Add. 3 (n 1)) provided as a mandatory rule that ‘upon extinction of a security right in an encumbered asset, a secured creditor in possession must return the asset to the grantor’. 28

128

Bénédict Foëx

regretted that Article 54 is a mandatory provision (see Article 3, paragraph 1 of the UNCITRAL Model Law), it must at least be conceded that the solution thus found makes it possible to take account of the cases in which the encumbered asset must have a fate other than being returned to the grantor (eg disencumbered asset having to be returned to the third-party owner of the asset, who had authorised the granting of the security right by the grantor (see Article 6, paragraph 1 of the UNCITRAL Model Law) provided that the asset be returned to it when the security right is extinguished; disencumbered asset to be kept henceforth in custody by the creditor on behalf of the grantor; disencumbered asset to be delivered to a third party for the purpose of securing a receivable of the latter towards the grantor, etc).

VII. The Secured Creditor’s Right to Use and Inspect the Encumbered Asset (Article 55) Article 55, paragraph 1 of the UNCITRAL Model Law provides that the secured creditor in possession of the encumbered asset is entitled to ‘make reasonable use of the asset’ and to ‘apply the revenues it generates to the payment of the secured obligation’ (subparagraph b); in addition, the secured creditor has the right ‘to be reimbursed for reasonable expenses it incurs’ in fulfilling its obligation to preserve the asset under Article 53 of the UNCITRAL Model Law (Article 55, paragraph 1, subparagraph a). Moreover, Article 55, paragraph 2 establishes in favour of the secured creditor not in possession a right to inspect the encumbered asset in the possession of the grantor. This Article 55 essentially incorporates Recommendation 113 of the Legislative Guide. It proposes non-mandatory rules,31 which Recommendation 113 expressly states.32 The first of the three rules enacted in Article 55 does not require much comment: it seems logical that the secured creditor is entitled to reimbursement of the ‘reasonable expenses’ it incurs in fulfilling its preservation obligation under Article 53 (Article 55, paragraph 1, subparagraph a). At most, it may be noted that the UNCITRAL Model Law could have provided in addition that the claim for reimbursement of these costs is secured by the encumbered asset. Given the silence of the legal text, it will be for the parties to insert in their security agreement a clause to that effect.33

31

See Art 3, para 1 of the UNCITRAL Model Law. the title preceding Recommendation 113: ‘Non-mandatory rules’; see also the text of Recommendation 113: ‘The law should provide that, unless otherwise agreed, the secured creditor is entitled …’ (ch VI, p 257). 33 See Art 6, para 3, sub-para b (the security agreement must describe the ‘secured obligation’) and Art 7 (‘A security right may secure one or more obligations of any type, present or future, determined or determinable, conditional or unconditional, fixed or fluctuating’) of the UNCITRAL Model Law. 32 See

The Rights and Obligations of the Parties to the Security Agreement 129 The second subparagraph of Article 55, paragraph 1 allows the secured creditor to ‘make reasonable use of the asset’ in its possession. This rule may a priori be surprising: it is not in principle the nature of the security right to provide the secured creditor with the right to use the encumbered asset. In this respect, it may be recalled that the UNCITRAL Model Law defines the security right as ‘a property right in a movable asset that is created by an agreement to secure payment or other performance of an obligation […]’ (Article 2, paragraph kk). In other words, a security right is above all a guarantee right,34 an enforcement right, and not a right of use; as the Legislative Guide notes, ‘at the heart of a secured transactions regime is the right of the secured creditor to look to the amount that can be realized upon the sale of the encumbered assets to satisfy the secured obligation’.35 In these circumstances, it would probably have been more logical to provide for the reverse of the rule in Article 55, paragraph 1, subparagraph b, stating that the secured creditor may not, unless otherwise agreed, make use of the encumbered asset. Incidentally, it will be recalled that this is the solution provided for by the OAS Model Law,36 whose Article 33 provides that ‘a creditor in possession of the collateral: […] III. May use the collateral only as provided in the security contract’. Be that as it may, the choice being made, it seems logical that Article 55, paragraph 1, subparagraph b specifies that the secured creditor in possession of the encumbered asset has the right ‘to apply the revenues’ that this asset generates ‘to the payment of the secured obligation’. The second paragraph of Article 55 reserves the right of the secured creditor not in possession to inspect the encumbered asset ‘in the possession of the grantor’. This rule is in a way the equivalent of the fact that when the grantor remains in possession of the encumbered asset, it has the obligation to preserve the asset (Article 53). That being so, one can imagine the difficulties of practical application that the exercise of such an inspection right is likely to pose. And it is precisely where the grantor does not regularly fulfil his obligation to preserve the asset that it may attempt to evade inspection, alleging that it is temporarily impossible or prejudicial to his interests, that it is occurring at an inopportune moment, etc. Prudent parties will, therefore, be careful to specify the terms of this right of inspection in the security agreement. They have the leisure to do so, since, as we have seen, Article 55 is a non-mandatory provision.37 In the absence of such stipulations, the terms and conditions of the inspection shall be determined according to the general standards of conduct of Article 4: the secured creditor and the grantor shall act ‘in good faith and in a commercially 34 See also Legislative Guide (n 11), ch II, p 65, para 1: ‘A security right under the Guide is a property right (as opposed to a personal right) in movable assets (as opposed to immovable property) created by agreement (as opposed to statutory or judgement rights) between the grantor and the secured creditor […]. This property right is meant to secure the performance of an obligation owed by the grantor or another person to that creditor’. 35 Ch VIII, para 6, p 276. 36 See n 19. 37 See n 31.

130

Bénédict Foëx

reasonable manner’. In particular, the more pressing the circumstances, the more the grantor will need to comply diligently and promptly with the request for inspection: as noted in the Draft Guide to Enactment of the draft Model Law on Secured Transactions,38 ‘in extreme cases, such as where the debtor is in default or the secured creditor has reason to believe that the physical condition of the collateral is in jeopardy or has been, or is about to be, removed from the jurisdiction, the secured creditor may be justified in demanding an immediate inspection’.39 That being so, an earlier version of Article 55, paragraph 2 provided that the right to inspect may be exercised not only in respect of the grantor but also generally in relation to ‘another person’, provided that the latter is in possession of the encumbered asset.40 It may be asked whether it would have been possible to impose a duty on any person in possession of the encumbered asset. In this regard, it should be borne in mind that the security right established by the UNCITRAL Model Law is a ‘property right’.41 The UNCITRAL Model Law does not define this notion. According to the Legislative Guide, however, ‘references to ‘property right’ are to be interpreted as referring to a right in an asset (right in rem) as opposed to a personal right (right ad personam)’.42 It is probably accurate to add that the security right, as a property right, is enforceable against third parties;43 if so, the right to inspect could be considered as one of the attributes of the property right and as constituting a corollary of the well-known maxim ubi rem meam invenio, ibi vindico. In other words, it could be considered that the secured creditor derives from its security right—insofar as it has been made effective against third parties44—an erga omnes claim allowing him to require from any party in possession to let it carry out the inspection of the encumbered asset. In any event, the current text of Article 55 no longer provides that the secured creditor is entitled to inspect the encumbered asset ‘in the possession of […] another person’. We can rejoice in this: it is not certain that this ‘appendix’ would have solved more problems than it would have caused.

38 Document UNCITRAL A/CN.9/WG.VI/WP.66 / Add.3, available on the UNCITRAL website at // documents-dds-ny.un.org/doc/UNDOC/LTD/V15/055/67/PDF/V1505567.pdf?OpenElement. 39 Draft Guide to Incorporation (n 38), p 5, para 15 (referring to Art 50 of the draft Model Law, now Art 55 of the UNCITRAL Model Law). 40 See Art 50, para 2 of the draft Model Law in its summer 2015 version, at www.uncitral.org/uncitral/en/commission/working_groups/6Security_Interests.html. 41 See in this respect the definition of security right: ‘A property right in a movable asset […]’ (Art 2, para ii of the UNCITRAL Model Law). 42 Legislative Guide (n 11), Introduction, No 17, p 5 and ch II, para 1, p 65. 43 See, for example, the Legislative Guide (n 11), ch III, para 7, p 104: ‘once the conditions for creation of a security right […] are satisfied, the security right becomes effective as between the grantor and the secured creditor (that is, inter partes; see recommendation 30). However, in order for the security right to affect the rights of third parties (erga omnes), the requirements for third-party effectiveness set out in this chapter must also be satisfied (see recommendation 29)’. 44 See Art 18 ff of the UNCITRAL Model Law; see also n 43.

The Rights and Obligations of the Parties to the Security Agreement 131

VIII. Right of the Grantor to Obtain Information (Article 56) The fifth and final general rule relating to the rights and obligations of the parties to a security agreement provides the grantor with the right to obtain, on written request to the secured party, (a) ‘a statement of the obligation currently secured’ and (b) ‘a description of the assets currently encumbered’ (Article 56, paragraph 1 of the UNCITRAL Model Law). It can be understood that the need for such information is felt: thus, by way of illustration, it is noted that the secured obligations and the encumbered assets can be defined in a particularly wide way by the parties in the security agreement,45 so that the grantor (and, perhaps, the secured creditor as well) may find it difficult to ascertain which assets are encumbered, as security for what obligations. However, the rule found in Article 56 seems self-evident: does it not already follow from the duty to behave ‘in good faith and in a commercially reasonable manner’ anchored in Article 4 of the UNCITRAL Model Law that the creditor must provide such information to the grantor if so required? For the rest, it is regrettable that the UNCITRAL Model Law goes one step further by resolving the issue of the costs of providing the requested information (Article 56, paragraphs 2 and 3). This is an issue which might equally well be settled by the parties themselves. Since these rules are not mandatory (Article 3 paragraph 1 of the UNCITRAL Model Law), the parties retain in any case the right to derogate from them, which is to be welcomed.

IX. Conclusion Having arrived at the end of this short contribution, it can be observed that Articles 52–56 of the UNCITRAL Model Law have been carefully drafted and are ready to pass the test of acquiring the force of law and being applied in an enacting State. The author regrets, however, that the UNCITRAL Model Law confers the character of a mandatory norms to Articles 53 and 54. Indeed, these are not provisions that are so fundamental in their values that they must be applied in all circumstances. The solutions they propose are harmonious, but it should remain open to the parties to prefer other ones.

45 See Art 7 (‘A security right may secure one or more obligations of any type, present or future, determined or determinable, conditional or unconditional, fixed or fluctuating’) and Art 8 (‘A security right may encumber: […] [d] All of a grantor’s movable assets’) of the UNCITRAL Model Law.

132

Bénédict Foëx

Most importantly, it is unfortunate that, as elsewhere in the UNCITRAL Model Law, the clarity of the rules laid down is somewhat obscured by the broad definition of the terms ‘grantor’ and ‘secured creditor’. It follows that the ‘grantor’ is not necessarily the person creating the security right, that the secured creditor is not necessarily the holder of the obligation secured by the security right and that the ‘mutual rights and obligations of the parties to a security agreement’ announced by Section I of Chapter VI of the UNCITRAL Model Law may bind persons who are not parties to such an agreement. The enacting State whose tradition is to draw up laws that are not intended solely for specialists may prefer to choose a less ambitious option, making it easier to read and understand the provisions concerned. By using these polysemic definitions, the UNCITRAL Model Law aims to ‘cast a wide net’; in doing so it runs the risk of ‘over-egging the pudding’. This seems to be the place to remember the wise words of Professor Roderick Macdonald: ‘In law reform, as in life, the perfect is the enemy of the good, and the quest for the perfect often undermines the good already accomplished’.46

46 R Macdonald, ‘A Model Law on Secured Transactions. A Representation of Structure? An Object of Idealized Imitation? A Type, Template or Design?’ (2010) 15(2) Uniform Law Review 419, 446.

9 Sûretés Mobilières et Stocks : ou l’Art et la Manière de Résoudre la Quadrature du Cercle JEAN-FRANCOIS RIFFARD* At first glance, the constitution of a security right on inventory represents a challenge. It seems there is an irreducible incompatibility between, on the one hand, the vital need for the grantor to freely dispose of its inventory in the ordinary course of its business, and, on the other hand, the protection of the creditor who must be sure to take and dispose of the encumbered assets in the event of default. How to reconcile these two contradictory aspirations? How to ensure this protection, while traditional tools such as the dispossession of the grantor or the ‘droit de suite’ are obviously inadequate in the case of security rights on inventory? Even though it is attempting the impossible, it is necessary for obvious economic reasons to answer this question. The most recent secured transactions reforms show that this challenge can be met. By enacting very few specific rules and making room for party autonomy, it is possible to take into account the specificities of the inventory within the framework of a unitary and integrated secured transactions regime as advocated by the UNCITRAL Legislative Guide and other international instruments. The purpose of this contribution is to analyse not only the content of these specific principles, such as the extension of the security rights on any asset received in exchange of the sold or manufactured inventory, but also their legal treatment as part of a reform. In other words, to find a way to square the circle … Le droit des sûretés est riche. Très riche. Cette richesse, notre collègue Roderick Macdonald avait parfaitement su en prendre la mesure, lui qui tout au long de sa carrière n’a eu de cesse d’en analyser et d’en explorer les multiples facettes. Dépassant l’apparente technicité – d’aucuns diraient aridité – de cette matière, il a su en tirer la substantifique moelle et mettre en lumière ses enjeux économiques, politiques voire philosophiques. Il appréciait la capacité du droit des sûretés en général, et celui des sûretés mobilières en particulier à se renouveler en faisant apparaître de nouvelles problématiques, sources de discussions passionnées et *

Professeur de droit privé, Ecole de Droit, Université d’Auvergne.

134

Jean-Francois Riffard

passionnantes. Nos débats et échanges dans le cadre des travaux de la Commission des Nations Unies pour le Droit Commercial International (CNUDCI) sur les opérations garanties en fournissent le meilleur exemple. Combien de questions que l’on pensait tranchées n’ont-t-elles pas, aux détours de quelques discussions informelles, rebondi en séance ? Combien de fois un point que l’on pensait a priori anodin ne s’est-il pas avéré dissimuler une question de fond des plus intéressantes ? Parmi ces points, il en est un qui, encore aujourd’hui constitue un terrain d’exploration négligé malgré ses implications pratiques et théoriques : les sûretés mobilières sur stocks. Fondamentalement, toute sûreté réelle va tirer sa force de la conjonction de ces deux facteurs essentiels que sont la valeur du bien grevé et le caractère prioritaire de l’affectation de cette valeur à la garantie de la créance. Ainsi, le caractère attractif d’une sûreté va dépendre du degré de certitude qu’aura le créancier de pouvoir, en cas de défaillance du débiteur, appréhender aisément et rapidement le bien afin de faire apparaître la valeur sur laquelle il exercera son droit de préférence. Partant, il n’est guère étonnant que l’hypothèque immobilière ait été pendant très longtemps considérée comme la Reine des sûretés réelles conventionnelles. Son assiette est fixe, immobile, et sa valeur ne risque guère de diminuer tout du moins sur le court et moyen terme. Les sûretés mobilières posent, en revanche, un défi. Le caractère mobile des biens expose le créancier au risque de voir le bien grevé être distrait, a fortiori s’il est resté en la possession du constituant. Les biens meubles, notamment corporels, sont par ailleurs souvent soumis à une dépréciation rapide. Ce défi est encore plus grand lorsqu’est envisagée la constitution d’une sûreté mobilière sur des actifs circulants 1 et plus spécifiquement sur des stocks. Ces stocks, qu’ils soient constitués de marchandises, de matières premières ou de produits en cours de fabrication, ont pour caractéristique première d’avoir une vocation temporaire dans le patrimoine professionnel puisqu’ils sont destinés à être vendus ou utilisés dans le cadre de l’activité économique du débiteur. 2 Dès lors, la constitution d’une sûreté mobilière sur ce type d’actifs apparaît de prime abord comme étant une gageure, tant sur le plan pratique que théorique. N’y a-t-il pas incompatibilité entre, d’un côté, l’impérieuse nécessité pour le constituant de pouvoir continuer à aliéner librement ses stocks dans le cours normal de ses affaires, et de l’autre, la protection du créancier qui doit avoir l’assurance de pouvoir appréhender et disposer des biens grevés en cas de défaillance ? Comment assurer la sécurité de l’un, sans obérer l’activité de l’autre ? 1 Y Blandin, «Sûretés et bien circulant, Contribution à la réception d’une sûreté réelle globale », thèse de doctorat, Paris II, 2014 sous la direction de A Ghozi. n°50 s. qui propose la définition suivante des actifs circulants : tous les biens présents temporairement au sein d’un patrimoine professionnel qui l’intègre par extraction ou acquisition et qui incorporé ou aliéné dans le cadre d’un cycle d’exploitation de l’entreprise est remplacé par d’autres, identiques ou semblables de par leur nature physique ou leur valeur pécuniaire. Pour cet auteur, les biens circulants peuvent être de nature mobilière comme immobilière. 2 Blandin (n 1) 41 n°51.

Sûretés Mobilières et Stocks

135

Vouloir répondre à cette question s’apparente alors à résoudre la quadrature du cercle. Pourtant, le défi mérite d’être relevé quand bien même les obstacles pratiques et théoriques semblent nombreux et importants. Nul ne saurait contester l’importance économique des stocks et l’utilité qu’il existe pour les commerçants, industriels ou autres agriculteurs de pouvoir les valoriser en les offrant en garantie. Car loin de constituer des poids morts comme le soutiennent nombre de gestionnaires adeptes du flux tendu, les stocks peuvent constituer une source de financement non négligeable et un atout pour le développement de l’entreprise. Ils représentent en ce sens, un nouvel et important enjeu de réforme du droit des sûretés. Cette brève contribution se propose d’envisager les problématiques spécifiques posées par les suretés mobilières sur stocks et les voies qui ont pu être suivies pour tenter de résoudre cette quadrature du cercle, tant par les législateurs nationaux notamment français ou américains que par les instances internationales à travers des instruments tels que le Guide Législatif sur les opérations garanties de 2007 de la CNUDCI, 3 sa Loi type sur les sûretés réelles mobilières 4 ainsi que les Principes Européens sur les sûretés réelles sur biens meubles (PEL. Prop. Sec.). 5 Il s’agira ainsi d’envisager dans leurs substances, ces principes qui visent à réconcilier l’inconciliable (I) avant d’examiner la question de la manière, c’est-à-dire de leur traitement légistique, question souvent négligée mais ô combien importante en pratique (II).

I. L’Art : Comment Réconcilier l’Inconciliable En matière de sûretés réelles, la protection du créancier est traditionnellement assurée à travers deux institutions : la dépossession et le droit de suite. Appliqués aux sûretés sur stocks, ces modes de protection s’avèrent toutefois manifestement inadaptés (A) ce qui rend nécessaire le recours à des principes plus respectueux de la spécificité de ces biens (B).

3 Consultable sur le site de la CNUDCI, www.uncitral.org/uncitral/fr/uncitral_texts/security/ Guide_securedtrans.html. Sur ce projet voir notamment : SV Bazinas, «The UNCITRAL draft legislative guide on secured transactions» (2005) 10(1–2) Uniform Law Review / Revue de droit uniforme (Roma) 141–154 ; HC Sigman, «L’influence du modèle américain sur le projet de guide législative de la CNUDCI: mythe ou réalité? » (2004) 97 Banque et droit (Paris) 35–42. 4 Loi type de la CNUDCI sur les sûretés réelles mobilières : www.uncitral.org/uncitral/fr/commission/working _groups/6Security_Interests.html . Les présents développements se fondent sur la version de la Loi type telle qu’adoptée par la Commission lors de sa session de juillet 2016 (A/CN.9/884 and Add 1–4). Sur ce projet, voir notamment B Foëx et alii, Le projet de loi type de la CNUDCI sur les operations garanties, Pourquoi et comment ? (Centre de droit bancaire et financier, Université de Genève, Schulthess ed. 2016). 5 U Drobnig et O Böger, Proprietary Security in Movable Assets (Selier European Law Publishers, 2015).

136

Jean-Francois Riffard

A. L’inadéquation des Modes Classiques de Protection du Créancier La mise en possession du créancier et le droit de rétention qui lui est attaché, assurent une sécurité maximale, puisque sa mainmise sur le bien grevé le protège contre toute aliénation ou dissipation du bien par le constituant. On sait que cette protection a été pendant longtemps érigée en principe fondamental du droit des sûretés mobilières par le législateur français. Pour preuve, le Code civil de 1804 ne reconnaissait comme seule sûreté mobilière que le gage avec dépossession. En fait, sinon en droit, une telle position aboutissait à condamner toute constitution d’une sûreté mobilière sur des stocks. Cette réticence ne fut pas l’apanage du législateur français puisqu’elle a été du moins jusqu’au milieu du XXème siècle partagée par de nombreux autres droits. Il en est ainsi de la Common Law notamment américaine. 6 Dans l’esprit des juges américains, la masse des biens grevés par une sureté mobilière devait rester fixe et ce dans l’intérêt des tiers. 7 C’est pourquoi, là encore, ils se montrèrent pour la plupart d’entre eux, hostiles à la constitution d’un chattel mortgage sur des stocks, le fait de conférer au constituant un jus disponendi sur ces biens, ayant toujours quelque chose de frauduleux. La dépossession du constituant n’est pas en matière de sûretés portant sur des stocks, une réponse adaptée aux attentes de la pratique. Comment admettre que le constituant immobilise ses stocks alors que leur vente constitue le cœur même de son activité ? Imagine-t-on le créancier, généralement une banque, accepter de prendre possession d’un volume de stocks relativement important, alors que matériellement, il n’est pas équipé pour cela ? Confrontée à l’hostilité du droit positif vis-à-vis des sûretés sans dépossession, la pratique fit alors preuve d’imagination afin de résoudre cette nouvelle quadrature du cercle ; respecter l’exigence de la dépossession tout en en limitant au maximum ses effets néfastes. Les commerçants et industriels français comme américains utilisèrent alors habilement la notion de dépossession symbolique ainsi que celle de tiers convenu et indépendant, à travers l’organisation d’entrepôts publics 8 dans lesquels le constituant était invité à déposer ses stocks. Si elle marquait un certain progrès, cette technique impliquait toujours le transport par le débiteur des marchandises dans lesdits entrepôts. Allant plus loin, la pratique eut l’idée de laisser les biens dans l’entreprise du constituant, dans un emplacement ou local clos loué par le créancier, et confié à la garde d’un tiers indépendant. Enfin, le recours à la notion d’instrument symbolique a pu de même permettre la mise en garantie de marchandises en cours de transport, notamment dans le cadre particulier du commerce international. 9 6 J-F Riffard, Le security interest ou l’approche fonctionnelle et unitaire du droit des sûretés mobilières : contribution à une rationalisation du droit français (LGDJ-Montchrestien 1999). 7 Edgell v Hart 9 N.Y. 213, 217 (1853). 8 Les public warehousing américains et les magasins généraux français. 9 L’idée était alors de gager non pas les marchandises elles-mêmes, mais le titre, généralement le connaissement maritime ou le warrant, qui permet au porteur d’en prendre possession auprès du tiers

Sûretés Mobilières et Stocks

137

Mais ces techniques ne constituent qu’un pis-aller et ne répondent que très partiellement aux besoins du monde commercial. Elles ne peuvent suffire à masquer la carence d’un instrument adapté, lequel ne pouvait se présenter que sous la forme d’une sûreté sans dépossession. La reconnaissance d’une telle sûreté, en tant que sûreté mobilière de droit commun, ne fut pas chose aisée, notamment en droit français. Alors que le législateur américain avait franchi le pas dès 1952 avec le Security Interest 10 du Livre 9 de l’Uniform Commercial Code (UCC Article 9), le législateur français choisit dans un premier temps de ne reconnaitre que ponctuellement de telles sûretés sans dépossession, à travers des instruments spéciaux et dans des domaines très spécifiques. 11 Ce ne fut qu’à l’occasion de la réforme du droit des sûretés intervenue avec l’ordonnance du 3 mars 2006 qu’il décida (enfin !) de consacrer de manière générale le gage sans dépossession. Cette reconnaissance tardive s’explique principalement par la réticence des législateurs à faire jouer, en matière mobilière, un rôle plein et entier au droit de suite du créancier, faute de pouvoir organiser une publicité efficace assurant une parfaite opposabilité aux tiers. Cette réticence a été vaincue et le droit de suite, pierre angulaire de toutes sûretés réelles mobilières sans dépossession est aujourd’hui spécifiquement consacré. L’article 2337 du Code civil français dispose désormais que les ayants cause à titre particulier d’un constituant ayant publié régulièrement son gage, ne peuvent se prévaloir de la règle selon laquelle en fait de meuble possession vaut titre. Mais cette reconnaissance soulève une nouvelle difficulté en matière de stocks. En application de la règle de l’article 2337, un client d’un professionnel qui aura gagé ses stocks va acquérir ces biens grevés de la sûreté. Certes l’article 2342 du Code civil apporte dans ce cas, un tempérament. Lorsque le gage sans dépossession a pour objet des choses fongibles tels que les stocks, le constituant peut les aliéner si la convention le prévoit à charge de les remplacer par la même quantité de choses équivalentes. Mais il n’en demeure pas moins qu’à défaut d’une telle autorisation, le droit de suite subsiste. Bien que protectrice des intérêts du créancier, l’application du droit de suite au cas des sûretés sur stocks n’est ni pertinente, ni réaliste. Il est inacceptable sur le plan social, comme sur le plan économique, de sacrifier les clients du constituant. L’essor du commerce impose que ces acheteurs puissent avoir confiance dans les transactions réalisées avec un commerçant. Ils ne peuvent accepter que flotte sur convenu Ce gage sur instrument symbolique a été consacré légalement en France par la loi du 23 mai 1863. De même les cours américaines ont admis dans le même temps de le transfert au créancier du bill of lading équivaut au transfert matériel des marchandises. 10 Sur cette question : Riffard, Le security interest (n 6); A Eigenmann, L’effectivité des sûretés mobilières, Etude critique en droit suisse au regard du droit américain et propositions législatives (Ed. Universitaires Fribourg Suisse, 2001). 11 C’est ainsi que furent consacrés notamment les warrants pétroliers (Art L524-1 à L524-21 C. com.), le nantissement du matériel et de l’outillage (L525-1 à L525-20 C. com.) et les warrants agricoles (L342-1 à L342-17 C. rural).

138

Jean-Francois Riffard

leur tête l’épée de Damoclès d’un éventuel droit de suite du créancier. A cet égard, l’argument tenant à la probabilité d’une autorisation donnée par le créancier au regard de l’article 2342, est à relativiser, car bien que forte, elle n’est pas absolue et elle ne permet pas de lever les doutes pour l’acquéreur. S’agissant des stocks de marchandises, il est au contraire nécessaire d’abandonner purement et simplement le principe du droit de suite et de consacrer, de manière générale, la règle selon laquelle l’acheteur prend toujours les biens libres de toute sûreté, et ce sans exception. Dès lors, une éventuelle connaissance qu’il pourrait avoir de l’existence d’une convention de sûreté, doit être à notre sens inopérante, la vente ne pouvant être considérée comme ayant été réalisée en fraude aux droits du créancier. C’est une solution de bon sens qui de plus correspond aux attentes des parties à la convention constitutive de sûreté. Il est en effet légitime et raisonnable de penser qu’en prenant une sûreté sans dépossession sur les stocks, le créancier accepte délibérément le risque inhérent à leur rotation. Il doit être présumé irréfragablement avoir accepté que les biens soient vendus, libres de toute sûreté, dans le cadre de l’activité commerciale habituelle du constituant, à charge d’être remplacés. S’il n’entend pas courir ce risque, il lui est alors tout à fait loisible de prendre une sûreté avec dépossession en ayant recours à l’ancienne technique des gages sans déplacement ou à celle des magasins généraux. Quant au constituant, il doit avoir l’assurance qu’il pourra aliéner les marchandises laissées en sa possession, sans risquer de voir sa clientèle être inquiétée. Bien entendu, cette protection des tiers acquéreurs se doit d’être strictement encadrée. Elle ne doit bénéficier qu’aux personnes qui achètent des marchandises à un commerçant dont d’activité commerciale habituelle et apparente consiste en la vente de choses pareilles. Cette condition peut être exprimée par référence à la notion d’acquéreur dans le cours normal des affaires du constituant, laquelle notion a été dégagée par le UCC Article 9 12 et reprise depuis par nombre d’autres législateurs, notamment de droit civil. 13 L’acclimatation de cette notion dans les pays de tradition civiliste est d’autant plus aisée qu’elle peut être rapprochée de celle du marchand de choses pareilles consacrée à l’article 2277 du Code civil français. De plus, on relèvera avec intérêt que cette protection des acheteurs dans le cours normal des affaires serait loin de constituer une hérésie dans la mesure où le droit Romain admettait déjà que l’hypothèque générale sur un commerce (taberna) n’emporta pas de droit de suite sur les marchandises vendues par le commerçant. Mais si l’acquéreur dans le cours normal des affaires prend les biens aliénés libres de toute sûreté, comment alors assurer la protection du créancier ?

12

RH Skilton, «Buyer in ordinary Course of Business under UCC » (1974) Wisconsin Law Review 1. 26 de la loi Belge du 11 juillet 2013 modifiant le Code Civil en ce qui concerne les sûretés réelles mobilières et abrogeant diverses dispositions en cette matière. 13 Art

Sûretés Mobilières et Stocks

139

B. La Prise en Compte de la Circulation des Stocks et la Protection du Créancier Pour répondre au défi lancé, il convient de se pencher sur la seconde caractéristique présentée par les stocks. Si ceux-ci doivent pouvoir librement circuler et être aliénés par le constituant, ils ont aussi et plus que tout autre bien vocation à être remplacés. Par définition, les stocks font l’objet d’une rotation régulière. 14 C’est alors dans cette gestion dynamique de l’assiette de la sûreté que peut et va résider la protection du créancier. En consacrant le principe de la continuation de la sûreté sur les stocks ou autres biens venant en remplacement de ceux vendus, il est alors possible de concilier les aspirations contraires du créancier et du constituant. Si la reconnaissance de ce principe peut être aisément consacrée (1°), il n’en demeure pas moins que son application va soulever de nouvelles questions parfois épineuses qu’il s’agisse de l’identification des biens venant en remplacement (2°), de l’opposabilité aux tiers de la sûreté étendue (3°), des nouveaux conflits de priorité engendrés (4°) voire du sort des biens transformés (5°).

i. L’évolution des Stocks : Transport ou Extension de la Sûreté ? La sûreté du créancier doit continuer à grever les biens venant en remplacement des stocks aliénés dans le cours normal des affaires. Cette règle ne souffre, en son principe aucune contestation tant elle apparaît comme une solution d’équité et de justice. 15 Afin d’expliquer ce phénomène, le premier réflexe est de faire appel à la notion de subrogation réelle, véritable clé de voute du droit des sûretés réelles. 16 Cette référence à la subrogation réelle est d’autant plus naturelle que les stocks de marchandises ou de matières premières sont des biens fongibles, c’est-à-dire équivalents en raison de leurs identités naturelles. 17 Comme le note un auteur, c’est de cette parfaite identité des biens fongibles que résulterait leur subrogation réelle. 18 La sûreté doit ainsi se transporter sur les biens de même nature venant en remplacement, de sorte que le créancier pourra réaliser sa sûreté sur ces biens et ce en dépit de toute individualisation. 19 Cette solution classique est celle admise en droit positif français tant au niveau du gage de droit commun que du gage des stocks. L’article L527-2 du Code de commerce rappelle très clairement que le privilège du créancier passe de plein droit des stocks aliénés à ceux qui leur sont substitués. Cette solution se justifie d’autant plus si l’on considère que les

14

Blandin (n 1) 40 n°50. P Veaux-Fournerie, Fongibilité et subrogation réelle en matière de gage commercial, in le Gage commercial, 126 s. 16 V Ranouil, La subrogation réelle en droit civil français (LGDJ, 1985) 218. 17 ibid. 18 Blandin (n 1) 147, n°192. 19 La même solution existe en matière de réserve de propriété sur biens fongibles. 15

140

Jean-Francois Riffard

stocks, biens fongibles, constituent en outre une université de fait. La référence au concept d’universalité de biens est particulièrement adaptée aux sûretés sur stocks et fournit une autre explication tout à fait utile au principe du report. D’ailleurs, le Code civil du Québec a en son article 2674 posé le principe que « l’hypothèque qui grève une universalité de biens subsiste mais se reporte sur le bien de même nature qui remplace celui qui a été aliéné dans le cours des activités de l’entreprise ». Mais le recours à la subrogation ou à l’universalité de fait, notions à forte connotation civiliste, est toutefois limité, n’apportant pas de justification satisfaisant au cas où les biens en remplacement sont d’une autre nature que ceux initialement grevés. C’est pourquoi, confrontés à la question, d’autres législateurs ont adopté une position légèrement différente en optant pour le principe de l’extension de la sûreté sur les biens issus de l’aliénation des stocks. Ainsi, la Loi type de la CNUDCI prévoit que la sûreté sur un bien grevé s’étend à son produit identifiable, y compris au produit du produit. 20 D’aucuns diront que la différence est minime, pour ne pas dire insignifiante. Elle ne serait que d’ordre terminologique. Elle est pourtant à notre sens, digne d’intérêt. Le recours à la notion d’extension, plutôt qu’à celle du report, a le mérite de poser le problème en termes d’assiette de la sûreté. Elle permet d’avoir une approche plus globale et non liée aux seuls biens fongibles. A travers cette notion, il est possible d’ériger en règle générale, le principe selon lequel l’assiette de la sûreté initiale portant sur les stocks comprend non seulement les biens initiaux mais tous les biens venant en remplacement. Raisonner en termes d’assiette de la sûreté et d’extension de celle-ci aux biens venant en remplacement conduit en fait à considérer ces derniers comme des biens futurs, inclus ab initio et de lege feranda dans l’assiette de toute sûreté mobilière. Ainsi, le créancier peut faire valoir sa sûreté dès que les produits apparaissent et intègrent le patrimoine du constituant. Or donc, le principe de l’extension de la sûreté ne sera pas limité aux seuls biens de même nature, mais doit porter sur tout bien venant en remplacement quelle qu’en soit la nature. Il s’agit là encore d’une solution de bon sens. Il se peut en effet que dans le cadre de son activité normale, le constituant aliène les stocks sans procéder à leur remplacement à l’identique. Il peut simplement encaisser les fonds, ou utiliser le prix de vente pour acquérir des biens de nature différente, sans pour autant que cela procède d’une volonté de sa part de frauder les droits du créancier. Bien plus, il nous apparaît que cette extension aux biens acquis en remploi ou transformés à partir du bien initial ne doit pas être cantonnée au cadre spécifique des sûretés sur stocks, mais doit constituer un principe général du droit des sûretés mobilières. Une sûreté mobilière est en effet une sûreté sur la valeur, et le créancier doit pouvoir appréhender cette valeur grevée quel que soit son support. Il est possible d’atteindre un tel résultat en combinant la règle de l’extension de la sûreté et la notion de « produits ».

20

Art 10 reprenant la Recommandation 19 du Guide législatif.

Sûretés Mobilières et Stocks

141

En effet, et au contraire du législateur français qui n’a pas jugé bon aller en ce sens, 21 le législateur américain a consacré de manière générale le principe de l’extension de la sûreté à travers la notion de proceeeds, notion qui recouvre tant les biens de même nature que d’autres quelle que soit leur forme. En vertu de l’articles 9-203 (f), 22 un security interest valablement constitué sur un bien va grever automatiquement tous les produits qui peuvent en découler dès que ceuxci apparaissent et demeurent identifiables. Cette notion de « produit », permet ainsi de consacrer de manière large le principe de l’extension de la sûreté à tout bien venant en remplacement, au-delà de l’hypothèse de la seule sûreté sur stocks. Il n’est guère étonnant que ce concept ait été repris à la fois dans les instruments de la CNUDCI et dans les PEL Prop. Sec.. Selon l’article 2 (bb) de la Loi type, le terme “produit” désigne tout ce qui est reçu en relation avec un bien grevé. Il englobe non seulement ce qui est reçu de la vente ou d’un autre acte de disposition, du recouvrement, de la location ou de la mise sous licence du bien grevé. Il englobe aussi les fruits naturels et civils ou les revenus, les indemnités d’assurance, les droits nés d’un vice, de l’endommagement ou de la perte du bien grevé, et le produit du produit. Quant aux PEL Prop. Sec., ils définissent la notion à l’article IX 1 :201 (11) comme représentant toute valeur issue d’un bien grevé telle que notamment le produit d’une vente. Si le principe de l’extension de la sûreté aux produits identifiables est acquis, reste à savoir si cette extension s’opère de plein droit ou si elle nécessite une stipulation expresse des parties. Sur cette question, des positions différentes ont été défendues. Si la Loi type opte pour l’extension automatique, sauf convention contraire des parties, les PEL Propr. Sec. adoptent une position plus nuancée puisque selon l’article IX 2-306 (3), les produits autres que les créances qui naissent en raison d’un vice, préjudice ou perte du bien initialement grevé 23 ne sont compris dans l’assiette de la sûreté que dans la mesure où les parties en conviennent ainsi. Là encore, il nous apparaît que la règle de l’extension de plein droit, sauf convention contraire des parties, aux biens identifiables venant en remplacement doit être approuvée, car plus conformes aux attentes légitimes et supposées de tous les acteurs, parties ou tiers.

ii. Le Défi du Report sur des Biens d’autre Nature : l’Identification des Produits Le principe de l’extension de la sûreté à tout bien venant en remplacement du bien initialement grevé connaît toutefois une limite pratique. Il n’a vocation et ne 21 Malgré une proposition qui avait été déjà faite par la Société d‘Etudes Législatives qui avait préconisé en 1939 de rajouté à l’article 2076 ancien du Code civil, un second alinéa disposant que si pendant la durée du gage, le bien remis en gage est vendu par le débiteur en accord avec le créancier, les droits de ce dernier sont transportés d’abord sur le prix et ensuite sur l’objet mobilier acquis en remplacement. 22 Renvoyant à 9-315 (a)(2). 23 Y compris les indemnités d’assurance.

142

Jean-Francois Riffard

peut s’appliquer qu’à la condition que le bien de remplacement soit et demeure identifiable. Cette exigence revient à caractériser de manière incontestable un lien direct entre le bien original et le bien issu de son aliénation, la charge de la preuve incombant au créancier. Plusieurs hypothèses sont à envisager. La plus simple est celle du remplacement des marchandises aliénées par des marchandises de même nature. En fait, cette question est d’autant plus aisée à solutionner, qu’elle ne se pose pas en réalité en matière d’extension de la sûreté. En effet, s’agissant de biens fongibles de même nature que ceux décrits de manière générique dans la convention constitutive de sûreté, ils seront de plein droit grevés sans qu’il soit besoin de s’interroger sur leur provenance. La seconde hypothèse est celle des produits se présentant sous la forme d’argent ou d’une créance de sommes d’argent suite à la vente des stocks. S’il ne fait pas de doute que la sûreté subsiste sur ces sommes, 24 la question de leur identification demeure entière. Si le prix de vente n’a pas encore été versé au débiteur, la sûreté s’étendra à la créance du prix de vente. La solution est classique et rejoint celle adoptée par le droit français en matière de clause de réserve de propriété. 25 Au cas où, à la suite de la vente du bien grevé, les produits se présentant sous la forme de monnaie scripturale ont été déposés sur le compte du constituant et mélangés avec d’autres avoirs, le problème de l’identification va prendre toute sa dimension. Si le UCC Article 9 américain fait ici une large place à la notion de « tracing », 26 il n’en est pas de même en droit français. En raison du fonctionnement même du compte bancaire du débiteur et de l’effet novatoire de l’entrée en compte, les sommes encaissées vont être fusionnées en un article unique, le solde, rendant juridiquement impossible toute identification. Pour qu’il en soit autrement deux voies sont envisageables. La première, d’ordre pratique, consiste à isoler le prix de vente dans le patrimoine du débiteur en permettant au créancier d’exiger la domiciliation du paiement du prix des marchandises grevées sur un compte spécial ouvert dans les livres du banquier du débiteur. Ce dernier pourrait de plus recevoir pour instructions de bloquer ces sommes tant que les biens n’auraient pas été remplacés dans le patrimoine du constituant. Souple puisque de nature conventionnelle, cette solution présente l’inconvénient de reposer sur l’entière collaboration du constituant. La deuxième solution pourrait s’inspirer du principe du solde intermédiaire le plus faible connu du droit américain. 27 Cette méthode repose sur l’idée que, d’une part, les sommes autres que les produits grevés sont présumées être retirées en priorité du compte, 28 et que d’autre part les dépôts ultérieurs de nouvelles sommes 24

UCC, § 9-306 (identifiable cash proceeds) et Art 2674 C civ. Q. 2372 C. civil : Le droit de propriété se reporte sur la créance du débiteur à l’égard du sousacquéreur ou sur l’indemnité d’assurance subrogée au bien. 26 UCC, § 931-5(a)(2) : «if the proceeds are not goods, to the extent that the secured party identifies the proceeds by a method of tracing, including application of equitable principles, that is permitted under law other than this article with respect to commingled property of the type involved ». 27 UCC, § 9-315 (b) (2). 28 Universal CIT Credit Corp v Farmers Bank of Portageville 358 F. Supp. 317 (1973). 25 Art

Sûretés Mobilières et Stocks

143

ne peuvent permettre de reconstituer le montant des produits grevés. En d’autres termes, les produits sont réputés être toujours inscrits en compte aussi longtemps que le solde reste supérieur ou égal à leur montant. S’il devient inférieur, les droits du créancier seront alors limités au montant du solde intermédiaire le plus faible affiché par le compte depuis l’entrée en compte des produits. Une solution identique a été adoptée dans le cadre de la Loi type de la CNUDCI. 29 Il est d’ailleurs à noter que cette technique n’est pas sans rappeler celle retenue par le législateur français dans un autre domaine. Confronté à la difficulté de l’identification des gains et salaires d’un époux déposés sur un compte joint, ces gains et salaires étant insaisissables par les créanciers de l’autre époux, l’article R162-9 du Code des procédures civiles d’exécution dispose que lorsqu’un compte, même joint, alimenté par les gains et salaires d’un époux commun en biens, fait l’objet d’une mesure d’exécution forcée ou d’une saisie conservatoire pour le paiement ou la garantie d’une créance née du chef du conjoint, il est laissé immédiatement à la disposition de l’époux commun en biens une somme équivalant, à son choix, au montant des gains et salaires versés au cours du mois précédant la saisie ou au montant moyen mensuel des gains et salaires versés dans les douze mois précédant la saisie. Enfin, le dernier cas de figure, le plus ardu est celui des biens d’une autre nature que les stocks cédés, acquis en remploi des sommes issues de leur aliénation. Il sera ici en pratique quasiment impossible d’établir le lien entre le bien initial et le bien acquis en remploi, à moins que les parties n’aient pris soin de prendre des mesures particulières. Outre la mesure envisagée supra consistant à obliger le constituant à déposer l’ensemble des fonds issus de l’aliénation des stocks initiaux sur un compte spécial, on peut penser à la déclaration de remploi par laquelle le constituant déclarera que les sommes utilisées pour l’acquisition du bien proviennent des sommes perçues. Toutefois, cette technique ne semble pas adaptée à la pratique commerciale. Mais si ces mesures peuvent favoriser l’identification des produits dans les relations entre les parties, et éviter les contestations de ce chef, elles s’avèrent insuffisantes à assurer la parfaite opposabilité aux tiers de la sûreté sur ces nouveaux produits.

iii. Le Défi de la Continuation de la Sûreté et l’Opposabilité aux Tiers Dans quelle mesure la sûreté grevant les produits des stocks initialement grevés demeure-t-elle être opposable aux tiers ? Cette question ne se pose par définition que dans le cas d’une sureté sur stocks rendue opposable par inscription d’un avis sur un registre des sûretés mobilières. Le produit pouvant prendre la

29 Art 10 (c) : Si à un moment quelconque après le mélange, la valeur des espèces mélangées ou du solde crédité sur le compte bancaire est inférieure à la valeur du produit immédiatement avant le mélange, la sûreté réelle mobilière grevant les biens mélangés se limite à la valeur la plus basse entre le moment où le produit a été mélangé et celui où la sûreté est revendiquée.

144

Jean-Francois Riffard

forme d’un bien d’une autre nature que le bien initialement grevé, il peut alors exister une distorsion entre la description du bien tel qu’elle figure dans l’avis d’enregistrement et le produit sur lequel va porter la sûreté. Ce cas pose un nouveau défi puisqu’il va s’agir de concilier deux objectifs antinomiques. Doit-on faire prévaloir l’information des tiers, et la fiabilité du registre en exigeant une réactualisation « en temps réel» de l’information publiée ? Doit-on, au contraire, privilégier la protection du créancier, lequel doit pouvoir opposer sa sûreté sur un produit identifiable quand bien même celui-ci se présenterait sous la forme d’un bien non visé dans l’avis d’enregistrement, et ce, au risque de saper les principes qui sous-tendent les règles d’opposabilité ? 30 Face à ce dilemme et pour résoudre cette nouvelle quadrature du cercle, une solution de compromis est envisageable comme l’atteste la solution retenue par la Loi type de la CNUDCI s’inspirant peu ou prou de celle dégagée par le droit américain. Cette solution de compromis distingue trois cas de figures. En premier lieu, si l’avis d’inscription vise expressément les biens dont le produit va prendre la nature, la question de l’opposabilité de la sûreté de se pose pas. Tel est le cas du créancier qui aura pris soin de viser dans la convention et l’avis d’inscription les stocks présents et futurs du constituant. Cela sera aussi le cas, lorsqu’il aura visé outre les stocks, les biens d’une autre nature pouvant résulter de l’aliénation des stocks initiaux. Si l’avis d’inscription vise par exemple les stocks présents et futurs ainsi que les équipements présents et futurs du constituant, ces équipements seront couverts qu’ils proviennent ou non du remploi du prix de vente des stocks initiaux. Le second cas de figure est celui de l’extension de la sûreté sur stocks aux prix de vente desdits stocks. Si les produits prennent la forme d’espèces, de créances, d’instruments négociables ou de droits au paiement de fonds crédités sur un compte bancaire, tant le droit américain que la Loi type 31 ont fait le choix de consacrer le principe de l’opposabilité de plein droit de la sûreté sur ces produits. Comme le relève avec justesse le Guide législatif, 32 l’idée ici est que l’absence d’un acte d’enregistrement supplémentaire pour ces types de produit ne pose pas un risque de préjudice majeur pour les tiers. En effet, les bénéficiaires du transfert ou les créanciers garantis ne risquent pas d’être lésés dans la mesure où ils prennent généralement les espèces ou les instruments libres de toute sureté. Le bénéficiaire du transfert de fonds provenant d’un compte bancaire les prend généralement libres de toute sureté, si bien que l’absence de publicité ne porte pas atteinte à ses droits non plus. S’agissant des cessionnaires (y compris des créanciers garantis) d’un droit au paiement des fonds crédités sur un compte bancaire, les réclamants concurrents sont censés savoir qu’ils risquent la subordination (voire la perte de la sureté lorsque les espèces sont retirées du compte dans des circonstances qui

30

Guide législatif sur les opérations garanties, §90 p. 131. Art 19 § 1 reprenant la recommandation 39 du Guide législatif. 32 Guide législatif, § 92–93 p. 132. 31

Sûretés Mobilières et Stocks

145

font qu’elles ne sont plus identifiables), à moins qu’ils ne se protègent en prenant le contrôle du compte. Enfin, le troisième cas de figure vise le cas où le bien venant en remplacement n’est ni des espèces, ni un bien dont la nature est visée dans l’avis d’enregistrement. C’est dans cette hypothèse que la recherche du compromis est la plus délicate. Pour y parvenir, il est généralement fait appel à la notion de « délai de grâce ». La sûreté sur le produit va rester de plein droit opposable pendant un certain délai fixé légalement, 33 ce délai devant permettre au créancier de régulariser la situation. Il s’agit pour lui d’enregistrer un avis de modification de l’enregistrement initial afin de faire apparaître, dans la description des biens grevés, le nouveau produit. A défaut, sa sûreté sur les biens acquis en remploi d’une autre nature que les biens d’origine ou des espèces, deviendra inopposable aux tiers. Cette solution constitue un compromis équilibré permettant de concilier les intérêts des tiers et du créancier.

iv. Des Conflits de Priorités Particuliers Une autre spécificité des sûretés mobilières sans dépossession sur stocks est de faire naître des conflits de priorités particuliers dont le traitement va nécessiter de déroger à la traditionnelle règle prior tempore, potior jure donnant priorité au créancier ayant rendu sa sûreté opposable en premier. Outre le conflit avec l’acquéreur dans le cours normal des affaires qui a été envisagé antérieurement, celui opposant le créancier initial et le créancier postérieur finançant l’acquisition des nouveaux stocks venant en remplacement revêt une importance pratique considérable. En effet, si la sûreté initiale sur les stocks s’étend à ceux venant en remplacement, il se peut que ceux-ci soient eux-mêmes grevés d’une sûreté au profit du créancier vendeur. Cette question se pose bien entendu avec acuité essentiellement dans le cadre d’une approche unitaire du droit des sûretés mobilières. Cette approche consiste à ne plus reconnaître la réserve de propriété en tant que telle, mais à l’assimiler alors à une sûreté mobilière de droit commun. Dans ce cas, le conflit n’est pas purement technique et ne se limite pas à un simple conflit entre deux créanciers titulaires chacun d’une sûreté de même nature sur un même bien, conflit devant alors se résoudre en fonction de la date à laquelle les sûretés ont été rendues opposables. Il se pose en des termes très politiques. Il s’agit en effet de savoir si le créancier vendeur est dans une situation telle qu’il doit bénéficier d’un traitement privilégié, à l’instar de la solution retenue dans l’approche non unitaire. La réponse g énéralement apportée est affirmative. Le purchase money interest (PMSI) américain 34 en est une très bonne illustration. Parce que le créancier vendeur a financé l’acquisition d’un 33 Pour droit américain : UCC, § 9-315 (c-e) prévoit que la sûreté régulièrement constituée sur ces biens se reporte automatiquement sur les produits identifiables issus de leur disposition et reste automatiquement opposable à tout tiers pendant une durée de vingt jours (1998). Pour la Loi type de la CNUDCI : Art 19 §2. 34 Riffard (n 6) 140 ; Eigenmann (n 10) 225, §773.

146

Jean-Francois Riffard

bien qui vient accroitre l’actif du débiteur, et in fine, l’assiette de la sûreté du premier créancier, il semble juste et logique de lui octroyer une superpriorité quand bien même sa sûreté aurait été rendue opposable postérieurement. De même on peut concevoir que cette priorité est avant tout une mesure de protection envers le débiteur. Il s’agit, en effet, d’éviter qu’il ne perde toute possibilité de financement et ne soit à la merci d’une décision arbitraire du créancier titulaire de la sûreté mobilière globale qui pourrait refuser toute nouvelle avance. Les conditions d’octroi de cette superpriorité font toutefois débat. Une chose est sûre, le créancier finançant l’acquisition doit inscrire, comme tout créancier garanti, sa sûreté au Registre des Sûretés pour la rendre opposable aux tiers. Mais cette formalité suffit-elle à lui conférer la superpriorié ou d’autres formalités sontelles nécessaires ? Des législateurs, et notamment le législateur américain, l’ont pensé, estimant que le créancier finançant l’acquisition devait en plus aviser les autres créanciers ayant une sûreté inscrite sur les stocks du constituant. L’existence de cette règle spéciale est justifiée par le fait que le créancier qui bénéficie d’une garantie sur les stocks de son débiteur s’attend à ce qu’il y ait une rotation de ces stocks, les nouveaux éléments venant en remplacement des éléments vendus dans le cours normal des activités du débiteur. La disposition des biens en stock ne représente pas un danger pour le créancier tant que s’opère un remplacement systématique de ces biens. Or, si les biens venant en remplacement sont déjà grevés au profit du vendeur, la position du créancier nanti sur stock se verra altérée, l’assiette de sa garantie diminuant. Il est alors juste de prévoir une information du créancier nanti afin qu’il puisse prendre toute disposition nécessaire à la protection de ses intérêts. Cette exigence peut s’avérer lourde pour le créancier. En contrepartie, le droit américain lui offre la possibilité d’un cross collateralization en cas de PMSI portant sur les stocks du débiteur. 35 Ainsi, le créancier titulaire d’un PMSI sur un bien, bénéficiera du même PMSI sur tous les biens qu’il sera amené à vendre par la suite à son débiteur en garantie de n’importe quelle créance. Est ainsi reconnue la possibilité de constituer un purchase-money security interest flottant sur stocks. 36 La solution américaine est loin de faire l’unanimité, notamment auprès des législateurs qui connaissaient un système de réserve de propriété opposable erga omnes et donc occulte. A leurs yeux, l’exigence de l’enregistrement de la sûreté en garantie du prix d’achat représente une charge suffisamment importante pour le créancier pour qu’il ne soit pas opportun de lui en imposer une supplémentaire. Chacune de ses justifications repose sur des considérations tout à fait légitimes. Aussi la Loi type n’a pu trancher et propose sur cette question une alternative reflétant les deux positions. 37 Compte tenu de la position du droit français tra35

UCC, § 9-103 (b) (2). Eigenmann (n 10) § 780. 37 Loi type Art 38. L’option A conditionne la superpriorité non seulement à l’inscription de la sûreté finançant l’acquisition au Registre des Sûretés, mais aussi à l’envoi d’un avis par le créancier finançant l’acquisition à tout créancier garanti ne finançant pas l’acquisition qui a inscrit au registre un avis relatif à une sûreté créée par le constituant sur des biens du même type. L’option B n’exige que l’inscription. 36

Sûretés Mobilières et Stocks

147

ditionnellement très favorable aux créanciers vendeurs et de l’attachement de la pratique à celle-ci, il nous semble raisonnable, au moins dans une perspective française, de retenir la première proposition et n’exiger que la seule inscription de la sûreté. Cette solution aboutit à un juste équilibre entre souplesse du régime et protection des tiers. Un dernier conflit de priorité spécifique, assez pointu, se pose lorsqu’une sûreté sur les stocks d’une part, et une sûreté en garantie du financement de ces stocks d’autre part, se sont toutes deux étendues aux produits. D’un côté, il est possible de soutenir qu’une sûreté sur le produit de stocks doit avoir la même priorité que la sûreté grevant ces biens en garantie du paiement de leur acquisition. Elle répondrait au même régime de sorte que le créancier garanti finançant l’acquisition devrait aviser les autres créanciers garantis du fait qu’il a, avant la naissance du produit, inscrit au Registre un avis concernant des biens du même type que le produit. Sous cette condition, le créancier vendeur conserverait sa superpriorité sur les produits. Mais, d’un autre côté, étendre la superpriorité du créancier vendeur à sa sûreté sur les produits des biens vendus est difficilement justifiable. Au contraire, il nous semble que le sacrifice imposé aux autres créanciers du fait de la superpriorité est trop important. Partant, le plus simple et le plus juste serait d’en revenir à la règle de droit commun en prévoyant que la priorité d’une sûreté grevant des stocks en garantie du paiement de leur acquisition, ne s’étend pas au produit de ces biens. Le conflit entre créanciers sur ces produits doit alors se résoudre par application de la règle générale prior tempore, potior jure.

v. Le Défi des Biens Transformés Les marchandises ne sont pas les seuls biens pouvant être détenus en stocks. Ces derniers peuvent aussi être constitués de matières premières ou de biens semifinis, tels que des pièces détachées. Dans ce cas de figure, la difficulté à laquelle va être confronté le créancier va être d’une autre nature. A la différence des marchandises, les matières premières et les produits semi-finis n’ont pas vocation à être vendus. Ils vont être mélangés à d’autres pour former une nouvelle masse ou transformés ou encore assemblés en un produit fini à travers un processus de fabrication. Quels sont alors les droits du créancier sur ces nouveaux biens ? Avant tout, il est important de noter que le problème des biens incorporés ne se pose que si le bien originellement grevé perd, à travers le processus de transformation, son identité. En effet, s’il demeure identifiable et aisément détachable du bien auquel il a été incorporé, le créancier pourra se prévaloir de sa sûreté initiale et la réaliser classiquement sans difficulté. Dans le cas contraire, une première question va naître quant à l’étendue de l’extension de la sûreté initiale. Dans quelles limites et proportions, cette extension va-t-elle s’opérer ? Le processus de transformation ne doit pas altérer les droits du créancier. Il ne doit pas être non lui conférer un avantage indu. Aussi, un premier réflexe serait de considérer que l’extension s’opère sur le produit fini dans la limite

148

Jean-Francois Riffard

de la valeur des biens grevés au moment de la transformation. Cette solution a été celle retenue par les rédacteurs du UCC Article 9. 38 Si cette limite peut être appliquée sans difficulté aux produits semi-finis, elle est toutefois plus discutable en cas de stocks de matières premières sous la forme de choses de genre (pétrole, blé etc). Lorsqu’une sûreté est prise sur un stock représentant une certaine quantité et une certaine valeur, et que ce stock est mélangé avec des choses de même nature et de même qualité, la question qui se pose alors est celle de savoir si l’on doit raisonner en termes de valeur ou en termes de quantité. Un exemple permettra d’illustrer ce point. Soit une sûreté prise sur un stock de 10.000 litres de kérosène, contenu dans une cuve. La convention constitutive de sûreté ne vise que ce stock initial et ne vise pas les stocks futurs. Par la suite, le constituant transvase ce stock dans une autre cuve contenant déjà 10.000 autres litres de kérosène. Quels sont alors les droits du premier créancier garanti sur ces 20.000 litres ? L’application de la règle de la valeur conduit à déterminer le prix des 10.000 litres au moment du mélange et à limiter les droits de créancier à cette valeur. Ce qui pourra conduire, notamment en cas de hausse du cours du pétrole, à une diminution de la quantité de pétrole qu’il pourra revendiquer au moment de la défaillance du débiteur. A l’inverse, si le cours diminue, il pourra saisir l’intégralité de la quantité au moment de la réalisation, si celle est d’une valeur égale ou inférieure à la valeur des stocks grevés au moment du mélange. Cette solution est a priori séduisante dans la mesure où elle protège le créancier contre la fluctuation des cours. Mais elle aboutit à créer une distorsion de traitement entre un créancier qui a une sûreté sur des biens fongibles non mélangés à d’autres qui subira de plein fouet la chute des cours, et celui qui a la chance de voir ses biens grevés être mélangés à d’autres non grevés. Dans ces conditions, et afin de prendre en compte les fluctuations des cours des produits de base, une solution alternative peut être proposée, sous la forme d’une règle proportionnelle. La sûreté prise sur le stock initial se limiterait alors à la valeur de la masse dans la même proportion que celle où les biens grevés et les biens non grevés ont contribué à la valeur de celle-ci, soit dans notre exemple 50% des 20.000 litres. Si les 20.000 litres se retrouvent au moment de la réalisation de la sûreté, le créancier pourra exécuter sa sûreté sur 10.000 litres quel que soit le cours du pétrole à cette date. En d’autres termes, s’il y a eu hausse des cours, le créancier bénéficiera d’une augmentation de la valeur de l’assiette de la sûreté. Il est à noter que cette question a été fortement débattue dans le cadre des discussions sur la Loi type. Le nouvel article 11 opère désormais une distinction entre les masses et les produits finis. Après avoir rappelé le principe de l’extension d’une sûreté grevant un bien mélangé à une masse ou transformé pour former un produit fini, sur cette masse ou ce produit fini, cet article distingue, s’agissant des limites de l’extension, les deux cas de figure. La sûreté qui se reporte sur une

38 UCC, § 9-336 f (2) : If more than one security interest is perfected under subsection (d), the security interests rank equally in proportion to the value of the collateral at the time it became commingled goods.

Sûretés Mobilières et Stocks

149

masse se limite à la même proportion de cette masse que la quantité de bien grevé représentait par rapport à la quantité de l’ensemble de la masse immédiatement après le mélange. 39 En revanche, la sûreté qui se reporte sur un produit fini se limite simplement à la valeur du bien grevé immédiatement avant qu’il n’ait été incorporé au produit fini. 40 Cette double approche nous semble plus conforme à la volonté des parties et à la réalité, même si, il faut bien le reconnaître, l’énoncé de la règle de l’article 11 § 2 ne brille pas par sa clarté. Le Guide sur l’incorporation devrait d’ailleurs contenir des exemples chiffrés à même d’illustrer cette règle. Dans le prolongement, la même problématique se rencontre en cas de conflit entre créanciers faisant valoir chacun une sûreté sur tout ou partie du produit fini. Ce cas peut être particulièrement délicat lorsque la valeur du produit fini est inférieure à la somme de la valeur des chacun des composants grevés. La solution la plus logique consiste là encore à appliquer une règle proportionnelle. C’est la voie choisie par la Loi type, en son article 33 qui dispose que les créanciers garantis ont droit à une part égale au rapport entre la valeur de leurs obligations garanties respectives et la valeur maximum cumulée des toutes les obligations garanties par les sûretés en concurrence. Il est alors précisé que pour l’application de cette dernière règle, toutes les limitations découlant de l’article 11 sont bien évidemment applicables. Enfin, la dernière difficulté tient à l’opposabilité de la sûreté aux tiers de la sûreté étendue au produit fini ou à la masse. Une inscription initiale qui ne viserait que les stocks de matières premières, est-elle suffisante pour assurer la pleine opposabilité de la sûreté sur le produit fini après transformation ? Comme en matière de produits, ce cas de figure présente une dimension politique car soulevant, là encore, des considérations antagoniques. D’une part, le principe de fiabilité de la publicité sur registre et la protection des tiers devrait conduire à imposer au créancier d’inscrire un avis portant sur les biens finis après transformation, à l’instar de la règle retenue pour les produits. D’autre part, on peut estimer qu’un tel enregistrement serait inutile. En effet, il est raisonnable de penser qu’un tiers qui procéderait à une recherche au nom du constituant et qui prendrait connaissance de l’existence d’une sûreté portant sur les seuls stocks de composants par exemple, aura pleinement conscience que cet avis inscrit porte également sur le produit fini obtenu à partir de ces composants puisqu’il s’agit de l’activité de fabrication du constituant. De plus, l’inscription n’est pas nécessaire pour protéger les acheteurs ultérieurs dans la mesure où les produits finis deviendront alors marchandises vendues dans le cours normal des affaires du constituant. Au vu de ces considérations, le Guide législatif a fait sienne cette dernière approche. 41 Lorsqu’une sûreté sur un produit fini ou une masse est opposable, la sûreté sur le produit fini qui en résulte, dans le cadre de l’activité habituelle de fabrication du constituant est de plein droit opposable sans autre formalité de la part du créancier garanti. 39

Art 11 § 2. Art 11 § 3. 41 Guide législatif, recommandation n°44. 40

150

Jean-Francois Riffard

II. La Manière : le Traitement Légistique du Gage des Stocks Les développements précédents nous ont permis de constater que les stocks présentent quelques particularités que le législateur se doit de prendre en compte dans le cadre de l’élaboration d’un régime de sûreté mobilière moderne et efficace. Mais ces spécificités justifient-elles la création d’une sûreté spéciale soumise à un régime autonome (A) ou est-il possible d’intégrer ces règles au sein d’un régime unique de sûreté mobilière selon le modèle aujourd’hui prôné et suivi par beaucoup (B) ?

A. La Tentation du Régime Spécial Instaurer un régime spécial applicable aux sûretés sur stocks est un réflexe assez naturel, qui s’explique d’abord par des raisons historiques. A l’époque de la prohibition des sûretés mobilières sans dépossession, le législateur français avait déjà consacré quelques sûretés spéciales sans dépossession dans des domaines particuliers. Cette tentation du régime spécial a été si forte que le législateur ne sut y renoncer dans le cadre de la réforme du droit des sûretés intervenue en 2006. Passant alors d’un extrême à l’autre, il consacra non pas une mais deux nouvelles sûretés sur les stocks. Alors que le nouveau gage sans dépossession de droit commun pouvait parfaitement porter sur les actifs circulants, le législateur décida contre toute attente de créer une nouvelle sûreté spéciale sous la forme du gage des stocks du Code de commerce. Régi par les articles L. 527-1 et suivants du Code de commerce, le gage des stocks n’a toutefois pas connu de succès en pratique, eu égard notamment à son régime tout à fait contraignant. Cet échec n’a pas empêché un auteur de continuer à prôner cette voie en proposant la création d’une nouvelle sûreté réelle spéciale, la sûreté globale, portant sur les actifs circulants, meubles comme immeubles et qui se caractériserait par sa nature hypothécaire et fluctuante. 42 De même, et alors qu’il était parfaitement conscient de l’échec du gage des stocks et de sa néfaste concurrence avec le gage sans dépossession du Code civil, le législateur français a choisi de réformer récemment le régime de ce gage en le conservant mais … en l’alignant sur le gage de droit commun ! 43 Quoi qu’il en soit, le recours à une sûreté spéciale doit être à notre sens rejeté, et ce pour deux raisons majeures. Outre le fait qu’il soit source d’insécurité juridique, un tel recours est surtout inutile.

42

Blandin (n 1) 337. n° 2016-56 du 29 janvier 2016 prise en application de la loi n° 2015-990 pour la croissance, l’activité et l’égalité des chances économiques du 6 août 2015, dite « loi Macron ». 43 Ordonnance

Sûretés Mobilières et Stocks

151

La coexistence de deux régimes de sûretés sur stocks, l’un général, et l’autre spécial, ne peut qu’engendrer une insécurité juridique préjudiciable aux créanciers, comme l’a montré l’exemple français. Le régime spécial du gage des stocks, dans sa version initiale, s’avérant plus lourd et contraignant que le régime général, s’est très rapidement posée la question de savoir si les parties pouvaient par le biais d’une clause expresse, décider de soumettre leur gage sur stocks aux seules dispositions du Code civil. Bien qu’une majorité de la doctrine soit favorable à cette possibilité, la Cour de cassation s’est dans un premier temps montrée réticente. 44 Mais la juridiction de renvoi est alors entré en résistance. Las, la Cour de Cassation a alors, par un arrêt de son Assemblée plénière du 7 décembre 2015 mis fin aux espoirs de la pratique en confirmant le caractère exclusif du régime spécial. 45 Cette jurisprudence fut toutefois à son tour condamnée par la réforme intervenue avec l’ordonnance du 29 janvier 2016, laquelle prévoit expressément que les parties demeurent libres de recourir au gage des stocks ou au gage de meubles corporels prévu aux articles 2333 et suivants du Code civil. 46 Cette liberté de choix, couplée avec le fait que le régime spécial est aligné avec le régime de droit commun, pose de plus fort la question de l’utilité de conserver une telle sûreté spéciale. De plus, l’existence d’un régime spécial suppose que son champ d’application soit parfaitement délimité, ce qui aboutit, en ce qui nous concerne, à définir de manière précise et incontestable la notion de stocks. Or, il est à craindre que des constituants peu scrupuleux n’hésitent pas à contester ou invoquer selon le cas, la nature de stocks de certains biens grevés afin de remettre en cause la sûreté conférée. Ce risque est d’autant plus grand si, à l’instar d’une proposition doctrinale récente, le champ d’application de la sûreté sur actifs circulant est défini non pas à partir de la nature des biens engagés mais de leur fonction. 47 Il sera alors assez tentant pour le constituant de contester ou de modifier cette fonction. En tout état de cause, il y a là un nid à contentieux incompatible avec la sécurité à laquelle tout créancier est en droit d’aspirer. Source d’insécurité, la solution de la sûreté spéciale est de plus à contre-courant de la tendance actuelle gouvernant la plupart des réformes modernes en matière de sûretés mobilières. Ces projets consacrent un régime unique de sûreté, suffisamment large pour couvrir l’hypothèse des sûretés sur stocks, rendant ainsi inutile le recours à un régime spécial autonome.

44 Cass. Com. 19 fév. 2013 n°11–21.763, RTD Civ. 2013 p. 418 obs. P. Crocq ; JCP G 2013, 539 note N Martial-Braz, Ch Gijsbers, « L’exclusion du droit commun du gage par le régime spécial du gage des stocks », RLCD avr. 2013 p. 26. 45 Cass. ass. plén., 7 déc. 2015, n° 14-18.435 : JurisData n° 2015-027117 ; JCP G 2016, 57, note J.-J. Ansault et C. Gijsbers ; Rev. Lamy dr. civ. févr. 2016, 134, note P Pailler ; JSS 13 févr. 2016, n° 12, note E Ronzier ; Gaz. Pal. 12 janv. 2016, 2, note S Piédelièvre. 46 Art L527-1 C. Com. 47 Blandin (n 1) 243, n°325.

152

Jean-Francois Riffard

B. La Voie des Règles Dérogatoires dans le Cadre d’un Régime Unique de Sûreté Indifférenciée Structurer le droit des sûretés mobilières conventionnelles autour d’une seule figure, soumise à un régime unique et capable de porter sur n’importe quel bien en garantie de n’importe quelle créance, est considéré comme la voie à suivre, notamment par la de Loi type. Il faut dire que ce modèle a déjà fait ses preuves comme en attestent les réussites des réformes américaine, québécoise, ou plus récemment, mexicaine, vietnamienne ou encore ghanéenne … 48 Il n’en demeure pas moins que la question du traitement des sûretés sur stocks semble poser un défi dans le cadre d’un tel régime. Comment prendre en compte leurs spécificités ? Ce défi peut être une nouvelle fois relevé en limitant d’une part les dispositions de nature législative au minimum (1°) et en faisant d’autre part une large place à la liberté contractuelle (2°).

i. Un Régime Légal Homogène Intégrant les Sûretés sur Stocks Les expériences du Guide législatif et de la Loi type de la CNUDCI démontrent que la spécificité des sûretés sur stocks peut parfaitement être prise en compte dans le cadre du régime de la sûreté mobilière unique. Car adopter une approche unitaire ne veut pas pour autant dire imposer un régime uniforme. Il est évident que la spécificité de certains biens meubles, tels que les stocks, implique l’adoption de certaines règles dérogatoires sur des points précis. Mais, à la lumière des développements qui ont précédé, on se rend compte que ces dérogations vont être somme toute limitées et ne remettent en cause nullement le caractère homogène du régime proposé. Force est au contraire de constater que la constitution d’une sûreté sur stocks ne présente aucune particularité par rapport au régime général, tant sur le plan du fond que de la forme, dans la mesure où le régime permet à celle-ci de porter sur un bien ou un ensemble de biens décrits de manière générique dans la convention constitutive de sûreté. De même, la publicité de la sûreté par inscription d’un avis au Registre général des sûretés est parfaitement adaptée, sans qu’il soit besoin de créer un registre spécialisé qui n’apporterait aucune plus-value et serait au contraire source de confusion. De même, nous avons pu démontrer que la règle de l’extension de la sûreté aux biens venant en remplacement des stocks grevés, fondamentale en cette matière,

48 Sur l’importante réforme mexicaine : DB Furnish, «Mexico’s Emergent New Law Of Secured Transactions: Recent Developments 2000 – 2010 » (2011) 28(1) Arizona Journal of International & Comparative Law 143 s. ; J Barry, «Secured Transaction Reforms in Mexico: In Pursuit of a Uniform System » (2012) 34(2) Houston Journal of International Law. Sur la réforme survenue au Ghana en 2008 et au Vietnam en 2012 : IFC, Secured transactions and Collateral registries, www.ifc.org/wps/wcm/con nect/793e79804ac10fff9ea69e4220e715ad/Secured+Transactions+and+Collateral+Registries+Broch ure-English.pdf?MOD=AJPERES.

Sûretés Mobilières et Stocks

153

peut parfaitement être formulée sous la forme d’une règle générale applicable aux produits du bien grevé. Enfin, s’agissant de la réalisation, les règles générales permettant au créancier à son choix de vendre le bien sous sa propre responsabilité, ou d’en solliciter l’acquisition à titre d’exécution de l’obligation garantie, sont parfaitement adaptées aux cas des stocks. A cet égard, la prohibition du pacte commissoire telle que consacrée par le législateur français dans le cadre du gage des stocks apparaît comme une anomalie difficilement justifiable, qui explique d’ailleurs en partie l’échec de cette sûreté. En définitive, les règles particulières sont limitées à trois points particuliers que sont 1°) le traitement du conflit opposant le titulaire d’une sûreté sur stocks à l’acquéreur dans le cours normal des affaires, 2°) celui l’opposant au créancier ayant financé l’acquisition des stocks, et 3°) le cas des biens mélangés. Formellement, ces trois peuvent être traités par le biais de dispositions relativement courtes et concises. Ainsi, le principe selon lequel l’acheteur d’un bien meuble corporel grevé vendu dans le cours normal des affaires du vendeur acquiert ses droits libres de la sûreté, peut-il être simplement exprimé sous la forme d’une exception à la règle générale en vertu de laquelle, en cas de transfert du bien grevé, un acheteur, un preneur à bail ou un preneur de licence sont soumis à toute sûreté rendue opposable grevant ledit bien. 49 S’agissant de la question de la priorité d’une sureté réelle mobilière grevant des stocks en garantie du paiement de leur acquisition, il est possible d’exprimer la règle particulière sous la forme d’un article disposant qu’une telle sûreté a priorité sur une sûreté concurrente non liée à leur acquisition et créée par le constituant, à condition que le créancier garanti finançant l’acquisition soit en possession desdits biens ou qu’il ait, avant la prise de possession ou l’acquisition des biens par le constituant, inscrit au Registre des sûretés mobilières un avis concernant la sûreté en garantie du paiement de l’acquisition. 50 Enfin, il est possible de s’inspirer des articles 11 et 33 de la Loi type afin de résoudre le cas des stocks mélangés ou incorporés pour former une masse ou un produit fini. Il est alors possible d’inclure dans le régime général de sa sûreté un article disposant que le gage sur un bien meuble corporel mélangé à une masse se reporte sur cette masse en se limitant à la même proportion de cette masse que la

49 Ainsi l’article 34 de la Loi type prévoit-il en son alinéa premier que « Si un bien grevé est vendu ou transféré d’une autre manière, loué ou mis sous licence et si une sûreté réelle mobilière grevant ce bien est opposable au moment de la vente ou du transfert, de la location ou de la mise sous licence, les droits qu’acquiert l’acheteur, le bénéficiaire du transfert, le preneur à bail ou le preneur de licence sont soumis à la sûreté, sous réserve des dispositions du présent article. ». L’alinéa 4 de ce même article dispose quant à lui que « L’acheteur d’un bien meuble corporel grevé vendu dans le cours normal des affaires du vendeur acquiert ses droits libres de la sûreté, à condition qu’au moment de la conclusion du contrat de vente, il ne sache pas que la vente viole les droits du créancier garanti découlant de la convention constitutive de sûreté ». 50 cf Art 38 option A de la Loi type.

154

Jean-Francois Riffard

quantité de biens grevés représentait par rapport à la quantité de l’ensemble de la masse immédiatement après le mélange. Le gage qui se reporte sur un produit fini se limite à la valeur des biens grevés immédiatement avant qu’ils n’aient été incorporés au produit fini. Lorsque plus d’une sûreté réelle mobilière se reporte sur la même masse ou sur le même produit fini et que chacune de ces sûretés grevait un bien meuble corporel distinct au moment du mélange, les créanciers garantis ont droit à une part égale au rapport entre la valeur de leurs sûretés respectives et la valeur maximum cumulée de leurs sûretés sur la masse ou le produit fini.

ii. Un Régime Légal Nécessairement Complété par le Recours à la Liberté Contractuelle Au-delà de ces principes de nature législative, rien ne fait obstacle à ce que les parties puissent aménager leurs relations conventionnellement. Ce recours à l’autonomie des parties doit même être encouragé en matière de sûretés sur stocks. On peut à cet égard s’étonner de ce que le législateur français ait choisi d’ériger, dans le cadre du régime du gage des stocks, en dispositions de nature législatives, des règles qui relevaient à l’évidence, du domaine contractuel. Sans doute était-ce là une manière assez artificielle de donner un peu plus de corps à ce régime … Ainsi en est-il de l’obligation pour le constituant de justifier que les stocks sont assurés contre les risques d’incendie et de destruction 51 ou encore de celle qui lui est faite de tenir à la disposition du créancier un état des stocks engagés ainsi que la comptabilité de toutes les opérations les concernant. Il est probable qu’un créancier diligent et soucieux de la défense de ses intérêts ait pris soin d’inclure dans la convention constitutive de sûreté. Le recours à la liberté contractuelle va permettre au créancier de renforcer sa position en encadrant l’utilisation des stocks grevés par le constituant. Sans doute, les parties auront-elles recours par exemple à la clause d’arrosage, laquelle est particulièrement fréquente en matière de nantissement de compte-titres. 52 En vertu de cette clause, le créancier peut convenir avec le constituant d’un niveau de stocks en deçà duquel le constituant ne saurait descendre. Pour le respect de cet engagement, les parties pourront aussi convenir des modalités du contrôle que pourra exercer le créancier (envoi d’état des stocks, visite des lieux, etc.). Quant aux sanctions pour le non-respect de cet engagement, la sanction sera là encore prévue par le contrat et pourra aller de la simple mise en demeure d’avoir à reconstituer les stocks dans un certain délai jusqu’à la déchéance du terme. 53 Le recours à la liberté contractuelle peut aussi, comme nous l’avons vu précédemment, s’avérer utile pour les parties afin d’aménager les règles r elatives à 51

Art L527-6 C. com. H Le Nabasque, JM Gaillard et M Baffreau, «L’assiette du nantissement de compte d’instruments financiers » (1998) juill-août RD Bancaire et bourse 132. 53 Il est à noter que ce mécanisme a reçu une consécration légale, à notre sens inutile, dans le cadre du gage des stocks du Code de commerce français (Art L527-7 alinéa 3 C. com). 52

Sûretés Mobilières et Stocks

155

l’identification des produits et notamment des biens acquis en remploi. Il appartient alors au créancier de prendre, en accord avec le constituant, toutes les mesures pratiques permettant de favoriser l’identification des produits et par là même de renforcer sa position. De manière générale, il est beaucoup plus raisonnable de laisser, sur l’ensemble de ces points, les parties trouver le meilleur accord possible, en fonction de leurs situations respectives, sans qu’une consécration législative apporte véritablement de plus-value. *** Et la montagne accoucha d’une souris ! Tel est le constat, frustrant, que l’on est tenté de dresser à l’issue de cette étude. Alors qu’il était a priori légitime de penser que la spécificité des stocks imposerait l’adoption d’un nombre important de règles complexes régissant la constitution d’une sûreté mobilière sur ces biens, il apparait in fine qu’il n’en est rien. Une telle conclusion est en fait rassurante. Elle démontre que les projets d’harmonisation du droit des sûretés mobilières ont aujourd’hui atteint un tel degré d’homogénéisation qu’il est tout à fait possible d’intégrer dans le régime unique proposé, les sûretés portant sur des biens aussi spécifiques que les stocks, tout en répondant parfaitement aux attentes des praticiens. Bien mieux en tout cas que nombre de législations pourtant spécialement dédiées, comme le gage des stocks français. Preuve une nouvelle fois, que l’unification du droit des sûretés mobilières, légitimement souhaitée aussi bien sur le plan national qu’international, est utile, efficace et à portée de main. Et notre ami Rod Macdonald n’y est pas pour rien….

156

 

10 Some Thoughts (and Facts) about the Process of Secured Transactions Law Reform, with Special Emphasis on Registration, the Key to Achievement of Reform’s Goals MAREK DUBOVEC* AND HARRY C SIGMAN**

I. Introduction By providing for a proprietary interest in movable assets that secures an obligation, the law enhances the value of such assets to their owners (by making them available for use as collateral) and facilitates and encourages the extension of credit.1 To maximise the achievement of that function, the law must keep up with developments on the ground. For a long period of history, the law formally

* Dr Marek Dubovec is the Senior Research Attorney of the National Law Center for Inter-American Free Trade in Tucson, Arizona. He is a delegate to UNCITRAL’s Working Group VI and has been consulting on a number of secured transactions projects, including in Colombia, Ghana, Honduras, Kenya, Malawi, Mexico, Mozambique, Pakistan, and Zambia. ** Harry C Sigman (JD, Harvard University, 1963) is a member of the California Bar, specialising in commercial law. He has taught and lectured on secured transactions at law faculties around the world, and has been involved in law reform projects around the world since the early 1990s, including serving as a member of the Drafting Committee in connection with the 1998 revision and 2010 amendment of UCC Art 9 and as a delegate of the United States to UNCITRAL’s Working Group VI in connection with the UN Convention on the Assignment of Receivables in International Trade and the UNCITRAL Legislative Guide on Secured Transactions and to the Convention on the Law Applicable to Certain Rights in Respect of Securities Held with an Intermediary (Hague Securities Convention). 1 We use the terminology adopted by the United Nations Commission on International Trade Law, The UNCITRAL Legislative Guide on Secured Transactions (United Nations, Vienna and New York, 2010) (the Legislative Guide) of security right, grantor, secured creditor and registration rather than security interest, debtor, secured party and filing. References in this chapter are to the UNCITRAL Model Law, as adopted at the 49th Commission Session, 27 June–15 July 2016, New York, and references to the draft Guide to Enactment are to the draft dated 12–18 April 2016 (A/CN.9/855 and Add 1–4, available at www.uncitral.org/uncitral/commission/sessions/49th.html, accessed 21 April 2017.

158

Marek Dubovec and Harry C Sigman

recognised pledges only if supported by dispossession of the grantor (to guard against the false appearance of wealth).2 However, the ‘real world’ required some method of providing security while leaving the collateral (encumbered asset) in the hands of the grantor for use or sale. A functional nonpossessory pledge created by a constructive delivery or physical possession (such as, by way of a bailment or lease) being in the hands of the pledgor that differed in nature from the contemporary European concept of pledge (which was dependent on continuous physical dispossession) was already known in Roman law.3 In some countries, the ability of the grantor to remain in possession of the collateral was enabled by use of ownership (title),4 constructive possession or another legal fiction, either judicially created or sanctioned. In some other countries, this ability was provided by special legislation for a specific type or range of transactions.5 The secured transactions law in most countries often consisted of a Civil Code, special legislation, some of which provided for registration of some kind, and/or judicial decisions frequently resulting in complexity simply as the result of the multiplicity of uncoordinated sources. These sources also often provided rights only in favour of certain classes of creditors (eg banks),6 or were applicable only to certain classes of assets or only to certain classes of grantors (eg registered companies).7 The first comprehensive secured transactions law, based on a unitary and functional approach, generally applicable to all types of secured creditors, all types of grantors and all types of assets (present or future, tangible or intangible), organised in a coherent fashion in a single comprehensive enactment, utilising a single registry, addressing the creation of a security right, its third-party effectiveness, priority vis-à-vis competing claims to the assets, enforcement of the security right, the application of the relevant law and transition from the old to the new framework, was Article 9 of the Uniform Commercial Code (UCC Article 9).8 Numerous countries have subsequently followed this approach, as have numerous model laws, and this approach has now become the standard promoted by experts

2

As declared in the classic French maxim, ‘il n’y a pas de gage sans dépossession.’ See WJ Zwalve, ‘A Labyrinth of Creditors: A Short Introduction to the History of Security Interests in Goods’ in EM Kieninger (ed), Security Rights in Movable Property in European Private Law (Cambridge, Cambridge University Press, 2004) at 41–42. 4 Whether title retention by a seller, eg eigentumsvorbehalt, developed by the German courts into a powerful device, especially in its verlängerte and erweiterte form, or by fiduciary transfer (sicherungsübereignung). 5 See, for instance, French statutes for financing of hotels, oil in pipelines. 6 See Loi Dailly in France. Law 81-1 of 2 January 1981 (now incorporated into the Monetary and Financial Code as Arts L. 313-23 to L. 313-35 C. mon. fin.). 7 See the English-law-inspired Companies Acts adopted in many African countries that limit the scope of their application to registered companies: L Gullifer, ‘Piecemeal Reform: Is it the Answer?’ in F Dahan (ed), Research Handbook on Secured Financing in Commercial Transactions (Edward Elgar, 2015) 421, 426. In France, the fiducie may not be constituted by an individual. 8 UCC Art 9 initially entered into force in Pennsylvania in 1954. See generally HC Sigman, ‘Security in Movables in the United States—Uniform Commercial Code Article 9: a basis for comparison’ in Kieninger (n 3) 54ff. 3

Process of Secured Transactions Law Reform

159

in the field and organisations that promote secured transactions law reform.9 A key element, often the very element that motivates the undertaking of a reform effort, is the establishment of a registry. It is apparent that the rules governing the registry are a crucial part of the substantive law, each of which reflects a policy determination that should be made by the legislature and not by the registry itself. The duties and powers of the registry should be limited to technologically and administratively executing the statutory rules. These rules should be provided in the utmost clarity and comprehensiveness so that they provide the certainty that induces the extension of the greatest amount of credit at the lowest possible cost. Many of these points are elaborated on below. After setting forth the primary characteristics of a modern efficient registry, this chapter will discuss several issues concerned with the process of secured transactions law reform and the UNCITRAL Model Law, focusing especially on these issues as they relate to the registration system.

II. Key Features of a Modern Efficient Registry We here set forth the current consensus among experts concerning the key features of a modern efficient registry. A registry should: (1) be fully automated and entirely electronic; (2) be available 24/7 for both searching and registering, with each of these functions accomplished virtually instantaneously; (3) be accessible to all persons who wish to search and to all registrants who establish accounts at (or are otherwise recognised by) the registry; (4) be simple and user-friendly, allowing it to be utilised by small and medium-sized enterprises (SMEs) without the need for assistance of counsel;10 (5) operate in a manner that does not subject notices submitted for registration to screening or review by registry personnel other than mechanical screening to confirm machine-readability and the existence of some content in each of the required fields; (6) utilise a single nationwide database built on a grantor-based (not an asset-based) index; (7) be able to accept notices covering substantially all forms of security rights (however denominated) over movables, including those covering title-based rights, and possibly also non-consensual liens; (8) be set up to require only a statutorily prescribed minimum amount of data, essentially limited to identification11 of the parties and 9 See, for instance, the Legislative Guide (n 1), and AJ Duggan and JS Ziegel, Secured Transactions in Personal Property, Cases, Text, and Materials (6th edn, Toronto, Emond Montgomery, 2013) 19. 10 B Whittaker, ‘The Review of the Personal Property Securities Act 2009—Final Report’ (Commonwealth of Australia, 2015) 158 highlighted that ‘Many users of the Register will have no legal training, and no experience working with complex legal concepts. This means that simplification is even more important for the Register than it is for the Act.’ 11 It should be noted that identification, with respect to the secured creditor, can include identification of a representative of the secured creditor. There may be valid reasons for the parties to wish

160

Marek Dubovec and Harry C Sigman

a description, as defined by the law, of collateral; (9) to the greatest extent possible, utilise numerical identifiers to minimise the risk of incorrect identification of the grantor; and (10) especially notably, be free of cost for both searching and registering, or subject only to minimal fees as prescribed by the law that are no higher than sufficient to cover the operating costs of the registry.12 Notwithstanding the frequent use of the term ‘collateral registry’ (eg the Bank of Ghana operates the Collateral Registry), what is registered is neither a security right (which need not exist at the time of registration) nor a particular document that creates or evidences a security right (eg the security agreement itself), but rather only a limited number of specific data elements that, taken together, constitute a notice of a possible security right, generally described as ‘notice filing’ rather than ‘document registration’. A recent World Bank study highlighted that without a well-functioning registry even the best secured transactions laws will be ineffective.13 Most of the recent reforms in Latin America (eg Honduras and Colombia) and Africa (eg Liberia and Malawi) have established efficient registries, although Guatemala by dint of its fee structure14 has made its reform of little value, as the registry is scarcely used.15 The establishment of an efficient registry underpinned by a fundamentally sound legal framework determines whether the reform will have the desired effect. While experience shows that a sound legal framework without an effective registry will have minimal, if any, economic impact, it remains unclear what the economic impact of a sound registry system which is built on a less than comprehensive substantive law may be.

to maintain confidentiality (eg when a seller and its customer wish to keep the fact of their business relationship secret). 12 A registry model reflecting these features has been implemented not only on a national level but also internationally under the Convention on International Interests in Mobile Equipment and Protocol Thereto on Matters Specific to Aircraft Equipment. Registration is not a requirement for the constitution of an international interest, nor is it a guarantee than an international interest exists; the function of registration is to give notice to third parties of the existence of an international interest if it does exist and preserve the priority of the holder of the interest: R Goode, Convention on International Interests in Mobile Equipment and Protocol Thereto on Matters Specific to Aircraft Equipment, Official Commentary (3rd edn, UNIDROIT, 2013) 76–80. Similarly, under Art IX—3:301 of the Draft Common Frame of Reference that might be the basis for a future European register of security rights, ‘the registration does not have a constitutive effect in relation to the creation of a security right or of a retention of ownership device’: see U Drobnig and O Böger, Proprietary Security in Movable Assets (Sellier, 2015) 433. 13 I Love, MS Martínez Pería and S Singh, ‘Collateral Registries for Movable Assets, Does Their Introduction Spur Firms’ Access to Bank Finance?’ Policy Research Working Paper 6477 (The World Bank, 2013) 3. 14 See the discussion of fees below in section VIII E. 15 We present examples primarily from Africa and Latin America, based on practical experiences in reforming secured transactions laws there (in which we were directly involved), because these are the sites of the most recent reform efforts and because, for the most part, they produced modern, efficient registries.

Process of Secured Transactions Law Reform

161

III. Reform Approaches: Comprehensive or Piecemeal In many countries, the circumstances may be such that a comprehensive reform is not practically or politically feasible. This may be because: (1) the entity in charge of the reform initiative has only limited powers in terms of the regulatory framework that it may adopt on its own (eg the Central Bank of Nigeria); (2) the federal parliament has the power to legislate in only certain subject-matter areas, while the power to enact and repeal some laws remains with provincial or state parliaments (eg Pakistan);16 or (3) a strong position is taken by an important stakeholder that wants to retain some aspects of the current legal framework (eg in Zambia, the Patents and Companies Registration Agency (PACRA) initially sought to limit the scope of the future secured transactions law to individuals and non-registered companies, intending to retain the existing system of fixed and floating charges for registered companies).17 Yet, these countries are pursuing limited-in-scope reforms driven by the desire to establish a registry. Other countries may have different constraints that could limit the registry to a particular set of transactions with the view that when the political climate is ripe for a comprehensive reform, a registration system will already be in place.18 For instance in Mexico, initially, financial leases and sales of accounts receivable were excluded from the secured transactions law, not made subject to registration and became effective on execution of the relevant agreements between the parties. Finally, the law may be deficient because of errors committed during the drafting process that affect its scope and the relationship with other existing laws and registries, such as in Ghana. These countries are examples of reform projects that are less than comprehensive, but in which the registry has played or will play a critical role in sowing the seeds for a subsequent systematic and comprehensive overhaul. Piecemeal reforms present a number of challenges and inevitably yield less legal and economic benefits than comprehensive reforms.19 A danger of this 16 A modern nationwide Personal Property Securities Act and a single nationwide registry became possible in Australia only after the constitutional federalism problem was solved by agreement among the states and the federal government. See A Duggan and D Brown, Australian Personal Property Securities Law (Chatswood, NSW Lexis Nexis Butterworths, 2012) 19. 17 On the recommendation of the Ontario Law Reform Commission, corporate securities were excluded from the original Ontario PPSA. Not surprisingly, the Ontario courts had difficulty distinguishing between corporate securities governed by the CSRA and other types of security rights given by corporations: JS Ziegel, ‘Canadian Perspectives on “How Far Is Article 9 Exportable?” The English Experience’ (1996) 27 Canadian Business Law Journal 232. 18 Some reforms may also be limited in other aspects, such as being limited to a particular kind of asset or transaction: HC Sigman and EM Kieninger, ‘The Law of Assignment of Receivables: in Flux, Still Uncertain, Still Non-Uniform’ in HC Sigman and EM Kieninger (ed), Cross-Border Security over Receivables (Munich, Sellier, 2009) 1. 19 Gullifer (n 7) 449–50.

162

Marek Dubovec and Harry C Sigman

approach is that policymakers in the country become satisfied with the already achieved benefits and fail to realise that the deficiencies are an obstacle to further growth.20 A registry for notices relating to security rights in movable assets is often the primary motivating factor for policymakers to entertain the idea of a legal reform.21 A typical demand made on donor agencies by policymakers and other stakeholders is that they assist the country in establishing registries rather than reforming the legal framework. These policymakers only later realise that the specifics of the establishment of a registry are the product of the adoption of a particular legal framework. Once the policymakers have committed themselves to establishing a well-functioning electronic registry, the door to examination of the country’s secured transactions framework is open. Studies and experiences from a number of countries indicate how much economic impact comprehensive secured transactions reforms can have.22 The policymakers will eventually recognise that limitations may inevitably exist with respect to some aspects of the secured transactions framework, such as when only regulated financial institutions may act as secured creditors or only non-registered companies and individuals may create security rights as grantors. However, that does not change their goal to establish as modern, effective, inexpensive to use and easily accessible a registry as possible. Whether countries pursue comprehensive overhauls or engage in limited reforms, the resulting substantive framework is not self-executing. Professor Macdonald reminded us that ‘Once the reform is proclaimed in force, governments have to invest23 in the legal infrastructure—in the instant case, primarily the registry’.24 When properly implemented, the impact of registries can be significant. For instance, a World Bank study found that a secured transactions reform increases access to loans by seven percentage points, reduces interest rates by an average of three percentage points and extends loan maturities by six months.25 These are the kinds of headlines that grab the attention of policymakers

20 M Dubovec and B Osei-Tutu, ‘Reforming Secured Transactions Laws in Africa: The First African Collateral Registry in Ghana’ (2013) 45 UCC Law Journal 77. 21 The desire to create an electronic registry was one of the main reasons that at first generated an interest in and then became a driving force behind a reform in New Zealand: D Brown, ‘The Personal Property Securities Act 1999’ in J De Lacy (ed), The Reform of UK Personal Property Security Law, Comparative Perspectives (Routledge Cavendish, 2010) 328, 330. 22 See Love, Pería and Singh (n 13). 23 Such investments may be relatively minimal, since the software often is procured with funding provided by a donor agency. In addition, the host agency may already have the hardware necessary to run a registry software application. By making their substantive laws parallel to that of other countries, developing countries can save costs that would otherwise be necessary to develop the entire registry software application from ‘scratch’, instead paying only for customisation of an existing software application. 24 RA Macdonald, ‘Article 9 Norm Entrepreneurship’ (2006) 43 Canadian Business Law Journal 240, 267. 25 Love, Pería and Singh (n 13) 12–14.

Process of Secured Transactions Law Reform

163

in d eveloping countries. Sometimes the experience and statistical data from the registries established in neighbouring countries induce these policymakers into taking some action, as was the case in Colombia that looked at the experiences in Honduras. The policymakers in these countries call for the establishment of a registry, whether or not the political landscape is receptive to substantive law reform, and the recipients of these calls are expected to take some action.

IV. The Quixotic Search for Brevity Given the comprehensive scope of a modern secured transactions law, its purpose cannot be accomplished in a few short articles and it is unlikely to be possible to present it in a classic Civil Code drafting style.26 Whatever virtue brevity may have in the abstract, brevity achieved by omitting rules that provide certainty in advance and that announce solutions to issues that otherwise would have to be provided by courts after litigation, will fail to allow secured creditors to properly assess the risks being undertaken and will prevent counsel from giving definitive advice and unqualified opinions. The secured transactions law has a very wide scope, covering many types of transactions, involving many types of tangible and intangible assets and affecting many types of competing claimants whose priorities must be specified. A few terse provisions simply cannot accomplish this task effectively and consistently within the goals of the law.27 Three examples of misguided efforts to achieve brevity (or the appearance thereof) in the UNCITRAL Model Law are: (1) the separate presentation28 of the

26 ‘Style’ means far more than fashion or technique. See Professor Macdonald’s penetrating discussion of this point in R Macdonald ‘A Model Law on Secured Transactions. A Representation of Structure? An Object of Idealized Imitation? A Type, Template or Design?’ (2010) XV Uniform Law Review 429–30. 27 An example of an extreme model is the Australian Personal Property Securities Act of 2009, which contains 343 sections. A model secured transactions law for the European Union—Book IX of the Draft Common Frame of Reference—contains 33 articles governing operations of the registry. 28 The registration provisions of the UNCITRAL Model Law are presented in a manner that invites enacting States to incorporate them separately from the other Chapters of the Model Law. At its 25th session in New York, 31 March—4 April 2014, Working Group VI decided to allocate registry provisions between the draft Model Law and an annex which were to contain those provisions that cover ‘technical registration-related issues’. Report of Working Group VI (Security Interests) on the work of its 25th session New York, 31 March–4 April 2014 (A/CN.9/802) paras 12–14. At its 27th session in New York, 20–24 April 2015, Working Group VI decided the registration-related provisions in the draft Model Law and the Annex should be reflected all together as a whole in the Annex, while Chapter IV could be reduced to a provision stating that a registry was established and enacting States should implement the registry-related provisions in the Annex’. Report of Working Group VI (Security Interests) on the work of its 27th session New York, 20–24 April 2015 (A/CN.9/836) para 120. The Committee of the Whole, established by the Commission, noted that ‘the registry-related provisions set forth in the draft Registry Act might be implemented in the law enacting the draft Model Law, a separate act, decree or regulation, or a combination thereof. On the understanding that the name and the location of the draft Registry Act would be discussed after the Committee had discussed all the registry-related provisions’.

164

Marek Dubovec and Harry C Sigman

registration provisions from the rest of the UNCITRAL Model Law with its own internal numbering (likely, in the authors’ view, to lead in a reform process to separate preparation and lack of coordination with the substantive law); (2) the absence of rules regulating serial-numbered goods (see Section VIII H below); and (3) the decision not to incorporate a proposed statutory elaboration of the collateral description provisions that would have answered questions that are instead left open. Article 11(1) of the Model Registry Provisions of Chapter IV of the UNCITRAL Model Law provides ‘The encumbered assets must be described in an initial or amendment notice in a manner that reasonably allows their identification.’ This rule, without further guidance, is too open-ended and imprecise to assure a prospective secured creditor that it has satisfied the standard, and leaves far too much room for judicial interpretation. In the early days of UCC Article 9, which provided a similarly-worded rule, courts in the US reached disparate conclusions— inconsistent with the underlying objective of a notice-filing system—holding, for instance, that descriptions such as ‘all farm equipment’ failed to identify the collateral.29 Additional provisions added in subsequent revisions of UCC Article 9 have helped to ameliorate this issue.30 There is now little doubt that the description in the notice (called a ‘financing statement’ in UCC Article 9) must only suffice to warn a searcher to make further inquiry and must embrace (not exclude) the agreed-upon asset. Reformers should consider adding the following provision to the substantive law: ‘(a) the description in the notice need not be identical to that in the security agreement; (b) to the extent that the description in the notice exceeds the description in the security agreement, the notice does not make effective against third parties a security right in such assets; (c) reference to an asset in a notice does not imply or represent that the grantor presently has, or in the future will have, rights in the asset; and (d) a description by type, category, quantity or computational formula meets the description standard.’

This brief addition would provide certainty in advance and ab initio announce solutions to issues that otherwise would have to be decided by courts after The Committee of the Whole did not get an opportunity to discuss the name and location of the draft Registry Act: Report of UNCITRAL, 48th Session (29 June–16 July 2015) Doc. A/70/17 para 169. 29 JC Miller, ‘Farm Collateral Under the UCC: “Those Are Mighty Tall Silos, Ain’t They Fella?”’ (1975) 20 South Dakota Law Review 514, 520. 30 UCC § 9-108(b) currently provides examples of reasonable identification, including description by specific listing, category, type defined in the UCC, quantity, computational or allocational formula or procedure, and any other method, if the identity of the collateral is objectively determinable. While an ‘all of the debtor’s assets’ super generic description is not a sufficient collateral description in a security agreement, such super generic indication of the collateral is sufficient in a financing statement under UCC, § 9-504; this was a political decision based on the assertion by certain consumer advocates that a consumer should not be expected to comprehend the meaning of ‘all.’ Thus, it is clear under UCC Art 9 that collateral consisting of a desk may be described as ‘desk’, ‘furniture’, ‘equipment’ (if the grantor holds the desk for use in its office), and need not be specific (eg need not indicate that the desk is made of wood or is of indicated dimensions).

Process of Secured Transactions Law Reform

165

litigation. Indeed, failure to include such elaborations would make more likely the need for amendments to the law in the future. These elaborations should be supplemented with clear instructions to design the registry notices with a field designated for a collateral ‘free text’ description in which the registrant would be able to enter description satisfying the above-mentioned standard rather than requiring the registrant to select one or more checkboxes that identify types and subtypes of collateral, a methodology that has generated a number of concerns.31 Brevity of a model law achieved by placing such elaborations in a supplementary instrument (eg the Guide to Enactment) that does not itself have the force of law is unwise. It is highly unlikely that draftsmen involved in the reform effort and later users of the enacted law (eg lawyers, judges, businesspeople) will have ready access to such supplementary materials, particularly as time goes by. Only an enacted law that can stand on its own legs can be deemed a truly successful reform. Many issues are more comprehensively covered not only in UCC Article 9, but also in the Personal Property Security Acts adopted by the Canadian Provinces, New Zealand and Australia. Recent experience in Africa has revealed that the New Zealand PPSA, even though it contains 201 sections, may be a better starting point for a number of these issues than a ‘brief ’ model law. Acting on the mandate from the Commission to prepare a concise model law, UNCITRAL’s Working Group VI often decided to avoid a topic (eg serial number descriptions of some collateral and related issues) or knowingly gave a topic scant coverage (eg unique numerical identifiers of grantors). A legislator must not act under such a constraint. See the discussion in Section VIII of this Chapter under the heading ‘Omissions From and Other Gaps in the UNCITRAL Model Law.’

V. Allocation of Functions We consider it crucial that the respective roles of law-making, supervision of the registry and the registry itself be clearly delineated. In particular, it is important that the registry not be empowered to make legal or factual determinations and that its role be limited to the purely operational (technological and administrative) execution of rules set by the secured transactions law based on policy determinations made by the legislature. This view is based on the fact that the registry contemplated by the reform discussed in this chapter differs so dramatically from the classic land registries that most countries are familiar with.32 It should also be kept in mind that the substantive law that is the goal of the reform effort is not the

31 For the difficulties caused by the design of the field designated for the collateral description in Australia, see Whittaker (n 10) 169–70. 32 See SV Bazinas, ‘The UNCITRAL Guide on the Implementation of a Security Rights Registry: Part IV in a Great UNCITRAL Saga’ (2014) 46(1) Uniform Commercial Code Law Journal 51, Art 2.

166

Marek Dubovec and Harry C Sigman

simple outgrowth of pre-existing law developed as a natural process indigenous to the enacting country and is instead based on policy goals not necessarily shared (or even understood) by often understaffed local courts. Neither the expression of the rules, nor the determination of their underlying policies, may be left with the registry personnel. The statutory rules must be comprehensive and a governmental authority designated by the legislature (eg a Central Bank or an Economic or Finance Ministry) must be the body that issues any necessary subordinate rules and supervises the conduct of the registry. References to this point are made throughout this chapter to illustrate its importance. Neither the UNCITRAL Model Law nor the draft Guide to Enactment spells out clearly: (1) the legal structure of the registration provisions; (2) the governance of the registration process; and (3) a standard for allocation of subject matter between the substantive law and any administrative regulations. A footnote to the Model Registry Provisions provides: The Model Registry Provisions are intended to take effect simultaneously with the enactment of the Model Law. They are presented as a separate component with their own internal numbering in order to give enacting States flexibility in their implementation. Depending on its drafting conventions, an enacting State may choose to: (a) incorporate all of the Provisions in its enactment of the Model Law as a separate chapter; (b) incorporate all of the Provisions in a separate statute or other type of legal instrument; or (c) incorporate some of the Provisions in its enactment of the Model Law and the rest in a separate statute or other type of legal instrument.

This formulation provides insufficient guidance on the allocation of registration provisions between the secured transactions law and subsidiary legislation and may result in the placement of some essential rules (eg those that affect third-party rights or priorities) in administrative regulations issued by the registrar. We do not think that administrative regulations should determine on what grounds a notice may be rejected, how the grantor must be identified, etc. These matters should be addressed either in a law or some legal instrument issued by a ministry that supervises the operations of the registry.

VI. Interrelationship between Policy Choices, the Statutory Rules Governing the Registration System and the Registry Design In designing a reform, whether comprehensive or piecemeal, the policymakers and legislators should understand the close connection between the substantive law and the provisions on registration, and why the shaping of the registration provisions must be done by the legislature, closely fitted to policy decisions. Particular registration rules will protect or undermine the rights of the secured creditor, and, consequently, will further detract from achievement of the goal of

Process of Secured Transactions Law Reform

167

enhancing the availability and reducing the cost of secured credit. Secured creditors will provide more credit at lower cost if they can rely on a coordinated regime of substantive priority rules and rules governing the conduct of the registry that gives them confidence in the extent and quality of their protection. Decisions on particular registration rules will favour registrants over searchers, or vice versa. Let us explore an example. Consider the situation that arises after a secured creditor has registered a notice of a security right and someone (possibly a mischievous grantor), without the secured creditor’s authorisation, submits a cancellation notice that purports to cancel the registered notice; a third party, then intending to lend to the same grantor against the same collateral, searches the registry. If that search discloses the existence of both registered notices, the registry has done its job by having raised a warning flag, since the later searcher can make inquiry about the significance and effectiveness of the apparent cancellation. On the other hand, if the registry has removed both notices from public view upon registration of the cancellation notice, the searcher, relying on the clean search report, will have a claim that he ought to be able to rely on that fact and have priority over the earlier secured creditor. In this situation, both lenders are blameless and neither could have done anything more to protect itself. This hypothetical situation makes clear the need for a comprehensive rule that answers the question presented and the need for policy determinations to be: (1) made; (2) expressed in the statutory rules governing priorities and those governing the registration system; and (3) implemented in the design of the registry. The legislature should ask questions such as whether more or lower cost credit will be supported by protecting a first-to-register secured creditor who has fully satisfied all applicable legal requirements against the risk that it will be primed by a later competing claimant who searches and finds a registered cancellation notice that has been wrongfully submitted by an unauthorised person or who finds nothing because of the effect of an erasure of the registered notices from the publicly accessible record and claims priority over the first secured creditor, even when the first-to-register secured creditor is unaware of that unauthorised cancellation notice, or, will protecting the later competing claimant induce more total credit? The law should include rules that prevent or at least reduce the frequency of the occurrence of the registration of an unauthorised cancellation and ameliorate its negative impact. One approach to reducing the occurrence of unauthorised amendments and cancelations would be to mandate the registry to include security techniques that would allow only the secured creditor named in the particular registered notice to submit amendments and cancellations thereto. At the Commission Session in July 2015, Article 6 of the draft UNCITRAL Model Law was modified to require the registry to provide security techniques.33 The reformer is then expected

33 See A/CN.9/WG.VI/WP.65/Add.1 (n 1). This provision is contained in Art 5(2) of the UNCITRAL Model Law.

168

Marek Dubovec and Harry C Sigman

to consult paragraph 24 of the draft Guide to Enactment, which provides a commentary on Article 5 of the Model Registry Provisions of the now adopted UNCITRAL Model Law, to find out what those security techniques might entail.34 The draft Guide to Enactment suggests that the registry shall provide for such security techniques, which may include either or both: (1) an exclusive access useraccount setup; (2) a secret code or password associated with the secured creditor identified in a particular registered initial notice; or (3) some other security technique preventing wrongful access to the registry for a submission from an unauthorised person. In our view, the security technique(s) to be adopted should be determined by the legislature and should accompany a provision in the law prohibiting access to the registry for the submission of both an amendment and a cancellation notice unless the registrant satisfies the security technique(s) adopted. Another matter for the legislature to consider in dealing with the hypothetical situation in which a registered notice is amended or cancelled without the secured creditor’s authorisation is the choice, between: (1) a rule that provides for a so-called ‘open drawer’, under which the registry simply accepts and stores (available for public search) all registered notices, and never removes anything until at or after expiration—a rule that provides maximum information to searchers and thus makes impossible the dilemma posed by the hypothetical; and (2) a rule of automatic and immediate deletion by the registry of the registered notice upon registration of a cancellation notice, which generates the risk of harming an innocent later searcher or the innocent earlier secured creditor. An ‘open drawer’ enables the later searcher to discover the full historical contents of the record and protect itself accordingly so that neither person is harmed. Alternatively (though less desirable), a ‘temporary open drawer’ could be effected by means of a rule that postpones the emptying of the ‘drawer’ by the registry until 30 days after registration of the cancellation notice.35 Such a feature, at least temporarily, provides a warning to the later searcher, allowing it to protect itself by making an appropriate inquiry, and gives the earlier registered secured creditor an opportunity to petition the court for the issuance of a protective order precluding the emptying of the drawer, or providing for reinstatement if it establishes entitlement to such an order. An additional protective technique (insufficient by itself), not contemplated in the draft Guide to Enactment, is a rule obliging the registry, immediately upon receipt, to inform the earlier registered secured creditor of all subsequent registrations relating to that registered notice. This technique should be implemented in combination with a security technique described above (eg secret code or exclusive access technique).

34

See A/CN.9/885/Add.1. example, the New Brunswick PPSA Regulations, s 79, provides that ‘On the expiration of thirty days after the discharge (cancellation) of a registration in the Registry, all data relating to that registration may be removed from the records of the Registry.’ 35 For

Process of Secured Transactions Law Reform

169

We think that the ‘open drawer’ approach, coupled with the security techniques mentioned above, is the solution that best implements the underlying policy goals.36 Only a minority of registered notices actually get cancelled,37 and electronic records do not involve the issues presented by paper records, so the argument that registry records would get cluttered if registered notices were to be retained after their cancellation simply does not carry much weight. The operational aspect of the registry would also be simplified as cancellation would not trigger a deletion. The ‘deletion’ approach might be acceptable only if the law mandates all the protective techniques described in the preceding paragraphs. The foregoing illustrates the interplay between priority, third-party effectiveness and rules governing the design and conduct of the registry, as well as the need for all such decisions to reflect policies adopted by the legislature, both of which are best accomplished by a single comprehensive legislative act. The UNCITRAL Model Law and the accompanying draft Guide to Enactment leave the impression that providing for one or another security technique is an administrative matter, while in fact the decision implicates an important policy choice as to which party should be protected and at what cost in complexity of the statutory text and of the software design. The registry’s role should be limited to implementing the policy choices made by the legislature; and in keeping with the goal of making the registry reliable and efficient, the registry should not be permitted to make determinations about the effectiveness or legitimacy of a notice submitted for registration.

VII. The Crucial (but Limited) Functions of a Secured Transactions Registry For the legislature to make the right policy choices, as well as provide a comprehensive legal framework, it must understand the nature of the registry; its functions as well as limitations. Often, the modern efficient registry model described above that is devoid of formalities and onerous requirements and is speedy and inexpensive, differs dramatically from the registries that legal systems are familiar with (eg the classic land registries).38

36 In certain countries, particularly where there are no credit reporting systems, an additional a rgument in support of the ‘open drawer’ is sometimes made—it is asserted that both grantors and secured creditors prefer the registry to maintain cancelled registered notices on record until they expire, so that the fact of prior credits having been paid off is shown on the record. 37 One study of UCC Art 9 filing systems found that about 30% of financing statements are terminated (cancelled) while 55% cease to be effective on their expiration: LM LoPucki and E Warren, Secured Credit, A Systems Approach (7th edn, Aspen 2012) 383. 38 See Bazinas (n 32).

170

Marek Dubovec and Harry C Sigman

The registry model proposed by reformers (and, for the most part described by the Legislative Guide): (1) provides a method for third-party effectiveness;39 (2) plays an important, but not exclusive, role in stating priority rules, especially by providing an objective (date certain, without the need for a notary) and efficient point of reference to establish priorities, particularly rules based on the time of registration; and (3) is also a limited source of information for third parties that contemplate dealing with the grantor’s assets.40 A number of factors determine whether the registry system will efficiently fulfil these purposes. Because the registry system reflects the secured transactions legal framework, if that framework is deficient (eg if it provides for the verification of information by the registrar), the registry itself will be slow, costly and otherwise inefficient. The registry must be designed with certain functionalities and operate in a fashion that corresponds to the needs of the local marketplace, responsive to the typical lending practices and transactions. The registry must take into account the common types of grantors, secured creditors and third parties that search it for information. Especially (but not only) if the typical lenders are less sophisticated financial institutions and microlenders catering to informal SMEs, as in Africa, the law must dictate a registry design that is easily accessible electronically, at a low or no cost and that is searchable according to numerical identifiers rather than common names of grantors. While certain differences among registries exist,41 for a registry to effectively and efficiently fulfil its purposes, both the policymakers and the registry’s future users must adequately understand what the registry does and does not do.42

A. Limited Alert Function of the Registry The essence of the notice-filing system is that it provides only skeletal information, raising a warning flag to alert the searcher about the possible existence of a security right in present or future assets of the grantor.43 It must be kept in mind that the registry is not the sole or even primary source of information about

39 Under UCC Art 9 and the PPSA a security right is effective against third parties once the requirements of attachment have been satisfied. Such third-party effectiveness is subject to the other provisions of these laws; in particular, the priority rules. 40 Legislative Guide (n 1) ch IV, The Purpose of the Registry System. 41 One such difference is the requirement to register notices of enforcement or subordination. The former is characteristic in Latin American countries. These are additional elements that add functions that may be deemed desirable in particular countries, but they are not elements essential for the fundamental purposes of a modern efficient registry. 42 For instance, those searching the International Registry established under the Aircraft Protocol to the Cape Town Convention must be aware that: (i) an asset may be encumbered by a pre-existing interest and thus outside the scope of the Convention under Art 60(1); (ii) an interest might have been given by a debtor who was not located in a Contracting State at the time of the agreement, etc: Goode (n 12) 79. 43 Registration essentially gives a clue that the searcher should inquire further: DG Baird, ‘Notice Filing and the Problem of Ostensible Ownership’ (1983) 12 Journal of Legal Studies 53, 60.

Process of Secured Transactions Law Reform

171

the grantor’s assets or obligations—this is obtained in the course of due diligence performed by the prospective credit provider from the financial information received from the grantor, loan applications, past tax returns, etc.44 Indeed, the presence of a registered notice usually serves merely to corroborate information from other sources. Often, this limited informational function is misunderstood and the policymakers in developing countries present a laundry list of information they would like a search of the registry to reveal, including such items as the estimated value of the collateral, the amount of the outstanding secured obligation, the day of execution of the security agreement, a detailed description of every asset over which a security right has been taken, etc. These requests are made in the mistaken belief that an interested party should be able, based solely on public record information, to form an opinion as to whether to take a security right or buy some assets described in the registered notice. This approach would result in imposing an undue burden on those who register, affect the efficiency of the registry, greatly increase the likelihood of subsequent litigation challenging the accuracy of these data elements, hinder the use of modern financing techniques such as the ‘floating lien’ to support ‘revolving credit’, increase the cost of credit for grantors and would fail, in any event, to provide reliable information to s earchers.45 Being limited in terms of the amount of information it provides, ‘a notice-filing system remains cost-efficient and conscious of protecting some confidential information relating to a secured transaction.’46

B. Registration as a Partial Determinant of Priority Registration plays a very important, but far from exclusive, role in the priority scheme of a secured transactions law (and when it does, it is the time of registration not the time of creation of the security right that provides the relevant time for priorities).47 But registration cannot be the sole determinant of priorities.48

44 SV Bazinas, ‘Intellectual Property Financing under the Supplement to the UNCITRAL Legislative Guide on Secured Transactions’ (2011) 43(2) Uniform Commercial Code Law Journal, Art 1 writes that ‘… who is the owner of an encumbered asset … is not a matter of secured transactions law but of the relevant property law …’ 45 For instance, the real value of most movables today will most likely differ from its estimated value six months earlier, the amount of the outstanding secured obligation is likewise a variable in many typical financing transactions, etc. 46 R Goode, Convention on International Interests in Mobile Equipment and Luxembourg Protocol Thereto on Matters Specific to Railway Rolling Stock, Official Commentary (2nd edn, UNIDROIT, 2014) 271. 47 In this respect, IFC and ADB highlight that the date of registration indicates the date by which competing claims to collateral are measured: Asian Development Bank, ‘Unlocking Finance for Growth, Secured Transactions Reform in pacific Island Economies’ (2014) 6. 48 One of the drafters of the original UCC Art 9 said the following: ‘Registry is not [the entire] priority system—Can a lender, for example, claim that a bank, which has taken cash proceeds in satisfaction of a note, holds such proceeds for the account of the lender because the bank is on “record” notice under a filing statute or because the bank had knowledge of the lender’s claim? Certainly record notice

172

Marek Dubovec and Harry C Sigman

A typical modern secured transactions law includes additional methods of thirdparty effectiveness that secured creditors may utilise, namely: (1) possession; (2) control; (3) temporary; and (4) automatic.49 Often, policymakers ask how these alternative methods compare with registration in terms of providing transparent and easily accessible public notice. Is placing collateral in a field warehouse or executing a control agreement public enough?50 A handful of enacted and proposed laws have designated registration to be the exclusive form of third-party effectiveness, resulting in the registry being the sole determinant of priorities.51 These laws are typically a reaction to a concern that a transparent and public method of registration compares favourably with a relatively non-transparent method such as constructive possession and control that some feel could easily be m anipulated.52 Of course, this type of framework should not even be considered if the country will operate an expensive and cumbersome-to-use registry. In any event, a registration-only approach is arguably appropriate primarily for relatively undeveloped economies with rudimentary types of secured transactions that primarily involve loans secured with household assets of micro-borrowers, farm products of family farmers and used equipment of SMEs. In contrast, if the typical transactions involve imports of goods subject to letters of credit for which a negotiable document functions as collateral or financial transactions with cash collateral, providing for the registry to be the sole method of third-party effectiveness may have a negative impact on their efficiency. Nonetheless, even in the systems with multiple methods to achieve third-party effectiveness, searching the registry will provide the kind of practical assurances that cannot be obtained without it.53

ought not to defeat the bank’s claim and it would seem that the expectation that a person in business will pay bills with proceeds of the business is so strong that a recipient of cash ought not to be defeated even if he knows of the lien’: A Dunham, ‘Inventory and Accounts Receivable Financing’ (1949) 62 Harvard Law Review 588, 605–606. 49

See ch III of the UNCITRAL Model Law (n 1). of these methods of third-party effectiveness have been incorporated in modern secured transactions statutes, and from there found their way into the UNCITRAL instruments. It is noteworthy that the 1952 Official Draft of UCC Art 9 required registration with respect to goods under a field warehousing or similar arrangement because in many situations true possession by the field warehouseman may be questionable. For instance, ‘one of the debtor’s employees is put on the warehouseman’s payroll as custodian; the debtor must have considerable dominion in order to process the collateral, and there is always a danger that the custodian may take orders from his former employer, the debtor, instead of from his present “employer,” the warehouseman’: PF Coogan, ‘Public Notice under the Uniform Commercial Code and Other Recent Chattel Security Laws, Including “Notice Filing”’ (1962) 47 Iowa Law Review 309–10. 51 For instance, s 10 of the Sierra Leone Borrowers and Lenders Act (2014) recognises registration as the exclusive method of third-party effectiveness. Inevitably, exceptions from this rule must be provided to protect buyers, lessees, purchasers and transferees of collateral. 52 ‘If the central purpose of perfection is providing public notice to third parties, it would seem more sensible to require filing in all cases and not to allow possession as an alternative means of perfecting a security interest’: RA Zadek, ‘The Uniform Commercial Code’s Misplaced Emphasis on Possession’ (1994) 38 Loyola Law Review 391, 399. 53 Coogan (n 50) 338. 50 Both

Process of Secured Transactions Law Reform

173

Another point that must be recognised is that a registry search result speaks as of a certain moment. The searcher must keep in mind that due to provisions in the law for grace periods, a notice registered after another notice but within an allowed grace period may nevertheless have priority. Such rules are typically necessary to reflect commercial practices and the nature of transactions. One may find numerous grace periods in the UNCITRAL Model Law, including those that allow the secured creditor to: (1) register a notice with respect to an acquisition security right in equipment a number of days after delivery of such equipment to the g rantor; (2) release negotiable documents of title, such as bills of lading, to the grantor for a specific purpose; or (3) amend the registered notice to add a new identifier of the grantor.54 The policymakers should, among other considerations,55 take the registry design into account when making a decision on the length of those grace periods. While 30 days may be appropriate when the registry is paper-based or otherwise not easily accessible, a shorter period (eg 10 days) may be justifiable when the registry is electronic and open for business 24/7. Different policy considerations come into play in the various situations; whether there should be a grace period in a particular situation and how long it should be is a matter of balancing the competing interests in the context of the particular country. The point here is that a ‘clean search report’ from the registry is not an assurance that an asset offered as collateral has not been previously encumbered.

VIII. Omissions from and Gaps in the UNCITRAL Model Law The UNCITRAL Model Law comprehensively addresses certain aspects of secured transactions but in some instances it does not provide sufficient guidance to draft a law or design a registry. As Professor Macdonald noted, by presenting certain rules in a model instrument while omitting others, the instrument reflects a particular policy choice as to what really matters.56 This section identifies certain gaps and areas in which there is a need for further guidance, including unique identification numbers of grantors, serial-numbered collateral, the potential for coverage of non-consensual liens and security rights in fixtures. The issue of registry fees is discussed separately in Section IX below. 54 See, for instance, Art 38(1)(b) under Option A that sets forth the priority rules for competitions between acquisition and non-acquisition security rights. 55 Although the grace periods are justified by the particular circumstances, it must be kept in mind that they do impose risk during their duration of the existence of a secret-but-priority security right, so a careful balance must be struck by the legislature. Sometimes, lenders go forward with their own registrations but delay actual funding of the loan until after the expiration of the applicable grace period and a search reveals no competing registration. 56 Macdonald (n 26) 426.

174

Marek Dubovec and Harry C Sigman

With respect to some of these issues, the draft Guide to Enactment attempts to provide further guidance. Such guidance in the form of an explanation is most likely to be insufficient for the local policymakers and legislators to actually craft adequate statutory text. Furthermore, coverage of these issues in a guide that accompanies the UNCITRAL Model Law is not a substitute for a clear and thorough model statutory text; while the Guide to Enactment may be the proper place for an explanation of a rule in the Model Law, it is not the proper place for additional or more detailed rules.57 In light of the narrow mandate given to Working Group VI to prepare a concise model law and a guide to enactment, the current draft Guide to Enactment contains cross-references to other instruments on security rights adopted previously by UNCITRAL. This presentation increases the burden on the future implementers of the UNCITRAL Model Law, who will have to navigate through the Model Law itself (having first mastered the Legislative Guide), its related Guide to Enactment, and the UNCITRAL Guide on the Implementation of a Security Rights Registry (Registry Guide) to find an answer to their questions or realise that none of these instruments sufficiently covers the subject-matter (eg actual provisions specially governing serial-numbered goods that they wish to include in their domestic secured transactions laws).

A. Grantor Numerical Identifiers The most important element of notice is the grantor identifier, since that is the basis for the indexing of the public record and thus is the key element for discoverability of the registered notice. It is, therefore, an element that requires especially great care in the reform process and it is also the one least likely to be susceptible of a single specific solution as it is likely to vary from country to country.58 We have in mind the use of numbers rather than names whenever possible, as this can strengthen the reliability of the registry and, at the same time, reduce the risk of errors rendering registered notices ineffective and reduce the likelihood of litigation.59 The UNCITRAL Model Law’s formulation of ‘grantor’s identifier’

57 ‘Many countries lack the capacity to prepare an effective and efficient model secured transactions law without guidance. This is no less true with respect to the law and the regulations relating to the registry’: HC Sigman, ‘Some Thoughts about Registration with Respect to Security Rights in Movables’ (2010) XV(2) Uniform Law Review 508. 58 Interestingly enough, these considerations appear not to have featured in the shaping of the Australian approach with respect to registrations and searches against serial-numbered consumer goods, which instead was motivated primarily by privacy concerns. The objective was to keep an individual grantor’s personal details off the registry unless absolutely necessary. The Australian approach can also be seen as a continuation of the approach taken in the pre-PPSA state laws governing registration of security rights in motor vehicles which provided for registration and search against the serial number only: A Duggan, ‘The Australian PPSA From A Canadian Perspective: Some Comparative Reflections’ (University of Toronto Law Working Paper Series No 2014-03) 18. 59 See further Sigman (n 57).

Process of Secured Transactions Law Reform

175

in Article 8(1) of the Model Registry Provisions and its definition as a ‘name’ in Article 9 relies on its accompanying draft Guide to Enactment to provide some guidance to legislators but such guidance may be inadequate to actually help fashion statutory text. Selecting a name as the grantor identifier, typically that stated in a specified official document, leaves many questions open and requiring further elaboration, including whether middle names and/or initials must be entered in a notice, the development of search logic and whether to allow variations from the search logic. Furthermore, the draft Guide to Enactment is limited to the implementation of the ‘name identifier’, and the explanations in paragraph 34 only briefly mention the possibility of a unique number to be designated as the grantor identifier for the purposes of indexing and searching the registry record. The law should carefully establish the most efficient identifier for grantors based on an in-country assessment of the availability of various types of identifiers for domestic and foreign grantors, individuals and legal entities. The more reliable (non-variable) and easy to determine and verify the grantor identifiers are, the more reliable the search result will be. The rules for the identification of grantors direct secured creditors to include a particular identifier of the grantor in a notice. Article 9 of the Model Registry Provisions stating that the grantor’s identifier is the ‘name’ that appears in a relevant official document may prove to be quite difficult to implement in many environments.60 The solution in a number of countries has been to require that grantors be identified by an officially issued unique number. For individual grantors, such numbers could be voter identification numbers, national identification numbers, or special biometric codes assigned by a central bank through regulated financial institutions. For entity grantors, unique numbers may be tax identification numbers or company numbers assigned to the company upon its registration in the companies registry. The task of the reformer should be to assess whether a reliable identifier for grantors, other than names, actually exists. Of course, there may be situations in which a unique numerical identifier exists but does not cover all or a vast majority of potential grantors. In Nigeria, four different unique numerical identifiers exist: (1) voter’s identification number (voter’s ID); (2) tax-payer identification number; (3) national identification number; and (4) bank verification number (BVN). The coverage for these numbers varies, from close to eighty million Nigerian citizens and other eligible voters having been issued with a voter’s ID, to only about two million having a BVN, as of March 2015. However, the political realities and the impending national elections made it impossible to select the voter’s ID as the grantor

60 During one visit to Ghana, one of us learned that the only identification of members of one particular village is the name of the village itself that is tattooed on the person’s forearm. Even if these villagers were required to obtain some official document as a condition of approving a loan, the task of a searcher would be very difficult, having to pore over multiple search results because of the commonality of names.

176

Marek Dubovec and Harry C Sigman

identifier. Careful examination of tax and national IDs also revealed a number of risks, so eventually the drafting group that prepared the secured transactions bill settled on BVN to be the sole identifier of individual grantors. BVNs are biometric-based numbers issued by regulated financial institutions.61 In Honduras, individual grantors may be identified by a unique number which is the: (1) national ID for citizens; (2) unique number assigned by the immigration authority upon the issuance of a residence permit to non-citizen residents; and (3) passport number for foreigners. With respect to entity grantors, corporate or non-profit bodies, their unique numbers are those assigned to them upon registration in the relevant companies or non-profits registries. While both the number and a name of the grantor must be entered in a notice, only the former determines its effectiveness. The risk of error in entering numbers and names for individual grantors who are Honduran citizens is minimised through a connection of the registry system to a database of eligible voters that contains the numbers and names. Accordingly, when a registrant enters a name and a number into a notice, the registry system automatically verifies the existence and accuracy of these two data elements against the voters’ database. In Mexico, all individuals and entities have unique numbers that must be provided in the relevant field (denominated as folio electrónico). Initially, registrations against entities in its secured transactions registry, Registro Único de Garantías Mobiliarias (RUG), could be made only against Mexican companies. Recently, the law changed and the RUG was redesigned to allow registrations against foreign entities. While both the unique number and a name of the grantor must be provided in a notice, only the search against a name determines whether or not the registered notice is effective against third parties. However the RUG has access to the database of unique numbers for Mexican entities and individuals so that their accuracy, after they have been entered into a notice by the registrant, may automatically be verified.62 Like in Honduras, this set-up is an effective errorpreventionmechanism that benefits secured creditors and searchers and does not delay the registration process. If the identifier is a name and there is no error-prevention mechanism, such as the ones described above in which the registry has a computerised access to a database that verifies numbers and names of grantors, a significant amount of litigation could be generated, particularly by bankruptcy trustees who may claim that a registration is ineffective because of an error in the name of the grantor that renders it unsearchable. In the name-rather-than-number situation, the law should provide, in a waterfall,63 a list of official documents (eg national identification 61

For more information on BVNs see www.bvn.com.ng, accessed 15 September 2015. registrants and searchers may verify that the individual grantor’s numerical identifier (known as CURP) matches a particular name independently of the RUG at www.consultas.curp.gob. mx/CurpSP/, accessed 21 April 2017. 63 It is crucial that this list is presented in such a manner that one reaches down the list of source documents to a lower category only if a higher category of the source document does not exist as to the particular grantor. In other words, only a single source can be applicable to the grantor. 62 Both

Process of Secured Transactions Law Reform

177

cards) to constitute the source of the correct name. Even this technique may not always produce a precise result and there is likely to be pressure to allow some degree of slippage. One solution is for the law (not a registry-issued regulation) to require an ‘exact match’, allowing no errors in grantor identifiers at all. In the context of a numerical identifier, this is quite reasonable—to require a registrant to proofread the information entered in the designated field before submitting the notice to the registry is not too great a burden, and this approach avoids a great deal of litigation. In the context of a name identifier, however, there is likely to be some pressure to allow a certain margin for error. Here, one rule that is objective and simple to use is to allow, solely, any error that does not prevent discovery by a search under the ‘correct’ name. This can be supplemented by using a search logic that permits certain specified, and publicised, deviations from the exact match (eg disregard of spacing, capitalisation, possibly disregard of certain elements of an entity designation, for example, allowing as a permissible error presentation of ‘corporation’ as ‘corp’). It should be noted that the registry merely publicises the search logic previously determined by the legislature.64 Another, far less satisfactory, solution identified as an option in Article 23 Option B of the Model Registry Provisions, is to validate a name that is ‘closely’ similar to the correct name. This is highly subjective, likely to vary from judge to judge and result in much litigation. Unique identification numbers have many advantages for secured creditors and searchers, but also for the drafters of the substantive law because they may eliminate or at least reduce the need for some substantive rules, such as the one reflected in Article 25 of the Model Registry Provisions that provides for a grace period allowing the secured creditor to amend its registered notice to add the changed identifier of the grantor. Further, the law would not need to specify the effect, as well as priority consequences, of amending or not amending the identifier within a statutory deadline. Since unique identification numbers are far less likely than names to change over time, the burden on secured creditors to monitor for changes in the identifier and consequently register amendments or potentially lose to competing claimants who deal with the grantor after the change in the identifier but before the registration of an amendment is minimised.

B. Serial-numbered Goods Many jurisdictions have included provisions in their secured transactions laws providing for indexing of registered notices according to serial numbers of some

64 Both searchers and registrants must know what the rules that determine the search logic are. Search logic is one of the many examples of how technology (software) can, if not carefully coordinated with policy-making and rules implementing the policy choices made, develop inconsistently from the policy decisions. It is critical for the legislature (not the registry) to first decide policy questions and only then the software developer implement them.

178

Marek Dubovec and Harry C Sigman

categories of encumbered assets, typically motor vehicles.65 Their registries were designed accordingly. An unfortunate omission from the UNCITRAL Model Law is provisions relating to serial-numbered goods. The draft Guide to Enactment mentions serial-number indexing and searching and then refers the reader to the discussion in paragraphs 131–134 of the Registry Guide. The UNCITRAL Model Law, the draft Guide to Enactment and the Registry Guide do not provide any guidance to legislators as to how to make important policy choices and how to actually craft the statutory text governing serial-number indexing/searching. Serial-number indexing/searching is of great importance, as financing of automobiles constitutes a major element of secured financing in both developed and developing countries. Indeed, registrations with respect to motor vehicles not held as inventory in many jurisdictions are likely to constitute 80 per cent or more of all registrations, with the result that this single class of financing transactions essentially supports the registry.66 Although every jurisdiction may require notices to identify motor vehicles with a reliable, legally-recognised manufacturer-applied permanent serial number, beyond that, which other items should receive special treatment will be a local variable and the list should be expanded beyond motor vehicles only with great caution. Coverage of serial-number goods would require a definition, some special provisions for content of the notice, special priority rules and a decision whether description by number will be mandatory or merely an alternative to a verbal description, but in all events would not require very much drafting difficulty or add much complexity. One of the issues that will require a policy choice is whether identification of serial numbers for defined categories of assets is mandatory or permissive; that choice, and its consequences, will then have to be expressed in the substantive secured transactions law. For instance, under the recently enacted Malawi PPSA Section 65, the failure to accurately provide a serial number in a notice will render it ineffective as against all third parties. The New Zealand PPSA follows the same approach. One of the consequences of this approach is that a registered notice

65 The scope of the definition of serial-numbered goods varies. For instance, s 2(1) of the Nigerian Registry Regulations includes an overbroad definition of serial-numbered goods that includes ‘such movable property that have a serial or identification number permanently marked on or attached to its principal part by the manufacturer including motor vehicles, planes, boats, plant and machinery.’ The Australian PPSA includes even intellectual property rights and their licenses within the definition of serial-numbered property. In the majority of the Canadian PPSA jurisdictions, the concept is limited to goods such as motor vehicles, trailers, mobile homes, aircraft, boats and outboard motors, and in Ontario and Yukon, the concept is even narrower, applying only to motor vehicles. The New Zealand PPSA includes motor vehicles and aircraft under its definition of serial-number goods; Whittaker (n 10) 179. 66 Notably the Costa Rican secured transactions law No 9246 (Ley de Garantías Mobiliarias), which took effect 20 May 2015, deviated from the laws and registries implemented in Colombia and Honduras in that it excluded security rights in vehicles from its scope, largely for political reasons. Such security rights continue to be governed by the pre-existing law and the instruments creating pledges and other interests must be registered in the registry of vehicles. Countries should contemplate such exclusions only if the existing laws and registration systems for security rights in vehicles are effective and efficient.

Process of Secured Transactions Law Reform

179

must be amended to add a new serial number whenever the grantor acquires a new asset that requires such description for the security right to become effective against third parties. In Australia, if the serial-numbered property is held as consumer goods, Section 153(1) of the PPSA requires the notice to set out only the serial number without any details identifying the grantor. For serial-numbered property held as equipment, the secured creditor may choose to indicate serial numbers or submit a notice otherwise describing such collateral, but this choice will have an impact on the priority of the security right as against competing claimants. Under Section 44 of the Australian PPSA, a registered notice that does not include serial numbers for equipment that is defined as serial-numbered property is not effective against buyers and lessees, but is effective against prospective secured creditors and execution creditors, including the bankruptcy trustee.67 Irrespective of the policy choice and its implementation in the statute, it is imperative that serial-number descriptions are not required for any asset when it is held as inventory.68 We urge countries to consider inclusion of provisions in their laws governing these aspects of serial-number descriptions in notices. The Australian and Canadian PPSAs provide superior guidance to that which is included in the UNCITRAL instruments.

C. Fixtures Financing secured with attachments to immovable property (fixtures) is not only complex but also greatly different from country to country. The Legislative Guide recognised its importance, but the UNCITRAL Model Law conspicuously has not incorporated any of the Legislative Guide’s recommendations. While it is true that fixture financing may not be susceptible to a one-size-fits-all solution to be presented in a model law, this gap should not leave local policymakers with an impression that no statutory text is needed to facilitate this kind of financing.69 Rather than expressly carving out fixtures from the scope of their secured transactions laws or saying nothing about fixtures, policymakers should look for guidance in the Legislative Guide or modern secured transactions laws that do deal with fixtures financing, such as UCC Article 9 and the Canadian PPSAs. The inclusion of fixtures within the scope of a secured transactions law will necessitate specific third-party effectiveness and priority rules but may also affect the design of the registry, particularly in regards to the requirement for the description of encumbered assets. 67

Duggan and Brown (n 16) 129. The New Zealand PPSA requires serial number descriptions only when a motor vehicle or aircraft is held as consumer goods or equipment: L Widdup, Personal Property Securities Act: A Conceptual Approach (3rd edn, LexisNexis NZ, 2013) 86. 69 The Legislative Guide and many secured transactions laws apply to security rights in fixtures, but a few laws, such as the Australian, Malawian and New Zealand PPSAs completely exclude them from their scope. 68

180

Marek Dubovec and Harry C Sigman

D. Non-consensual Liens Another area in which the UNCITRAL Model Law fails to give guidance is the possibility of using the registry to publicise and order priorities with respect to non-consensual liens (eg judgment, statutory and tax liens). Article 37 of the UNCITRAL Model Law provides a priority rule limited to judgment creditors, leaving the status of other types of liens uncertain. Paragraph 25 of the draft Guide to Enactment, explaining the function and operation of Article 37, simply provides that enacting states may designate registration of a notice as the step necessary for the judgment creditor to acquire rights in the encumbered asset. There is no guidance on how to implement this provision if the country in fact designates registration as such a step. Many questions, including whether a special type of access and notice separate from those utilised for the registration of notices relating to consensual interests would be required, remain unanswered. It is noteworthy that most of the recently established African and Latin American registries do include notices of non-consensual liens.70 The decision as to whether the secured transactions law should apply to third-party effectiveness and priority of non-consensual liens is policy-driven. Furthermore, these policies may differ depending on the type of lien. Policymakers faced with these decisions should take into account that the registration of notices of tax liens: (1) allows the government to protect its fiscal needs efficiently without demanding an absolute priority that would strongly discourage financing; (2) allows ordinary financing practices to take place efficiently; and (3) allows potential secured creditors to more easily determine the existence of competing claims.

IX. Registry Fees71 An issue that is often, at least initially, viewed as an administrative matter that need not be in the secured transactions law and need not be determined by the

70 In the United States, the states have taken a variety of approaches. The California Secretary of State’s Office accepts filings with respect to judgment liens and certain California tax liens. The Federal Tax Lien Act provides for filing of federal tax liens in the UCC records as part of a carefully delineated priority scheme that enables secured creditors to protect themselves from being primed without warning in the context of routine financing relationships. See SL Harris and CW Mooney, Security Interests in Personal Property, Cases, Problems and Materials (5th edn, Foundation Press, 2011) 545–46. 71 Fees are discussed because the topic illustrates many of the points we have stressed throughout the Chapter, particularly the allocation of functions between the legislature and the registry as to who should make policy decisions, the effect of decisions on achievement of the goal of reform, the inadequacy of the guidance provided by the UNCITRAL Model Law, the relation of policy decisions to software programming, etc.

Process of Secured Transactions Law Reform

181

legislature but may be left to the registry is the matter of fees.72 However, the matter of fees is directly relevant to the achievement of the goal of making more secured credit available and at a lower cost, especially for SMEs and low-value transactions. The setting of registry fees implicates important policy decisions. Of course, the initial fee schedule adopted by the legislature may need to be modified from time to time; such modification should nevertheless not be set by the registry (which may have an incentive to collect high fees to be used for a variety of purposes, including to engage in unnecessary activities, hire unnecessary personnel, etc), but rather by a policy-making body like the Ministry of Justice, the Central Bank or some other governmental entity charged with supervising the registry, as authorised in the secured transactions law. Reform of secured transactions law is intended to bring economic benefits to the country and encouragement of registration is an important component of that public purpose. The registry is not simply a private benefit for grantors and secured creditors. Thus, expenditure for the establishment of the registry and support for ongoing operation (the cost of which should be minimal in an all-electronic environment), should be treated as a public investment, not as a new source of fee revenue and not as a service for the benefit of private parties. In all events, any fee charged for any registry activity should not exceed cost recovery. It must be kept in mind that the cost of fees is not borne by the secured creditor but rather is passed on to the grantor. Two instructive examples should be kept in mind: Mexico’s excellent all-electronic registry charges no fees whatsoever for registering or searching. In contrast, the Guatemala and El Salvador registries charge a fee based on the size of the transaction, resulting in such high fees that the registries are hardly used. The UNCITRAL Model Law presents little guidance on the options of charging fees and not charging fees, with further variants of these two options set out in paragraphs 133–135 of the draft Guide to Enactment that relate to Article 33 of the Model Registry Provisions. The authors strongly urge that countries should not impose fees and encourage use of the registry for searching as well as registering, the approach that has, inter alia, the obvious virtues of brevity and simplicity.73 We suggest the following, as an illustrative text, for countries

72 The draft Guide to Enactment (n 1) para 129 states: ‘The Registry Guide, sets forth in an indicative way three options, namely a cost-recovery option, a no-fee or fee-below cost-recovery option and an option leaving fees to be determined in a subsequent instrument’. 73 In Art 33 of the UNCITRAL Model Law, Option A provides for fees for all registry services albeit only at a cost-recovery level (not necessarily easy to calculate, especially in advance) or lower. Adoption of a policy of charging fees has several possible variants. One variant is to limit fees to registration services and to provide that all searching services are free of charge. The advantage of this option is that it will encourage and facilitate the due diligence that potential financiers must do and will reduce risks and the frequency of disputes. Another variant is to limit fees to the registration of an initial notice and to provide that the registration of any subsequent notice and searching services are free of charge. This option has the benefit of ensuring that the enacting State will receive the revenue that it desires to derive at the earliest time. In addition, this variant will remove fee issues from, and thus simplify, all subsequent transactions and searches. Moreover, this variant will encourage registrants to register cancellation notices and relieve grantors from incurring time and expense to initiate proceedings to force

182

Marek Dubovec and Harry C Sigman

that choose not to impose fees: ‘There shall be no fees charged for registration of initial or subsequent notices of any type or for searches of the registry records or transmission or certification of search results’.

X. The Need for Education A crucial element of the reform process is education and practical training. Information about the new substantive law and about the role of the new registry must be dispersed starting at the very outset of the reform process. Intensive education programs must be provided to the small number of individuals (usually consisting of legislative and appropriate policy officials from the finance ministry or the central bank, practitioners with experience and expertise in financing transactions, and possibly academics) who are directly involved in the development of the new secured transactions law. In addition, training programs must be delivered to many future users and those potentially affected by the application of the new law. It is not only large commercial banks that must understand the effect of the new framework, but also ordinary SMEs that buy equipment from one another, as well as individuals who may buy motor vehicles. This educational component should consist of not only formal presentations on the legal and business aspects of secured lending along with practical training on the use of the registry, but also include a media campaign (eg in Liberia, the campaign included large billboards, appearances on radio shows, newspaper articles, brochures, short TV and radio advertisements, etc). Often, the registry software is designed and the registration of cancellation notices. Another variant for States that enact option B or C of Art 14 of the Model Registry Provisions (allowing a registrant to select the period of effectiveness), is to charge fees on a sliding scale, depending on the period selected by the registrant in an initial notice and any amendment notice that extends the period of effectiveness of a registered notice. This approach has the advantage of discouraging registrants from entering an excessive period in a notice (see Registry Guide, para 277). Countries that prefer to charge some fees might consider adopting one of the following variants: (1) there shall be no fees for searches of the registry records or transmission or certification of search results; there shall be fees charged for registrations; (2) there shall be fees for registration of initial notices but no fees for registration of any amendment notices, cancellation notices, searches of the registry records or transmission or certification of search results; (3) there shall be fees for registration of initial and amendment notices but no fees for registration of cancellation notices, searches of the registry records or transmission or certification of search results; or (4) the following fees are payable for the following services of the registry: (a) Registration of a notice (i) submitted electronically $_____ (ii) submitted on paper $_____ (b) Search requests (i) submitted electronically $_____ (ii) submitted on paper $_____ (c) Certificates: (i) provided electronically $_____ (ii) provided on paper $_____

Process of Secured Transactions Law Reform

183

programmed by a foreign company that, after the launch of the registry, is retained for a limited period of time to maintain and upgrade the software. It is essential that these necessary knowledge and technological skills be transferred to the local IT department supporting the registry so that it becomes self-sustainable and is enabled to keep operational costs at a low level.

XI. Conclusion This chapter describes the primary characteristics of a modern efficient registry, highlighting the key role of such a registry in a secured transactions law reform. We then discuss some of the significant issues involved in the reform process, including the drafting of the law itself, again focusing on those matters relating to the registration system, stressing the critical need for clear and comprehensive rules that provide predictability and minimise dependence on subsequent judicial resolution, as well as the need to draw a clear line between the functions of the legislature and those of the registry, the latter being limited to executing the choices made by the legislature. We then analyse whether the UNCITRAL Model Law (even with its draft Guide to Enactment) provides sufficient statutory guidance for a country to adequately transform it into enacted local law. Given the gaps and insufficient guidance on some key issues in these draft instruments, we encourage legislators to learn from the experiences of those countries that have successfully reformed their secured transactions laws and established effective registries, although we urge caution in experimenting.74 Finally, we close by inviting readers to study the cogent analysis (both theoretical and practical) of our esteemed colleague and dear friend, Professor Macdonald, who boldly expressed his conclusions (negative) about the feasibility of crafting a global model secured transactions law.75 We note that he observed that ‘no model law is ever enactable without modification, and of those enacted in an unmodified form, none has ever taken root.’76

74 On the encouragement to experiment see G McCormack, Secured Credit and the Harmonisation of Law (EE 2011) 35. The current review of the Australian PPSA has identified many gaps, inconsistencies and redundancies, and recommended rather significant amendments to the statute. Furthermore, it must be kept in mind that uniformity (even if not precise identity) diminishes the need for resort to conflict of laws rules. 75 Macdonald (n 26) 444. See also the views of Professor Cohen, concluding that, at least at the time of his writing, UNCITRAL should not attempt to draft a model law: NB Cohen, ‘Should UNCITRAL Prepare a Model Law on Secured Transactions’ (2010) XV(2) Uniform Law Review 325. 76 Macdonald (n 26) 444.

184

 

11 Conflict-of-Laws Rules on Security Rights in Non-Intermediaries Securities MICHEL DESCHAMPS* This chapter examines the conflict-of-laws rules that apply to security rights in non-intermediated securities in Canada and the United States of America. On occasion, the chapter also discusses the merits or disadvantages of these rules in comparison with other possible rules, including those proposed by UNCITRAL. Professor Roderick Macdonald passed away before the consideration of the UNCITRAL conflict-of-laws rules for security rights in non-intermediated securities. The legal community would have greatly benefited from his contribution to the work of UNCITRAL on these matters. The first part of this chapter deals with concepts and terminology. The second part is an overview of the role and scope of conflict-of laws rules as well as the policy considerations that should guide legislators in the selection of the most appropriate rules. The third part reviews the conflict-of-laws rules that apply to the creation, effectiveness against third parties and priority of a security right in non-intermediated securities in Canada and the US. The fourth part addresses the conflict-of-laws rules of the same States on the effectiveness against the issuer of a security right in these securities as well as on its enforcement generally. The fifth part analyses the UNCITRAL provisions on the issues dealt with by the Canadian and US rules. On common issues, the conflict-of-laws rules on security rights in nonintermediated securities must be aligned with the rules applicable to consensual outright transfers of these securities (eg, sales). Similarly, whenever possible, the rules on non-intermediated securities must be consistent with those on intermediated securities in order to achieve a level playing field for interested parties. These observations will be taken into account in the various parts of this chapter. * The author is member of the Canadian law firm McCarthy Tétrault LLP and is associate professor at the Law Faculty of the University of Montreal where he teaches banking law. He participates as Canadian delegate in law reform projects in the area of secured transactions sponsored by UNCITRAL and UNIDROIT.

186

Michel Deschamps

I. Concepts and Terminology In general, this chapter uses the terminology proposed by the UNCITRAL Model Law on Secured Transactions. Canadian or US equivalent (or corresponding) terms are often different and are noted. On certain matters, reference is also made to the Geneva Securities Convention (GSC) or the Hague Securities Convention (HSC). With respect to Canada, the relevant rules are essentially found in the Securities Transfer Act (STA) of each the Canadian provinces or territories and in the Personal Property Security Act (PPSA) of the relevant common law province or territory in the Civil Code of Quebec (CCQ). With respect to the US, the relevant rules are essentially found in Articles 8 and 9 of the Uniform Commercial Code in effect in each US state. UCC variations from one US state to another are not relevant for our purposes. The term ‘non-intermediated securities’ is used in contrast with the term ‘intermediated securities’. In common parlance, these terms refer respectively to securities ‘directly held’ with the issuer (for non-intermediated securities) and securities ‘indirectly held’ through an intermediary (for intermediated securities). Referring to directly-held or indirectly‑held securities is not however always accurate, as these expressions do not properly express in all holding systems the legal character of the rights and interests of an ‘indirect holder’ or its relationship with the issuer. The UNCITRAL Model Law defines non-intermediated securities as ‘securities other than securities credited to a securities account and rights in securities resulting from the credit of securities for a securities account’.1 The words that follow ‘other than’ reproduce the definition of ‘intermediated securities’ in the GSC.2 The term ‘securities account’ is defined by the UNCITRAL Model Law and the GSC as ‘an account maintained by the intermediary to which securities may be credited or debited’.3 In the GSC, a person who holds intermediated securities is designated as an account holder.4 The HSC uses the term ‘securities held with an intermediary’ (instead of ‘intermediated securities’) and defines the term as meaning ‘the rights of an account holder resulting from a credit of securities to a securities account’.5 In the Canadian STAs and in the UCC, the term that corresponds to intermediated securities is ‘security entitlement’. It follows from these definitions that non-intermediated securities are securities that are not held in a securities account maintained by a securities intermediary. This statement is not however, sufficient to describe the principal feature of nonintermediated securities, namely that the holder thereof has a direct relationship

1

Art 2(w) of the UNCITRAL Model Law. Art 1 of the GST. Art 2(jj) of the UNCITRAL Model Law and Art 1 of the GSC. 4 Art 1 of the HSC. 5 Art 1 of the GSC. 2 3

Conflict-of-Laws Rules on Security Rights

187

with the issuer. Moreover, the holder of non-intermediated securities generally needs to be registered in the books of the issuer in order to be able to exercise the rights attached to the securities, except for bearer securities. This chapter sometimes uses the term ‘security holder’ to refer to the holder of non-intermediated securities and the term ‘registered holder’ to refer to a security holder who is registered in the books of the issuer. Indeed, the word ‘holder’ does not necessarily entail that the person so designated has physical possession of the securities (or more accurately, of a certificate representing same); obviously, dematerialised non-intermediated securities are not capable of physical appropriation. It is worth noting that a fact pattern involving intermediated securities also generally involves non-intermediated securities, at least at one level. If issuer A issues and registers in its books securities in favour of intermediary B who, in turn, credits the securities (or a portion of them) to a securities account of C, then B will hold non-intermediated securities whereas C will hold intermediated securities. This example shows that securities may be viewed as being at the same time non-intermediated (at the A-B level) and intermediated (at the B-C level), to the extent that under the applicable law account holder C is considered as enjoying a property right in the underlying securities issued by A. On the other hand, if intermediated securities were to be characterised as a mere contractual (or personal) right against the intermediary (namely, a right analogous to the right of a bank’s customer arising from a bank deposit), then in the example the credit of the intermediated securities to the account of C would not be treated as a credit of the same securities as the non-intermediated securities held by intermediary B. As mentioned above, the term ‘security entitlement’ is used in Canada and the US as the term corresponding to intermediated securities. In the French version of the Ontario STA,6 ‘security entitlement’ is rendered by ‘droit intermédié’. Is it worth noting that the Quebec STA7 refers to ‘titre intermédié’ in the French version and to ‘security entitlement’ in the English version. The term ‘security entitlement’ carries in some way the implication that the account holder is not an indirect holder of the underlying securities but has nonetheless a form of property right in same (which could be described as a sui generis property right). The UCC Official Comments describe a security entitlement as a package of personal rights and property interests.8 The Canadian STAs and the UCC do not have one single term that would be the equivalent of non-intermediated securities. However, most of their provisions dealing with ‘securities’ essentially apply to the direct holding system, namely to ‘non-intermediated securities’. Where credited to a securities account, securities are rather described by the Canadian STAs and the UCC as ‘financial assets’.9

6

Securities Transfer Act, 2006, SO 2006, c 8 (hereafter in the footnotes, the OSTA). Act respecting the transfer of securities and the establishment of security entitlements, SQ 2008, c 47. 8 Comment 17 to UCC, § 8–102. 9 See OSTA, s 1(1) and s 95(1); UCC, §§ 8–102(a)(9) and 8–501(b). 7 An

188

Michel Deschamps

The GSC seeks to take a neutral approach on the legal nature of intermediated securities and avoids characterising them as a property right or a contractual right, or a combination of both. Therefore, the GSC does not describe intermediated securities as indirectly-held securities. In addition, the HSC attempts not to take a stand on the characterisation of intermediated securities, which, as mentioned above, are referred to in the HSC as ‘securities held with an intermediary’.10 This expression conveys, nonetheless, the idea that the account holder has an entitlement to those securities and not a mere contractual (or personal) right. It is also implicit from the terms used by the GSC and the HSC that the term ‘directly-held securities’ is not a substitute for non-intermediated securities. Still, the legal relationship between the issuer and the registered holder of non- intermediated securities is a direct relationship. Therefore, this chapter on occasion will refer to non-intermediated securities as ‘directly-held’; a reference to ‘holding directly’ is also found in the Canadian STAs and in the UCC in relation to securities which are not treated as security entitlements.11 Other concepts or terms relevant to this chapter are ‘grantor’, ‘security right’, ‘securities’, ‘certificated securities’, ‘uncertificated securities’, ‘transfer’, and ‘competing claimant’. As just noted, in Canada and the US, non-intermediated securities are generally referred to as ‘securities’; they are either ‘certificated’ or ‘uncertificated’. The UNCITRAL Model Law uses the same terms to describe the two categories of non-intermediated securities. The term ‘grantor’ means, under the UNCITRAL Model Law, a person who grants a security right; that person may be the debtor of the secured obligation or a third party. The CCQ also uses the term grantor, while the PPSAs and the UCC use the term ‘debtor’. ‘Security right’ is defined as follows in Article 2(ll) of the UNCITRAL Model Law: (i) a property right in a movable asset that is created by an agreement to secure payment of other performance of an obligation, regardless of whether the parties have denominated it as a security right, and regardless of the type of asset, the status of the grantor or secured creditor, or the nature of the secured obligation; and

10 An elaboration of this neutral approach is found in the article of C Bernasconi and T Keijser, ‘The Hague and Geneva Securities Convention: a Modern and Global Legal Regime for Intermediated Securities’ (2012) Uniform Law Review 549. See also the discussions on the nature of intermediated securities in the GSC Commentary, para 9–3, in PH Cnac, U Segna and L Thévenoz (eds), Intermediated Securities (Cambridge, Cambridge University Press, 2013), in particular ch 1 (L Thévenoz), ch 2 (P Paech), ch 4 (P Dupont), ch 11 (U Segna) and ch 12 (F Garcimartin), and in T Keijser (ed), Transnational Securities Law (Oxford, Oxford University Press, 2014), in particular ch 5 (E Micheler), ch 8 (CW Mooney and G Morton) and ch 10 (F Garcimartin and F Guillaume). 11 See OSTA, s 95(3) and UCC, § 8–501(d). The Official Comments to UCC Art 8 often make a distinction between a direct and indirect holding system.

Conflict-of-Laws Rules on Security Rights

189

(ii) the right of the transferee in an outright transfer of a receivable by agreement.12 This term broadly corresponds to the concept of security interest under the PPSAs and the UCC and, to some extent, to the concept of ‘hypothec’ under the CCQ.13 For purposes of the analysis in parts III and IV of this chapter of the Canadian and US conflict rules, ‘security right’ must be read as referring to a security interest or a hypothec under the relevant Canadian or US rule. The Canadian STAs define ‘security’ (or in the plural, ‘securities’) as follows:14 ‘security’ means an obligation of an issuer or a share, participation or other interest in an issuer or in property or an enterprise of an issuer: (a) that is represented by a security certificate in bearer form or registered form, or the transfer of which may be registered on books maintained for that purpose by or on behalf of the issuer, (b) that is one of a class or series, or by its terms is divisible into a class or series, of shares, participations, interests or obligations, and (c) that, (i) is, or is of a type, dealt in or traded on securities exchanges or securities markets, or (ii) is a medium for investment and by its terms expressly provides that it is a security for the purposes of this Act;

This definition comes from a similar definition in Article 8 of the UCC. In addition, both the Canadian STAs and the UCC contain additional provisions that qualify the above definition of ‘security’ (by expanding or narrowing its scope). For example, a share or other equity interest issued by a corporation is a security, regardless of whether the requirements of that definition are met. Conversely, an instrument subject to negotiable instruments laws is not a security and is outside the scope of the conflict rules on securities. It should be noted that the UNCITRAL Model Law has a definition of ‘securities which is in part similar to that quoted above but without the qualifications found in the Canadian STAs and the UCC. On the other hand, both the GSC and the HSC do not attempt to define the various categories of rights or interests to be considered as ‘securities’ for the purposes of harmonized substantive or conflict-of-laws rules on intermediated securities. Under Article 1(a) of the GSC, the term generally includes shares, bonds or other financial instruments on assets capable of being credited to a securities account. Certificated securities are securities represented by a certificate issued by the issuer. In that case, possession of the certificate is normally required to permit

12 The second paragraph of the definition is intended to expand the scope of the UNCITRAL Model Law to consensual outright assignments of receivables so as to subject them to the rules governing true security rights. The PPSAs and the UCC are to the same effect. 13 Under the CCQ, a consensual outright assignment of a receivable is not assimilated to a security right for purposes of the application of the CCQ secured transactions regime and the CCQ conflict rules. 14 See OSTA, s 1(1).

190

Michel Deschamps

the security holder to fully exercise the rights conferred by the related securities. Certificated securities may be in registered or bearer form. They are in a registered form where the certificate specifies the name of the person entitled thereto. They are in a bearer form where the certificate states that the bearer thereof is the security holder.15 Uncertificated securities are securities issued in favour of a person who is then registered in the books of the issuer as the holder of the securities, but which are not represented by a certificate. This term is preferred to the term ‘dematerialised securities’, as in common parlance the latter sometimes encompasses intermediated securities. The term ‘transfer’ under the Canadian STAs and the UCC includes both an outright (or absolute) disposition and a transaction whereby a security right or other limited interest is granted in non-intermediated securities.16 The conflictof-laws rules on security rights must in effect take into account that third-party effectiveness or priority issues may involve a dispute between an outright transferee and a secured creditor. The term ‘competing claimant’17 is used in the UNCITRAL Model Law in the context of disputes between a secured creditor and another person claiming rights in the asset subject to the security right of the secured creditor. The other person may be another creditor (secured or not), an outright transferee or an insolvency administrator in insolvency proceedings relating to the grantor. Priority disputes very often arise in the area of secured transactions. Therefore, the discussion on priority issues in part III of this chapter will mainly focus on disputes involving a creditor to whom a security right in securities has been granted.

II. Role and Scope of Conflict Rules and Policy Considerations Section A of this part presents an overview of the role and scope of the conflictof-laws rules in the area of secured transactions. As this chapter on occasion discusses the merits and weaknesses of the conflict rules in effect in Canada and the US, section B below highlights policy considerations for the selection by legislators of the most appropriate rules.

15 For instance, a bond payable to bearer. Many corporate laws prohibit the issuance of shares in bearer form. 16 In the same vein, the term ‘security interest’ is used in the GSC in a broad sense to denote ‘the grant of an interest other than full ownership in … securities for the purpose of securing the performance of … obligations’. The quoted language comes from the definition of ‘security collateral agreement’ in Art 31 GSC. 17 The definition of ‘competing claimant’ is found in Art 2(e) of the UNCITRAL Model Law.

Conflict-of-Laws Rules on Security Rights

191

A. Role and Scope of Conflict Rules Conflict-of-laws rules in relation to security rights determine the State whose substantive law will apply to issues such as the creation, effectiveness against third parties, priority and enforcement of a security right. Accordingly, in order to ascertain the substantive law to apply in a State on any of these issues, one must first look at the conflict rules of that State; these rules will determine whether the law applicable to the issue is the law of that State or the law of another State. For example, if litigation takes place in State A between two creditors claiming priority over the same non-intermediated securities, the courts of State A will use the conflict rules of State A to determine whether the dispute is to be resolved under the law of State A or the law of another State. Ascertaining the applicable substantive law is made through factors that connect the situation to the State whose law will apply. In the above example, if under the conflict-of-laws rules of State A the connecting factor is the place where the certificate representing the securities is physically located and that place is in State B, then the courts of State A will resolve the priority dispute by applying the law of State B. A State with two or more territorial units may have different laws that will be applicable on an issue, depending on the allocation of legislative powers in that State. For instance, in federal States such as Canada and the US, secured transactions laws generally fall under the legislative authority of the Canadian provinces (or territories) or the US states.18 Thus, each province or US state has substantive laws and conflict rules in that area. To accommodate the situations where the conflict rules or the substantive laws are those of a territorial unit of a country (instead of the country as a whole), this chapter uses ‘jurisdiction’ as a neutral term also referring to the territorial unit (eg, a Canadian province or a US state) whose laws may be applicable. In the same vein, under the UNCITRAL Model Law and the CCQ, if a conflict rule refers to a State with two or more territorial units in which the law applicable to the issue falls under the legislative authority of a territorial unit, the reference to the law of that State is construed as a reference to the law of the relevant territorial unit, as determined on the basis of the applicable connecting factor.19 The conflict-of-laws rules on security rights are mandatory and this private international law principle is expressly recognised by the UCC and the UNCITRAL Model Law.20 The parties cannot be permitted by a choice of law clause to avoid the substantive provisions of the legal system to which a conflict rule refers. The reason is that security rights are property or real (in rem) rights and affect

18 For convenience, ‘state’ (lower case) is used to differentiate the United States of the America as a country and the states that form part thereof. 19 See Art 95 of the UNCITRAL Model Law and Art 3077 of the CCQ. 20 UCC, § 1–301(2) and Art 3 of the UNCITRAL Model Law.

192

Michel Deschamps

third parties. Allowing the parties to select the applicable conflict rule would also defeat one of the main purposes of the conflict-of-laws rules, which is to identify the State whose substantive law will apply in the event of a dispute between competing claimants. In a priority dispute between secured creditor X and secured creditor Y, it would be impossible to identify the law applicable to the resolution of the dispute if each of X and Y had been permitted to choose in their security agreement with the grantor a different governing law for the ranking of their respective security right. One would then not know what substantive law to apply. The principle that conflict-of-laws rules for security rights are mandatory is confined to the property aspects of a security right. The law governing the contractual obligations of the parties is determined by the conflict rules on contractual obligations (which rules, subject to certain constraints, generally permit the parties to select the law governing their contractual relationship). The application of the conflict rules relating to security rights should not be conditioned on a prior determination that the situation presents an international element. Whenever a conflict rule refers to the law of a State, that reference should not be refused on the ground of the absence of true ‘internationality’ in the situation.21 Otherwise, courts might disregard a conflict rule enacted by a State by deciding that the case is not sufficiently international on the basis of discretionary criteria that are not part of the conflict-of-laws rules of that State. In other words, if in any particular situation a conflict rule of State A points to the law of State B, it must be presumed that the legislator in State A has considered that the situation of itself presents an international element. In special circumstances where additional criteria would be a prerequisite for the application of a conflict rule of a State, these criteria should be spelled out in the conflict-of-laws rules of that State. Indeed, most legal systems recognise exceptions to the application of their conflict rules such as where a rule would lead to a result manifestly contrary to the public policy of the forum State. The UNCITRAL Model Law is to the same effect but specifies that public policy considerations cannot displace conflict rules on the third-party effectiveness or priority of a security right. These exceptions, however, are not premised on the notion of internationality; moreover, they are rarely relevant to the laws on secured transactions.

B. Policy Considerations The secured transactions substantive law of a State should achieve certainty and permit the parties to a transaction (and the other persons affected by it) to easily predict its results. The achievement of this goal requires that the conflict-of-laws rules of a State also provide certainty in the determination of the applicable law. As explained in section A of this part, certainty would be impossible to attain if

21

See n 20 for the UCC and UNCITRAL Model Law.

Conflict-of-Laws Rules on Security Rights

193

the conflict rules on the property aspects of a security right were not mandatory. Certainty is also better achieved if the doctrine of renvoi is excluded from the conflict-of-laws rules. Under the doctrine of renvoi, a conflict rule of State A pointing to the law of another State refers not only to the substantive law of the other State (eg, State B) but also to the conflict-of-laws rules of that other State (eg, State B). For example, if litigation takes place in State A and if the conflict rules of State A incorporate renvoi and designate the law of State B as the applicable law, renvoi would unduly complexify the analysis and thereby undermine predictability as the conflictof-laws rules of State B might point to the law of a third State (eg, State C). Identifying the applicable law could prove to be an almost insurmountable task if the conflict rules of State C also include renvoi, meaning that the applicable law might be that of a fourth State (eg, State D) and not of State C. Excluding renvoi therefore makes the application of the conflict rules simpler and reduces the transaction costs associated with the investigation of the applicable foreign substantive law. Another objective of a sound conflict-of-laws framework is that the connecting factors should correspond to the reasonable expectations of those who are party to or may be affected by the transaction. The connecting factor that identifies the applicable law should have a factual relation to the situation to be governed by that law. This is the reason why conflict rules for security rights in tangible property generally point to the place where the property is located (lex rei sitae or lex situs), at least for priority issues not involving mobile assets (eg, aircraft). For intangible property, there is, however, no consensus as to the factors best connected with the situation. For example, there is a diversity of approaches as to the factor to be used for the determination of the law governing a security right in a receivable: the place where the receivable is payable, law of the contract under which the receivable arises, law governing the contractual relationship between the grantor and the creditor (eg, law governing the assignment for an assignment by way of security), location of the debtor of the receivable or location of the grantor. There are not as many approaches for a security right in securities but there is still a variety of conflict rules in the different legal systems on the law applicable to a security right in intermediated or non-intermediated securities. Among the considerations relevant to the choice of the most appropriate factors, simplicity in the application of the selected factor will sometimes prevail over a rule more closely connected to the situation; for example, in some States, the statutory head office of the grantor of a security right in a receivable is a connecting factor even if it is not closely connected to the situation. Another consideration to take into account is that the selection of the law applicable to security rights must correspond to the law governing transfers of ownership for security purposes. Security rights and ownership rights or interests are closely related and, in many legal systems, certain security rights are a form of ownership (eg, a mortgage); as well, ownership is often used as a security device (eg, retention of title or leasing). It is also worth noting that the UNCITRAL Model Law, the Canadian PPSAs and the UCC characterise all such security devices as security rights.

194

Michel Deschamps

Quebec does not do so although it is arguable that such characterisation should be used for purposes of the Quebec conflict rules on security rights. Therefore, the scope of the conflict-of-laws rules of a State should be broad enough to encompass not only security rights in the strict sense but also transactions whereby ownership is retained or conveyed for security purposes even if such transactions are not characterised in that State as security rights. This is especially important in the event of a priority contest between a secured creditor in the strict sense (eg, a pledgee) and a person holding an ownership right for security purposes. If both of these rights are not subject to the substantive law of the same State, a court might be unable to resolve the dispute. For the same reason, the conflict rules on security rights should coincide with those governing outright transfers of ownership (that is, absolute transfers not made for security purposes), at least for matters which affect third parties. This approach has been adopted by the UN Convention on the Assignment of Receivables in International Trade (the UNAssignment Convention), the provisions of which apply to outright transfers of receivables (eg sales) as well as security assignments (however designated).22 The conflict rule of the UN Assignment Convention points to one single law for the priority of any type of assignment, with the result that the same law will determine the respective priorities of a creditor holding or security right in a receivable and a purchaser of the same receivable. The same approach is generally found in the conflict rules of the Canadian and US jurisdictions on intermediated and non-intermediated securities.23 Consistency between the conflict rules of a State on security rights and its conflict rules on ownership matters may be described as internal harmonisation. Whenever possible, it is also desirable that the conflict rules of a State in the area of secured transactions be harmonised with those of other States sharing the same economic environment. If a transaction may give rise to the exercise of recourses in several States, a secured creditor seeking full protection will want to have its security right recognised in all such States. For example, if a lender lends money to an investor located in State A on the security of non-intermediated securities issued by an issuer located in State B, the lender will need to inquire about the conflict-of-laws rules of States A and B. This inquiry will permit the lender to ascertain the substantive law to be complied with in each of State A and State B in order for its security right to be effective in both States. If the conflict rule of State A does not point to the same substantive law as the conflict rule of State B for the creation, third-party effectiveness and priority of the lender’s security right, the lender will have to fulfil the requirements of two different laws to ensure recognition of its security in the two States. This would be the case if in State A the connecting factor were to be the location of the grantor (State A) while in State B the connecting factor were to be the location of the issuer and then lead to the application of the 22

Art 2 of the Convention. OSTA, ss 44–46 and OPPSA, s 7.1; UCC, § 8–110 and UCC, § 9–305. Quebec has adopted the same approach. 23 See

Conflict-of-Laws Rules on Security Rights

195

law of State B. Conversely, one single substantive law would apply to the effectiveness and priority of the lender’s security right if both States have the same conflict rule. Harmonising conflict-of-laws rules among States therefore minimises the costs and legal risks associated with cross-border transactions. The view is sometimes expressed that conflict-of-laws rules would be unnecessary if all legal systems had uniform laws in the area of secured transactions. Whether or not such uniformity is a desirable objective or could practically be achieved, it is worth noting that there would still be situations requiring the interested parties to determine the State whose substantive law will govern a transaction. In many States, registration is a condition precedent to the thirdparty effectiveness of most non-possessory security rights. Accordingly, uniform laws would not obviate to the necessity of identifying the State of the relevant registry.

III. Creation, Effectiveness against Third Parties and Priority The term ‘creation’ refers to the requirements to be met in order that property be encumbered or charged by a security right. ‘Validity’ or ‘attachment’ as used in the PPSAs, the CCQ or the UCC are similar concepts. Form requirements prescribed by the law of a jurisdiction for a security agreement to be valid are matters normally falling under the notion of creation. A security right that has been created is effective (or enforceable) as between the grantor and the secured creditor. The term ‘effectiveness against third parties’ refers to a security right that is effective not only between the grantor and the secured creditor but also against a competing claimant. The applicable substantive law may provide that a security right, once created, is automatically effective against third parties (without any further action being required) or that certain requirements must be fulfilled in order for the security right to become effective against a competing claimant. The term ‘priority’ is used in the context of a dispute between a secured creditor and a competing claimant. A security right that has become effective against third parties does not necessarily enjoy a priority status as regards a competing claimant other than an insolvency administrator. Priority is therefore a separate issue. The conflict rules in Canada and US on the creation, third-party effectiveness and priority of a security right in non-intermediated securities are generally different depending on whether the securities are certificated or uncertificated.24

24 On the conflict-of-laws rules in effect in Canada and the US in the area of secured transactions, see generally for Canada RCC Cumming, C Walsh and RJ Wood, Personal Property Security Law (2nd edn, Irwin Law, 2013); for the US, NB Cohen and EE Smith, ‘International Secured Transactions and Revised Article 9’ (1999) 3 Chicago, Kent Law Review 1191.

196

Michel Deschamps

A. Certificated Securities i. General Rule in Canada In Canada (both under the PPSAs and the CCQ), the creation, effectiveness against third parties and priority of a security right in certificated securities is determined by the law of the jurisdiction of the location of the certificate at the relevant time.25 For example, if a security certificate is located in jurisdiction A at the time a security right is intended to be created in the securities represented by the certificate, a court in Canada will apply the law of jurisdiction A to ascertain whether the security right has been validly created between the parties. If, after the creation of the security right, the certificate is moved to jurisdiction B, then the effectiveness and priority of the security right will be governed by the law of the new location of the certificate. The rationale for this rule is that a security certificate is assimilated to a tangible asset and the conflict rules on security rights in tangible assets generally point to the law of the location of the asset (lex situs). As applied to securities, the lex situs could be viewed as objectionable as it allows a secured creditor in possession of the certificate to easily change the applicable law without the consent of the grantor.

ii. General Rules Under the UCC Under the UCC, the third-party effectiveness and priority of a security right in certificated securities are determined as under the Canadian jurisdictions by the law of the location of the certificate. For creation, the location of the certificate is not however a connecting factor. The UCC rather refers to the law governing the agreement whereby the grantor has agreed to grant the security right.26

iii. Exceptions Applying the Law of the Location of the Grantor In Canada and the US, the third party effectiveness of a security right in certificated securities may be achieved by possession of the certificate (with or without endorsement by the grantor) or by a registration or filing in a security right registry. If under the law of the location of the grantor, third-party effectiveness may result from a registration or filing, then the Canadian and US conflict rules will

25

See the Ontario PPSA, s 7.1(1)(a) and the CCQ, Art 3108.8. Official Comment 2 to UCC, § 9–301(1) for the law applicable to attachment and validity (which are UCC notions intended to be captured by the term ‘creation’ as used in this chapter). As attachment is a component of perfection under the UCC, it is, however, arguable that the form requirements to fulfil for attachment to occur are governed by the law governing perfection and not by the law governing the security agreement. The latter would only govern the consensual aspects of creation (eg whether the grantor has effectively agreed to the grant of the security right). However, the better view is probably that attachment is also a matter falling under the law governing the security agreement. 26 See

Conflict-of-Laws Rules on Security Rights

197

apply the law of the location of the grantor to the question of whether a security right in certificated securities has been properly registered (and not the law of the location of the certificate). The location of the grantor is not, however, determined under the same criteria in all jurisdictions. In Canada, there are three approaches (Ontario, the other common law jurisdictions and Quebec). The definition of grantor’s location in the Ontario PPSA was replaced on 1 January 2016 and the new definition is relatively complex.27 Under the new definition, the location of a grantor generally depends on its legal status. Typical transactions which are captured by this new definition most often involve a grantor whose location is determined in accordance with one of the following criteria: (i) if the grantor is a corporation incorporated or a limited partnership organized under the law of a province or territory of Canada or the law of a US state, the grantor is located in that province, territory or US state. For example, if the grantor is incorporated under the Business Corporations Act of Ontario, Ontario is the grantor’s location; (ii) if the grantor is a corporation incorporated under a federal law of Canada, the grantor is located in the jurisdiction where the registered office (ie, the statutory head office) of the grantor is situated. For example, if the grantor is incorporated under the Canada Business Corporations Act (a federal statute) and its registered office is in Alberta, Alberta is the grantor’s location. As pointed out below, the registered office of a corporation may be situated at a place different from that of its place of central administration or chief executive office; (iii) if the grantor is a general partnership (as opposed to a limited partnership) and the partnership agreement states that the agreement is governed by the law of a province or territory of Canada, the grantor is located in that province or territory; (iv) if the grantor is an individual (for example, an individual carrying on business under a trade name), the grantor is located in the jurisdiction where the individual’s principal residence is located; (v) if the grantor is a corporation incorporated outside Canada or the US, then the location of the grantor is the place where the grantor has its chief executive office. Other rules apply in Ontario to determine the location of a grantor whose legal status does not fall under the categories described in (i) to (v) above (for example, a grantor which is a trust). The aim of the Ontario rules is to achieve greater harmonisation with the UCC definition of grantor’s location. There are scenarios however where the Ontario and UCC definitions will not point to the same location, including the frequent scenario where the grantor has been incorporated in Canada. Consider for example a corporation incorporated under a law of the 27

Ontario PPSA, s 7(3).

198

Michel Deschamps

Province of Quebec that has its chief executive office in Ontario. In such case, a court in Ontario would determine that the corporation is located in Quebec while, as explained below, a court in the US would find that the corporation’s location is in Ontario. The Ontario PPSA rules for the determination of the location of the grantor are not in force in the other Canadian common law provinces. In these provinces, the location of a grantor who is a business enterprise is still (as it was before 2016 in Ontario) the jurisdiction where the grantor has its place of business; if the grantor has places of business in different jurisdictions, then the grantor’s location is in the jurisdiction where it has its chief executive office. The UNCITRAL Model Law adopts the same approach as the Canadian common law provinces other than Ontario, but instead of referring to ‘chief executive office’, it refers to the place where the central administration of the grantor is exercised. It is worth noting that the chief executive office of a corporation is a concept different from ‘registered office’. For instance, a corporation incorporated under the Canada Business Corporations Act (a federal statute) may have its registered office (the place designated as such in its corporate documents) in Alberta but its senior officers may be based in and its business may be managed from an office in Toronto. In such case, for Ontario conflict rule purposes, the location of the grantor is Alberta, while for Alberta conflict rule purposes, the location of the grantor is Ontario. In the event of litigation in Ontario, under the new Ontario rules, a court in Ontario will look to Alberta law for matters governed by the law of the location of the grantor (including on priority issues). Prior to 2016, the court would have looked to Ontario law on those matters. On the other hand, a court in Alberta will look to Ontario law on those matters. In Quebec, the location of the grantor is its legal domicile. The domicile of a corporation is defined under Quebec law as the place where the corporation has its legal ‘head office’ (statutory seat), which, under many corporation laws, means the ‘registered’ office of the corporation.28 Under the UCC, the location of a grantor who is a corporation is determined pursuant to a cascade of criteria, the most important of which in the context of this chapter are the following:29 (i) if the corporation is incorporated under the laws of a state of the US, the corporation’s location is in that state; (ii) if the corporation is incorporated outside the US, it is located at its place of business or, if it has more than one place of business, at the place of its chief executive office; (iii) however, criterion (ii) applies only to the extent that the jurisdiction of the place of business (or the place of the chief executive office) is in a jurisdiction whose law generally provides that a non-possessory security right must 28 29

Art 307 of the CCQ. UCC, § 9–307.

Conflict-of-Laws Rules on Security Rights

199

be the subject of a registration or filing in a public registry in order to obtain priority against a ‘lien creditor’ (which under the extended definition of that term includes an insolvency administrator);30 if the law of the relevant jurisdiction does not provide for such a public registry, then the grantor is deemed to be located in the District of Columbia. Consider a corporation incorporated outside the US which has its chief executive office in the Province of Ontario. Presumably, the Ontario registry for security rights is a public registry that satisfies the requirements set out in (iii) for the application of the rule in (ii). Under the UCC, the grantor will then be located in the province of Ontario as its chief executive office is in the province of Ontario. As a result, if litigation occurs in a US state, a court of such state will apply Ontario law to determine if effectiveness against third parties of a security right in certificated securities has been achieved by registration.

B. Uncertificated Securities i. General Rule in Canada and the US For uncertificated securities, the conflict rule in Canada and the US generally points to the law of the issuer’s jurisdiction for the creation (except for the UCC which refers to the law of the security agreement),31 effectiveness against third parties and priority of a security right in uncertificated securities.32 The term ‘issuer’s jurisdiction’ normally refers to the jurisdiction under whose law the issuer has been constituted or formed. If the issuer has been constituted under a law of a Canadian province (or territory) or US state, the issuer’s jurisdiction in that province (or territory) or US state. In Canada, an entity that issues securities may have been constituted or formed under a federal law of Canada on under a law of a Canadian province (or territory). If the issuer has been constituted under provincial law, the issuer’s jurisdiction is that province. As secured transactions laws in Canada are not federal (in general for issues relevant to this chapter), it has been necessary to designate the provincial law that would be treated as the issuer’s jurisdiction where the issuer has been constituted under a federal law; otherwise, it would have been impossible to identify the province (or territory) whose law would be applicable. Therefore, where the issuer is a federally constituted entity, the issuer’s jurisdiction is the province or territory in which ‘the issuer has its registered or head office’.33 Although Canadian statutes are not entirely clear on the issue, ‘registered’ and ‘head office’ seem to have been intended to be equivalent terms and to refer to the place specified in the 30

UCC, § 9–102(a)(52). See n 25. 32 See Ontario PPSA, s 7.1(1)(b) and CCQ, Art 3103.8. 33 See OSTA, s 44(5). 31

200

Michel Deschamps

constitutive documents of the entity as its ‘statutory seat’. Certain federal corporate laws use the term ‘head office’ and others ‘registered office’ to designate that place. In the US, ‘issuer’s jurisdiction’ is usually the jurisdiction under which the issuer of the securities is organised. The UCC definition also addresses through subrules a scenario where the issuer has been organised under a US federal law. The above criteria are easy to apply. However, the definitions of the STAs and the UCC also provide that the issuer’s jurisdiction may be a jurisdiction other than as determined above if so ‘specified by the issuer’ to the extent this is permitted by the law of the constitution of the issuer. The conflict rules are silent on the manner in which (or where) the designation of another law can be made and this creates a certain level of uncertainty. Policy considerations dictate that the designation of a law other than the law of the constitution of the issuer should be made in the instrument providing for the issuance of the securities; it would have been preferable to clarify the relevant rules on this matter. The option to so designate another law is probably intended to accommodate in particular an issuance of debt securities where the parties wish that the governing law be a law other than the law of the State of the constitution of the issuer. For example, the law governing the contractual rights and obligations of the issuer and the holders of the securities could be a law other than the law of the State of the constitution of the issuer; in such case, the parties may want to provide that a security in the securities will be governed by the same law.

ii. Exception Applying the Law of the Location of the Debtor As for certificated securities, whether third-party effectiveness may be (or has been) achieved by registration or filing is a matter for the law of the location of the grantor.

IV. Effectiveness against the Issuer and Enforcement The effectiveness against the issuer of a security right in non-intermediated securities and its enforcement are related issues: a disposition of securities in the course of the enforcement process will often require the involvement of the issuer. For example, if securities are sold by the secured creditor or pursuant to a court order, the purchaser will usually want to become the registered holder of the securities. Indeed, this necessitates an action by the issuer; the law applicable to the effectiveness of a security right against the issuer will determine the conditions to be satisfied in order for the issuer to be bound to register the transfer in its books. Effectiveness against the issuer may also be relevant to the effectiveness against third parties of a security right in non-intermediated securities. For example, in Canada and the US, the secured transactions laws provide that third party

Conflict-of-Laws Rules on Security Rights

201

effectiveness of a security right in uncertificated securities may be achieved by the secured creditor becoming the registered holder of the securities. As well, under the same substantive laws, another mode of achieving third-party effectiveness is to give to the secured creditor control of the securities through a control agreement with the issuer. The law applicable to the effectiveness of the security right against the issuer will determine whether the issuer may be compelled to become a party to such an agreement. This illustrates that certain matters relating to the effectiveness against third parties of a security right may be governed by the law applicable to the effectiveness of the security right against the issuer.

A. Effectiveness against the Issuer In all jurisdictions which are the subject of this Chapter, the conflict rules point to the law of the issuer’s jurisdiction as the law applicable to the effectiveness against the issuer of a security right in non-intermediated securities. The criteria which identify the issuer’s jurisdiction have been described in section III B i above.

B. Enforcement against the Grantor Enforcement refers to the measures or actions that may be taken by a secured creditor to realise on its security right after default under the security agreement. In the context of securities, enforcement will usually lead to a sale of the securities in order that the proceeds of the sale can be applied to the payment of the secured obligations. Depending on the applicable substantive law, enforcement may also include the exercise by the secured creditor of the rights attached to the securities (such as collecting income therefrom or exercising voting rights). The applicable substantive law may in addition empower the secured creditor to become the owner of the property subject to the security right in total or partial satisfaction of the secured obligations. The law applicable to enforcement will determine the remedies available to the secured creditor. In the Canadian and US jurisdictions (with the exception of Quebec), the conflict-of-laws rules refer to the law governing the security agreement as the law applicable to enforcement.34 Quebec does not have any express or implicit conflict rule on enforcement and there is no case law on the matter. It is arguable that a Quebec court should apply the law governing priority as enforcement has an impact on the claims that other creditors may have against the assets charged by the security right. The UNCITRAL Model Law applies that law to enforcement against intangible assets generally, but uses other connecting factors for non-intermediated securities.35 34 35

See Ontario PPSA, s 8(1) and Official Comment 2 to UCC, § 9–301. See section V of this chapter.

202

Michel Deschamps

As the law applicable to enforcement determines the remedies available to the secured creditor, that law will also indicate whether these remedies may be exercised extra-judicially (self-help, in common legal parlance) or through the intervention of the courts. Indeed, conflict rules are rules of the forum and if enforcement takes place in a non-Canadian or non-US jurisdiction, the other jurisdiction will not necessarily recognise the availability of the remedies permitted by the law applicable under the US or Canadian conflict rules. This issue may arise in particular if the security agreement provides for self-help but the asset is in a jurisdiction where the applicable law does not permit self-help. As noted above, enforcement usually gives rise to a transfer of the securities to another person and a transfer requires the intervention of the issuer. The law of the issuer’s jurisdiction will be applicable to the recognition of the transfer by the issuer and such law might be different from the law applicable to enforcement. Finally, procedural matters are outside the scope of the above conflict rules. The forum will apply its own procedural rules.

V. The UNCITRAL Rules The conflict rules of the UNCITRAL Model Law on security rights in nonintermediated securities are essentially found in Article 100, which reads as follows: 1. The law applicable to the creation, effectiveness against third parties, priority and enforcement of a security right in non-intermediated equity securities, as well as to its effectiveness against the issuer, is the law under which the issuer is constituted. 2. The law applicable to the creation, the effectiveness against third parties, the priority and the enforcement of a security right in non-intermediated debt securities, as well as to its effectiveness against the issuer, is the law governing the securities.

As Article 100 establishes one rule for equity securities and another rule for debt securities, each of them will be reviewed separately. At the outset, three observations must be made: (i) first, the distinction between equity securities and debt securities is an approach different from that of the UCC and the Canadian regimes, which rather distinguish between certificated securities and uncertificated securities for the designation of the law applicable to a security right in nonintermediated securities; (ii) second, unlike some other conflict rules of the UNCITRAL Model Law, each paragraph of Article 100 uses one single connecting factor for all issues relevant to a security right in the same type of asset (creation, effectiveness against third parties, priority, enforcement and effectiveness against the obligor—the obligor being here the issuer of the securities);

Conflict-of-Laws Rules on Security Rights

203

(iii) third, Article 100 applies regardless of whether the securities are publicly traded or not. For example, the rule on equity securities is applicable to shares of a ‘private’ company. The general conflict rules of Article 100 of the UNCITRAL Model Law are subject to one exception: Article 98 provides that the law of the location of the grantor applies to the third-party effectiveness by registration of a security right in certificated non-intermediated securities if the law of the grantor’s location recognises registration as a method of achieving third-party effectiveness of the security right. This exception will be discussed after the review of the general rules.

A. General Rule on Equity Securities For equity securities, the general conflict rule of the UNCITRAL Model Law designates the law of the constitution of the issuer for the creation, third-party effectiveness, priority, enforcement and effectiveness against the issuer of a security right in these securities. This solution provides greater certainty in the determination of the applicable law as referring to one single law for all issues avoids the difficulties arising in circumstances where there could be an overlap between some issues (eg, enforcement and effectiveness against the issuer) and where the conflict rules do not point to the same law for all such issues. Two questions must be examined: the concept of equity securities and the notion of the law of the constitution of the issuer (in a shorthand, the issuer’s law).

i. The Concept of Equity Securities Although the term equity is not defined in the UNCITRAL Model Law, it should be understood as referring to participation rights or interests in the capital of the issuer. For a corporation or a similar legal person, equity securities consist of the shares in its capital (eg, capital stock for corporate laws using this term). Similarly, for an entity which is not a legal person under its constitutive law (such as a general or limited partnership in many jurisdictions), equity securities should refer to the rights or interests of the persons (eg, the partners) who are entitled to receive upon the liquidation of the entity the residual value of its assets after payment of its liabilities. It must be noted that shares commonly referred to as preferred shares should be considered as equity securities even if under accounting rules they are classified as liabilities. For example, accounting rules could treat as liabilities shares carrying a preferential right to dividends and redeemable at the option of the shareholder after a certain period of time. Under most corporate laws, these shares are still part of the capital of the corporation. The test for the distinction between equity and debt securities should be based on their corporate law characterisation and not on accounting rules. Likewise, subordinated debt securities (eg, debt payable in insolvency only after satisfaction of obligations owing to certain creditors such as

204

Michel Deschamps

lenders or trade creditors) should be treated as debt securities even if subordinated debt may be viewed as equity from the perspective of lenders making credit available to the issuer.

ii. The Law of the Constitution of the Issuer The law of the constitution of the issuer is the law under which it has been formed. For a corporation, this is easy to be ascertain (eg, the law under which it has been incorporated). For a partnership, this should be the law under which the partnership has been created. In federal States where the issuer may be constituted either under a federal law or a law of one on its territorial units, the UNCITRAL Model Law does not provide specific guidance on the relevant applicable law in a scenario where the issuer’s law is a federal law and the law on secured transactions is that of a territorial unit. As mentioned in section III B i of this chapter, the UCC and the Canadian regimes have sub-rules on this matter. However, under Article 95 of the UNCITRAL Model Law, the internal conflict rules of the federal State (or of the territorial unit which is the forum) will determine the territorial unit’s law to be applicable to the issues falling under Article 100 where some of these issues are not dealt with by the federal law of the constitution of the issuer. A State enacting the UNCITRAL Model Law should then ensure it has (or will adopt) internal conflict rules to that effect.

B. General Rule on Debt Securities For non-intermediated debt securities, the general rule of the UNCITRAL Model Law applies the law governing the securities for the creation, third-party effectiveness, priority, enforcement and effectiveness against the issuer of a security right in these securities. As pointed out in the introductory paragraph of section V A, greater certainty is achieved by designating one single applicable for all such issues. The concepts of debt securities and the law governing the securities will now be examined.

i. The Concept of Debt Securities The term debt securities is not defined in the UNCITRAL Model Law. The word debt is however well understood in most legal systems and denotes a payment obligation. In the context of debt securities, the obligation is generally to make payment of a sum of money. Bonds, debentures and promissory notes are debt securities, to the extent they come under the definition of securities in Article 2(ii) of the UNCITRAL Model Law. The obligation of a borrower to a lender under a credit facility would not qualify as a debt security: it is not captured by that definition. Such an obligation is rather a receivable and a security right therein is subject to the conflict rules on receivables.

Conflict-of-Laws Rules on Security Rights

205

The concept of debt securities raises two questions: the characterisation of convertible debt securities and the law applicable to a security right in these securities. Convertible debt securities are debt securities that are convertible into equity securities at the option of the security holder or the issuer or upon the occurrence of a specified event. Where the conversion is at the option of one of the parties, the option may usually be exercised only after the expiry of a pre-determined timeperiod (eg, five years after the issuance of the securities). Other events may trigger a conversion including an event the occurrence of which is outside the control of the parties. For example, the ‘bail-in’ legislation now in effect in many States (such as a number of member States of the European Union) permits supervisory bodies for financial institutions to convert into equity certain categories of obligations (including debt securities) of an institution whose level of solvency no longer meets required standards. Convertible debt securities should be characterised as debt securities because they constitute payment obligations as long as they are not converted into equity. This means that upon their issuance and until conversion, the contractual law governing these securities will be the law applicable to the creation, third-party effectiveness, priority, enforcement and effectiveness against the issuer of a security right in such securities. What should be the characterisation of convertible debt securities for conflictof-law purposes if and once they are converted into equity? Logic dictates that the connecting factor become the law of the incorporation or formation of the issuer. Therefore, upon being converted into equity, the law applicable to a security right in convertible debt securities will be the law of the State under which the issuer has been constituted. A consequence of the change in the conflict rule (change from the law governing the securities to the issuer’s law) is that a security right in debt securities made effective against third parties under the law governing the securities might become ineffective against third parties after the change. In any event, a prudent secured creditor will make its security right effective against third parties under both the law governing the securities and the issuer’s law if the two laws are different. The UNCITRAL Model Law addresses in Article 23 the impact of a change in the applicable law or in Article 91 a change in the connecting factor (change in the location of the encumbered asset or the grantor). These Articles are not, however, strictly applicable to a change in the nature of non-intermediated securities. A sophisticated legislator might want to draw from Articles 23 and 91 and adopt rules dealing with the change on the basis of principles similar to those that underlie Articles 23 and 91.

ii. The Law Governing the Securities The law governing the securities is the law selected by the parties as the law governing their contractual rights and obligations. In the absence of a choice of law (which is extremely rare for debt securities), the forum will determine the

206

Michel Deschamps

applicable law under its own conflict rules. The UNCITRAL Model Law does not deal with the question of whether the parties may select a governing law which has no connection with the issuance of the securities. This matter is left to the conflict rules on contractual obligations of the forum State. There could be an inconsistency between the law governing the securities and the corporate law provisions of the issuer’s law on the question of whether the issuer should be compelled to give effect of a transfer by a secured creditor in the course of an enforcement process. If this scenario is not addressed in the instrument providing for the issue of the securities, the parties will likely be presumed to have implicitly referred to the issuer’s corporate law on the question.

C. The Law of the Location of the Grantor Article 98 of the UNCITRAL Model Law makes an exception to the conflict rules reviewed in section V A and B above. Article 98 reads as follows: If the law of the State in which the grantor is located recognizes registration of a notice as a method for achieving effectiveness against third parties of a security right in a negotiable instrument, negotiable document, right to payment of funds credited to a bank account or certificated non-intermediated securities, the law of that State also is the law applicable to the third-party effectiveness of the security right in that asset by registration.

The purpose of Article 98 is to allow a secured creditor to rely on the law of the location of the grantor to make its security right effective in a wide range of assets by registration under the law of grantor’s location, even if for some these assets the applicable law is otherwise different. In many cases, achieving third-party effectiveness by registration would only yield to a lower ranking priority in the case of dispute with a competing creditor having made its security right effective against their parties by another method; however, in insolvency, the secured creditor would still be entitled to set up its security right against the insolvency administrator. It should be emphasised that with respect to non-intermediated securities, Article 98 only captures certificated securities. Uncertificated securities are outside the scope of Article 98 and, therefore, the issuer’s law (and not the grantor’s location law) will be the law applicable to the third-party effectiveness by registration (if permitted by the issuer’s law) of a security right in uncertificated securities. The rule of Article 98 is different in Canada and the US, which refers to the grantor’s location law for both certificated and uncertificated securities with respect to third-party effectiveness by registration. Another feature of Article 98 must be stressed. This Article refers to the grantor’s location law for third-party effectiveness by registration only if that law recognises registration as a method of achieving such effectiveness. If this is not the case, then the general rules of Article 100 will apply.

12 Comparative Approaches to the Enforcement of Secured Credit in Insolvency JOSÉ M GARRIDO* AND EDWIN E SMITH**

I. Introduction This chapter highlights a number of selected issues that arise when a creditor obtains a security right in an encumbered asset, the grantor of the security right is the subject of an insolvency proceeding, and the creditor seeks to enforce its security right. We will address these issues by looking at comparative approaches to resolving them. In this regard, we will refer to international standards in the regulation of secured credit and insolvency, and we will compare the approach followed by different insolvency regimes. There is a wide variety of solutions for the problems examined here, and it would not be possible to be exhaustive in describing them all. The objective of the analysis is to show evolutionary trends and differences in the use of legal techniques in the face of similar problems. When we refer to secured credit, we will be referring to the extension of credit secured by a security right, ie, an interest in property that secures the payment of a credit. When we refer to insolvency, we will be referring to judicial proceedings designed to address the situation of businesses (incorporated or unincorporated) in debt distress, therefore including both bankruptcy liquidations and

* José M Garrido is Professor of Commercial and Corporate Law at the University of Castilla-La Mancha (Spain) and Senior Counsel at the Legal Department of the International Monetary Fund. This paper represents his personal views and research and does not necessarily reflect the views of the IMF, its Executive Board, or IMF management. ** Edwin E Smith is a partner in the New York City and Boston offices of Morgan, Lewis & Bockius LLP. He was a member of the US delegation to the UNCITRAL working group that formulated the UN Convention of the Assignment of Receivables in International Trade and the UNCITRAL Legislative Guide on Secured Transactions.

208

José M Garrido and Edwin E Smith

rehabilitations.1 Both ‘secured credit’ and ‘insolvency’ are understood in line with the usage of these terms in international texts, such as the UNCITRAL Legislative Guide on Secured Transactions2 and the UNCITRAL Legislative Guide on Insolvency Law.3 The issues that we will discuss arise largely from the competing goals of secured credit and insolvency. Generally, secured credit enables a person to borrow funds against the value of its assets. If the grantor is unable to pay the amounts owed under the loan, the secured creditor will be able to look to those assets to satisfy the credit, generally in priority to the claims of other creditors.4 As a result, the secured creditor is often more willing to extend credit to the grantor or to charge lower rates on the credit.5 But the function of secured credit is very different from the function performed by the insolvency system. Insolvency regimes are designed to create collective procedures in which all creditors of the grantor participate with the objective of increasing the returns to all creditors—minimising their losses, and achieving a

1 We exclude the bankruptcy of natural persons from this analysis. Issues of enforcement of secured credit (especially, mortgages over immovables) in the insolvency of natural persons are particularly sensitive. It is arguable that the rules and principles at play are different from those that apply to business insolvency. 2 UNCITRAL Legislative Guide on Secured Transactions (2007), available at www.uncitral.org/pdf/ english/texts/security-lg/e/09-82670_Ebook-Guide_09-04-10English.pdf. For convenience, we have used terms in this paper terms, such as ‘security right’ and ‘encumbered assets’, which are used and further defined in the UNCITRAL Legislative Guide on Secured Transactions. 3 UNCITRAL Legislative Guide on Insolvency Law (2004), available at www.uncitral.org/pdf/ english/texts/insolven/05-80722_Ebook.pdf. The UNCITRAL Legislative Guide has been updated in 2012 and 2013. Other relevant documents on insolvency law include the World Bank’s Principles and Guidelines for Effective Insolvency and Creditor Rights Systems (2001, subsequently revised) available at www.worldbank.org/ifa/ipg_eng.pdf; and the International Monetary Fund’s Orderly and Effective Insolvency Procedures (1999), available at www.imf.org/external/pubs/ft/orderly/. 4 The basic elements of secured credit are, therefore: (1) a credit relationship; (2) encumbered assets, in the form of tangible or intangible assets; (3) a link between the encumbered assets and the credit relationship—the encumbered assets remain connected to the credit relationship; (4) the right of the creditor, under certain circumstances, to enforce its security right by, among other things, seizing the encumbered assets, collecting on encumbered assets consisting of payments rights and selling or otherwise disposing of the encumbered assets; and (5) the satisfaction of the secured creditor with the proceeds of the encumbered assets with priority over other creditors of the grantor. To these characteristics, there is another one that may be added, which is the possibility of seizing the encumbered assets even when the encumbered assets have been transferred to a third party. However, this principle knows some important exceptions in modern secured transactions law (especially regarding certain categories of movable assets, such as inventory). For a classic definition of the principles of secured credit in American law, see G Gilmore, Security Interests in Personal Property (Boston, Little, Brown & Co, 1965) passim. Under an English law perspective, see R Goode, Goode on Legal Problems of Credit and Security, 5th edn (Sweet & Maxwell, 2013). From a civil law perspective, see JM Garrido, Garantías reales, privilegios y par condicio (Madrid, Centro de Estudios Registrales, 1999) 25ff. 5 See CW Calomiris and others, ‘How Collateral Laws Shape Lending and Sectoral Activity’ (February 2016), available at www0.gsb.columbia.edu/faculty/mlarrain/papers/CLLS.pdf. Anecdotal evidence and practical experience suggest that the most important effect of an efficient secured transactions regime is the increase in access to credit (see H Fleissig, M Safavian and N De La Peña, Reforming Collateral Laws to Expand Access to Finance (Washington DC, World Bank, 2006)); the reduction of interest rates is not a necessary consequence in all cases.

Enforcement of Secured Credit

209

better outcome for the creditor group than in the situation where each creditor pursues its own individual rights. Avoiding a destructive race among creditors and imposing cooperation mechanisms represent essential features of the insolvency regime. In the end, insolvency has the ultimate objective of preserving enterprise value for the benefit of creditors, the debtor, third parties and society in general. This means that there is a tension between the objectives of the secured transactions regime and the objectives of the insolvency system. The result of the collision of the goals of secured credit law and insolvency law is a tension between the secured creditor desiring to look to the encumbered assets for an early repayment of the secured claim, and the creditor group, and other stakeholders, desiring to keep the grantor’s assets together with a view to obtaining a better return from an orderly liquidation or a rehabilitation of the grantor’s business. Recognising these competing goals, we will discuss a number of issues relating to the enforcement of secured credit in insolvency proceedings generally, followed by certain issues relating to the treatment of secured credit more specifically in rehabilitation proceedings. Our analysis is based on the texts that establish best international practice in the area of secured transactions and insolvency, and the study of relevant solutions in a number of legal systems. Drawing comparisons between different systems represents an invaluable tool to understand and assess the functions of the legal rules in this area.6

II. The Enforcement of Secured Credit in Insolvency Proceedings Generally It is widely recognised that it is precisely in the case of the insolvency of the grantor where a secured creditor needs most protection.7 Security rights are designed to protect the secured creditor generally against the risk of default of the loan by the debtor, but there is no question that the risk of default materialises with the insolvency of the borrower—or the grantor of the security right, when they are not the same person. Indeed, the insolvency of the grantor is the main event against which the creditor has sought the protection of the security right. The legal techniques by which the security right is protected determine the effectiveness of the security right itself, and also the effectiveness of the insolvency

6 See J Gordley, ‘The Functional Method’ in PG Monateri (ed), Methods of Comparative Law (Edward Elgar, 2012) 107ff. In addition, a functional comparison of rules that confronts specific legal systems tends to be more useful than a discussion around the idea of ‘families of laws’, that eventually needs to be qualified by the consideration of ‘mixed’ systems: see KGC Reid, ‘The Idea of Mixed Legal Systems’ (2003) 78 Tulane Law Review 5. 7 See É Garaud, ‘Sûretés et procédures collectives: l’articulation des réformes’ (2009) 120 Les Petites Affiches 4.

210

José M Garrido and Edwin E Smith

proceedings. In particular, the question of whether the encumbered assets are part of the insolvency estate, and whether the secured creditor is subject to a stay on enforcement actions, stand out as fundamental points that determine the level of protection of secured creditors and the possibilities of keeping the insolvent business as a going concern. The other issues that we have decided to cover are the participation of the secured creditor in the insolvency process, the priority and ranking of the secured creditor’s claim, and the payment of secured loans in a liquidation of the grantor’s estate.

A. Are Encumbered Assets Included or Excluded in the Insolvency Estate? The issue of the inclusion or exclusion of encumbered assets in the insolvency estate is crucial: regimes that exclude encumbered assets from the insolvency estate provide the most intense protection to secured creditors, but do not allow insolvency procedures to fulfil their function of preserving going concern value for the benefit of creditors, debtors, and the economy in general. Coupled with the possibility of creating security rights that cover most or all the productive assets of an enterprise, the exclusion of encumbered assets from the insolvency estate results in credit enforcement systems where insolvency proceedings play a merely residual role. This is because the preferred method of reacting against a debtor’s default under such a regime is by individual enforcement of the secured creditor’s claim, often by appointment of a receiver who takes possession of all the enterprise assets. The international insolvency standard recommends that encumbered assets be included in the insolvency estate to allow a better realisation of the enterprise value of the debtor’s business: Including encumbered assets in the estate and thus limiting the exercise of rights by secured creditors on commencement of proceedings may assist not only in ensuring equal treatment of creditors, but may be crucial to the proceedings where the encumbered asset is essential to the business. (…) There will be advantages, in particular in reorganization and where the business is to be sold as a going concern in liquidation, in having all assets of the debtor available to the insolvency estate from the time of commencement.8

In contrast, not including encumbered assets in the insolvency estate represents the strongest protection for secured creditors. Some legal systems have adopted the solution of leaving secured creditors unaffected by insolvency proceedings, leaving secured creditors free to pursue their individual remedies to enforce their legal and contractual rights. The rationale for excluding encumbered assets from the insolvency estate is premised on the distinction between rights in personam and rights in rem. 8

UNCITRAL Legislative Guide on Insolvency Law (n 3) 77.

Enforcement of Secured Credit

211

In traditional insolvency analysis, the insolvency process represents the forum in which rights in personam—credit rights—are enforced, whereas rights in rem, rights over specific assets, are left unaffected by that process.9 This distinction is evident when comparing the position of the owner of an asset that is in the possession of the insolvent debtor, with the position of the unsecured creditor for services rendered to that same insolvent debtor. The owner of the asset will recover its property, entirely unaffected by the insolvency proceedings; and the unsecured creditor will have to participate in the insolvency proceedings in order to recover its claim, sharing the proceeds of the liquidation with other creditors. In numerous legal systems, the position of the owner of an asset that is in the possession of the debtor at the time of insolvency is not specifically regulated; the owner of the asset will be able to use the same legal remedies as are available in the absence of the insolvency proceeding. However, there are also numerous legal systems, as discussed below, that have instituted a special legal action to be used by a person who claims the ownership of an asset in the possession of the insolvent debtor. The situation is less clear when secured creditors are considered in contrast to owners. In the case of a secured creditor, there is a combination of a right in in personam (the credit) and a right in rem (the security right). The mixed legal nature of secured credit is what explains the existence of different approaches:10 if prevalence is given to the right in rem, the result is that the secured creditor is unaffected by the insolvency process and the encumbered assets are not included in the insolvency estate. It is also possible, as the international standard recommends, that the secured claim is included in the bankruptcy process, ie, as a credit right, and therefore the encumbered assets are also part of the insolvency estate, subject to the ranking of claims over those assets. In the systems where encumbered assets are considered to be unaffected by the commencement of the grantor’s insolvency proceeding on account of the existence of rights in rem over the encumbered assets, it follows logically that a secured creditor is free to enforce its rights over the encumbered assets. The different procedural mechanisms for the exercise of the rights of secured creditors have evolved in a parallel way to those designed for the protection of owners.

9 See R Goode, Goode on Proprietary Rights and Insolvency in Sales Transactions, 3rd edn (Sweet & Maxwell, 2009). 10 The existence of differences between the treatment of an owner and the treatment of a secured creditor also represents a factor in the trend to adopt the use of ownership as security. This is clear in the German and French systems, where reservation of title and other structures based on ownership offer better protection than security rights: see J Stoufflet, ‘L’usage de la propriété à des fins de garantie’ in A Bruyneel et A-M Stranart (eds) Les sûretés (Paris, FEDUCI, 1984) 320; F Barrière, ‘La fiduciesûreté en droit francais’ (2013) 58 McGill Law Journal 869. This also explains the evolution of Brazilian law and practice in Brazil: the use of fiduciary sales (alienacao fiduciaria) is so widespread mainly because in the event of the insolvency of the debtor, the creditor-owner may require the insolvency administrator to turn over to the creditor the possession of the asset. The creditor will have to sell the asset and recover its claim in accordance with the ranking of creditors, but without being subject to the insolvency process. Conditional sales are also used in Brazil, with similar effects.

212

José M Garrido and Edwin E Smith

There are systems—such as many of the traditional common law regimes—in which there is no specific action for the enforcement of the rights of a secured creditor with respect to encumbered assets; secured creditors are free to enforce their rights by using their generally applicable remedies even though an insolvency process affecting the grantor has been commenced. In many civil law systems, there is a special action that was developed to allow a secured creditor to claim its encumbered assets from the insolvency administrator. This is the so-called ‘separatio ex iure crediti’ (separation based on a credit right), or Absonderungrecht, in the Germanic systems.11 This special remedy provides the secured creditor with a cause of action to recover the possession of the encumbered assets and then proceed to the enforcement of the secured claim. It assumes that the encumbered assets do not belong to the insolvency estate because of the preeminence of the secured creditor’s right in rem to the encumbered assets. On the other hand, the existence of a separation right assumes that an action by the secured creditor is necessary to exclude the encumbered assets from the insolvency estate. In some systems, though, there is no separation right in favour of secured creditors. The existence of a separation right in favour of owners can be used by creditors if they are able to structure the protection for their loans by using ownership of an asset as a security device (for instance, using reservation of title or fiduciary transfers). The case of Brazil is one of the most interesting ones: the insolvency estate comprises encumbered assets, but this means that only assets encumbered in traditional security rights (pledge, mortgage) are included in the estate.12 However, assets that are the object of a fiduciary sale or a conditional sale are not part of the insolvency estate, and the owner-creditor can recover the possession of the asset in order to satisfy its claim. The exclusion of encumbered assets may have the advantage of generally enhancing the availability of credit and lowering its cost because a secured creditor would be reassured that its enforcement and other rights would not be adversely affected by the commencement of its grantor’s insolvency proceedings. It is increasingly accepted, however, that this general advantage to an economy is outweighed by the advantages to be derived from including encumbered assets in the insolvency estate and restricting the exercise of a secured creditor’s rights as noted above. In most legal systems there has been an evolution that has resulted in the inclusion of encumbered assets in the insolvency estate, and the corresponding reduction or elimination of separation actions. In common law systems, the evolution has resulted in the inclusion of encumbered assets in the insolvency estate, thus eliminating the possibility of individual enforcement action by secured creditors without approval of the insolvency tribunal.

11 See §§ 49–52 of the German Insolvency Law. The concepts of Absonderung and Aussonderung are also present in other Germanic systems, such as the Austrian one. 12 See Arts 80–83 of the Brazilian Civil Code. See also Arts 48(3) and 85 of the Brazilian Bankruptcy Law (Law 11,101/2005).

Enforcement of Secured Credit

213

In any case, insolvency regimes around the world vary in responding to the question of the inclusion of encumbered assets in the insolvency estate. In most of the modern regimes, like the one in the US under the US Bankruptcy Code,13 encumbered assets are included in the insolvency estate.14 The policy rationale is that, absent special circumstances, assets of the grantor, whether encumbered or unencumbered, should be held together with a view to obtaining a greater recovery for the creditor group as a whole. Even though an asset may be encumbered, the asset may be an important component of the overall value of the grantor’s business, the separate liquidation of which would decrease the recovery for other creditors of the grantor. There are very few exceptions to this general principle. This is the same position as in most insolvency regimes, like France, Russia,15 and Spain. Once again, the international standard for insolvency and creditor rights recommends that encumbered assets be included in the insolvency estate. Most countries either comply with this recommendation or are moving gradually towards it. It is true that some systems, especially those where creditor protection is more intensive, avoid including encumbered assets in the insolvency estate. This results not only in a strong protection of secured creditor rights, but also in a lack of balance between the protection accorded to secured creditors and the protection of other creditors or the debtor. The systems that still establish that encumbered assets are not part of the insolvency estate correspond, for the most part, to the old common law tradition (for instance, Australia,16 Malaysia, Singapore), and some civil law countries (for instance, the Netherlands).

B. Even if an Encumbered Asset is Included in the Insolvency Estate, is Enforcement of the Secured Creditor’s Security Right Stayed in the Insolvency Proceeding? Of course, the mere inclusion of encumbered assets in the insolvency estate, as most legal systems prescribe, does not necessarily mean that secured creditors are prevented from enforcing their security rights. There is an additional question of whether secured creditors will be subject to a stay of creditors’ enforcement actions. The stay of creditors’ enforcement actions is one of the defining characteristics of an insolvency process: a stay of creditors’ enforcement actions ensures that creditors do not engage in a destructive race to seize the assets of the insolvent debtor.

13

11 USC § 101 ff (hereinafter the ‘US Bankruptcy Code’). US Bankruptcy Code § 541. 15 See Art 131 of the Russian Bankruptcy Law. 16 See, for Australia, s 471C of the Corporations Act 2001, which isolates secured creditors from the effects of the insolvency process. Secured creditors remain entitled to enforce their security rights despite the commencement of the liquidation. 14

214

José M Garrido and Edwin E Smith

Freezing creditors’ positions is the most effective way to ensure that the debtor’s business is kept as a going concern and this, in turn, provides better opportunities to maximise the returns of the insolvency process for all participants.17 Insolvency systems generally provide for a stay of creditors’ enforcement actions: in most cases, this stay is automatic and is part of the effects of commencement of an insolvency process.18 There is also a possibility that a stay of creditors’ enforcement actions may be requested by the insolvency administrator to defend the integrity of the insolvency estate. In this case, the stay is not automatic but would require, instead, a separate order from the insolvency tribunal restraining specific creditors from taking enforcement action against the insolvency estate. In this context, it is important to address whether the stay of creditors’ enforcement actions must also affect secured creditors in an insolvency regime in which encumbered assets are included in the insolvency estate.19 Finding the right balance between the interests of a secured creditor and the interests of the insolvency process may prove difficult.20 The inclusion of encumbered assets in the insolvency estate may increase the value of the insolvent business as a going concern. But recall that secured creditors obtain security rights to maximise recovery on their claims when their debtors or grantors default. That objective is achieved when there is a prompt realisation on the value of the encumbered assets.21 In the end, the inclusion of the encumbered assets in the insolvency estate should respect the value of the secured creditor’s security right,22 which may be put at risk by different factors, including the mere passage of time. The international standard provides a nuanced response to the question of the applicability of the stay of creditors’ enforcement actions to secured creditors and to the enforcement of secured credit within the insolvency process. In discussing 17 The stay of creditor actions is a fundamental feature of the insolvency process. According to the UNCITRAL Legislative Guide on Insolvency Law (n 3), Recommendation 7e, it is necessary to establish the protection of the insolvency estate against the actions of creditors, the debtor itself and the insolvency representative. The Guide goes on to say that ‘where the protective measures apply to secured creditors, the manner in which the economic value of the security interest will be protected during the insolvency proceedings’. 18 See, for example, the automatic stay in the French insolvency system (Art 47 of the law of 25.1.1985). 19 Of course, when encumbered assets are not included in the insolvency estate, secured creditors will not be affected by a stay of creditors’ enforcement actions. The insolvency tribunal may have no right, as a jurisdictional matter, to issue a stay against enforcement of security rights in encumbered assets. 20 See UNCITRAL Legislative Guide on Insolvency Law (n 3) 84: ‘The balance that is difficult to achieve in liquidation proceedings is between the competing interests of secured creditors, who will often hold a security right in some of the most important assets of the business and wish to enforce that security right, and unsecured creditors, who may benefit from retention of that asset to facilitate sale of the business as a going concern’. 21 cf UNCITRAL Legislative Guide on Insolvency Law (n 3) 87. 22 In bankruptcy law theory, it is generally accepted that the bankruptcy process transforms the contents of the procedural rights of creditors (such as the secured creditor’s right to enforcement), but the bankruptcy process respects the value of the substantial right of the creditors, including their relative ranking: see generally TH Jackson, ‘Translating Assets and Liabilities to the Bankruptcy Forum’ (1985) 14 Journal of Legal Studies 73.

Enforcement of Secured Credit

215

the inclusion of assets in the insolvency estate, the UNCITRAL Legislative Guide on Insolvency Law provides: Where encumbered assets are included in the insolvency estate, a number of insolvency laws provide certain protections, such as those relating to maintaining the value of the encumbered asset or the secured portion of a creditor’s claim and to specified situations where the encumbered asset may be separated from the estate. An insolvency law should make it clear that such an inclusion will not deprive secured creditors of their rights in the encumbered assets, even if it does operate to limit the exercise of those rights (eg postponement by operation of the stay) and should specifically ensure the protection of the rights of secured creditors in encumbered assets.23

The position, therefore, is that secured creditors may be affected by a stay of creditors’ enforcement actions, but there must be appropriate safeguards in place to protect the rights of secured creditors in encumbered assets and, especially, the value of those rights. In cases where the application of the stay to secured creditors does not generate any substantial benefit to the insolvency estate (for instance, when the encumbered assets will not fetch a higher value in a going-concern sale of the assets in insolvency estate or will otherwise have no effect on going concern value), there may be an exception to the general rule that secured creditors are subject to the stay. In those cases secured creditors may be allowed to enforce their security rights in the encumbered assets. Some insolvency laws establish that the function of the insolvency administrator is solely to liquidate, ie, to collect and realise on assets of the debtor and distribute proceeds among creditors by way of dividends. In those circumstances, a secured creditor may be permitted to freely enforce its rights against the encumbered assets to satisfy its claim without affecting the liquidation of other assets.24 Since liquidation of encumbered assets could be accomplished either by the insolvency administrator or secured creditors, secured creditors are excluded from the application of the stay. In one of the most influential systems, that of English law, the holders of fixed charges (as opposed to floating charges) are entitled to initiate enforcement actions despite the commencement of liquidation proceedings. This is also the position in numerous countries, including India, Mexico, Thailand25 and China.26 In other systems, even when encumbered assets are not

23

UNCITRAL Legislative Guide on Insolvency Law (n 3) 77. See UNCITRAL Legislative Guide on Insolvency Law (n 3) 88. 25 See Art 110 of the Thai Bankruptcy Act. The only limitation that secured creditors may suffer is that they need to allow the Official Receiver to inspect the encumbered assets. 26 A peculiar system is the one of the insolvency law of China, where the rights of secured creditors have been reinforced as compared with the previous regime. In the Chinese system, there is a general stay of creditors’ enforcement actions, affecting also secured creditors, but it only lasts for the time necessary for the insolvency administrator to take over the debtor’s assets. The enforcement actions of secured creditors can be resumed—or initiated—afterwards. The insolvency administrator may recover the encumbered assets either by satisfying the secured claims or by providing alternative collateral which is acceptable to the creditor (Art 37 of the Chinese bankruptcy law). 24

216

José M Garrido and Edwin E Smith

technically part of the insolvency estate, the position of secured creditors is further isolated from the potential effects of the stay (Malaysia). Secured creditors are entitled to apply to the insolvency tribunal for leave to initiate legal action to enforce their security rights, when a liquidation procedure (winding-up) has been commenced. In some of these systems, as an alternative to the stay, a practice has developed of holding pre-commencement negotiations between the debtor and secured and other creditors to achieve agreement on how to proceed. Systems that exempt all secured creditors from the scope of the stay may also establish exceptions that allow the insolvency administrator to suspend the enforcement of the secured creditor’s enforcement actions when it can be proved that the joint liquidation of assets can yield a better result for the general body of creditors.27 This is the case in Mexico, where the insolvency administrator may obtain an injunction against the enforcement action of a secured creditor within the first 30 days from the commencement of the insolvency process if the encumbered assets are needed,28 and a valuation of the encumbered assets is determined by order of the insolvency tribunal. If the encumbered assets have a higher value than the secured loan, including any accrued interest, the insolvency administrator will pay the full amount of the loan. If the valuation is less than the full amount of the loan, the insolvency administrator will only pay the amount determined in the valuation. The creditor may challenge the valuation before the insolvency tribunal. In other systems (Canada), the general rule is that the actions of secured creditors are not stayed in a liquidation proceeding, but the insolvency tribunals retain the discretion to consider an application of the insolvency administrator to stay the actions of secured creditors to preserve the value of the insolvency estate. Russian law offers secured creditors the possibility of electing separate, in-court enforcement of encumbered assets in the phases of an insolvency procedure that actually precedes the commencement of formal insolvency proceedings. In Germany, the secured creditors can exercise their separation actions to enforce their security rights; sometimes those security rights are based on ownership techniques, such as reservation of title and fiduciary transfers. The exceptions to this general rule are more procedural than substantive, providing the insolvency administrator with the possibility of selling the encumbered assets instead of allowing the secured creditor to conduct the sale—but this does not translate into a stay of enforcement actions by the secured creditor. The Japanese system

27 According to the UNCITRAL Legislative Guide on Insolvency Law (n 3) 88, ‘Some insolvency laws thus except secured creditors from the scope of the stay. Where that approach is adopted, however, some flexibility may be needed in cases where the insolvency representative may be able to achieve a better result that maximizes the value of the assets for the collective benefit of all creditors if the stay is applied to secured creditors. This may be particularly relevant where the business can be sold as a going concern in the context of the liquidation proceeding. It may also be true in some cases where even though assets are to be sold in a piecemeal manner, some amount of time is needed to arrange a sale that will give the highest return for the benefit of all unsecured creditors.’ 28 Art 214 of the Mexican Ley de Concursos Mercantiles (LCM).

Enforcement of Secured Credit

217

also recognises the rights of secured creditors to enforce their claims when the debtor is in a liquidation proceeding. The same regime applies to reservation of title and fiduciary transfer of assets. A similar position is that of the Korean system, where the insolvency process does not affect the enforcement powers of secured creditors. There are numerous other examples where the enforcement rights of secured creditors are not affected by liquidation proceedings (Sweden). Essentially, legal systems where encumbered assets are included in the insolvency estate approach the question of the application of the stay to secured creditors from two opposing starting points that operate as general rules (inclusion of the secured creditors from the scope of the stay; and exclusion of secured creditors from the scope of the stay), but the existence of exceptions to these general rules converge to a point in which specific situations may be resolved in a very similar way.29 Systems that follow closely the international standard establish a stay of creditor actions in liquidation, with appropriate protections to preserve the position of secured creditors. The function of the stay is to allow the insolvency administrator to take possession of the enterprise assets and determine the best method to achieve an economically beneficial sale of the assets.30 It is frequent that the sale of a business as a going concern generates more proceeds than a sale piecemeal of the assets included in the insolvency estate. The objectives of an insolvency regime support the existence of a limited stay that allows the insolvency administrator to arrange the sale of the enterprise as a going concern (examples would include France, Spain, Italy, or Switzerland). This stay can be effected in a number of different ways: a limited period of time—for instance, the French system establishes a three-month period,31 or an unlimited period of time with the possibility that secured creditors may obtain relief from the stay, or a stay that applies only to secured creditors if the encumbered assets are essential to achieve the sale of the business as a going concern. An example is the Spanish system, where the secured creditors’ actions are only stayed in the cases where the encumbered assets are indispensable for the continuation of the business activity. Although it would appear that this application of the stay is tied to the existence of a rehabilitation procedure, the Spanish system is based on a single unitary procedure (encompassing

29 In addition, there is a principle whereby certain encumbered assets securing financial contracts tend to be excluded from the effects of the stay of creditors’ enforcement actions in all legal systems. 30 See UNCITRAL Legislative Guide on Insolvency Law (n 3) 93–94. The Guide refers to a stay as a brief period, giving 30 or 60 days as examples of possible periods for a stay of secured creditor actions in the context of a liquidation proceeding. However, if the encumbered assets are essential for the sale of the business, the stay could be extended. Conversely, the stay could be lifted when the encumbered assets are not needed for a more beneficial sale of the other enterprise assets. 31 In the French insolvency system, secured creditors can pursue individually the encumbered assets if the liquidator has not taken action within a three-month period from the start of the liquidation (Art 161 of the Law of 25.1.1985). Creditors that legitimately retain possession of one of the debtor’s assets may obtain payment of its debt against restitution of the retained asset, if the asset is necessary to the debtor’s business, or be granted full ownership of the asset in payment of its debt.

218

José M Garrido and Edwin E Smith

both liquidation and rehabilitation).32 The Polish system also includes a broad stay of creditor actions for both liquidation and rehabilitation proceedings. Even though a stay on creditors’ enforcement actions may be imposed in liquidation proceedings, a stay of creditors’ enforcement actions and the application of the stay to secured creditors are even more important in rehabilitation proceedings. When the goal of the proceeding is to achieve the rehabilitation of the debtor’s business, there may be an imperative need to preserve encumbered assets that are essential for the continuing operation of the enterprise. In this case, the safeguards in favour of secured creditors need to be reinforced. The rehabilitation may require that the encumbered assets be included in the insolvency estate for a longer period of time than in the case of a liquidation proceeding. The fact that the encumbered assets may be needed to continue the business also implies that the risks of asset loss, asset substitution and asset depreciation are considerably higher than in a liquidation of the enterprise. Therefore, it is interesting to observe that legal systems that have introduced rehabilitation procedures in their insolvency systems tend to include secured creditors in the stay of creditors’ enforcement actions. Any exclusion would threaten the effectiveness of the rehabilitation. At the same time, those rehabilitation procedures establish stronger safeguards for the secured creditors affected by the stay.33 Whether an insolvency proceeding involves a liquidation or a rehabilitation, different legal systems have put in place different measures to protect secured creditors from the negative effects of a stay in the insolvency proceeding. These measures may relate, typically, to the duration of the stay, the protection of the value of the encumbered assets, payment of interest on the secured loan, and, ultimately, the provision of relief from the stay where the encumbered assets are not sufficiently protected or where they are not necessary to the sale of the entire business or of a productive part of it. There are also different factors that may be relevant when considering whether a secured creditor deserves protection for its security right or, ultimately, whether the secured creditor is entitled to relief from the stay to enforce its security right. These factors may include the comparison between the value of the secured claim and the value of the encumbered assets, the depreciation of the encumbered assets, the cost of providing protection to the secured creditor, and, as indicated above, the need of the encumbered assets for a more beneficial or efficient liquidation or rehabilitation of the insolvency estate.34 The measures adopted in protection of secured creditors depend on the specific circumstances affecting the encumbered assets and the secured loan. In many cases, if the value of the encumbered assets exceeds the amount of the loan, the

32 See Art 56 of the Spanish insolvency law (Ley Concursal) of 2003. In the Spanish system, it is possible that the stay of secured creditors’ actions last for one full year, even if the process results in a liquidation, and not in a rehabilitation. 33 See below, section III. 34 cf UNCITRAL Legislative Guide on Insolvency Law (n 3) 95.

Enforcement of Secured Credit

219

secured creditor may be protected by receiving the payment of interest on the loan despite the commencement of the insolvency process, limited by the value of the encumbered assets (see, for instance, the Spanish system,35 or the Swiss system). By contrast, if the value of the encumbered assets is less than the amount of the loan, it is possible that the insolvency administrator relinquishes the encumbered assets to the secured creditor. In other cases, the secured creditor may be consulted on the use and sale of the encumbered assets,36 with the possibility of adopting other protective measures. The protection against the depreciation in the value of encumbered assets can adopt different forms.37 It is possible that additional or substitute assets are provided as collateral for the secured claim.38 In some countries, it is also possible that the insolvency administrator makes payments in favour of the secured creditor that correspond with the diminution, through depreciation or otherwise, of the value of the encumbered assets.39 If a secured creditor does not receive proper protection, the legal system may provide that the creditor is entitled to seek relief from the stay. There may be additional circumstances that justify the relief from the stay. A typical example under the US Bankruptcy Code would be where the amount of the secured credit exceeds the value of the encumbered asset and either the grantor’s assets are being liquidated or, if the grantor is being rehabilitated, the encumbered asset is not necessary for the rehabilitation.40 The secured creditor may also have some grounds to seek an exemption from the stay if the grantor is using the encumbered asset in such a way that its value is diminishing and the secured creditor is not being compensated.41 This circumstance could arise when encumbered assets are being sold, leased or used by the grantor. In such a case, the encumbered asset is being sold, leased or used for the eventual benefit of the grantor’s other creditors seeking a higher recovery, but the secured creditor, being bound by the stay, is unable to enforce its security right against the encumbered asset at a time when its value is being diminished. In such a situation, it seems fair that the secured creditor be compensated for the diminishment of the value of the encumbered asset. Any other result would discourage the secured creditor from relying on the encumbered asset in extending credit in the first place, thereby reducing the availability or increasing the cost of credit ex ante.

35

See Art 59.1 of the Ley Concursal. See UNCITRAL Legislative Guide on Insolvency Law (n 3) 95. 37 Protective measures are needed as long as the value of the encumbered assets is lower than the value of the secured loan. In the opposite situation, ie, when the value of the encumbered assets is higher than the value of the secured loan, there is a ‘cushion’ that protects the secured creditor against the diminution in value of the encumbered assets. 38 The possibility of substituting collateral is expressly recognized in the French system: see Art 78 of the law of 25.1.1985. 39 See UNCITRAL Legislative Guide on Insolvency Law (n 3) 96. 40 US Bankruptcy Code § 362(d)(2). 41 US Bankruptcy Code § 362(d)(1). 36

220

José M Garrido and Edwin E Smith

Under the US Bankruptcy Code, the compensation to the secured creditor for the grantor’s sale, lease or use of the encumbered asset is referred to as ‘adequate protection’ and may be provided by monetary payments or additional collateral.42 However, if the value of the encumbered asset exceeds the amount of the credit, that excess may itself be adequate protection for the secured creditor unless and until that excess value has eroded. If the secured creditor is entitled to adequate protection and, even with the approval of the insolvency tribunal, is not receiving it, the reason is often because the debtor does not have the resources to provide the adequate protection. In that case, the secured creditor is entitled to an exemption from the stay on account of adequate protection not being provided to the secured creditor.43 Once the exemption is granted, the secured creditor may enforce its rights against the encumbered asset under non-insolvency law. In other insolvency regimes, the balance of interests between the secured creditor and the insolvency estate is achieved differently. The international standard on insolvency law provides examples of different measures to protect secured creditors,44 including the possibility of an application for relief from the stay, when the value of the encumbered asset is diminishing as a result of the commencement of insolvency proceedings and the secured creditor is not protected against that diminution of value. In general, legal systems have provided for more protective measures in favour of secured creditors in the case of rehabilitation proceedings, and protections in the case of stays applicable in liquidation proceedings tend to be less developed. It could be argued that the best protection for secured creditors in liquidation proceedings is that the liquidation proceeds swiftly to provide secured creditors with a prompt payment of their claims. If the liquidation phase is not prolonged, the risks of losses to secured creditors are greatly diminished.

C. How Does the Secured Creditor Participate in the Insolvency Proceeding? As the main goal of the insolvency proceeding is the maximisation of the going concern value of the estate so as to minimise the losses to unsecured creditors, there are numerous systems in which the participation of secured creditors is entirely secondary. In the systems where the encumbered assets do not belong to the insolvency estate, the secured creditor cannot participate in the insolvency estate. In the systems where secured creditors can exclude the assets from the estate, their participation usually consists in the exercise of a separation action to request the delivery of the encumbered assets. Finally, in the systems where secured creditors 42

See US Bankruptcy Code § 361. See US Bankruptcy Code § 362(d). 44 The UNCITRAL Legislative Guide on Insolvency Law (n 3) Recommendations 50–51, specify the various ways in which the secured creditor can receive ‘adequate protection’. 43

Enforcement of Secured Credit

221

are not subject to the stay of creditor actions, creditors tend to exercise their noninsolvency enforcement rights and thereby to avoid the insolvency process. This is particularly true in cases where the liquidation procedures are cumbersome and can be prolonged for many years—India is a case in point. In fact, the participation of secured creditors in the insolvency process is secondary in most systems. There are some systems in which a secured creditor cannot even request the commencement of an involuntary insolvency proceeding unless it is able to prove that the value of the encumbered assets is insufficient to cover the full amount of the claim. In some legal systems (Thailand, Ghana) only an undersecured creditor may commence an involuntary insolvency proceeding. It is considered that fully secured creditors and oversecured creditors have enough protection so that they do not need to resort to the insolvency process. In certain systems, therefore, the participation of the secured creditor is regarded as an exceptional event, and it is normally motivated by the desire of the secured creditor to participate in the insolvency proceeding for the unsecured portion of its loan. The Thai system provides an interesting example of this approach. A secured creditor has the following alternatives regarding its participation in the insolvency process, in case of filing a claim:45 (1) the secured creditor may surrender the encumbered assets for the benefit of all creditors, in which case the secured creditor can then file its claim for the full amount of the debt; (2) the secured creditor may enforce its security right against the encumbered assets, in which case the secured creditor can then file its claim for the balance of the unpaid debt; (3) the secured creditor may ask the insolvency administrator to sell the encumbered assets by public auction and apply the proceeds to the secured obligation, in which case the secured creditor can then file for the balance of the unpaid debt; (4) the secured creditor may arrange for the encumbered assets to be appraised, in which case the secured creditor can then file its claim for the deficit between the appraised value and the total amount of debt. In the Indian system, where secured creditors have an option to participate in the insolvency process, the liquidator will value the encumbered assets and, in consultation with secured creditors and the debtor, will sell encumbered assets at a public auction. A secured creditor may object to the sale of the encumbered assets, but the insolvency tribunal has wide discretion as to whether it accepts or rejects such an objection. In practical terms, there are serious drawbacks for secured creditors. The process is slow, and the assets will depreciate over time. Another detriment for the secured creditor is that the ranking of its secured claim in the liquidation may be lower than the ranking applicable outside a liquidation proceeding.

45

See s 96 of the Thai Bankruptcy Act.

222

José M Garrido and Edwin E Smith

The situation is similar in Malaysia. Secured creditors may rely on their security rights and not prove their claims; realise on their security rights and prove for the balance; surrender their encumbered assets and prove for the whole debt; or estimate the value of their encumbered assets and prove for the balance of their debt.46 However, secured creditors are not entitled to any interest in respect of their debts after the making of the receiving order if they do not realise their encumbered assets within six months from the date of the order. The same rule applies in Singapore. Under most insolvency regimes where encumbered assets are part of the insolvency estate, secured creditors will need to submit a claim for the amount of the secured credit. Under the US Bankruptcy Code, the claim would be treated as a secured claim entitled to preferential treatment based on the value of the encumbered assets. But if the value of the encumbered assets is less than the amount of the claim, only the portion of the claim equal to the value of the encumbered assets will be treated as secured; the balance of the claim will be treated as a separate general unsecured claim.47 This principle of connecting the value of the claim with the value of the encumbered assets should operate as a universal principle, resulting in the conceptual categories of undersecured creditors, fully secured creditors and oversecured creditors.48 A further question arises as to what elements constitute the secured creditor’s claim, whether secured or unsecured. If the claim is related to a loan, the claim would consist of the principal of the loan and any interest on the loan unpaid up to the time of the commencement of the insolvency proceeding. The question often arises, though, as to whether the creditor may claim as well the interest accruing after the commencement of the insolvency proceeding and the collection costs. There is an even finer legal point, as this question can be further split in two: one issue is whether secured creditors can claim interest and costs at all; and the second question is whether secured creditors can claim interest and costs with the same priority as the principal of the loan. Under the US Bankruptcy Code, a secured creditor is generally entitled to claim post-commencement interest and reasonable collection costs but only to the extent that the value of the encumbered asset exceeds the secured creditor’s pre-commencement claim.49 If the pre-commencement claim exceeds the value of the encumbered asset, the secured creditor is not entitled to post-commencement interest but may perhaps claim enforcement costs as part of its pre-commencement general unsecured claim. In other insolvency regimes, a similar situation emerges. In the Spanish insolvency regime, the rule on non-accrual of interest is only excepted for secured 46 See schedule c of the Bankruptcy Act 1967 of Malaysia; see B (Kam) Kamarul, ‘Insolvency Law in Malaysia’ in R Tomasic (ed), Insolvency Law in East Asia (Aldershot, Ashgate, 2006) 321, 327. 47 US Bankruptcy Code § 506(a)(1). 48 See JM Garrido, ‘Oversecured and Undersecured Creditors in Cross-Border Insolvencies’ (2014) 5 International Insolvency Law Review 375. 49 US Bankruptcy Code § 506(b).

Enforcement of Secured Credit

223

creditors up to the value of their encumbered assets. However, expenses and costs are not normally included in the priority of the secured creditor and could only be claimed as part of a general unsecured claim, if at all.50 Of course, of critical importance to the secured party is the priority of its claim, at least as far as the claim is covered by the value of the encumbered assets. The claim of the secured creditor will be given preferential treatment on account of the existence of the security right under most insolvency regimes. That preferential treatment raises more complex issues when a secured creditor has a broad security right encompassing a substantial portion of the grantor’s assets. Spain recently reformed its law to clarify that the priority of secured credit extends to the value of encumbered assets—but no further.51 This is a puzzling reform, as the law as reformed seemed to be the position under the old law, too. It does not make sense to infer that because a creditor happens to have a security right, the priority of the security right would extend to the whole amount of the loan, even if the amount of the loan is much higher than the value of the encumbered assets. The reform of the law has introduced references to the valuation of the encumbered assets as an important part of the procedure. This is a significant development, since the valuation of the encumbered assets is central to the determination of the rights of secured creditors, especially in all the cases where there is a rehabilitation plan that may represent a solution to the insolvency process without a sale of the encumbered assets. Some systems merely enunciate the principle of the priority of secured creditors. For instance, China’s concise bankruptcy law recognises the priority of secured creditors without providing any more precision to this general principle.52 In other countries, such as India, Russia, or Mexico, the priority of the secured creditor’s claim is not absolute. In the case of India, when secured creditors decide to participate in the liquidation process, the proceeds of the sale of the encumbered assets will be allocated to the estate and will be available for distribution to creditors according to the insolvency ranking of claims. That ranking establishes that secured creditors will receive payment after the cost of administering the estate and pari passu with claims of workers for the last four months of wages. This represents a disincentive for the participation of secured creditors in the insolvency process. Indeed, one of the most problematic features of an insolvency system is the existence of differences in the ranking or in the extent of the priority. Such is the Russian case, where the secured creditor who elects to pursue a separate enforcement action will recover the full value of the encumbered assets. However, if the

50 There is controversy as to whether unsecured creditors can claim for their expenses and costs in the insolvency of the borrower, and the position of secured creditors seems similar in this regard. 51 Art 90.3 of the Spanish insolvency law, as modified by Royal Decree-Law 11/2014 and its confirming law. 52 Art 109 of the Chinese insolvency law (Secured creditor is entitled to obtain payment in priority over debtor’s specific asset).

224

José M Garrido and Edwin E Smith

creditor decides to participate in the insolvency process, it will only obtain 80 per cent or even 70 per cent of the proceeds to its claim; the balance of the secured claim will occupy a lower position in the ranking of claims, immediately above the claims of unsecured creditors.53 In other systems, such as in Mexico or France, the issue is that the secured creditor does not enjoy an absolute priority over the proceeds of the encumbered assets. Other claims, namely the claims of workers for wages, have priority over the secured claims. This is often referred to as ‘super-priorities’, as opposed to regular statutory priorities, which provide creditors with a preference over unsecured creditors only. Super-priorities reduce the effectiveness of secured credit and may have a negative impact on the availability or cost of credit for businesses. Of course, the effect of reduction of credit or its cost depends on the amount of the claims that these super-priorities can protect.54 A more difficult issue arises when considering that in modern secured transactions systems it is possible to create security rights over substantially all the assets of a grantor. In such cases, the debate is on the protection of other creditors and the function of the insolvency regime. The Finnish regime was the first one to react against the priority of the secured creditor over all the assets of the debtor. It established the principle that the secured creditor holding a security right over all the debtor’s assets is entitled to a priority claim only up to 50 per cent of the proceeds of the sale of encumbered assets.55 A similar argument over the excessive control and preeminence of creditors with floating charges resulted in the reform legislation introduced in England by the Enterprise Act (2002). According to this legislation, there is a ‘prescribed part’ of the proceeds of a floating charge which must be made available to unsecured creditors, and which must have a ₤10,000 minimum value, and a ₤600,000 maximum value. The prescribed part of the proceeds is calculated on the scale of 50 per cent of the first ₤10,000 and 20 per cent of higher amounts, until reaching the maximum.56 In the US, notwithstanding an intense academic debate, there have been no legal changes reducing the extent

53 The amount the secured creditor is entitled to receive is 70% of the proceeds of the sale of the encumbered assets (80% if the secured claim is a loan). The rest of the proceeds will be distributed as follows: 20%–15% for the first and second priority in the ranking of creditors (tort claims and labor claims); and 10–5% of the proceeds will be allocated to costs. 54 For instance, the French super-priority is much more restricted that the super-priority in Mexican law: whereas the French insolvency law provides a super-priority for workers’ wages corresponding to the last 60 days of work (Arts L625-7 y L625-8 Ccom), Mexican law provides for a priority covering the last two years of wages. See other examples in W Huaiyu, ‘An International Comparison of Insolvency Laws’ in OECD, Legal and Institutional Reforms of Asian Insolvency Systems (2006) [some of the information in the paper is no longer accurate]. 55 In Finland the floating charge gives only 50% priority, see C Bergström, T Eisenberg, and S Sundgren, ‘On the design of Efficient Priority Rules for Secured Creditors: Empirical evidence from a change in law’ (2004) 18 European Journal of Law & Economics 273. 56 See s 176A of the Insolvency Act 1986, as drafted by the Enterprise Act 2002, and the Insolvency Act 1986 (Prescribed Part) Order 2003.

Enforcement of Secured Credit

225

of the priority of secured creditors, even when those security rights may cover virtually all assets of an enterprise.57 A related question is whether some deduction must be made to the proceeds obtained in the sale of encumbered assets for the expenses of the realisation of those assets and as a general contribution to the expenses of the insolvency process. Germany introduced a contribution by secured creditors to the costs of the insolvency proceedings. The contribution is conventionally fixed at 9 per cent of the proceeds of the realisation of the encumbered assets.58 In other systems, there is no fixed contribution for the expenses of the insolvency proceeding but the secured creditor is responsible for the costs incurred by the insolvency administrator in preserving the value of the encumbered assets, and for the costs of organising and carrying out the sale, in the cases where the sale is conducted by the insolvency administrator. For instance, under the US Bankruptcy Code, the costs of preserving the value of the encumbered assets at the request and for the benefit of the secured creditor may be charged against the encumbered assets, but not otherwise.59 Most insolvency laws follow this approach, with the salient exception of the German system described above. The preferential right of the secured creditor includes a right to the proceeds of the encumbered assets. However, the amount of the proceeds may depend on the effectiveness of the arrangements for the sale of the encumbered assets. In this regard, there are different approaches in insolvency systems. In the systems where encumbered assets are not part of the insolvency estate or in systems where the secured creditors have the right to separate the assets from the insolvency estate, secured creditors will normally enforce their security rights separately from the insolvency process. The proceeds will be used to pay off the secured loan, and any surplus is handed to the insolvency administrator. In many systems, the insolvency administrator has the possibility of participating in the enforcement process conducted by the secured creditor. There are mainly two reasons for this: the first is the general one that the insolvency administrator protects the interests of the general body of creditors. It is in the interest of all creditors that the sale of the encumbered assets fetches the highest possible price. In contrast, the secured creditor may be content with a price that covers the amount of the secured loan but has no incentive in making more efforts to obtain a higher value. The second reason for the participation of the insolvency administrator exists in the case of super-priorities. The insolvency administrator

57 In 1996, Professor Elizabeth Warren submitted to the American Law Institute a proposal to ‘carve out’ 20% of the proceeds of the sale of encumbered assets in bankruptcy for the benefit of unsecured creditors. This sparked a heated debate in US academia and legal practice: see L Bebchuk and J Fried, ‘The Uneasy Case for the Priority of Secured Claims in Bankruptcy’ (1996) 105 Yale Law Journal 857; WJ Woodward, ‘The Realist and Secured Credit: Grant Gilmore, Common-Law Courts, and the Article 9 Reform Process’ (1997) 82 Cornell Law Review 1511; LM LoPucki, ‘Should the Secured Credit Carve-Out Apply Only in Bankruptcy? A Systems/Strategic Analysis’ (1997) 82 Cornell Law Review 1483. 58 See § 171 of the German Insolvency Act. 59 US Bankruptcy Code § 506(c).

226

José M Garrido and Edwin E Smith

must collect the amounts that the super-priority creditors are entitled to recover in precedence to the secured creditor. Even if the insolvency administrator does not participate in the process, secured creditors are under the obligation to pay these amounts, and super-preferential creditors can request the payment of these debts from secured creditors. In other systems, when the secured creditor is subject to a stay of enforcement actions, there are typically several options as to how to proceed with the sale of the encumbered assets. A summary of these possibilities includes the following: (1) sale of the encumbered assets as part of the sale of the business as a going concern; (2) sale of the encumbered assets by the insolvency administrator; and (3) sale of the encumbered assets by the secured creditor; and, in some systems, appropriation of the assets by the secured creditor. The first case, the sale of the business as a going concern, is the best justification for the stay of secured creditors’ enforcement actions in a liquidation proceeding, but it is also the case in which there may be more difficulties in establishing the rights of parties. The sale of the business as a going concern is a complex transaction, and it frequently collides with the interests of the secured creditors. In some systems, it is possible to sell the business with the condition that the assets remain encumbered by the security rights and the acquirer becomes responsible for the repayment of the secured claims.60 This procedure usually requires the consent of the secured creditor, and it also tends to reduce the probabilities of obtaining a favourable price for the sale of the business. In Japan, for instance, a sale of encumbered assets conducted in the insolvency process does not automatically release the encumbered assets from a security right in the encumbered assets. The insolvency administrator can negotiate with the secured creditor for the release of the security right and can even obtain an order from the insolvency tribunal for the release of the security right when the secured creditor is opposing the sale, especially if the secured creditor has a junior priority security right in the encumbered asset and is wholly unsecured after taking into account the senior security rights. The insolvency tribunal may cancel the security right subject to certain conditions established by the insolvency tribunal, and the sale can proceed afterwards, free of encumbrances. Another alternative consists in the redemption of the encumbered assets by the insolvency administrator. The difficulty in this approach is that the insolvency administrator may face challenges in raising the necessary funds to pay the secured claim so that the encumbered assets could be redeemed. In the absence of a nonconsensual release of the security right or the redemption of the encumbered assets, most systems that foresee the sale of the business as a going concern place the responsibility of conducting the sale on the insolvency administrator, with the insolvency administrator distributing the proceeds of the 60

See Art 93 of the French law (loi 25.1.1985).

Enforcement of Secured Credit

227

sale in accordance with the ranking of creditors’ claims. The technical difficulty derives from the fact that normally the secured creditor will have a priority right over specific assets or categories of assets in the insolvency estate, but not over all the assets that are included in the sale of the business as a going concern. The usual method to solve the problem is based on a valuation of the encumbered assets and the assignment of a percentage share in the proceeds of the sale attributable to the value of the encumbered assets that were sold as part of the overall sale.61 In this way, an eventual shortfall in the sale is distributed between the secured creditor and the other creditors.62 In the case of the sale of specific assets, the issue is whether the sale is controlled by the insolvency administrator and the insolvency tribunal or whether a secured creditor can use other enforcement mechanisms, including out-of-court enforcement and judicial adjudication of the encumbered assets to the creditor (France),63 or even appropriation of the encumbered assets, where these methods are legally available (for example, in Poland). Most insolvency laws recognise the power of the insolvency administrator to sell the assets included in the insolvency estate, even when those assets are encumbered, and establish either that secured creditors preserve their preferential rights over the proceeds of the sale (Brazil, France),64 or that secured creditors are entitled to receive those proceeds. The insolvency administrator has the right incentives to seek a higher price that benefits all creditors, and not just the secured creditor. The asset is sold free of encumbrances, and the secured creditor is entitled to receive the proceeds that are sufficient for the repayment of the secured claim, after payment of the expenses of the sale (Germany). The specific amounts that the insolvency administrator will deduct from the proceeds before paying the secured creditors depend on the characteristics of the legal system, including its ranking of claims in insolvency.65 Under the US Bankruptcy Code, the secured creditor is generally entitled to all of proceeds of the sale less any expenses incurred by the insolvency administrator in effecting the sale. If, after application of the sale proceeds, the secured credit has been fully paid, and surplus proceeds remain, the surplus must, subject to the rights of any junior secured creditors, be

61

Art 93 of the French law (loi 25.1.1985). The secured creditors will be also unsecured creditors for the unsatisfied amount of their secured claims. 63 See Art L.642-20-1-1º and 2º Ccom and Art 159 of the French insolvency law (loi 25.1.1985). In the liquidation, the creditors protected by fiduciary arrangements can also enforce their security rights: see P Crocq, ‘L’ordonnance du 18 décembre 2008 et le droit de sûretés’ (2009) JCPE 1313. The possibility of judicial adjudication of the enforcement of security rights is only recognized for certain assets. Appropriation of the encumbered assets is forbidden in French insolvency procedures: see L622-7-I-3º Ccom. 64 See Art 78 of the French law (loi 25.1.1985). 65 For instance, in Brazil, the secured creditor will be entitled to receive the proceeds of the sale, but only after the deduction of the expenses of the liquidation and the super-priority of workers’ claims, which covers a maximum of 150 minimum wages per worker (see Art 83, II, of the Law 11,101/2005 of Brazil). 62

228

José M Garrido and Edwin E Smith

turned over to the insolvency administrator for the benefit of the insolvency estate. This is a common feature of most insolvency systems: if there are any proceeds left after paying the secured creditor’s claim, the insolvency administrator will distribute the rest of the proceeds among the other creditors according to their respective positions in the ranking of claims. If the solution favoured by the legal system is the opposite, ie, the sale of the encumbered asset by the secured creditor, the system must include safeguards to ensure that best efforts are employed for the procedures for the sale and that the proceeds are distributed according to the ranking of claims. In systems where secured creditors have the possibility of enforcing their claim separately, it is frequent that secured claims occupy the first place in the ranking of creditors. The main issue then is to ensure that the surplus after the repayment of the secured claim is turned over to the insolvency administrator so that the surplus can be distributed among the creditors placed in lower positions of the hierarchy of claims. After secured creditors receive their payment, they may continue to be involved in the insolvency process if there is a part of their claim that was not satisfied with the proceeds of the sale. In the system in the US, if the encumbered asset is sold or leased or, in the case of rights to payment, collected, whether in the insolvency proceeding or, with an exemption from any stay being granted, under non-insolvency law, and the secured credit has not been fully paid from the proceeds of the sale, lease or collection, the secured creditor will typically be permitted to continue to participate in the insolvency proceeding with a submitted general unsecured claim for the remaining unpaid credit.66 This is also the position in most insolvency systems. It is understood that secured creditors enjoy a preferential right that is limited to the proceeds of the sale of the encumbered assets, but the outstanding balance of the secured loan after receiving a partial payment is characterised as an unsecured claim. This is why many secured creditors, even in systems where those creditors have the possibility of enforcing their claims outside the insolvency process, decide to participate in the collective proceedings: when creditors are undersecured, they tend to become involved in the process, as the unsecured portion of the claim may be significant, or even more substantial than the secured portion.

D. How is the Secured Claim Ranked and Paid in a Liquidation of the Grantor? As discussed above in the case of the sale of a single encumbered asset, if the secured creditor’s claim is entitled to first priority under non-insolvency law and the encumbered asset is sold or collected in an insolvency proceeding, all of the proceeds of the sale or collection, less any expenses incurred by the insolvency administrator in effecting the sale or collection, would be paid to the secured creditor for application to the secured credit. This is the general trend around 66

US Bankruptcy Code § 506(a)(1).

Enforcement of Secured Credit

229

the world, with the exception of systems where ‘super-priorities’ are established (Brazil, Mexico, and France, among others). Where these ‘super-priorities’ exist, they tend to protect the claims of workers or, less frequently, the claims of public creditors, such as the tax authorities.

III. The Treatment of Secured Credit in Rehabilitation Proceedings The position of secured creditors in rehabilitation proceedings raises distinct legal issues that deserve particular analysis and careful consideration. The goal of rehabilitation proceedings is the preservation of the business as a going concern, providing for the satisfaction of creditors’ claims through methods that avoid the liquidation of the debtor’s insolvency estate. The goal and methods of rehabilitation proceedings require significant changes to the rights of secured creditors and, in turn, these changes also require intense measures to protect the economic value of the interests of secured creditors from the negative effects of the different procedural regime. The first and most important of the modification of secured creditor rights in rehabilitation proceedings is the imposition of a longer—and riskier—stay of creditors’ enforcement actions. Although many insolvency regimes, as described above, do not include a stay of secured creditor enforcement actions in liquidation proceedings, or, following international best practice, incorporate a limited and targeted stay, the situation in rehabilitation proceedings worldwide is substantially different. The imposition of a stay is considered to be more important in the rehabilitation context than in the liquidation context because the goal of the rehabilitation is to keep the debtor’s business viable as a going concern. A stay of secured creditor enforcement actions can be instrumental in achieving that goal, especially when the encumbered assets are key elements of the enterprise, such as essential productive assets. In this regard, the stay may be absolutely necessary in a rehabilitation process, and not just merely convenient as in a liquidation proceeding, where the stay is justified solely in terms of the economic advantage that may bring a sale of the business as a going concern. In the case of a rehabilitation proceeding, it is not just a question of mere economic advantage, but an essential mechanism to achieve the ultimate goal of the process. This is a basic economic assumption in the rehabilitation procedures of most countries (including, for instance, the US, France,67 or Spain). The safeguards for secured creditors in the stay for rehabilitation proceedings may be similar, if more intense, than those adopted in the

67 In France, there are two rehabilitation procedures: ‘procedure de sauvegarde’ (including ‘sauvegarde acceleree’ and ‘redressement judiciaire’. The stay and other protective measures apply to both procedures, with some different rules.

230

José M Garrido and Edwin E Smith

case of the stay of creditor actions in a liquidation proceeding: duration of the stay, the protection of the value of the encumbered assets, payment of interest on the secured loan, and, ultimately, the provision of relief from the stay where the encumbered assets are not sufficiently protected or where they are not necessary to the sale or rehabilitation of the entire business or a productive part of it. These factors explain why even systems where secured creditor rights are strongest, a stay imposed by the system nevertheless accommodates the requirements of modern and effective rehabilitation procedures (Canada,68 China,69 Germany,70 Japan, Korea,71 the Netherlands,72 Singapore,73 or Thailand).74 One clear example is the law of Austria, which subjects secured creditors to a stay in the rehabilitation procedure. A reform in 2009 suspended the right of separation of secured creditors, for a maximum of six months, when the return of the encumbered assets to the secured creditor would jeopardise the continuation of the debtor’s business. There are other similar examples: Belgium also establishes a six-month stay period for the enforcement of secured claims during judicial rehabilitations. One of the problems typical of the stay of secured creditor enforcement actions that modern rehabilitation procedures seek to address is the length of the stay. A rehabilitation procedure may require a longer period in order to negotiate and agree on a rehabilitation plan, while the business must continue its activities in order to preserve its going concern value. However, an unlimited stay of creditor enforcement actions would impact negatively the availability or cost of credit.75 To avoid the application of the stay for an uncertain or unnecessarily lengthy period and encourage a speedy resolution of the rehabilitation proceedings, it may be reasonable to limit the application of the stay to the time that it may reasonably take for a rehabilitation plan to become effective, provided that that does not take a significant period of time and the proceedings are not allowed to continue for

68 The stay of secured creditor actions applies, in different degrees and with different conditions, to the two rehabilitation procedures existing in Canada, the BIA rehabilitation and the CCAA rehabilitation. 69 See Art 75 of the Chinese law. 70 The peculiarity is that Germany, like Spain, does not have a separate procedure for liquidations and rehabilitations, but a single unified procedure with two tracks, for rehabilitation and for liquidation. The protection of the estate is necessary as long as there is a possibility of implementing a rehabilitation. 71 In Korean rehabilitation proceedings, the enforcement of secured claims is stayed with the commencement order and the stay remains effective until the confirmation of the rehabilitation plan. 72 In the Netherlands, despite the absence of a modern rehabilitation procedure, the suspension of payments procedure allows a stay of secured creditors’ actions if the insolvency tribunal orders a so-called ‘cooling-off ’ order that affects them. 73 In Singapore, secured creditors have their enforcement rights suspended in judicial management unless there is leave of the insolvency tribunal to proceed with the enforcement action. 74 See s 90/28 of the Thai Bankruptcy Act. 75 UNCITRAL Legislative Guide on Insolvency Law (n 3) 87, discussing that such a stay ‘may ultimately undermine not only the autonomy of the parties in their commercial dealings and the importance of observing commercial bargains, but also the availability of affordable credit; as the protection provided by security rights declines, the price of credit may need to increase to offset the greater risk’.

Enforcement of Secured Credit

231

years without a rehabilitation plan being proposed and approved.76 Such a limitation may also have the advantage of providing secured creditors with a degree of certainty and predictability as to the duration of the period of postponement of their rights and the treatment of those rights in the rehabilitation plan. Alternatively a fixed time period might be specified. The difficulty with that approach, however, is that the time period may not always be sufficiently long, depending on the size and complexity of the rehabilitation and the plan required, and may be difficult to enforce. The systems of Belgium and Austria are indicative of this approach. Against the risks derived from a long period before the procedure is concluded, numerous insolvency systems have established a maximum limit in rehabilitation proceedings for the stay of secured creditors’ enforcement actions. Belgium and Austria have established a six-month limit, and there are other countries which have set a similar limit (Brazil).77 In a number of legal systems, the stay period can be longer (for instance, in England and Spain the stay can last for twelve months). Other possible approaches include establishing clear time limits, with the possibility of extension (Canada),78 or providing for relief from the stay in certain circumstances. The riskier systems connect the length of the stay with the adoption of other decisions within the process, with the possibility that delays in the procedure would translate into longer stay periods.79 In any case, it is important that the overall design of the insolvency law encourage speedy and efficient progress of the rehabilitation proceedings, enabling the period for application of the stay to secured creditors to be minimised. Another risk that the secured creditor faces is the loss or depreciation of the encumbered assets. This risk is much higher in rehabilitation proceedings, because the stay is longer than in a liquidation proceeding, and also because frequently the encumbered assets will be used in the productive activities of the business. Many legal systems have introduced, as a protective measure, the right of the secured creditor to request that the stay is lifted when the encumbered assets are not required for the rehabilitation (see for instance, Thailand,80 or France). 76

See, in this regard, UNCITRAL Legislative Guide on Insolvency Law (n 3) 93. Brazilian system establishes a maximum of 180 days for the stay period (Art 6 of the Law 11,101/2005). 78 In the Canadian CCAA rehabilitation process, the stay can be granted for a 30-day period, but this period can be extended. 79 This is the case in the French system, where the stay of secured creditors’ actions last for the length of the observation period and until the approval of the plan (see Art L. 622-23-1 Ccom). Belgian law also ties the stay of secured creditors’ action to the time necessary for the approval of the rehabilitation plan. 80 In Thailand, secured creditors can obtain permission from the insolvency tribunal to enforce where the collateral is not needed for the rehabilitation or when secured creditors are not being adequately protected (s 90/28 of the Thai Bankruptcy Act). There are differences between systems where the fact that the asset is not needed for the rehabilitation justifies that the stay is lifted, from other systems that require that the asset is essential for the rehabilitation in order to maintain the stay (Spain, Art 59 of the Insolvency Law, Ley Concursal; French law uses the similar concept of ‘indispensable’ assets, see Art 70 of the loi 25.1.1985). 77 The

232

José M Garrido and Edwin E Smith

The most advanced rehabilitation systems, such as the United States of America, have adopted a flexible framework to provide ‘adequate protection’ to secured creditors affected by the stay. As stated above, adequate protection may be provided by the mere existence of an ‘equity cushion’, ie, the excess of the value of the encumbered assets over the amount of the loan. But there are several other alternatives that can adequately protect the secured creditor, such as monetary payments or additional or substitute collateral. In other systems, adequate protection lacks flexibility, and only some forms of protection are explicitly foreseen in the law (for instance, Spanish law only considers the possibility of accruing interest in favour of the secured creditor when the value of the encumbered assets exceeds the amount of the loan). In cases where the interests of secured creditors are at risk and there are no measures that adequately protect their position, numerous systems allow secured creditors to apply to the insolvency tribunal for leave from the stay in order to enforce their security rights, thus minimizing the damage suffered. Some systems are more restrictive in granting this relief to creditors (Korea,81 China),82 but other systems recognise broad grounds to obtain the exemption from the stay (Japan, Canada).83 As the rehabilitation procedure may conclude with a rehabilitation plan that implies the continuation of the business as a going concern, it may result in a solution to the debt distress problem of the debtor that does not involve the sale of encumbered assets. If that is the case, the position of secured creditors needs to be protected against the risk of not receiving the economic equivalent of what creditors would obtain by liquidating the encumbered assets. A secured creditor may be willing to participate in the rehabilitation plan, especially if offered in the rehabilitation plan a restructured debt obligation secured by the same encumbered assets or in substitute encumbered assets of at least equivalent value. However, the question often arises of whether the secured creditor may be forced to participate in the plan over the objection of the secured creditor. Under the US Bankruptcy Code, the answer to the question of the participation of the secured creditor in the rehabilitation plan will first depend upon whether the secured creditor’s claim is impaired. The claim is not impaired if the secured creditor retains all of the rights that it had pre-commencement, or all pre- commencement defaults are cured, the original maturity of the debt is reinstated, 81 In the Korean system, secured creditors may obtain relief before the confirmation of the rehabilitation plan as long as the insolvency tribunal finds that such relief does not harm the rehabilitation process. In practice there are few instances in which relief from the stay is granted to secured creditors. 82 Under Art 75 of the Chinese Bankruptcy Law, when the collateral ‘is at the risk of being destroyed or its value is at the risk of being reduced dramatically to the extent of impairing the rights of secured creditors, the secured creditors may request the People’s Court to be exempt from the stay’. 83 The insolvency tribunal in a Canadian BIA rehabilitation process enjoys wide discretion, as the law establishes as grounds that the creditor is likely to be materially prejudiced by the stay or that it is otherwise equitable that the stay be lifted. The other rehabilitation procedure, the CCAA rehabilitation, does not contain specific provisions on lifting the stay, but insolvency tribunals also enjoy the discretion of lifting the stay.

Enforcement of Secured Credit

233

the secured creditor is compensated for damages, and all of the secured creditor’s pre-commencement are otherwise retained.84 Usually the secured creditor will be treated in a separate class of creditors;85 it takes a majority in number and two-thirds in amount of each class to accept the plan for that class.86 If the secured creditor’s claim is unimpaired, then the secured creditor is considered to have accepted the plan.87 If the secured creditor’s claim is impaired, it is entitled to reject the plan. Even though the secured creditor’s claim is impaired and it rejects the rehabilitation plan, the secured creditor may still be forced to participate in the plan (be ‘crammed down’) if: (1) at least one impaired class of claims does accept the plan; (2) under the plan the secured creditor will retain its security right in the encumbered assets or, alternatively, will receive the value of the security right, the proceeds of the encumbered assets if the encumbered assets are sold or what the US Bankruptcy Code refers to as the ‘indubitable equivalent’ of its secured claim; (3) the secured creditor will receive under the plan at least as much as it would have received on a straight liquidation of the debtor’s assets (the so-called ‘best interest of creditors’ test); (4) junior claims or interests do not receive anything under the plan until senior claims or interests are paid in full (the so-called ‘absolute priority’ rule); and (5) certain other statutory requirements are met.88 Often the outcome will be determined by what value is placed upon the encumbered assets and what discount rate applies to the stream of payments that the secured creditor is to receive under the plan. Both of these issues are often hotly negotiated and may well be contested if there is no agreement among the affected parties. The sophisticated set of protections that the system in the US affords tends to work in a balanced way, and in a climate of negotiation among the parties. Bearing in mind this example, it is interesting to look at some of the elements of protection of secured creditors in rehabilitation plans and how these elements are implemented in other legal systems around the world. One of the basic elements for the protection of secured creditors in rehabilitation procedures is the requirement that rehabilitation plans be voted by classes of creditors. If the system would require a majority of creditors, it is quite possible that a majority of unsecured creditors would be able to impose a plan on higher ranked creditors, notwithstanding the differences in risk and potential returns in the rehabilitation. Voting by classes avoids that creditors with disparate

84

US Bankruptcy Code § 1124. See US Bankruptcy Code § 1122. US Bankruptcy Code § 1126(c). 87 US Bankruptcy Code § 1126(f). 88 US Bankruptcy Code § 1129. 85 86

234

José M Garrido and Edwin E Smith

i nterests impose their will on smaller groups of creditors. Indeed, voting by classes has become part of the international standard and is recognised as a basic principle in numerous systems, including those with a short tradition in rehabilitation proceedings.89 The general principle in a system of voting by classes is that creditors should only be part of the same class when there is some common element. In most systems, secured creditors form a separate class, reflecting their higher position in the ranking of claims. In certain cases, the existence and specific requirements of classes are set by the law;90 in other cases, the system allows considerable leeway in the formation of classes, as under the US Bankruptcy Code. Voting by classes represents a protection for secured creditors, but the system often requires balancing the protection afforded to secured creditors with the achievement of the goals of the rehabilitation procedure. To that effect, the law allows, under certain circumstances, the confirmation of a rehabilitation plan even against the vote of the class of secured creditors. Otherwise, creditors who hold security rights over key productive assets would be given a veto over the rehabilitation plan. This possibility of approving a rehabilitation plan without the consent of secured creditors (‘cram-down’ of secured creditors), which was referred to above in the context of the system in the US, is a feature that is extending to other legal systems in different parts of the world. Approving a rehabilitation plan against the will of secured creditors, and with binding effects over them, requires special protection. Some of these protections vary greatly from system to system (a required number of classes voting in favour of the plan, a required majority of creditors within each class,91 or substantive contents of the plan),92 but the concepts of ‘absolute priority’ and the ‘best 89 In a new system, where there was no tradition of rehabilitation procedures, such as the Chinese system, the law also prescribed a system of voting by classes (see Art 82 of the Chinese bankruptcy law). There are few exceptions to the vote by classes in rehabilitation in comparative insolvency law—the most noticeable exception may be Belgium, where all creditors vote in one single creditors’ meeting. 90 The Thai system protects secured creditors by introducing specific requirement in the formation of classes, which are the following: (1) a separate group for each secured creditor with secured debt equal to not less than 15% of the total amount of debts; (2) a group comprising all secured creditors not classified in group (1); (3) unsecured creditors may be divided into several groups, with unsecured creditors having the right to claim or benefit in essentially the same matter or in a similar matter being classified in the same group; and (4) a group comprising all subordinated creditors (s 90/42 bis of the Thai Bankruptcy Act). In Brazil, creditors are separated in three classes of claims: labour claims, secured claims and a third class which comprises claims entitled to special privilege, claims entitled to general privilege, unsecured claims and subordinate claims (Arts 41–46 and 58 of the Law 11,101/2005). These classes are established by law and may not be modified. 91 Spain has recently introduced the cram-down of secured creditors, and the protection mechanism for secured creditor is based in the rules for formation of classes and in the need of obtaining a supermajority in each class, which may be of 60% or 75% within each class, depending on the contents of the plan (severe haircuts require the higher supermajority). See Art 134.2 of the Ley Concursal, reformed in 2015. 92 The Belgian system allows for a peculiar cram-down: secured creditors are not protected by a vote by classes, and can be bound by the rehabilitation plan, but there is substantive protection in the contents of the plan, which can only impose an additional stay of two years on secured creditors, including payment of interest during the stay period. In the Brazilian system, the law requires that the

Enforcement of Secured Credit

235

i nterests of creditors’ test’ are key elements in the articulation of the protection of secured creditors in rehabilitation plans. Absolute priority, in this context, implies that the hierarchy of claims cannot be altered by the rehabilitation plan to favour the interests of other claimants or stakeholders at the expense of the rights of secured creditors. The ‘best interests of creditors’ test’ means, also in this context, that the payments to be received by a secured creditor under a rehabilitation plan cannot be lower than the payments that the secured creditor would have received by liquidating the encumbered assets. Although these principles are incorporated in the international insolvency standard, they are not universally respected.93 For instance, there is no respect of the absolute priority rule under r ehabilitation procedures in Brazil. A rehabilitation plan in Brazil approved by the required majorities may provide for the payment of unsecured creditors even if the secured creditors are not fully paid. Finally, there is another topic that deserves consideration in the analysis of the position of secured creditors in rehabilitation proceedings: the regulation of postcommencement finance. Between the time of commencement of the insolvency proceeding and the conclusion of the rehabilitation plan, the grantor’s business may need to be financed. The grantor may want to sell or otherwise dispose of encumbered assets and use any cash proceeds of the sale or other disposition to run its business. In the practice developed under the US Bankruptcy Code, if the grantor is able to continue operations on the basis of using its cash receipts from the sale or other disposition of the encumbered assets, the secured creditor and the grantor will often enter into a ‘cash collateral’ stipulation, whereby the grantor uses the cash receipts in exchange for granting to the secured creditor of a security right in replacement encumbered assets. The replacement encumbered assets will often consist of a security right in new inventory, accounts receivable, and cash receipts of the grantor created post-commencement, and may often include security rights in unencumbered assets or junior security rights in encumbered assets—all granted with notice to creditors and insolvency tribunal approval. If the secured creditor and the grantor are not able to reach agreement as to such a stipulation, the grantor will usually apply to the insolvency tribunal for an following conditions for a cram-down are met: (a) holders of the majority of the total claims present at the meeting of creditors vote for approval; (b) two of the three classes of claims vote to approve the plan; and (c) at least one-third of the creditors and claims in the dissenting class vote to approve the plan. The insolvency tribunal may only approve a rehabilitation plan under these cram-down conditions if the plan does not entail a discriminatory treatment among creditors of the class which rejected it (see Art 58 of the Law 11,101/2005). 93 There are many systems in which these rules are not explicitly stated in the statutory regime (Canada is an example). In some cases, paradoxically, the lack of the best interest of creditors’ test can provide undue protection to secured creditors. That has been the case of Serbia, where the practice of coexisting multiple security rights over the same assets results in the proliferation of ‘secured’ creditors with no expectation of recovery from the proceeds. Nevertheless, the secured creditors can block many of the solutions of the insolvency process due to their classification as secured creditors, notwithstanding the fact that these creditors would receive less under a liquidation.

236

José M Garrido and Edwin E Smith

order permitting use of the cash proceeds of the encumbered assets. The insolvency tribunal will then hold a ‘contested cash collateral’ hearing at which the insolvency tribunal will decide whether to grant the grantor’s request and permit the use of cash proceeds, albeit with the secured creditor receiving what the tribunal views as ‘adequate protection,’ usually by way of replacement security rights. If at the end of the insolvency case the replacement security rights are not sufficient to compensate the secured creditor for the loss of the cash proceeds, the secured creditor is entitled to a claim in the insolvency proceeding for the loss, and that claim will be superior to all other claims of creditors in the insolvency proceeding after secured creditors’ claims are satisfied from their encumbered assets. However, the grantor may not be able to operate its business on the basis of cash receipts without borrowing additional funds. In that case, the grantor will often borrow additional funds from its existing secured creditor. But if the existing secured creditor is not willing to advance additional funds or if the terms offered by the existing secured creditor appear onerous to the grantor, the grantor may seek financing from a third party lender. The third party lender will often require, as a condition to advancing funds to the debtor, a first security right on the encumbered assets of the existing secured creditor. Under the US Bankruptcy Code, usually the insolvency tribunal will not permit such a ‘priming’ security right unless the insolvency tribunal determines that, whether by virtue of residual value in encumbered assets or otherwise, the existing secured creditor is ‘adequately protected.’94 In either case—whether the grantor is obtaining funds from its existing secured creditor or from a new third party lender (so called, ‘debtor in possession financing’)—the grantor must demonstrate what efforts it has made to obtain financing and why the financing chosen by the grantor is in the best interest of the grantor’s creditors. Furthermore, the lender will typically condition the extension of new credit to the debtor on the debtor granting to the lender, with insolvency tribunal approval, a security right in any new post-commencement assets of the grantor and in unencumbered or under-encumbered assets. Sometimes the grantor negotiates debtor in possession financing arrangements even though it has little expectation that it will actually need the new funds. Rather, it wants to indicate to trade suppliers that, if the grantor’s cash receipts are not sufficient, the grantor has a ready and available source of additional funding to pay for goods shipped or services provided postcommencement. Such debtor in possession financing commitments, often well above anticipated funding levels, are fairly common. Post-commencement financing is, arguably, the weakest link in the rehabilitation proceedings of most systems. This is evidenced by the lack of solid rules and robust practices in most systems around the world, with the exception of the system in the US. In many legal systems where rehabilitation procedures have been recently introduced, there are only laconic rules indicating that ‘new finance’ enjoys

94

US Bankruptcy Code § 364(d).

Enforcement of Secured Credit

237

priority (China),95 or that post-commencement finance is considered a cost of the rehabilitation procedure (Russia, Mexico).96 Obtaining post-commencement finance is notoriously difficult, and the legal framework leaves numerous questions unanswered. Some systems have succeeded in developing mechanisms for post-commencement finance that are effective and protect secured creditors. In Canada, for instance, post-commencement financing is allowed under the CCAA rehabilitation process, which requires the consent of secured creditors in the case that post-commencement finance would enjoy a priority over encumbered assets.

IV. Conclusion The regulation of the enforcement of security rights in insolvency represents a major intersection of secured transactions law and insolvency law. The protection of secured credit must be reconciled with the general goals of the insolvency process. Achieving the right balance between the interests of a secured creditor, on one hand, and the interests of the other creditors and the grantor, on the other, is an aspiration for all legal systems. The need for the balance is manifested in issues such as the inclusion of the encumbered assets in the insolvency estate, and the application of the stay of creditor actions to secured creditors, both in liquidation and in rehabilitation proceedings. A complex set of technical questions arises from the use of the encumbered assets, their depreciation and use, and whether costs and interest of the secured loan are covered by the priority of the secured claim. The essence of the security right is the priority afforded to the secured claim, and this priority must be respected in a sale or use of the encumbered assets, in the sale of the business as a going concern, and also in the event of a rehabilitation plan. Legal systems have developed mechanisms to protect the economic value of security rights without preventing insolvency procedures from attaining their goal of value maximisation. The creation of protective mechanisms of security rights in rehabilitation procedures represents one of the most sophisticated developments in insolvency law. The complexity of the treatment of the enforcement of security rights in insolvency goes hand in hand with its importance. The balance between the protection of secured credit and the effectiveness of insolvency procedures is difficult to achieve and benefits from a comprehensive analysis supported by the study of comparative experience.

95

See Art 75 of the Chinese Bankruptcy Law. has an interesting regime for post-commencement finance. Post-commencement loans are considered costs of the proceeding and are paid as they fall due (Art 224 LCM). However, payment should not prejudice the secured creditor unless the post-commencement loan benefited directly the encumbered assets (Art 225 LCM); preferential claims of workers, for their salaries of the last two years, always have priority (Art 224.1 LCM). 96 Mexico

238

 

13 Comparative Study on Indian Secured Transactions Law and the UNCITRAL Legislative Guide on Secured Transactions MADHUKAR R UMARJI*

In law reform, as in life, the perfect is the enemy of the good and the quest for the perfect often undermines the good already accomplished. Let this not be the case with the UNCITRAL Legislative Guide on Secured Transactions.1 Professor Roderick A Macdonald

I. Introduction The UNCITRAL Legislative Guide on Secured Transactions (the Legislative Guide) recommends that States should aim to enact legislation that is comprehensive in scope and that, as far as possible, embraces all forms of secured transactions, all categories of grantor (borrower) and secured creditor, and all types of movable assets and secured obligations. While it is not disputed that the model recommended by UNCITRAL needs to be adopted by enacting States, each State has its own legal system, which is long-standing and deep rooted and which serves the needs of trade and commerce. Enacting States need to carefully consider, weigh and balance any changes to be introduced in a well-established legal system for the purpose of adopting the recommendations of the Legislative Guide. The com-

* The Author is at present Partner, Alliance Corporate Lawyers and former Chief Legal Adviser, Indian Banks Association. 1 UNIDROIT, Uniform Law Review NS, Vol XV 2010, 2, 446.

240

Madhukar R Umarji

parative study of Indian secured transactions law with the Legislative Guide in this chapter illustrates the difficulties that may be encountered by enacting States in adopting the recommendations of the Legislative Guide in their totality. India has a legal system based on the common law of England and has statutory commercial laws that have been operative for more than a century. The basic principles of law of contract, including sales of goods, property, negotiable instruments, intellectual property rights, etc, are well established, and parties have full autonomy to enter into contracts and decide the terms of contracts as they deem fit, subject to the general principles of commercial law. The law recognises non-possessory security rights over movable property. In addition, there are multiple registration systems operative for registration of encumbrances on different kinds of property rights. Moreover, India has enacted laws for e-commerce (Information Technology Act, 2000) and the Arbitration & Conciliation Act, 1996, adopting the model laws of UNCITRAL. India has also enacted special laws for the recognition, transfer and creation of security rights over warehouse receipts, and a new law entitled the Factoring Regulation Act, 2011, for assignment of receivables in favour of factors. The secured transactions law in India is contained in various specific laws relating to property rights. Thus, security rights over immovable property are provided by the Transfer of Property Act, 1882 and possessory security by pledge of movable property is provided by the Indian Contract Act, 1872. In addition to the general laws relating to immovable and movable property, there are laws relating to specific types of property that incorporate provisions relating to security interests over various types of property, for example: (1) the Motor Vehicles Act, 1988, under which a security interest in a motor vehicle is recorded on the certificate of registration of the vehicle and no transfer of interest in such a motor vehicle is permitted without the consent of the secured creditor; (2) the Merchant Shipping Act, 1958, which makes provision for mortgage of ships and provides for registration of such mortgages and rights and obligations of the mortgagor and the mortgagee; (3) intellectual property laws, such as the Patents Act, 1970, the Designs Act, 2000, the Copyright Act, 1957, and the Trade Marks Act, 1999, which provide for the creation of a security interest in intellectual property rights and their registration. However, the primary object of the intellectual property registration system is to record title to and encumbrances on intellectual property rights. Under the general laws applicable to security interests over movable property there was no provision specifically recognising non-possessory security, but the concept of a floating charge over all movable property of a company is recognised by the Companies Act. Furthermore, courts also recognise hypothecation of goods and book debts (receivables) as security for loans and advances given by banks

Comparative Study on Indian Secured Transactions Law

241

and other financial institutions, and such security over movables can be created by agreement between the parties. While the laws recognise security interests over various kinds of property, the Indian law contained no provision for the treatment of a non-possessory security as a transfer of interest in property and for the enforcement of a security interest by a secured creditor without court intervention. Consequently, to seize and sell encumbered assets in order to pay the amount owed under a loan in default, a secured creditor had to obtain a court order, which took an unduly long time and thus resulted in the deterioration in the value of the encumbered assets.

II. Historical Background of New Secured Transactions Law in India In the early 1990s, India launched financial sector reforms based on the policies of deregulation and liberalisation. As a part of such reforms, banks and other financial institutions were directed to undertake asset classification based on income recognition principles. On account of delays in recovery of defaulted loans, banks and financial institutions had to make provisions for probable losses impacting their profitability and ability to maintain capital adequacy as required under the prudential norms prescribed by the central bank. It was therefore necessary to confer special powers of recovery of defaulted loans on banks and other financial institutions. As a part of those financial sector reforms, the Government of India considered many suggestions for new legislation, including the following: (1) out-of-court repossession and sale of assets given as security for credit for the payment of amounts owing under loans in default; (2) securitisation of financial assets by banks and other financial institutions through the issuance of debt instruments by a special purpose vehicle in favour investors; and (3) establishment of an asset reconstruction fund for the acquisition of non-performingassets of banks and other financial institutions. To expedite the process of reform, the Government decided to implement all of the above three proposals in a single law, the Securitization and Reconstruction of Financial Assets and Enforcement of Security Interest Act, 2002 (the ‘SARFAESI Act’). It is interesting to note that the Indian law on secured transactions was enacted much before the finalisation of the Legislative Guide. Still, there are many provisions in the Indian secured transaction law that, in effect, adopt the recommendations of the Legislative Guide. A comparative analysis of various provisions of the Indian secured transactions law and the relevant recommendations of the Legislative Guide is included in following sections.

242

Madhukar R Umarji

III. Comparative Analysis A. Objectives of Secured Transactions Law under the Legislative Guide The fundamental justification for secured transaction law lies in the premise that the total net wealth of an economy will increase if more secured credit is available as a complement to unsecured credit. The objectives of secured transactions law under the Legislative Guide are: (a) to promote secured credit; (b) to allow utilisation of the full value inherent in a broad range of assets to support credit in the widest possible array of secured transactions; (c) to enable parties to obtain security rights in a simple and efficient manner; (d) to provide for equal treatment of diverse sources of credit and of diverse forms of secured transactions; (e) to validate security rights in assets that remain in the possession of the grantor; (f) to enhance predictability and transparency with respect to rights serving security purposes by providing for registration of a notice in a general security rights registry; (g) to establish clear and predictable priority rules; (h) to facilitate enforcement of creditors’ rights in a predictable and efficient manner; (i) to balance the interests of affected persons; (j) to recognise party autonomy; and (k) to harmonise secured transactions laws, including conflict-of-laws rules.

B. Objectives and Reasons of the Indian Secured Transactions Law The financial sector has been one of the key drivers in India’s efforts to achieve success in rapidly developing its economy. While the banking industry in India is progressively complying with the international prudential norms and accounting practices, there are certain areas in which the banking and financial sector do not have a level playing field as compared to other participants in the financial markets in the world. There is no legal provision for facilitating securitisation of financial assets of banks and financial institutions. Further, unlike international banks, the banks and financial institutions in India do not have power to take possession of securities and sell them. Our existing legal framework relating to c ommercial transactions has not kept pace with the changing commercial practices and

Comparative Study on Indian Secured Transactions Law

243

nancial sector reforms. This has resulted in slow pace of recovery of defaulting fi loans and mounting levels of non-performing assets of banks and financial institutions. Narasimham Committee I and II and Andhyarujina Committee,2 constituted by the Central Government for the purpose of examining banking sector reforms, have considered the need for changes in the legal system in respect of these areas. These committees, inter alia, have suggested enactment of a new legislation for securitisation and empowering banks and financial institutions to take possession of the securities and sell them without the intervention of the Court. Acting on these suggestions, the Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest Ordinance, 2002 was promulgated on 21 June 2002, to regulate securitisation and reconstruction of financial assets and enforcement of security interest and for matters connected therewith or incidental thereto. The provisions of the Ordinance would enable banks and financial institutions to realise long-term assets, manage problem of liquidity, asset liability mismatches and improve recovery by exercising powers to take possession of securities, sell them and reduce non-performing assets by adopting measures for recovery or reconstruction. Upon approval by the Indian Parliament, the Ordinance was passed as an Act of Parliament, effective from the date of the Ordinance. Although the Indian secured transactions law was not enacted with the object of facilitating growth of credit, in effect this objective is achieved because with the comfort of speedy recovery of amounts owed under loans in default, credit particularly to micro-, small- and medium-sized enterprises and to individual borrowers grew steadily. Furthermore, the Indian secured transactions law also achieved the other objectives of facilitating the creation of security interests over a broad range of assets in a simple and efficient manner, and the recognition of non-possessory security interests and party autonomy. One other significant feature of the Indian secured transactions law is the provision for setting up asset reconstruction companies (ARCs) for the purpose of acquiring non-performing loans from banks and other financial institutions with a view to restructuring the debt and reviving business enterprises in financial distress. Any company registering with and obtaining a licence from the Bank of India can set up an ARC. Such ARCs acquire non-performing loans from banks and other financial institutions with debt instruments, which they issue and pay upon realisation of the non-performing loans. Unlike the asset reconstruction funds set up in some other jurisdictions, in India ARCs do not have any support or guarantee from the Government. Currently, there are 14 ARCs operating in India.

2

The author was a member of this Committee.

244

Madhukar R Umarji

C. Scope of Application of Secured Transactions Law under the Legislative Guide The secured transactions law recommended in the Legislative Guide applies to: (1) security rights in all types of movable asset, tangible or intangible, present or future, including inventory, equipment and other tangible assets, contractual and non-contractual receivables, contractual non-monetary claims, negotiable instruments, negotiable documents, rights to payment of funds credited to a bank account, rights to receive the proceeds under an independent undertaking and intellectual property; (2) security rights created or acquired by all legal and natural persons, including consumers, without, however, affecting rights under consumer protection legislation; (3) security rights securing all types of obligations, present or future, determined or determinable, including fluctuating obligations and obligations described in a generic way; and (4) all property rights created contractually to secure the payment or other performance of an obligation, including transfers of title to tangible assets for security purposes, assignments of receivables for security purposes, the various forms of retention-of-title agreements and financial leases. The law should also apply to security rights in proceeds of encumbered assets.

D. Types of Property Over Which a Security Interest Can be Created under the Indian Secured Transactions Law The Indian secured transactions law defines ‘property’ in broad terms and includes tangible and intangible, as well as immovable, property. Property is defined as follows: ‘property’ means: (i) (ii) (iii) (iv) (v)

immovable property; movable property; any debt or any right to receive payment of money, whether secured or unsecured; receivables, whether existing or future; intangible assets, being know-how, patent, copyright, trade mark, licence, franchise or any other business or commercial right of similar nature.

It is therefore clear that a security interest can be created in any kind of property including future receivables. Rights to payments under independent undertakings are not included in the definition of property because such rights are coupled with other credit availed from banks. To illustrate this point, an exporter avails credit from a bank and sells his product. The buyer’s bank establishes a letter of credit in favour of the exporter

Comparative Study on Indian Secured Transactions Law

245

and the payment to be made by the buyer’s bank under the letter of credit has to be routed through the exporter’s bank for repayment of the credit availed. In such a case, the exporter cannot be allowed to borrow money against the security of the receivable under the letter of credit. If such rights to payment are segregated from linkages with credit availed from banks and treated as separate and distinct property rights, the lending activities of the banks will be adversely affected. However, rights to receive money, including a right to payment under an independent undertaking, is a property right and hence a security interest can be created in such rights. However in India borrowing against security in funds laying in any bank account or a right to payment under an independent undertaking is not permitted without the consent of bank or other financial institution holding the deposit, or the bank or other financial institution issuing the independent undertaking.

i. Possessory Security Interests The basic law of pawn/pledge is contained in the Indian Contract Act, 1872 and is well settled law. Principles of law relating to the creation of security interests in movable property by giving possession of the property to the lender are applicable to loans against the pledge of shares and other securities. With the adoption of new technology making possible the holding of securities in dematerialised form in a depository, the pledge or hypothecation of dematerialised securities is foreseen in the Depositories Act, 1996. To facilitate creation of possessory security interests in agricultural produce, commodities and other goods, warehouse receipts can be issued as negotiable documents making it possible for the lender to obtain a security interest in the goods by taking possession of the goods covered by the warehouse receipt under the Warehousing (Development and Regulation) Act, 2007. Similar provisions have been enacted for declaring a multi-modal transport document a negotiable document incorporating title to the goods covered by the document. As they are covered in special legislation, the SARFAESI Act, that is, the secured transactions law in India, excludes from its scope possessory security interests in securities and negotiable documents.

ii. Assignment of Receivables Rights to payment under negotiable instruments are governed by the Negotiable Instruments Act, 1881. Under this Act, it is permissible to create a security interest in a right to payment under a negotiable instrument. With regard to other rights to receive payments, a separate law, the Factoring Regulation Act, 2011, has been enacted. This Act provides for both the outright assignment of, and the creation of, a security interest in receivables for borrowing against such receivables. Banks and non-banking financial institutions registered as factors with the Reserve Bank of India can undertake factoring. Like the SARFAESI Act, the Factoring Regulation Act applies only to assignment of receivables by a factor or other financial institution. But with regard to the defini-

246

Madhukar R Umarji

tion of ‘receivable’ and the exclusion of certain categories of receivables, Indian law has adopted the principles and concepts contained in the UN Convention on the Assignment of Receivables in International Trade.3 Generally, although Indian law is not totally in conformity with the Convention, it has adopted a substantial part of the Convention and thus facilitates the assignment of receivables. As a receivable is a property right that can be assigned in the context of important financing transactions, it is worthwhile noting the salient features of the Indian Factoring Regulation Act, 2011. Under the Act: (1) ‘receivables’ means all or part of or undivided interest in any right of any person under a contract including an international contract where either person under a contract including an international contract where either the assignor or the debtor or the assignee is situated or established in a State outside India; to payment of a monetary sum whether such right is existing, future, accruing, conditional or contingent arising from and includes, any arrangement requiring payment of toll or any other sum, by whatever name called, for the use of any infrastructure facility or services; (2) rights to receive payment cannot be assigned, if they arise under: (i) a merger, acquisition or amalgamation of business activities or a sale or change in the ownership or legal status of the business; (ii) any transaction on a stock exchange or commodities exchange regulated by the Securities and Exchange Board of India Act of 1992 (Act 15 of 1992) or by the Forward Market Commission under the Forward Contracts (Regulation) Act of 1952 (Act 74 of 1952), respectively; (iii) any financial contract governed by a netting agreement, except a receivable owned on the termination of all outstanding transactions; (iv) any foreign exchange transaction except receivables in a foreign currency; (v) any inter-bank payment system, inter-bank payment agreement or clearance and settlement system relating to securities or other financial assets or instruments; (vi) any bank deposits; (vii) a letter of credit or independent guarantee; (viii) any rights and obligations of a person under the law governing negotiable instruments or negotiable documents (including negotiable warehouse receipts), instruments that are; (ix) any sale of goods or services for personal, family or household purposes; (x) any assignment of loan receivables by a bank or non-banking financial institution to another bank or non-banking financial institution;

3

New York, 2001.

Comparative Study on Indian Secured Transactions Law

247

(xi)

(3)

(4)

(5) (6)

(7)

(8)

any securitisation transaction (including the assignment of receivables to special purpose vehicles or trusts that issue securities against such receivables, bought from a single debtor or single group of debtors). the assignment of receivables can be effected by agreement in writing between any business enterprise in favour of any factor, bank or non-banking financial institution; upon assignment of a receivable, the assignee can demand payment from the debtor only if notice of such assignment is served on the debtor by the assignor; upon payment of a receivable to the assignee after the debtor is notified of the assignment by the assignor, the debtor is discharged; if the debtor makes payment to assignor, the assignor is required to hold the proceeds in trust for the benefit of the assignee and pay forthwith the assignee; upon receiving notice of assignment, the debtor is required to intimate to the assignee any deposits or advances paid by him to the assignor and the debtor cannot get a discharge unless payment is made to assignee; an assignment agreement cannot alter the terms of contract between the debtor and the assignor.

iii. Warehouse Receipts Under the Warehousing (Development & Regulation) Act, 2007, a warehouse receipt is a negotiable document. Thus, a security interest in goods covered by a warehouse receipt may be created by endorsement and delivery of the warehouse receipt.

iv. Intellectual Property Rights Under Indian secured transactions law, an intellectual property right is covered by the definition of property. Thus, a security interest may be created in an intellectual property right in the same way as in any other type of movable property.

v. Immovable Property The Legislative Guide is not applicable to security interests in immovable property. This is so because in many federal States, immovable property is the subject of state, rather than federal, law. In India immovable property law is state law, but during British Rule, certain commercial laws such as the Transfer of Property Act of 1882, the Indian Contract Act of 1872, the Negotiable Instruments Act of 1881 applied throughout India. After the adoption of the Constitution of India, such laws continued to apply, and legislative power in regard to such laws can be exercised only by the Central (Federal) Government. India therefore has commercial laws that are uniform in all provinces, despite the fact that it is

248

Madhukar R Umarji

a federal State and each province has separate and distinct legislative p owers. As immovable property is included in the definition of ‘property’, Indian secured transactions law applies to security interests in immovable property also and, among other things, confers powers of extra-judicial enforcement of such security interests on secured creditors, defined as banks or other financial institutions.

vi. Movable Property As mentioned above, with few exceptions, all types of movable property are included in the definition of ‘property’ in the Indian secured transactions law, which thus applies to possessory and non-possessory security interests in movable property. Such movable property includes raw material, semi-finished and finished goods, receivables arising from the sale of goods or services, vehicles, plant, machineries, equipment, crops, live-stock etc. Furthermore, company law recognises a floating charge over all movable property and receivables of a company. Company law also provides for registration of such floating charge with the Company Registry. Company law also provides that, if such charge is not registered, the creditor cannot claim rights as a secured creditor in the event of winding up (liquidation) proceedings against the company. But the registration system under company law is not applicable to non-corporate borrowers, who avail credit against the security in movable property. Generally, Indian secured transactions law is in conformity with the recommendations of the Legislative Guide except with regard to the registration of security interests, which is dealt with separately below.

vii. Applicability to All Legal and Natural Persons Indian secured transactions law was enacted for the purpose of empowering banks and other financial institutions to enforce security interests to recover amounts owing under loans in default. The definition of secured creditor under Indian law covers only banks and other financial institutions; the law is not applicable to any other lenders that are not banks or other financial institutions as defined under the SARFAESI Act. This is one major deviation from the recommendations of the Legislative Guide. In order to make Indian law conform to Legislative Guide standards it is necessary to extend the Indian law to all legal and natural persons who are secured creditors. Such reforms are not likely in the near future, because of concerns that lenders other than banks and other financial institutions might abuse powers of extra-judicial enforcement. However, it is possible that the law will be amended to recognise secured credit extended by all persons, and to extend the registration system to such credit and the rights of priority over all other creditors, without giving the power of extra-judicial enforcement. In India prior to the enactment of the SARFAESI Act there was no law defining security interest in movable property and the rights of the creditor were not clearly defined. With the enactment of the SARFAESI Act, 2002, a generic, unitary and exclusive concept of security interest, and also of property, has been

Comparative Study on Indian Secured Transactions Law

249

defined, thereby recognising transfer of interest in movable property to secure due repayment of money advanced or to be advanced. The new law extends to major lenders, ie banks, housing finance companies and public financial institutions and covers nearly 90 per cent of secured lending transactions.

E. Types of Obligation In terms of the Legislative Guide, security interests securing all types of obligation, present or future, should be covered by the secured transactions law. In India, a security interest over personal property can be created to secure any credit granted or to be granted, and also for any contingent liabilities, such as guarantees that become effective in the event of default. Thus the law in India is in conformity with the Legislative Guide.

F. Limitations on Scope i. Under the Legislative Guide The Legislative Guide provides that the Guide shall not apply to: (1) aircraft, railway rolling stock, space objects and ships, as well as other categories of mobile equipment in so far as such asset is covered by a national law or an international agreement to which the State enacting legislation based on these recommendations is a party and the matters covered by this law are addressed in that national law or international agreement; (2) intellectual property in so far as the provisions of the law are inconsistent with national law or international agreements, to which the State is a party, relating to intellectual property; (3) securities; (4) payment rights arising under or from financial contracts governed by netting agreements, except a receivable owed on the termination of all outstanding transactions; and (5) payment rights arising under or from foreign exchange transactions.

ii. The Position under Indian Law Certain categories of property rights are not covered by Indian secured transactions law. These are: (1)

4 5

a lien on any goods, money or security given by or under the Indian Contract Act, 18724 or the Sale of Goods Act, 19305 or any other law for the time being in force; 9 of 1872. 3 of 1930.

250

Madhukar R Umarji

(2)

a pledge of movables within the meaning of section 172 of the Indian Contract Act, 1872;6 (3) a security interest in any aircraft as defined in clause (1) of section 2 of the Aircraft Act, 1934;7 (4) a security interest in any vessel as defined in clause (55) of section 3 of the Merchant Shipping Act, 1958;8 (5) a conditional sale, hire-purchase or lease or any other contract in which no security interest has been created; (6) a right of an unpaid seller under section 47 of the Sale of Goods Act, 1930;9 (7) a security interest in any asset that is not subject to attachment (excluding asset specifically charged with the debt recoverable under this Act or a sale under the first proviso to subsection (1) of section 60 of the Code of Civil Procedure, 1908.10 (8) a security interest for securing repayment of any financial asset not exceeding 100,000 rupees; (9) a security interest created in agricultural land; (10) a security interest securing an amount that is less than twenty per cent of the principal amount and interest thereon. Limitations on the applicability of the secured transaction law in India are along the lines of the limitations on scope recommended in the Legislative Guide. However, in addition to the recommended limitations on the scope, the following property rights are excluded under Indian law: (1) security interests in agricultural land: Indian law excludes security interests in agricultural land from the applicability of the secured transactions law. The reason for such exclusion is that agricultural land and security interests over such land are subjects in the State list and the parliament has no legislative power to enact a law on agricultural land; (2) conditional sale, hire-purchase or lease—acquisition financing: In terms of the Legislative Guide, hire-purchase, financial lease or any other transaction aimed at providing finance for the acquisition of assets amounts in substance to a secured transaction under which a security interest is created in the asset given on hire or lease or sold with a repayment schedule specified by the seller. While the recommendations of the Legislative Guide with regard to acquisition financing are based on sound legal principles and cannot be disputed, in India they are kept outside the purview of secured transactions law for other valid reasons, including the following: (a) hire-purchase, financial lease or retention-of-title sales on credit are recognised as title transactions and in the event of default, the hirer, 6

9 of 1872. 24 of 1934. 44 of 1958. 9 3 of 1930. 10 5 of 1908. 7 8

Comparative Study on Indian Secured Transactions Law

251

lessor or the seller is permitted to exercise his ownership rights to repossess the assets, without the intervention of the courts. Financial institutions engaged in such business are not financial institutions for the purposes of secured transactions law and cannot exercise powers of enforcement of securities. But the courts recognise that those institutions can repossess assets owned by them on the basis of title retention.11 (b) The law applicable to acquisition financing applies to all types of creditor and is not restricted to banks and other financial institutions.

G. The Registry System i. Under the Legislative Guide The Legislative Guide provides that the purpose of provisions on the registry system is to establish a general security interest registry and to regulate its operation. Thus, the purpose of the registry system is to provide: (1) a method by which an existing or future security interest in a grantor’s existing or future assets may be made effective against third parties; (2) an efficient point of reference for priority rules based on the time of registration of a notice with respect to a security interest; and (3) an objective source of information for third parties dealing with a grantor’s assets (such as prospective secured creditors and buyers, judgment creditors and the grantor’s insolvency representative) as to whether the assets may be encumbered by a security interest. To achieve this purpose, under the Legislative Guide, the registry system should be designed to ensure that the registration and searching process are simple, timeand cost-efficient, user-friendly and publicly accessible.

ii. The Position Under Indian Law Indian law contains provisions for the purpose of the registration of security interests in movable property. However, when the law was enacted in 2002, no registry was established and the law became operative without requiring registration. The registry as provided in law was established in March 2011, but initially only mortgages were subject to registration and what was actually registered was the

11 Decisions of the Supreme Court on the issue include Trilok Singh v Satya Deo Tripathi AIR 1979 SC 850; Charanjit Singh Chadha v Sudhir Mehra (2001) 7 SCC 212; Orix Auto Finance (India) Ltd v Shri Jagmandar Singh (2006) SCCL.COM82.

252

Madhukar R Umarji

mortgage loan document. Hence, the secured transactions law in India is operative without a registration system. The reasons for this are as follows: (1) there are title-based registries operating in the system under other laws for registration of title to property and any encumbrances thereon, for immovable property, motor vehicles, intellectual property rights and properties belonging to companies; (2) the secured transactions law provides that the validity and priority of security interests shall be determined by the respective laws and the registration system under the secured transaction law shall not affect such validity or priority; (3) the legal regime of security interests in movable property (goods or receivables) operates on trust and recognises the rights of a grantor (borrowers) to sell such property in the ordinary course of business; (4) after the enactment of the secured transactions law, secured creditors are exercising powers of repossessing movable property, such as goods, plant, equipment, machineries, etc. However, if there are any sales outside the ordinary course of business, secured creditors do not pursue buyers to assert their rights in assets that are in the hands of third parties. This is so because making such claims against buyers involves litigation in the courts and judicial intervention, which takes a long time; (5) where security interest in movable property such as motor vehicles, corporate registry or intellectual property rights are registrable in a special registry, secured creditors are in a position to demand satisfaction of their claims before the transfer of title to buyers.

iii. Reforms Required in the Registration System of India After the establishment of a Central Registry for recording security interests created in favour of only banks or other financial institutions, it was accepted that the registry record is incomplete. It is necessary that security interest in any type of movable property created by any person needs to be registered, so that any encumbrance on property is disclosed to persons dealing with such property. The Central Registry in India, with the assistance of the World Bank, appointed a Legal Working Group to suggest further reforms in the registration system. The Working Group12 has made recommendations for legislative and other administrative reforms to extend the registration system to all secured creditors, as well as to security interests created on all kinds of property rights. The recommendations are being considered by the Government of India.

12

The author of this chapter was the Chairman of the Legal Working Group.

Comparative Study on Indian Secured Transactions Law

253

H. Third-party Effectiveness and Priority to Secured Creditor There are no provisions in the secured transactions law of India dealing with methods of making a security interest effective against third parties and establishing its priority. But the general property law recognises that the creation of a security interest in movable property is effective from the date of creation. Thus, a secured creditor has priority over all other claimants, with the exception of claimants’ rights that were created prior to creation of the security interest of that secured creditor and claimants with priority established by law. With regard to the priority of a security interest secured creditors over claims of the State, the position in Indian law is as follows: (1) generally, dues to the Government, ie tax dues, land revenue, etc, have priority over ordinary debts, ie unsecured creditors. A security interest has priority over the claims of unsecured creditors as well as claims of the State; (2) certain tax or other claims of the State are, by law, given priority over security interests; (3) in cases of insolvency, assets subject to a security interests are not part of the insolvency estate and a secured creditor has a right to enforce its security interest in individual proceedings, but has to bear the costs of preservation of the assets.

I. Extra-judicial Enforcement by Secured Creditors i. Under the Legislative Guide The Legislative Guide recommends that the exercise of rights and performance of obligations in the context of enforcement shall be in good faith and in a commercially reasonable manner. Such obligation cannot be waived by agreement between parties. The Guide also provides that any person aggrieved by the action of a secured creditor in the context of enforcement can claim damages and any other relief.

ii. The Position Under Indian Law Since the focus of Indian law is to empower banks and other financial institutions to exercise powers of extra-judicial enforcement, the above recommendations are incorporated in the law. Moreover, since the law extends extra-judicial powers, procedural requirements for the sale of encumbered assets are prescribed by rules to protect the interests of grantors (borrowers), which are as follows: (1) action for the enforcement of a security interest can be taken in respect of a loan in default (classified as non-performing loans, as per the directions of the Reserve Bank of India);

254

Madhukar R Umarji

(2) a 60-day notice calling upon the grantor to repay the loan has to be issued by the secured creditor; (3) if the grantor/borrower raises any objection to the notice, the secured creditor is required to reply to such objection; (4) if the grantor/borrower fails to pay, the secured creditor can issue a further notice calling upon the borrower to hand over possession of the encumbered assets; (5) the secured creditor can take possession of an encumbered asset if it is handed over voluntarily and peacefully by the grantor/borrower. Otherwise, the secured creditor has to file an application before the District Magistrate or Chief Metropolitan Magistrate to take possession and hand over the encumbered assets to the secured creditor. Public notice of repossession has to be published in English and local language newspapers; (6) the grantor/borrower has the right to file an application against the action of the secured creditor before the Debt Recovery Tribunal and, if the Tribunal decides that possession of encumbered assets was taken wrongfully, it can restore possession and also award damages for wrongful repossession; (7) after taking possession, the secured creditor shall obtain a valuation and set the price of the encumbered assets; (8) the secured creditor can sell the encumbered assets by public auction issuing an advertisement and inviting offers from the public. Private sales are also permitted, but only with the consent of the grantor/borrower; and (9) if any Tribunal or other authority is not acting in accordance with the law and in a fair and reasonable manner, the grantor/borrower also has the right to file a writ petition in the High Court seeking directions to the concerned authority. The right to file writ petitions for enforcement of fundamental rights under the Indian Constitution and other statutory rights is provided by the Constitution and is an effective remedy against any unfair and illegal action by any authority acting under a statute.

IV. Modifications Required in Indian Secured Transactions Law to Conform with the Legislative Guide In India, secured transactions law has achieved the following objectives: (1) banks and other financial institutions have the comfort of extra-judicial recovery powers for the enforcement of security interests, which creates an environment conductive to an increase in available credit; (2) there is better discipline in repayment of loans, thanks to the powers of extra-judicial enforcement available to secured lenders;

Comparative Study on Indian Secured Transactions Law

255

(3) there are procedural safeguards to ensure that creditors act in good faith and in a commercially reasonable manner. While Indian secured transactions law has certain unique features, adapted to suit the Indian legal system, major modifications are required to bring it in conformity with basic principles recommended by the Legislative Guide. The necessary modifications include: (1) the applicability of the law needs to be extended to all secured creditors so that all secured transactions are governed; (2) the registration system needs to be extended to all security interests created over any kind of movable property (except ships, aircraft, etc, excluded by the Legislative Guide), so that data of all encumbrances on movable property is available in the Registry for the benefit of any person dealing with such property; (3) security interests over any type of movable property should be effective from the date and time of registration; (4) powers of secured creditors to enforce security interests without the intervention of the courts need to be linked to registration, which should be made optional; (5) all existing registration systems need to be integrated with the Central Registry under the secured transactions law; (6) security interests should be given priority over all unsecured claims, including claims of the State and employees. Any exceptions should be clearly specified by law.

V. Conclusion In conclusion, the extent to which Indian secured transactions law conforms with the recommendations of the Legislative Guide, and suggestions for reforms, are set out below in tabular form. Sr. Recommendation No. of the Legislative Guide 1

Law to extend to all forms of secured transactions.

Status of Indian Law

Indian law defines ‘security interest’ as any right, title or interest of any kind whatsoever upon property, created in favour of any secured creditor (s 2(1)(zd)).

Deviation from Suggestion for the Legislative Reforms Guide No deviation

No change required.

(continued)

256

Madhukar R Umarji

Sr. Recommendation No. of the Legislative Guide

Status of Indian Law

Deviation from Suggestion for the Legislative Reforms Guide

2

Law should adopt a functional approach, which should cover rights of sellers or financial lessors under various forms of retention of the title agreements and financial leases.

Indian law excludes retention of title agreements from the applicability of secured transactions law (s 31(e)).

Title retention contracts and sale on credit are not treated as equivalent to secured credit

Indian law needs to adopt the functional approach and cover retention of title agreements as secured transactions.

3

Law extends to all categories of grantors (borrowers) and all secured creditors

Indian law defines a secured creditor as any bank or financial institution (s 2(1)(zd)) and does not cover secured creditors who are not banks or financial institutions.

Indian law is not comprehensive and does not extend to all secured creditors.

Reforms need to be undertaken to make the law comprehensive.

4

The law should cover security rights over all type of movable assets and obligations

Indian law defines No deviation. property in wide terms to include immovable, movable and intangible property and also includes any receivable present or future, any debt or any right to receive payment

No change required.

5

A registry system may be established for registration of notices of security rights in existing or future assets of the grantor (borrower) and provide priority to the secured creditor based on date and time of registration of notice, and make the security rights effective against third parties.

Indian law makes provision for establishment of a Central Registry to record notices of creation of security interests. But the secured transactions law is operative without any registration requirement, except registration of mortgage by deposit of title deeds effective from 31 March, 2011.

Indian law needs to be modified to adopt the modern registration system and introduce provisions for third party effectiveness and priority to secured creditors from the date and time of registration.

Secured transactions law in India is governed by the general law of contracts and transfer of property in regard to third party effectiveness, registration requirements and priority to secured creditors.

(continued)

Comparative Study on Indian Secured Transactions Law Sr. Recommendation No. of the Legislative Guide

Status of Indian Law

257

Deviation from Suggestion for the Legislative Reforms Guide

6

Law should provide powers of extrajudicial enforcement to secured creditors to sell or otherwise dispose of lease or license encumbered assets.

Indian law provides for extra-judicial enforcement including powers for take-over of management of the business of the borrower, and also restructure the debt coupled with conversion of part of the debt into equity.

No deviation.

No change required.

7

Law should provide that enforcement of rights and performance of obligations by parties under the provisions on enforcement should be in good faith and in a commercially reasonable manner.

Indian law provides all No deviation. the procedural safeguards to ensure that secured creditors act in a fair and reasonable manner.

No change required, except extending the powers of extra-judicial enforcement to all secured lenders.

8

Any person aggrieved by actions for enforcement should have the right to approach judicial authority for relief against secured creditor.

Indian law provides for No deviation. application to the Debt Recovery Tribunal and an appeal to Appellate Tribunal against any action of secured creditor from the stage of taking possession of secured assets.

No change required.

Lastly, enacting States need to appreciate that secured credit is the driving force of the economy and a blessing. To promote the growth of credit, enacting States must focus on protecting the interests of secured lenders and recognise the superpriority of secured creditors over all other claimants, including claims of the Revenue (Crown Debts). If a legal environment conducive to secured lending is in place, it will create wealth, generate employment and result in overall growth in the economy.

258

Madhukar R Umarji

VI. Recent Reforms in India—an Addendum The secured transactions law in India has recently (August 2016) been amended to widen the scope of the law and introduce major reforms that are in conformity with the recommendations of the UNICITRAL Legislative Guide. The reforms undertaken are as follows: (1) hire purchase, financial lease and conditional sale transactions are included in the definition of security interest and other consequential changes are made in the law; (2) a debenture trustee in whose favour a security interest is created to secure redemption of debt securities on due dates by the company raising funds is also included in the definition of secured creditor; (3) secured claims have priority over all other claims, including tax claims of the Central and State Governments and local authorities; (4) the registration of a security interest is made effective from the date and time of registration. Failure to register is not an offence and registration is made optional. However the right of enforcement of security without the intervention of the court is available only to registered security interests of banks and financial institutions. (5) orders of attachment of assets obtained from any court or issued by any tax authority may also be registered under the law. The secured transactions law in India, the Securitization and Reconstruction of Financial Assets and Enforcement of Security Interest Act, 2002, has become an effective tool for recovery of loans in default for banks and financial institutions. Such remedies result in better loan repayment discipline amongst borrowers and also trigger creation of wealth, generating employment and growth in the economy.

14 Security Interests over Intellectual Property Rights in Italy: Critical Analysis and Reform Proposals ANDREA TOSATO*

I. Introduction I met Professor Roderick Macdonald for the first time in 2008, at the 14th session of UNCITRAL Working Group VI. He was a long-serving member of the Canadian delegation, while this was my first experience as part of the Italian contingent. In this context, I had the privilege of working with him for six years. The Working Group sessions are brief and industrious. Nevertheless, on these occasions, I had plenty of opportunities to converse, discuss, debate and laugh with Professor Macdonald. Above all, I had the opportunity to learn from him. Professor Macdonald’s mastery of the law was formidable. He possessed both a systematic command of commercial law and specialised knowledge of security interest law. This was combined with a keen sensitivity for public policy issues and a sophisticated understanding of the diverse challenges faced by both civil and common law jurisdictions. One of the most esteemed participants of Working Group VI, his interventions were always precise and insightful. He was often pivotal in bridging diverging views and achieving consensus. Professor Macdonald’s human virtues, however, were even more outstanding than his legal expertise. He had a jovial sense of humour and an infectious laugh. He was passionate about his views, yet always open to others’ ideas. Approachable and generous with his time, I personally sought his counsel on many occasions, and benefited from his intellectual honesty and candour. He was a truly inspiring figure. I count myself fortunate to have been his colleague and friend. As I first met Professor Macdonald during the negotiations for the UNCITRAL Supplement on Security Rights in Intellectual Property,1 I thought it fitting to *

Assistant Professor, University of Nottingham. The law is stated as it stood on 31 April 2015. by UNCITRAL in July 2010, Report of UNCITRAL on the work of its forty-third session, A/65/17, para. 227. This was a supplement to the UNCITRAL Legislative Guide on Secured Transactions, adopted by UNCITRAL in December 2007, Report of UNCITRAL on the work of its 1 Adopted

260

Andrea Tosato

return to the topic of security interests over intellectual property rights (henceforth ‘IPRs’) for my contribution to this volume. As knowledge increasingly becomes the key factor of production in developed economies, IPRs are emerging as some of the most valuable property owned by businesses. Against this landscape, there is a burgeoning interest in deploying these assets as collateral to reduce the cost of credit.2 Taking security over IPRs presents atypical legal challenges.3 Personal property security law conceptually arose in ancient times and flourished around tangible goods, documents of title and receivables over the past two centuries.4 By contrast, intellectual property law originated in the seventeenth to eighteenth century, focusing predominantly on requirements for protection, scope of protection, enforcement and, to a limited extent, transfer of title. Sophisticated proprietary dealings in IPRs, such as the taking of security, did not receive equal attention from legislatures and suffered from doctrinal underdevelopment.5 Consequently, the intersection of personal property security law and intellectual property law demanded by modern commercial realities gives rise to a panoply of conceptual difficulties. This chapter endeavours to analyse the manner in which Italian law addresses these issues. It will commence by identifying the relevant normative elements of the Italian real consensual security framework and intellectual property statutes. Subsequently, it will chart the fashion in which these two bodies of rules intersect. This will be followed by a critical assessment of the resulting Italian legal regime for the taking of security over IPRs, focusing in particular on legal certainty and commercial palatability. Lastly, the possibility of legislative reform will be considered, contemplating interventions affecting both the extant general personal property security law framework and the specific intellectual property statutes. The scope of this enquiry will be subject to three limitations. Firstly, it will not cover non-consensual security interests. Secondly, it will not extend beyond resumed fortieth session, A/62/17 (Part II) paras 99–100. For in depth commentary see O Akseli, International Secured Transactions Law: Facilitation of Credit and International Conventions and Instruments (Routledge, 2011); on the UNCITRAL Supplement on Security Rights in Intellectual Property see S Bazinas, ‘Intellectual Property Financing under the Supplement to the UNCITRAL Legislative Guide on Secured Transactions’ (2011) UCC Law Journal 601. 2 For an empirical study see M Bezant and R Punt, ‘The Use of Intellectual Property as Security for Debt Finance’ (1997) IPQ 297; I Davies, ‘Technology-based small firms and the commodification of intellectual property rights’ in J de Lacy (ed), The Reform of UK Personal Property Security Law: Comparative Perspectives (Routledge-Cavendish, 2010) 308. 3 This chapter will not address the valuation difficulties that IPRs raise or other difficulties that emerge at the credit appraisal stage. For an overview of these matters, see R Ghafele, ‘Accounting for IP?’ (2010) Journal of Intellectual Property Law & Practice 521; T Hberden, ‘Intellectual Property Valuation and Royalty Determination’ in A Liberman, P Chrocziel, R Levine, International Licensing and Technology Transfer: Practice and the Law (Wolters Kluwers, 2011) ch 4. 4 WJ Zwalve, ‘A labyrinth of creditors: a short introduction to the history of security interests in goods’ in EM Kieninger (ed), Security Rights in Moveable Property in European Private Law (Cambridge, Cambridge University Press, 2004) 38. 5 For a common law perspective, see L Bently and B Sheeman, The Making of Modern Intellectual Property Law (Cambridge, Cambridge University Press, 1999); for a civil law perspective see R Franceschelli, Trattato di diritto industriale (Giuffrè, 1973).

Security Interests over Intellectual Property Rights in Italy

261

patents, trademarks6 and copyright. Analysing the taking of security over all types of IPRs recognised by Italian law would require an extremely lengthy contribution that is best reserved for a monographic study. Nevertheless, this exclusion must not be interpreted as a suggestion that IPRs not included herein are commercially unpalatable collateral or legally ill suited. Thirdly, it will not encompass the law governing agreements that achieve the functional equivalent of granting security over IPRs without actually creating a security interest in them, as these transactions present issues different and ulterior from those under consideration.7

II. Legal Framework for Taking Security over Intellectual Property Rights Italy is a civil law jurisdiction, the private law of which is characterised by Roman law roots, a marked influence of the Napoleonic Code Civil and assonances with the German Bürgerliches Gesetzbuch.8 The history of Italian personal property security law commences with the 1865 post-unitary civil code, in which real consensual security devices echoed those present in the 1804 Napoleonic Code Civil. This legal framework was subsequently amended by the currently in force 1942 Italian Civil Code, yet the pre-existing legal edifice was not radically restructured.9 Italy has historically regulated IPRs by enacting dedicated statutes. Interestingly, legislation dating back to the nineteenth century already contained provisions addressing aspects of the taking of security over these assets.10

A. Real Consensual Security Framework Italian personal property security law is founded upon two cardinal axioms.11 First, debtors are liable with all their assets (present and future) for the satisfaction

6 The locution ‘trademark’ is used throughout this paper to refer to registered trademarks exclusively. 7 An example of such a transaction is the transferring of IPRs into a special purpose vehicle, followed by the offering of the shares of this special purpose vehicle as collateral. On this issue see A Tosato, ‘Security interests over intellectual property’ (2011) Journal of Intellectual Property Law & Practice 93. 8 A Padoa Schioppa, ‘Dal Code Napoléon al codice civile del 1942’ (1993) Riv Dir Civ 531. 9 For an overview of Italian personal property security law see G Ferrarini, ‘Changes to personal property security law in Italy: a Comparative and Functional Approach’ in R Cranston, Making Commercial Law: Essays in Honour of Roy Goode (Oxford, Oxford University Press, 1997) 477; P Wood, Comparative Law of Security Interests and Title Finance (Sweet & Maxwell, 2007) 451. 10 A Tosato, ‘Garanzie Reali e Diritti IP: Per Un Censimento Dei Materiali’ (Real Consensual Security Interests and IP Rights: A Survey of the Relevant Legal Sources in Italy) (2009) IDA 557. 11 For an exhaustive systematic overview see A Veneziano, ‘Italy’ in H Sigman and EM Kieninger, Cross-Border Security over Tangibles (Sellier, 2007) 159.

262

Andrea Tosato

of their outstanding obligations.12 Secondly, creditors are to be satisfied pari passu upon default of their common debtor (par condicio creditorum).13 The Italian Civil Code provides three real security instruments that outflank the pari passu principle: pegno, ipoteca and privilegio.14 While pegno is purely consensual in nature, ipoteca and privilegio can be both consensual and non-consensual. These security devices are typified and a numerus clausus. Parties are granted minimal freedom to diverge from statutory rules governing pegno, ipoteca and privilegio by way of agreement, and cannot contractually create ulterior security interests.15 These security devices differ significantly from each other, and are regulated by dedicated statutory provisions. Nevertheless, they are all built on a common normative foundation, as they are underpinned by a set of general provisions that apply to all consensual security devices equally. Thus, it is necessary first to consider these shared principles and rules, prior to analysing the specific regime for privilegio, pegno and ipoteca. First, the Italian Civil Code structures real consensual security interests as rights in rem.16 They are effective erga omnes and award a right of suit to their holders, subject to limited exceptions.17 Secondly, real consensual security interests are always accessory to an obligation.18 A security interest can be created to secure any type of obligation (monetary and non-monetary), provided that it is existing, future, or conditional. A relevant corollary of this principle is that if the secured obligation is extinguished, the associated security interest suffers the same fate. Thirdly, real consensual security interests can only be taken over specifically identified goods, both present and future. The Italian Civil Code does not allow for the taking of security over an indeterminate array or a class of goods. A significant corollary of this rule is that it is not possible for a person (legal or natural) to grant a security interest over all their assets.19 Fourthly, real consensual security interests permeate the totality of each one of the encumbered assets, for the entire value of the secured obligation. It is not possible to grant a security interest over a fraction of a specific undivided asset.20 Thus, security interests can be described as indivisible.

12

Art 2740 Italian Civil Code. Art 2741 Italian Civil Code. 14 Art 2741.2 Italian Civil Code. 15 Art 2741.1 Italian Civil Code. 16 This principle was not always accepted unanimously. One view was that security interests only awarded a right to recover a determinate monetary sum from a specific asset. A different view was that security interests were only procedural rights. 17 Art 2808 Italian Civil Code; for an exhaustive analysis see Veneziano (n 11) 162. 18 This principle is inferred from Arts 2808, 2843, 2852 Italian Civil Code. 19 This is not the case for non-consensual security instruments, Art 2746 Italian Civil Code. 20 This principle is inferred from Art 2809 Italian Civil Code. 13

Security Interests over Intellectual Property Rights in Italy

263

Beyond these fundamental principles, the Italian Civil Code also contains a set of provisions that enact a common regime for all creditors. The following are of particular significance, for the purpose of the analysis at hand. Article 1186 of the talian Civil Code establishes that a secured creditor is entitled to demand immediate discharge of the secured obligation if the grantor has caused a diminution of the collateral. Under Article 2743 of the Italian Civil Code, if the collateral perishes or deteriorates, even due to fortuitous circumstances, to an extent that renders it unsuitable to secure the obligation in question, the secured creditor has a right to demand for it to be replaced with another asset of adequate value. If the grantor is unable to comply, the secured creditor is entitled to request immediate discharge of the secured obligation. Imposing a general limit on freedom of contract, Article 2744 of the Italian Civil Code establishes that any security agreement by virtue of which the property in an encumbered asset is transferred to the secured creditor in the event of the grantor defaulting is void.21 Italian courts have given a strict and far-reaching interpretation of this prohibition. Thus, any contract that functionally transfers ownership of an asset for security purposes is void, regardless of its form, type or the express intention of the parties.22 The Italian Civil Code also entitles all creditors, both secured and unsecured, to take action to preserve the financial wholeness of their debtor. Article 2900 of the Code provides that if a debtor neglects to exercise rights vis-à-vis third parties, creditors can do so in its stead. This is subject to two conditions: the rights in question must not be personal in nature and the inertia of the debtor must have resulted in a financial loss. Similarly, under Article 2901 of the Code, a creditor can seek the judicial annulment of any legal transactions entered into by the debtor that prejudices its financial positions, provided that specific conditions are met. Having described the common fabric to all consensual security devices under Italian law, attention can turn to pegno, ipoteca and privilegio. The following analysis will only address the rules and principles that are relevant to taking security over IPRs, as an exhaustive exploration of these devices is beyond the scope of this paper.

i. Pegno Pegno is a possessory, consensual security device. The progeny of the Roman Law pignus and the gage in the Napoleon Civil Code, it is nowadays regulated by Articles 2784–2807 of the Italian Civil Code. 21 Art 1523 of the Italian Civil Code expressly allows for reservation of title in a sale of goods; however, save, for few exceptions (eg Art 1154), these transactions are not widespread, due to the possession vaut titre principle enshrined in Art 1153 of the Italian Civil Code. 22 Notably, Italian legislation implementing the EU legal framework for financial collateral arrangements (Directive 2002/47/EC of the European Parliament and of the Council of 6 June 2002 on financial collateral arrangements) rules out the application of Art 2744 of the Italian Civil Code.

264

Andrea Tosato

Article 2784 of the Code establishes that a pegno can be granted both by the person who owes the obligation to be secured or by a third party. It further specifies the types of assets that can be the object of this security device: movable goods, receivables and rights other than receivables, provided that they relate to movable goods.23 A pegno can be granted on both existing and future goods.24 Under Article 2791, if the object of a pegno yields proceeds, the secured creditor is entitled to appropriate them, unless otherwise agreed. Appropriated proceeds must be counted towards repayment of the secured obligation. Though no statutory definition is provided, courts and commentators concur that this rule only applies to proceeds naturally stemming from the collateral; revenues flowing from dealings in the encumbered asset are excluded.25 The source of a pegno can be either a security agreement between grantor and secured creditor or an instrument unilaterally issued by the grantor. The two following requirements must be satisfied for the security interest to be created and become effective against third parties. First, the security agreement or instrument must identify the grantor, the secured creditor, the obligation that is being secured and the asset being encumbered.26 The law does not impose requirement of form. However, if the secured obligation exceeds a predetermined statutory value,27 the security agreement must be dated and in writing.28 Secondly, when dealing in tangible assets, the grantor must transfer possession of the collateral either to the secured creditor or to a third party custodian.29 By way of exception, departing from the archetypal possessory nature of this security device, Article 2806.1 of the Italian Civil Code establishes that when granting a pegno over a right ‘other than a receivable’, transfer of possession is not necessary. Instead, it is only required that the security agreement complies with the substantive and formal requirements normally mandated for the transfer of the right in question. Article 2806.2 further holds that a pegno over rights other than receivables is subject to any asset-specific statutory provisions applicable to the encumbered asset in question. Notably, the travaux preparatoires of the Italian Civil Code state that Article 2806.2 was included specifically to defer to statutes such as those regulating IPRs.30

23

Art 2784 Italian Civil Code. this paper, the expression ‘future goods’ is used to identify both goods not yet in existence and goods not yet owned by the grantor. 25 E Gabrielli, Il pegno (UTET, 2005) 152; G Gorla and P Zanelli, Del pegno (Zanichelli, 1992) sub art 2791. 26 Art 2787 Italian Civil Code. 27 Art 2787.3 Italian Civil Code. Notably, this is a small amount standing at €2.52. This is due to the fact that this value was set in 1942 and never subsequently revised. 28 Art 2787.3 Italian Civil Code. 29 Art 2786 Italian Civil Code. Exceptionally, there are specific classes of assets for which transfer or possession is not required. Inter alia, see Legge 24 July 1985 Art 1 regulating security interests over cured ham. 30 D Grandi, ‘Relazione al duce’, in Lavori preparatori del codice civile 1939–1941 (Istituto Poligrafico dello Stato, 1942) I 64. 24 Throughout

Security Interests over Intellectual Property Rights in Italy

265

If the aforementioned creation and perfection31 requirements are not satisfied, the security agreement only gives rise to rights in personam between the grantor and creditor. However, commentators and courts recognise that if the security agreement fails to identify either the obligation to be secured or the asset to be encumbered, it is entirely void.32 These provisions regulating creation and perfection have significant implications in relation to future goods. For tangible assets, the security interest comes into existence and becomes effective against third parties only after possession has been transferred. With regard to intangible assets governed by Article 2806 of the Italian Civil Code, however, the security interest arises and is perfected as soon as the asset either comes into existence or is acquired by the grantor.33 Once a pegno has been created and perfected, the secured creditor in possession of the collateral cannot use it or dispose of it in any manner, unless otherwise agreed by the parties.34 The law allows for multiple rights of pegno over the same asset. However, with regard to tangible collateral, once the first pegno has been granted, further security interests can only arise with the consent of the first secured creditor. This is a corollary of the possessory nature of this security device. For a second pegno to be created and perfected, the first secured creditor must positively assent to hold possession both for himself and on behalf of the second secured creditor. Conversely, though commentators are divided on this issue, when Article 2806.1 of the Italian Civil Code applies, multiple rights of pegno can be granted without the consent of the first secured creditor, as transfer of possession of the encumbered asset is not necessary.35 In the event of multiple pegno over the same asset, priority among secured creditors will be established pursuant to the first in time rule.36 Under Article 2789 of the Italian Civil Code, the secured creditor is entitled to protect the value of the collateral by bringing actions normally available to a person lawfully in possession of movable goods and those available to the grantor as proprietor of the encumbered asset. The secured creditor is also entitled to exercise any defence that is available to the grantor as proprietor of the collateral. Article 2789 goes beyond the aforementioned Article 2900, as it requires neither inertia on the part of the grantor, nor evidence of the eventus damni. Completing this body of rules, Article 2795 of the Italian Civil Code provides that a secured creditor can ask judicial authorities for authorisation to sell the

31 Italian law does not contemplate the possibility of a security interest coming into existence and yet not being effective against third parties. Under Italian law, creation and perfection are not severable and are treated as a unitary, indivisible concept. For further observations concerning this general issue, see L Gullifer (ed), Goode on Legal Problems of Credit and Security (4th edn, Sweet & Maxwell, 2013) para 2–02. 32 Gorla and Zanelli (n 25) sub art 2786. 33 ibid. 34 Art 2792 Italian Civil Code. 35 Gorla and Zanelli (n 25) sub art 2806. 36 ibid.

266

Andrea Tosato

collateral if it deteriorates to a degree that raises the prospect of its value becoming insufficient to cover the secured obligation. Coextensively, the grantor can ask authorities to sell the collateral or substitute it with different assets in the event of deterioration or loss of market value. This provision complements Article 2743, as it allows a secured creditor to take action as soon as there is a risk of the collateral deteriorating to the point of becoming insufficient to satisfy the secured obligation in question.

ii. Ipoteca Ipoteca is a non-possessory consensual security device. It is the descendent of the Roman law hypotheca and closely mirrors the French hypothec of the Napoleon Civil Code. It is regulated by Articles 2808–2899 of the Italian Civil Code. An ipoteca can be granted either by the debtor or by a third party.37 Article 2810 of the Italian Civil Code establishes that an ipoteca can only be granted over specific types of goods: immovables, rights over immovables, Italian treasuries, cars, aeroplanes and ships. An ipoteca can be granted on present and future goods. A right of ipoteca does not extend to the proceeds of the encumbered asset.38 An ipoteca can be granted either by written agreement or by virtue of a written instrument unilaterally issued by the grantor.39 Crucially, the security interest is created and becomes effective against third parties only upon registration on the relevant register; prior to this occurrence, the secured creditor only has a right in personam against the grantor to obtain registration of the interest in question.40 Registration of an ipoteca is based on transaction filing and requires the following documents to be deposited with the relevant registry. First, the original or an authenticated copy of the instrument granting the ipoteca. Secondly, a registration notice detailing, inter alia, the following information: the identity of the grantor and the secured creditor, a description of the encumbered asset, and the value of the secured obligation.41 Registration renders an ipoteca effective for 20 years, but it can be renewed by the secured creditor for a further two decades at any given moment.42 A notable consequence of the function of registration in relation to the creation and perfection of an ipoteca is that when future goods are involved, the security interest arises at the time of registration. This act is only possible after the goods have come in existence or have been acquired by the grantor.43

37 An ipoteca can also come into existence non-consensually. These forms of ipoteca, however, fall outside of this work, as they are non-consensual in nature. 38 Art 2811 Italian Civil Code. 39 Art 2821 Italian Civil Code. 40 On the notions of creation and perfection under Italian law see n 31. 41 Arts 2838–2839 Italian Civil Code. 42 Arts 2847, 2851 Italian Civil Code. 43 Art 2823 Italian Civil Code.

Security Interests over Intellectual Property Rights in Italy

267

Multiple ipotecae can be granted over the same asset. Priority is established based on the order of registration.44 In similar vein, time of registration serves as the criterion to solve disputes among competing claimants45 with proprietary interests. The secured creditor can seek judicial intervention if the grantor or another person is responsible for conduct that diminishes the value of the encumbered asset.46 If the secured obligation is assigned or otherwise transferred by the original secured creditor, the right of ipoteca securing that obligation follows suit. All such transfers are ineffective vis-à-vis third parties until registered.47

iii. ‘Privilegio’ and ‘Privilegio Speciale’ The Italian Civil Code originally structured privilegio as a purely non-consensual security device.48 The law established that determinate contractual and noncontractual obligations were supported by a security interest over assets of the obligor, owing to the nature of the obligation itself. Nevertheless, several statutory instruments enacted after the Code introduced special types of privilegio that were consensual and non-possessory in nature. Notably, these were ad hoc devices intended to facilitate the securing of loans extended by regulated credit institutions to enterprises operating in specific industrial sectors.49 This piecemeal assemblage of laws was abolished in 1994 by the D.lgs 1 S eptember 1993, n 385 ‘Testo unico delle leggi in materia bancaria e creditizia’ (henceforth ‘TUB’).50 Herein, Article 46 TUB introduced a privilegio speciale (henceforth ‘PS’) that consolidated and rationalised all the aforementioned pre-existing consensual forms of privilegio.51 Under Article 46.1 TUB, a PS can only be granted by enterprises52 to secure obligations owed to regulated credit institutions, for medium and long term loans.53 The grantor can either be the debtor under the loan being secured, or a third party. 44

Art 2852 Italian Civil Code. this paper, ‘competing claimant’ is used to identify a creditor of a grantor that is competing with another creditor of the grantor having a security right in the same encumbered asset of the grantor, as well as buyers and other transferees of proprietary interests. 46 Art 2813 Italian Civil Code. 47 Art 2843 Italian Civil Code. 48 For a comparative analysis of security devices of this nature in European personal property security law, see U Drobnig, ‘Legal Principles Governing Security Interests’, Annex to the ‘Report of the Secretary-General: Study on Security Interests’ (1976) 7 UNCITRAL Yearbook 175. 49 In depth, see Tosato (n 10) 572 ff. 50 This statute consolidated pre-existing Italian banking legislation and meaningfully reformed the extant legal framework. For an exhaustive analysis see D Siclari (ed), Italian Banking and Financial Law (Palgrave McMillan, 2015); specifically addressing legal issues involving the taking of security over personal property see Veneziano (n 11) 159 ff. 51 For an exhaustive analysis of this device see Chapter 15 of this book, by G Castellano; Ferrarini (n 9) 486–88. 52 In Italian law, the notion of ‘undertaking’ is inferred from that of ‘entrepreneur’ defined in Art 2082 of the Italian Civil Code. 53 Art 46.1 TUB. 45 Throughout

268

Andrea Tosato

The types of goods that can form the object of a PS are limited to movable goods, present or future, used by the grantor enterprise in the course of its business, provided that they are not subject to a specialised registration system.54 The PS does not automatically extend to the proceeds of the encumbered assets. The source of a PS must be a written agreement, detailing the identity of the grantor and financing creditor, all relevant terms of the loan being secured, the value of the security interest, and the encumbered assets.55 A PS is created and becomes effective against third parties upon registration on special registers held by geographically competent administrative authorities.56 Registration is based on transaction-filing, requiring for an original or authenticated copy of the security agreement to be filed.57 Crucially, Article 46.5 TUB establishes that a registered PS can be enforced against third parties, save for those who have acquired possession of the encumbered asset in good faith, for value; this is an application of the general principle possession vaut titre established in Article 1153 of the Italian Civil Code. In such cases, the security interest created by the PS automatically shifts to the proceeds of the transaction that resulted in the possession of the collateral being transferred. Notably, the significance of the rule established in Article 46.5 TUB evanesces in relation to intangible assets that are ontologically incapable of being reduced into possession. Multiple PS can be granted over the same asset. Priority will be established based on the time of registration.58 In like vein, if the same asset is encumbered by both a PS and a pegno, priority will be established pursuant to the prior in tempore potior in iure principle.59

B. The Intellectual Property Law Framework Conventionally, IPRs are treated as a homogenous category, due to the conceptual, structural and historical affinities shared by these exclusive rights. Nevertheless, with regard to the taking of security, the statutes governing registered (patent and

54

Art 46.1, 46.1-bis TUB. Art 46.2 TUB. 56 Art 46.3 TUB. The competent registers are those identified in Art 1524 of the Italian Civil Code. 57 Art 46.2 TUB. 58 Art 46.4 TUB. 59 Art 46.4 TUB, Arts 2748.1, 2777.3 Italian Civil Code; Court of Cassazione 17 February 1996, n 1238. However, under Art 46.5 TUB, if a creditor taking security by way of pegno acquires possession in good faith, an earlier PS is unenforceable. The majority of commentators unreservedly support the application of the prior in tempore principle to establish priority between a pegno and a PS; see P Piscitello, ‘Pegno e beni dell’impresa’ (2001) BBTD 155; M Rescigno, ‘Il privilegio per I finanziamenti bancari a medio e lungo termine a favour delle imprese, con particolare riguardo alla rotativita’ del suo oggetto’ (1999) BBTD 583. Nevertheless, L Ghia, C Piccinini, G Severini, Gli organi del fallimento e la liquidazione dell’attivo (UTET, 2010) 284 have argued that if the same asset is encumbered by both a PS and a pegno, the latter always has priority over the former, even if the pegno was created after the PS. 55

Security Interests over Intellectual Property Rights in Italy

269

trademarks) and unregistered (copyright) IPRs establish markedly different legal regimes and thus are best considered separately.

i. Registered Intellectual Property Rights Patents and trademarks are governed by the D.lgs 10 February 2005, n 30 ‘Codice della Proprietà Industriale’60 (CPI). Therein, Articles 137–140, 195–196 CPI contain norms relevant to dealings in these assets that involve the transfer of ownership, security interests and the issuing of licences. Article 104.1 CPI touches upon issues relative to the creation of security interests over registered IPRs, providing that they can only be used as collateral to secure monetary obligations. This is a notable departure from the Italian Civil Code general rules, which do not contain such a limitation.61 Article 138 CPI requires that transactions transferring ownership, awarding licences and creating a security interest over a trademark or patent must be registered on the relevant specialist intellectual property register to be effective vis-à-vis third parties.62 This provision is complemented by Articles 195–196 CPI, which establish the documents that need to be filed for a valid registration. In relation to a security interest, firstly, the original or an authenticated copy of the instrument granting the security interest must be deposited.63 Secondly, a registration notice must be filed, recording the identity of the secured creditor and of the grantor, a description of the encumbered IPR, the monetary obligation being secured and the value of the security interest.64 Cancellation of a security interest from the register is subject to the deposit of a document evidencing the consent of the secured creditor.65 Article 140.2 CPI implicitly acknowledges the possibility of multiple security rights being granted over the same registered IPR, establishing that priority among conflicting security interests is ascertained pursuant to the order of registration. This same criterion applies when determining priority with regard to all competing claimants. Although indirectly, Articles 78–79 CPI also affect the legal regime of security interests over registered IPRs. These provisions establish that title-holders of patents and trademarks cannot renounce them wholly or in part without the express consent of any person who holds property rights in them, including secured creditors.

60 The CPI regulates several other IPRs, alongside trademarks and patents. In principle, Arts 138–140 CPI apply unaltered to these other types of IPRs; however, such matters fall outside the scope of this chapter. 61 See Section IIA. 62 Art 139.1 CPI. 63 Art 138.3 CPI. 64 Arts 195, 196 CPI. 65 Art 140.3 CPI.

270

Andrea Tosato

ii. Unregistered Intellectual Property Rights Copyright is governed by Legge 22 April 1941, n. 633 ‘Protezione del diritto d’autore e di altri diritti connessi al suo esercizio’ (henceforth ‘LA’).66 The LA provides that copyright can be acquired, transferred and dealt with pursuant to the general rules established by the Italian Civil Code for movable goods.67 Notably, courts and commenters agree that the nemo dat principle applies strictly to all transactions involving property rights in copyright. Thus, conflicting claims are resolved on the basis of the first in time criterion; there is no bona fide purchaser exception applicable in relation to dealings in copyright.68 The LA does not expressly regulate the taking of security. The sole exception is Article 111 LA in which it is stated that a copyright cannot be subject to distraint, judicial sequestration or a pegno, as long as it is owned by the original author. The LA imposes no formalities for the coming into existence of copyright. Nevertheless, it does establish a general register for copyrighted works and two special registers for software and films respectively.69 These registers have an ancient history and have served different legal functions over the course of time.70 In their present incarnation, they allow for the registration of the existence of a protected work, the identity of the titleholder and any dealings that transfer ownership (whole or partial) or create property rights in the registered works, including security interests.71 Crucially, registration is merely discretionary, only serves as prima facie evidence of the information recorded therein, and has no further legal effects.

III. Taking Security over Intellectual Property Rights Having identified the relevant elements of the Italian personal property security law and IP legal frameworks, attention turns to the intersection of these two normative streams when security is taken over IPRs. 66 The LA also regulates rights neighbouring copyright. Commentators unanimously agree that, under Italian law, the legal regime applicable to the taking of security over copyright also extends to its neighbouring rights by analogy. Nevertheless, these IPRs fall outside of the scope of this paper and thus will not be considered. 67 Art 107 LA. 68 Seminally, G Oppo, ‘Creazione intellettuale, creazione industriale e diritti di utilizzazione economica’ (1969) Riv dir civ 1–45; LC Ubertazzi, M Ammendola, Il diritto d’autore (UTET, 1993) 69; P Greco, P Vercellone, I diritti sulle opere dell’ingegno (UTET, 1974) 332. 69 Art 103 LA. Notably, in 1994, legislation was introduced to a establish a registration system for films that would have been analogous to that in place for registered IPRs; however, this reform was never enacted. In depth, A Tosato, ‘sub Art 103 LA’ in LC Ubertazzi (ed), Commentario breve alle leggi su proprietà intellettuale e concorrenza (5th edn, CEDAM, 2012). 70 See A Tosato, ‘sub Art 104 LA’ in Ubertazzi (n 69). 71 Art 104 LA.

Security Interests over Intellectual Property Rights in Italy

271

A. Security Interests over Registered Intellectual Property Rights The CPI leaves no doubt that registered IPRs can be the object of security interests, as it contemplates this possibility in multiple provisions. Having recognised that registered IPRs can be used as collateral, it is necessary to determine which consensual security devices are suitable for taking security over these assets. In this respect, CPI provisions are of no assistance, as they generically refer to ‘security interests’ without ever mentioning pegno, ipoteca or the PS. The latter can be confidently excluded due to its incompatibility with assets subject to a registration system.72 Conversely, it is not immediately apparent whether the locution ‘security interests’ in the CPI should be construed as comprising both pegno and ipoteca or only referring to one of these devices. Commentators exclude the possibility that both these devices are concurrently suitable for taking security over registered IPRs and that the choice is left to party autonomy. This inference is based on the systematic observation that the Italian Civil Code structures pegno and ipoteca as jointly exhaustive and mutually exclusive. Further support for this view is adduced from the empirical observation that there is no asset class that can be subject to both these devices.73 Based on this assumption, two alternative theses have been put forth. The first is that security interests over registered IPRs are governed by the rules of pegno.74 More specifically, the applicable regime is that for pegno over rights other than receivables,75 together with the special norms established in the CPI. Proponents of this view argue that pegno is the only suitable device to take security over registered IPRs, as they are movable goods falling within the scope of Article 2784.1 of the Italian Civil Code. Moreover, they state that these assets are not included in the list of goods that can be subject to ipoteca, in Article 2810. Lastly, they point to the travaux preparatoires of the Italian Civil Code76 and submit that they express a general predisposition of the legislature towards pegno as the device to be used to take security over IPRs. The second thesis places greater emphasis on the CPI. Proponents of this view observe that the legal regime for proprietary dealings in registered IPRs echoes that of cars, ships and aeroplanes, which are all linked to ipoteca. They further emphasise that the registration system articulated in Article 139–140 CPI for security interests over registered IPRs closely mirrors that established by the Italian

72

Art 46.1 TUB. Chianale, ‘La funzione dei registri pubblicitari nella costituzione di garanzie reali su privative titolate’ (2009) AIDA 116. 74 P Greco and P Vercellone, Le invenzioni e i modelli industriali (UTET, 1968) 279; Gorla and Zanelli (n 25) sub art 2806; Gabrielli (n 25) 208ff; G Piepoli, ‘Autonomia privata e garanzie reali sulla proprietà industriale’ (2009) AIDA 294. 75 Art 2806.1 Italian Civil Code. 76 See n 30. 73 A

272

Andrea Tosato

Civil Code for ipoteca. Consequently, they infer that security interests over registered IPRs fall under the rules of ipoteca, combined with those of the CPI.77 Notably, the general norms of the Italian Civil Code for all consensual security interests will apply, regardless of this issue.78 With regard to patent and trademarks applications, the CPI offers no guidance concerning whether these assets can be the object of security interests. The majority view among commentators is that they can be used as collateral. There are, however, diverging opinions in relation to the law governing the taking of security over these assets. One thesis is that the legal regime for security interests over registered IPRs should be extended to applications by analogy.79 An alternative view is that, absent specific wording in the CPI, these assets should be treated as a right other than a receivable, security over which must be taken by way of pegno or PS. Notably, security interests over patent or trademark applications would not extend to the potentially resulting IPR, as this is an entirely distinct asset.80

B. Security Interests over Unregistered IPRs The LA implicitly acknowledges that copyright can be the object of a security interest by fleetingly contemplating this possibility in Articles 104 and 111 LA. Building on these scarce legislative indications, commentators suggest that there are two possible options for taking security over copyright. The first is by way of pegno. The provisions regulating attachment, perfection and priorities will be those set out for a pegno over rights other than receivables.81 The only exceptional rule to be considered is that in Article 111 LA, establishing that copyright cannot be used as collateral as long as the original author is the titleholder. The second option is to take security by way of PS. This device is suitable, as copyright falls within the subject matter described in Article 46.1 TUB and is not covered by any of the exclusions detailed therein.82 The rules of the PS concerning attachment, perfection and priorities apply unaltered; by contrast, the rule in Article 46.5 TUB is inconsequential, as copyright is incapable of being reduced into possession.83 With regard to the specific rules in the LA, it should be noted that Article 111 LA will be immaterial when taking security over copyright by way of PS; this is because Italian law only recognises individuals as original authors of copyright works, while this security device can only be granted by a company.

77 F Ferrara, L’ipoteca mobiliare (SEFI, 1932) 308; D Rubino, L’ipoteca immobiliare e mobiliare (Giuffrè, 1956) 221; Chianale (n 73) 118. L Nivarra, ‘Le garanzie reali su privative titolate’ (2009) AIDA 110. 78 See Section IIA. 79 For extensive bibliographical references see Nivarra (n 77) 111. 80 ibid. 81 Art 2806 Italian Civil Code. 82 Contrary to registered IPRs. 83 See Section IIA3.

Security Interests over Intellectual Property Rights in Italy

273

The general norms of the Italian Civil Code for all consensual security interests will apply, regardless of whether security is taken by way of pegno or PS.84

C. Security over IP Licences Neither the CPI nor the LA contain guidance on whether intellectual property licences can be the object of a security interest. Absent legislative indications, commentators observe that licences can be transferred subject to the express consent of the licensor and argue that the same regime should be applied to the taking of security by analogy.85 Nevertheless, beyond this preliminary inference, they do not specify which rules would govern these security interests. In principle, the rules for taking security over movable goods should be followed, as intellectual property licences fall within this asset class. Accordingly, the provisions for pegno over rights other than receivables should apply, pursuant to Article 2806 of the Italian Civil Code. The PS should equally be a viable option for taking security over licences, provided that all legislative requirements concerning the subjects involved and obligation being secured are met.86 However, with regard to licences for registered IPRs an alternative view is possible. Under Article 138 CPI, licences must be registered in the specialist IP registers in order to be enforceable vis-à-vis third parties.87 Relying on this legislative foothold, it can be argued that dealings involving registered IPRs licences should equally be subject to registration, by analogy. In turn, the syllogistic conclusion of this argument would be that the legal regime for security interests over registered IPRs should extend to the taking of security over licences of registered IPRs.

D. Conflict of Laws Rule The Italian private international law framework is articulated in the Legge 31 May 1995, n. 218 ‘Riforma del sistema italiano di diritto internazionale privato’ (henceforth ‘PILR’).88 Article 51 PILR generally establishes that the law applicable to property rights over movables is lex rei sitae. This provision further specifies that the same criterion applies to the instrument on the basis of which such property rights are acquired, save for transfers by way of contract, succession and family law. Article 54 PILR refines this general rule, stating that the law applicable to IPRs is that of the State for which protection is sought (lex loci protectionis); notably, 84

See section IIA. Ex multis Greco and Vercellone (n 74) 297–99. 86 Art 46 TUB. 87 Art 138 CPI. 88 This statute comprehensively reformed the Italian private international law legal framework. It establishes the limits of Italian jurisdiction, the criteria on the basis of which Italian courts determine the applicable law to a specific matter, and the enforceability of foreign rulings. 85

274

Andrea Tosato

this article does not establish a connecting factor to determine the law applicable to the instruments transferring or creating property rights in these assets.89 Against this statutory background, commentators unanimously agree that the connecting factor for security interests over IPRs is lex loci protectionis of the encumbered IPR.90 Therefore, the law of the protecting State must be followed to determine whether the IPR in question can be encumbered, which security devices are suitable to this purpose, the relevant substantive and form requirements for creation and perfection, and the relevant priority rules. Conversely, the mutual rights and obligations of the parties under a security agreement are governed by the law of the contract.91

IV. Critical Analysis of the Italian legal Regime for the Taking of Security over Intellectual Property Rights The preceding analysis has charted the intersection between Italian personal property security law and intellectual property law. This offers fertile ground for a critical assessment of the resulting legal regime. Attention will be focused primarily on two aspects: ‘legal certainty’92 and ‘commercial palatability’.93

A. Issues Affecting Security Interests over All Intellectual Property Rights Tangible assets and, to a lesser extent, receivables are the archetypal class of collateral around which the Italian personal property security law framework is built. 89 Art 54 PILR mirrors the normative choice made at international and regional level in all major multilateral instruments dealing with IPRs. See JJ Fawcett and P Torremans, Intellectual Property and Private International Law (Oxford, Oxford University Press, 2011); J Basedow and others (eds), Intellectual Property and the Conflict of Laws (Mohr Siebeck, 2005). 90 See S Bariatti, ‘The law applicable to security interests over intellectual property rights’ (2010) JPIL 395. 91 As per Art 3 of Regulation (EC) No 593/2008 of the European Parliament and of the Council of 17 June 2008 on the law applicable to contractual obligations (Rome I). 92 Legal scholarship has produced a veritable constellation of different definitions for the locution ‘legal certainty’. In this chapter, it is used in a positivist sense to indicate the degree to which it is possible to predict reliably the legal consequences that flow from a particular conduct, in relation to a specific provision. Ex multis, for a broad analysis of the different conceptualisations of legal certainty, see J Raitio, The Principle of Legal Certainty in EU Law (Kluwer, 2003) 267–340; S Bertea, ‘How NonPositivism Can Accommodate Legal Certainty’ in G Pavlakos (ed), Law, Rights and Discourse: Themes from the Legal Philosophy of Robert Alexy, (Hart, 2007) 69–82; HWR Wade, ‘The concept of legal certainty a preliminary skirmish’ (1941) MLR 183. 93 In this chapter, this locution is used to indicate the degree to which the policy choices realised by a provision are amenable to the interests and needs of secured creditors.

Security Interests over Intellectual Property Rights in Italy

275

While the cardinal elements of this body of rules possess sufficient elasticity to adapt to the intangible morphology of IPRs, tensions arise when applying more extensively articulated norms. This is particularly evident when considering the Italian Civil Code provisions that are applicable to all security interests. With regard to Article 1186 of the Italian Civil Code,94 establishing what constitutes a diminution of the collateral caused by the grantor presents challenges in relation to IPRs. Two issues need to be settled: the standard of causation on the basis of which the conduct of the grantor will be assessed, and the parameters relevant in the appraisal of whether a diminution has occurred. It is submitted that the first issue must be resolved by applying the same criterion used for tangible assets by analogy: any conduct on the part of the grantor that is a necessary cause of the diminution falls with the remit of Article 1186. With regard to the second issue, it is submitted that diminution should be appraised in terms of the economic value of the encumbered IPR in relation to the secured obligation. Thus, Article 1186 will be triggered if the grantor is responsible for the IPR becoming extinguished wholly or partially.95 The same will be true in the event of a failure to pursue infringers, provided that the consequent reduction in value is meaningful. It is more questionable whether unsuccessful commercial exploitation of the IPR could also fall under Article 1186. Inertia and negligence on the part of the grantor are likely to fall foul of this provision, while averse market movements will not.96 Article 2743 of the Italian Civil Code presents similar conceptual difficulties, as the rights attributed to the secured creditor therein are conditional on the collateral perishing or deteriorating.97 The former notion is not problematic: an IPR will be considered to have perished if it becomes void, for any one of the reasons contemplated by the law. Deterioration of an IPR, by contrast, is not an immediately obvious concept. In relation to tangible collateral, courts construe Article 2743 deterioration as physical degradation.98 Naturally, this is unsuitable for IPRs, due to their intangible nature. Thus, one possibility is that deterioration occurs if the value of the encumbered IPR falls due to market conditions. An alternative hypothesis is that deterioration of an IPR takes place if the scope of the exclusive right in question is reduced at law, for example due to partial annulment. It is submitted that this latter construction is preferable, as the secured creditor should be considering value oscillations when taking security.99 Similarly, Article 2900 of the Italian Civil Code presents construction difficulties in relation to IPRs.100 Commentators argue that this provision entitles a secured creditor to both bring actions against infringers of the encumbered IPR 94

See Section IIA. For example due to neglecting to pay renewal fees. Commentators are divided on this issue. See Piepoli (n 74) 300; Nivarra (n 77) 112. 97 See Section IIA. 98 There are a few commentators, however, who argue that market value should also be considered; see L Barbiera, Responsabilità patrimoniale (Giuffrè, 2010) sub art 2743. 99 Notably, this construction should also be extended to the concept of deterioration at the heart of Art 2795 of the Italian Civil Code. 100 See Section IIA. 95 96

276

Andrea Tosato

and demand payment from licensees. They suggest, however, that Article 2900 does not vest a secured creditor with the power to grant new licences or terminate existing licences in the event of a repudiatory breach.101 Respectfully, this thesis is not entirely convincing. In particular, it is unclear why Article 2900 would entitle the secured creditor to enforce certain rights (ie right to compensation) under the licence contract but not others (ie right to terminate). It is submitted that this distinction is artificial and unsubstantiated. Under Article 2900, a secured creditor must be allowed to enforce all the rights of a grantor under a licence, provided that they are not strictly personal in nature. Analogous difficulties also arise in relation to the specific rules established by the Italian Civil Code for pegno and ipoteca. Many of the relevant provisions either do not suit the intangible character of IPRs or are entirely incompatible with it. This is particularly evident with regard to the numerous provisions of pegno that assume that the collateral is reducible into possession.102 These elements of uncertainty are problematic, but not fatal, as they do not compromise irremediably the normative mechanics underpinning the taking of security over IPRs. Nevertheless, they evidence legal tensions between personal property security law and IP law, the effect of which is to reduce the reliability of the entire framework. Looking to commercial palatability, the legal regime for security interests over IPRs suffers from several structural flaws that are likely to deter parties from considering such assets as collateral. Firstly, there are limitations to party autonomy that hinder the use of IPR as collateral. In particular, the legislative prohibition on taking security over a class of assets of the grantor is especially problematic. Parties are not at liberty to create a security interest over all IPRs relative to a specific product or service. It is necessary to identify each specific IPR, significantly complicating the credit appraisal stage of the transaction. Secondly, there is a lack of a reliable, consolidated source of information relevant to creditors potentially interested in taking security over IPRs. It is necessary to investigate multiple IP registers, the register for the PS, as well as conduct ad hoc due diligence for those IPRs that have no registration system. Thirdly, the existence of different legal regimes regulating the taking of security over registered and unregistered IPRs complicates the existing legal framework. This divergence finds its root cause in the disparate legal history of copyright, patent and trademarks. At present, however, the existence of profoundly differing security interests regimes for these IPRs feels artificial and unnecessary, in the face of a commercial landscape in which copyright, trademarks and patents are inextricably intertwined in services, products and business development.

101 102

Piepoli (n 74) 302. For an extensive list of these provisions see Gorla and Zanelli (n 25) sub art 2786.

Security Interests over Intellectual Property Rights in Italy

277

B. Issues Specific to Security Interests over Registered IPRs The provisions specifically regulating the taking security over registered IPRs do not spawn major legal complications. Nevertheless, there are two issues that engender difficulties. First, it is unclear whether the taking of security over registered IPRs is subject to the rules of pegno or ipoteca. This uncertainty is particularly problematic with regard to the creation of security interests over these IPRs. If provisions for pegno are applicable, the interest arises as soon as the security agreement is concluded.103 By contrast, if the rules for ipoteca are to be followed, the security interest is not created until registration has been effected.104 However, beyond the creation stage, it is submitted that the issue at hand loses significance. This is because the provisions of the CPI combined with the intangible nature of IPRs affect the regimes for pegno and ipoteca, eliding almost all meaningful differences between the two. Notably, perfection,105 priorities,106 formalities,107 extension of the security interest to proceeds,108 and the granting of multiple security interests in the same asset109 are regulated in the same manner, regardless of whether one applies the rules of pegno or ipoteca. Thus, while the doubts with regard to the applicability of the rules for pegno or ipoteca are indicative of underlying systematic disorder, its practical impact is limited. This is reinforced by the observation that secured creditors typically seek to register their interest forthwith to perfect it and acquire priority,110 effectively confining the uncertainty concerning time of creation to the narrow interval between conclusion of the security agreement and registration. The second source of uncertainty is the absence of statutory guidance for security interests over applications for trademarks and patents. Moreover, the destabilising effect of this lacuna is exacerbated by the chasm between the alternative legal regimes that have been hypothesised for the assets in question. Against this unsettled landscape, it is submitted that subjecting applications to the same regime applicable to security interests over registered IPRs is the preferable view. Firstly, it is consonant with the legislative choice of the EU legislature for community trademarks.111 Secondly, it is systematically enticing, as it avoids the unpalatable outcome of diverse legal regimes for applications and the IPRs to which they relate. 103

Art 2806.1 Italian Civil Code. See Section IIA2. Art 138 CPI. 106 Art 140 CPI. 107 Both the rules pegno over rights other than receivables and ipoteca require a written security agreement or a written legal instrument for the creation of the security interest. 108 Ipoteca does not extend to proceeds and pegno does not extend to non-natural proceeds, such as those eventually yielded by IPRs. 109 Both the rules for ipoteca and pegno over rights other than receivables entitle the grantor to grant multiple security interests over the same asset 110 As per Arts 138–140 CPI. 111 Arts 24, 19 Council Regulation (EC) No 207/2009 of 26 February 2009 on the Community trademark L78/1. 104 105

278

Andrea Tosato

The practical significance of this issue should not be underestimated. Admittedly, lenders will plausibly shun applications that generally have brief examination times, preferring instead to take security over the eventually-resulting IPR. By contrast, applications with lengthy processing times, for highly valuable IPRs, such as pharmaceutical patents, are likely to be deemed attractive collateral by both grantors and lenders. With regard to commercial palatability, it is submitted that Articles 138–140 CPI decisively buttress the legal framework for the taking of security over registered IPRs. These provisions establish a registration system that provides a reliable source of information for all proprietary dealings and licences involving patents and trademarks. Moreover, the CPI establishes clear rules for the perfection of security interests and for the acquisition of priority. The sole cacophonic note emanates from Article 140.1 CPI, restricting the use of registered IPRs as collateral exclusively to secure monetary obligation.

C. Issues Specific to Security Interests over Unregistered IPRs The taking of security over copyright does not suffer from acute legal deficiencies. The provisions governing pegno and PS apply virtually unfettered, as the LA is almost entirely devoid of idiosyncratic norms for security interests. Article 111 LA is the only notable exception,112 marking a departure from general principles, yet not causing systematic failures. Nevertheless, this body of rules is markedly fragmented, establishing three distinct rulesets depending on the legal nature of the person who owns the copyright to be encumbered. Individual authors are entirely prevented from granting security interests over their own copyrights, pursuant to Article 111 LA. Natural persons who have derivatively acquired copyrights can grant security over them by way of pegno, but have no recourse to the PS, pursuant to Article 46.1 TUB. Enterprises have access to both pegno and PS.113 This differentiation appears to be more the product of uncoordinated legislative interventions, rather than the realisation of definite policy objectives. Despite this relatively unproblematic legal framework, the regime governing the taking of security over copyright suffers from flaws that significantly hinder its commercial palatability. Firstly, reconstructing the chain of title for these assets is extremely challenging. This uncertainty stems from the intangible nature of copyright, combined with the lack of a registration system that exhaustively documents proprietary dealings. Admittedly, this issue does not affect secured transactions any more than it impacts assignments and licences of copyright. The difference is that a prospective creditor typically lacks the willingness to investigate these matters and undertake the related risks in the same manner as a hypothetical purchaser or licensee. 112 113

See Section IIB2. See Section IIIB.

Security Interests over Intellectual Property Rights in Italy

279

Secondly, even if the uncertainty affecting the chain of title is overcome, a potential lender has no reliable source of information to determine whether a specific copyright is already encumbered. As previously mentioned, the registration system for the PS only provides partial information, as it does not record existing rights of pegno, while that enacted by the LA is ontologically not exhaustive with regard to security rights. Thirdly, the availability of both pegno and PS to take security over copyrights does not yield positive outcomes. In principle, the availability of two security devices should offer flexibility and versatility. In practice, however, these instruments are largely overlapping, rather than catering to different economic or legal circumstances. Therefore, this duality generates further uncertainty, exacerbating the complexity of an already intricate legal framework. These flaws are not intractable. Extensive due diligence and sophisticated contract drafting are likely to overcome the aforementioned obstacles in most instances. Nevertheless, this process will generate significant transactional costs that will invariably be passed on to the grantor, markedly diminishing the value proposition of the entire transaction.

D. Issues Specific to Security Interests over Intellectual Property Licences The legal regime for security interests over intellectual property licences suffers from a significant degree of legal uncertainty. Firstly, it is unclear whether these assets can be used as collateral, due to the absence of specific legislative indications in the statutes for both registered and unregistered IPRs. Nevertheless, as licences are transferrable and there are no legislative limitations relative to security interests, it is submitted that there is no reason why they should unsuitable to be used as collateral. Secondly, assuming that licences can be used as collateral, there is uncertainty surrounding the applicable rules for the taking of security over licences of registered IPRs. This state of uncertainty is particularly problematic, as the possible alternative legal regimes are markedly different.114 It is submitted that security interests over registered IPRs licences cannot be brought under the regime applicable to the taking of security over registered IPRs. Although this proposition is appealing from a systematic perspective, its weakness lies in the fact that Article 138 CPI only provides for the registration of licence agreements and not for subsequent dealings in the licences themselves. The registration regime established by Articles 138–140 CPI is exceptional in nature and its scope cannot be expanded beyond the letter of the law by way of interpretation. Thus, security interests over licences of registered IPRs must be brought under the general rules generally applicable to intangible movable goods.115 114 115

See Section IIIC. See Section IIIB.

280

Andrea Tosato

With regard to commercial palatability, the body of rules governing security interests over IP licences is hardly attractive. Registered IPRs licences make for lacklustre collateral due to the haziness surrounding their legal regime. The taking of security over unregistered IPRs licences suffers from flaws conceptually similar to those considered for copyright. A conscientious lender will be confronted with the daunting task of investigating the chain of title relative to the licence in question, the legitimacy of the licensor, and whether other security interests have already been granted.

E. Conflict of Laws Issues The Italian conflict of laws rule governing security interests over IPRs does not raise legal uncertainties. The lex loci protectionis approach is intrinsically uncomplicated, subjecting all proprietary aspects of the taking of security to the law of the State in which the encumbered IPR is protected.116 The simplicity of this rule, however, is purchased at the expense of its commercial palatability. The lex loci protectionis criterion is not problematic when taking security over a single IPR. However, in the vast majority of cases, a grantor will be looking to encumber a portfolio of IPRs awarded by multiple States; even when dealing with a single invention, creative work or distinctive sign, the grantor will typically seek to encumber the relative IPRs in all the jurisdictions in which they have been granted. In these scenarios, lex loci protectionis requires the grantor and secured creditor to investigate and comply with the laws of every jurisdiction in which the IP asset in question is protected. Thus, in secured transactions involving IPRs from multiple jurisdictions, the lex loci protectionis rule is timely, costly, burdensome and possibly capable of dissuading a lender from considering this type of collateral.

V. Legal Reform The preceding analysis has shown that the Italian legal regime for taking security over IPRs suffers from a measure of legal uncertainty and lacks commercial palatability to a varying degree. In light of these shortcomings, legal reform is a desirable prospect. There are two intersecting levels at which legislative intervention can be hypothesised: firstly, reform of Italian personal property security law as a whole, and secondly, reform of the IP statutory provisions that articulate asset-specific rules for IPRs. 116 The parties’ mutual rights and obligations, by contrast, will be governed by the law of the contract pursuant to Art 3 of Regulation (EC) No 593/2008 of the European Parliament and of the Council of 17 June 2008 on the law applicable to contractual obligations (Rome I) L 177/6.

Security Interests over Intellectual Property Rights in Italy

281

A. Real Consensual Security Framework Reform Over the past 50 years, scholarly debates concerning the possible reform of the Italian personal property security law framework have surfaced cyclically.117 Commentators have advocated for the need to modernise this body of rules, arguing that it has ossified and does not cater adequately to modern commercial realities. Nevertheless, these entreaties have fallen on unsympathetic ears, as this body of rules has remained unaltered, save for a few sectorial legislative interventions and minor judicial concessions.118 In the recent past, partly inspired by the work of UNCITRAL, the Italian legislature has embarked on a timid exploration of the possibility of more substantive reform.119 Concrete action, however, does not appear to be forthcoming in the immediate future. Over the past century, two alternative strategies have been prevalent in jurisdictions that have reformed their personal property security law. The first is to set aside typified security devices and replace them with a security interest law regime based on a functional approach and a notice-filing general security interest register. This approach was championed by the USA with UCC Article 9 and followed by Canada, New Zealand, and Australia. The second approach is to preserve existing personal property security devices, yet enact interventions aimed at rendering them more flexible and capable of taking of security over all types of assets: present, future, tangible, intangible, determinate and indeterminate This has been the path chosen in France, Germany and Spain. Recently, a third strategy has been suggested. Its premise is that reform initiatives must shift ‘the focus from domestically defined categories to the operational rules that allow secured transactions to perform their economic function of managing credit risk’.120 The suggestion is that it is possible to reform personal property security law by implementing a registration system on the model recommended in the UNCITRAL Legislative Guide on Secured Transactions, ‘without clashing with those categorisations that are embedded in national legal systems’.121 Discussion of the strengths and weaknesses of these alternative reform approaches lies outside the scope of the present enquiry. However, lessons can be learnt from scrutinising the manner in which legislatures that have reformed

117 For an overview G Tucci, ‘Towards a transnational commercial law for secured transactions: the preliminary draft UNIDROIT Convention and Italian law’ (1999) Uniform Law Review 371; Ferrarini (n 9); Veneziano (n 11) 159ff. 118 Ferrarini (n 9) 477ff. 119 See Disegno di legge recante ‘Delega al governo recante disposizioni per l’efficienza del processo civile, la riduzione dell’arretrato, il riordino delle garanzie mobiliari, nonché altre disposizioni per la semplificazione e l’accelerazione del processo di esecuzione forzata’ 12 February 2014. This draft proposal aimed to introduce a new non-possessory security device compatible with all types of asset classes and supported by a general register. 120 G Castellano, ‘Reforming Non-Possessory Secured Transactions Laws: A New Strategy?’ (2015) MLR 612. 121 ibid.

282

Andrea Tosato

their national personal property security law have dealt with IPRs. In some jurisdictions, hardly any steps were taken to coordinate the pre-existing intellectual property statutes and registers with the newly-introduced personal property security law.122 In others, attempts were made to amalgamate the reformed personal property security law with the pre-existing IP framework, yet insufficient care was devoted to this process, resulting in an unsatisfactory regime.123 These approaches are unsatisfactory, as they frustrate rather than facilitate the use of IPRs as collateral. It is submitted that any reform initiative of the Italian general security interest law framework should give a prominent role to these assets. This would require, firstly, articulating a set of rules that is not purely designed for tangible goods and negotiable instruments but also caters to the intangible nature of IPRs. Secondly, a careful analysis of the pre-existing IP legal framework should be conducted to identify and amend norms that are discordant with the new personal property security law framework. This coordination exercise should not be approached in cavalier fashion. Nevertheless, it is not inconceivable that all provisions contained in intellectual property statutes for the taking of security over IPRs might be repealed, in favour of extending to these assets the same regime introduced for all other intangibles. In similar vein, there should be no reluctance towards establishing information-sharing mechanisms between a newlyintroduced security interest general register and the pre-existing IP registers.124

B. Intellectual Property Law Framework Reform Legislative reform exclusively aimed at amending the IP legal framework would necessarily be less ambitious than a general rebasing of the entire personal property security law system. Nevertheless, such an intervention could significantly improve the extant state of the law. With regard to registered IPRs, three minimally invasive amendments are suggested. Firstly, security interests over applications and licences should be expressly regulated and subjected to the same regime applicable to registered IPRs. This reform would reduce legal uncertainty, simplify the body of rules governing registered IPRs and further strengthen the specialist IP registers as a comprehensive source of information for potential creditors. Secondly, the law should perspicuously establish whether the rules of pegno or ipoteca are applicable to security 122 For examples see M Dubovec, ‘UCC Article 9 Registration System for Latin America’ (2011) Arizona Journal of Comparative Law 139–140. In England, in the Law Commission Report 296, Company Security Interests (2005), para 3.235 it was recommended that ‘where the legislation that establishes the specialist registry contains rules determining the priority of competing charges, those should apply in place of the general rule of priority from date of filing’. 123 For a detailed analysis of the law in Australia see R Burrell and M Handler, ‘The PPSA and Registered Trade Marks: When Bureaucratic Systems Collide’ (2011) UNSWLJ 600. 124 UNCITRAL ‘Guide on the Implementation of a Security Rights Registry’ adopted by UNCITRAL in 16 December 2013, Report of UNCITRAL on the work of its sixty-eighth session, A/68/108.

Security Interests over Intellectual Property Rights in Italy

283

interests over registered IPRs. This intervention would definitively resolve the uncertainty surrounding the time of creation for security interests over IPRs. Thirdly, the rule in Article 140.1 CPI that limits the use of registered IPRs as collateral exclusively to secure money obligations should be repealed. This norm descends unaltered from the first Italian post-unitary statute governing patents; in that commercial context, it had a discernible rationale.125 At present, by contrast, it is hardly justifiable. For unregistered IPRs, by contrast, slightly more penetrating interventions are recommended. Firstly, registration should be elevated to a mandatory requirement to acquire third party effectiveness for all proprietary dealings in copyright.126 This reform would not be as radical as it might initially appear, as the Italian 1925 copyright law provided for exactly this regime.127 Such a change in legislation would not be incompatible with the Berne Convention, as registration would not be imposed as a condition for economic exploitation of copyright.128 Moreover, it would have the effect of markedly aligning the rules for the taking of security of registered and unregistered IPRs. Secondly, this new registration regime should not be confined to proprietary dealings, but also extended to copyright licences, due to their impact on legal and economic impact on the underlying IPR. Thirdly, priority among competing claimants should be determined on the basis of the time of registration. This would markedly enhance certainty and transparency for all proprietary dealings and licences. Fourthly, the limitation to party autonomy established in Article 111 LA should be abolished. This provision has ancient roots. It enshrined the policy objective that authors should never be stripped of the fruits of their creative labour against their own will, even in the event of personal bankruptcy. At present, however, the limitation imposed by Article 111 LA has become artificial and merely forces authors to transfer their copyrights into a special purpose vehicle in order to be able to encumber them. Finally, reform of the conflict of law rule applicable to security interests over both registered and unregistered IPRs should also be considered. This is a topic that has been the object of a flourishing body of scholarship129 and

125

For a historical analysis see Tosato (n 10) 561 ff. Notably, if this reform were enacted, registration of a PS would have to be reformed so that it would need to be effected in the copyright register; alternatively, an information sharing mechanism between PS registers and the new copyright register would need to be introduced. 127 See Tosato (n 10) 585. 128 See Art 5(2) The Berne Convention for the Protection of Literary and Artistic Works (Paris Text 1971); for an exhaustive analysis of this provision see S Ricketson and J Ginsburg, International Copyright and Neighbouring Rights: The Berne Convention and Beyond (2nd edn, Oxford, Oxford University Press, 2006). 129 S Weise, ‘The financing of intellectual property under the revised UCC article 9’ (1999) ChicagoKent Law Review 1077; L Brennan, ‘Financing Intellectual Property under Revised Article 9: National and International Conflicts’ (2001) Hastings Communication & Entertainment Law Journal 309; Bariatti (n 90) 395. 126

284

Andrea Tosato

international initiatives.130 The view increasingly prevalent among commentators is that enacting a choice of law rule that mandates a single law for all aspects of the taking of security over IPRs yields unsatisfactory results. Accordingly, the argument is advanced that the dépeçage technique should be adopted, establishing distinct connecting factors for different aspects of these transactions.131 Specifically, the taking of security over IPRs should be conceptually divided into matters that concern personal property security law and the debtor-creditor relationship on one hand, and matters that pertain to the IPR collateral as such, on the other. The former should be governed by the law of the grantor. The latter should be regulated by lex loci protectionis. Both Recommendation 248 of the UNCITRAL Supplement on Security Rights in Intellectual Property132 and Articles 3:801–3:802 of the European Max-Planck Principles on Conflict of Laws in Intellectual Property,133 though not identical, embody this approach. These choice of law model rules offer an appealing compromise between the interests of the parties to the secured transaction and the need to maintain systematic coherence within the intellectual property law framework. It is submitted that they should be given thorough consideration in the event of legislative reform in Italy.

130 See UNCITRAL, Supplement Recommendation 248; European Max-Planck-Group for Conflict of Laws in Intellectual Property (CLIP), Principles on Conflict of Laws in Intellectual Property (CLIP Principles) Arts 3:801, 3:802, 3:803; American Law Institute, Intellectual Property: Principles Governing Jurisdiction, Choice of Law, and Judgment in Transnational Disputes (American Law Institute Publishers, 2008). 131 For a detailed analysis of this technique with regard to security interests over IPRs see C Walsh, ‘Mobilisation of Intellectual Property in Secured Financing: Managing the Intersection between Territorialism and Globalism’ in Intellectual Property at the Edge: New Approaches to IP in Transsystemic World—Meredith Lectures, 342. 132 This recommendation states that ‘The law should provide that: (a) The law applicable to the creation, effectiveness against third parties and priority of a security right in intellectual property is the law of the State in which the intellectual property is protected; (b) A security right in intellectual property may also be created under the law of the State in which the grantor is located and may also be made effective under that law against third parties other than another secured creditor, a transferee or a licensee; and (c) The law applicable to the enforcement of a security right in intellectual property is the law of the State in which the grantor is located’. 133 Art 3:801 provides that the mutual rights and obligations of the parties arising from a contract to create or transfer a security right in intellectual property shall be governed by the law chosen by the parties. In the absence of a choice, the law of the State where the grantor of the security has her/ his habitual residence at the time of conclusion of the contract, unless the contract is manifestly more closely connected with another State, the law of that other State shall apply. Art 3:802 establishes that the law of the State for which protection is sought shall apply to ‘a) the existence, validity, scope and all other issues concerning the intellectual property right as such which is used as security …; b) the ownership of the intellectual property right which is used as security, unless otherwise provided in Sections 2 and 4, (c) the bona fide acquisition of an intellectual property right or security interests in such rights; (d) any registration requirements in intellectual property registers of the State of protection and the effect of registration or non-registration in such registers; (e) the priority and the thirdparty effects of security rights in intellectual property’.

15 Reverse Engineering the Law: Reforming Secured Transactions Law in Italy GIULIANO G CASTELLANO*

I. Introduction In the last decades, several international initiatives have been advanced to both modernise and harmonise domestic laws governing non-possessory secured transactions, a process crowned with the completion, in July 2016, of the UNCITRAL Model Law on Secured Transactions.1 However, the implementation of these projects into national legal contexts and, in general, the process of reforming this ambit of the law are arduous tasks. Overcoming these hurdles implies the definition of reform strategies that, by at large, are intended to design a simplified legal regime that is conducive to private negotiations to establish a secured credit market. Also in this field the contribution of Roderick Macdonald, as a scholar and a law reformer, has established an enduring legacy. This chapter contributes to this debate by acquiring a ‘law in action’ approach that examines various reform routes adopted in different jurisdictions. Through an inductive method, secured transactions law reforms are reverse engineered and deconstructed into their key components to determine the most effective route. Ultimately, this leads to considering the viability of a new reform route, whose theoretical foundations have been elaborated elsewhere,2 and testing its potential application within a European jurisdiction that has been struggling to ensure broader access to credit * Dr Giuliano G Castellano is an Assistant Professor of Law, University of Warwick, School of Law and a Research Associate, i3-CRG, École polytechnique, CNRS (Université Paris-Saclay). He serves as a member of the Italian delegation at UNCITRAL, Working Group VI (security interests). The views expressed in this chapter does not reflect any official opinion. The author gratefully acknowledges the support of the Economic and Social Research Council (ESRC) via the University of Warwick’s Impact Acceleration Account (ESRC0018-2015). 1 See UNCITRAL Model Law on Secured Transaction (UN Doc A/CN.9/884 and Add 1–4). 2 GG Castellano, ‘Reforming Non-Possessory Secured Transactions Laws: A New Strategy?’ (2015) 78 Modern Law Review 611.

286

 Giuliano G Castellano

via secured transactions, ie the Italian legal system. Through these lenses the new non-possessory pledge recently introduced in Italy is examined. Reverse engineering is a practice used in various technical and industrial domains; it studies the design or functioning of artefacts (or processes) to extract the knowledge that allows their functioning.3 It is a scientific, inductive method that deconstructs in detail man-made products, rather than natural phenomena, to acquire new knowledge about their functioning. This practice is used for various reasons. Military and commercial espionage use this method with the intent of enhancing security or designing competing products.4 In computer science, reverse engineering is also used to overcome the problem of ‘system obsolescence’, which occurs when a system cannot be maintained any longer, eg due to excessive costs, inefficiencies or lack of spare parts, although its functionality is still essential. Hence, reverse engineering aims at preserving the function of the system while its core components are redesigned in a more efficient fashion with new technologies. This practice resonates with the legal discourse on secured transactions law reforms on several levels. First, the legal system could be conceived of as a manmade phenomenon in which different legal rules are enacted to address societal issues. Therefore, legal reforms are understood here as processes aiming at changing legal rules to tackle new societal issues or existing issues in a more effective fashion. This implies understanding law reforms as a process of legal design.5 Second, the legal regimes governing secured transactions laws have been developed over centuries to ensure a legal structure that allows access to credit through collateral. The numerous national and international reform debates are an evident signal that the legal technology deployed suffers, in many countries, from an obsolescence problem. This means that such a technology—supporting a system that is intended to facilitate access to secured credit—is no longer adequate, vis-à-vis the variety of new legal solutions developed and the new nature of societal issues represented by the economic needs of international finance. In many legal systems, a stratified and complex apparatus of legal categories hardwired in the domestic legal culture, are resilient to change, even in circumstances where legal reform appears much desired. The resulting scattered international scenarios have been deemed to render impossible a process of national legal reforms and international harmonisation.6 However, there are numerous examples of successful 3

E Eilam, Reversing: Secrets of Reverse Engineering (Indianapolis, Wiley Publishing, 2005) 3. eg N Lee, Counterterrorism and Cybersecurity: Total Information Awareness (New York, Springer, 2013) 109. 5 The idea of that the law is a discipline concerned with design, presenting relevant commonalities with engineering has been also advanced by D Howarth, Law as Engineering—Thinking About What Lawyers Do (Cheltenham, Edward Elgar Publishing, 2013). 6 U Drobnig, ‘Study on Security Interests’ (1977) 8 UNCITRAL Yearbook 171 (UN Doc A/CN.9/ SER.A/1977). In support of these considerations Ulrich Drobnig also mentioned three failed projects: the UNIDROIT proposal of 1968; the research of the Service de Recherches Juridiques Comparatives of the French Centre National de la Recherche Scientifique (CNRS) in 1972; and the European Community’s proposal (pending at the time of Drobnig’s study) to establish a European registry for security interests. 4 See,

Reverse Engineering the Law

287

national reforms that have either implemented international standards or, by simplifying national regimes, offer useful insights for national law reformers.7 Against this backdrop, the process of reverse engineering secured transactions law reforms aims at identifying the core elements that could resolve the problem of system obsolescence faced by various national legal regimes for secured transactions laws. The reverse engineering approach to law reforms is developed in four parts through this chapter. The second part identifies a core of implicit questions that law reformers must address in regard to reforming secured transactions laws and isolates two main reform routes adopted in different jurisdictions. The third part isolates the core elements of a legal reform process in this realm of the law. The fourth and final part adopts the knowledge gained through the reverse engineering process to advance an alternative reform route. This stage consists of a process of ‘re-engineering’ the reform process within a specific legal context. It entails examining a possible reform of the Italian legal system—one of the most complex, and still most in need of reform, secured transactions legal regime in Europe. This represents the ultimate test for the argument advanced here, and it is performed by presenting the possible implementation of UNCITRAL’s provisions on registries in the Italian legal system. The recent introduction in the Italian legal system of a new legal category to take security over company assets is considered. Albeit marking a step forward in the process of legal modernisation, the new non-possessory pledge appears to add a further layer of complexity, hindering the objective of facilitating access to credit via the banking system. In contrast, reforming publicity rules appears to be a more suitable and rational reform route.

II. Secured Transactions Law Reforms: Common Roots, Different Routes Domestic legal categories—developed through time and entrenched in national legal cultures—define the structure of the system governing secured t ransactions that is often resilient to change. In fact, a change in such a structure poses r elevant problems, as it implies the modification—if not eradication—of deeply rooted conceptual frameworks. The understanding of the core elements of such a structure that has influenced various legal regimes within Western legal traditions is a necessary step for this analysis, as it traces the perimeters within which law reforms are often constrained. In particular, the focus is on the development of non-possessory secured transactions and the tension they pose to the system of rules largely grounded in the requirement of dispossession. 7 N Cohen, ‘Internationalizing the Law of Secured Credit: Perspective from the US Experience’ (1999) 20 University of Pennsylvania Journal of International Law 423.

288

 Giuliano G Castellano

A. The Perimeters of Legal Reforms and the Origins of Domestic Categorisations The original proprietary nature of security rights constitutes the mainstay upon which secured transactions have developed. The fundamental categories of pignus and hypotheca, contained in the Justinian’s Curpus Iuris Civilis, filtered, during the Middle Ages, into various legal systems of continental Europe.8 Under Roman law, the distinction between pignus and hypotheca (or pignus conventum) only affected the mode of creation of a security interest, the former requiring dispossession and the latter simply requiring consensus.9 Hence, non-possessory security interests in movable assets were allowed through constructive delivery under the Justinianean hypotheca.10 It was not until the fifteenth century that lawyers advocated more forcefully for a regime based on dispossession to create a security interest in movable property, defining the roots of the modern pledge, referred as to pegno in Italian, gage in French, and Faustpfand in German.11 The idea has been progressively absorbed in the codification processes, marking the origin of the bifurcation between the legal regimes for security interests over movable and immovable property. With the lack of a reliable publicity legal regime based on registration, the old Dutch Civil Code (1838), the French Code Civil (1804), the German Bürgerliches Gesetzbuch (1900), and the old Italian Civil Code (1865) constrained the applicability of the Roman hypotheca exclusively to immovable property, whereas security rights over movables were validly established only through a transfer of possession. Similarly, in England, security interests in personal property without dispossession were considered fraudulent acts.12 In English common law, non-possessory security rights originally developed under the principles of equity, as reflected in the fundamental distinction between legal and equitable security interests in English law. The resulting legal regimes reflected the needs of a mercantile economy, in which the diffusion of limited liability partnerships constituted the primary source of capital, and the certainty over the existence (and non-existence) of proprietary entitlements was necessary

8 For a complete historical account see WJ Zwalve, ‘A Labyrinth of Creditors’ in E-M Kieninger, Security Rights in Movable Property in European Private Law (Cambridge, Cambridge University Press, 2004). 9 As illustrated in this section, the Roman pignus and hypotheca (or pignus conventum) should not be confused with the modern ‘pledge’ and ‘hypothec’. According to Ulpian the distinction between pignus and hypotheca has to be identified in the fact that pignus required delivery of possession: ‘Proprie pignus dicimus, quod ad creditorem transit, hypothecam, cum non transit nec possessio ad creditorem’; Ulpianus, D.13,7,9,2. 10 The term ‘hypotheca’ derives from Greek (ύπουήκη) and in the post-classic era has been adopted to identify the pignus conventum (or conentum) according to which the debtor, through constructive delivery (constitutum possessorium), could pledge on properties without traditio (delivery), otherwise required by the pignus datum. 11 See Zwalve (n 8). 12 For early references see the Fraudulent Conveyance Act of 1571 and Twyne’s [1601] 76 ER 809.

Reverse Engineering the Law

289

to ensure the circulation of goods.13 Hence, dispossession may be considered as the most basic form of publicity. Alternative consensual arrangements developed to circumscribe the requirement of dispossession. To accommodate the new financing needs following the Industrial Revolution and, in particular, the need to use as collateral assets that are functional to the production of goods, alternative legal arrangements have been developed. Retention of title clauses, sale and lease-back agreements, hire and purchase contracts as well as fiduciary arrangements (drawing from the pre-Justinian fiducia cum creditore) and equity principles have been deployed to establish a non-possessory secured transaction. The acceptance of these instruments, and the consequential relaxation of the requirement of dispossession, varied greatly among legal systems and still influences modern national regimes for non-possessory secured transactions. The resulting variety of approaches to accommodate new financing techniques ultimately appears to be the source of the existing discrepancies amongst national laws. Hence, it is possible to argue that national idiosyncrasies do not appear to be an embedded feature of archetypical categorisations, whose core features and aspirations were similar before and after the codification process. Rather, they appear to be the outcome of subsequent adaptations and reforms aimed at relaxing or tightening those categories. In particular, the scattered nature of secured transactions legal regimes appears to stem from a new, common issue, which is the increasing use of non-possessory secured transactions as critical financing techniques. The different approaches deployed to address this issue define the core features of different reform routes. The more recent developments in American and English common law are emblematic of the two main routes. In the US, first with the Uniform Trust Receipts Act 1933 (UTRA) and then with Article 9 of the Uniform Commercial Code (UCC), dispossession was no longer a requirement to constitute a security interest in movable assets.14 Moreover, UCC Article 9 included in the general c ategory of ‘security interests’ a wider range of consensual arrangements to establish a secured transaction over movables. In the UK, although the requirement of d ispossession was progressively relinquished, the original set of legal categories, with their different corresponding legal treatments, has been maintained. There are more profound differences between the two legal regimes.15 Nonetheless, this sketch allows the introduction of the reform routes available to policymakers.

13 A thorough comparison between UK and Germany in the nineteenth century reveals that access to equity markets was the primary driving force for economic growth and development, see C Gerner-Beuerle, ‘Law and Finance in Emerging Economies: The Case of Germany 1800–1913’ (2017) 80(2) Modern Law Review 263. 14 The UTRA replaced dispossession with a registration precisely to allow the use as collateral goods that are instrumental to the production of other goods. 15 For a complete comparative analysis between the two systems see G McCormack, Secured Credit under English and American Law (Cambridge, Cambridge University Press, 2004).

290

 Giuliano G Castellano

B. One Common Question and Two Reform Routes Beneath the granular elements defining the evolution and the features of national legal regimes, a common question sets the tone, more or less explicitly, for any reform process: should secured transaction law adopt a unitary approach offering the same legal treatment for all consensual arrangements establishing an entitlement over personal property to secure an obligation in the event of the debtor’s default? In a more visual and rather prosaic fashion, the question has been often referred as to whether a given legal system should ‘import’ or ‘follow’ the US model by adopting a unitary approach to classify and govern the variety of consensual devices used to secure an obligation with collateral.16 Currently, reference to the US appears accurate only to the extent that, almost half a century ago, UCC Article 9 adopted a unitary approach with the category of ‘security interests’.17 From that moment, however, a number of national reforms have adopted and expanded the unitary approach. Although openly inspired by the rationale of UCC Article 9, for instance, the Personal Property Security Acts implemented in Canada, Australia and New Zealand further enlarged the scope of the ‘security interests’ category. Thus, in current reform debates, the key question rests more accurately in choosing whether, and to what extent, a unitary approach should be adopted. The answer to this fundamental question leads to the pursuit of different reform paths that may ultimately be regrouped into two main routes. A negative answer—or no actual answer—to this core question implies that change would occur through a process of incremental changes, in which judicial interpretations and targeted legislative reforms expand the perimeters of secured transactions law without questioning the building blocks of the regime. In contrast, a positive answer would imply a conscious reform effort aimed at reshaping the core categorisations in favour of a unitary route. As further illustrated below, intimately related to this fundamental question is the issue of the interpretative technique to be adopted in order to determine whether a given consensual arrangement may be qualified, in law, as a secured transaction. At the root of national legal reforms, hence, resides the necessity of resolving what is commonly intended as a clash between approaches based on ‘formalism’, requiring us to ascertain the legal nature of the security agreement in accordance to national legal categories, and ‘functionalisms’, requiring an investigation over the economic function performed by a given security agreement. It is not surprising that 16 For a recent analysis demonstrating the centrality of the question for Continental European legal systems see T Tajti, ‘Could Continental Europe adopt a Uniform Commercial Code Article 9-type Secured Transactions System? The Effects of the Differing Legal Platforms’ (2014) 35 Adelaide Law Review 149; for Latin America, see M Dubovec, ‘UCC Article 9 Registration System for Latin America’ (2011) 28 Arizona Journal of International and Comparative Law 117; for France see J-F Riffard, Le Security Interest ou l’Approche Fonctionnelle et Unitaire des Sûretés Mobilières. Contribution à une Rationalisation du Droit Français (Paris, Presses Universitaires de Clermont-Ferrand, 1997). 17 Between the first release of the UCC in 1952 and the Official Version of 1962, all states (except for Louisiana) adopted the UCC Art 9.

Reverse Engineering the Law

291

non-unitary approaches rely mostly on formalism to determine the legal treatment for a given security device, whereas unitary approaches adopt an interpretative and ordering criterion functionalism. The formalistic and dogmatic understandings of secured transaction laws—attributed to European legal traditions—appear to collide with the functionalist ethos typically embraced by North American jurisdictions, influencing many reform projects around the globe and embraced by the UNCITRAL Legislative Guide and the UNCITRAL Model Law. Distinguished scholars, moving from thorough comparative analyses, have unravelled the conceptual structures beneath these different understandings of the law.18 Upon these premises, Roderick Macdonald devised a number of practical solutions resolving such a cultural dichotomy and designing effective legal reforms, where the legal precision of formal definitions has been paired with the systematic organisation offered by a unitary category governed through functionalism.19

C. Reform Through Incremental Changes A negative (implicit or explicit) response to the fundamental question results in reforms that seek to modernise non-possessory secured transactions law through incremental changes. Such changes generally occur through a mixture of direct legislative interventions, generally limited to specific transactions or classes of assets, and judicial interpretations that progressively broaden access to secured credit without reshaping the building blocks of the legal apparatus pertaining to secured credit and proprietary entitlements. Existing legal categories, codified or developed through judicial interpretation, represent the basic structure upon which new forms of consensual security interests are added, often in the form of exceptions enshrined in special laws or as a result of judicial interpretations. The gradual development of secured transaction laws in European legal systems from the Industrial Revolution to modern times mostly followed this route by progressively relaxing dispossession requirements. More recently, this approach characterises reforms that have occurred, among other legal systems, in France, Germany, Italy and the UK. For instance, under German law, notwithstanding some reluctance and the fact that the statutory pledge (Faustpfand) had been the main security instrument,20 different forms of non-possessory security rights in movables developed primarily as fiduciary transfers of property rights.21 In a similar vein, Italian scholars have regrouped new forms of non-possessory secured

18 MG Bridge and others, ‘Formalism, Functionalism, and Understanding the Law of Secured Transactions’ (1999) 44 McGill Law Journal 567. 19 R Macdonald, ‘Article 9 Norm Entrepreneurship’ (2006) 43 Canadian Business Law Journal 240. 20 See RG 24 Sept 1880, RGZ2, 173. 21 German legal scholarship regrouped various transactions that performed similar functions under the category of besitzlose Mobiliarsicherheiten.

292

 Giuliano G Castellano

transactions in the concept of pegno anomalo (literally ‘anomalous pledge’).22 Under this concept, defined as ‘anomalous’ precisely because it is outside the codified category of ‘pledge’, delivery of possession is not always required, eg, when the encumbered assets is instrumental to the production of goods. Proprietary entitlements started to be constituted over the value represented by certain assets— rather than an identified item of property—allowing for a change, defined as a ‘rotation’, within the pool of encumbered assets as long as the value covered by the security was maintained.23 Italian courts have progressively accepted these interpretations of the statutory provisions based on the idea that dispossession could have been substituted by an adequate level of control over the encumbered assets and the value was sufficient to identify fungible assets.24 In general terms, incremental changes occur naturally through the interpretative efforts of courts and scholars. This is the case, for instance, for security interests in intellectual property, such as patents or trademarks, whose insertion into the existing legal framework has been often problematic.25 Judicial interpretation, however, is not always sufficient (alone) to ensure broader access to credit by overcoming dispossessory requirements, and a more direct legislative intervention has been required. Examples of direct legislative interventions are emerging from the analysis offered in this chapter on different national laws and they are often aimed to facilitate the use of specific movable assets or security devices. At the international level a similar trend may be noted, for instance with reference to the special regimes for security interests created over financial and negotiable instruments at the international level, according to which proprietary entitlements are created through a signature (or endorsement).26 Although a reform route based on incremental changes appears to offer immediate benefits by avoiding costly reforms and unlocking new financing opportunities for debtors and creditors, a series of drawbacks may surface over time. Incremental changes tend to design legal regimes that are composed of general rules and specific provisions that coexist and overlap with possible ambiguities over the legal treatment attributed to a given security device. The resulting uncertainty of such an approach emerges, for instance, from one of the main risks affecting English

22 On the development of these new forms of secured transactions see G Tucci, Garanzia sui crediti dell’impresa e tutela dei finanziamenti (Milan, Giuffrè, 1974); E Gabrielli, Il pegno ‘anomalo’ (Padova, Cedam, 1999). 23 See, eg C Mancini, ‘La riconosciuta normalità del pegno rotativo’ (1998) 6 Giurisprudenza commerciale 3. 24 See, for instance, the landmark decision Corte di Cassazione, 28 May 1998, n. 5268. On the evolution in Italy of non-possessory security interests see A Veneziano, Le garanzie mobiliari non possessorie (Milan, Giuffrè, 2000). 25 For an analysis of the issue in English common law see A Tosato, ‘Security Interests over Intellectual Property’ (2011) 6 Journal of Intellectual Property Law & Practice 93. 26 See, eg United Nations Convention on International Bills of Exchange and International Promissory Notes (concluded 9 December 1988 in New York) (1989) 28 ILM 177, Art 22; and the Convention Providing a Uniform Law for Bills of Exchange and Promissory Notes (signed 7 June 1930 in Geneva) 143 LNTS 247, Art 19.

Reverse Engineering the Law

293

law, ie the risk of ‘recharacterisation’, according to which a court may ascertain that a security agreement has a legal nature that differs from the one originally attributed by the parties and therefore apply a legal treatment that differs from the one originally envisaged by the parties. Leaving aside the need to curb fraudulent behaviours, the phenomenon is dreaded by lawyers (and their clients) in the City of London in regard to drafting a fixed-charge agreement, which may risk being turned into a floating charge, with notable material implications on the priority ladder.27 In fact, under English common law, registered floating charges rank below registered fixed charges despite the order of creation.28 The risk of recharacterisation is particularly pronounced when different categories exists and the characterisation of a specific transaction is both a point of priority and a matter of law for the court to determine. In such a situation, further definitions or interpretative criteria should be established to ensure an adequate level of certainty. In English common law, courts have established as a primary criterion, in the lack of a clear-cut definition, that a charge is fixed when control of the collateral is left to the charge holder.29 However, when a fixed charge is established over book debts, receivables, or other intangible assets—which require being managed by the secured debtor (the chargor)—the problem is particularly salient, leaving room for interpretative ambiguities.30 In this circumstance, incremental changes driven by judicial interpretation aimed at adjusting the perimeters of longstanding legal categories to fit new economic needs translated into a higher level of uncertainty, until some clarity is brought by the Court, as occurred with the landmark decision in Re Spectrum.31

27 For a thorough analysis over the difference between floating and fixed charges see, especially RJ Mokal, ‘Liquidation Expenses and Floating Charges—The Separate Funds Fallacy’ (2004) Lloyd’s Maritime and Commercial Law Quarterly 387; and E Ferran, ‘Floating Charges—The Nature of Security’ (1988) 47 Cambridge Law Journal 213. For a detailed account of the different legal treatments in insolvency proceedings see R Goode, Principles of Corporate Insolvency Law (4th edn, London, Sweet & Maxwell, 2011) 417. 28 Insolvency Act 1986, s 175. Moreover, according to s 176 of the same Act, a floating charge is also subordinated to preferential creditors and a portion of charged assets must be portioned for unsecured creditors. 29 In the Court of Appeal decision, Romer LJ identified the three characteristics of a floating charge: ‘(1.) [i]f it is a charge on a class of assets of a company present and future; (2.) if that class is one which, in the ordinary course of the business of the company, would be changing from time to time; and (3.) if you find that by the charge it is contemplated that, until some future step is taken by or on behalf of those interested in the charge, the company may carry on its business in the ordinary way as far as concerns the particular class of assets I am dealing with’: Re Yorkshire Woolcombers’ Association Ltd [1903] 2 Ch 284 at 295. 30 For a thorough analysis covering the evolution of English common law in this ambit see L Gullifer, Goode on Legal Problems of Credit and Security (5th edn, London, Sweet & Maxwell, 2013) 136ff. 31 Re Spectrum Plus Ltd [2005] UKHL 41. For a complete analysis of the case and its impact see S Worthington, ‘Floating Charges: the Use and Abuse of Doctrinal Analysis’ in J Getzler and J Payne (eds), Company Charges: Spectrum and Beyond (Oxford, Oxford University Press, 2006). Recharacterisation may occur also in reverse direction, when a floating charge is in law a fixed charge: see The Russell Cooke Trust Company Limited v Elliot [2007] EWHC 1443 (Ch).

294

 Giuliano G Castellano

What is rightfully approached as a matter concerning both the drafting of s ecurity agreements and the strategy to be adopted in the case of dispute flags the existence of a deeper problem with a system that developed through incremental change. The problem could be more generally ascribed to the simultaneous existence of different legal categories whose contours are not always clear and where formalism is the primary ordering criterion. This overlap may also occur when new statutory changes are introduced. For instance, the French Commercial Code introduced the legal regime for a ‘pledge of inventory’, which only applies when the lender is an authorised credit institution. However, in such a circumstance, there is an overlap with the provisions contained in the French Civil Code. Further judicial interpretation was required in such a circumstance, given that the novel legal category overlaps with existing ones.32 A similar phenomenon may be predicted to manifest where new legal categories are introduced in addition to existing ones. It is argued here that this reform route not only increases the discrepancies amongst legal systems, but it determines a situation of path-dependency, aggravating the complexity and the dogmatic structure of various European legal regimes. In fact, the complexity of many European regimes originates precisely from reforms designed to avoid more drastic changes. Moreover, the rationale for a reform based on gradual adjustments is safeguarding basic categorisations while new financing practices are allowed. This must occur through interpretative efforts that ensure consistency between new interpretations and statutory rules and the fundamental building blocks defining secured transactions law. Hence, novel interpretations and new rules often affirm the existence of an archetypical legal construction, defining security interests in a fashion that is well depicted by the adage, attributed to Cicero, exceptio probat regulam in casibus non exceptis.33 As a result, comprehensive reforms become more arduous to attain, given the increased strength of archetypical categorisations. This results in a paradoxical situation, in which countries more in need of a radical change are those that find this task more difficult and therefore are less likely to pursue comprehensive reforms.

D. The Unitary Route A second route, implying an affirmative answer to the fundamental question, results in a more drastic legal change by establishing a unitary category, such as ‘security interest’ or ‘security right’, that encompasses various forms of non-possessory 32 The Cour d’Appel of Paris established that in such a circumstance, when a credit institution is involved, parties have to choose whether to establish a non-possessory pledge either under the Civil Code (which allows self-attribution right) or following the Commercial Code; see Cour d’Appel de Paris, 3 May 2011, n 10/13656. 33 The original meaning of this principle (translated as ‘the exception confirms the rule in cases not excepted’) indicates that if legislators provide for an exception, then it is reasonable to assume that the general rule applies to all other circumstances.

Reverse Engineering the Law

295

secured transactions. The primary benefit of this route resides in the possibility of rationalising the complexity created through centuries of incremental changes.34 A unitary route provides for a cohesive approach that ensures uniformity, and therefore more certainty, in the legal treatment of non-possessory secured transactions. Within this framework, registration regulates priority among competing claimants and failure to register a transaction does not affect the creation of the right in rem. The unitary categorisation also implies an overarching interpretative technique to determine whether a given transaction falls within the newly-established category. This is performed through a ‘functionalist’ or ‘functional equiva lence’ approach, according to which consensual arrangements serving a similar function—ie establishing an interest in personal property to secure payment or performance of an obligation—are included in the same category and therefore are subject to the same legal treatment. Retention of title clauses, hire and purchase, and non-possessory pledges are thus subjected to the same publicity requirements, priority rules, and enforcement procedures. Although this approach is often labelled ‘substance over form’, it appears more accurate to define it as an approach that prioritises economic substance over legal nature. The functionalist interpretation focuses on the economic effects of a security agreement, rather than on its legal nature as it occurs, for instance, under English law.35 This results in a relative simplicity for the parties involved in a legitimate agreement to determine ex ante the priority ladder and enforce the right in the case of default. Within this framework, the characterisation of the right (and its recharacterisation) does not have consequences on the preferential status originally envisaged by the parties, being creation, perfection and priority regulated through similar legal principles. Although the debate over functionalism is a keystone of secured transactions law reforms, it is not an exclusive of this ambit of the law. The adoption of a rationale that investigates the economic effect of an agreement to ascertain its legal nature is common in various domains of financial law and developed as an alternative to the more formal approaches, such as numerus clausus,36 which relies on conceptual categorisations defined a priori to determine the perimeters of a legal regime.37 For instance, in the US, whether a financial instrument or an investment falls within the scope of the Securities Act of 1933 and the Exchange 34

This is particularly true for European countries, where different categories coexist. this regard it can be noted that English law focuses on the legal substance rather than the economic substance. See Gullifer (n 30) 4. 36 For a critique see B Rudden, ‘Economic Theory of Property Law: the Numerus Clausus Problem’ in J Eekelaar and JS Bell (eds), Oxford Essays in Jurisprudence (3rd series, Oxford, Clarendon Press, 1987) 242. 37 As noted elsewhere, the focus on the economic substance rather than on a formal classification allows for a greater flexibility of legal provisions when the underlying economic practices may take novel forms that are hard to be grasped by static conceptualisations; in the context of ‘securities’ as financial instruments see GG Castellano, ‘Towards a General Framework for a Common Definition of “Securities”: Financial Markets Regulation in Multilingual Contexts’ (2012) 12 Uniform Law Review 449, 458. 35 In

296

 Giuliano G Castellano

Act of 1934 is determined by the so-called ‘economic reality test’ that investigates the nature of the transaction based on its economic effects.38 When countries have adopted more formal categorisations, they have been exposed to the risk of regulatory arbitrage practices aimed at designing a transaction to fall outside the formal definition, thus avoiding the legal requirements attached therewith.39 A unitary, functionally based approach is also followed by international legal standards, elaborated by UNCITRAL and contained in the Legislative Guide on Secured Transactions (the Legislative Guide),40 the Supplement on Security Rights in Intellectual Property,41 the Guide on the Implementation of a Security Rights Registry (the Registry Guide),42 and the UNCITRAL Model Law on Secured Transactions, as currently drafted.43 International legal standards, as those elaborated in the UNCITRAL Model Law, aim at creating a clearly defined set of legal categories to be implemented in various jurisdictions presenting different levels of sophistication and complexity.44 Hence, the categorisation and the interpretative framework put forward should not cling to formal definitions. The use of neutral legal terminology translated into various languages is often a necessary step to reach an international consensus by mitigating the political implications of adopting legal constructions attached to a specific legal system.45 Such neutral terminology is designed to capture ‘synthetic legal concepts’,46 which are reflected in legal categories that include features of different legal systems. The reliance on the economic effect of the transaction may provide a useful ground to bypass national idiosyncrasies and identify the object of the standardisation process. For instance, the term ‘charge’ adopted by the European Bank for Reconstruction and Development (EBRD) differs substantially from the English law concept, by presenting features that resonate with the civil law concepts of ‘non-possessory pledge’

38

See the landmark decision SEC v WJ Howey Co 328 U.S. 293 (1946). This is the case of the ‘securities’ in Hungarian law; see T Tajti, ‘Central European Contribution to the American Debate on the Definition of “Securities” or Why does the Definition of “Security” Matter?’ (2005) 15(1) Transnational Law & Contemporary Problems 109. 40 In 2001 UNCITRAL entrusted a Working Group to define ‘a set of core principles for an efficient legal regime governing secured transactions to be inserted into a legislative guide’; UN, Official Records of the General Assembly, Fifty-Sixth Session, Supplement No 17, Doc A/56/17, para 357 (2001). See SV Bazinas, ‘The Work of UNCITRAL on Security Interests: An Overview’ (2015) 15 Uniform Law Review 315. 41 See SV Bazinas, ‘Intellectual Property Financing under the UNCITRAL Guide’ (2011) 43 UCC Law Journal 601. 42 On the need for specific registry recommendations see HC Sigman, ‘Some Thoughts about Registration with Respect to Security Rights in Movable’ (2010) 15 Uniform Law Review 507. 43 For a complete account of on the UNCITRAL’s works see Chapter 4 of this work, by S Bazinas. 44 For an overview in this field see O Akseli, International Secured Transactions Law (Oxford, Routledge, 2011). 45 For an account of the legal and linguistic dimension underlying the UNCITRAL’s works on secured transactions see S Bazinas, ‘Multilingualism in UNCITRAL’s Work on Security Interests’ (2012) 17 Uniform Law Review 413. 46 K Pistor, ‘The Standardization of Law and Its Effect on Developing Economies’ (2002) 51 American Journal of Comparative Law 97, 110. 39

Reverse Engineering the Law

297

or ‘hypothec’ (on movables), and ‘retention of title’.47 Similarly, the adoption of a functional equivalence approach in the Legislative Guide and the UNCITRAL Model Law to define the unitary category of ‘security rights’ is a necessary premise to establish a consensus in international negotiations.48 With regard to the actual implementation of this reform route, either through the adoption of international standards or through a purely domestic impetus, a series of practical difficulties may adumbrate its outright merits of offering a systematic organisation of secured transactions law regimes. A first set of considerations relates to the scope of a new category and to the functional test implemented to determine the perimeters of a new legal regime. For instance, although UCC Article 9 pioneered this reform route by adopting a unitary category of ‘security interests’, the Canadian Personal Property Security Acts takes this approach even further by subjecting long-term true leases to registration, which is not a requirement under UCC Article 9.49 A second set of practical issues relates to the dynamic nature of legal reforms. Implementing a unitary route based on functionalism implies, in various legal systems, a drastic shift from an existing condition—often governed by multiple categories and conceptual formalism—to a new conceptualisation of secured transactions. This translates into a chasm between the starting point, represented by the pre-reform legal regime, and the final results that a given reform process wishes to attain. The breadth of this chasm varies depending on the legal system under reform and on the final result. Nonetheless, national reforms have to factor in these constraints, which can be summarised as follows: (i) the resilience of existing legal categories and interpretative structures, when rooted in the legal culture; (ii) the complexity of the existing legal regime under reform; and (iii) its interconnectedness with related ambits of the law, such as property and insolvency law.50 The first strand of considerations focuses on the optimal definition of a unitary category. This aspect is critical, as the reform may ultimately deliver less than it promises. Cogent analyses of the Canadian PPSA and UCC Article 9 noted that

47 The EBRD Model Law Art 6.1 institutes three types of charges: (i) ‘registered charge’ according to which registration is an essential requirement to create the charge (Art 6.2.2); (ii) ‘possession charge’ created by possession (Art 6.4.2); and (iii) ‘unpaid vendor’s charge,’ which is a retention of title mechanism (Art 9). It is clear that these categories are a synthesis of various legal traditions without fully embracing any specific domestic regime. 48 The UNCITRAL Model Law defines security right as: ‘a property right in a movable asset that is created by an agreement to secure payment or other performance of an obligation, regardless of whether the parties have denominated it as a security right, and regardless of the type of asset, the status of the grantor or secured creditor, or the nature of the secured obligation’: UNCITRAL Model Law, Art 2. 49 See the Saskatchewan PPSA, s 3(2), which provides for registration to a lease with a term of more than one year. In general, on the law of secured transactions in Canada, see generally RCC Cuming, CA Walsh, and RJ Wood, Personal Property Security Law (2nd edn, Toronto, Irwin Law, 2012). A similar approach is also adopted in Australia and New Zealand, whereas the Cape Town Convention takes functionalism a step further and subjects any leases (financial or true) of any duration to its regime. 50 For the relationship between secured transactions law and insolvency law see Chapter 12 of this work, by J Garrido and E Smith.

298

 Giuliano G Castellano

when overarching categories lack a clear definition of what is a ‘security interest’, discretion and uncertainty are still present in the legal regime.51 In that sense, Roderick Macdonald noted that the functional approach could be conceived of as a device to compensate for the inability of common law to provide a clearcut definition of security interests,52 rather than a cultural rejection to formalism. When a functional approach governs a unitary category, an investigation into the nature of the interest is still required at least to mark the distinction between what is owed and what is owned.53 This implies that law reformers have to establish whether different legal treatments should govern a proprietary entitlement that ensures a priority right over an asset to a seller, who originally owned that asset, and the interests in the collateral held by a secured debtor, whose preferential status responds to a financing need. Once this distinction is made, the resulting legal framework may still require an investigation into the legal nature of a security. Similarly, the reluctance of different legal systems towards adopting functionalism as a guiding principle for a unitary categorisation appears to be more a scepticism towards its more extreme version, which is represented by the North American models. However, a marriage between functionalism and formalism—defined here as nuanced functionalism—may be achieved and more accurately reflect the menu of options available to law reformers. The approach has been well illustrated by Roderick Macdonald who, drawing from his direct experience as a law reformer, noted that the legal reforms in Ukraine and Quebec demonstrate how a regime may follow the unitary route, based on functionalism, while relying on a conceptual canvas that guides interpretative efforts.54 These considerations highlight that a unitary route could be constructed by deploying both a definition and a functional analysis. Although a nuanced functionalism may resolve these first issues, other difficulties, represented by the aforementioned practical constraints, still stand in the way of this reform route. This is well illustrated by the rich literature on legal change and transplants.55 Successes as well as failures constellate the history of legal changes through the plain appropriation of foreign ideas,56 as well as the convergence of laws through the adoption of international standards.57 New rules

51

See Bridge and others (n 18) 579. Macdonald, ‘Three Metaphors of Norm Migrations in International Context’ (2009) 34 Brooklyn Journal of International Law 604, 609. 53 Macdonald (n 19) 283. On this point see also I Davies, ‘The Reform of English Personal Property Security Law: Functionalism and Article 9 of the Uniform Commercial Code’ (2004) 24 Legal Studies 295, 301. 54 Macdonald (n 19) 281. 55 See M Graziadei, ‘Comparative Law as the Study of Transplants and Receptions’ in R Zimmermann M Reimann (eds) The Oxford Handbook of Comparative Law (Oxford, Oxford University Press, 2008). 56 A Watson, Legal Transplants: An Approach to Comparative Law (Edinburgh, Scottish Academic Press, 1974). 57 J Braithwaite and D Drahos, Global Business Regulation (Cambridge, Cambridge University Press, 2001). 52 R

Reverse Engineering the Law

299

may not be applied correctly, given the difficulty of understanding concepts originating outside the legal context in which they are ultimately implemented,58 or their introduction may ‘irritate’ existing legal concepts.59 A phenomenon of ‘normative latency’, to borrow from Friedrich von Hayek’s studies,60 and the resilience of ‘informal institutions’, as noted by Douglas North,61 may undermine the social and economic change sought by the implementation of new norms. This problem is exacerbated when legal change occurs through the transfer of legal rules from one cultural context to another or through the implementation of synthetic concepts contained in international standards.62 Hence, when convergence and legal change are pursued through norms that are exogenous to the recipient legal system, frictions with the existing legal and cultural environment might surface, as posited by Pierre Legrand with his provocative thesis on the impossibility of convergence among legal systems.63 National reform projects that follow international standards, although benefiting from further guidance, still face the problems related to the domestic reception of alien legal categories. International legal standards, such as those elaborated by UNCITRAL or by the EBRD, construct legal categories that are naturally alien (being ‘synthetic’) to the legal culture within which they are to be implemented and may require a drastic reorganisation of the legal regime originally in place. The UNCITRAL Model Law follows the ‘economic substance’ rationale reflected in the functional equivalence approach.64 Hence, for instance, the distinction between gage and nantissement (under French law) or retention of title, fixed and floating charge (under English law) becomes irrelevant vis-à-vis their legal treatment. As a result the chasm between those categories and the approach advanced by the Model Law is wide and its implementation, no matter how desirable, may encounter political and practical resistance. The UNCITRAL Legislative Guide addresses, at least to a certain extent, this issue. It does so through a distinction that acknowledges the existence of different legal categories to qualify functionally similar instruments. The Legislative Guide leaves to implementing countries the adoption of recommendations alternatively within a ‘unitary’ or a ‘non-unitary’ approach to classify security rights. When a

58

See Pistor (n 46). Teubner, ‘Legal Irritants: Good Faith in British Law or Unifying Law Ends Up in New Divergences’ (1998) 61 Modern Law Review 11. 60 See FA Hayek, Law Legislation and Liberty. Rules and Order (Chicago, University of Chicago Press, 1973). 61 DC North, Institutions, Institutional Change and Economic Performance (Cambridge, Cambridge University Press, 1991). On the conceptual closeness between neo-institutional economics and comparative law see I Biglino, ‘Formants and Institutions: Intellectual Meeting Points between Rodolfo Sacco and Douglass North’ (2011) 11(2) Global Jurist. 62 See R Sacco, ‘Mute Law’ (1995) 43 American Journal of Comparative Law 464. 63 See P Legrand, ‘Comparative Legal Studies and Commitment to Theory’ (1995) 58 Modern Law Review 262; and P Legrand, ‘The Impossibility of ‘Legal Transplants’ (1997) 4 Maastricht Journal of European and Comparative Law 111. 64 See UNCITRAL Model Law, Art 2. On the functional ethos of UNCITRAL see Bazinas (n 40). 59 G

300

 Giuliano G Castellano

country adopts a non-unitary approach, the UNCITRAL recommendations might be adopted and national legal categories retained, provided that the legal treatment of functionally similar transactions is the same.65 As Roderick Macdonald noted, the recognition that countries might adopt a non-unitary approach marks the most evident separation between the Legislative Guide and the North American model, translating into practice the idea that a project of international harmonisation must positively consider a certain degree of legal pluralism.66 From the above, it is possible to note that a reform route that provides a comprehensive (re)organisation of the law pertaining to secured transactions is to be preferred. Canvassing different security devices under an overarching category, governed by a functionally-based interpretative technique, represents a logical choice for legal reforms that aim at facilitating access to secured credit. This route may be intended as the most direct strategy to overcome the ‘obsolescence’ (borrowing from the reverse engineering terminology) of many secured transactions legal regimes, where multiple categories and different legal treatments for functionally-similar security instruments coexist, with material consequences on the credit protection sought by the parties to a security agreement. Nonetheless, such a route may not deliver what is expected. The reverse engineering perspective reveals that a number of constraints may hinder its correct implementation. As a result, the need for reform sought in many jurisdictions and the practical difficulty related to the overhaul of longstanding conceptual frameworks may ultimately drive national reformers to a sub-optimal strategy that adds new categories and ultimately locks reforms in this field into a path dependence loop. Hence, where a unitary route is not practicable, a new reform strategy must be devised. Such a route should ensure the same benefits of a unitary approach based on (pure or nuanced) functionalism—represented by a simplified legal framework whereby similar legal treatment for functionally similar security devices is ensured—while retaining the advantages of a non-unitary route, by enacting a reform that respects the constraints imposed by the existing legal system.

III. From Legal Functionalism to Reverse Engineering I have argued elsewhere that legal reforms could follow an alternative reform strategy.67 The idea is grounded in the fact that functionally-oriented publicity rules, based on modern registration standards, could be directly implemented without clashing with national idiosyncratic categorisations, while achieving the

65

The UNCITRAL Model Law, however, does not maintain this option. On the influence of UCC Art 9 in defining international secured transactions law, see Macdonald (n 19). 67 See Castellano (n 2). 66

Reverse Engineering the Law

301

same benefits as a more comprehensive reform strategy.68 To push this argument further, the analytical lenses of reverse engineering adopted in this chapter leads to considering the teleological aspirations and the societal function(s) performed by secured transactions. It is then possible to verify whether the proposed novel strategy maintains the legal components allowing secured transactions to perform their role while avoiding the aforementioned pitfalls of the two main reform routes just considered. To this aim, it is worth first drawing a distinction between functionalism as a method and functionalism as an interpretative technique.

A. The Functional Method in Secured Transactions Laws A functionalist methodology in comparative legal studies extends from the perimeters of functionalism in secured transactions law. In the latter context, functionalism represents an interpretative technique that allows both identifying and treating a variety of consensual arrangements used to secure the performance of an obligation through collateral. In comparative law, functionalism represents a methodology to compare different legal systems, prized by many for its practicality and dismissed by others for its lack of sufficient theoretical maturity.69 Notably, Konrad Zweigert and Hein Kötz defined the functional method as ‘the basic methodological principle of all comparative law’.70 Distinguished commentators have elaborated on the benefits of this methodology as well as on its limits.71 Methodological functionalism focuses on societal phenomena, which are factual effects of rules rather than the dogmatic structures represented by rules per se. Hence, the objects of a comparative enquiry are understood and analysed in relation to the function they perform in societies. It follows that legal norms and institutions, notwithstanding their formal classification in any given legal system, are comparable if they perform similar functions, which may be intended as the aspiration of tackling similar societal phenomena.72 It is clear that such a methodological approach is implicit in international projects aiming at harmonising secured transaction laws, and it constitutes an essential step for any meaningful comparative study in this field. Leaving aside the political

68 ibid.

69 The issue has been thoroughly considered from an interdisciplinary angle by R Michaels, ‘The Functional Method of Comparative Law’ in R Zimmermann and M Reimann (eds), The Oxford Handbook of Comparative Law (Oxford, Oxford University Press, 2008) 339. 70 K Zweigert and H Kotz, An Introduction to Comparative Law (Oxford, Oxford University Press, 1998) 33. 71 Most notably M Reinmann, ‘The Progress and Failures of Comparative Law in the Second Half of the Twentieth Century’ (2003) 50 American Journal of Comparative Law 671; J Reitz, ‘How to do Comparative Law’ (1998) 46 American Journal of Comparative Law 617; and H Collins, ‘Methods and Aims of Comparative Contract Law’ (1991) 11 Oxford Journal of Legal Studies 396. 72 In turn, this leads to adopting functionalism as an evaluative criterion, ie a benchmark to assess which legal arrangement(s) may better cope with the societal issue(s) under scrutiny; see Michaels (n 69) 342.

302

 Giuliano G Castellano

necessity of eliciting neutral terminology based on synthetic legal concepts, it is impossible to imagine that a group of national delegations, composed of legal experts, and international officers, would be able to engage in dialogue without identifying common ground. This common ground is represented by a cognitive framework that transcends domestic categories and tackles, in meta-legal terms, both a common goal, such as access to credit through secured transactions, and a shared understanding of the societal phenomena to be regulated, ie consensual arrangements establishing a preferential entitlement over an encumbered asset as a way to ensure an alternative form of repayment should debtors default on their obligations. Such a logical step is a prerequisite also for comparative studies in this area of the law. In fact, even opponents of functionalism as an interpretative technique find themselves adopting functionalism as a comparative law method when they consider how secured credit is regulated in different jurisdictions. For instance, an English lawyer may well oppose putting retention of title clauses, financial leases, and (fixed or floating) charges in the same bucket. Indeed, it may be argued that including these instruments in the same category could result in a mere juxtaposition of security devices with different legal natures. In turn, the appreciation of these diversities is pivotal for a consistent interpretation and grasp of national legal regimes. Nonetheless, such classifications are meaningless in foreign legal systems, where retention of title clauses, to keep with our example, are subject to publicity rules similar to those applied for full-fledged security rights, as is the case, for instance, in Italy, or true financial lease are included in the security interest category, as established by the PPSAs.73 Through the comparative prism, the pursuit of internal consistency is relinquished in favour of a methodological approach that allows us to transcend the idiosyncratic categorisation offered by any given legal system.74 Thus, a meaningful comparative investigation in this field must resort to a functional method to identify a common societal phenomenon and isolate the legal arrangements adopted in various jurisdictions to address it.75 Upon these premises, it is worth précising the societal phenomenon that secured transactions law reforms aim at tackling and therefore their core aspirational function. In industrialised economies, secured credit, and in particular non-possessory security interests, represent a key component for sustained growth and economic development.76 Through non-possessory secured transactions borrowers—in 73

See n 49 and accompanying text. The pursuit of internal consistency has been considered ‘the fetish of municipal lawyers’ and it has to be abandoned by any meaningful comparative analyses, see R Sacco, ‘Legal Formants: A Dynamic Approach to Comparative Law’ (1991) 39 American Journal of Comparative Law 24. 75 Alternative methodologies may lead to an arbitrary selection of domestic categories and their mere juxtaposition. 76 See G Gilmore, Security Interests in Personal Property Vol 1 (Boston, Little, Brown & Co, 1965) at 50. More recently Sir Roy Goode noted that the lack of an adequate legal regime for secured transactions hinders economic development: RM Goode, ‘Security in Cross-Border Transactions’ (1998) 33 Texas International Law Journal 47, 50. Ulrich Drobnig also pointed that the advancement of market economy implies a greater demand for secured credit: U Drobnig, ‘Secured Credit in International Insolvency Proceedings’ (1998) 33 Texas International Law Journal 53. 74

Reverse Engineering the Law

303

particular, small- and medium-sized enterprises—may gain access to credit by collateralising tangible or intangible assets that are integral to the operation of their businesses while retaining control and continuing to profit from those same assets. The underlying economic rationale may be understood as the possibility for a secured creditor of establishing ex ante an alternative repayment mechanism in case of default. Such a mechanism manifests its function in insolvency proceedings, when the right of priority established with a secured transaction bypasses the effects of the pari passu principle,77 according to which all creditors are treated equally and share pro rata the assets of the insolvent debtor.78 Notwithstanding the possible benefits related to a broader use of non-possessory secured transactions, credit institutions hesitate to accept non-real estate as collateral.79 With the establishment of a legal framework that is conducive to secured credit, legal reforms are intended to unlock the potential of secured transactions, by allowing security to be taken on a broader pool of assets. Empirical studies have noted that secured transactions law changes lending practices in the secured credit market with a direct impact on the assets used as a collateral and the availability of credit.80 Hence, moving from the possible economic benefits, secured transactions laws establish a set of legal arrangements that aim at: (i) enlarging the class of collateral over which security may be taken without dispossessing the debtor; (ii) ensuring clear enforcement powers to secured creditors; (iii) providing for a preferential claim and clear priority ladder for secured creditors over unsecured creditors and amongst competing claimants; and (iv) establishing a mechanism to publicly disclose the existence of security interests.81

77 The pari passu principle might be found in various legal systems, eg égalité des créanciers as established by Art 2093 of the French Civil Code, and par condicio creditorum contained in Art 2741 of the Italian Civil Code. 78 The very existence of the pari passu principle has been questioned to be the general rule; RJ Mokal, ‘Priority as Pathology: The Pari Passu Myth’ (2001) 60 Cambridge Law Journal 581, 585–587 and V Finch, ‘Security, Insolvency and Risk: Who Pays the Price?’ (1999) 62 Modern Law Review 633. For an early discussion on fairness in secured transactions, see RM Goode, ‘Is the Law too Favourable to Secured Creditors?’ (1983/84) 8 Canadian Business Law Journal 53. A civil law perspective over the ‘myth’ of the equal protection of creditors is presented by A Gambaro, ‘Il diritto di proprietà’ in A Cicu, F Messineo, L Mengoni (eds), Trattato di diritto civile e commerciale (Milan, Giuffrè, 1995) 15. 79 Many jurisdictions suffer from a mismatch between the assets held by small- and mediumsized enterprises and the assets lenders are willing to accept as collateral; see generally H Fleisig, MS Safavian, N de la Peña, Reforming Collateral Laws to Expand Access to Finance (Washington, World Bank Press, 2006); and MS Safavian, ‘Firm-Level Evidence on Collateral and Access to Finance’ in F Dahan and J Simpson (eds) Secured Transactions Reform and Access to Credit (Cheltenham, Edward Elgar, 2009) at 110. 80 Empirical study relating legal reforms to access to credit is well developed, for an overview see J Armour and others, ‘How Do Creditors Rights Matter for Debt Finance? A Review of E mpirical Evidence’ in F Dahan (ed) Research Handbook on Secured Financing in Commercial Transactions (Cheltenham, Edward Elgar, 2015); and CW Calomiris and others, ‘How Collateral Laws Shape Lending and Sectoral Activity’ (2015) Columbia Business School Research Paper No 14–58. 81 J Armour, ‘The Law and Economics Debate About Secured Lending: Lessons for European Lawmaking?’ in H Eidenmüller and E-M Kieninger (eds), The Future of Secured Credit in Europe (ECFR special volume, De Gruyter, 2008) 14.

304

 Giuliano G Castellano

B. The Function of Non-possessory Secured Transactions and the Role of Law Notwithstanding the variety of legal solutions deployed in domestic laws to address those items, a secured transactions law offers to secured creditors a level of protection against credit risk that is higher than the one normally accorded to unsecured creditors. Therefore, secured transactions serve the purpose of curbing credit risk, allowing borrowers to assume a less-risky position and consequently facilitate access to new credit. Through this lenses, non-possessory secured transactions are understood as risk mitigation devices that allow access to new credit without limiting the production capacity of the collateral.82 In light of the reverse engineering method here adopted, once the core function of secured transactions is isolated, the subsequent logical step is to determine what legal arrangements would ensure the development of that function. Nonpossessory secured transactions require a legal regime that reduces the uncertainty related to the credit-debt relationship to a level that is manageable, thereby making the financing operation appealing to the parties involved. Any investment presents an element of risk, given that being repaid or obtaining a return is inherently uncertain. Through secured credit, the lender is placed in the position of having a greater protection against that risk. To ensure that mechanism, the law should tackle two critical issues: first, the condition of asymmetric information that characterises any creditor-debtor relationship must be resolved; second, the law should ensure that uncertainty is kept to a level that still allows non-possessory secured transactions to function as risk-mitigation devices. Part of the law and economics literature identifies in the ability of resolving asymmetries of information the rationale, and therefore the justification, for having a secured transactions law regime in the first place.83 It must be noted that this may lead to a logical fallacy based on the confusion between the phenomenon and its function, or more accurately, between the nature (ontos) that justifies the existence of a given law and the tools through which that legal regime performs its

82

Castellano (n 2) 617. For an overview of the law and economics approach see: TH Jackson and AT Kronman, ‘Secured Financing and Priorities among Creditors’ (1979) 88 Yale Law Journal 1143; A Schwartz, ‘The Continuing Puzzle of Secured Debt’ (1984) 37 Vanderbilt Law Review 1051; JJ White, ‘Efﬁciency Justiﬁcations for Personal Property Security’ (1984) 37 Vanderbilt Law Review 473. For a more advanced understanding of secured transactions law see RE Scott, ‘A Relational Theory of Secured Financing’ (1986) 86 Columbia Law Review 901; A Schwartz, ‘A Theory of Loan Priorities’ (1989) 18 Journal of Legal Studies 209; RC Picker, ‘Security Interests, Misbehavior, and Common Pools’ (1992) 59 University of Chicago Law Review 645. For a specific elaboration over the role of asymmetries of information see GG Triantis, ‘Secured Debt under Conditions of Imperfect Information’ (1992) 21 Journal of Legal Studies 225; BE Adler, ‘An Equity-Agency Solution to the Bankruptcy-Priority Puzzle’ (1993) 22 Journal of Legal Studies 73. 83

Reverse Engineering the Law

305

function (telos). Once the nature of secured transactions is understood in terms of risk mitigation, the resolution of asymmetries of information represents one of the functions the law should attain. This point is key to determining what role the law should play in secured transactions. A thought experiment may clarify this point. If the resolution of asymmetries of information were the essence of secured transaction laws, in an ideal world without transaction costs or information asymmetries, secured transactions would not be necessary tout court. However, even in such a scenario, is not difficult to imagine that a prudent creditor would require an asset to secure the obligation as a form of insurance against default and business risks. In fact, the uncertainty over the creditor’s likelihood of full satisfaction is inherently unknown and does not depend upon the existence of asymmetries of information. Hence, secured transactions would still be required to manage such a risk. This is also in line with empirical findings showing that secured credit tends to be used mostly in connection with riskier firms.84 Alas, we do not live in such an idyllic world, and asymmetries of information, together with transaction costs, constitute a feature that hinders the ability of a secured transaction to be an effective device to mitigate credit risk. Where asymmetries of information are high, the distinction between a secured and unsecured creditor becomes thinner. Lenders without a preferential claim or lenders whose preferential claim cannot be enforced are subjected to the level of risk of any unsecured creditor. Similarly, the risk of a change in the priority ladder due to a recharacterisation of the right increases uncertainty and renders secured transactions a less effective risk-mitigation device. It appears that there are two legal elements preserving this mechanism. On the one hand, the functional equivalence (intended as interpretative technique) minimises the risk of recharacterisation, and thus it facilitates secured transactions to manage credit risk. On the other hand, publicity rules tackle the c ondition of asymmetric information between borrowers and lenders.85 Even more, when registration is a point of priority, the negative consequences of recharacterisation are further limited, given that the legal nature of the security instrument does not affect the priority ladder. From this perspective, legal institutions should fill the information gap by both requiring public disclosure of the existence of security rights (largely intended) and governing priority in order to ensure their enforceability.

84 See SS Chen, GH Yeo, KW Ho, ‘Further Evidence on the Determinants of Secured Versus Unsecured Loans’ (1998) 25 Journal of Business Finance & Accounting 371. 85 To continue with the mental experiment just introduced, it can be imagined that without asymmetries of information only a mechanism that makes known the existence of a security interest would be unnecessary.

306

 Giuliano G Castellano

C. Publicity Rules as the Cornerstone of Secured Transactions Law The reverse engineering approach reveals that publicity rules are not only a critical element of secured transactions laws,86 they also constitute the central element for a new reform route.87 In particular, registry rules are here understood as devices that, by correcting asymmetries of information, allow secured transactions to function as risk-mitigating devices, without limiting the production capacity of the collateral. Building on this understanding, reforming secured transactions laws could be accomplished well by reforming publicity rules, which alone may meet all the aforementioned aims of secured transactions legal reforms. First, a registry system that encompasses all the security instruments, regardless to their formal classification, automatically enlarges the class of collateral over which security may be taken without dispossessing the debtor. It is clear that retention of title clauses, financial leases or any other form of non-possessory pledge, such as charges, would converge within the same publicity requirements. Second, from a lender’s perspective, this allows for a complete outlook over the existing claims over an encumbered asset, and therefore, the enforcement powers may be more easily administered. Third, if registration is a point of priority, rather than the date of the agreement, a clear priority ladder for secured creditors is constructed. Fourth, transaction costs are reduced and negotiations are facilitated by a mechanism that publicly discloses the existence of security interests, and thus rendering the negotiations more transparent. The implementation of such a system would occur without modifying the categorisations that are embedded in national legal systems as long as the registry rules list the transactions that should be registered and how this affects priority. This avoids the legal transplant issues encountered with more drastic reforms, given that the building blocks of property, contract and even insolvency laws are not touched. Drawing from the Kahn-Freund distinction between ‘organic’ and ‘mechanical’ legal rules,88 publicity rules may be conceived of as ‘mechanical’ rules because they do not embed any cultural value while serving operational considerations.89 Even more, the application and interpretation of publicity rules— when not linked to the legal nature of the security—requires a cognitive process

86 See International Finance Corporation (IFC), Secured Transactions Systems and Collateral egistries (Washington, IFC Publishing, 2010) at 50. More generally see HC Sigman, ‘Some Thoughts R about Registration with Respect to Security Rights in Movable’ (2010) 15 Uniform Law Review 507. For an economic analysis see F Lopez-de-Silanes, ‘Turning the Key to Credit: Credit Access and Credit Institutions’ in F Dahan and J Simpson (eds), Secured Transactions Reform and Access to Credit (Cheltenham, Edward Elgar, 2008) 43. For a thorough analysis over the role of registries in law reforms see Chapter 10 of this work, by M Dubovec and HC Sigman. 87 Castellano (n 2) 632. 88 O Kahn-Freund, ‘On Uses and Misuses of Comparative Law’ (1974) 37 Modern Law Review 1. 89 This consideration primarily relies on the fact that publicity rules are loosely linked to legal culture sustaining the (organic) rules of property or contract laws; see Castellano (n 2) 638.

Reverse Engineering the Law

307

that does not create friction with the legal organic elements of the legal system.90 Hence, transplant issues may be avoided. In practical terms, the point is demonstrated in the subsequent section of this chapter.

IV. From Reverse Engineering to Re-Engineering: Implementing the UNCITRAL Registry System in the Italian Legal System To test the argument advanced here, it is necessary to identify the specific publicity rules and the legal system in which they could be implemented. This leads to a ‘re-engineering’ phase, where, after having deconstructed the legal regime to isolate its key components, an alternative route is constructed. To this aim, the main features of UNCITRAL’s Registry Guide, as also contained in the UNCITRAL Model Law, are briefly introduced and are considered as a standalone device for legal reforms that could be implemented without reforming the whole system governing secured transactions law at the national level. This reform route is then tested in the Italian legal system, one of the most complex, and most in need of reform, legal regimes currently operating in Europe.

A. Understanding the UNCITRAL Registry System The registry system envisaged by UNCITRAL deploys nine key elements: (i) the establishment of a general and centralised single registry for all security rights; (ii) open access to registry services (registration and search) with minimum fees; (iii) a notice-filing system, which does not require submitting a copy of the agreement, but rather a simple form; (iv) the registrar’s lack of scrutiny over the access to registry services, whether registration or search; (v) priority based on a first-to-file rule and effectiveness determined from the time the notice is accessible to searchers; (vi) the possibility of registering present and future security rights; (vii) the sufficiency of a single notice with minimum formalities to register multiple security rights between the same parties; (viii) the recognition that registration uniquely serves to grant effectiveness against third parties (perfection rule); and (ix) the possibility of searching through a grantor identifier (grantor-based).91 The functional equivalence approach lies at the core of the registry system, immediately affecting its scope. As mentioned earlier, regardless of whether a country opts for a unitary or non-unitary classification of secured transactions, functionally similar devices to establish a non-possessory security interest are 90 In order to be ‘mechanical’, publicity should not be conceived as a mechanism to protect false wealth or a gatekeeper for public faith. See Castellano (n 2) 619. 91 These features are reflected in the UNCITRAL Model Law at Ch IV.

308

 Giuliano G Castellano

s ubject to the same publicity regime and are included in a single registry system.92 The adoption of the functional equivalence approach—in addition to immediately enlarging the scope of the registry—results in a set of less obvious consequences defining the relationship between publicity regime and the category of ‘security rights’. By privileging (economic) substance over (legal) form, U NCITRAL envisaged a system in which registration does not carry formalities. It follows that the UNCITRAL registry system follows the perfection rule, according to which registration is a requirement only to ensure effectiveness against third parties.93 The effects of registration manifest as soon as the notice is available to search, according to a priority rule that follows a first-to-file rationale.94 The perfection rule represents a step towards a system that separates the enforceability of the security against third parties from the requisites designing the nature of the agreement. Hence, registration is detached from the ‘organic’ rule and performs the two key aforementioned functions of reducing asymmetries of information and attributing priority. Contrary to the regime in place in many national legal systems using the transaction-filing approach, but in line with the international practice,95 the UNCITRAL registry system does not follow a ‘public faith’ rationale. The registry only serves to flag the possible existence of a security right.96 In this scenario, it is left to the parties to further investigate the legal nature of the security right and determine its validity in accordance with national categories. The lack of formalities in the registration process establishes a set of incentives to expand the use of the registry—and therefore its reliability—without having an adverse impact on contractual freedom.97 In addition, the system allows advance registration to eliminate the gap between the creation of the right and its perfection against third parties.98 92 It follows that title-retention rights and lease arrangements, as they serve the same economic function identified in the definition of security rights, enter in the scope of the UNCITRAL registry. Even though the UNCITRAL Model Law does not envisage a non-unitary system, the same result is achieved; see Art 2. 93 Legislative Guide, Recommendation 33. 94 Registry Guide, Recommendation 11. In electronic-based registries, the moment a notice is submitted coincides with the moment when the information is searchable and thus effective. In paper-based registries, the information should be available as soon as practicable, within a short period of time. 95 The idea of reducing formalities is increasingly characterising modern registries since the notice filing system was adopted in 1930s in the US, with the UTRA and then absorbed in the UCC Art 9. Moreover, Art 8.4 of the EBRD Model Law on Secured Transactions, states that a ‘registration statement’ suffices. 96 According to the Legislative Guide a security right is created as soon as minimal formalities are met, such as the consent among parties (written form is an evidentiary matter); Legislative Guide, Recommendations 13–15. The registration per se does not constitute evidence over the existence of the underlying security right; UNCITRAL, Registration of Security Rights in Movable Assets, Doc A/CN.9/ WG.VI/WP.46 (New York, April 2011) at para 53. 97 The information required for an initial notice are basic and serve to allow the correct identification of: grantor, secured creditor, encumbered asset(s) and—if enacting states require it, the period of effectiveness and the maximum pecuniary value secured. Registry Guide, Recommendation 23. 98 Registry Guide, Recommendation 14. This approach minimises the risk of conflicts that will be resolved according to the first-to-file principle adopted to rank priority rights.

Reverse Engineering the Law

309

Finally, the effectiveness of a registry system to reduce transaction costs relies on both the availability on the information recorded therein and the ease of retrieving it. Without a single registry system, searches are conducted relying on assets descriptions, and creditors have to run a number of searches (in one or more registries or for multiple assets) to identify whether priority rights were established on specific assets. For this reason, the UNCITRAL registry system deploys a grantor-based indexing method,99 which reaffirms the role of registries as tools to facilitate private negotiations and the use of secured transactions to effectively manage credit risk. The key features of the Italian legal regime pertaining to secured transactions are introduced next in order to investigate the possible implementation of this apparatus of rules within an existing legal system, without modifying the existing legal categories.

B. The Italian Secured Transactions Law Regime: A Case of Path-dependency100 The complexity of the Italian laws governing secured transactions is the result of a tortuous historical development, epitomic of adjustments that have occurred through gradual legal changes. The premises of the Italian legal regime were set in the old civil code (1865) and subsequently assimilated in the Italian Civil Code (codice civile) of 1942, according to which a pledge might be created ‘by means of the delivery of the chattel or documents granting the right to dispose of the pledged assets to the creditor.’101 Moreover, according to the Civil Code, a security interest may be only created on assets that could be identified; hence, a security right over future assets takes effect once the collateral comes into existence and possession is surrendered.102 Legislative interventions as well as judicial and scholarly interpretations have progressively relaxed these rigid statutory principles.103 The ultimate result is an intricate legal environment that stifles access to credit.

99

Legislative Guide, Recommendation 54(h) and Registry Guide, Recommendation 34. The law is stated as it stood on 1 June 2016. 101 Art 2786 Italian Civil Code. 102 Art 2023 of the Italian Civil Code establishes the general rule according to which registration on future goods cannot be made until their existence. As exception to the general rule contained in Art 2023, the Italian maritime and navigation code (codice della navigazione) at Art 565 (concerning ships) and at Art 1027 (for aircraft) permit the registration of a security rights over ships and aircraft under construction or to be constructed. Conversely, motor vehicles follow the general rule. 103 The lack of a comprehensive general regime for security interests over movables has been often remarked in the last decades by Italian scholars; among the most notable contributions, see G Tucci, Garanzie sui crediti dell’impresa e tutela dei finanziamenti (Milan, Giuffrè, 1974); G Ferrarini, ‘Changes to Personal Property Security Law in Italy: A Comparative and Functional Approach’ in R Cranston (ed) Making Commercial Law: Essays in Honour of Roy M Goode (Oxford, Oxford University Press, 1997) at 477; and M Bussani, ‘Il modello italiano delle garanzie reali’ (1997) Contratto e Impresa 163. More rencently, see A Veneziano, Le garanzie mobiliari non possessorie (Milan, Giuffrè, 2000); and A Candian, Garanzie mobiliari, modelli e problemi nella prospettiva europea (Milan, Giuffrè, 2001). 100

310

 Giuliano G Castellano

This is also reflected in the World Bank’s rankings,104 and in a recent Country Report drafted by the International Monetary Fund, where it has been noted that a reform of secured transactions laws in Italy would benefit the credit environment, contributing to curbing the risks associated with non-performing loans.105 The process of gradual adjustment, absent a systematic reorganisation or amendment of the provisions contained in the Civil Code, renders a comprehensive legal reform more arduous, thus reinforcing a path-dependency loop, whereby direct legislative interventions render the system more complex and novel reforms are repeatedly required. This is demonstrated by the recent introduction of a new security instrument that is canvassed in a novel legal category, ie the ‘non-possessory pledge’.106 It follows that the route taken by Italian law reformers to facilitate the recourse to non-possessory secured transactions is anchored on the adoption of special laws that add new security instruments to the corpus of rules enshrined in the Civil Code, with the intent of overcoming its rigid provisions. Therefore, before examining the latest addition and advancing a possible alternative strategy, it is worth introducing the main traits of the Italian legal regime pertaining to consensual secured transactions over movables. The Civil Code, under Article 2741, defines three categories of consensual security rights attributing a priority right over movable assets to secured creditors. These are: pegno, which is a possessory pledge over movables; ipoteca (literally, ‘hypothec’), which is a non-possessory security right over both immovable assets and selected registered movables, such as motor vehicles, ships, and aircraft; and privilegi (literally, ‘privileges’), which are preferential claims, established by the Civil Code or enacted through special laws and affecting specific movable assets. Against this backdrop the Italian legal regime pertaining to secured transactions presents the following key features: (i) Delivery of possession represents the basic form of publicity and is a necessary element to create a pledge (pegno) pursuant to Article 2786 of the Civil Code. It is precisely in consideration of the possessory nature of the pledge that the Civil Code also attributes to this instrument the highest level of priority. Hence, Article 2748 of the Civil Code establishes that, as a general principle, pledges have priority over preferential claims and are immediately

104 The Doing Business report on Italy in 2013 registered an increased difficulty to access to credit in an already difficult framework. Hence, in 2013 Italy has been downgraded from position 97 (as of 2012) to position 104, where position 185 represent the system with the highest difficulty in accessing to credit and 54 is the EU average. The WB & IFC, Doing Business in Italy 2013, Washington (2013), 4. More recently, the Doing Business score attributed to Italy for ‘getting credit’ has improved, reaching position 97 in 2016, as it was in 2012; see www.doingbusiness.org/data/exploreeconomies/italy (last accessed April 2017). 105 International Monetary Fund, ‘Italy—Selected Issues’ (July, 2015) IMF Country Report No 15/167, in particular see Legal Recommendation 42 at pp 66–67. 106 As further illustrated below, the new category has been introduced, without amending the existing regime, with D.l. n 59 3rd May 2016, published in Gazzetta Ufficiale n.102 on 3 May 2016.

Reverse Engineering the Law

(ii)

(iii)

(iv)

(v)

311

after statutory debts, such as court expenses (Articles 2751bis and 2777 Civil Code)—unless otherwise provided by the law. With regard to security rights on registered movable assets, known as ipoteche mobiliari (literally, ‘hypothecs on movables’), registration is intended as a substitute for dispossession. Registration is thus a necessary formality to create a security right over those movables.107 Such a category could be defined as a form of chattel mortgage, whose legal treatment has been moulded around the scheme for mortgages on real estate, known as ipoteca immobiliare (literally, ‘hypothec on immovable assets’). Hence, the Italian Civil Code considers pegno as a security right created through dispossession; whereas ipoteca is a security right created through registration. To overcome the narrow contours of this categorisation, different nonpossessory security instruments have been established through special laws, mostly in the form of privilegi, such as those concerning security rights established over certain types of dairy and cured-meat products, as well as the preferential claims that credit institutions may establish over the pool of companies’ assets. These instruments rest in between the categories of non-possessory pledges (pegno) and chattel mortgages (ipoteca mobiliare) and their constitutive laws generally provide for them to be preferred, as an exception to the aforementioned general principle, to the possessory pledge. For preferential claims established through these special laws, registration is required to render the security instrument enforceable against competing claims and, in some circumstances, also to create the security right (see below). In general, for non-possessory security rights over movables, registration is the most common publicity requirement. It follows a transactionfiling approach, imposing a requirement to file a (notarised) copy of the agreement establishing the security right. Nonetheless, different publicity regimes apply for each instrument and various, uncoordinated registry systems coexist. Finally, the role of notaries is deemed essential, as they are considered gatekeepers safeguarding public faith.108 Notaries, under their professional code of conduct, have a duty to verify the legality of the agreement before registering it. This implies that the registrars do not have to control the information submitted, given that an a priori control is ensured through this specialised legal profession. Given the costs associated with notarial activities,

107 Art 2810, para 2 of the Civil Code extends the regime for mortgage on land to a limited number of movable goods, which are easily identifiable as they are registered in special registers originally established for licensing and regulatory purposes: Registro Aeronautico Nazionale (RAN) for aircrafts; Pubblico Registro Automobilistico (PRA) for motor vehicles; and Registro delle Imbarcazioni da Diporto (RID) for ships. 108 See generally U Mattei, Regole sicure. Analisi economico-giuridica comparata per il notariato (Milan, Giuffrè, 2006).

312

 Giuliano G Castellano few legislative interventions have attempted to limit the role of notaries to facilitate access to credit at a lower cost.109 However, in line with the Italian civil law tradition, notaries are entrenched in the Italian legal system, and a change in their function would demand a broader reform.

The resulting legal regime is incredibly complex, especially when compared to the international standards defined by the UNCITRAL or EBRD. Within this framework functionally similar security instruments are subjected to different legal treatments, with material implications on the priority ladder. More in-depth considerations over the existing instruments to take security over movable assets are required in order to unravel possible ambiguities and overlaps with the recent reform. Following the Italian culinary tradition that appreciate ham and cheese as an amuse-bouche, the analysis of the Italian publicity regimes for secured transactions should start from the special laws dealing with security interests in those assets. In 1985, a special law was enacted to regulate security rights over hams classified with ‘denominazione di origine controllata e garantita’ (docg),110 which is the Italian predecessor of the EU quality schemes.111 In 2001, the regime for hams was extended to cheeses also classified as either docg or Protected Designation of Origin (PDO).112 The rationale for these laws establishing a consensual preferential claim (privilegio) lies in the fact that traditionally cured meat and cheese are products that requires initial investments that are high for local producers, while the value of those assets is expected to increase over time. Furthermore, rigorous maintenance operations are necessary to conform with the requirements demanded for PDO certification. While the need of special laws for dairy and cured-meat products appears a singular choice, a closer look reveals that direct legislative interventions were intended to preserve the local and national economy. For instance, parmigiano reggiano cheese represents a valuable market that sustains a local economy involving 50,000 people in the production process with a business turnover in the global retail market of more than €2 billion (2013 p roduction).113 However, the financing schemes offered by the archetypical categories of Italian secured transactions laws were inadequate to sustain the credit needs of the 109 For instance, the so called ‘decreto Bersani’ (Legislative Decree n 223, 4 July 2006) abolished the exclusive competence of notaries for security rights on cars, allowing public officials managing the register and civil servants of municipalities to certify the authenticity of the agreements and register the transaction. Similarly, the statutory requirements establishing the ‘bank charge’ mandatorily halves the fees of notaries (see below). 110 Law 24 July 1985 n 401. 111 See Regulation (EU) 1151/12 of the European Parliament and of the Council on Quality Schemes for Agricultural Products and Foodstuffs [2012] OJ L343/1. For a commentary on EU quality schemes see A Tosato, ‘The Protection of Traditional Foods in the EU: Traditional Specialities Guaranteed’ (2013) 19 European Law Journal 545. 112 Art 7 of the Law 27 March 2001 n 122. 113 Moreover, according to the consortium of producers of parmesan cheese the 2013 production generated a business turnover in 2014 worth more than €1 billion; see www.parmigianoreggiano.it (last accessed April 2017).

Reverse Engineering the Law

313

local businesses producing this dairy product. Each loan generally finances from 70 per cent to 80 per cent of the value of the encumbered asset, which is a wheel weighing in average forty kilograms and a wholesale value fluctuating constantly and worth approximately €7/kg to €9/kg in 2014.114 Therefore, the requirement of dispossession imposes on the financing industry the establishment of specific infrastructures as well as the acquisition of skills and knowledge in order to ensure the completion of a delicate process of aging.115 An ad hoc legislative intervention substituted dispossession through a publicity system that requires both the registration in a special registry—kept by the producer and publicly accessible through the Registro delle Imprese, the Company Registry Book (CRB)—and a seal engraved on each item, as a way to ensure transparency against third parties as well as the traceability of the encumbered assets. The legal consequences if only one of the two requirements (registration or sealing) is met are not clear. Due to the lack of developed case law, a strict interpretation suggests that both requirements are necessary not only to make the right in rem effective against third parties but also to create the security interest. This special regime is emblematic of a legislative technique that aims at maintaining the core conceptual structure established in the Civil Code, while very specific provisions are enacted to broaden access to credit for a sector of interest to the national economy. Other, more general instruments are available to establish and publicise nonpossessory secured transactions (broadly conceived). According to Article 1524 and following of the Civil Code, the agreement establishing a retention of title clause on machinery with a value of €15.49 or more must be publicly recorded. Following an asset-based approach, a notarised copy of the transaction must be filed with the registrars located in one of the 167 First Instance Courts (Tribunali). The exact registry is determined in accordance with the location of the encumbered asset at the time when the retention of title clause is established, following the rules defining the jurisdiction of the correspondent court. Registration is required only to claim ownership against sub-buyers, whereas an ascertained date is necessary to make the agreement effective against third-party claimants.116 There are no provisions regulating the circumstance in which the encumbered property is moved, even within the Italian territory. In practical terms, it follows that to determine the existence of a priority right (established through retention of title), a search through 167 paper- and asset-based registries is required. Non-possessory security rights on assets held in use by companies are normally established under Article 46 of the Italian banking law, Testo Unico Bancario

114 ibid.

115 The financing model of a local bank, Credito Emiliano SpA (Credem), has been presented as a case study at the Harvard Business School. The authors noted how the bank adopted a unique approach, requiring the use of special vaults to keep in custody approximately 300,000 wheels of cheese that are regularly controlled to avoid their deterioration; see N Trichakis, G Tsoukalas, E Moloney, ‘Credem: Banking on Cheese’ (March 2015) Harvard Business School Case 615. 116 Art 1524 Civil Code first paragraph.

314

 Giuliano G Castellano

(TUB),117 which encompasses not only machinery but also present and future assets explicitly enumerated in the law.118 Article 46 TUB is the most commonlyadopted device to access secured credit for operations of acquisition financing. It offers the possibility for regulated credit institutions to protect a loan through a ‘security package’ that includes a large portion of companies’ assets. This general security instrument has been introduced pursuant to Article 2777 of the Italian Civil Code, to which Article 46 TUB explicitly refers. Hence, Article 46 TUB is a consensual preferential claim (privilegio speciale) that, from a practical standpoint, works as a company charge over the pool of assets that are functional to the business activities. However, to distinguish it from the company charge in use under other national and international laws, the locution ‘bank charges’ is adopted here, given that Article 46 TUB applies exclusively to charges made between companies and regulated credit institutions, for medium- and long-term loans, ie lasting more than 18 months. Bank charges concern assets that are not subjected to registration in special registries; thus chattel mortgages (ipoteche mobiliari) are left outside the scope of the provision, together with security interests over registered intellectual property rights, such as patents.119 Bank charges should be registered in the same registries established for retention of title clauses. However, a different publicity regime is applied. Registration for bank charges renders the agreement effective against third parties, and it is a point of priority, following a first-to-file rule. In an attempt to simplify the search and provide a partial solution to the scattered registry system dispersed among different local courts, notarised copies of the title establishing the bank charge must be filed both in the registry kept in custody by the Tribunale that is territorially competent in the secured creditor’s residency and in the registry where the lending credit institution has its seat. The fees for registration are not high. However, the borrower ultimately bears the costs of the notary services, which are calculated on the basis of the value of the transaction. The Italian legislators attempted to reduce the cost of secured credit by mandatorily establishing that ‘notary fees shall be reduced by half ’.120 The registration procedure, however, is further complicated by the requirement of an exact description of the collateral.121 Such a provision reflects the prudential regulation requirements contained in the Basel Accords that impose to regulated credit institutions a detailed description of tangible assets

117

D.lgs. n 385 1 September 1993, with subsequent modifications and amendments. particular: (a) existing or future plants or works, licenses, capital goods or any asset that is instrumental to the business; (b) raw materials, semi-manufactured products, inventory, finished products, crops, fruit, livestock and merchandise; (c) any goods acquired with the loan secured by the charge; (d) receivables, including future receivables, deriving from the sale of goods referred to in the preceding categories; Art 46(1) TUB. 119 For a treatise on the issues related to security rights on intellectual property rights under Italian law see Chapter 14 of this work, by A Tosato. 120 Art 46(6) TUB. 121 Along with: the details of lender(s), borrower(s), third-party grantor(s) (if any); the exact amount of the loan; and the exact amount secured. 118 In

Reverse Engineering the Law

315

taken as collateral in order to benefit from reduced capital charges.122 However, because Article 46 TUB allows to encumber the pool of present and future assets, distinguished commentators have argued that a sufficiently detailed description enabling the identification the encumbered assets should suffice.123 Pursuant to Article 2777 of the Civil Code and Article 46(4) TUB, bank charges are subordinated to statutory debts, ie court expenses and employees’ claims, as well as to dispossessory pledges established before the (double) registration of the agreement creating the charge. There are, however, some ambiguities that, absent a sufficiently developed case law, give rise to different scholarly interpretations and flag the lack of a rational organisation of the Italian regime for secured transactions over movables. First, courts have indicated that a correct interpretation of the Civil Code requires possessory pledges to be considered as preferred to employees’ preferential rights.124 Nonetheless, it is commonly acknowledged that Article 46(4) subordinates bank charges to both employees’ statutory claims and to pledges created prior to the registration of the charge.125 It is therefore uncertain how to construct the priority ladder among consensual and statutory claims when a pledge is created over an asset that is also included in the pool of collateral encumbered by a bank charge. Second, if a pledge has been validly constituted, even after the registration of the bank charge, the Civil Code—by favouring rights in rem that are manifested through physical possession—protects the subsequent pledgee in good faith, as it does for sub-buyers in good faith.126 Albeit the norm provides for bank charges to follow the encumbered assets if they are transferred, Article 46(5) TUB also stipulates that, if the preferential claim cannot be enforced against subsequent sub-buyers, because in good faith, the charge is automatically passed onto the proceedings resulting from the disposal of the assets. It is therefore debated whether a prior registration of a bank charge offers a sufficient ground to sustain, should a controversy arise, that a subsequent pledge (over assets already included in the pool encumbered by the bank charge) could be considered as c reated against good

122 See Basel Committee on Banking Supervision (BCBS), ‘International Convergence of Capital Measurement and Capital Standards—A Revised Framework’ (Bank for International Settlements, 2004, revised 2006), para 522. The provision has not been amended with the adoption of Basel III and is contained also in the new Capital Requirements Regulation, which implements Basel III in the European Union; see Article 210(d) 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, [2013] OJ L 176/1. 123 See A Veneziano, ‘La garanzia sull’intero patrimonio dell’imprenditore della nuova legge bancaria italiana al confronto con i modelli stranieri: una riforma a metá?’ (1996) Diritto Commerciale Internazionale 938. A more recent decision delivered by one of the first instance courts, however, has reaffirmed the principle of exact description, see Tribunale of Mantova, 23 November 2006. 124 See Corte di Cassazione, 17 February 1996, n 1238. 125 See G Tucci, ‘Garanzie mobiliari non-possessorie’ in F Capriglione and G Alpa (eds) Commentario al Testo Unico delle Disposizioni in Materia di Intermediazione Finanziaria (Milan, Giuffrè, 2012) at 546. 126 The principle, referred to as possession vaut titre, is enshrined in Art 1153 of the Civil Code, which is not applied for registered movables, whereby registration substitutes possession under Art 1156 of the Civil Code; see also Ferrarini (n 103) 483.

316

 Giuliano G Castellano

faith.127 If this is not the case, a subsequent pledge may be validly constituted and the bank charge would be de facto defeated by a subsequent security right. These ambiguities are a direct effect of a special publicity regime established for bank charges. The priority status of bank charges is established through a registration mechanism that operates in the shadow of a longstanding (and codified) principle that considers dispossession as the paramount form of publicity. The Italian bank charge is epitomic of a reform route that, instead of offering a systematic (re)organisation of a legal regime that is not up to the new societal and economic necessities, adds further layers of complexity. This is done by merging different rationales, such as the logic of preferential claim under the provisions of the Civil Code, capital adequacy standards for credit institutions requiring a detailed description of encumbered assets, the registry mechanism originally established for retention of title clauses, and special dispositions to reduce notary fees. Further overlaps appear to be a natural effect of such a reform route. Given the amplitude of the provision contained in Article 46 TUB, the regime for bank charges overlaps with other provisions regulating security interests over receivables, retention of title clauses, financial leases admitted under Italian law (which do not require registration),128 and, of course, the aforementioned provisions over ham and cheese. This results in the coexistence of different regimes (bearing distinct legal consequences) for functionally equivalent security interests. In general terms, security rights over receivables follow specific publicity provisions that do not require registration in public records, in line with relevant European legislation.129 For instance, under Italian law, a debenture may be encumbered via a security transfer (cessione a scopo di garanzia) that, being a transfer of ownership, does not require registration. Alternatively, a pledge on claims (pegno su crediti) may be created. In this case, a special form of registration is required for taxation and prudential regulatory purposes.130 Shares and

127 Some scholars noted that registered bank charges have priority over any subsequent security interest, including possessory pledges; see Tucci (n 125) 542. 128 Financial lease (leasing finanziario) in Italy until the 1970s was considered to be against the provisions on retention of title. The practice is now becoming accepted, thanks also to the UNIDROIT Convention on International Financial Leasing of 1988 that entered into force in Italy in 1995 with the Law 259 of 14 July 1993. Giorgio De Nova, Il contratto di Leasing (1995); M Bussani, Proprietàgaranzia e contratto. Formule e regole nel leasing finanziario (Trento, Università degli studi di Trento, 1992). Sale-and-leaseback arrangements are still considered against the law, as contrary to the prohibition of pactum commissorium contained in Art 2744 of the Italian Civil Code. Nonetheless, when exact estimation of the value of the asset at the time of enforcement is determined courts tend to recognise the legitimacy of the agreement. Cass. 16 October 1995, N. 10805, in 1 Foro It. 3492 (1996), commentary of Alberto Monti. 129 See notably, Directive, 2009/44/EC of the European Parliament and of the Council of 6 May 2009 amending Directive 98/26/EC and Directive 2002/47/EC OJ L 146/37 of 10.6.2009. On security interests over receivables see Chapter 14 of this work. 130 Pledges on claims are regulated by the codice civile at Art 2800ff, which coordinates with the general discipline on pledges on movables established by Art 2784 cc and requires a written form and a notification to the debtor with ascertained date to be opposable against third parties. In the lack of one of these formalities the pledge is not opposable against third parties.

Reverse Engineering the Law

317

stocks may also be pledged (pegno su azioni).131 In the case of security interests in shares of unlisted companies, registration must be made in the CRB.132 In the case of security interests in stocks of listed companies, given the dematerialisation of these titles since 1998,133 financial instruments must be recorded both in the CRB and in a special account managed by the intermediary named conto vincoli (liabilities account).134 However, securities in receivables and certain financial instruments (if part of the overall pool of assets of a company) may fall within the discipline of bank charges, which requires registration to establish priority and be effective against third parties. A second matter of concern regards the overlaps between retention of title clauses and Article 46 TUB. Let us consider a practical scenario. An agreement of retention of title with an ascertained date has been validly constituted but not registered. Subsequently, a bank charge on the same asset is established in favour of another secured creditor. In such a circumstance, it is unclear whether the retention of title with an ascertained date that precedes a registered bank charge on the same asset would prevail.135 The registration of the agreement establishing a retention of title only creates an interest with priority against sub-buyers, whereas the ascertained date determines priority among the competing creditors. Nonetheless, if the view that the ascertained date prevails over registered bank charges is adopted, then Article 46 TUB would be stripped of its meaning, given that through a retention of title clause, a secured transaction could be created without being searchable and enjoying a higher priority status. Any attempt to reform such a complex system, no matter how desirable, encounters a series of practical difficulties. This has been witnessed by other ambitious attempts to reshape the pillars of property law in Italy that are still waiting to gain political momentum.136 However, incremental adjustments aimed at facili-

131 Originally, pledges over shares needed to be registered in the Libro Soci (Shareholders’ Ledger), which has been abolished by Legislative Decree n 185 29 November 2008, and now absorbed in the Registro delle Imprese, where pledges over security have to be registered. 132 See Royal Decree n 239 of 29 March 1942. A similar provision also applies to nominative stocks of shares following Articles 1997, 2024 and 2026 of the Italian Civil Code. Moreover, a pledge over a nominative stock can be created with ‘double annotation’ (doppia annotazione) or with the ‘security endorsement’ (girata in garanzia). 133 The dematerialisation—regulated by D.lgs. 24 June 1998, n 213 (and its subsequent amendments) on the introduction of the Euro into the Italian legal system—had had a significant impact on pledges over financial instruments, see F Sartori, ‘La dematerializzazione degli strumenti finanziari’ (1999) Rivista di Diritto Civile 275. 134 Art 87 of the consolidated law on finance, Testo Unico sulla Finanza (TUF), D.lgs 24 febbraio 1998, n 58 and its subsequent amendments. 135 M Rescigno, ‘Le garanzie rotative convenzionali: fattispecie e problemi di disciplina’ (2001) 1 Banca, borsa e titoli di credito 21. According to Giuseppe Tucci the registration of a bank charge would prevail any time is registered prior to the title retention; see Tucci (n 125) 533. 136 With the intent to change the definition of property rights in the Italian Civil Code in order to expand their social dimension, in 2007 the Minister of Justice established a Commission chaired by Professor Stefano Rodotà. Notwithstanding the cultural movement generated by the project, the reform proposal did not become law.

318

 Giuliano G Castellano

tating access to secured credit directly contributed to the intricacy of the system. This path emerges also from the newly-added security device. Recently, a new, non-possessory security instrument has been adopted to take security over movable assets, without overhauling the legal framework just presented. On 3 May 2016, the Italian Government presented a law decree, named ‘banks’ decree’ (decreto banche), that introduced a ‘non-possessory pledge’ (pegno mobiliare non possessorio) in Italy.137 The decree has been presented and drafted by the Government following the urgency procedure for delegated legislative acts pursuant to Article 77 of the Italian Constitution. The urgency character has been attributed to the core provisions contained in the decree, the primary intentions of which have been to: expedite the enforcement procedures to satisfy banks’ claims; reduce the impact of non-performing loans in the Italian banking system; and strengthen the protection of investors from losses resulted from recent banking resolutions. In its first chapter, probably to sweeten the pill of what has been perceived as another emergency measure to heal the Italian banking sector, the decree introduces new measures to facilitate access to credit for companies and small businesses, a task that appears to be chiefly mandated to the first article of the decree that introduces a new legal category, ie the non-possessory pledge.138 The article, comprised of ten succinct paragraphs, has to be praised for its simplicity in attaining the long-awaited introduction of a general non-possessory security instrument. According to the new provision, entrepreneurs registered in the Company Registry Book may establish a security right over present and future assets that are functional to their business operations, with the exclusion of registered movables.139 The encumbered assets may be disposed and sold in the ordinary course of business and the security right is automatically transferred upon the proceedings.140 To be validly created, a non-possessory pledge must be in written form and be registered in a new, electronic registry system,141 which is held by the Italian Tax and Revenue Agency (Agenzia delle Entrate) and is regulated through delegated acts enacted by the Ministry of Finance.142 Priority is determined following the first-to-file rule, without explicitly requiring a detailed description of the encumbered assets. If the secured debtor is in default, the

137 D.l. n 59 3rd May 2016. On June 2016 the Italian Parliament converted into law the decree; Law 30 June 2016 n 119/2016 [GU n 153, 2 July 2016]. In particular, while keeping the overall structure and ethos of the decree introduced by the Government, the Senate introduced some modifications on the function of registration. For an analysis of the final text governing non-possessory pledges in Italy see GG Castellano, ‘The New Italian Law for Non-possessory Pledges: A Critical Assessment’ (2016) 31 Butterworths Journal of International Banking and Financial Law 31, 542. 138 ibid Art 1. 139 ibid paras 1 and 2. 140 ibid para 2. 141 ibid paras 3 and 4. Albeit registration appeared to be a mandatory requirement in the decree, the amendments introduced by Parliament seem to indicate that registration is a requirement that renders the pledge effective against third parties. Yet there might be some interpretative ambiguities; see Castellano (n 137) 543–44. 142 ibid para 6.

Reverse Engineering the Law

319

c reditor may sell the encumbered assets, also through out-of-court procedures.143 Furthermore, if p rovided in the security agreement and after fulfilling specific formalities, the creditor may lease or even take possession of the encumbered assets.144 It is preserved the right of the debtor to seek judiciary remedies, in the form of compensatory damages, within three months from the beginning of the enforcement procedures, should the enforcement procedures not be compliant with legal requirements contained in the norm.145 Finally, under insolvency law, the non-possessory pledge has the same legal treatment attributed to the ordinary, possessory pledge.146 If read discretely from the corpus of the aforementioned stratified provisions, the new norm appears to mark an important achievement in the reform process of secured transactions laws in Italy, and one that is in line with the spirit of the UNCITRAL Model Law. However, the norm does not operate in a vacuum and it does not amend any of the existing security instruments. This runs the risk of adding further ambiguities, which are considered next, to a system already daunted by an excessive complexity. The qualification of the new legal category as non-possessory pledge (pegno non possessorio) is already problematic. According to the Italian Civil Code, as illustrated above, pledges are characterised by dispossession, whereas ipoteche (hypothecs) require registration to be validly established and to be enforceable against competing claims. It appears that the new instrument would be better qualified as ipoteca mobiliare (literally, an hypothec on movables), given that registration is a substitute for dispossession and without such a formality a security right pursuant the new norm would not be validly created. The issue is not merely taxonomical. In fact, it contributes to interpretative ambiguities in case of dispute, where reference to the category of ‘pledge’ may lead to question the role of publicity and its priority status vis-à-vis preferential claims (privilegi) and possessory pledges. Few examples of potential conflicts would clarify this point. Bank charges enjoy a higher degree of priority than subsequently established possessory pledges that are not constituted in good faith.147 It stands to reason that the same applies to the new, non-possessory pledge. However, this opens up a series of new ambiguities that may undermine the efficacy of both bank charges and the newly introduced security instrument. If a non-possessory pledge is established and registered in the new registry system after the creation of a bank charge, it is arguable which one of the two concurring security instruments should prevail in terms of priority. Because registration for a non-possessory pledge equates to dispossession, it is to be determined whether the same provisions protecting a subsequent dispossessory pledge created in good faith apply, by analogy, to a

143

ibid para 7(a). ibid para 7(c) and (d). ibid para 9. 146 ibid para 10. 147 See above n 127 and accompanying text. 144 145

320

 Giuliano G Castellano

non-possessory pledge. If this is the case, it is to be ascertained whether a nonpossessory pledge could be constituted in good faith, even if a prior bank charge encumbers the same assets. There are two possible interpretative solutions that may be advanced. First, following a systematic interpretation of Italian law, it could be noted that the publicity regime for the newly-introduced non-possessory pledge has a stronger legal ground than the one governing bank charges. In fact, in the former case, registration is an essential formality to render the security right valid between the parties, whereas in the latter case, registration is required uniquely for priority and enforcement purposes. Drawing from the Italian legal regime concerning the publicity requirements for chattel mortgages, it might be argued that the new non-possessory pledge always prevails, even against bank charges that have been established beforehand.148 Hence, it could be argued that, by introducing a nonpossessory pledge that follows a similar logic to create a right in rem, the Italian legislator has placed the onus to search the registry manged by the Tax and Revenue Agency upon potential lenders. Therefore, if a search in the newly-established register does not reveal any proprietary entitlement, the assets may be presumed to be unencumbered. As an important corollary, it could be sustained that conducting a search in those records satisfies (per se) the threshold of good faith. As a result, following this interpretation, previously established bank charges, being registered in another registry system and under a different publicity regime, may be superseded by a subsequently established non-possessory pledge. A second and different interpretation may lead one to consider a bank charge to prevail over a subsequently established non-possessory pledge, even if created in good faith. With reference to Article 46(5) TUB, it could be stated that priority follows a first-to-file principle that affects also security rights that are not subjected to the same publicity regime.149 It may be also argued that the protection given to rights in rem created in good faith cannot be analogically extended to the non-possessory pledge, even if formally qualified as a ‘pledge’, because, by definition, delivery of possession is excluded. If such an interpretation is accepted, the new norm would impose on lenders a requirement to conduct a search also into the paper-based registry system held by the courts of first instance. All in all, the complexity of the Italian legal regime for secured transactions does not appear resolved; if anything, it appears deepened. Regardless of the interpretation eventually accepted by Italian courts, the uncertainty created by the new norm may open it up to controversies, slowing an enforcement process that was designed to be swift. Moreover, two, uncoordinated registry systems, designed to govern priority, coexist. As a result, the objective of facilitating access to credit by approximating the Italian law for secured transactions to the most advanced

148 In general, when registration substitutes dispossession, rights in rem are validly established only upon registration in the public records indicated by the law. 149 As advocated by Tucci (n 125).

Reverse Engineering the Law

321

international standards is hindered. In this regard, it is worth noting that the UNCITRAL Model Law provides for the establishment of a registry system that is centralised and consolidated.150 Furthermore, the attribution of the role of registrar to the Italian Tax and Revenue Agency represents a singular policy choice, and one that has not been encountered, to the best of our knowledge, in any other legal system. It is acknowledged that the costs associated with the use (registration and search) of a secured transactions registry have to be minimal, in order to ensure that access to secured credit is not distorted by hidden costs or fees. The Legislative Guide explicitly stipulates that registration and search fees should not be used to raise revenue: rather they should be set uniquely to recover the costs of constituting and operating the registry.151 Hence, from a practical perspective, having the ‘tax man’ involved in the process of extending credit to the Italian economic sector goes again, per se, such a general principle. In general terms, this choice could send a misleading signal over the nature of publicity requirements for secured transactions. The fees for registration may be considered as a form of tax, discouraging in turn the recourse to the new instrument. Moreover, unless registry services are offered free of charge,152 such an institutional design may indeed result into a new fi scal levy on registered secured transactions. If this is the case, fees and taxes over secured transactions could significantly deter the utilisation of the registry, thus hindering the success of the new law. Given the existence of a widely used, e lectronic and centralised public registry for entrepreneurs and companies operating in Italy, ie the aforementioned Company Registry Book held by the Chamber of Commerce, the rationale for creating a novel mechanism operated by the Tax and Revenue Agency does not appear to be a sound policy choice. It should be also noted that the new non-possessory pledge, albeit contained in a ‘bank decree’, not only involves only banks, and it allows security to be taken for loans that are also short-term. However, whether regulated credit institutions would adopt this instrument largely depends on whether a given lending operation is deemed less risky under internationally harmonised capital requirements. This form of credit is commonly offered by non-banking institutions, such as factoring and leasing companies, that face higher costs of funding, due to the fact that they are not authorised to take deposits. Typically, non-banking institutions fund their activities by, for instance, issuing bonds, securitising receivables or raising new equity capital; thus, they typically charge a premium that is higher than the interest rate applied by banks.153 These operations often involve loans and 150 See UNCITRAL Model Law, Art 26. Explicit reference to a centralised and consolidated registry system is advanced in the Legislative Guide, ch IV, paras 21–24, in particular, Recommendation 54(e). 151 Legislative Guide, ch IV, para 37, and Recommendation 54(i). A principle reflected also in the Registry Guide, para 274. 152 A possibility contemplated in the UNCITRAL Model Law; see Ch IV, Section G, Art 34, Option B, according to which ‘[t]he Registry may not charge any fee for its services’. 153 See OECD, New Approaches to SME and Entrepreneurship Financing: Broadening the Range of Instruments (Paris, OECD, 2015).

322

 Giuliano G Castellano

orrowers that are deemed too risky for bank loans.154 It follows that by adopting b a more flexible security instruments, without changing the overall body of rules to take security over movables, access to secured credit may be broadened, but the overall costs of credit may not be reduced. Indeed, empirical studies have noted that an increase in interest rates is often connected to collateralised loans.155 Moving from these considerations, it is possible to note that the insertion of a new security instrument into the Italian legal regime for secured transactions may implicitly incentivise the development of the non-banking sector— ironically through a statutory intervention intended to facilitate the banking sector. On the one hand, the ability of Italian credit institutions to offer secured credit is constrained within the provisions established by capital requirements, and banks use Article 46 TUB as a primary tool to secure loans. On the other hand, the new non-possessory pledge equips non-banks with a more agile security instrument that calls into question the priority ladder vis-à-vis possessory and non-possessory security instruments, including bank charges. It follows that the primary beneficiaries of the new norm are not regulated banks. Facilitating the expansion of the non-banking sector necessitates careful considerations over the benefits and dangers related to the phenomenon of ‘shadow banking’. According to the Financial Stability Board, shadow banking is an activity of credit intermediation occurring outside, or partially outside, the regulated banking system, while implying dynamics that, in terms of leverage and maturity transformation, are similar to those of traditional banking activities.156 Following the global financial crisis, shadow banking has become a primary concern for national and international regulators, given that the uncontrolled expansion of credit outside the banking system may ultimately pose relevant concerns for the stability of the financial system as whole.157 It follows that, absent a rational and systematic reorganisation of the existing legal regime, the introduction of a new security instrument may lead to unintended consequences, undermining its primary policy objectives of fostering access to credit and strengthening the stability of the Italian financial system.

154 Asset-based lending has traditionally been considered as the last resort for financially distressed firms, ibid. 155 See SA Davydenko and JR Franks, ‘Do Bankruptcy Codes Matter? A Study of Defaults in France, Germany and the UK’ (2008) 63 Journal of Finance 565; JR Booth and LC Booth, ‘Loan Collateral Decisions and Corporate Borrowing Costs’ (2006) 38 Journal of Money, Credit and Banking 67; and SS Chen, GH Yeo and KW Ho, ‘Further Evidence on the Determinants of Secured Versus Unsecured Loans’ (1998) 25 JBFA 371. 156 Financial Stability Board (FSB), ‘Strengthening Oversight and Regulation of Shadow Banking: Recommendations of the Financial Stability Board’ (Basel, 2011). 157 ibid. On the risk of shadow banking, see SL Schwarcz, ‘Regulating Shadow Banking’ (2012) 31 Review of Banking and Financial Law 619 and S Ghosh, I Gonzalez del Mazo, İ Ötker-Robe, ‘Chasing the Shadows: How Significant Is Shadow Banking in Emerging Markets?’, The World Bank Economic Premise n. 88 (September 2012).

Reverse Engineering the Law

323

C. Reforming Secured Transactions in Italy: E Pluribus Unum From the above, it appears that the main impediments to the development of a secured credit market in Italy are to be attributed to the complexity of the legal regime governing secured transactions. The strategy followed by Italian law reformers in the last decades, rather than offering a comprehensive reform, comprises a series of incremental changes designed to address contingent issues affecting the national economic and financial system. Conflicts among competing claims and different security instruments have not been resolved, leaving room for interpretative ambiguities between general provisions and a plethora of special laws. The concomitant existence of various categories that are governed by different provisions to create, enforce, and register security rights, results in a legal regime for secured transactions that it is not in line with the ethos of the UNCITRAL Model Law. A different strategy is thus required. The reverse engineering method here advanced reveals that the complexity of the Italian regime pertaining to non-possessory security interests may be solved with the implementation of a comprehensive, public, and nationwide registry system. Such a registry system could follow the publicity rules contained in the UNCITRAL’s texts—in particular, in the Model Law and Registry Guide—and may be included within the existing CRB. The implementation of such a reform route would create a series of benefits without incurring the problems related to more drastic reforms. First, a consolidated registry system governing priority for every security instrument would dramatically reduce the complexity created by the existence of multiple and overlapping legal categories. Second, it would offer an effective remedy to the problem of asymmetries of information affecting the creditor-debtor relationship. A single registry system would imply that searches could be conducted effectively without having to search within various public records. Third, once registration is a point of priority following the first-to-file principle, non-possessory secured transactions—notwithstanding their formal categorisation—would offer an effective device to manage credit risk. Secured lenders would be in a position to identify prior entitlements over a collateral and therefore they could assess more accurately the riskiness of their operation. Fourth, the impact of recharacterisation on the priority ladder of a given security instrument would be substantially reduced. The various consensual security instruments (and their corresponding legal categories) would be subjected to the same publicity regime and registration requirements. Retention of title clauses, bank charges and non-possessory pledges could be governed by the same publicity requirements and priority rules, which are determined ex ante. In other words, rather than implementing a new legal category, the existing security instruments, including the newly-established non-possessory pledge, should be subjected to a common publicity regime. As a consequence, the aforementioned conflicts affecting Italian law would be elegantly resolved. To wit, the route presented here could be defined as a unitary approach to

324

 Giuliano G Castellano

publicity that delivers the benefits of a strategy creating a functionally-based, unitary legal category, without incurring the resistance to and costs of such a drastic reform. For this route to be effective, a legal reform of publicity rules should be re-engineered in consideration of the constraints imposed by the legal framework within which it has to operate. In practical terms and with reference to the Italian legal system, this implies that a new secured transactions law implementing a consolidated registry system should dovetail with the core publicity principles enshrined in the Italian Civil Code. This would require the publicity regime for chattel mortgages (ipoteche mobiliari) to be extended to existing and future movable assets that are used by companies to conduct their business operations. To determine which transactions should be registered, law reformers may implement a functional equivalence approach, a listing technique—whereby registrable security instruments are enumerated in the law—or a combination of both under a nuanced form of functionalism. Different procedures for enforcement may be maintained, if justified by specific policy considerations, eg to distinguish security interests from retention of title clause. Nonetheless, certainty over the existence of prior entitlements, their ranking, and enforceability would be ensured. To this end, existing registration mechanisms should be collapsed into one registry system. Given the central role of the Company Registry Book in keeping records of the life cycle of Italian businesses, it is only natural to imagine that a secured transaction registry could be contained therein.

V. Conclusion The reverse engineering method adopted in this chapter represents an attempt to develop an approach to study law reforms. A wide array of analyses in international commercial law are concerned with the definition of an optimal legal regime and the possibility of transposing (or transplanting) such a legal regime into national legal systems. Nonetheless the two intellectual activities seldom communicate in a constructive fashion. The method adopted here attempts to fill this gap. To this aim, it focuses on the national contexts in which legal reforms are desired and whose objectives are sufficiently clear, but where efforts to pursue the traditional reform routes have failed. Hence, a new route must be devised. Through a process of reverse engineering, the main routes of legal reforms have been deconstructed. This allowed the advantages and the difficulties encountered in the implementation of substantive laws to be elicited, moving from the experiences of various legal systems. The knowledge gained through this process has been then used to (re)construct a reform route that overcomes the impediments in implementing new legal rules. The alternative reform strategy that emerged is by no means a one-size-fits-all solution. It is, however, an alternative on the menu of options for law reformers.

Reverse Engineering the Law

325

The alternative reform route is designed to overcome the difficulties encountered with a unitary, functionally-based approach, without relying on disjoined incremental changes. The route consists of implementing directly a publicity regime that is based on UNCITRAL’s prescriptions. The case of Italy offers a practical example of the viability of this route that in practice achieves the same benefits as a more comprehensive reform, without incurring in the same costs. In light of the new security instrument recently adopted in Italy, the implementation of a single registry system appears to be a more effective strategy to broaden access to credit. In fact, notwithstanding the commendable efforts of Italian law reformers, the introduction of a new non-possessory pledge, without amending the existing legal regime, runs the risk of deepening the ambiguities inherent in the Italian legal regime governing secured transactions over movable assets.

326

 

INDEX

absolute priority 180, 224, 233–35 account debtors 107–8 account holders 70–71, 74, 186–88 accounts 8, 10–11, 13, 21–22, 159, 161, 185–87 sales of 10, 22, 161 acquéreurs 138, 145, 153 acquisition 69–70, 102, 113–14, 143, 145–47, 153, 246 financing 16, 23, 34, 50–51, 113, 250–251, 314 security 29, 173 security interests/rights 26, 29, 51, 59, 70 actifs circulants 134, 150–151 administrative costs 22, 79 Africa 160, 165, 170 after-acquired assets 26, 29 agreements control 65, 70–71, 74, 172, 201 licence 98–100, 279 netting 92, 246, 249 written 61, 266, 268 aircraft 64, 67, 193, 249–50, 255, 310 Alberta 197–98 alert function 170–171 aliénation 136–38, 140, 143–44 ambiguities 292–93, 312, 315–16, 319, 323, 325 anti-assignment clauses, see non-assignment clauses applicability 214, 248, 250, 255–56, 277, 288 applicable law 74–76, 91, 191–93, 195–96, 201–6 appropriation 226–27 physical 187 archetypical categorisations 289, 294, 312 aspirations 8, 101, 133, 139, 237, 289, 301 asset reconstruction companies (ARCs) 243 asset-based lenders 103–5, 107–9, 112 asset-based lending 103–4, 112, 117 cross-border, see cross-border asset-based lending asset-based loans 104–5 asset-based registries 313 assets 56–58, 60–64, 121–31, 210–213, 224–27, 260–269, 311–15 after-acquired 26, 29 debtor’s 224, 233

encumbered, see encumbered assets financial 187, 241–43, 246, 250, 258 grantor’s 170–171, 209, 219, 223, 251 immovable 310–311 intangible 96–98, 193, 201, 256, 265, 268, 293 leased 25–26, 29, 31–33, 35–39 movable 96, 121, 124, 126, 288–89, 310–312, 324–25 non-performing 243 productive 210, 229, 234 tangible 74–75, 98, 121, 196, 244, 264–65, 274–75 unencumbered 219, 223, 235 assiette 139–41, 146, 148 assignability, free 80, 86, 91 assignees 79, 81–85, 89–91, 246–47 assignment 11, 74, 77–86, 88–92, 107–8, 193–94, 245–47 effectiveness 78, 83, 85, 89, 92 prohibition on 83–84 of receivables 77–78, 85, 87–90, 92, 102, 240, 244–47 reasons for regulation 86–89 assignors 79, 81–87, 89–92, 246–47 assistance financial 113–14 technical 44 asymmetric information 304–5, 323 attachments 14, 58, 86, 179, 195, 250, 258 auctions, public 221, 254 Australia 2, 105, 109, 111, 179, 281, 290 Austria 230–231 avis 143–45, 147, 149, 152–53 bail-in legislation 205 bailments 19–21, 158 bank accounts 65, 70, 73–75, 90, 112, 206, 244–45 bank charges, see banks, charges registered 316–17 bank lending 38, 78, 87, 92 bank verification numbers, see BVNs banking, shadow 322 bankruptcy 19–20, 90; see also insolvency legislation 18, 20, 223 personal 4, 283

328

Index

banks 13, 240–241, 243–49, 251–54, 256, 318, 321 central 161, 166, 175, 181–82, 241 charges 314–17, 319–20, 322–23 depositary 65, 70–71, 74 bargaining power 33, 84, 86–87, 90–91 bearer securities 187, 189–90 Belgium 102, 105, 230–231 best interests of creditors test 233, 235 biens transformés 139, 147–50 bona fide purchasers 80, 270 bonds 58, 76, 189, 204, 321 book debts 12–13, 22, 78, 80, 240, 293 borrowers 61–62, 103–6, 108–10, 112–16, 252–54, 256–58, 304–5 receivables 104, 107–8, 112 borrowing availability 106, 108–10, 112–13 base 104, 107–10, 112 consolidation 110–112 Brazil 212, 227, 229, 231, 235 brevity 163–65, 181 British Columbia 19–20 buyers 5, 12, 15–18, 21, 97–98, 120–122, 251–52 BVNs (bank verification numbers) 175–76 Canada 19–20, 37–38, 111, 185–91, 194, 196–201, 230–232 cancellation 59, 68, 127, 167–69, 182, 269 capital 78, 92, 114, 203, 288 adequacy 241, 316 care, reasonable 59, 72, 123–25 cars 266, 271 carve-outs for unsecured creditors 109–10 cash 9, 64, 77, 172, 235–36 collateral 172, 235 dominion 112–13 identifiable 142 receipts 235–36 central banks 161, 166, 175, 181–82, 241 centralisation of receivables 106–7 certificated non-intermediated securities 121, 203, 206 certificated securities 71, 188–90, 196–97, 200, 202, 206 certificates 6, 10, 71, 182, 187, 189–91, 196–97 conclusive 6, 10 characterisation 26, 30–32, 38, 188, 205, 293, 295 disputes 31–32, 35, 39 of leases 27, 36, 38 charge, instruments of 6–7, 9–10 chargees 6, 10, 23 charges 6–9, 11–13, 17–18, 22–23, 47, 50, 314–15 company 5, 8–9, 19, 314 fixed 11, 113, 215, 293

floating 4, 6, 9, 11, 109, 224, 248 general enterprise 3 pre-existing 17 chattel mortgages 11, 136, 311, 314, 320, 324 China 215, 223, 230, 232, 237 circulation des stocks 139–41 civil law countries 51, 105, 116, 213, 261 claims preferential 69, 109–10, 303, 305, 310–312, 314–16, 319 ranking of 211, 224, 227–28, 234–35 retention-of-title 106, 108 secured 210, 222–23, 228, 230, 232–33 unsecured 222–23, 228, 255 clawback measures 3–4 co-borrowers 110–111 codification 47–48, 51, 288–89 co-existence, transactional, see transactional co-existence coherence 21, 43, 48, 284 collateral 10–12, 14–15, 17–18, 103–5, 263–66, 274–76, 278–80 cash 172, 235 tangible 265, 275 Colombia 157, 160, 163 commercial palatability 260, 274, 276, 278, 280 commingled property 142 common law jurisdictions 5, 26, 33, 35, 79, 95–97, 197–98 company charges 5, 8–9, 19, 314 company liquidators 19 Company Registry Book, see CRB comparative priority entitlement 22 compensatory damages 91, 319 competing claimants 69–70, 167, 177, 179, 188, 195, 267 competing creditors 21, 23, 206, 317 complexity 231, 237, 287, 294–97, 316, 320, 323 comprehensive reforms 161, 294, 323, 325 compromises 45, 50–51, 276, 284 conclusive certificates 6, 10 conditional sales 13, 18, 20, 23, 50–51, 250, 258 conditional sellers 11, 18–19, 23 confidentiality 66–67, 160, 171 conflict-of-laws rules on security rights 59–60, 74–75, 185–206, 242 concepts and terminology 186–90 creation, effectiveness against third parties and priority 195–200 effectiveness against issuers 200–202 role and scope 190–195 UNCITRAL 202–6 conflits de priorités 145–47 connecting factors 191, 193–94, 196, 201–2, 205, 274, 284 consensual arrangements 289–90, 295, 301–2 consensual security devices 262–63, 271

Index consensual security interests 262, 272–73, 291, 310 consensus 43, 45, 47–48, 50–51, 159, 288, 297 consent 74, 77, 234, 237, 240, 265, 269 consignments 10–11, 14, 20–22 consignors 21–22 consistency 43, 194, 294, 302 constituants 134, 136–40, 142–44, 146, 149, 151, 153–55 constructive delivery 158, 288 constructive notice 9 constructive possession 158, 172 consumer lessees 27, 32–33, 35 consumer protection 32–33, 35, 56–57, 61, 244 consumers 32–33, 39, 56–57, 90, 244 defaulting 33, 35 contracts 77–81, 83–86, 88–93, 96, 240, 246–47, 273–74 freedom of contract, see contractual freedom contractual freedom 1, 4, 23, 77, 82–84, 87, 89 contractual rights 78, 84, 96–97, 99, 200, 205, 210 control 99, 105, 112, 172, 292–93, 303, 311 agreements 65, 70–71, 74, 172, 201 convergence 298–99 conveyances, fraudulent 4 copyright 244, 261, 269–70, 272, 276, 278–80, 283; see also intellectual property corporate groups 104–5, 107, 110–112, 115 costs 68, 102–3, 108, 169–71, 222–25, 321–22, 324–25 administrative 22, 79 lower 68, 104, 167, 181, 312 transaction 82, 93, 193, 279, 305–6, 309 court expenses 311, 315 CRB (Company Registry Book) 313, 317–18, 321, 323–24 créances 134, 141–44, 146, 152 créanciers 134–49, 151, 153–55 créanciers garantis 144, 146–49, 153–54 créanciers vendeurs 145, 147 credit facilities 102, 106, 110–112, 114–17, 204 asset-based 109, 113 syndicated 114–17 cross-border 115–16 credit institutions 303, 311, 316 regulated 267, 294, 314, 321 credit risk 87, 104, 281, 304–5, 309, 323 creditor actions, stay of 217–18, 221, 229–30, 237 creditor remedies 1, 7 creditors 125–29, 207–16, 218–19, 221–25, 227–29, 232–37, 262–63 competing 21, 23, 206, 317 hypothecary 34–35 involuntary 3, 18 judgment 63, 180, 251 lien 18, 199

329

oversecured 3, 221–22 secured, see secured creditors undersecured 221–22 unsecured, see unsecured creditors cross-border asset-based lending 101–17 current issues 104–17 cultures, legal 22, 43, 47, 53, 286–87, 297, 299 damages 59, 62, 79, 91, 232–33 compensatory 91, 319 liquidated 33 punitive 35 débiteurs 134, 136, 141–42, 146, 148 debt, parallel 115–17 debt distress 207, 232 debt securities 76, 200, 202–5, 258 non-intermediated 202, 204 subordinated 203 debtors 14–15, 17–18, 20–23, 79–86, 89–91, 209–11, 246–47 account 107–8 assets 224, 233 business 210, 214, 218, 229–30 insolvent 36, 211, 213, 303 secured 293, 298, 318 sovereign 92 vulnerable 39 debts 1–2, 111, 116, 221–22, 226, 243–44, 256–57 book 12–13, 22, 78, 80, 240, 293 statutory 311, 315 subordinated 204 défaillance 134, 148 defaulting consumers 33, 35 defaulting lessees 35 defences 72, 81, 265 delivery 57, 61, 64, 70, 173, 220, 247 constructive 158, 288 dematerialised securities 187, 190, 245 deposit 13, 245, 247, 256, 269, 321 depositary banks 65, 70–71, 74 dépossession 135–38, 145, 150 description of encumbered assets 67, 179, 316 deterioration 241, 263, 266, 275 diligence, due 79, 82, 108, 171, 276, 279 direct legislative interventions 291–92, 310, 312 discount rate 22, 233 discretion 35, 216, 221, 298 dispossession 27, 133, 288–89, 292, 311, 316, 319 requirement 287, 289, 313 disputes 19, 26, 59, 82, 190–192, 194–95, 206 diversity 4, 31, 193, 302 dividends 203, 215 dominion cash 112–13 springing 112–13

330 Drobnig, Ulrich 41, 267 droit américain 142, 144–46 droit commun 137, 139, 145, 147, 150–151 droit de suite 133, 135, 137–38 due diligence 79, 82, 108, 171, 276, 279 EBRD (European Bank for Reconstruction and Development) 45, 296, 299, 312 economic benefits 117, 161, 181, 303 economic substance 26–30, 33–39, 295, 299 economic value 83, 229, 237, 275 education 182–83 effectiveness 63, 65, 67–68, 90–92, 195–96, 199–206, 307–9 against issuers 200–202 of assignments 78, 83, 85, 89, 92 period of 66, 182 third-party 63–65, 69, 74–75, 172, 194–96, 253, 256 electronic registries 68, 162, 181, 318 employees 13, 109, 255, 315 employers 4, 81–82 encumbered assets 59–63, 72–73, 96–98, 120–131, 209–33, 235–37, 262–68 description 67, 179, 316 exclusion 210, 212 inclusion 212–14 obligation of secured creditors to return 126–28 obligation to preserve 123–26 rights of persons with contractual rights to 97–99 rights to claims supporting 96–97 sale 224–26, 232, 253 secured creditors’ right to use 128–30 tangible 98–99 value 219, 223 endorsement 196, 247, 292 enforceability 57–58, 111–12, 305, 308, 324 enforcement 59–60, 75–76, 200–205, 209–13, 241–43, 253–54, 257–58 actions 210, 212–16, 218, 226, 229, 231 expenses/costs 33, 222 extra-judicial 59, 72, 248, 253–54, 257 in-court 216 in insolvency 207–37 judicial 59, 72 out-of-court 227 powers 217, 303, 306 process/procedures 200, 206, 225, 295, 318–20 rights 27, 32, 36, 39, 217, 221 of security interests/rights 3, 72–74, 90, 237 entrepreneurs 267, 318, 321 equitable interests 9, 16, 288 equity 21, 35, 203–5, 257, 288 securities 76, 202–3, 205

Index errors 68, 161, 176–77 risk of 66, 174, 176 estates 17, 210, 212, 215, 220, 223, 229 European Bank for Reconstruction and Development, see EBRD exemptions 8–9, 219–20, 228, 232 expectations 20, 84, 172, 236 expropriation, unfair 28, 35 extra-judicial enforcement 59, 72, 248, 253–54, 257 powers 248, 253–54, 257 factoring 77–78, 87, 89, 93, 245, 321 facultative 21 non-notification 70 failure to register 18, 258, 295 fairness 3, 39 Faustpfand 288, 291 fees 68–69, 109, 181–82, 314, 321 notary 314, 316 registry 68, 173, 180–182 fiduciary transfers 212, 216–17, 291 financial assets 187, 241–43, 246, 250, 258 financial assistance 113–14 financial assistance laws 102, 113–14, 117 financial institutions 13, 170, 241–43, 245–48, 251–54, 256, 258 regulated 162, 175–76 fixed charges 11, 113, 215, 293 fixtures 10, 21, 173, 179 floating charges 4, 6, 9, 11, 109, 224, 248 formalism, functional, see functional formalism fraudulent conveyances 4 free assignability 80, 86, 91 freedom of contract 1, 4, 23, 77, 82, 87, 263 functional formalism 25–39 functional test 19, 297 functionalism 1–26, 290–291, 295, 297–98, 300–302, 324 methodological 301 nature and limits 10–22 and purchase-money security Interests 17 substance test 10–14, 19 functions 157, 159, 169, 208–10, 215, 301, 303–6 alert 170–171 allocation 165–66 economic 281, 290 gage 136–39, 141, 150–151, 153–55, 263, 288, 299 garantie 134–36, 143, 146–47, 152–53 general enterprise charges 3 Germany 105–6, 108, 111, 116, 225, 227, 230 Ghana 152, 157, 160–161, 221 good faith 16–17, 124, 127, 129, 131, 315, 319–20 grace periods 104, 106, 173, 177

Index grantor identifiers 67, 174–77, 307 grantors 59–63, 71–75, 120–133, 175–77, 192–201, 235–37, 263–69 assets 170–171, 209, 219, 223, 251 categories 239, 256 insolvency 63, 105, 211, 251 location 75–76, 197–98, 203, 206 right to obtain information 131 guidance 42, 49, 58, 164, 173–75, 178–81, 272–73 Guide to enactment 55, 166, 168–69, 174–75, 178, 180–181, 183 harmonisation 39, 45, 57, 72, 76, 194, 197 head offices 193, 197–200 Honduras 157, 160, 163, 176 Hong Kong 105 hypothecary creditors 34–35 hypothecs 34, 56, 189, 297, 310–311, 319 movable 27, 34, 311, 319 identification 139, 141–43, 155, 159, 164, 175, 178 identification des produits 141–43 identifiers 66–67, 175–77 grantor 67, 174–75, 177, 307 numerical 160, 165, 170, 175, 177 IMF, see International Monetary Fund immovable assets 310–311 immovable property 58, 66, 240, 244, 247–48, 252, 288 implementation 174–75, 285, 287, 296–97, 299, 306, 323–25 incentives 35, 181, 225, 227, 308, 322 inclusion of encumbered assets 212–14 inconsistencies 48, 57, 64, 206 indexing 67, 174–75, 177–78 India 239–58 comparative analysis 242–54 historical background 241 modifications required for conformity with UNCITRAL provisions 254–55 individuals 24, 89, 161–62, 175–76, 182, 272 inductive method 285–86 inertia 263, 265, 275 information 66–67, 131, 144, 146, 170–171, 304–6, 308–9 asymmetric 304–5, 323 asymmetries 305, 323 reliable sources 278–79 right to obtain 131 inscription 143–44, 147, 149, 152 insécurité juridique 150–151 insolvency administrators 105, 107, 109, 212, 214–17, 219, 225–28 enforcement in 207–37

331

estates 109, 210–218, 220, 222, 225, 227–28, 237 grantors 63, 105, 211, 251 law 6–7, 208–9, 215, 225, 227, 231, 237 proceedings 190, 209–11, 213, 216, 220, 225, 303 process 210–214, 216–17, 219, 221, 223–26, 228, 237 regimes/systems 207–9, 213–14, 218, 220, 222–25, 228–29, 231 tribunals 212, 214, 216, 220–221, 226–27, 232, 235–36 insolvent debtors 36, 211, 213, 303 inspection 6, 10, 121, 129–30 institutions, financial 13, 170, 241–43, 245–48, 251–54, 256, 258 instruments of charge 6–7, 9–10 intangible assets 96–98, 193, 201, 256, 265, 268, 293 intellectual property 75–76, 99, 249, 259–61, 268–78, 280, 282–84 encumbered IPRs 269, 274–75, 280 rights 67, 99–100, 240, 247, 252, 314 Italy 259–84 registered 269, 271, 314 unregistered 270, 272, 276, 278–80, 283 intellectual property rights, registered 269, 271–73, 277–80, 282–83, 314 intentions 30, 35, 73, 84, 263 interest rates 162, 321–22 interests equitable 9, 16, 288 legal 16–17 security 10–16, 55–58, 60–67, 69–76, 243–45, 247–55, 258–83 intermediated securities 185–90 International Monetary Fund (IMF) 207, 310 international standards 207, 287, 296–99, 312, 321 interpretation 11, 37, 56, 60, 85, 279, 320 judicial 164, 290–294 scholarly 309, 315 interpretative techniques 290, 295, 300–302, 305 inventory 61–63, 67, 104, 107–10, 112, 133, 178–79 new 112, 235 investors 194, 241, 318 invoices 70, 77–78, 89, 107 involuntary creditors 3, 18 ipoteca 262–63, 266–67, 271–72, 276–77, 282, 310–311, 314 IPRs, see intellectual property issuers 63, 65, 71–72, 76, 185–90, 194, 199–206 jurisdiction/law 199–206 Italy critical analysis 274–80 intellectual property rights 259–84

332

Index

legal framework 261–70 ‘privilegio’ and ‘privilegio speciale’ 267–68 reform 280–325 Jamaica 49 Japan 216, 226, 230, 232 judgment creditors 63, 180, 251 judicial enforcement 59, 72 jurisdictions 105, 107–8, 114–16, 177–78, 195–203, 280–282, 302 knowledge 68–69, 71, 91, 98–101, 286–87, 321, 324 Korea 230, 232 labels 13–14, 105 language 6, 10, 21–22, 43, 296 legislative 46–47, 76 leased assets 25–26, 29, 31–33, 35–39 leases characterisation outside secured transactions law 36–38 formal characterisation under Quebec Civil Code 34–36 security, see security leases true, see true leases leasing 23, 30, 38, 193 legal categories 286, 289, 291, 293–94, 296, 299, 323–24 existing 297, 309 new 287, 294, 310, 318–19, 323 overlapping 323 legal cultures 22, 43, 47, 53, 286–87, 297, 299 legal frameworks 160–162, 267, 270, 278–79, 282, 298, 303 legal functionalism, to reverse engineering 300–307 legal interests 16–17 legal persons 203 legal reforms, see reform legal systems 191–93, 217–21, 236–37, 239–40, 286–91, 296–302, 306–9 legal uncertainty 279–80, 282 Legislative Guide on Secured Transactions, UNCITRAL 41–53, 55–63, 65–69, 74–76, 95–98, 239–42, 247–51 legislative interventions 78, 88–89, 93, 280–281, 309, 312 ad hoc 313 direct 291–92, 310, 312 uncoordinated 278 legislative powers 191, 247–48, 250 legislative styles 47–48, 51 legislators 32, 37, 165–66, 174–75, 178, 190, 192 lender-friendly countries 105–6

lenders 103–6, 108–10, 112–16, 194, 204, 248–49, 304–5 asset-based 103–5, 107–9, 112 potential 279, 320 lending, asset-based 103–4, 112, 117 lessees 15, 17, 19–20, 25–26, 29–33, 35–39, 98–99 consumer 27, 32–33, 35 defaulting 35 lessors 15, 19–20, 23, 26–27, 29–33, 35, 37 letters of credit 44, 58, 63, 244–46 lex loci protectionis 273–74, 280, 284 lex rei sitae 193, 273 lex situs 193, 196 liabilities 33, 59, 66, 68, 91, 111, 203 Liberia 160, 182 liberté contractuelle 152, 154–55 licence agreements 98–100, 279 licences 98–100, 153, 243–44, 269, 273, 278–80, 282–83 registered IPRs 273, 279–80 unregistered IPRs 280 licensees 73, 76, 98–100, 276, 278 non-exclusive 98–99 licensors 70, 98–100, 273, 280 lien creditors 18, 199 liens 142–43, 172, 180, 249 floating 171 non-consensual 159, 173, 180 tax 180 limited partnerships 197, 203, 288 liquidated damages 33 liquidation 203, 210–211, 215–18, 220–221, 226, 228–31, 237; see also insolvency insolvent 4, 13 procedures 216, 221 proceedings 215, 217–18, 220, 229 liquidators 5, 8, 20, 221 litigation 163, 165, 174, 176–77, 191, 193, 198–99 loans 12–13, 104–5, 110–115, 218–19, 221–23, 240–241, 253–54 asset-based 104–5 non-performing 243, 253, 310, 318 secured 210, 216, 218, 225, 228, 230, 237 location 26, 28, 74–75, 193–94, 196–98, 200, 205–6 of title 10–11 losses 114, 220, 231, 236, 241, 263, 266 Luxembourg 111, 116 Macdonald, Roderick 39, 41–42, 96–97, 100–101, 183, 259, 298 Malawi 157, 160, 178 Malaysia 213, 216, 222 marchandises 134, 136–39, 142, 147, 149 market value 37, 266 maturity 22, 162, 232, 301, 322

Index medium-sized enterprises 92, 101, 159, 243, 303 mergers 102, 114, 246 methodological functionalism 301 meubles corporels 151, 153–54 Mexico 152, 157, 161, 176, 181, 215–16, 223–24 middle-market companies 103, 117 model laws 42–53, 55–76, 102, 106, 165–66, 174, 179 misgivings 41–53 reasons for preparation 42–43 UNCITRAL 55–76 modern secured transactions regimes 3–17, 68, 102, 107–8 mortgagees 1, 240 mortgages 9, 11, 13, 23, 240, 251, 256 chattel 11, 136, 311, 314, 320, 324 motor vehicles 178, 182, 240, 252, 310 movable assets 96, 121, 124, 126, 288–89, 310–312, 324–25 movable goods/property/movables 240–241, 244–45, 247–53, 255, 264–65, 288–89, 310–311 movable hypothecs 27, 34, 311, 319 multiple security rights 269, 307 mutual rights and obligations 120, 122–23, 132, 274 natural persons 56, 244, 248, 278 negotiable instruments/documents 65, 70–71, 73–74, 96–98, 121, 172–73, 244–47 Netherlands 102, 105–7, 111, 114, 116, 213, 230 netting agreements 92, 246, 249 New Zealand 2, 11, 18, 105, 165, 281, 290 Nigeria 161, 175 non-assignment clauses 77–93, 106–8 definition and reasons for use 79–82 legal effect 82–86 UNCITRAL approach 89–92 non-banking institutions 321 non-consensual liens 159, 173, 180 non-exclusive licensees 98–99 non-intermediated debt securities 202, 204 non-intermediated securities 65, 71–73, 75–76, 185–88, 190–191, 193–95, 200–202 certificated 121, 203, 206 security interests in 58, 65, 71 non-performing assets 243 non-performing loans 243, 253, 310, 318 non-possessory pledges 286–87, 295–96, 306, 310–311, 318–21, 323, 325 non-possessory security interests/rights 26, 195, 198, 240, 288, 310–311, 313 non-unitary approach 23, 51, 291, 299–300 notaries 170, 311–12 notary fees 314, 316

333

notice, constructive 9 notice registration 65–66 numerical identifiers 160, 165, 170, 175, 177 obligations 10–12, 59, 71–72, 119–29, 131–32, 203–5, 262–65 contractual 192, 206 fluctuating 61, 244 performance 7, 10–11, 244, 253, 257, 295, 301 preservation 123, 128 secured 60, 73, 127–29, 188, 201, 262–64, 266–67 obligors 11, 71, 97, 202, 267 secondary 96–97 third-party 71–72, 74 obsolescence 286–87, 300 omissions 31, 165, 173, 178 one-size-fits-all approach 44, 48, 179, 324 Ontario 11, 30–32, 187, 197–99 opérations garanties 134–35, 144 opposabilité 137, 139, 143, 143–45, 149 Organization of American States (OAS) 45–46, 124 oversecured creditors 3, 221–22 ownership 4–5, 14, 18, 25–26, 28–30, 193–94, 211–12 rights 36, 193, 251 transfers 38, 194, 263, 269–70, 316 Pakistan 157, 161 palatability, commercial 260, 274, 276, 278, 280 parallel debt 115–17 pari passu principle 223, 262, 303 parties definition 120–122 secured 10, 14, 79, 131, 142, 223 sources of reciprocal rights and obligations 122–23 partnerships 24, 197, 204 limited 197, 203, 288 party autonomy 27, 59, 120, 123, 133, 242–43, 271 limitations to 276, 283 patents 64, 66–67, 244, 261, 268–69, 272, 276–78; see also intellectual property path-dependency 300, 309–10 payment 33–35, 73–74, 82–86, 88–89, 111–12, 114–16, 244–47 monetary 220, 232 obligations 92, 204–5 rights to 228, 244–45, 249 pegno 262–65, 268, 270–273, 276–79, 310–311, 316–17, 319 perfection 6–7, 18, 265–66, 272, 274, 277–78, 307–8 performance 7, 10–11, 84–85, 96–97, 253, 257, 295

334

Index

personal bankruptcy 4, 283 personal property 1, 10, 21, 28, 288, 290, 295 Personal Property Security Acts, see PPSAs personal property security law 260, 276, 281, 284 personal rights 96, 187 physical appropriation 187 physical degradation 275 physical possession 158, 187, 315 pledges 11, 13, 158, 245, 309–10, 315–16, 319–20 non-possessory 286–87, 295–96, 306, 310–311, 318–21, 323, 325 possessory 310–311, 315–16, 319 subsequent 315–16, 319 policy choices 45, 166, 169–70, 173, 178–79 policymakers 162–63, 166, 170–174, 179–80, 289 possession constructive 158, 172 delivery of 292, 310, 320 physical 158, 187, 315 secured creditors in 123, 126, 128–29, 196, 265 transfer of 264–65, 288 possessory pledges 310–311, 315–16, 319 post-commencement finance 235–37 powers 14, 159, 161, 242–43, 252, 255, 257 enforcement 217, 303, 306 extra-judicial enforcement 248, 253–54, 257 legislative 191, 247–48, 250 PPSAs (Personal Property Security Acts) 25–30, 32–35, 37–39, 179, 186, 188–89, 195–96 III.extension to long-term true leases 29–33 predictability 13, 39, 80, 183, 193, 231, 242 preferential claims 69, 109–10, 303, 305, 310–312, 314–16, 319 prices 17, 30, 34, 89, 225–27, 254 purchase 28, 30, 50, 64, 108 priorité 139, 142, 145–47, 153 priority 69–71, 194–96, 202–5, 222–24, 252–53, 267–69, 316–20 absolute 180, 224, 234–35 conflicts 14, 17, 69, 71 ladder 293, 295, 305, 312, 315, 322–23 registration as partial determinant 171–73 rules 1, 4, 7, 9–10, 14, 16, 179–80 absolute 233, 235 of security interests 69–71 special 26, 29, 34, 50, 57, 178 status 195, 316–17, 319 private companies 102, 114, 203 privilegio 262–63, 267, 310–312, 314, 319 speciale 267–68, 271–73, 276, 278–79 privity 77–78 procedural safeguards 255, 257 productive assets 210, 229, 234

produits finis 147–49, 153–54 property 20, 82–83, 95–97, 99–100, 240–241, 244–45, 252 commingled 142 definition 244, 247 encumbered 98, 313 immovable 58, 66, 240, 244, 247–48, 252, 288 intangible 193, 256 intellectual, see intellectual property law 14, 16, 119, 253, 317 personal 1, 10, 21, 28, 288, 290, 295 rights 26–28, 96–98, 129–30, 187–88, 244–46, 269–70, 273–74 to claims supporting encumbered assets 96–97 of licensees to intellectual property 99–100 of persons with contractual rights to encumbered assets 97–99 tangible, see tangible assets treatment of concepts by UNCITRAL 95–100 protective measures 89, 219–20, 231 prudential norms 241–42 public auctions 221, 254 publicity 26–28, 34, 289, 310, 316, 319, 324 regimes 308, 311, 314, 320, 323–25 requirements 29, 34, 295, 306, 320–321, 323 rules 287, 300, 302, 305–7, 323–24 punitive damages 35 purchase price 28, 30, 50, 64, 108 purchase-money security interests 7–8, 15–17, 22, 29, 146 purchase-money security Interests, and functionalism 17 Quebec 25–26, 35, 39, 187, 194, 197–98, 201 Quistclose trusts 2 ranking 192, 210–211, 221, 223–24, 227–28, 234, 324 reasonable care 59, 72, 123–25 receivables 21–22, 77–78, 85–90, 92, 104, 106–9, 244–48, 271–74 assignability 84, 86, 92 borrowers 104, 107–8, 112 financing 58, 77, 105, 107–8 sovereign 90 transfer 58, 121–22, 194 recharacterisation 15–17, 293, 295, 305, 323 risk 293, 305 recovery 1, 21, 213–14, 219, 241, 243, 258 economic 102 redemption 34, 226, 258 re-engineering, reverse engineering to 307–24

Index reform 6–8, 157–83, 255–58, 280–283, 286–89, 297–98, 323–24 comprehensive 161, 294, 323, 325 comprehensive or piecemeal 161–63 Italy 280–325 legal functionalism to reverse engineering 300–307 processes 46, 164, 174, 182–83, 287, 290, 297 reverse engineering to re-engineering 307–24 routes 285, 287, 287–301, 307, 316, 323–25 régime spécial 150–151 régime unique de sûreté indifférenciée 152–55 registered bank charges 316–17 registered holders 187–88, 200–201 registered intellectual property rights 269, 271–73, 277–80, 282–83, 314 registered offices 197–98, 200 registrable particulars 9–10 registrants 68, 159, 165, 167–68, 176–77, 182 registrars 6–7, 10, 166, 170, 311, 313, 321 registration 6–9, 64–71, 157–83, 266–70, 305–8, 313–17, 319–21 documents 66, 160 notice 65–66 as partial determinant of priority 171–73 process 66–67, 166, 176, 308 requirements 22, 34, 108, 256, 323 statutory rules 166–69 systems 66, 159, 248, 252, 255–56, 271, 278–79 registries 65–66, 68–69, 159–63, 165–74, 176–83, 308–9, 314 asset-based 313 design 166, 170, 173 electronic 68, 162, 181, 318 fees 68, 173, 180–182 functions 169–73 key features 159–60 special 252, 313–14 Registry Guide, UNCITRAL 55–56, 65, 67–68, 76, 174, 178, 182 registry records 169, 175, 182, 252 registry systems 65–69, 170, 176, 251, 306–7, 320–321, 323–24 consolidated 323–24 single 308–9, 323, 325 règle proportionnelle 148–49 règles dérogatoires 152–55 regulated credit institutions 267, 294, 314, 321 regulated financial institutions 162, 175–76 rehabilitation 208–9, 218–19, 229–31, 233 plans 223, 230–235, 237 procedures 217–18, 229–30, 232–37 proceedings 209, 218, 220, 229–31, 234–37 reinstatement 32, 168 reliability 105, 174, 276, 308 relief 217–20, 231–32, 253, 257

335

relitigation 45, 47, 49–51, 53 remedies, creditor 1, 7 remplacement 138–42, 145–46, 152 remploi 140, 143–45, 155 renvoi 151, 193 reorganisation 36–37, 210, 299, 310, 322 repayment 13, 22, 226–28, 245, 249–50, 254, 264 replacement 235–36 repo 2, 22 repossession 35–37, 73, 241, 254 resale 7, 20–21, 32–33 reservation of ownership 26, 29–30, 34 of title 7–9, 11, 16–18, 21, 23, 212, 216–17 reserves 108–9, 113, 129 residual value 28, 33, 64, 203, 236 resilience 286–87, 297, 299 restitution 126–27 retention of title 193, 251, 297, 299, 313, 317 agreements 256 clauses 289, 295, 302, 306, 313–14, 316–17, 323–24 sales 56–57, 64, 250 transactions 50–51 retirement benefits 57–58 revenues 68, 128–29, 257, 264, 321 reverse engineering legal functionalism to 300–307 to re-engineering 307–24 rights in personam 210–211, 265–66 risk 26, 35, 45, 47–49, 231–33, 292–93, 304–5 of error 66, 174, 176 mitigation 304–6 of recharacterisation 293, 305 of relitigation 53 Roman law 158, 261, 263, 266, 288 Russia 213, 216, 223, 237 safeguards 10, 72–73, 215, 218, 228–29 procedural 255, 257 saisie 143 sales of accounts 10, 22, 161 conditional 13, 18, 20, 23, 50–51, 250, 258 retention of title 56–57, 64, 250 Saskatchewan 10, 19, 79 searchers 66, 164, 167–68, 170–171, 173, 176–77, 307 searches 167, 170–171, 182, 307–9, 313–14, 320–321, 323 searching 67, 159–60, 172, 175, 178, 181, 251 secured claims 209–12, 218–19, 221–24, 226–28, 230, 232–33, 237 secured credit 207–9, 222–24, 227–29, 237, 242, 302–5, 321–22 cost 167, 314 enforcement 209, 214

336

Index

secured creditors 61–74, 97–100, 125–32, 166–68, 208–37, 252–57, 263–67 authorisation 68, 167–68 competing 5, 8 consent 234, 237 definition 248, 258 obligation to return encumbered assets 126–28 in possession 123, 126, 128–29, 196, 265 prospective 164, 179, 251 right to use encumbered assets 128–30 rights 211, 213, 215, 217, 223, 229–30, 235 secured debtors 293, 298, 318 secured loans 210, 216, 218, 225, 228, 230, 237 secured obligations 60, 73, 127–29, 188, 201, 262–64, 266–67 secured transactions 41–42, 55–57, 67–69, 284–90, 300–307, 309–10, 320–323 reform 41–43, 101, 103, 133, 162 secured transactions regimes, modern 3–17, 68, 102, 107–8 securities accounts 58, 186–87, 189 certificated 71, 188–90, 196–97, 200, 202, 206 dematerialised 187, 190, 245 intermediated 185–90 non-intermediated 65, 71–73, 75–76, 185–88, 190–191, 193–95, 200–202 uncertificated 71, 188, 190, 199, 201–2, 206 securitisation 93, 241–43, 321 security agreements 59–61, 71–73, 98–100, 119–32, 201–2, 263–65, 293–95 security devices 193, 262, 264–65, 267, 272, 291–92, 300 consensual 262–63, 271 non-consensual 267 security instruments 291, 300, 305–6, 311–12, 314, 319, 323 non-consensual 262 non-possessory 311, 318, 322 security interests 10–16, 55–58, 60–67, 69–76, 243–45, 247–55, 258–83 acquisition 26, 29, 51, 59, 70 applicable law 74–76 consensual 262, 272–73, 291, 310 creation under UNCITRAL Model Law 60–63 definition 248, 258, 294, 298 enforcement 3, 72–74, 90, 237 non-consensual 260 non-possessory 26, 195, 198, 240, 288, 310–311, 313 perfected 14–15 possessory 245 priority 69–71

purchase-money 7–8, 15–17, 22, 29, 146 third-party effectiveness 63–65 unperfected 15, 18 security leases 19–20, 26–27, 30–33, 35–39 and true leases 26–27, 29–31, 36, 38–39 security purposes 35, 56, 58, 64, 193–94, 242, 244 security rights 50–51, 103–6, 114–17, 188–94, 213–16, 224–26, 244 conflict-of-laws rules on 59–60, 74–75, 185, 187, 189–97, 199–206, 242 enforcement 3, 90, 237 junior 235 multiple 269, 307 sellers 5, 15–16, 18, 21–23, 29–30, 34, 250–251 conditional 11, 18–19, 23 separation actions 212, 216, 220 serial-numbered goods 67, 164, 173, 177–79 set-off 13, 70–71, 81, 91 shadow banking 322 shares 58, 102, 104, 113–14, 189, 203, 316–17 ships 64, 67, 240, 249, 255, 266, 271 simplicity 123, 181, 193, 280, 295, 318 Singapore 1, 102, 105, 114, 213, 222, 230 single registry system 308–9, 323, 325 small businesses 77–78, 80, 84, 87, 89, 92–93, 318 software 63, 169, 183, 270 solde intermédiaire 142–43 sovereign debtors 92 sovereign receivables 90 Spain 207, 213, 217, 219, 222–23, 229, 231–32 special priority 26, 29, 34, 50, 57, 178 special purpose vehicles 241, 247, 283 special registries 252, 313–14 springing dominion 112–13 stability 322 social 101 stakeholders 161–62, 209, 235 standardisation process 296 top-down 1 standards 38, 59–60, 101, 129 capital adequacy 316 international 207, 287, 296–99, 312, 321 statutory debts 311, 315 stay of creditor actions 217–18, 221, 229–30, 237 of enforcement actions 216, 226 limited 217 period 230–231 unlimited 230 stocks 133–55 circulation des 139–41 de marchandises 138–39 sûretés sur 135, 137, 140, 150–152, 154 sub-buyers 12, 20, 313, 315, 317

Index subordinated debt 203–4 subrogation 139–40 subsequent pledges 315–16, 319 subsidiaries 103–4, 106–7 substance test economic 27 functionalism 10–14, 19 substantive law 159, 163–66, 177, 182, 191–95, 201, 324 applicable 191, 193, 195, 201 substitute assets 219 substitute collateral 232 superpriorité 146–47 super-priority 8, 226 suppliers 79, 87, 89, 106, 108–9 sûretés mobilières 133–55 sûretés sur stocks 135, 137, 140, 150–152, 154 Sweden 109, 217 Switzerland 105, 217 syndicated cross-border credit facilities 115–16 syndication 114–17 tangible assets 74–75, 98, 121, 196, 244, 264–65, 274–75 tangible encumbered assets 98–99 tax 13, 37, 57, 114, 171, 176, 320–321 authorities 13, 37, 229, 258 liens 180 technical assistance 44 technical proficiency 52 territorial units 191, 204 Thailand 215, 221, 230–231 third parties 69–71, 83–85, 194–96, 199–202, 205–6, 251–53, 263–69 and priority 69, 74–76, 180, 185, 194–96, 199, 253 third party lenders 236 third-party effectiveness 63–65, 69, 74–75, 172, 194–96, 253, 256 third-party obligors 71–72, 74 tiers 136–39, 141, 143–47, 149 time of commencement 75, 210, 235 title 4–5, 9–11, 16–17, 19–21, 252, 255, 314 chain of 278–80 location 10–11 reservation of, see reservation, of title retention of, see retention of title transfer of 17, 19, 50, 56, 244, 252, 260 unencumbered 17 tort claimants 3–4 tortious liability 91 trademarks 67, 261, 269, 272, 276–78, 292; see also intellectual property transaction costs 82, 93, 193, 279, 305–6, 309 transactional co-existence 1–23 models 22–36 transaction-filing approach 308, 311

337

transfer fiduciary 212, 216–17, 291 of possession 264–65, 288 of receivables 58, 121–22, 194 of title 17, 19, 50, 56, 244, 252, 260 transferees 62, 69–71, 73–74, 76, 98–99, 120–122, 189–90 transferors 69, 120–122 transparency 4, 30, 38, 63, 68, 242, 283 true leases and security leases 26–27, 29–31, 36, 38–39 trustees-in-bankruptcy 5, 15 and vesting 18–20 trusts 11–13, 34, 83, 116, 197, 247, 252 Quistclose 2 uncertainty 10, 12, 125, 276–77, 279, 292–93, 304–5 legal 279–80, 282 UNCITRAL (United Nations Commission on International Trade Law) Legislative Guide on Insolvency Law 208, 215 Legislative Guide on Secured Transactions 41–53, 55–63, 65–69, 74–76, 95–98, 239–42, 247–51 Model Law creation of security interests 60–63 enforcement of security interests 3, 72–74, 90, 237 implementation of registry system 307–24 law applicable to security interests 74–76 objectives and fundamental policies 56 omissions and gaps 173–80 priority of security interests 69–71 registry system 65–69 rights and obligations of parties 119–32 rights and obligations of parties and third-party obligors 71–72 scope and general provisions 56–60 third-party effectiveness of security interests 63–65 Registry Guide 55–56, 65, 67–68, 76, 174, 178, 182 undersecured creditors 221–22 unencumbered assets 219, 223, 235 unintended consequences 8, 322 unitary approach 21, 23, 51, 290–291, 294–95, 297–98, 300 United Kingdom 12–13, 16–18, 78, 84–85, 108–9, 111–14, 288–89 United States 103–7, 110–112, 115–17, 185–88, 190–191, 195–200, 232–34; see also Table of Legislation unregistered IPRs 270, 272, 276, 278–80, 283

338 unsecured claims 222–23, 228, 255 unsecured creditors 4–5, 18, 57, 109–10, 211, 224, 303–5 carve-outs for 109–10 validity 86, 195, 252, 308 valuation 216, 223, 227, 254 value 3–4, 213–16, 218–23, 225, 232–33, 265–69, 312–14 economic 83, 229, 237, 275 market 37, 266 residual 28, 33, 64, 203, 236

Index vesting and trustees-in-bankruptcy 18–20 vulnerable creditors 3 vulnerable debtors 39 wages 89, 109, 223–24 warehouse receipts 240, 245–47 withholding taxes 114–15 Working Groups 47–48, 50–52, 55, 96–100, 157, 252, 259 World Bank 49, 160, 162, 252, 310 Zambia 157, 161