Contents of Commercial Contracts: Terms Affecting Freedoms 9781509930494, 9781509930524, 9781509930517

Citation preview

CONTENTS OF COMMERCIAL CONTRACTS Freedom of contract is a great strength of English law: it is a key reason why English law is often the law of choice. But the terms of commercial contracts often restrict freedom of action. In this book, leading commentators consider those terms by taking stock of recent developments, such as increased reliance on good faith and the rise of smart contracts. In so doing, they make original contributions to ongoing debates concerning the limits to parties’ freedom of contract. This important subject will interest drafters of commercial contracts keen to ensure that contracts are clear and enforceable; litigators disputing the meaning, scope and validity of terms; and academics interested in the purpose and nature of the exercises involved. Volume 35 in the series Hart Studies in Private Law

ii

Contents of Commercial Contracts Terms Affecting Freedoms

Edited by

Paul S. Davies and

Magda Raczynska

HART PUBLISHING Bloomsbury Publishing Plc Kemp House, Chawley Park, Cumnor Hill, Oxford, OX2 9PH, UK 1385 Broadway, New York, NY 10018, USA HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2020 Copyright © The editors and contributors severally 2020 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/, 1998–2020. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication data Names: Contents of Commercial Contracts: Terms Affecting Freedoms (2019 : University College, London, Faculty of Laws)  |  Davies, Paul S. (Law teacher), editor.  |  Raczynska, Magda (Law teacher), editor. Title: Contents of commercial contracts : terms affecting freedoms / edited by Paul S Davies and Magda Raczynska. Description: Oxford ; New York : Hart, 2020.  |  Series: Hart studies in private law ; volume 35  |  “The chapters in this book follow a conference held at UCL on 9 and 10 May 2019.”—ECIP acknowledgements.  |  Includes bibliographical references and index. Identifiers: LCCN 2020011136 (print)  |  LCCN 2020011137 (ebook)  |  ISBN 9781509930494 (hardcover)  |  ISBN 9781509930500 (Epub) Subjects: LCSH: Commercial law—England—Congresses.  |  Contracts—England—Congresses. Classification: LCC KD1629.A2 C66 2020 (print)  |  LCC KD1629.A2 (ebook)  |  DDC 346.4207—dc23 LC record available at https://lccn.loc.gov/2020011136 LC ebook record available at https://lccn.loc.gov/2020011137 ISBN: HB: 978-1-50993-049-4 ePDF: 978-1-50993-051-7 ePub: 978-1-50993-050-0 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.

FOREWORD LADY ARDEN Private ordering reordered

This book contains valuable studies on many different aspects of freedom of contract, and their effect on commercial contracts. If we step back in time, we find that the traditional justifications for freedom of contract are that it facilitates the private ordering of arrangements by the parties, encourages entrepreneurialism and enables the parties to allocate the risk of loss in an agreed fashion. Parties should allocate the risk of loss themselves, for good or ill, and state and judicial intervention should be kept to the minimum. Lord Diplock in Federal Commerce and Navigation Co Ltd v Tradax Export S A, The Maratha Envoy [1978] 1 AC 1 at pages 8-9, turned a deaf ear to pleas that, where things have gone wrong with the parties’ allocation of risk, the courts should interfere to produce a fairer result: It is an important function of a court … to provide parties to a contract with legal certainty at the negotiation stage as to what it is that they are agreeing to. And if there is that certainty, then when occasion arises for a court to enforce the contract or to award damages for its breach, the fact that the members of the court themselves may think that one of the parties was unwise in agreeing to assume a particular misfortune risk or unlucky in its proving more expensive to him than he expected, has nothing to do with the merits of the case or with enabling justice to be done. The only merits of the case are that parties who have bargained on equal terms in a free market should stick to their agreements.

Precisely a century ago, in 1920, Professor Anson published the fifteenth edition of his well-known work, Law of Contract. There was little discussion of freedom of contract. It was just assumed to be the desideratum and effect of contract law. It was part of the warp and weft of the law. But things have changed over the last century. This book demonstrates many of the reasons for this. One of them to my mind was Lord Denning’s pushing of the envelope in so many ways but notably in the unsuccessful development of the concept of inequality of bargaining power in the case of Old Herbert Bundy of Yew Tree Farm (Lloyds Bank v Bundy [1975] 1 QB 326, cf National Westminster Bank v Morgan [1985] 1 AC 686). But the winds of change were blowing. There was growing dissatisfaction with the way law treated consumers leading to the intervention of Parliament in

vi  Foreword the Unfair Contract Terms Act 1977, and other legislation. There was growing dissatisfaction with the small print which contracts often contain; this led Lord Denning, once again, to suggest a ‘red hand’ rule that notice of some exemption clauses would only be provided if they were printed in red or with a red hand pointing towards them (J Spurling Ltd v Bradshaw [1956] 1 WLR 461). Many other examples could be given. In addition, there developed a different way of looking at the rationale for the rules of contract law. Rules of contract law were there not simply to give effect in law, where possible, to what the parties had agreed. They also had to be fashioned to give effect to the reasonable expectations of honest persons. So, if people in commerce would not expect the law to produce a particular result, the development of the law to produce results which more nearly approximated to what they would anticipate could be justified as being their reasonable expectation (see, for example, Transfield Shipping lnc of Panama v Mercator Shipping Inc of Monrovia, The Achilleas [2009] AC 61). The winds of change have blown on commercial contracts law as well as consumer law. The activity has taken several forms. One form is academic study, which raises important questions about coherence or principle. Another is judicial intervention – leading to academic acclamation or criticism. Yet another is statute law (such as the Insurance Act 2015, discussed in this book). A further form of activity is the painstaking and invaluable work of the Law Commissions, which leads either directly or indirectly to change, and there are several examples in this book. A special feature of the Law Commissions’ work is that the Law Commissions look at a given subject as a whole and with a particular emphasis on its practical implications. The Law Commissions consult widely with users of the law, and others who are concerned about the way the law under review works. This book is one of the first dedicated specifically to freedom of contract in this jurisdiction. Apart from PS Atiyah’s authoritative study, The Rise and Fall of Freedom of Contract, written largely from a historical perspective and published as long ago as 1985, there are few other monographs on this subject in the UK. So, this work fills that gap and brings many aspects of contract law up to date. It contains detailed studies of developments in many areas, such as (to name but a few) contract law’s attitude to no oral modification clauses (or NOMs), where the law has recently been clarified by the Supreme Court, the area of penalty clauses where recent developments in the law, as described in this book, mean that to be enforceable penalty clauses must now not be disproportionate, and the increasingly discussed, but highly contentious, area of good faith obligations and their role in the negotiation, interpretation and performance of contractual obligations. The new debate about good faith can be seen as a response to the strictness of the rules of contract law. The book notes the scope for different results to be achieved by negotiation and drafting. The book maps other changes, in particular the growing number of ‘smart contracts’, that is, contracts made between pre-programmed computers without

Foreword   vii any form of human intervention. This development raises an acute issue as to whether the rules of contract law should apply to these contracts in the same way that they apply to contracts made in the conventional manner, whatever the outcome, or whether the rules should be adapted in order to produce comparable outcomes to those which are achieved in conventional cases. The recent decision of the Court of Appeal of Singapore in Quoine Pty Ltd v B2C2 Ltd [2020] SGCA (1) 02 illustrates the challenges in this area. The remarkable thing is that, when contract law is placed under the microscope, it is found not to be a lifeless pre-set series of examples of freedom of contract, and limitations on that freedom, but a writhing mass of activity. I heartily congratulate the many contributors on their different chapters and the editors for their skill in the process of compilation. It is not necessary to detail the chapters as they are admirably summarised in Chapter One, with a response to them in the final Chapter, and the fact that I have not mentioned them all in this Foreword is no reflection whatever of the importance which I attach to them. The book was preceded by a conference at which many of these papers were presented for the first time, and a lively debate took place at that event. Inspired by the level of discussion, the paper-writers have skilfully honed and polished their papers and made them even more relevant. The book is a cutting-edge collection of contributions from leading experts in contract law: readers will find in them much valuable material and many new and stimulating ideas. The Rt Hon Lady Arden DBE Justice of the Supreme Court of the United Kingdom March 2020

viii

ACKNOWLEDGEMENTS The chapters in this book follow a conference held at UCL on 9 and 10 May 2019. The conference was generously supported by the UCL Faculty of Laws ‘Back to Bentham’ Fund, to celebrate our return to Bentham House. We are also very grateful for the financial support and general enthusiasm for the project from Norton Rose Fulbright LLP, and especially Richard Calnan and Tara Pichardo-Angadi. At the conference, each paper was commented upon by a practitioner. For their thoughtful comments, we would like to thank: Sushma Ananda, Sir Jack Beatson, Robin Brooks, Sarah Crowther QC, Michael Godden, Alexander Gunning QC, Marie Louise Kinsler QC, Peter MacDonald Eggers QC, Sandy Phipps, Tara Pichardo-Angadi, Georgia Quenby, Graham Penn, Adam Sanitt, Andrew Sheftel, Catrina Smith, Joanna Smith QC, Sudhanshu Swaroop QC, Aaron Taylor and Philip Wood QC (Hon). Each session was chaired by a member of the judiciary, and we are very grateful to: Lady Arden, Sir Jack Beatson, Dame Sara Cockerill, Lord Leggatt, Sir Bernard Rix and Lord Sales. Lady Arden was especially generous in chairing sessions on both days of the conference, and we are delighted that she has written the Foreword to the book. The conference was very well-attended by both practitioners and academics, and the papers benefited greatly from the lively discussion that each session generated. We would also like to thank Jacqueline Cook for offering to write an epilogue to the book from a practitioner’s perspective; Abe Chauhan and Martin Fischer for excellent research assistance in preparing the manuscript for publication; Catherine Minahan for efficient copyediting; and Sinead Moloney, Tom Adams and all the team at Hart Publishing for supporting the project so strongly from start to finish.

x

CONTENTS Foreword by Lady Arden����������������������������������������������������������������������������������������������v Acknowledgements������������������������������������������������������������������������������������������������������ ix List of Contributors��������������������������������������������������������������������������������������������������� xiii 1. Freedom of Contract and Terms Affecting Freedoms�������������������������������������������1 Magda Raczynska and Paul S Davies 2. Binding Our Future Selves�����������������������������������������������������������������������������������13 Robert Stevens 3. Anti-Oral Variation Clauses: Rock-Solid or Rocky?�������������������������������������������35 Andrew Burrows 4. Controlling Contractual Interpretation���������������������������������������������������������������51 Richard Calnan 5. Good Faiths and Contract Terms������������������������������������������������������������������������65 Magda Raczynska 6. Excluding Good Faith and Restricting Discretion����������������������������������������������89 Paul S Davies 7. The New Override of Bans on Assignment of Receivables��������������������������������113 Hugh Beale 8. The Boundaries of a Borrower’s Freedom to Act: Negative Covenants in Loan Agreements�������������������������������������������������������������������������������������������139 Louise Gullifer and Graham Penn 9. ‘Ethical Clauses’ in Global Value Chain Contracts: Exploring the Limits of Freedom of Contract���������������������������������������������������������������������163 Lucinda Miller 10. Smart Contracts�������������������������������������������������������������������������������������������������191 Sarah Green and Adam Sanitt 11. Disproportionate Penalties in Commercial Contracts��������������������������������������211 William Day 12. Opting for ‘Documentary Fundamentalism’: Respecting Party Choice for Entire Agreement and Non-Reliance Clauses���������������������������������������������239 Jonathan Morgan

xii  Contents 13. Planning for Failure: Contract Design, Ineffective Bargains and Restitution���������������������������������������������������������������������������������������������������267 Niamh Connolly 14. ‘All Watched Over by Machines of Loving Grace’? The Inevitable Conflict between Contract Law and Free Speech in Cyberspace����������������������293 Nicholas J McBride 15. Private Law and Public Concerns: Non-Disclosure Agreements in English Contract Law������������������������������������������������������������������������������������315 Catharine MacMillan 16. Professional Ethics and NDAs: Contracts as Lies and Abuse?�������������������������339 Richard Moorhead 17. Choice of Court and Choice of Law Agreements: Freedom of Freedom of Contract����������������������������������������������������������������������������������������������������������363 Alex Mills 18. Illegality in English Arbitration Law after Patel v Mirza���������������������������������381 Uglješa Grušić and Manuel Penades Fons 19. The Reform of Insurance Warranties: Looking Beyond the Past����������������������403 John Lowry and Rod Edmunds 20. The Right to Delivery of Goods under Contracts of Carriage��������������������������421 Melis Özdel 21. The Contents of Commercial Contracts: Terms Affecting Freedoms – A Response��������������������������������������������������������������������������������������441 Jacqueline Cook Index��������������������������������������������������������������������������������������������������������������������������453

LIST OF CONTRIBUTORS Hugh Beale is Professor Emeritus of the University of Warwick, a Senior Research Fellow at Harris Manchester College, and a Visiting Professor at the University of Oxford. Lord Burrows FBA, DCL is a Justice of the Supreme Court of the United Kingdom. Richard Calnan is a Partner at Norton Rose Fulbright LLP and a Visiting Professor at University College London. Niamh Connolly is a Lecturer in Law at University College London. Jacqueline Cook is a Senior Professional Support Lawyer, Finance, at Stephenson Harwood LLP. Paul S Davies is Professor of Commercial Law at University College London. William Day is a Barrister at 3 Verulam Buildings, London, and Morgan Fellow in Law at Downing College, Cambridge. Rod Edmunds is a Senior Lecturer in Law at Queen Mary University of London. Sarah Green is the Law Commissioner for Commercial and Common Law for England and Wales. Uglješa Grušić is an Associate Professor in Law at University College London. Louise Gullifer is the Rouse Ball Professor of English Law and a Fellow of Gonville and Caius College, University of Cambridge. John Lowry is an Emeritus Professor of Commercial Law at University College London and Visiting Professor of Commercial Law, HKU. Catharine MacMillan is Professor of Private Law at the Dickson Poon School of Law, King’s College London. Nicholas J McBride is a Fellow of Pembroke College, Cambridge. Lucinda Miller is an Associate Professor in Law at University College London. Alex Mills is Professor of Public and Private International Law at University College London. Richard Moorhead is a Professor of Law and Head of Exeter Law School at the University of Exeter.

xiv  List of Contributors Jonathan Morgan is a Reader in English Law and Fellow of Corpus Christi College, University of Cambridge. Melis Özdel is a Lecturer in Law at University College London and Director of the UCL Centre for Commercial Law. Manuel Penades Fons is a Lecturer in International Commercial Law at King’s College London. Graham Penn is Professor of International Finance Law at University College London and Co-Head of Sidley Austin LLP’s Global Finance practice. Magda Raczynska is an Associate Professor in Law at University College London. Adam Sanitt is Head of Disputes Knowledge, Innovation and Business Support, Europe, Middle East and Asia, at Norton Rose Fulbright LLP. Robert Stevens is the Herbert Smith Freehills Professor of Private Law at the University of Oxford.

1 Freedom of Contract and Terms Affecting Freedoms MAGDA RACZYNSKA AND PAUL S DAVIES

Freedom of contract is a principle of paramount importance. It underpins the existence of all kinds of commercial activity, ranging from more traditional commercial practice areas such as trade, carriage and insurance, through to finance and technology. Freedom of contract embodies the idea that the content of the parties’ contractual obligations is a matter of choice for the parties, rather than imposed by the law,1 and that the obligations voluntarily undertaken by the parties are enforced by courts.2 The principle is probably most famously expressed in the following words of Sir George Jessel MR in Printing and Numerical Registering Co v Sampson: It must not be forgotten that you are not to extend arbitrarily those rules which say that a given contract is void as being against public policy, because if there is one thing which more than another public policy requires it is that men of full age and competent understanding shall have the utmost liberty of contracting, and that their contracts when entered into freely and voluntarily shall be held sacred and shall be enforced by courts of justice. Therefore, you have this paramount public policy to consider – that you are not lightly to interfere with this freedom of contract.3 1 S Smith, Contract Theory (Clarendon Law Series, OUP 1993) 59. It is useful to distinguish, as Smith does, freedom of contract from the freedom of individuals to enter into contracts (labelled as ‘freedom to contract’), although some authors run both under the same heading of freedom of contract: see eg H Collins, The Law of Contract, 4th edn (CUP 2008) 25. 2 This second limb is sometimes expressed as the principle of sanctity of contract. Sanctity of contract is, logically, separate from the principle of freedom of contract when freedom of contract is understood as the freedom to choose the terms of the contract, but nothing turns on this distinction here. For practical purposes, in this chapter, freedom of contract includes the sanctity principle, because there is little point in freedom to choose terms unless those terms are enforceable in the way parties intend. On sanctity of contract, see generally D Parry, The Sanctity of Contracts in English Law (Stevens & Sons 1959); JH Baker, ‘From Sanctity of Contract to Reasonable Expectation?’ (1979) 32 CLP 17. 3 Printing and Numerical Registering Co v Sampson (1874–75) LR 19 Eq 462, 465 (Sir George Jessel MR), quoted more recently in eg Johnson v Moreton [1980] AC 37 (HL) 66 (Lord Simon), Esso Petroleum Co Ltd v Harper’s Garage (Stourport) Ltd [1968] AC 269 (HL) 305 (Lord Morris). See also Homburg Houtimport BV v Agrosin Private Ltd (The Starsin) [2003] UKHL 12 [57], [2003] 1 CLC 921,

2  Magda Raczynska and Paul S Davies This passage, like most discussion about freedom of contract,4 directs our attention to the limits of the principle. Party autonomy is clearly not absolute. For example, an agreement to commit a crime is unenforceable,5 and a contract concluded at gunpoint, or induced by a misrepresentation, is voidable. But to say that interference with party autonomy should not be made lightly is questionbegging (when is an interference ‘light’? what should the effect of interference be?) and offers little guidance on when courts should (not) interfere. This book examines a range of commercial contract terms purporting to affect the freedoms of parties that have recently proved problematic. In so doing, it illuminates why there are limits to freedom of contract, and what role contracts and contract law play generally. For the purposes of this volume, we interpret ‘terms affecting freedoms’ broadly, without any attempt to construct a conceptual or technical legal category. Such terms are interesting to us for two main reasons. First, it is in relation to these sorts of terms that courts or Parliament have tended to consider whether freedom of contract should be interfered with. Looking at such terms helps us to see how much freedom of contract English law has. Second, examining how such clauses operate in practice informs our understanding of the role of contract law; it teaches us what freedom of contract is for. The discussion of those issues is particularly interesting in the context of commercial contracts6 because of the importance generally attached to freedom of contract by commercial parties. ‘Terms affecting freedoms’ can be understood in various ways. For present purposes, they can be separated into three loosely-formulated groups: (a) Terms that restrict parties’ freedoms to act, by requiring parties to behave or not behave in a particular way. Examples of typical clauses that fall into this group are: no-oral-variation clauses,7 non-assignment clauses,8 nondisclosure clauses9 and negative covenants.10 Less obvious clauses that are also 944 (Lord Steyn): ‘Commercial men must be given the utmost liberty of contracting. They must be left free to decide on the allocate [sic] commercial risks.’ 4 See MJ Trebilcock, The Limits of Freedom of Contract (Harvard University Press, 1997); P Atiyah, The Rise and Fall of Freedom of Contract (OUP 1979, reprinted 2003); R Brownsword, Contract Law. Themes for the Twenty-first Century, 2nd edn (OUP 2006) ch 3. 5 Goodinson v Goodinson [1954] 2 QB 118 (CA) 120 (Somervell LJ), and the same result would be reached under the multifactorial approach adopted by Lord Toulson in Patel v Mirza [2016] UKSC 42, [2017] AC 467. An agreement to commit a crime is itself an offence: Criminal Law Act 1977, s 1(1). 6 It is of course difficult to define ‘commercial contracts’ (see eg Cavendish Square Holding BV v Makdessi [2015] UKSC 67, [2016] AC 1172 (Makdessi) [168] (Lord Mance)) but consumer contracts clearly raise different considerations, which are not discussed in this book. 7 R Stevens, ‘Binding Our Future Selves’ ch 2 in this volume; A Burrows, ‘Anti-Oral Variation Clauses: Rock-Solid or Rocky?’, ch 3 in this volume. 8 H Beale, ‘The New Override of Bans on Assignment of Receivables’, ch 7 in this volume. 9 C MacMillan, ‘Private Law and Public Concerns: Non-Disclosure Agreements in English Contract Law’, ch 15 in this volume; and R Moorhead, ‘Professional Ethics and NDAs: Contracts as Lies and Abuse?’, ch 16 in this volume. 10 L Gullifer and G Penn, ‘The Boundaries of a Borrower’s Freedom to Act: Negative Covenants in Loan Agreements’, ch 8 in this volume.

Freedom of Contract and Terms Affecting Freedoms  3 addressed in this book are ethical clauses11 (eg requiring a party to pay fair wages to its workers, or not to manufacture products using ozone-depleting substances or processes), and, in the context of services providing speech platforms, clauses implicitly requiring users to exercise their free speech in a particular way by enabling the service operator to suspend its users.12 Clauses controlling interpretation,13 express or implied duties of good faith and controls of contractual discretions,14 as well as clauses excluding good faith,15 are also discussed as part of this category.16 (b) Terms that seek to limit or extend a party’s liability. Examples of such clauses are liquidated damages clauses,17 warranties in insurance contracts18 and clauses limiting the liability of the carrier for misdelivery.19 Entire agreement clauses and non-reliance clauses20 and clauses affecting the availability of unjust enrichment claims21 can also fall within this category since their effect is often to limit the extent to which one party may hold the other liable. (c) Terms that seek to limit or preclude the parties’ freedom to seek the usual assistance from the courts in exercising controls available under English law, or indeed any legal control at all over contracts. An innovative example of an attempt to exclude any legal control may be clauses purporting to ensure that smart contracts be enforceable without the need for the structures of a legal system.22 Arbitration clauses23 and choice-of-law as well as choice-of-forum clauses24 could also be included in this category. However, in the context of private international law, party autonomy may be understood as a freedom to choose between different freedoms under different laws, which serves a function different from contractual party autonomy.25

11 L Miller, ‘“Ethical Clauses” in Global Value Chain Contracts: Exploring the Limits of Freedom of Contract’, ch 9 in this volume. 12 N McBride, ‘“All Watched Over by Machines of Loving Grace”? The Inevitable Conflict between Contract Law and Free Speech in Cyberspace’, ch 14 in this volume. 13 R Calnan, ‘Controlling Contractual Interpretation’, ch 4 in this volume. 14 M Raczynska, ‘Good Faiths and Contract Terms’, ch 5 in this volume. 15 PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6 in this volume. 16 Admittedly, it would also be possible to locate these clauses in group (b), illustrating that the boundaries between these categories is porous. 17 W Day, ‘Disproportionate Penalties in Commercial Contracts’, ch 11 in this volume. 18 J Lowry and R Edmunds, ‘The Reform of Insurance Warranties: Looking Beyond the Past’, ch 19 in this volume. 19 M Özdel,‘The Right to Delivery of Goods under Contracts of Carriage’, ch 20 in this volume. 20 J Morgan, ‘Opting for “Documentary Fundamentalism”: Respecting Party Choice for Entire Agreement and Non-Reliance Clauses’, ch 12 in this volume. 21 N Connolly, ‘Planning for Failure: Contract Design, Ineffective Bargains and Restitution’, ch 13 in this volume. 22 S Green and A Sanitt, ‘Smart Contracts’, ch 10 in this volume. 23 U Grušić and M Penades Fons, ‘Illegality in English Arbitration Law After Patel v Mirza’, ch 18 in this volume. 24 A Mills, ‘Choice of Court and Choice of Law Agreements: Freedom of Freedom of Contract’, ch 17 in this volume. 25 ibid.

4  Magda Raczynska and Paul S Davies Traditionally, the limits on freedom of contract imposed by courts and Parliament have been greater in relation to clauses in group (b) than in group (a). But the essays in this book show that there is some movement towards greater limitation of party autonomy in relation to clauses that restrict parties’ freedoms to act (group (a)). Before we look at specific clauses, we should consider four broad trends in English contract law that might suggest the principle of freedom of contract is becoming more limited. The first relates to the way in which courts interpret contracts and thus determine the content of commercial contracts. Ever since Lord Hoffmann’s judgment in Investors Compensation Scheme v West Bromwich Building Society (No 1),26 it has been clear that agreements should be interpreted contextually. If the content of the contract is based on courts’ contextual interpretation, and somewhat vague notions of ‘commercial common sense’, then the meaning of the contract is not just a question of the language the parties have chosen. The extent to which courts should resort to ‘background factors’ in the interpretative process has been controversial and generated a great deal of ­litigation.27 In an effort to provide greater certainty to contracting parties, Richard Calnan has argued that parties should consider making use of a clause controlling contractual interpretation.28 The primary purpose of such a clause is to limit the amount of context the court takes into account when determining the content of the contract. The second trend to note is the rise in cases where contracts are characterised as relational and terms of good faith implied into contracts. This looks surprising given that English law is generally thought not to require parties to exercise their contractual rights and powers in ‘good faith’.29 Closely related is the third trend, which concerns the control of contractual discretions (via ‘Braganza-type’ duties).30 Magda Raczynska shows that the debate on good faith in English law conflates various meanings of good faith and risks impelling this jurisdiction to travel in a wrong direction. A distinction needs to be made between the general power of the courts to apply and develop existing legal principles and doctrines (which in some instances have a similar effect to the operation of good faith doctrines in other jurisdictions, suggesting that English law already has a doctrine of good faith), and the power to impose a particular standard of behaviour on the parties.31 English courts do not and should not have a general power of the 26 Investors Compensation Scheme Ltd v West Bromwich Building Society (No 1) [1998] 1 WLR 896 (HL) (ICS). 27 Arnold v Britton [2015] UKSC 36, [2015] AC 1619 suggested a ‘rowing back’ from ICS (n 26). See eg Lord Sumption, ‘A Question of Taste: the Supreme Court and the Interpretation of Contracts’ (2017) 17 Oxford University Commonwealth Law Journal 301, criticised by Lord Hoffmann, ‘Language and Lawyers’ (2018) 134 LQR 553. The position is still unclear: see eg Wells v Devani [2019] UKSC 4, [2019] 2 WLR 617, although that case concerned an oral, as opposed to written, contract. 28 Calnan, ‘Controlling Contractual Interpretation’ (n 13). 29 See eg White & Carter (Councils) Ltd v McGregor [1962] AC 413 (HL). 30 Socimer International Bank Ltd v Standard Bank London Ltd (No 2) [2008] EWCA Civ 116, [2008] Bus LR 1304; Braganza v BP Shipping Ltd [2015] UKSC 17, [2015] 1 WLR 1661. 31 Raczynska, ‘Good Faiths and Contract Terms’ (n 14).

Freedom of Contract and Terms Affecting Freedoms  5 latter kind, although one specific area where controls should apply are contractual discretions. Paul Davies addresses some concerns that imposing a ‘good faith’ standard entails and argues that parties might consider expressly excluding terms from their contracts.32 Similar to terms controlling contractual interpretation, ‘no-good-faith’ clauses would be based on party autonomy. Davies argues, however, that exclusion of some controls relating to contractual discretions is likely to be much harder. The fourth trend is different in kind, and results from the developing concept and use of smart contracts. In the extreme, it probably appears at first as limiting the role of contract law whilst expanding party autonomy. Sarah Green and Adam Sanitt identify various types of smart contract, ranging from those where a regular contract contains a short section of a self-executing code that implements a single clause to those that consist only of code and expressly exclude any other language and all contractual oversight.33 Smart contracts of the latter kind seek to remove all contractual oversight but, as Green and Sanitt argue, they are not likely to eliminate it. The effect of smart contracts on the limits of party autonomy is likely to attract greater attention in the future. Turning to specific terms that restrict one or both parties’ freedoms, it is important to engage with normative arguments in favour of overriding certain clauses, sometimes as a result of particular policies and sometimes as a result of a desire to protect the parties’ future freedom of contract. Anti-oral variation clauses have recently been given effect on the basis of respect for party autonomy in MWB Business Exchange v Rock Advertising:34 parties have no freedom to vary the contract unless and until they have complied with the formalities they themselves set out.35 The decision raises a rare instance of conflict between two principles of autonomy: does our freedom to contract include a freedom to bind our future selves (and therefore restrict our future party autonomy)? The Supreme Court in Rock said ‘Yes’,36 but this is problematic. From the perspective of contract theory, Stevens conceptualises freedom to contract as an unrestricted power to restrict our future choices, and also argues that the law should secure for us freedom (independence) from the choices of others.37 He points out that the principle of equal freedom cannot justify a power to contract that restricts our freedom from the choices of others. Stevens contends that an effective exercise of a power to vary a contract requires consideration if variation involves increasing contractual rights, but requires no consideration where variation decreases contractual rights. As for non-oral variation clauses, on Stevens’ analysis, they should be ineffective.

32 PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6 in this volume. 33 Green and Sanitt, ‘Smart Contracts’ (n 22). 34 MWB Business Exchange v Rock Advertising [2018] UKSC 24, [2019] AC 119 (Rock). 35 Unless there is an estoppel: ibid [16]. 36 Although Lord Briggs in Rock (n 34) [26] took a much more cautious approach than the majority of the Supreme Court. 37 Stevens, ‘Binding Our Future Selves’ (n 7).

6  Magda Raczynska and Paul S Davies Looking at non-oral variation clauses from the perspective of legal principle, Burrows arrives at the same conclusion as Stevens.38 He argues that the startling consequence of Lord Sumption’s reasoning in Rock is that the parties deprive themselves of the ability to enter into a valid agreement. A better position, Burrows argues, is that adopted by Beatson LJ in Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd,39 that parties can vary the contract orally notwithstanding the anti-oral variation clause. Of course, that is not the approach adopted by the Supreme Court of England and Wales, but it appears to represent the law in other major common law jurisdictions. Of the terms that restrict parties’ freedoms mentioned in group (a) above, only non-assignment clauses are clearly overridden for public policy reasons. As Hugh Beale argues,40 the purpose behind the introduction of the Business Contract Terms (Assignment of Receivables) Regulations 2018 (SI 2018/1254) was to override the effect of a clause that prohibited or limited assignment. Prior to the Regulations, the effect of such clauses in all contexts had been that an attempted assignment of such a right had no effect.41 Although in practice workarounds have developed, their effectiveness has been somewhat uncertain. For example, the effectiveness of a declaration of trust over a contractual right in the face of a non-assignment clause depends on the fact that a non-assignment clause is not also a no-trust clause, and that the characterisation of the substance of assignment is different from trust. Beale, having been at the ‘front-line’ of the reform discussions, argues that the policy driving it was to protect access to finance for small and medium-size business (SMEs) with regard to ‘bog standard’ receivables. This purpose was achieved, but, as Beale recognises, the way in which it was done leaves much to be desired. Three chapters in this book argue for greater control of party autonomy in certain contexts in order to protect other legal principles and policies. The first two relate to non-disclosure agreements. Catharine MacMillan argues that non-disclosure clauses in some contexts should be made unenforceable.42 She considers that existing ways to strike down such clauses (such as duress, mistake, misrepresentation and illegality) are problematic and that the time has come to expand the public policy grounds on which clauses can be made unenforceable to include the contravention of human rights. In a similar vein, Richard Moorhead sounds a stark warning to lawyers who advise their clients to include unenforceable clauses, as lawyers continue to face serious professional misconduct charges for ‘cheating’ on behalf of their clients.43 Moorhead argues that the threat of 38 Burrows, ‘Anti-Oral Variation Clauses: Rock-Solid or Rocky?’ (n 7). 39 Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd [2016] EWCA Civ 396, [2017] 1 All ER (Comm) 601. 40 Beale, ‘The New Override of Bans on Assignment of Receivables’ (n 8). 41 Contrast, for example, with a breach of a non-disclosure clause. 42 MacMillan, ‘Private Law and Public Concerns: Non-Disclosure Agreements in English Contract Law’ (n 9). 43 Moorhead, ‘Professional Ethics and NDAs: Contracts as Lies and Abuse?’ (n 9).

Freedom of Contract and Terms Affecting Freedoms  7 ethical misconduct adds a public interest dimension to the debate on limits of party autonomy. The third chapter considers online social media platforms such as Facebook, Twitter and YouTube. Nicholas McBride argues that the terms of those online services generally entitle the service providers to suspend users from using the service for any reason, thus affecting the users’ exercise of their freedom of speech.44 This is particularly problematic where the suspension occurs because a user’s views on what ‘human flourishing’ entails are not congruent with the general views on this endorsed by the service provider. McBride examines the various ways in which the service user could try to win in a dispute with the service provider, but suggests that the existing avenues offered under private law are not likely to succeed and are therefore unsatisfactory. Instead, he argues, a contractual term that allows A to impose a sanction on B because B adopts (or questions) a particular view of what human flourishing involves should be invalid on grounds of public policy. McBride further suggests that there should be tortious liability for denial of the service by A. While the chapters described above set out arguments for increased restrictions of contractual terms affecting freedoms based on public policy, the chapter by Uglješa Grušić and Manuel Penades Fons45 shows a new side to existing public policy restrictions. They argue that in relation to contracts with arbitration clauses, the multi-factorial approach to illegality developed by Lord Toulson in Patel v  Mirza46 (read together with Les Laboratoires Servier v Apotex Inc47) makes it possible to take into account illegality other than that provided by the proper law or the law of the place of performance. This suggests that, in some cases, contracts with arbitration clauses may become unenforceable in a wider range of circumstances than was the case before Patel v Mirza. We now turn to chapters that focus on terms limiting or excluding liability. In determining whether a term is a valid liquidated damages clause or an unenforceable penalty, the Supreme Court in Cavendish Square Holdings BV v Makdessi48 departed from the old test of whether the stipulated sum was a ‘reasonable preestimate of loss’49 in favour of a test that looks at whether the clause that stipulates the sum to be paid ‘imposes a detriment on the contract-breaker out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation’.50 William Day argues that the new test is best understood as a rule against disproportionate penalties, not a rule against penalty clauses per se, 44 McBride, ‘“All Watched Over by Machines of Loving Grace”? The Inevitable Conflict between Contract Law and Free Speech in Cyberspace’ (n 12). 45 Grušić and Penades Fons, ‘Illegality in English Arbitration Law After Patel v Mirza’ (n 23). 46 Patel v Mirza [2016] UKSC 42, [2017] AC 467 [120]. 47 Les Laboratoires Servier v Apotex Inc [2014] UKSC 55, [2015] 1 AC 430. 48 Makdessi (n 6). 49 Dunlop Pneumatic Tyre Co Ltd v New Garage & Motor Co Ltd [1915] AC 79 (HL) 87–88 (Lord Dunedin). 50 Makdessi (n 6) [32] (Lord Neuberger and Lord Sumption).

8  Magda Raczynska and Paul S Davies and that it reflects a wider public policy in English law against disproportionate punishment.51 Day argues that this new test strikes a good balance between freedom on contracting and freedom after contracting, which could not be said of the previous test. Discussing entire agreement clauses and non-reliance clauses, Jonathan Morgan concedes that such clauses may, in substance, operate as exclusion clauses.52 He shows that whilst the courts have some tools to review such terms, these are limited or applied in a limited way. For example, there is a principle of public policy that a party cannot exclude liability for its own fraud, but it is unclear whether this extends to clauses excluding liability for one’s agent’s fraud; there are dicta going both ways.53 And even where courts apply statutory controls,54 the clauses may well meet the statutory test of reasonableness, especially in contracts between commercial parties.55 Morgan also dismisses broader (academic) objections raised against the enforceability of such clauses, arguing that such terms are an important tool for the parties to signal their preference for a textual and formalistic regime over a more contextual (‘relational’) approach. Niamh Connolly in turn shows that parties have considerable freedom to control the potential ways in which restitutionary liability for unjust enrichment can arise when a contract is found to be ineffective or invalid.56 Given the somewhat crude requirement that consideration must fail totally for a claim in unjust enrichment to succeed, Connolly argues that it is important that those drafting agreements make use of the freedom afforded to them by contract law to ensure greater choice, control and certainty when restitutionary scenarios arise. For example, parties could identify the core of the bargain and what each act of performance is conditional upon, or explicitly state that the performance is divisible and indicate how different parts of the consideration correspond to different parts of the performance. There are two specific contexts in which clauses excluding liability of a party have attracted stricter controls. One relates to insurance contracts, the other to contract of carriage. In relation to insurance, John Lowry and Rod Edmunds examine the development of the long-settled rule that breach of warranty in the insurance policy automatically discharges the underwriter’s liability as from the date of the breach.57 They show that the rule, along with the strict liability approach, was based on the idea that the underwriter was the vulnerable party

51 Day, ‘Disproportionate Penalties in Commercial Contracts’ (n 17). 52 Morgan, ‘Opting for “Documentary Fundamentalism”: Respecting Party Choice for Entire Agreement and Non-Reliance Clauses’ (n 20). 53 HIH Casualty and General Insurance Ltd v Chase Manhattan Bank [2003] UKHL 6, [2003] 1 CLC 358. 54 eg Misrepresentation Act 1967, s 3; Unfair Contract Terms Act 1977, s 3(2)(b)(i). 55 See eg First Tower Trustees Ltd v CDS (Superstores International) Ltd [2018] EWCA Civ 1396, [2019] 1 WLR 637. 56 Connolly, ‘Planning for Failure: Contract Design, Ineffective Bargains and Restitution’ (n 21). 57 Lowry and Edmunds, ‘The Reform of Insurance Warranties: Looking Beyond the Past’ (n 18).

Freedom of Contract and Terms Affecting Freedoms  9 in need of protection. Lowry and Edmunds argue that the Insurance Act 2015 in respect of insurance warranties represents a re-balancing of the law, away from its protectionist instincts. For instance, the statute substitutes the long-settled rule with the notion of suspensory liability, such that breach of a term merely results in cover being suspended until it is remedied. Freedom of contract is balanced with what we might view as a wider public policy in English law against disproportionate punishment, which we saw also underpinned penalty clauses. In the context of carriage, Melis Özdel brings to our attention the fact that clauses excluding a carrier’s liability to those interested in the cargo are struck down under Article III, rule 8 of the Hague-Visby Rules, which can apply mandatorily or by contract.58 She demonstrates that the historical reason behind this has been the one-sided nature of contracts of carriage contained in or evidenced in the bill of lading. However, Özdel argues that in practice there is a range of circumstances when the liability for misdelivery can arise outside the scope of the Rules, so the restrictions on the carrier’s exclusion of liability do not bite. Nevertheless, the carrier’s exclusion of liability clauses do not tend to cover liability arising for breach of bailment or conversion. Another main message that emerges from this book relates to the role of contracts and contract law. Stevens asks why we need contract rights as opposed to merely powers to enforce them, and answers that litigation is not the only forum in which contract law is important.59 The truth of this statement is seen in the chapter by Louise Gullifer and Graham Penn.60 They argue that the purpose of negative covenants in loan agreements is to align the duties of the directors in deciding on the activities of the borrower with consideration of the lender’s interests. The fact that contract rights do not exist exclusively for the purpose of their enforcement means that a party cannot be compelled to enforce any term. This can be a problem in the context of global value chain contracts that contain ethical clauses (eg to pay workers fair wages, or to manufacture products in an environmentally sustainable way) examined by Lucinda Miller.61 She shows that contracts with such ethical clauses assume a public or regulatory role. Public policy is generally used as an argument to restrict the validity or enforceability of certain clauses, but Miller in her chapter directs our attention to a new question: whether public policy could be used to restrict a party’s choice in enforcing (or not) a particular clause. The chapters by Stevens, Gullifer and Penn, and Miller inform our understanding of the principle of freedom of contract. Their analysis shows that an important ingredient of party autonomy after contracting is the power to decide whether and how to enforce the other party’s obligations. 58 Özdel, ‘The Right to Delivery of Goods under Contracts of Carriage’ (n 19). 59 Stevens, ‘Binding Our Future Selves’ (n 7). 60 Gullifer and Penn, ‘The Boundaries of a Borrower’s Freedom to Act: Negative Covenants in Loan Agreements’ (n 10). 61 Miller, ‘“Ethical Clauses” in Global Value Chain Contracts: Exploring the Limits of Freedom of Contract’ (n 11).

10  Magda Raczynska and Paul S Davies Mills’ chapter reveals that freedom of contract is not the only freedom that can be the basis of terms.62 In arguing that choice of law and choice of forum clauses are based not on freedom of contract but on broader party autonomy, he shows that freedom of contract is not the sole principle aimed at protecting parties’ choice. Having looked at the main themes that emerge from the discussion on contractual terms in this book, some questions emerge about how these themes fit with the ongoing theoretical debates about contracts.63 Why should freedom of contract be respected in a legal system? To the extent that freedom of contract is limited in law, how should those limitations be justified? Various justifications for the existence of freedom of contract have been given. But one possible view is that freedom of contract cannot be justified. This is often associated with the reliance-based approach to contract law.64 The reliance-based view of contract law conflicts with the classical view of freedom of contract: the latter emphasises that the content of contracts is a matter of the parties’ choice,65 whereas the former suggests that the content of the parties’ contract (their obligations) is a matter of law, specifically the law’s response to protect reliance on a representation. However, the reliance view is not supported by case law66 and the element of reliance, if it exists, inevitably seems somewhat thin. A better approach to party autonomy is that the contents of contracts are a matter of the parties’ agreement, and their meaning is ascertained objectively.67 Courts apply an objective approach when determining the contents of contracts, and consider how the words or conduct would appear to a reasonable person in the position of the addressee.68 Through this process it is possible to determine, objectively, the parties’ intention. If freedom of contract is based upon the objective will of the parties, it is necessary to explain why it exists and why it should be limited. The importance of autonomy has a long history in moral philosophy. Kant saw autonomy as the foundation of morality, and a moral good that facilitates self-fulfilment.69 If the parties’ objectively manifested will is based on this deontologically libertarian understanding of autonomy, the result in practice is that the 62 Mills, ‘Choice of Court and Choice of Law Agreements: Freedom of Freedom of Contract’ (n 24). 63 See Smith, Contract Theory (n 1) ch 8. 64 Atiyah, The Rise and Fall of Freedom of Contract (n 4) 731–33, 742, 744–47. 65 See text to n 1. 66 See eg Hasham v Zenab [1960] AC 316 (PC), where a party was bound to a contract despite tearing up the document within minutes of signing it, and despite the fact that the other party had no chance to rely on it. 67 See eg Stevens, ‘Binding Our Future Selves’ (n 7); Smith, Contract Theory (n 1) 273. 68 ICS (n 26) 114; Destiny 1 Ltd v Lloyds TSB Bank Plc [2011] EWCA Civ 831; Hartog v Colin & Shields [1939] 3 All ER 566. 69 E Kant, Groundwork of the Metaphysics of Morals in M Gregor (trans, ed), Practical Philosophy (CUP 1996, originally published in 1785) 440, defining autonomy as ‘the property of the will by which it is a law to itself (independently of any property of the objects of volition)’. See also the discussion of party autonomy in Mills, ‘Choice of Court and Choice of Law Agreements: Freedom of Freedom of Contract’ (n 24), section III.A.

Freedom of Contract and Terms Affecting Freedoms  11 limitation of party autonomy needs to be justified but the autonomy itself does not.70 That this thinking appears to underpin freedom of contract under English contract law might perhaps be gleaned from Sir George Jessel MR’s dictum in Printing and Numerical v Sampson, cited at the beginning of this chapter.71 However, the discussion of no-oral-variation clauses,72 non-disclosure clauses,73 ethical clauses74 or terms terminating online social platform service agreements75 suggests, in effect, that there are problems with deontological libertarianism. The reason that parties enter into contracts is not to realise the value of their abstract will but in order to pursue their wishes. While the law cannot protect every whim of the parties, it can recognise that the parties’ rights and powers are protected by law so that the parties can realise their desired practical benefits. This conception of party autonomy draws upon the thinking of Rudolf von Jhering, a nineteenth-century jurist, who has been called the German Bentham.76 Jhering developed his argument in opposition to the ‘will jurisprudence’,77 which valued an abstract will. He drew attention to the fact that human will is moved by practical benefits, not a categorical imperative. He understood a right as a means to an end, as a legally protected interest: the substance of a party’s right was the practical benefit (‘interest’) that the right conferred.78 Jhering understood the concept of legal interest broadly, encompassing both economic and non-economic benefits; for example, business profits and creativity were both legal interests under his approach. The main problem with Jhering’s philosophy is that he did not provide any objective criteria for identifying which purposes the law ought to serve, and for determining the interests worthy of protection.79 Jhering’s ideas are useful in so far as they direct our attention to the existence of different interests parties might pursue through contracting. In the context of our discussion of terms affecting freedoms, Jhering’s insights help us to see that both party autonomy and its restrictions should be justified, so consideration of interferences with autonomy inevitably becomes a balancing exercise.

70 Terminology taken from Mills, ‘Choice of Court and Choice of Law Agreements: Freedom of Freedom of Contract’ (n 24). 71 Printing and Numerical (n 3). 72 Stevens, ‘Binding Our Future Selves’ (n 7) and Burrows, ‘Anti-Oral Variation Clauses: Rock-Solid or Rocky?’ (n 7). 73 MacMillan, ‘Private Law and Public Concerns: Non-Disclosure Agreements in English Contract Law’ (n 9) and Moorhead, ‘Professional Ethics and NDAs: Contracts as Lies and Abuse?’ (n 9). 74 Miller, ‘“Ethical Clauses” in Global Value Chain Contracts: Exploring the Limits of Freedom of Contract’ (n 11). 75 McBride, ‘“All Watched Over by Machines of Loving Grace”? The Inevitable Conflict between Contract Law and Free Speech in Cyberspace’ (n 12). 76 W Seagle, ‘Rudolf von Jhering: or Law as a Means to an End’ (1945) 13 The University of Chicago Law Review 71, 71. 77 eg F von Savigny, System des Heutigen Römischen Rechts (Veit 1849). 78 R von Jhering, Der Zweck im Rechts (Britkopf & Härtel 1877), translated into English by I Husik as Law as a Means to an End (The Boston Book Company 1913) (arguing that the essence of any right is not the absolute will but the interest or purpose for the sake of which the right exists). 79 Seagle, ‘Rudolf von Jhering’ (n 76).

12  Magda Raczynska and Paul S Davies None of this undermines the status of freedom of contract as a crucial legal principle. If in doubt, the interests protected under freedom of contract should outweigh other interests. The balancing approach to freedom of contract and its limitations can be viewed as an invitation to more rigorous discussion of which interests are being protected in given circumstances and why. The balancing approach might naturally lead to different results in different circumstances. For example, in relation to the discussion of non-disclosure clauses in this book, the balance may be tipped against the sanctity of the clause in favour of other interests; but in other contexts, such as entire agreement and non-reliance clauses, the balancing exercise might help to emphasise the importance to both parties of achieving certainty via ‘documentary formalism’.80 The examination of terms affecting freedoms in this book sheds new light on the extent of limitations on freedom of contract under English law, the role of contract terms and contract law, and the role of freedom of contract. The lessons learnt from this book allow us to see that freedom of contract can be understood as capturing the idea that the content of the parties’ contractual obligations is a matter of choice for the parties, rather than imposed by the law (which can be referred to as ‘freedom on contracting’); and that the obligations voluntarily undertaken by the parties are enforced by courts, although the parties have the freedom to choose whether to enforce them in courts (‘freedom post-contracting’). Moreover, although the dictum of Sir George Jessel MR in Printing and Numerical Registering Co v Sampson quoted at the start of this chapter81 indicates that freedom of contract requires no justification, and that interferences with that freedom must be explained as a matter of ‘paramount public policy’, that is largely unhelpful. The preferable approach is to balance the interests protected by freedom of contract against interests that would be protected if freedom of contract were to be restricted, although there is a strong presumption that the balance be tipped towards freedom of contract.

80 To borrow Jonathan Morgan’s expression: Morgan, ‘Opting for “Documentary Fundamentalism”: Respecting Party Choice for Entire Agreement and Non-Reliance Clauses’ (n 20). 81 Printing and Numerical v Sampson (n 3).

2 Binding Our Future Selves ROBERT STEVENS

I.  Power to Contract ‘Freedom of contract’ is an inaccurate label for the idea that it seeks to capture. The law very rarely restricts the freedom of people to enter into any agreements they choose (agreements to commit criminal offences are a rare example). If we wish to agree to a non-disclosure agreement, or an agreement containing a penalty clause, or an agreement excluding my liability for negligently injuring you, or an agreement not to enter future agreements save in particular ways, or an agreement that we will never again enter into an agreement, or (nowadays) an agreement to maintain a lawsuit with the funder sharing in successful proceeds, or most other agreement the human mind can imagine, we are perfectly at liberty to do so. What is ordinarily meant by ‘freedom to contract’ is not this absence of any restraining duty to behave in this way, but instead the parties’ unrestricted power to contract. Contracts create rights between the parties entering into them that did not exist before. Contracts remove freedoms (not to work, not to pay money, not to deliver goods, etc) that we would otherwise have. Further, those rights are often enforced through the power of the state (either through the compelling of their performance, or through the enforcement of an obligation to pay damages for breach). Contracts restrict our future choices. The mystery that requires explanation is not why there may be limits placed upon our power to constrain ourselves, but why we should be able to do so at all. In order to understand the limits of our power to bind our future selves, it is first necessary to understand how and why we should have this power at all, and what business it is of the law to compel us to stand by what we have agreed, constraining our independence. This chapter proceeds in five stages. First, it will be argued that what is of primary importance are the rules of contract law, not the ability to enforce them. Second, it will be argued that agreements are a necessary condition of the creation of contractual rights, promises being insufficient. Third, that it is important to distinguish between agreements, and enforceable contracts, and that unenforceable agreements are not a (legal) nullity. Fourth, that the doctrine of consideration is in English law a substantive condition of there being an enforceable contract,

14  Robert Stevens and not a mere formality requirement. Fifth, that the ratio of the leading case of Foakes v Beer1 is more narrow than commonly supposed, that principle requires a distinction between increasing and decreasing pacts, and that the Supreme Court’s failure to review the area in Rock Advertising Ltd v MWB Business Exchange Centres Ltd2 was unfortunate.

II.  Why Do We Need Contract Rights? The ability of human beings to create duties between one another, and thereby coordinate their lives, is without doubt extremely useful, but it may be doubted whether the law is central to our engaging in this activity. Indeed, in some cases it may be positively counter-productive. The law’s irrelevance may be illustrated by the difference between gaming agreements and contracts of insurance. Until relatively recently,3 gaming a­ greements were, as a matter of law, void, whereas insurance contracts have always been valid and enforceable. This did not, however, prevent there being a large and t­hriving gaming industry. The structure of the two kinds of agreement is very similar (a stake/premium, with a countervailing payout on the happening of a contingency). The security the insured/gambler had of a ‘winning’ payout was factually very similar, despite the lack of any legal rights. Why? The answer is that a bookmaker who did not pay would immediately destroy his reputation in the business. This provided (and provides) a far greater incentive to comply with the obligations created by an agreement than does the threat of legal action on an individual bet. The bookmaker did not need recourse to law against the player (who may be a one-off participant in the market and so careless of his reputation) because he required payment (a stake) in advance. If anything, respect for agreements, and their diligent performance, has been less of a feature of the insurance agreements that have always been enforceable than of gaming agreements that were not. This phenomenon should be familiar to all of us as consumers. Why should a builder or plumber seek to provide work of acceptable quality to a consumer who employs him? The threat of legal action if the work is unsatisfactory is, in most cases, remote. Even for those well versed in the law, the time, trouble and expense of legal proceedings, even before an informal small claims court, is often out of all proportion to the size of any claim. For traders, their reputation, now given a valuable online public score by those employing them on websites such as Checkatrade, is far more important as an inducement to perform agreements than is the threat of legal action.



1 Foakes

v Beer (1884) 9 App Cas 605. Advertising Ltd v MWB Business Exchange Centres Ltd [2018] UKSC 24, [2018] 2 WLR 1603. 3 Gaming Act 2005, repealing the Gaming Act 1845. 2 Rock

Binding Our Future Selves  15 Performance by consumers is less easily incentivised without the threat of law, because they may not be repeat players in the market, but even here peer-to-peer Internet hubs such as AirBNB and Uber can provide a ranking for those who are, giving traders some assurance, and for the reputable the ability to pick and choose with whom they deal. More useful than legal proceedings (which in the case of small sums will usually not be worth pursuing against consumers) will be diversification of risk (extending smaller sums as credit to a large range of people over short periods, only a small fraction of whom may default) and requiring stage payments on larger jobs (with the consumer who does not pay consequently left in the lurch). In the context of business to business and business to consumer contracts, therefore, the importance of legal recourse has always been small, and has declined as online reputation indicators have now become readily available. When going on holiday abroad, even lawyers will check a hotel’s Tripadvisor score, not the terms and conditions of the standard form contract and the cost of proceeding before the local courts. Whilst legal recourse is of small, and declining, importance in the vast majority of contracts, even in a world of angels we need contract law to give certainty to the rights of the parties. What happens in the case of an unexpected event? What do the terms mean in the case of ambiguity? What ought one party to do instead if he fails to perform? When can one party refuse to perform because of the failure of the other to do so? In a game of Monopoly between friends, we need a set of rules so that each player knows what she must do, even absent any external coercive power to make us comply. As it is in games, so it is in contracting. We need a set of rules, agreed to apply in advance, so that we know what it is that we must do. The rules of contract are, therefore, far more important than the ability to enforce them through court action. It is easy for lawyers, especially common lawyers, to place too much emphasis on civil recourse. We are taught contract law through cases, where one party has sued another. Judges, and litigation lawyers, then naturally continue to think of contract from the perspective of litigation. Transaction lawyers know this provides a distorted picture. Most contracting parties want to perform their contractual obligations. They just want to know what they are, and to be able to determine for themselves what they are in advance so that they may be accurately priced. Enforceability before a court is not central to contract law.

III.  Promises and Agreements The problem with Charles Fried’s classic Contract as Promise4 is apparent in its title. Contracts are not promises. Good people keep their promises. Indeed, good

4 C

Fried, Contract as Promise: A Theory of Contractual Obligation (Harvard University Press 1981).

16  Robert Stevens people have a reason to comply with the vows that they privately make in their own heads. Being resolute is a virtue, and we ought to keep our vows, subject to countervailing reasons on the other side of the moral ledger. Promises are a fortiori vows. Someone who has made a promise to another invites that person to trust him, and, all other things being equal, acts in a blameworthy way when that trust is disappointed. The key divide that Fried failed to cross is that between duties of virtue and duties of right. It is not enough to justify imposing legal duties to act (or not act) that good people behave (or do not) in such a way. Should the law be in the business of forcing us to be virtuous, or (perhaps less dramatically) in prompting us to conform with our good reasons for action? The idea that it should is what, in earlier times in England, and still today in some jurisdictions, prompts law makers into enforcing various religious norms. If you believe in your heart that those who leave parts of their body uncovered, or eat particular kinds of food or play games on certain days of the week, are going to eternal hell, then the case for enforcing these duties, or at least providing carrots and sticks to prompt people not to behave in such a way, may appear overwhelming. Even for non-believers, there may be thought to be good reasons to pass laws restricting the calories we can consume or requiring us to take daily exercise. Why not employ the law to make us be good? Some of a more liberal persuasion might take the view that only our moral duties that concern other people ought to be enforced, and that we should be left to decide for ourselves whether to brush our teeth or read improving books. However, that a moral duty concerns another, cannot alone turn it into a duty of right or entitlement, in the sense that the positive law is then justified in ­enforcing it. The most obvious illustration of the difference between duties of right and duties of virtue is the case of easy rescue. The baby drowning in an inch of water. Do good people pick up drowning babies when they could easily do so? Yes of course. We all have good, indeed overwhelming reasons for performing cost-free, easy rescues such as this. If the law were in the business of requiring us to conform with our good moral reasons for action, it should certainly do so here. For the utilitarian and other consequentialists, there is no (easy) distinction between drowning a baby and not saving one. The end result, one dead baby, is the same. Compelling people to be good by operation of law is wrong because it is to use people as a means to an end. This places limits upon the kinds of duties the law may legitimately impose. If duties of right are instead based upon our entitlement to independence from the choices of others, the drowning baby should have no right to be saved. If the baby had a legal right that I save it, this would be to use me as a means to an end, and be inconsistent with my independence. It would compel me to use my body for another. In a world of absolute freedom, we would all be free to stab one another in the back, burn down a farmer’s crops, tell strangers that our acquaintances are axe

Binding Our Future Selves  17 murderers and lock others in our basements. If we start from the premise that law should secure for each of us the same ability to lead our lives independent of the choices of others, from this first principle of independence we can at least deduce rights not to be injured, locked up, defamed and lied to. Independence alone does not justify rights that others cure us when we are ill, speak well of us when it is justified, repair our broken bicycles or educate us in things we do not know. Most of the law of torts can be justified in this way. In the common law it is wrong to drown babies, but not to fail to pick them up. But, without further work, a principle of equal freedom cannot alone justify the power to contract, which restricts our future independence from the choices of others. Perhaps the most that can be said is that it is part of being human that we can form bonds of duty between one another.5 So, just as if we would have no need for property rights if we were hyper-intelligent sharks (as unable to possess or make use of things external to ourselves), if we were unable to communicate with one another we would have no contractual rights (as unable to form the bonds between each other). That (fortunately) is not the world we find ourselves in. Given that rights that others do things are a (marvellous) feature of our human lives, enabling coordination that would otherwise be impossible, how can such rights be created compatibly with our all possessing equal freedom one from another? Absent a system of slavery, agreement seems to be the only legitimate method available. The practical person may ask why it matters that the law is concerned with rights bilaterally created by agreement between persons, and not merely (unilateral) duties of personal morality, such as arise from bare promises? For the hard-headed lawyer unconcerned with the above (brief, inadequate) foray into theory, it helps us to explain some core features of our law.

A. Acceptance It is possible to impose moral duties upon ourselves to act in particular ways unilaterally. A vow in our minds alone creates a good reason for action, and a promise ought to be kept as soon as it is made. Rights are different. Ordinarily we cannot confer rights upon another without that person’s acceptance. If I want to transfer my right to my camel to you, or my right to a portion of land, it is wrong if I am able to do so regardless of your consent. The camel owner will usually feel obliged to feed and care for it. Ownership of land transferred may be subject to a duty to incur environmental clean-up costs. Even my paying you $1m may put you into a higher income bracket, subjecting you to taxes you do not want to pay. Although such consent to acquisition on the part of the recipient may

5 Children can form such bonds too, although we protect them from the enforcement of the agreements they enter into.

18  Robert Stevens as a matter of fact be capable of being inferred by conduct (eg whoops of delight), there seem to be good arguments as to why we should not be able to impose rights upon others that they do not want. As a result it is (generally) not possible to transfer title to things to others without their acceptance.6 If I own a camel, I cannot make you the owner of it without your cooperation. I cannot pay you cash without your acceptance; throwing notes on to your front lawn will not do. I may be able to pay money into your bank account without your personal cooperation, but that is because banks are agents for acceptance on behalf of their customers. The agent’s acceptance is that of the principal. As it is for the transfer of rights, so it is for their conferral by contract. A unilateral promise is not enough; the law requires an offer and an acceptance. Agreements are the doing of both parties. If our sole concern were the protection of the promisee, the putative right holder, a wholly subjective uncommunicated assent should suffice. In law it does not. Another alternative to the position of the positive law that acceptance is required would be to say that the promise alone confers the right immediately but to give the promisee the power of rejection. Why should this not suffice, obviating the need for any acceptance? A focus on the morality of promising fails to capture the mutuality of contracting, that it is done by both together not by one alone.7 It is a power that cannot be exercised unilaterally. Further, a contract mutually made cannot be unilaterally ended. If a promise is the creation of a unilateral act of the promisor, it could be said that his own subsequent unilateral act could bring it to an end, however wanting in virtue that may be. As a contract is the creation of both parties, through an offer and acceptance communicated by both, one alone should not be able to withdraw.

B. Objectivity By contrast, for those who would see contracts as promises created by the will of the promisor, or in combination with the will of the promisee, binding upon us for the same reason as promises more generally are, the existence and content of a promise ought to be determined by the subjective intention(s) of the parties making them. Although versions of this explanation for the law of contract as based upon the (subjective) morality of promise-keeping has never fitted very

6 In the modern law there are some exceptions, such as where title can be transferred by registration. These are no doubt convenient, and we give the transferee the option to renounce the transfer, thereby treating it as if it never happened. 7 ‘Unilateral’ contracts are also mutual in this sense: they require offer and acceptance. They are labelled unilateral because only one party comes under any obligations. Mrs Carlill was under no contractual obligation to do anything (Carlill v Carbolic Smoke Ball Company [1893] 1 QB 256).

Binding Our Future Selves  19 well with the positive law as it is, it has persisted from the nineteenth century until the present day.8 Two tactics are commonly employed to explain the positive law of contract’s divergence from promissory theory. Neither is persuasive. First, it is argued that the law uses an objective standard for evidentiary reasons: it promotes certainty and prevents fraud.9 Second, that because the promisee will have relied upon the promise, or that there is at least a risk that she will have done so, this provides a good reason for departure from what was actually intended.10 But neither of these contingent reasons need be operative, and still the law applies an objective standard. If promises were solely a matter of personal virtue, then what the promisor subjectively intended would be determinative of both the existence and content of the promise, just as the existence and content of a vow is determined by what goes on in our heads. However, unlike a vow, we cannot make an agreement on our own in a darkened room. I cannot simply will an agreement into existence, however much I try. Even if two parties intend precisely the same promise from one to another to exist, that alone is not an agreement. Agreements are actions in the world, an offer and an acceptance. They are acts of communication; indeed they are the archetypal example of speech acts. Both the existence and the content of an agreement are determined objectively because the existence and content of communications are determined objectively.11 Nevertheless, for an act to impose obligations upon me it must be my doing; I must be responsible for it. If a fraudster impersonates me and forges my signature on a contractual document, I am not bound even if the promisee believes that the promise was mine and any reasonable person in his shoes would have believed the same. It is not the case, however, that I am responsible only for the consequences of my actions that I intend. If I negligently knock over a Ming vase, I am responsible for its destruction, and in law must pay for it, regardless of the fact that that result was not what I intended. The same is true of words. If I promise to sell to you my white cat for $100 that is what I have done, regardless of whether in my head I intended to offer to buy your black dog for $50. The narrow doctrine of non est factum is a nice illustration of the point. If a celebrity is asked to sign his autograph and it is later discovered that the document is in fact a contract for the sale of his home, he may plead that the document is not his. If, however, someone is careless in signing a document, such as someone

8 eg FC von Savigny, System des Heutigen Römischen Rechts (first published 1840); Fried, Contract as Promise (n 4) 12–13, 60–63; TM Scanlon, ‘Promises and Contracts’ in P Benson (ed), The Theory of Contract Law: New Essays (CUP 2001) 104. 9 Fried, Contract as Promise (n 4) 62–66. 10 ibid 62–63; P Atiyah, An Introduction to the Law of Contract, 5th edn (Clarendon 1995) 82. 11 See T Endicott, ‘Objectivity, Subjectivity and Incomplete Agreements’ in J Horder (ed), Oxford Essays in Jurisprudence (4th series, OUP 2000).

20  Robert Stevens who signs a document containing blanks that are later filled in otherwise than in accordance with his instructions, he may be bound.12 This does not mean that what the contract means is to be determined by a rigid literalism, anymore than this is true of the use of words generally. ‘Many people, including politicians, celebrities and Mrs Malaprop, mangle meanings and syntax but nevertheless communicate tolerably clearly what they are using the words to mean. If anyone is doing violence to natural meanings, it is they rather than their listener.’13 An important illustration of the irrelevance of what the parties intended (either as a matter of subjective fact, or what a reasonable person would taken them to have been) is the exclusion from consideration in interpreting an agreement of their subsequent conduct.14 If our concern were to determine what the parties themselves intended their agreement to mean,15 or what a reasonable person would have thought they intended, how the parties subsequently operated their agreement would be extremely strong probative evidence, and its exclusion would make no sense. If, however, we are interested in discovering what the words they have used mean, what they thought they had meant, or what a reasonable person would think that they intended, is neither here nor there. Subsequent conduct is excluded because it would not tell the court anything at all about what it is looking for: the meaning of the words in the agreement. This is immutably fixed at the time they are agreed. A strong parallel here is with the interpretation of statutes. So for many years in the UK, the rule was that the proceedings in Parliament were inadmissible in interpreting a statute. If this rule, and its analogous contract restrictions, is seen as a rule of evidence, it makes little sense in an era when the finder of fact is a judge as opposed to a jury. Why can the judge not be trusted to give the evidence the weight it deserves? This seems to have been the idea behind the decision of the House of Lords in Pepper v Hart16 to admit statements made in the House of Commons or House of Lords during the passing of a Bill in interpreting the resultant Act. However, an Act is an act. What the words used mean is a quite different question from what the promoters of a Bill intended them to mean. The goal of interpretation is not to discover what was intended by anyone, and this is even more obvious in relation to Acts of Parliament than contractual terms. The liberalisation of Pepper v Hart has proved to be a damp squib,17 with the judiciary forced to clarify that although statements in the passing of a Bill might (often) tell 12 Gallie v Lee [1971] AC 1004. 13 ibid. 14 L Schuler AG v Wickman Machine Tool Sales Ltd [1974] AC 235 (HL). 15 There are, of course, many loose judicial statements that that is what is being searched for, eg Arnold v Britton [2015] UKSC 36, [2015] 2 WLR 1593 [15]. These cannot be taken too seriously. 16 Pepper v Hart [1992] AC 593. 17 R v Warwickshire CC, ex parte Johnson [1993] 1 All ER 299; Robinson v Secretary of State for Northern Ireland [2002] UKHL 32, [2002] NI 390; Wilson v Secretary of State for Trade and Industry [2003] UKHL 40, [2004] 1 AC 816.

Binding Our Future Selves  21 us something about the intentions of those enacting it, the court is searching for something else: the meaning of the words enacted. It may be objected that there are occasions when the courts look to the parties’ subjective states of mind in finding the existence of a contract. If one party subjectively makes a mistake as to the terms of the deal, and the counterparty knows of that mistake, there is no binding agreement.18 A famous example is the so-called ‘­snapping up’ case of Hartog v Collin Shields.19 The defendant offered to sell Argentine Hare skins at a price quoted ‘per pound’. This was a mistake. It was intended to offer them ‘per piece’ and there were about three hare skins to the pound. As the plaintiff knew that the defendant was making a mistake, he could not enforce the contract at the low price stated in the offer. This rule is usually said to show that the question of agreement is not a wholly objective one but that it also contains subjective elements.20 However, that this is not so may be shown by considering what the position would have been if the market had moved dramatically, so that the deal had become a good one for the seller. Could the buyer have insisted that there was no agreement? The answer is surely no. There was objectively a bargain, but one that only the seller could enforce. Why could the buyer not rely upon the agreement? Because the buyer’s subjective consent to the bargain was vitiated, he was not responsible for it. The non-conformity of the parties’ subjective intentions with what the agreement they have objectively created may provide a good reason for not enforcing what they have done, but there is no justification for enforcing the parties’ subjective intentions, even where shared.

C. Privity If we consider contracts to be binding either because that is what is subjectively intended by the promisor or because another has or may rely upon the promise, then the doctrine of privity of contract (which is not, unlike the doctrine of consideration, a common law idiosyncrasy) appears anomalous. An unholy alliance of will theorists and reliance theorists have long denounced privity of contract. If the parties intend a third party to get a contractual right, it would be to frustrate their intentions for the law not to confer one upon that third party. A third party may expect performance, rely upon it and have a factual interest in its being rendered. If, however, we think in terms of rights, the doctrine of privity of contract follows naturally. The only person who can enforce a right is the right holder, and 18 Smith v Hughes (1870–71) LR 6 QB 597 has come to stand for this proposition. Of course, it will be a rare case where the promisee knows this when a reasonable person would not know of the mistake, but it is perfectly possible, and if the promisee is honest he will admit it. 19 Hartog v Collin & Shields [1939] 3 All ER 566. 20 E Peel, Treitel: The Law of Contract, 13th edn (Sweet & Maxwell 2012) [1-002].

22  Robert Stevens persons who are made worse off because of the infringement of someone else’s right do not have standing to sue. The doctrine of privity of contract is one form of these starting points, which are found across private law. So, if a buyer of goods suffers loss because the goods delivered have been damaged by the carelessness of a carrier employed by the seller, he will have no claim against the carrier if he did not own the goods at the time of damage.21 If a husband suffers loss because of the defamation of his wife that results in her losing her job, he will have no standing to sue. If an elderly widow is deceived into making worthless investments by a rogue, her heirs will have no claim in their own names. If a nuisance is created by interfering with the quiet enjoyment of land, those with a proprietary interest in the land may sue, but others who suffer a loss, such as contractual licensees, cannot claim.22 If a parent is injured by a driver’s carelessness, any dependent children who suffer a loss because of the victim’s loss of earnings have no right of action.23 Third parties are not parties to the agreement, and they do not therefore acquire any contractual rights. They therefore have no claim if they are left worse off than they otherwise would be where a contract is breached. Nothing more needs to be, or can be, said. There are sometimes said to be a number of common law exceptions to this rule, but almost all are nothing of the kind. In England, however, the Contracts (Rights of Third Parties) Act 1999 has created a large exception to the general rule.24 Further, the rights it creates do not need to be accepted by the third party, and are not, at least on the face of the legislation, capable of being disclaimed by him. In principle this seems incorrect.

IV.  Rights and Recourse Most of the restrictions on our power to contract do not relate to the parties’ ability to create rights through agreement between each other (which I have claimed is generally what matters). Rather they concern the ability to create enforceable contractual rights (ie there are restrictions on the civil recourse that the law will provide). The distinction between the rights created by contract law (indeed by private law in general) and the power to sue to enforce those rights is the same as the distinction between the obligations of private law and the liability to be sued to have those obligations enforced. Four practical reasons for distinguishing between those agreements (indeed of all rights) that are unenforceable and those that are a nullity, sufficient even

21 Leigh & Sillivan Ltd v Aliakmon Shipping Co Ltd (The Aliakmon) [1986] AC 785 (HL); Robins Dry Dock v Flint 275 US 302 (1927) (Holmes J). 22 Hunter v Canary Wharf Ltd [1997] AC 655 (HL). 23 Best v Samuel Fox & Co Ltd [1952] AC 716 (HL) 734. 24 For arguments supporting the Act, see Law Commission, Privity of Contract: Contracts for the Benefit of Third Parties (Law Com No 242, 1996). For criticism, R Stevens, ‘The Contracts Rights of Third Parties Act 1999’ (2004) 120 LQR 292.

Binding Our Future Selves  23 to satisfy Oliver Wendell Holmes’ bad man only interested in whether he will be subject to a sanction,25 may be given. First, as an unenforceable legal obligation is not a nullity, its performance provides a good justifying reason why it is rendered and so cannot be recovered. As Lord Mansfield said in Moses v Macferlan26 of the action for money had and received: This kind of equitable action, to recover back money, which ought not in justice to be kept, is very beneficial, and therefore much encouraged. It lies for money which, ex aequo et bono, the defendant ought to refund; it does not lie for money paid by the plaintiff, which is claimed of him as payable in point of honour and honesty, although it could not have been recovered from him by any course of law; as in payment of a debt barred by the Statute of Limitations, or contracted during his infancy, or to the extent of principal and legal interest upon an usurious contract, or, for money fairly lost at play: because in all these cases, the defendant may retain it with a safe conscience, though by positive law he was barred from recovering.27

Second, if we ignore the distinction between liability and obligation, contractual exclusions of liability, which constitute a defence, would be incoherent. If legal obligations were synonymous with legal liabilities, a contract that purported to undertake obligations to another whilst excluding legal liability for noncompliance would make no sense. On that view, a contracting party could not in the same breath say ‘I undertake to deliver the goods to you on Friday, no liability for late delivery is accepted.’ Just as the parties may make their agreement wholly unenforceable (‘Subject to Contract’), there is nothing mysterious about their agreeing to make it partially unenforceable, civil recourse not being a defining feature of what it is to undertake a contractual obligation. The important distinction between terms that define the parties’ obligations and those that exclude or limit their liability for breach is reflected in legislation that controls the latter but not the former.28 Third, it should not matter for the tort of procuring breach of contract, nor for the action of causing loss by unlawful means where such means constitute a breach of contract, that the contract that is breached is unenforceable. That the contracting party may exceptionally not be liable (because, for example, the contract was not in a requisite form) does not mean that its breach was non-wrongful, and so the promisor’s immunity from contractual suit should not affect the non-contractual actions for loss suffered. Fourth, the continuing existence of the underlying legal obligations can be seen where the parties have contractually excluded or limited their liability for damages. This is nicely illustrated by the decision of the Court of Appeal in AB v CD.29 The defendants were the owners of intellectual property rights for an

25 OW

Holmes, ‘The Path of Law’ (1897) 10 Harvard Law Review 457. v Macferlan (1760) 2 Burr 1005, 97 ER 676. 27 ibid 680–81. 28 Misrepresentation Act 1967, s 3. 29 AB v CD [2014] EWCA Civ 229, [2015] 1 WLR 771. 26 Moses

24  Robert Stevens Internet-based platform for the provision of goods and services relating to the mining business. They granted the claimants a licence to market this ‘eMarketplace’ in the Middle East. The defendants purported to terminate the contract, and the claimants sought an interim injunction to restrain them from doing so. At first instance Stuart-Smith J refused to grant an injunction because the claimants had an adequate remedy in damages. The Court of Appeal allowed the appeal. It did so on the basis that the licence agreement excluded liability for loss of profits. As a result of this clause, the defendants were, to that extent, immune from a suit to enforce their obligation to pay damages reflecting the actual harm their breach caused. As Underhill LJ stated: The primary commercial expectation must be that the parties will perform their obligations. The expectations created (indeed given contractual force) by an exclusion or limitation clause are expectations about what damages will be recoverable in the event of breach; but that is not the same thing.30

The effect of the clause was that a court’s order of damages ceased to be an adequate remedy, as no longer reflecting the legal obligation to pay compensation. As a result, the Court ordered an injunction that it would, absent the exclusion or limitation, have refused.

V.  Formalities, Nullity and Unenforceability The range of reasons why the law may conclude that it is inappropriate to enforce contractual obligations is different from, and much wider than, the range of reasons why it may conclude that there are no obligations at all. So, for example, limitation periods are created for reasons of policy and enable all parties to close the books on when they will be sued. Lapse of time cannot on its own nullify an agreement, but it may justify non-enforcement for its breach. Similarly, legal systems sometimes (but increasingly rarely) require certain formal steps to be taken before some kinds of agreements will be enforced. A requirement of form can be said to serve a number of goals. First, it provides evidence of the existence of the agreement. Second, it provides evidence of the content of its terms. Third, it demonstrates that the agreement was intended to be serious. Fourth, it helps to prevent, although it cannot eradicate, rash and foolish agreements from being enforceable, as the feckless promisor must go through the formalised process and is thereby given a period for reflection. The categories of agreement to which formality requirements applied under the Statute of Frauds 1677 were contracts for the sale of land, contracts for the sale of goods above a certain amount, contracts that cannot be completed in less than one year and contracts of guarantee. In England, the Statute of Frauds has

30 ibid

[28].

Binding Our Future Selves  25 been largely repealed, save in the case of guarantees.31 It is possible to prove the existence of an agreement without satisfying any formality requirement; such rules appropriately go to the enforceability and not the validity of the rights created (‘Noe Action shall be brought …’). This may be contrasted with the law of wills. There is no such thing as an informal will. If a document does not fulfil the formal requirements for a will, it is not a will. In that context, formality goes to validity, not enforceability. An informal guarantee may not be an enforceable contract but it is still a valid agreement. Although formality requirements for contracts for the sale of land have been retained, today in England they inappropriately render the contract wholly invalid, and not merely unenforceable, when not complied with.32 Whether this legislation is to be taken seriously, so that fully executed but informal contracts for the sale of land can be unwound without more, remains unclear.33 Other hurdles that an agreement must cross before it becomes a contract include certainty, capacity and intention to create legal relations. All, in principle, go towards the enforceability, not the existence, of the agreement.

VI. Consideration Today, formality requirements are rare and so place little constraint upon the parties’ power to contract. The doctrine of consideration is universally34 applicable to all contracts at common law, and so places at least a theoretical constraint upon our being able to bind our future selves. It still has the potential to trip up the unwary, and this gives rise to the difficult question of whether lack of consideration results in the legal nullity or unenforceability of the agreement. Hobhouse J, one of the finest common lawyers of the twentieth century, said that ‘ultimately the question of consideration is a formality, as is the use of a seal’.35 Lord Mansfield thought that consideration ought to be dispensed with, at least in commercial cases, where a promise was evidenced in writing, which, whilst not adopted,36 has been proposed as a basis for reform.37 Again, this sees consideration as fulfilling the same sort of function as a requirement of writing. If it is a formality requirement, it should render the agreement merely unenforceable.

31 Law Reform (Enforcement of Contracts) Act 1954. 32 Law of Property (Miscellaneous Provisions) Act 1989, s 2. A result of Law Commission, Formalities for Contracts for Sale of Land (Law Comm No 164, 1987). The Report contains no discussion of the substantive difference between an invalid and an unenforceable agreement. 33 See Sharma v Simposh Ltd [2011] EWCA Civ 1383, [2012] 3 WLR 503. 34 Or nearly. The doctrine never seems to have applied to letters of credit, possibly because any bank that took the point would instantly destroy its reputation and hence its business. 35 Vantage Navigation Corporation v Suhail & Saud Bahwan Building Materials Plc (The Alev) [1989] 1 Lloyd’s Rep 138, 147 (Hobhouse J). 36 Pillans v Van Mieron (1765) 3 Burr 1663 rejected Rann v Hughes (1778) 7 TR 350 n, 4 Bro PC 27. 37 Law Revision Committee, Sixth Interim Report (Cmnd 5449, 1937).

26  Robert Stevens Seen as a requirement of form, consideration appears to make little sense, and this may be one source of its current unpopularity. It is very poor at fulfilling any of the goals a requirement of form seems to fulfil. First, consideration comes from the wrong party. Consideration comes from the promisee and is consequently of poor probative force in showing what the promisor intended. This is particularly so where the consideration provided is subsequent to the promise, as it will be in the case of a unilateral contract. How did Mrs Carlill’s use of the smoke ball provide any evidence of what the Carbolic Smoke Ball company intended? Consideration may, in some circumstances, provide evidence of what the promisee thought was being promised and how seriously she thought it was intended, but that is all. If I can show that I have provided you with consideration, this does not provide evidence of what has been promised in return. Second, an oral exchange of promises, whilst enforceable, is no easier to prove than a gratuitous promise. If one party denies that he has made a promise, how does the existence of an oral counter-promise make it easier to prove? Third, it is sometimes claimed that gratuitous promises that are intended to be legally binding are inherently unlikely. It is, however, doubtful whether this is so. Even if we did think that such promises are unlikely, that is not a sufficient reason for refusing to enforce them when they can be genuinely proven by other means. If consideration merely provides evidence of seriousness of intent, why are gratuitous promises not enforceable where the promisor concedes that at the time of making the promise he intended to be bound, or where such intent is proven by other means?38 Fourth and conversely, where consideration can be established this provides little or no evidence of an intention to be legally bound. Every day most of us enter into informal agreements that are not intended to be enforceable but where we have promised something capable of being good consideration (‘I’ll wash the car if you mow the lawn’). Mowing lawns, washing cars and the promises to do so are all capable of being good consideration, but agreements of this kind are not intended to be legally binding. Most agreements that we enter into in our daily lives as a matter of fact are of this kind. A requirement of intent to be legally bound may assist in preventing the enforcement of rash promises but consideration certainly does not (‘I’ll pay you half my lottery winnings if you do the washing up’). The requirement of consideration provides no breathing space or protection for the rash and foolish promisor. Fifth if consideration is a requirement of form, why are certain categories of contract required to be put into the form of a deed, such as a lease for more than three years. Why does the formality of consideration not already suffice? Finally, it is inherently unlikely that consideration was introduced as a requirement of form. A formal requirement exists for reasons of policy. So where

38 cf

Rann v Hughes (n 36).

Binding Our Future Selves  27 someone fails to comply with a formal requirement, such as in drawing up a will, our refusal to give effect to the testator’s demonstrable intent is justified on the basis not that this is what is fair so far as the testator and his intended beneficiaries are concerned, but rather on the basis that it will encourage future testators to go through the requisite forms. Pour encourage les autres we sacrifice the intended bequests of some testators for the common good. Just as when we cut out a cancer we sacrifice some of the surrounding tissue to ensure that we have caught every case. For the price of injustice to some, we are trying to eradicate a wider evil. Formality rules are not concerned with what is fair and just as between two individuals. They need to be publicised so that as few of the unwary as possible will be caught out. Creation of formal requirements, such as that contained in the Statute of Frauds, is the province of legislatures not courts. Courts notoriously have a problem with giving blanket effect to requirements of form, as is reflected in their bypassing them with devices such as the secret and half-secret trust, where the formal requirements will lead to injustice for particular individuals. Consideration in contract law is a judge-made rule, and judges are concerned with the justice as between the two parties to the dispute, not with a wider policy agenda. The alternative, and preferable, explanation is that consideration is a substantive not a formal requirement. It provides an important reason in addition to the agreement for the creation of a contractual obligation. If A lends B $100, which it is agreed that B will repay on 1 January, one reason B is obliged to pay $100 to A on that date is that he has agreed to do so. In addition, there is the fact that B will be $100 better off and A $100 worse off if B does not keep his promise. This provides a further reason of justice why B is so obliged. A is more deserving than someone who has been gratuitously promised $100. It may be objected that this is simply a restatement of the rule rather than a justification for it. The truth, however, is that the force behind the doctrine of consideration is axiomatic, and that the search for further policy justification or function for the rule confuses more than it enlightens. Consideration is just one of the reasons going towards what is fair as between the two parties, which also counts in favour of the law’s providing civil recourse. Although legal systems outside of the common law do not have a doctrine of consideration, it is not an irrational choice, especially when we consider the fragility of the reasons justifying enforcing agreements. Despite consideration’s not being a formality requirement, agreements that are unsupported by consideration are not a nullity. The better view is that they are merely unenforceable. If we see consideration as a substantive requirement of enforceability, its rules become more defensible. Five examples follow.

A.  Unilateral Contracts If consideration is seen as a requirement of form, it is very difficult to justify the enforcement of unilateral contracts. If A promises B that he will pay B $100

28  Robert Stevens if B walks to York, B’s subsequent walking to York is of little probative force in establishing what A originally intended by the promise. Performance is both too late to evidence what was intended at the time the promise was made and from the wrong party to serve a formal function.39 The inconsistency with the enforcement of unilateral contracts and the formal requirement of consideration has led Professor Steve Smith to conclude that there is no consideration in the case of unilateral contracts. The courts do not agree and have consistently found consideration in the context of unilateral contracts.40 If, however, consideration is a requirement of substance, unilateral contracts are a core example of where consideration is shown. If the walk to York or use of a smoke ball is in exchange for the promise, it provides an additional reason for the enforcement of the promise just as much as if it had been promised in advance.

B.  Past Consideration By contrast with unilateral contract, the conferral of a benefit before the making of a promise may provide excellent evidence of both the existence and the seriousness of a promise. If B cares for A for 20 years, A’s subsequent promise of a house is readily explicable and probably seriously meant. If consideration were a requirement of form, there appears to be no reason why past consideration should not suffice – indeed it would be a central case where consideration made sense. The rule that past consideration does not count is readily defensible if consideration is a requirement of substance. If the benefit conferred or reliance incurred is prior to the making of the promise, it cannot be said to be caused by the promise. It does not, therefore, provide an additional reason for the enforcement of the promise. B cannot say, ‘I would not have cared for you without the promise.’ B’s motivation for the care cannot have been the promise. The different results in relation to past consideration and unilateral contracts also demonstrate the primacy of the promise as the source of the obligation. The consideration must be as a result of and reinforce the promise, not the other way around.

C.  Third Parties If consideration merely plays a formal role in evidencing the promisor’s intentions, it should not matter whether the consideration comes from the party seeking to enforce the contract or not. This was the reasoning of the Law Commission in



39 cf

40 eg

S Smith, Contract Theory (OUP 2004) 219. Carlill v Carbolic Smoke Ball Company (n 7).

Binding Our Future Selves  29 its Consultation on the doctrine of privity of contract. If, therefore, a promisee provided consideration, it would be no objection that the third party had failed to do so. Similarly, where there are joint promisees, if consideration is a formal requirement, both should be able to enforce, even though only one of them provided consideration. If, however, consideration is a substantive requirement, it should be essential that those who wish to enforce contractual rights have provided consideration. If the consideration has come from someone else, the party who wishes to enforce the promise is missing an essential element of a contractual right. The Contracts (Rights of Third Parties) Act 1999 has not abolished the rule that consideration must move from the party seeking to enforce the contract but merely created an exception to it. If the promisee does not provide consideration no enforceable contract will come into existence, even if another party in reliance upon the promise has suffered a detriment or conferred a benefit upon the promisor. So if I promise you $100 if someone finds my cat, and a third party relying upon the promise returns the straying feline, the promisee has no claim because he has provided no consideration, whilst the cat finder has no contractual right because he has been promised nothing.

D.  ‘In Exchange for the Promise’ That the consideration provided must be at the request of the promisor or ‘in exchange for’ the promise reinforces the substantive nature of consideration. An example: A promises to pay B $50 the following week. B, delighted by A’s generosity, cleans the windows of A’s house by way of thanks, but without having been requested to do so by A. If B seeks the enforcement of the promise to pay, is his claim as strong as that of someone who was requested to clean the windows for a price of $50? Has A benefited from B’s work? A can assert that he never wished to have his windows cleaned. He may prefer dirty windows, or clean his windows himself. He can, therefore, deny that he is any better off as a result of B’s work. If, however, he had requested that the work be done, he cannot deny that the work is of benefit to him. This is so regardless of how idiosyncratic A’s request may have been. An accountant might state that A’s balance-sheet position is not improved by the cleaning of the windows if he would not have otherwise paid for the work to be done, but the law takes a broader view of benefit. Chocolates wrappers or long-distance walks, if given in exchange for a promise, are a benefit to the promisor if that is what he bargained for. A promisor can deny that unrequested or unbargained-for goods or services are of no benefit to him. The claim that ‘consideration need not be of benefit to the promisor’ takes an unnecessarily narrow view of benefit. Many of the benefits we bargain for leave us in no better financial position, but we cannot deny that we are benefited by those things we have bargained for. B’s claim is substantively more deserving where the work was done at the request of A than where it is not.

30  Robert Stevens

E.  Equity will not Assist a Volunteer That provision of consideration is not a requirement of form is also shown by its impact upon the remedies we have and its application to the bona fide purchaser defence. So, specific performance, ordering the performance of a primary obligation other than that to pay a sum of money, is only available if the promise has provided consideration for the promise made to him. Here consideration is given a narrower meaning, so that a promise alone is insufficient. Promises contained in bare deeds cannot be specifically enforced therefore. Further, in applying the defence of bona fide purchase, a purchaser is treated differently from someone who has merely received a misdirected asset. Why? Because he has provided consideration. He is in a morally significant different position

VII. Variation If consideration were serving a formal function, evidencing the parties’ seriousness of intent, it may be argued that once they have demonstrated that they intend their relationship to be given legal effect, they should not be required to do so again when they come to vary the deal. Once we have shown that the parties do intend their relationship to have legal effect, there seems little point in requiring them to show it again when it comes to changing the terms of what has been agreed. Why not therefore give effect to variations that are not supported by consideration? Viewed as a requirement of substance, however, if a promise is to be enforced then a contractual variation would need to be supported by consideration in the same way as any other. So, if I promise to fulfil a duty I am already under, this should be good consideration where the duty is owed to someone other than the promisee. The promise provides an additional reason that constitutes a detriment to me, which did not exist prior to its being made. Similarly where I am already under a duty to the promisee that is not contractual, for example my duty to pay tax to HM Revenue and Customs on the pitifully small sum of money I am paid, a promise to fulfil this duty should be capable of being good consideration, as it provides an additional reason, over and above my obligation to pay tax, why I am bound. In principle, therefore, a promise to do what I am already obliged to do can be good consideration. However, where I promise to do what I have already promised to do, at that point I confer no further benefit. After the second promise, the promisee finds himself in exactly the same position as he was in before. He just has a promise. So if I promise my contractual creditor to pay him money I already owe, or if I promise my employer to do work I am already obliged to carry out, at that point I confer no further benefit upon the creditor or employer. Debtors who promise to pay

Binding Our Future Selves  31 what they already owe, or sailors who promise to sail ships when they are already contractually obliged to do so, confer no benefit by so doing. However, the same is not necessarily true of the performance of a promise I am already contractually obliged to perform. If I owe you $1,000, and you promise to release me if I pay half, if I do so pay when I would not have paid without your agreeing to release me, you have received a benefit in exchange for the promise of release. A bird in the hand is worth more than one in the bush. The performance of a promise is worth more than the promise alone. Famously, the decision of the House of Lords in Foakes v Beer stands in the way of concluding that part payment can constitute good consideration. In principle, however, it should be capable of doing so if the payment would not otherwise have been received. If, for example, it could be demonstrated that but for the agreement to accept part payment the debtor would have gone into bankruptcy and the creditor have received nothing but a dividend, a factual benefit has been conferred. An agreement to accept less cannot in principle, alone, constitute good consideration even after part payment has been made, if such payment would have been received in any event. What would have happened if the agreement to accept less had never been entered into was never pleaded or proven in Foakes. Where what is promised is the provision of a service the promisor is already obliged to perform, it was established in Williams v Roffey that where this confers a ‘practical benefit’ upon the promisee, the agreement to pay more for work the carpenter was already obliged to perform was enforceable. What exactly this benefit constituted was not authoritatively considered by the Court of Appeal, because it was conceded by counsel that there was such a benefit. Again, however, at the moment the variation was agreed, the employer had received nothing more than he already had: a promise to work. If the work would not have been carried out but for the variation, consideration has been provided once such work is done. It is easier for the promisor to establish that he would not otherwise have done the work than it is for the debtor to show that he would not have paid, because promises to provide services are not capable of being specifically enforced, unlike a promise to pay a sum of money. That a promise of something may not be good consideration but its performance can, may be counter-intuitive but is commonplace. So, for example, a promise to negotiate may have no contractual force in England because it is too uncertain, but performing it by actually carrying out negotiations can. The developments in Williams v Roffey make a principled defence of consideration easier, not harder, to make.

VIII.  Increasing and Decreasing Pacts The conclusion to be drawn from the foregoing is that there are two necessary substantive conditions of a party’s having an enforceable contractual right.

32  Robert Stevens First, there must be an agreement. Second, consideration must have been provided. If, therefore, A agrees to provide B with a service in exchange for $100,000, if B subsequently agrees to pay $120,000 for the same work, A should not without more acquire an enforceable contractual right to be paid that greater sum if the work is done (unless the doctrine of consideration is to be abolished altogether, which is a defensible choice for the law to make). If you wish to increase such contractual rights as there are, consideration should be required. It does not follow, however, that if A agrees to be paid less ($80,000) for the same work, that does not correspondingly reduce the contractual right that he has. Remember that in order to possess an enforceable contractual right two necessary independent conditions need to be satisfied (agreement and consideration). Although there was once an agreement between the parties that A would be paid $100,000, after the fresh offer and acceptance between them that is no longer true. The subsequent agreement between the two of them means that the first condition of there being an enforceable contractual right to be paid $100,000 (that there is an extant agreement between the two of them to that effect) is unsatisfied. The fact of what the parties have agreed can always be changed by them. That there was once in the past before the time comes for a court judgment an agreement on different terms between the parties is no reason for giving effect to that agreement when that is no longer true. In order to reduce the enforceable contractual obligations they are under (by agreement) there should therefore be no need for the parties to provide fresh consideration. All they need to do is change their agreement. Consideration is a substantive requirement going to the creation of contractual rights, its existence is not a condition of there being no contractual rights. If you wish to decrease such contractual rights as there are, an agreement to that effect should suffice, as the law should only be in the business of enforcing contracts to the extent that there is an extant agreement between the parties to that effect. Seeing the agreement and the contract as separate enables us to understand why this is so. Increasing and decreasing pacts must be distinguished.41 It will be recalled that Rock Advertising Ltd v MWB Business Exchange Ltd concerned an agreement to reduce a contractual licence fee. There should be no need to prove fresh consideration in such a case. That there was, at some point in the past, an agreement on different terms should not suffice for the enforcement of that earlier agreement when it has ceased to exist at the time of judgment. There should have been no need to search for fresh consideration to be provided, as the Court of Appeal wrongly supposed, and the danger is that the doctrine of consideration will be degraded by a search for speculative benefits that may accrue to the promisor, such that it ceases to be meaningful in cases where consideration ought



41 See

G Treitel, Some Landmarks of Twentieth Century Contract Law (OUP 2002) 11.

Binding Our Future Selves  33 genuinely to be required. Destruction of contractual rights (by agreement alone) should be easier than their creation (by contract). This analysis provides an alternative, and superior, explanation for the result Denning J wished to reach in Central London Property Trust Ltd v High Trees House Ltd.42 An agreement to take a lower rent should have sufficed to reduce the landlord’s contractual entitlement, without recourse to any notion akin to estoppel, even where the new agreement had not yet been acted upon.43

IX. Conclusion The aforementioned distinction between increasing and decreasing pacts does not lead to a different result in the leading decision of the House of Lords in Foakes v Beer. There, it will be recalled, Mrs Beer had obtained a judgment in her favour against Dr Foakes of £2090 19s. Her entitlement to be paid was not, therefore, any longer solely based upon Dr Foakes’ promise to pay. As her right to payment was, after judgment, independent of any agreement between the two of them, a change in the agreement could not suffice. A contract, with its condition of consideration, was required to vary the judgment debt.44 The mere promise by Dr Foakes to do what he had already promised gave Mrs Beer nothing, and so did not suffice. We should have the power to unbind ourselves from contracts, whether wholly or in part, through agreement alone, without the need for consideration. If the parties’ agreement is no longer what it once was, at the time of the court judgment one of the two necessary conditions of an enforceable contract on those earlier terms is missing. One of the reasons why the entire doctrine of consideration has a bad reputation in England is that our leading ultimate appellate court case on the doctrine has been misunderstood as having a wider ratio than it in fact has. In cases of variation of a contract there should be no need to search for fresh consideration, save in the case of increasing pacts. The performance of a promise already made should also suffice as consideration for purposes of variation where it would not have been rendered absent the subsequent agreement, and this should be true regardless of whether the performance rendered is a payment, a service, the delivery of goods or anything else.

42 Central London Property Trust Ltd v High Trees House Ltd [1947] KB 130. 43 For criticism of the expansive notion of ‘estoppel’ employed, see R Stevens, ‘Not Waiving but Drowning’ in A Dyson, K Goudkamp and F Wilmot-Smith (eds), Defences in Contract (Hart Publishing 2017) 125. 44 See also Re Selectmove [1995] 1 WLR 474, which concerned an agreement to reduce tax liability, and Collier v P & M J Wright (Holdings) Ltd [2007] EWCA Civ 1329, which concerned an agreement to vary a consent order to pay a judgment debt.

34  Robert Stevens One problem with the unsupportable decision of the Supreme Court in Rock Advertising Ltd v MWB Business Exchange Centres Ltd45 was that the court’s strange conclusion that contracting parties are able to (partially?) limit the power of their future selves to contract in the ordinary way, meant that the scope of Foakes v Beer did not arise for decision. Lord Sumption was at least correct in describing the rule that it is usually taken as standing for as ‘ripe for re-examination’.46 Unbinding our current selves should be easier than binding our future selves.



45 See

A Burrows, ‘Anti-Oral Variation Clauses: Rock-Solid or Rocky?’, ch 3 in this volume. Advertising (n 2) [18].

46 Rock

3 Anti-Oral Variation Clauses: Rock-Solid or Rocky? ANDREW BURROWS*

I. Introduction Since the turn of the century, there has been uncertainty in this jurisdiction as to the validity of ‘no-oral-modification’ (NOM) clauses, otherwise known as antioral variation clauses. That uncertainty was essentially removed by the decision of the Supreme Court in 2018 in MWB Business Exchange Centres Ltd v Rock Advertising Ltd.1 But two significantly different lines of reasoning were adopted by the Supreme Court: that taken by Lord Sumption, which was agreed with by three other Justices (Lady Hale, Lord Wilson and Lord Lloyd-Jones); and that taken by Lord Briggs. Examining and assessing those two lines of reasoning is not only of great interest in its own right but also helps us to understand fully the issues involved. Moreover, as a practical matter, courts in other common law jurisdictions will need to decide whether to follow the Supreme Court or not. A third significantly different approach will also be considered: that favoured by the Court of Appeal in Rock,2 in which the judgment on this issue was given by Kitchin LJ (with whom Arden and McCombe LJJ agreed). The essential reasoning of Kitchin LJ was to apply the reasoning in obiter dicta of the Court of Appeal in Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd.3 The obiter dicta were * I am grateful to Paul Davies and Magda Raczynska for inviting me to give a paper at the conference that was the basis for this book; and to the participants at the conference for their helpful comments. At the time of completion of this essay, I was Professor of the Law of England and a Fellow of All Souls College, Oxford. I became a justice of the Supreme Court of the UK on 2 June 2020. 1 MWB Business Exchange Centres Ltd v Rock Advertising Ltd [2018] UKSC 24, [2018] 2 WLR 1603. See the excellent case-notes by PS Davies, ‘Varying Contracts in the Supreme Court’ (2018) 77 CLJ 464 and by R Calnan, ‘Contractual Variation Clauses’ (2018) 33(8) Journal of International Banking and Financial Law 487. 2 MWB Business Exchange Centres Ltd v Rock Advertising Ltd (Rev 1) [2016] EWCA Civ 553, [2017] QB 604 [19]–[36]. 3 Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd [2016] EWCA Civ 396, [2017] 1 All ER (Comm) 601. Having described in detail, [2016] EWCA Civ 553, [2017] QB 604 [22]–[29], the reasoning of Beatson LJ, Kitchin LJ said, at [34], ‘I respectfully agree with Beatson LJ that the decision

36  Andrew Burrows largely those of Lord Justice Beatson, with whom Underhill and Moore-Bick LJJ agreed. I shall therefore refer to the third view as that of Lord Justice Beatson. The argument of the chapter is that although the issues are difficult and not clear-cut, and there are powerful arguments that can be, and have been, put in favour of each of the three different approaches, there is much to be said for preferring the reasoning of Lord Justice Beatson as a matter both of principle and policy. So as to have in mind an immediate picture of the differences between the three approaches, let us imagine a scale representing a judicial willingness to allow the parties by agreement to depart from a NOM clause. At one end there is Lord Sumption, taking the view that a NOM clause cannot be departed from. In the middle is Lord Briggs, taking the view that a NOM clause can sometimes be departed from but only where it is clear that the parties had the NOM clause in mind and intended to depart from it. At the other end is Lord Justice Beatson, taking the view that the parties can always depart from a NOM clause whether they had the clause in mind or not. Again, if one were to compare the three different approaches on a scale representing whether a NOM clause is effective, we see the following. At one end is Lord  Sumption, with his view that a NOM clause is effective and cannot be departed from by subsequent agreement. In the middle is Lord Briggs, for whom a NOM clause can be effective but may sometimes be departed from by subsequent agreement. At the other end is Lord Beatson, for whom a NOM clause is ineffective and can always be departed from by subsequent agreement.

II.  MWB Business Exchange Centres Ltd v Rock Advertising Ltd: Facts and Decision The facts of the Rock case provide a simple and clear illustration of what we are talking about here. MWB entered into a contract with Rock by which it licensed out to Rock office space in central London for 12 months. Clause 7.6 of the contract provided: This Licence sets out all of the terms as agreed between MWB and Licensee. No other representations or terms shall apply or form part of this Licence. All variations to this Licence must be agreed, set out in writing and signed on behalf of both parties before they take effect.

Rock (whose managing director was Mr Idehen) fell into arrears (in excess of £12,000) in paying the agreed monthly licence fee to MWB. MWB’s of this court in World Online Telecom Ltd v I-Way Ltd [2002] EWCA Civ 413 was correct and should be followed for the reasons he gave’.

Anti-Oral Variation Clauses  37 credit-controller, Ms Evans, an employee acting with apparent authority,4 orally agreed with Mr Idehen to allow Rock to stay in the premises, with the arrears to be paid off over the course of a year under a revised schedule of payments. This would be less advantageous to MWB because the payments were initially to be below the monthly licence fee, albeit that, later in the year, they would be above the monthly fee. Rock made the first revised payment on the day of the oral agreement (27 February). But two days later (on 29 February), having spoken to her finance director, Ms Evans sent an email to Mr Idehen in which she withdrew her agreement to a revised payment schedule. A month later, MWB gave notice of termination of the licence and locked Rock out of the premises. The essential question was whether MWB was entitled to do so. In the Court of Appeal, three issues of law were considered. First, was the oral agreement binding given that, contrary to the NOM clause 7.6, it was not set out in writing and signed? Second, was the oral agreement supported by consideration? Third, did promissory estoppel operate here so that MWB could not go back on its promise? The Court of Appeal decided that, first, it did not matter that the agreement was oral and did not comply with the NOM clause: the view of the Court of Appeal on such clauses in Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd was approved. Second, the agreement was supported by consideration because MWB was obtaining a practical benefit. The oral variation agreement was therefore binding and MWB could not depart from it. It followed that the third issue did not arise. However, the Court of Appeal’s view in obiter dicta was that, had the agreement not been binding, promissory estoppel would not here have assisted Rock because, on these facts, it was not inequitable for MWB to revert to its legal rights under the original contract. More specifically, this was because Rock had suffered no detriment by making the first payment of the revised schedule; and, moreover, it was only two days after the oral agreement was made that MWB, through the email of 29 February sent by Ms Evans, had given notice that it was withdrawing that agreement, so that promissory estoppel here merely suspended MWB’s rights for a short period of time and did not operate to extinguish those rights. MWB appealed to the Supreme Court on the first two issues. Allowing the appeal, the Supreme Court held, on the first issue, that the NOM clause prevented the oral agreement’s being binding.5 Unfortunately, for the development of the law, there was therefore no need to consider the second ‘consideration’ issue.

4 Kitchin LJ, Rock (CA) (n 2) said the following at [8]: ‘The judge heard evidence from Miss Evans and Mr Idehen. He concluded that Miss Evans did agree to accept the terms and schedule proposed by Mr Idehen. Moreover, she had at least ostensible authority to commit MWB to an agreement of this kind.’ 5 Although the precise legal effect was not discussed – and Lord Sumption used the language of ‘ineffective’ at [4] and ‘invalid’ at [15] and [17] – it would seem that, using the standard language of contract law, the oral variation should be regarded, strictly speaking, as ‘void’.

38  Andrew Burrows

III.  The Reasoning of Lord Sumption (for the Majority) The decision of Lord Sumption was that the NOM clause was effective and could not be departed from. Although not expressed in quite this way, his Lordship’s reasons can be set out in the following seven points: (a) Respecting the parties’ autonomy in this context means that the parties’ intentions, as expressed in a NOM clause in the contract, should not be overridden. (b) Just as statutes can and do prescribe that certain types of contract shall be in a certain form,6 which cannot be departed from, so the parties should be able to achieve the same by a term of a contract. (c) There are at least three good commercial reasons why parties include NOM clauses in contracts and why they should be upheld. The first is that they prevent written agreements from being undermined informally, which can give rise to abuse, for example in raising defences to summary judgment. Second, they avoid disputes about whether a variation was intended and its precise terms. Third, such a clause stops those without authority binding a company to an informal variation of the contract. (d) There is a close analogy between entire agreement clauses, which are routinely upheld, and NOM clauses. Both promote contractual certainty as to the terms agreed by rendering invalid what the parties orally agreed outside the contract in question. (e) The parties to a NOM clause are not locking themselves into a contract, without the ability to change their minds, because it is an easy matter for them to comply with a NOM clause. All they have to do is to ensure that the variation complies with the specified formalities (eg in Rock, by putting the variation in writing and signing it). (f) An amendment that does not comply with a NOM clause may be binding to some extent by reason of an estoppel. Although Lord Sumption did not say what type of estoppel he had in mind, presumably he was principally thinking of promissory estoppel, at least where the relevant oral modification – as on the facts of this case – involved the promisor’s giving up rights (eg accepting a reduced or late payment) so that the estoppel would be invoked as a defence and not as a cause of action.7 But it may be that estoppel by convention was 6 For example, by s 2 of the Law of Property (Miscellaneous Provisions) Act 1989, subject to exceptions, a contract for the sale or other disposition of an interest in land is void unless it is made in writing and signed; by s 4 of the Statute of Frauds 1677, a contract of guarantee is unenforceable unless it is evidenced in writing and signed. 7 It is trite law that promissory estoppel can only be used as a defence not a cause of action: see, eg, Combe v Combe [1951] 2 KB 215 (CA); Baird Textiles Holdings Ltd v Marks and Spencer plc [2001] EWCA Civ 274, [2002] 1 All ER (Comm) 737. In practice, this means that it applies to a promise to give up an existing right but not to a promise to confer a new right.

Anti-Oral Variation Clauses  39 also in mind (which would apply where both parties have assumed that the oral agreement was binding, provided it would be unfair for the relevant party to resile from that assumption).8 Whichever the type of estoppel, Lord Sumption emphasised that the estoppel exception would not, and should not, swallow up the basic rule that a NOM clause is effective. I return to consider the scope of the estoppel exception later in this chapter.9 (g) It is significant that a NOM clause is effective (subject to an estoppel exception) under the Vienna Convention on Contracts for the International Sale of Goods (CISG), Article 29(2), and the UNIDROIT Principles of International Commercial Contracts (2016), Article 2.1.18. The former provision (although note that the UK has still not ratified the CISG) reads as follows: A contract in writing which contains a provision requiring any modification or termination by agreement to be in writing may not be otherwise modified or terminated by agreement. However, a party may be precluded by his conduct from asserting such a provision to the extent that the other party has relied on that conduct.

The wording of the latter (which is not a legally binding instrument) is very similar: A contract in writing which contains a clause requiring any modification or termination by agreement to be in a particular form may not be otherwise modified or terminated. However, a party may be precluded by its conduct from asserting such a clause to the extent that the other party has reasonably acted in reliance on that conduct.

There is no doubt that, through these seven points, Lord Sumption set out a powerful case for the effectiveness of NOM clauses. However, as a matter of principle, the consequences are, to my mind, startling and should cause us, at the very least, to pause for thought.10 Let us assume, for example, that A and B enter into a contract that contains a clause preventing a variation unless in writing and signed by both parties. A and B subsequently agree that, despite their explicit recognition of that no-oral-modification clause, they are making an oral variation (or, indeed, a written but unsigned variation) and do so. According to Lord Sumption’s reasoning, their variation agreement – even though the parties have expressly made reference to their wish to depart from the NOM clause – is invalid. They cannot

8 The precise elements and scope of estoppel by convention are not easy to pin down. For important cases, see Amalgamated Investment and Property Co Ltd v Texas Commerce Int Bank Ltd [1982] QB 84 (CA); The Vistafjord [1988] 2 Lloyd’s Rep 343 (CA); Prime Sight Ltd v Lavarello [2013] UKPC 22, [2014] AC 436. 9 In section VI at (b). 10 One also wonders about the knock-on effect on clauses that are similar to, but distinct from, NOM clauses. Take a clause that prevents a variation of the contract unless agreed not only by the parties to the contract (eg the borrower and lender under a loan agreement) but also by a third party (eg a guarantor). Does the decision in Rock mean that the contracting parties cannot agree a variation of the loan agreement without the consent of the third party?

40  Andrew Burrows depart from the NOM clause and, unless they comply with the formality laid down, are locked into it.

IV.  The Reasoning of Lord Briggs Although coming to the same decision on the facts – that the oral variation agreement was invalid – Lord Briggs disagreed with the reasoning of Lord Sumption (and the majority). While his Lordship did not express himself in quite this way, my interpretation of his reasoning can be set out in the following points: (a) As a matter of principle, the relevant intentions of the parties are their latest intentions and not some earlier inconsistent intentions. Put another way, parties should always be free to change their minds by a variation that satisfies the legal requirements for a contract. The parties have, and should have, autonomy not only to bind but to release themselves. To deny this is to embrace a conceptual impossibility (or inconsistency). In Lord Briggs’ words: What is … conceptually impossible is for the parties to a contract to impose upon themselves … a scheme (of particular formalities], but not to be free, by unanimous further agreement, to vary or abandon it by any method, whether writing, spoken words or conduct, permitted by the general law.11

(b) It is incorrect to draw an analogy between NOM clauses and statutes imposing formalities. Those formalities are imposed by the legislature, as part of ‘the law of the land’,12 in respect of particular agreements for reasons of policy. Where the statutes apply, all contractual agreements (unless specified to the contrary) are caught by the provisions, including contractual variations. In other words, it would contradict the purpose of the statute for parties to be free to vary the contract without complying with the statutory formalities. But there is no law of the land or statutory policy – or, although Lord Briggs did not say this, no policy laid down by the common law13 – in play in respect of NOM clauses.

11 Rock (SC) (n 1) [26]. 12 ibid. 13 The common law might have imposed formalities for contracts but, as it happens, this is now entirely a matter of statute (although, of course, contracts may be made by deed and there is also a view that the doctrine of consideration has a formality role: see, eg, L Fuller, ‘Consideration as Form’ (1941) 41 Columbia Law Review 799). One might say that what Lord Sumption had in mind was that, just as statute can impose formalities, so there can be a newly recognised common law formality rule to the effect that, for modifications, the courts apply the formality agreed, initially, by the parties. But for reasons set out in detail in section VI, the better view is that principle and policy do not support the recognition of such a common law rule. Put another way, reasoning by analogy requires that the analogy is a good one, and here it is not a good one so that the analogy is an illegitimate one to draw. Note that, in any event, a common law rule that deferred to the formality initially agreed by the parties would

Anti-Oral Variation Clauses  41 (c) It is unhelpful to draw an analogy with entire agreement clauses. Those clauses are concerned to establish what the agreement is and to promote legal certainty in deciding that. They have nothing to do with the parties’ changing their minds having made an agreement. (d) However, Lord Briggs was not saying that a subsequent variation that complies with the standard requirements for a contract always overrides a NOM clause. It is central to his reasoning that the contracting parties can only depart from a NOM clause if it is clear that they had in mind the NOM clause and intended to depart from it. It is not enough merely that the parties have agreed a variation that contradicts the NOM clause, because, according to Lord Briggs, parties may commonly simply overlook or be ‘blissfully unaware’14 of a NOM clause. What is required is that the parties have expressly, or ‘by strictly necessary implication’,15 agreed to do away with the NOM clause. There is no strictly necessary implication that they meant to do away with the NOM clause merely because they have agreed a variation that contradicts the NOM clause. In taking this mid-position, Lord Briggs considered that there was a powerful analogy with the way in which the law treats negotiations that are ‘subject to contract’. Where parties agree to negotiate ‘subject to contract’, they will not be considered to have entered into a binding contract until they have made a formal written contract, or expressly or by necessary implication agreed to dispense with the ‘subject to contract’ restriction. Lord Briggs also stressed that his mid-position approach of sometimes allowing a departure from a NOM clause represents merely an incremental development from the traditional common law approach – which always allows a departure from a NOM clause – as applied, for example, in the United States, Australia and Canada.16 In contrast, he saw Lord Sumption’s approach as involving a ‘more radical solution’ that constitutes ‘a clean break’ with international common law.17 Lord Briggs has set out a powerful case for taking this mid-position. Although, in practice, it will be rare for his approach to reach different results from that of Lord Sumption, one might say that the approach of Lord Briggs is more principled. run into the fundamental problem that the parties might specify ludicrous formalities (for examples, see n 22). I would like to thank Justice Edelman of the High Court of Australia for discussion of this point. 14 Rock (SC) (n 1) [30]. 15 ibid [31]. 16 Lord Sumption at [8] had referred to the following as supporting the view at common law that the parties can always depart from a NOM clause: in the USA, Beatty v Guggenheim Exploration Co (1919) 225 NY 380, 387–88 (Cardozo J); in Australia, Liebe v Molloy (1906) 4 CLR 347 (High Court); Commonwealth v Crothall Hospital Services (Aust) Ltd (1981) 54 FLR 439, 447 et seq, and GEC Marconi Systems Pty Ltd v BHP Information Technology Pty Ltd (2003) 128 FCR 1; and in Canada, Shelanu Inc v Print Three Franchising Corpn (2003) 226 DLR (4th) 577, para 54 (Weiler JA) citing Colautti Construction Ltd v City of Ottawa (1984) 9 DLR (4th) 265 (CA) (Cory JA). 17 Rock (SC) (n 1) [32].

42  Andrew Burrows What it precisely means in practice can be illustrated by the following example. Let us assume that A and B enter into a contract that contains a clause preventing a variation unless in writing and signed by both parties. Without any reference to that NOM clause, and forgetting all about it, A and B subsequently agree an oral variation (or a written but unsigned variation). According to Lord Briggs’ reasoning, their variation agreement is invalid. They have not expressly (or by strictly necessary implication) agreed to depart from the NOM clause. In contrast, had they orally stated that they were making an oral variation, which they knew did not comply with the NOM clause, that variation would be valid.

V.  The Reasoning of Lord Justice Beatson in the Globe Case This view adheres to the point of principle, as set out at (a) under Lord Briggs’ reasoning, without going on to qualify that principle in the way that Lord Briggs did. This was at the heart of Beatson LJ’s reasoning in the Globe case.18 Although not cited by Beatson LJ (albeit that it was cited by Kitchin LJ in the Court of Appeal in Rock19 in approving Beatson LJ’s reasoning), the principle was perhaps most famously and fully articulated by Cardozo J in Beatty v Guggenheim Exploration Co: Those who make a contract, may unmake it. The clause which forbids a change, may be changed like any other. The prohibition of oral waiver may itself be waived … What is excluded by one act, is restored by another. You may put it out by the door; it is back through the window. Whenever two men contract, no limitation self-imposed can destroy their power to contract again …20

According to this approach, therefore, a NOM clause is ineffective. The parties can always depart from it by a variation that complies with the normal rules of contract variation and would be binding had there been no NOM clause. In contrast to Lord Briggs, it is irrelevant to go on to ask whether the parties had in mind the NOM clause and have expressly, or by strictly necessary implication, intended to depart from it. So had the Globe case not been decided on another ground, the reasoning of Lord  Justice Beatson would have led him (along with Moore-Bick and Underhill  LJJ) to the result that the NOM clause in that case was ineffective. Similarly, as shown by the Court of Appeal’s decision in the Rock case – where 18 Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd (n 3) [95]–[113]. In addition to the central issue of principle, he also considered some issues of policy (eg certainty and issues of proof in relation to false allegations of oral modifications), as well as referring in some detail to the inconsistent relevant cases on this issue in this jurisdiction. 19 Rock (CA) (n 2) [34]. 20 Beatty v Guggenheim Exploration Co (n 16) 387–88.

Anti-Oral Variation Clauses  43 Kitchin LJ applied Beatson LJ’s reasoning – if this approach had been applied to the facts in Rock, it would have led to the different decision that the oral variation was valid, provided there was consideration.

VI.  Which is the Best Approach? In assessing which is the best approach, one might at first sight think that Lord Justice Beatson is favouring principle at the expense of the good commercial reasons for NOM clauses, which have been set out in point (c) of Lord Sumption’s reasoning in section III. In other words, one might say that the underlying conflict here is between policy or pragmatism, on the one hand, and principle, on the other. And one might perhaps therefore expect that practitioners would tend to support Lord Sumption’s approach, while the academics might side with Lord Justice Beatson. However, the position is not as simple as that dichotomy would suggest. Indeed, Lord Sumption relied on distinguished academic writing in support of his approach.21 But the important point is that there are considerable commercial and practical advantages that flow from applying the principled approach. In other words, it is not the case that the commercial policy considerations are all against Lord Justice Beatson’s approach. On the contrary, one can articulate the following arguments of policy (set out in eight points) in support of the principled approach: (a) The law is rendered significantly more certain if one treats NOM clauses as ineffective. The simple legal rule would be that NOM clauses are ineffective. Parties and legal advisers would know that the parties are always free to vary a contract by the normal rules of variation irrespective of a NOM clause. Indeed, if that were the law, NOM clauses might wither away (although the continued use of exemption clauses that are invalid under the Unfair Contract Terms Act 1977 and the Consumer Rights Act 2015 suggests that, if parties find it commercially useful to continue to include such clauses – even though of no legal effect – they will do so). (b) The law as set out by Lord Sumption is, in practice, far from clear. This is because, despite a NOM clause, Lord Sumption accepted that such a clause applies only to stop variations and does not extend to stopping the application

21 Rock (SC) (n 1) [9], where he cited: J Morgan, ‘Contracting for Self-Denial: On Enforcing “No Oral Modification” Clauses’ (2017) 76 CLJ 589; E McKendrick, ‘The Legal Effect of an Anti-oral Variation Clause’ (2017) 32 Journal of International Banking Law and Regulation 439; J O’Sullivan, ‘Unconsidered Modifications’ (2017) 133 LQR 191. cf E Peel, Treitel on the Law of Contract, 14th edn (Sweet & Maxwell 2015) para 5-036, indicating (prior to Rock) that NOM clauses could be departed from: see also (prior to Rock) Chitty on Contracts, 32nd edn (Sweet & Maxwell 2015) para 22-045, fn 196, that ‘the better view would appear to be that it is possible for parties to waive compliance’. Prior to Rock in the Court of Appeal there had been very little discussion of this issue in the academic literature, but that was perhaps because the general view was that the parties could always depart from a NOM clause.

44  Andrew Burrows of promissory estoppel and perhaps other forms of estoppel.22 Very commonly, as on the facts of Rock itself, one will have arguments about whether an oral variation is binding to some extent under the doctrine of promissory estoppel (or estoppel by convention). In some less common situations, it might also be open to a party to argue that the old contract has been orally rescinded and a new contract has been entered into: and no one is suggesting that a NOM clause can stop that. The essential point, however, is this. The idea that NOM clauses assist summary judgment, by stopping spurious arguments about oral variation, seems over-optimistic. A party will commonly be able to resist summary judgment by putting forward an argument that the oral variation is binding under the doctrine of promissory estoppel (or estoppel by convention). Before moving on to the third point, it is important to explore further how Lord Sumption sought to deflect this fundamental point about the uncertainty of the estoppel exception. He said this: This is not the place to explore the circumstances in which a person can be estopped from relying on a contractual provision laying down conditions for the formal validity of a variation. The courts below rightly held that the minimal steps taken by Rock Advertising were not enough to support any estoppel defences. I would merely point out that the scope of estoppel cannot be so broad as to destroy the whole advantage of certainty for which the parties stipulated when they agreed upon terms including the No Oral Modification clause. At the very least, (i) there would have to be some words or conduct unequivocally representing that the variation was valid notwithstanding its informality; and (ii) something more would be required for this purpose than the informal promise itself …23

With respect, this does not solve the uncertainty problem. If we focus on promissory estoppel, the normal operation of promissory estoppel – which can be used as a defence but not as a cause of action24 – is grounded on three elements.25 First, there must be a clear and unequivocal promise or representation. Second, there must be reliance or, perhaps, detrimental reliance by the promisee. And third, the doctrine may sometimes be suspensory only, rather than extinctive, in which case the promisor can revert to its original rights by giving reasonable notice. The operation of the doctrine is notoriously factspecific. It is anything but a certain doctrine. To meet this, Lord Sumption

22 Another way of expressing the uncertainty is to think about the scope of the restriction on the parties’ powers to bind themselves by a NOM clause. So, for example, can the parties specify any type of formality for a variation (‘must be signed in invisible ink in a boat in the mid-Atlantic’, or ‘must be signed and witnessed by 50 bishops’)? And what about going beyond formalities for variations to clauses whereby the parties agree that there shall be no estoppels between them or no future agreements between them? I am grateful to Fred Wilmot-Smith for drawing my attention to these concerns over the scope of the Rock decision. See also n 10. 23 Rock (SC) (n 1) [16]. 24 See n 7. 25 See A Burrows, A Restatement of the English Law of Contract (OUP 2016) 74–77.

Anti-Oral Variation Clauses  45 appears to be suggesting a narrowing of the normal doctrine so that it applies only where the promisor makes clear that it is not merely waiving its substantive rights under the contract but is also waiving its procedural rights (ie the right to insist on a written modification) under the no-oral-modification clause. In other words, Lord Sumption appears to require the promisor to say something like, ‘I am promising to waive my rights under the contract including my right not to be bound because this promise is oral not written’. But the whole point of the estoppel exception is to avoid injustice to the party who has relied or detrimentally relied on the oral promise. To narrow it in the way suggested by Lord Sumption would, in nearly all cases (because it will be rare for the promisor to refer explicitly to the no- oral-modification clause), leave in play the injustice to the promisee that one is seeking to avoid. If one is to recognise an estoppel exception so as to avoid injustice, one should surely be applying the ordinary principles of promissory estoppel (or estoppel by convention). So Lord Sumption is caught on the horns of a dilemma. In reality he comes close to saying that there is no estoppel exception and that, because we want as a matter of policy to uphold no-oralmodification clauses, we simply have to ignore the consequent injustice caused to the promisee. Indeed Lord Sumption’s citing of Actionstrength Ltd v International Glass Engineering SpA26 tends to reinforce this point, because in that case, although the House of Lords confined its d ­ ecision – ­rejecting estoppel by convention – to the particular facts, the reasoning indicated that it will almost always be the case that allowing an oral guarantee to be enforceable through an estoppel by convention will undermine the statutory rule (laid down in section 4 of the Statute of Frauds 1677) that guarantees must be evidenced in writing. (c) It would appear that Lord Briggs, like Lord Sumption, would recognise the continued role of estoppel,27 so that one can make the same criticism about the uncertainty of his approach as has just been made in relation to Lord Sumption. But Lord Briggs’ approach adds another particular uncertainty. This is because it will often require the courts to apply his difficult test of whether the parties intended to depart from the NOM clause by ‘strictly necessary implication’, which appears to take an unprecedented step beyond ‘necessary implication’. (d) As a matter of policy, it can be strongly argued that if informality in variations is a significant problem, it should be tackled by legislation. But on the face of it, it is hard to see, in terms of policy, why, if a contract can be made informally, one would want a higher level of formality to apply to a variation of that contract. (e) As was recognised in the Supreme Court, it would appear that in other common law jurisdictions NOM clauses can always be departed from.

26 Actionstrength 27 See

Ltd v International Glass Engineering SpA [2003] UKHL 17, [2003] AC 541. the reference to estoppel in Lord Briggs’ judgment at Rock (SC) (n 1) [30].

46  Andrew Burrows In  FHR European Ventures LLP v Cedar Properties LLC,28 Lord Neuberger, giving the judgment of the Supreme Court (in deciding that bribes taken in breach of fiduciary duty should be held on constructive trust), said ‘it seems to us highly desirable for all [common law] jurisdictions to learn from each other, and at least to lean in favour of harmonising the development of the common law round the world’. In other words, there is much to be said for a presumption of maintaining uniformity in the common law across the common law world, unless there are sufficiently strong reasons for rebutting that presumption by taking a different approach. Lord Sumption’s view would represent a clean break, and Lord Briggs’ mid-position would represent an incremental departure, from that common law harmony. In contrast, Lord Justice Beatson’s approach has the merit of maintaining alignment with what appears to be the common law in the rest of the common law world. (f) There is no convincing analogy between NOM clauses and statutory formalities or entire agreement clauses, essentially for the reasons set out by Lord Briggs (see his reasoning at (b) and (c) in section IV). However, with respect to Lord Briggs, his reliance on the ‘subject to contract’ analogy is also ultimately unconvincing. That area of the law is concerned with the parties’ clarifying that they have no intention to create legal relations. One will need something to make clear that they do in fact have an intention to create legal relations. In other words, that area is to do with one particular required element of a contract. There is no direct equivalent in relation to a NOM clause. One is not saying that the parties have no intention to create legal relations unless they comply with the specified formality. Rather one is saying that, although all the normal requirements for a contractual variation are present, including intention to create legal relations, that contractual variation is not binding simply because of an earlier NOM clause. (g) The justification for NOM clauses based on a fear of false allegations of oral modifications (see Lord Sumption’s second commercial reason for NOM clauses that we have looked at in section III) seems misplaced. There is always the potential for a fabricated claim that a party has orally made an agreement when it has not done so. But this potential problem is dealt with in the law of contract (as indeed, more generally, throughout civil law) by the application of the normal burden and standard of proof (the claimant must prove the agreement on the balance of probabilities). There is nothing to suggest that this normal approach to proof does not work satisfactorily in the context of oral modifications. (h) It is submitted that the important policy that is validly effected by NOM clauses is not the uncertainty as to what the parties have agreed but rather that one does not want those without authority binding a principal.29 28 FHR European Ventures LLP v Cedar Properties LLC [2014] UKSC 45, [2015] AC 250 [45]. 29 Of course, there is no such policy in play where the relevant person has actual authority to bind the principal. Actual authority is purely an internal matter as between the principal and agent. It almost always arises from a contract between the principal and agent.

Anti-Oral Variation Clauses  47 In other words, by reason of the doctrine of apparent authority – which rests on a representation by P to the other contracting party that, even though A has no actual authority, A has authority to bind P30 – the real problem is that the law on agency may result in a junior person in a company or other business organisation agreeing to a variation which is then binding on the company/business. This was the third of the commercial reasons for NOM clauses given by Lord Sumption that we have looked at in section III. On the facts of Rock, it would appear that this was the primary reason for the clause. It seems likely that Ms Evans, as credit controller, did not have actual authority to bind MWB to a variation of the payment terms of the contractual licence, but importantly it was decided that she did have apparent authority to do so.31 But if narrowing the scope of the law on agency by apparent authority is the policy goal, the focused way in which drafters can achieve that is by inserting a clause that makes clear to the other contracting party that only certain people in a company/business have authority to agree to a variation on behalf of the company/business; or perhaps more obviously that certain people do not have that authority. In other words, such a clause would operate to negate a representation by the principal of the relevant (junior) person’s authority that would otherwise arise. Once such a clause has been inserted into the contract, it is hard to see how, in general,32 it could be successfully argued that the relevant (junior) person has apparent authority to bind the company/business. So what is needed to reflect this valid policy goal is not a wide-ranging NOM clause but a narrower clause making clear that particular people do not have authority to bind the company/business. It is a ‘no agency’, not a NOM, clause that would reflect the sound policy in play and to which, in contrast to a NOM clause, there can be no objection of principle or policy. Indeed, rather than using a contractual clause, P might alternatively negate A’s apparent authority by a non-contractual ‘no agency’ representation. All that is required is for P to make clear to the other contracting party that, despite a representation that would otherwise arise, A does not have authority to agree amendments to the contract. I am not aware of any cases dealing with clauses expressly restricting a person’s authority to make oral variations.33 But there are cases dealing with clauses expressly restricting a person’s authority to make representations or warranties. Perhaps the best known is Overbrooke Estates Ltd v Glencombe Properties Ltd.34 Here the defendant buyers successfully bid for the claimants’ 30 A leading case is Freeman & Lockyer v Buckhurst Park Properties (Mangal) Ltd [1964] 2 QB 480 (CA). 31 See n 4. 32 Of course, the principal might expressly represent that the agent has authority, which would override the contractual term. Put another way, the principal would need to be astute to ensure that it does nothing to render it reasonable for the third party to rely on a representation by the principal as to the agent’s having authority despite the contractual term to the contrary. 33 Although Mendelssohn v Normand [1970] 1 QB 177 comes close to this: see n 35. 34 Overbrooke Estates Ltd v Glencombe Properties Ltd [1974] 1 WLR 1335 (Ch).

48  Andrew Burrows property in London at auction. One of the conditions of sale stated, ‘The vendors do not make or give and neither the auctioneers nor any person in the employment of the auctioneers has any authority to make or give any representation or warranty in relation to [the property].’ In response to an inquiry from the defendants, the auctioneers (Willmotts) had represented to them that the relevant local authority had no plans for the property and were not interested in compulsory purchase of it. In the week following their successful bid, the defendants were informed by the local authority that there was a chance the property would shortly be included in a slum clearance programme. The defendants therefore refused to pay for the property and the claimants sought specific performance. The defendants argued that they were entitled to escape from the contract because of the auctioneers’ misrepresentation. On the assumption that there had been a misrepresentation by the auctioneers inducing the contract, it was held that, because of the aforementioned condition, the claimants were not liable for the misrepresentation. The condition negated the ostensible authority of the agent (the auctioneers) to make representations on behalf of its principal (the claimants). It was further held that section 3 of the Misrepresentation Act 1967 did not apply. Specific performance was therefore granted. On the ‘no agency’ issue, Brightman J said the following: [I]n the case before me it is not, in my judgment, possible for the defendants to assert that Willmotts had ostensible authority to make the representations said to have been made by them, for these reasons – before any contract was made and, indeed before any representation was made, the defendants were obviously in possession of a document which, in terms, negatived any such authority … It seems to me that it must be open to a principal to draw the attention of the public to the limits which he places upon the authority of his agent and that this must be so whether the agent is a person who has or has not any ostensible authority. If an agent has prima facie some ostensible authority, that authority is inevitably diminished to the extent of the publicised limits that are placed upon it. When [the representative of the defendants] telephoned Willmotts he knew, or ought to have known, that Willmotts had no authority to make or give any representation or warranty in relation to this property. When he bid for the property five days later, he did so on the basis that nothing which had been said to him by Willmotts amounted to a representation or warranty binding upon the plaintiffs.35

35 ibid 1341. Brightman J’s reasoning was approved by the Court of Appeal in Collins v Howell-Jones (1981) 259 EG 331: for a useful note on the latter case, see John Murdoch (1981) 97 LQR 518. It was also agreed with, but distinguished, by Scott J in Museprime Properties Ltd v Adhill Properties Ltd (1991) 61 P&CR 111. In the Overbrooke case, Brightman J distinguished Mendelssohn v Normand [1970] 1 QB 177, in which a car park attendant was held to have apparent authority to bind his employer to an oral promise constituting a collateral contract despite a ‘no oral variation clause’ that read as follows: ‘No variation of these conditions will bind the garage proprietors unless made in writing signed by their duly authorised manager.’ See generally Peel, Treitel (n 21) para 16-027. I am grateful for an email interchange on this issue with Professor Francis Reynolds.

Anti-Oral Variation Clauses  49

The importance of Brightman J’s helpful exposition for my theme is that it shows that, provided a clause is drafted so as expressly to limit or negate a person’s apparent authority, it should be effective to do so.

Overall, it follows from these eight points that the approach of Lord Justice Beatson not only has the merit of being principled but can also be supported as a matter of policy. For those reasons, it is submitted that it constitutes the best approach.

VII. Conclusion The issues raised by NOM clauses are difficult. The Supreme Court has committed the law in this jurisdiction to uphold such clauses, and for many practitioners that will be very welcome news. Yet, in terms of principle, the latest agreement of the parties (assuming it would otherwise be binding) ought to prevail, whether written, oral or made by conduct. Moreover, what I have suggested here is that the apparent enhancement of certainty by the decision in Rock may be misleading, that the analogies drawn by the Supreme Court are unhelpful, that maintaining harmony in the common law world is desirable, that there are no particular problems of proof in this context and that the policy of real importance concerns the apparent authority of agents, not formality as such. It follows that, for reasons of policy as well as principle, the approach of Lord Justice Beatson in the Globe case is, with respect, to be preferred to that taken by Lord Sumption and Lord Briggs. This may be ‘water under the bridge’ in this jurisdiction but, for other common law jurisdictions, watch this space.

50

4 Controlling Contractual Interpretation RICHARD CALNAN*

English lawyers are not shy about writing long contracts. One of the distinctions between civil lawyers and common lawyers has been the length of their respective documents. Civil lawyers have tended to write relatively short documents, relying on their law codes to fill the gaps.1 Common lawyers have been more prescriptive. That is partly because they do not usually have easily accessible law codes, but perhaps more important is a desire for the contract itself to regulate as many of the features of the intended relationship between the parties as it is practicable to prescribe. It is not that the common law does not have underlying legal rules that would apply in the absence of agreement. It is just that we want to express the rules ourselves. They can then be tailored to the needs of the particular parties concerned. If there is a dispute, that avoids the necessity to rummage through the undergrowth of the underlying law, and enables us to concentrate on the meaning of the words used by the parties in their contract. Hence the pre-eminent position of contractual interpretation in the common law of contract.2 Lawyers spend a great deal of time negotiating the terms of contracts, with a view to making them as complete and clear as possible. But, whilst the terms of the contract are spelled out in detail, little thought is given to how those terms will be interpreted. Most contracts do contain interpretation clauses, but they tend to deal with minutiae. Few contracts deal with the key issues of interpretation. In practice, there are two important questions the contract could clarify: • How much background information should the interpreter be able to take into account when interpreting the contract?

* My thanks to all those who commented on a draft of this contribution. It was presented as a paper at the UCL Conference, ‘The Contents of Commercial Contracts: Terms Affecting Freedoms’ (9 and 10 May 2019). I am particularly grateful to Sandy Phipps of One Essex Court, who commented on the paper at the Conference and whose insights have shaped my approach to drafting. The usual caveats apply. 1 Although they are getting longer, to some extent under the influence of the common law approach. 2 A Burrows, A Restatement of the English Law of Contract (OUP 2016) 85.

52  Richard Calnan • To what extent should context or common sense be able to override the express words of the contract? The purpose of this chapter is to consider whether it is legally possible to control interpretation in this way, whether it is desirable to do so and what an interpretation clause might look like.

I.  The Context Like any debate concerning interpretation, this needs to be put in context. Why is it appropriate to consider this issue now? Perhaps the most important development in the law of contract over the last 25 years has been the explosion in the number of cases concerning contractual interpretation, and the resulting increase in discussion of the principles of contractual interpretation by judges, academics and practitioners. Although there are pre-echoes in earlier cases,3 it was the decision of the House of Lords in Investors Compensation Scheme v West Bromwich Building Society4 in 1997 that signalled the start of the recent trend to analyse the nature of the interpretation process. The debate may no longer be conducted with the intensity of the first decade of the twenty-first century, but it does still continue. There are two issues that have dominated the discussion, and they are to some extent interlinked. The first is the extent to which information from outside the contract itself should be taken into account in interpreting the words used in the contract. The second is the extent to which the interpreter should be able to use that context or common sense to ­override the express words of the contract. For much of the period since 1997, the prevailing orthodoxy was to encourage the use of the background facts to influence the meaning of the words used in the contract, and to make use of those facts and of common sense to override the express words of the contract. What is interesting about this phenomenon is that it was mostly confined to our highest court. For almost 20 years, the House of Lords and then the Supreme Court strained the meaning of the words of commercial contracts in order to give effect to what they considered was the real intention of the parties. There are numerous examples of these cases,5 but the following give a flavour of the approach: • In 1997, the House of Lords decided that ‘any claim (whether sounding in rescission for undue influence or otherwise)’ actually meant ‘any claim sounding in rescission (whether for undue influence or otherwise)’.6 3 Particularly in Charter Reinsurance v Fagan [1997] AC 313. 4 Investors Compensation Scheme v West Bromwich Building Society (No 1) [1998] 1 WLR 896 (HL). 5 They are discussed in more detail in R Calnan, Principles of Contractual Interpretation, 2nd edn (OUP 2017) ch 7. 6 Investors Compensation Scheme v West Bromwich Building Society (n 4).

Controlling Contractual Interpretation  53 • In 2009, the House of Lords decided that, in order to calculate ‘23.4% of the price achieved for each residential unit in excess of the minimum guaranteed residential unit value less the costs and expenses’, it was necessary to deduct the costs and expenses from the price achieved for each residential unit, then calculate 23.4% of that figure and then deduct the minimum guaranteed residential unit value from the result.7 • Also in 2009, the Supreme Court decided that the requirement in a security document that ‘During the Realisation Period the Security Trustee shall so far as possible discharge on the due date therefor any Short Term Liabilities’ of a company did not apply if the company was insolvent.8 • In 2010, the Supreme Court decided that a requirement to value a property at ‘the full market value of [the property] as at the date of entry for the proposed purchase … of agricultural land or open space suitable for development as a golf course’, did not require the property to be valued only on the basis that it was agricultural land or open space suitable for development as a golf course but enabled account to be taken of its development value.9 In 2015, the Supreme Court put a break on this judicial distortion of the English language10 in two cases – Arnold v Britton11 and Marks and Spencer v BNP Paribas.12 In Arnold v Britton, Lord Neuberger said that ‘the reliance placed in some cases on commercial common sense and surrounding circumstances … should not be invoked to undervalue the importance of the language of the provision which is to be construed’.13 In spite of Lord Hodge’s comment in Wood v Capita14 in 2017, that the ‘recent history of the common law of contractual interpretation is one of continuity rather than change’,15 I would suggest that the last quarter century has seen the pendulum swing first one way and then the other. In fact, it has been change that has been the order of the day. This should come as no surprise. It is not a new phenomenon. As Joanna McCunn has demonstrated,16 there was a period during the mid-sixteenth century when purposive interpretation held sway, and that was followed by a return to a more textual approach in succeeding generations. 7 Chartbrook v Persimmon [2009] UKHL 38, [2009] 1 AC 1101. It is hard to think of a case that more cried out for the use of a formula. 8 Re Sigma Finance [2009] UKSC 2, [2010] 1 All ER 571. 9 Multi-Link v North Lanarkshire Council [2011] 1 All ER 175. 10 To borrow an expression used by Lord Diplock in Photo Production v Securicor [1980] AC 827, 851. 11 Arnold v Britton [2015] UKSC 36, [2015] AC 1619. 12 Marks & Spencer Plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2015] UKSC 72, [2016] AC 742. 13 Arnold v Britton (n 11) [17]. 14 Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173. 15 ibid [15]. 16 J McCunn, ‘Revolutions in Contractual Interpretation: A Historical Perspective’ in S Worthington, A Robertson and G Virgo (eds), Revolution and Evolution in Private Law (Hart Publishing 2018) ch 8.

54  Richard Calnan The one thing that is certain about contractual interpretation is that the outcome of any particular dispute is often far from certain. As David McLauchlan has pointed out, the outcome of cases concerning the interpretation of contracts is difficult to predict.17 Once it is understood that interpretation is essentially a matter of judgement,18 it becomes much easier to understand why there is so much uncertainty. Is it possible to draft contracts in a way that will produce more certainty? At one level, the answer is obvious. Of course we can create more certainty by drafting more clearly. But the question I want to consider here is a different one: Is it possible to create more certainty through the drafting of an appropriate interpretation clause? Most commercial contracts do contain interpretation clauses. They are normally an adjunct to the definitions clause. But they add little to the sum of human knowledge. By and large, they content themselves with mundanities, such as requiring the singular to include the plural and the masculine the feminine. There is another type of clause that is perhaps more pertinent: the entire agreement clause. It is common for a commercial contract to contain a clause stating that the contract (together with any other relevant transaction documents) constitutes the entire agreement between the parties in relation to the transaction concerned, and therefore that any prior written or oral arrangements between the parties do not form part of the contract.19 In Inntrepreneur Pub Co v East Crown,20 Lightman J said that an entire agreement clause ‘constitutes a binding agreement between the parties that the full contractual terms can be found in the document containing the clause and not elsewhere’. This conclusion has not gone without criticism.21 But there are good commercial reasons why the parties do want to make clear the extent of their contract,22 and there is no reason to doubt that the courts will give effect to entire agreement clauses in accordance with their terms.23

17 D McLauchlan, ‘Contract Interpretation: What is it about?’ (2009) 31(1) Sydney Law Review 5. 18 See, for instance, the comment of Robert Walker LJ in John v PricewaterhouseCoopers [2002] EWCA Civ 899 [94]. 19 For discussion of entire agreement clauses, see J Morgan, ‘Opting for “Documentary Fundamentalism”: Respecting Party Choice for Entire Agreement and Non-Reliance Clauses’, ch 12 in this volume. 20 Inntrepreneur Pub Co v East Crown [2000] 2 Lloyd’s Rep 611, 614. 21 See Law Commission, Law of Contract: Parol Evidence Rule (Law Com No 154, 1986) para 2.15; and D McLauchlan, ‘The entire agreement clause: conclusive or a question of weight?’ (2012) 128 LQR 521. 22 See the observations of Lightman J in Inntrepreneur Pub Co v East Crown (n 20) 614. 23 There are more problems with clauses that purport to contract out of non-contractual remedies for misrepresentation. They are interpreted ‘strictly’ (see Axa Sun Life v Campbell Martin [2011] EWCA Civ 133, [2012] Bus LR 203 [94]) and there are statutory limits on their effectiveness, for instance under s 3 of the Misrepresentation Act 1967 and the Unfair Contract Terms Act 1977. Jonathan Morgan points out in Morgan ‘Opting for “Documentary Fundamentalism”’ (n 19) that the latter may also affect certain entire agreement clauses.

Controlling Contractual Interpretation  55 But there is a limit on the effectiveness of an entire agreement clause. Its purpose is to establish what the contract is, not what it means. Once the parties have established what the terms of the contract are, it is then up to the court to interpret the meaning of those terms. An entire agreement clause has nothing to say about that. Which leaves open the question: what if the parties do want to provide in their contract for how the contract should be interpreted? Are they entitled to do so?

II.  Can We Use Interpretation Clauses? It should come as no surprise that there is little authority on this point. Basic ­principles are rarely challenged. Hence the dearth of authority. So we have to go back to principle. In principle, there is nothing to prevent the parties from agreeing the basis on which their contract should be interpreted.24 In part this is because (as with entire agreement clauses) there is no relevant rule of law that prevents the parties from agreeing how the contract will be interpreted.25 But perhaps even more importantly, it is because the fundamental principle of contractual interpretation is to establish the objective intention of the parties from what they have written, said and done.26 And what could establish this more clearly than to explain it in the contract? This was the conclusion reached by Catherine Mitchell: If we accept that commercial parties can choose their contracting partners and their obligations, why shouldn’t the parties also choose the interpretative theory that will be applied to their contract? Or, in other words, why shouldn’t the contract tell the court not to look at the context? Contracts frequently do include ‘interpretation’ clauses and there is no more objective statement of the parties’ intentions than the terms they commit to writing.27

III.  Should We Use Interpretation Clauses? The main reason to use an interpretation clause in our contracts is to take control of the process of interpretation.

24 See the Scottish Law Commission, Review of Contract Law: Discussion Paper on Interpretation of Contract (Scot Law DP 147, 2011) paras 7.23–7.25. 25 As one of my colleagues used to say, ‘That’s just an assertion, Boss.’ That is true, but it is difficult to prove a negative. 26 See, for instance, Lord Steyn’s comment in Deutsche Genossenschaftsbank v Burnhope [1995] 1 WLR 1580, 1587. 27 C Mitchell, ‘Entire Agreement Clauses: Contracting out of Contextualism’ (2006) 22 Journal of Contract Law 222, 236–37. And see C Mitchell, Interpretation of Contracts (Routledge Cavendish 2007) ch 5.

56  Richard Calnan One of the great strengths of the common law tradition is that contracts do, as far as is practicable, spell out the rights and duties of the parties in all conceivable (or at least in all conceived) circumstances. Commercial lawyers do not like leaving things to chance. But to draft the terms of the contract in detail and not to explain how those terms are to be interpreted leaves the job half done. Having set out the terms, the natural next step is to explain how those terms should be interpreted in the event of a dispute. The events of the last 25 years have demonstrated how uncertain the interpretation process can be, and therefore how important it is for the contract to control the process. Nothing can alter the fact that interpretation will always ultimately be a matter of judgement. But what the draftsman can do is to steer the outcome in a particular direction. He or she can ensure that the judgement is exercised within the constraints imposed by the contract. I suggest that there is a compelling argument in favour of contracts being more prescriptive about how they should be interpreted. The problem comes in persuading lawyers to change the way in which they have always done things. We tend to follow precedents. We change our approach where necessary to reflect statutory changes,28 but otherwise we leave well alone. If there is no compulsion to change then why bother? That is a difficult question to answer in the abstract. The advantages of using an interpretation clause in our contracts can best be judged by seeing what such a clause might look like.

IV.  What Should an Interpretation Clause Say? It will be for the parties to decide what the interpretation clause should say. The reason for using such a clause is to enable the parties to take control of the process of interpretation. But in practice, I would suggest that the main requirement of commercial parties and their lawyers will be to increase the certainty of the contract. There are two ways in which an interpretation clause can help to achieve this. First, it can limit the amount of extrinsic evidence that is available when interpreting the contract. Second, it can reduce the ability of the interpreter to twist the meaning of the words in the contract.

A.  Background Facts Let us start with the context. In my lifetime, there has been a fundamental change in the way that the English courts have approached the interpretation of contracts. 28 For instance, contracts were immediately changed to exclude the application of the Contracts (Rights of Third Parties) Act 1999.

Controlling Contractual Interpretation  57 It is now generally recognised that interpretation is necessarily contextual. Words in a legal document need to be read not only in the light of the rest of the document, but also in the context in which that document is set. One can accept the truth of the principle that contracts must be read in context whilst at the same time deprecating the tendency to deploy vast amounts of (largely irrelevant) background information to determine the meaning of the words in the contract. The problem with the current approach to interpretation is that the background facts include ‘absolutely anything which would have affected the way in which the language of the document would have been understood by a reasonable man’.29 The approach of English law to this issue has been to start with a basic principle that all background facts are relevant, and then to exclude specific types of background fact.30 This comes at the problem from the wrong direction. It should not be the case that absolutely anything is relevant unless it is excluded. A better approach would be to limit at the outset the extent of the extrinsic evidence that can be used. The current approach encourages parties to trawl through the undergrowth in order to find something that might assist them in their argument. This simply extends the amount of time required to decide what the contract means and increases the costs of the exercise, as well as potentially prejudicing assignees. What is needed is a way of limiting the identity of the background facts to those that are really relevant to the process of interpretation. Easy to say; much more difficult to draft. How to identify facts that are relevant to the meaning of the words in the contract, whilst at the same time excluding anything else? Background facts should only be used if it can clearly be demonstrated that they are necessary in order for a reasonable person to understand the way in which the words in issue have been used. But how do you draft for what is essentially a qualitative judgement? The clearest approach is for the draftsman to identify the relevant background facts in the agreement. Indeed, this is already done in practice in many agreements. Recitals are commonplace. Modern contracts frequently describe them as ‘Background’ – which rather gives the game away. These are the background facts necessary to understand the contract. There is no necessity to go outside the contract to find them. The draftsman has already done the job. It is therefore suggested that the best approach in many types of commercial contract is simply to exclude background facts that do not appear from the transaction documents themselves. The transaction documents as a whole are likely to contain all the background relevant to interpreting their terms. That is not to say that the interpretation exercise becomes non-contextual. The words and expressions used in the contract need to be understood in the light of

29 Investors Compensation Scheme v West Bromwich Building Society (n 4) 192; Bank of Credit and Commerce International v Ali [2001] UKHL 8, [2002] 1 AC 251 [39]. 30 Such as pre-contractual negotiations and subsequent conduct.

58  Richard Calnan the transaction documents as a whole and of the background facts described in those documents. That may exclude from consideration some things a tribunal might conceivably have found useful. But the advantage of this approach is that it enables contracts to be interpreted more quickly, and with a greater degree of certainty. In most cases, the relevant context of an agreement is clear from its text. A good example is The Aragon.31 A charterparty only allowed the charterer to use the vessel within the ‘USA East of Panama Canal’. The question at issue was whether the US Gulf was within that limit. In purely geographical terms, the US Gulf is to the west of the meridian of longitude on which the Panama Canal stands. But the court had no difficulty in deciding that the charterer was entitled to use the vessel within the US Gulf. The purpose of the charter was to carry out a round trip from Europe to the East Coast of North America and then back. In this context, what was meant by the words ‘USA East of Panama Canal’ was a port in the United States of America that did not entail going through the Panama Canal. The court was able to interpret the words concerned in context without trawling through a great deal of background information. The relevant context was that the voyage started in Europe – to the East of the US Gulf – and that the Panama Canal is a means of heading West into the Pacific Ocean. Once that was understood, the contextual meaning of the words ‘East of Panama Canal’ – and the reason why they had been used – became clear. All the context required was provided by the terms of the agreement itself.

B.  Ordinary Meaning There is something arguably even more important than context. An interpretation clause can also limit the ability of the interpreter to twist the meaning of the words used in the contract. Under the general law, if the court considers that the parties cannot objectively have intended the words they have used to have their ordinary meaning, they will be given the meaning the parties must objectively have intended. The more unreasonable the result, the more unlikely it is that the parties could have intended it.32 Allied to this is the tendency for courts to interpret certain types of clause in a way different from the rest of the contract. This has been a particular issue with exclusion and limitation clauses, termination clauses and guarantees. Clauses of this kind are often read ‘strictly’ (whatever that may mean). Although there are some indications in the recent case law that the courts are open to the idea that all

31 Segovia Compagnia Naviera v R Pagnan & Fratelli (The Aragon) [1977] 1 Lloyd’s Rep 343 (CA). 32 Investors Compensation Scheme v West Bromwich Building Society (n 4) 913; Schuler v Wickman Machine Tool Sales [1974] AC 235, 251.

Controlling Contractual Interpretation  59 terms of commercial contract should be interpreted in the same way,33 there is still doubt about precisely how clauses of this kind will be interpreted.34 What can the draftsman do to clarify the position? One approach – which has a great deal of practical merit – is not to draft the contract so unreasonably that the courts will want to find a way round it. Another approach – perhaps rather less practical – is to add the words ‘and we really mean this’ at the end of every controversial clause. In practice, the best course is probably for the contract to state that it will be interpreted by reference to the ordinary meaning of the words used. That still leaves scope for the exercise of judgement, but it should at least limit the wilder flights of fancy seen in some of the recent cases. The question then arises as to how to deal with obvious drafting errors. Even in a well-drafted contract, ‘typos’ occur. ‘Landlord’ may be used instead of ‘tenant’, when it is obvious what was meant. The word ‘not’ may be inserted or omitted by accident. It would be sensible to allow the interpreter to rectify an obvious mistake by interpretation. The interpretation clause could also help to deal with ambiguities. With the best will in the world, it is impossible to avoid all trace of ambiguity when drafting. If words are ambiguous, the interpreter will decide which of the possible meanings is the most likely objectively to have been intended. In practice, the best evidence of what the parties are likely to have intended will come from the rest of the document. Failing that, the interpreter needs to turn to his or her own view of business common sense. In the leading modern case, Rainy Sky v Kookmin Bank,35 I would suggest that the Supreme Court gave effect to its understanding of what was commercially sensible in preference to reading the words in the context of the agreement as a whole. An interpretation clause could make it more difficult for that to be done. It could require the ambiguity to be resolved primarily by reference to the terms of the agreement as a whole.

C.  An Interpretation Clause Bringing these strands together, the interpretation clause could look something like this: Interpretation 1.1 This agreement will be interpreted in accordance with the provisions of this clause. 1.2 The words and expressions used in this agreement will be given their ordinary meaning in the context of the Transaction Documents as a whole. 33 See, for instance, Static Control Components v Egan [2014] EWCA Civ 392, [2004] 2 Lloyd’s Rep 429 (guarantees); and Persimmon v Ove Arup [2017] EWCA Civ 373, [2017] BLR 417 (exclusion clauses). 34 See, for instance, the discussion by the Court of Appeal in Harvey v Dunbar Assets [2013] EWCA Civ 952 [28]–[32] in relation to guarantees. 35 Rainy Sky v Kookmin Bank [2011] UKSC 50, [2012] Bus LR 313.

60  Richard Calnan 1.3 This is the case irrespective of any commercial or other considerations; and it applies to all the provisions of this agreement, whatever their nature. 1.4 The only exception is where there is an obvious typographical error and it is clear what was intended by the parties, in which event effect will be given to that clear intention. 1.5 If a particular word or expression is reasonably capable of more than one ordinary meaning, it will be given the meaning which is most consistent with the Transaction Documents as a whole. 1.6 Apart from the Transaction Documents, nothing else will be taken into account in the interpretation of this agreement. 1.7 The Transaction Documents means this agreement and [others].

V.  What Else Could an Interpretation Clause Do? The points explored in sections IV.A and B are the fundamentals of an interpretation clause. It would be possible for the clause to do other things as well. This, again, is a matter for the parties, but there are certain issues that cause problems in practice, which could be dealt with expressly in the contract. These include: • the implication of terms, particularly good faith provisions;36 • limitations on the exercise of contractual discretions; and • the use of canons of construction.

A.  Good Faith and Implied Terms In recent years the English courts have flirted with the idea of implying a general duty of good faith into certain types of contract. This was most famously done by Leggatt J in Yam Seng v International Trade Corporation.37 It received a lukewarm reception, but it has proved to be quite persistent, as the judgment of Fraser J in Bates v Post Office (No 3)38 illustrates. He decided that a duty of good faith is implied into ‘relational’ contracts.39 One might ask what is wrong with requiring the parties to act in good faith? Surely that is what they would expect? And why would a lawyer want his or her client to enter into an agreement that did not contain such a requirement? As so often with English law, the reason is a pragmatic one. Good faith is one of those vague concepts which it is impossible to pin down with any degree of certainty. 36 For the discussion of terms excluding good faith and limiting the exercise of contractual discretions, see PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6 in this volume. 37 Yam Seng v International Trade Corporation [2013] EWHC 111 (QB). 38 Bates v Post Office (No 3) [2019] EWHC 606 (QB). 39 ibid [711]. The discussion of good faith is to be found in [702]–[741].

Controlling Contractual Interpretation  61 The difficulty can be seen from those cases that have attempted to establish what is meant by an express duty of good faith. If the parties insert a provision into their contract that requires them to act in good faith, a court must give effect to it. But the difficulty of establishing precisely what it means can be seen from CPC Group v Qatari Diar.40 In that case, Vos J considered various definitions of what is meant by good faith. They included: • It underwrites the spirit of the contract and supports the integrity of its character. • It is a duty to recognise and to have due regard to the legitimate interests of both parties. • It emphasises faithfulness to an agreed common purpose and consistency with the justified expectations of the other party. The problem with giving legal effect to concepts like these is manifest. If parties feel the need for this sort of thing, they can choose a civil law jurisdiction that gives effect to it. But parties tend to choose English law because the rights and duties of the parties are clearly set out in the contract and the court will give effect to them. There is a real concern that implied duties of good faith impinge on the certainty of English law. Each party is entitled to expect that the other parties will not act dishonestly. But if they want other protections, they should be written into the contract clearly enough for the parties to understand what they can do, and what they cannot do. An interpretation clause can solve this problem. As Fraser J made clear in Bates v Post Office (No3),41 an express term of the contract can prevent any duty of good faith from being implied. In the light of the decision in that case, it is quite possible that express provisions of this kind will find their way into contracts – particularly into long-term contracts. The interpretation clause could provide as follows: 1.8 No term will be implied into this agreement which requires a party to act in good faith.

This raises a broader question about the implication of terms. The Supreme Court has recently emphasised in Marks & Spencer v BNP Paribas42 how difficult it is to imply a term into a lengthy commercial contract. But if the interpretation clause is to deal with the implication of terms of good faith, it could also tackle the question of implication more generally. It could provide as follows: 1.9 No other term will be implied into this agreement unless the agreement will not work without it.

40 CPC

Group v Qatari Diar [2010] EWHC 1535 (Ch). v Post Office (No 3) (n 38) [725]–[726]. 42 Marks & Spencer v BNP Paribas Securities Services Trust (n 13). 41 Bates

62  Richard Calnan

B.  The Exercise of Discretions Commercial agreements frequently authorise one of the parties to exercise a contractual discretion. If it is intended that the exercise of a discretion should be limited, the contract can provide for this. Where it does not do so, it is frequently intended that the discretion should be absolute. But the courts have intervened in this area in order to restrict the ability of a party freely to exercise a contractual discretion. This has been done by interpretation, by the implication of a term and by analogy with public law principles of unreasonableness.43 If parties do want discretions to be capable of being exercised freely, it is possible to provide for this in the contract.44 The interpretation clause could contain a provision along the following lines: 1.10 If a provision of this agreement has the effect of conferring a discretion on a party, that party may (except to the extent that the provision expressly provides to the contrary) exercise that discretion in its own unfettered interests, take into account anything it considers appropriate, and ignore anything else.

C.  Canons of Construction Lawyers tend to use canons of construction in order to persuade a court to reach a result that overrides the natural meaning of the words. We like to feel that we are deciding matters of interpretation on the basis of legal principle, rather than simply by intuition. But we actually use canons of construction to create our own private legal language from which non-lawyers are excluded, in order to reach a result not justified by the words used. Perhaps one of the most egregious of the canons of construction is the eiusdem generis rule. If there is a list of things, followed by general words, the general words will be read in the context of the list; but not if the general words contain the word ‘whatsoever’. Devlin J was critical of this approach in Chandris v Isbrandtsen-Moller Co,45 but the rule is still used in practice.46 What it fails to recognise is that the parties often use general words with the very purpose of expanding the scope of the more specific words, and that to limit them defeats the object of the exercise.

43 See, for instance, Braganza v BP Shipping [2015] UKSC 17, [2015] ICR 449. 44 I am grateful to my colleague Michael Sinclair for making this point and explaining how best to draft round it. 45 Chandris v Isbrandtsen-Moller Co [1951] 1 KB 240, 245. 46 Cosco Bulk Carrier Co v Team-Up Owning Co [2010] EWHC 1340 (Comm), [2010] 1 CLC 919.

Controlling Contractual Interpretation  63 An obvious answer is to avoid the use of lists, or to state the general words first and then use the list as illustrations. But there is no reason why the parties should not simply contract out of the rule altogether, and of similar rules, such as contra proferentem. The interpretation clause could provide as follows: 1.11 The contra proferentem and eiusdem generis rules will not be used in the interpretation of this agreement.

D.  Rectification, Estoppel by Convention and Fraud If the parties have used the wrong words in the contract, it can be rectified. But this requires clear evidence, and in practice it is difficult to satisfy the requirements. The interpretation clause should not attempt to alter this. Rectification is a useful safety valve where the drafting has gone wrong. But the necessity to plead rectification and to prove that the document as signed is not what the parties had actually agreed at that time means that it will only be relevant in a limited number of cases. For similar reasons, the interpretation clause should not impinge on the law concerning estoppel by convention. And it should also be clear that it does not affect fraud (in the sense of dishonesty). The interpretation clause can therefore provide as follows: 1.12 Nothing in this clause affects the law concerning rectification, estoppel by convention or fraud.

The full text of the suggested interpretation clause is set out at the end of this chapter.

VI. Conclusion The purpose of this chapter has been to start a debate. I have suggested that there are good reasons why commercial contracts should contain a clause that establishes how the contract is to be interpreted. And I have proposed that such a clause would contain two key elements. First, it would reduce the extent to which the interpreter could use information from outside the contract in order to interpret the words in the contract. Second, it would make it more difficult for the interpreter to override the express words of the contract. In addition, it could clarify particular contractual issues that have created problems in practice, such as duties of good faith and the exercise of discretions. If this chapter encourages others to consider this issue, it will have served its purpose.

64  Richard Calnan INTERPRETATION CLAUSE

Interpretation 1.1 This agreement will be interpreted in accordance with the provisions of this clause. 1.2 The words and expressions used in this agreement will be given their ordinary meaning in the context of the Transaction Documents as a whole. 1.3 This is the case irrespective of any commercial or other considerations; and it applies to all the provisions of this agreement, whatever their nature. 1.4 The only exception is where there is an obvious typographical error and it is clear what was intended by the parties, in which event effect will be given to that clear intention. 1.5 If a particular word or expression is reasonably capable of more than one ordinary meaning, it will be given the meaning which is most consistent with the Transaction Documents as a whole. 1.6 Apart from the Transaction Documents, nothing else will be taken into account in the interpretation of this agreement. 1.7 The Transaction Documents means this agreement and [others]. 1.8 No term will be implied into this agreement which requires a party to act in good faith. 1.9 No other term will be implied into this agreement unless the agreement will not work without it. 1.10 If a provision of this agreement has the effect of conferring a discretion on a party, that party may (except to the extent that the provision expressly provides to the contrary) exercise that discretion in its own unfettered interests, take into account anything it considers appropriate, and ignore anything else. 1.11 The contra proferentem and eiusdem generis rules will not be used in the interpretation of this agreement. 1.12 Nothing in this clause affects the law concerning rectification, ­estoppel by convention or fraud.

5 Good Faiths and Contract Terms MAGDA RACZYNSKA*

I. Introduction The classical view of contract law is that commercial (non-consumer) parties are ‘independent and equal actors, concerned primarily with their own self-interest’.1 The essence of this ethic of individualism2 is that one party’s self-interest is distinct from that of another,3 and each party is ‘free to pursue their own self-interest not only in negotiating but also in performing contracts provided they do not act in breach of a term of the contract’.4 While an individualist accepts5 that some limitations to this freedom are possible and desirable, for example protection from misrepresentation or duress, these are not to undermine but to facilitate the pursuit of one’s interest.6 Alive to the individualism ethic in English law, courts have traditionally been reluctant to police parties’ bargain through adoption of a general principle that would require parties to be loyal to their bargain, or to one another, such as the principle of good faith.7 For example, in contract negotiation a duty of good faith * I am grateful to everyone who commented on a draft of this paper when it was presented at the International Conference on Contracts (KCON XIV) at Tulane Law School (8 and 9 March 2019), and at UCL Conference, ‘The Contents of Commercial Contracts: Terms Affecting Freedoms’ (9 and 10 May 2019), and to Bartosz Kucharski for his helpful observations. I am particularly grateful to Tara Pichardo-Angadi for her insightful commentary at the UCL Conference. All errors mine. 1 Canson Enterprises Ltd v Boughton & Co (1991) 85 DLR (4th) 129, 154 (McLachlin J), cited with approval as reflecting English law in AIB Group (UK) plc v Mark Redler & Co Solicitors [2014] UKSC 58, [2015] AC 1503 [83] (Lord Reed), Main v Giambrone & Law (A Firm) [2017] EWCA Civ 1193, [2018] PCLR 2 [50]–[51] (Jackson LJ with whom Underhill and Moylan LJJ agreed). 2 D Kennedy, ‘Form and Substance in Private Law Adjudication’ (1976) 89 Harvard Law Review 1685, 1713 and see 1713–22 (contrasting individualism with altruism). For a juxtaposition with cooperativism, R Brownsword, Contract Law: Themes for the Twenty-First Century, 2nd edn (2006 OUP) 57. 3 Kennedy, ‘Form and Substance in Private Law Adjudication’ (n 2) 1713. 4 Yam Seng Pte Ltd v International Trade Corp Ltd [2013] EWHC 111 (QB) [124], [2013] 1 All ER (Comm) 1321 [123] (Leggatt J), citing E McKendrick, Contract Law, 9th edn (Palgrave 2011) 221–22. 5 Unlike a pure egotist, who considers any limitations impossible or undesirable: Kennedy, ‘Form and Substance in Private Law Adjudication’ (n 2) 1714. 6 Kennedy, ‘Form and Substance in Private Law Adjudication’ (n 2) 1715. This qualifies the quotation given, cited in the text to n 4. 7 This is notwithstanding the broad statements in early cases that good faith applies in all contracts: see Carter v Boehm (1766) 3 Burr 1905, 1909; 97 ER 1162, 1164 (Lord Mansfield); Mellish v Motteux

66  Magda Raczynska was rejected as ‘inherently repugnant to the adversarial position of negotiating parties’;8 in contract performance it was said to undermine the predictability of outcome and certainty, and to go against the ability to require the exact contractual performance.9 Any ‘demonstrated problems of unfairness’ have become curable through ‘piecemeal solutions’ such as vitiation of unconscionable bargains, regulation of exemption clauses (through statute),10 the rule against penalties11 or estoppel,12 not a general principle of good faith.13 In recent decisions the courts’ concern about parties’ pursuit of self-interest has abated. First, courts have become more willing to imply duties of good faith in contract out of necessity for business efficacy, and have even done this ‘more readily’ (albeit controversially) in a broad, emerging category of long-term (‘relational’) contracts.14 Second, contractual discretions are now controllable through the concept of good faith, alongside irrationality, capriciousness or ­arbitrariness.15 Consistent with those developments, courts also seem to be prepared to give effect to express terms to perform the contract in good faith, even terms to negotiate in good faith.16 While the approach to interpretation of such clauses

(1792) Peake 156, 157; 170 ER 113, 113–14 (Lord Kenyon). See, too, M Lobban, ‘Contractual Fraud in Law and Equity, c1750-1850’ (1997) 17 OJLS 441, 467–68 (noting the development of the seller’s duty in Mellish to disclose all latent defects known to him into an implied term of fitness for purpose in contract of sale of goods). 8 Walford v Miles [1992] 2 AC 128 (HL) 138 (Lord Ackner). 9 See, eg, J Mummery, ‘Commercial Notions and Equitable Potions’ in S Worthington (ed), Commercial Law and Commercial Practice (Hart Publishing 2003) 29; M Bridge, ‘Doubting Good Faith’ (2005) 11 New Zealand Law Quarterly 430, 448 (noting that introducing good faith would be like ‘letting a bull loose in china shop’). 10 See, eg, Insurance Act 2015, discussed by J Lowry and R Edmunds, ‘The Reform of Insurance Warranties: Looking Beyond the Past’, ch 19 in this volume, and Hague Visby Rules, discussed by M Özdel,‘The Right to Delivery of Goods under Contracts of Carriage’, ch 20 in this volume. 11 Cavendish Square Holding BV v Makdessi [2015] UKSC 67, [2016] AC 1172. For discussion of penalty clauses, see W Day, ‘Disproportionate Penalties in Commercial Contracts’, ch 11 in this volume. 12 MWB Business Exchange Centres Ltd v Rock Advertising Ltd [2018] UKSC 24, [2019] AC 119 [16] (Lord Sumption). 13 Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd [1998] All ER 348, [1989] QB 433 (CA) 439 (Bingham LJ). See also R (European Roma Rights Centre) v Immigration Officer at Prague Airport [2004] UKHL 55, [2005] 2 AC 1 [59] (Lord Hope); MSC Mediterranean Shipping Co SA v Cottonex Anstalt [2016] EWCA Civ 789, [2017] 1 All ER (Comm) 483 [45] (Moore-Bick LJ). 14 Yam Seng (n 4) [142]. See section IV.B. 15 Abu Dhabi National Tanker Co v Product Star Shipping (The Product Star) (No 2) [1993] 1 Lloyd’s Rep 397, 404 (Leggatt LJ, with whom Balcombe and Mann LJJ agreed); Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2) [2001] EWCA Civ 1047, [2001] 2 All ER (Comm) 299 [67] (Mance LJ); Paragon Finance plc v Nash [2001] EWCA Civ 1466, [2002] 1 WLR 685 [39]–[41] (Dyson LJ); Socimer International Bank Ltd v Standard Bank London [2008] EWCA Civ 116, [2008] Bus LR 1304 [60] (Rix LJ); Braganza v BP Shipping Ltd [2015] UKSC 17, [2015] 1 WLR 1661 [18] (Baroness Hale, with whom Lord Kerr agreed). See also Property Alliance Group Ltd v Royal Bank of Scotland plc [2018] EWCA 355 [163]. Note R Hooley, ‘Controlling Contractual Discretion’ (2013) 72 CLJ 65 considers that good faith in this context is an umbrella concept capturing the implied controls. 16 See, eg, Petromec Inc v Petroleo Brasileiro SA (‘Petrobas’) [2005] EWCA Civ 891 [106]–[121] (Longmore LJ) (obiter); Emirates Trading Agency LLC v Prime Mineral Exports Private Ltd [2014] EWHC 2104 (Comm), [2015] 1 WLR 1145 [53]. Cf Morris v Swanton Care and Community Ltd [2018] EWCA Civ 2763 and Emagine Films Ltd v Smith Entertainment Ltd [2019] EWHC 2085 (Ch) [54]–[59]

Good Faiths and Contract Terms  67 is somewhat cautious17 and restrictive,18 the fact that they are not struck down as fundamentally objectionable to the nature of a contract is significant.19 At the same time, the question whether English law should adopt a general duty (or principle) of good faith stimulates interest. An important argument in its favour is that other jurisdictions have already explicitly recognised it.20 As Leggatt J (as he then was) observed in his influential first instance judgment in Yam Seng Pte Ltd v International Trade Corp Ltd,21 ‘[i]n refusing … if indeed it does refuse, to recognise any such general obligation of good faith [in the performance of contracts], this jurisdiction would appear to be swimming against the tide’.22 The purpose of this chapter is to consider to what extent English law incorporates a duty of good faith in contract performance through interpretation and implication of terms, when the contract does not expressly provide for good faith.23 In so doing, it asks ‘To what extent is English law going with (or against) the flow?’ To answer this question, the chapter begins by explaining how the concept of good faith is conceived elsewhere in the world, by identifying its different models (section II). It shows, to use Lord Leggatt’s fluvial metaphor, that there are different ‘tides’. This section should be useful independently of its immediate purpose, which is to see, in the remainder of the chapter, how English law measures up against these models. The broader utility of this part lies in showing that, contrary to the popular view, it is wrong to simply think of good faith as setting out a required standard of behaviour, for example a duty to take interests or expectations of another into account, or to act reasonably. Instead, the models ought to be seen as conceptions of good faith that are methodologically different from one another. The remainder of the chapter considers the tools available in the judicial arsenal when interpreting and implying contract terms under English law, with a view to evaluating how, if at all, they might correspond to the

(Kelyn Bacon QC sitting as a Deputy Judge of the High Court). See also G Leggatt, ‘Negotiation in Good Faith: Adapting to Changing Circumstances in Contracts and English Contract Law’ [2019] Journal of Business Law 104. 17 Berkeley Community Villages Ltd v Pullen [2007] EWHC 1130 (Ch) [87], [97] and [110] (Morgan J); CPC Group Ltd v Qatari Diar Real Estate Investment Co [2010] EWHC 1535 (Ch) [246] (Vos J); Mid Essex Hospital Services NHS Trust v Compass Group UK and Ireland Ltd (t/a Medirest) [2013] EWCA Civ 200, [2013] BLR 265 [112] (Jackson LJ). 18 Mid Essex Hospital v Compass (n 17) [106]–[108] (Jackson LJ); Sainsbury’s Supermarkets Ltd v Bristol Rovers (1883) Ltd [2016] EWCA Civ 160 [26], [98]–[99] (Floyd LJ, with whom Laws and McCombe LJJ agreed). 19 See M Bridge, ‘The Exercise of Contractual Discretion’ (2019) 135 LQR 227, 235. 20 See section II. 21 Yam Seng (n 4). The decision was noted by S Whittaker, ‘Good Faith, Implied Terms and Commercial Contracts’ (2013) 129 LQR 463; E Granger, ‘Sweating over an Implied Duty of Good Faith’ [2013] LMCLQ 419; S Bogle, ‘Disclosing Good Faith in English Contract Law’ (2014) 18 Edinburgh Law Review 141. 22 Yam Seng (n 4) [124]. 23 The existence of express terms of good faith militates against implying a more general good faith duty: see Russell v Cartwright [2020] EWHC 41 (Ch) [89] (Falk J). For discussion of express duties of good faith, see PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6 in this volume.

68  Magda Raczynska conceptions of good faith. These tools are: interpretation of contracts with reasonable expectation of honesty (section III), implication of duties of good faith, both in fact and in law (section IV), and implied controls in the exercise of discretionary powers (section V). This part of the chapter shows that while in English law there is no general duty of good faith, there are doctrines that already resemble good faith duties in other jurisdictions, not only functionally but also in the way they operate. Where this happens, the proper (and perhaps surprising) conclusion is that English law already has its own (subtle) doctrines of ‘good faith’ in contract performance. The corollary is that English law has no need for a general principle of good faith.

II.  Four Models of Good Faith While it is interesting to try to examine the various contexts in which ‘good faith’ is used in other jurisdictions24 in contract performance, and then compare them with the developments in English law, this chapter posits that it is more productive for current purposes to try to identify basic models of the duty (or principle) of good faith depending on the way in which this concept responds to parties’ intentions and self-interest. The models are based on the way in which good faith principle in contract operates in various jurisdictions, but they are not meant to capture any one system.25 Before turning to the models, it is useful to observe that in a number of jurisdictions good faith plays an important role in contract law. For example, under Article  1104 of the French Civil Code, ‘[c]ontracts must be negotiated, formed and performed in good faith’. The German Civil Code (BGB) sets out the requirements to perform26 and also to interpret27 contracts ‘in good faith, taking customary practice into consideration’.28 The US Uniform Commercial Code (UCC) similarly provides in §1-304 UCC that ‘every contract or duty within this Act imposes an obligation of good faith in its performance or enforcement’.29 A unanimous Supreme Court of Canada in Bhasin v Hrynew held recently that ‘good faith contractual performance is a general organizing principle of the 24 See R Zimmerman and S Whittaker (eds) Good Faith in European Contract Law (CUP 2000). 25 For a different, and more abstract, mapping of good faith models, depending on the extent to which good faith is required, see Brownsword, Contract Law (n 2) 130–34. 26 §242 BGB. 27 §157 BGB. The two BGB provisions are often cited jointly and without much discrimination. See also discussion in section II.D. 28 See also the Dutch Civil Code, Art 6:2; Italian Civil Code, Art 1137. 29 Note that §1-201 UCC defines good faith as ‘honesty in fact and the observance of reasonable commercial standards of fair dealing’, and that, similarly, §2-103(1)(b) UCC uses ‘good faith’ to mean ‘in the case of a merchant honesty in fact and the observance of reasonable commercial standards of fair dealing in the trade’. See also Restatement (Second) of Contracts, §205, and Wigand v Bachmann-Bechtel Brewing Co (1918) 222 NY 272 (NY Crt of App) 277: ‘Every contract implies good faith and fair dealing between the parties to it’.

Good Faiths and Contract Terms  69 common law of contract’.30 At the international level, various instruments also incorporate a duty of good faith and fair dealing.31 Not all jurisdictions adopt a good faith duty, at least not explicitly. In Scotland, for instance, good faith is not ‘overtly recognised’ but is said to play a substantial role nevertheless.32 In Australia, the status of good faith is more problematic. At the federal level, ‘[t]he question whether a standard of good faith should be applied generally to contracts has not been resolved’.33 State courts have shown some, albeit varied, enthusiasm for a general good faith duty.34

A.  Interpretation Model This model looks at good faith as a factor considered when contract is interpreted. It is intended to describe the way in which to determine the scope of parties’ obligations under the contract, that is, how parties themselves have chosen to limit their self-interest. One version of this model is that courts take a particular standard (such as honesty) into account when they construe express terms and the contract as a whole.35 Proponents of such an understanding of good faith, Carter and Peden, say ‘[a] term in the form “X must act in good faith” should never be implied, for the simple reason that contract already requires that’.36 Good faith here means honesty: a party must act honestly in negotiating and performing the contract, and in exercising discretions and rights. Honesty does not require parties to behave reasonably, except in the sense that parties must not act in a way that no reasonable person would regard as permissible in the circumstances.37 The fact that this view, expressed in the context of Australian contract law, contradicts a number of

30 Bhasin v Hrynew [2014] SCC 71 [33]. Note, however, that in Quebec civil law the fundamental role played by good faith had been recognised for some time: National Bank of Canada (Canadian National Bank) v Soucisse [1981] 2 SCR 339, codified in 1994 in Québec Civil Code, Art 1375. 31 Principles of European Contract Law, Arts 1:201 and 1:202; Principles of International Commercial Contracts, Art 1.7. See also the United Nations Convention on Contracts for the International Sale of Goods 1980, Art 7(1). 32 H MacQueen, ‘Good Faith in the Scots Law of Contract: An Undisclosed Principle’ in ADM Forte (ed), Good Faith in Contract and Property Law (Hart Publishing 1999) 5; H MacQueen and S O’Byrne, ‘The Principle of Good Faith in Contractual Performance: a Scottish-Canadian Comparison’ (2019) 23 Edinburgh Law Review 301, 328. 33 Commonwealth Bank of Australia v Barker [2014] HCA 32 (High Court of Australia) [107] (Kiefel J); and similarly, Royal Botanic Gardens and Domain Trust v South Sydney City Council [2002] HCA 5, 76 ALJR 436 [40], [88] and [155]. 34 See subsequent discussion, text to nn 68–71. 35 E Peden, Good Faith in the Performance of Contracts (Lexis Butterworths Sydney 2003) [6.01]; see also E Peden, ‘When Common Law Trumps Equity: The Rise of Good Faith and Reasonableness and the Demise of Unconscionability’ (2005) 21 Journal of Contract Law 226. 36 JW Carter and E Peden, ‘Good Faith in Australian Contract Law’ (2003) 19 Journal of Contract Law 155, 171. 37 ibid 168.

70  Magda Raczynska decisions in Australia38 is not problematic for current purposes. However, what is difficult with this version of the model is to explain on what basis contract parties are understood to behave honestly. A possible explanation, based on individualist ethic, might be that honesty is necessary for parties to co-exist and for each to pursue its own self-interest. A more refined version of the interpretation model is that good faith is a particular attitude that parties adopt to the obligations undertaken. Such an understanding of good faith has been suggested by Daniel Markovits.39 He considers the duty of good faith to be an attitude of taking contractual duties appropriately seriously.40 Each contracting party is allowed to be self-interested and is only under a duty to take the other party’s expressed intentions into account – but only at face value. This requires a party only to ‘go the distance and along the path’ the contract specifies,41 and does not determine the content of parties’ duties.42 On this view, as on that of Carter and Peden, the duty of good faith does not add to the obligations under the contract and it does not alter the terms of contract to fit any ideal. A possible criticism is that this model of good faith says nothing more than that contracts must be kept. However, just because a conception of good faith is narrow does not necessarily mean it is unhelpful.43 The advantage of the refined version of the model is that it explicitly accommodates good faith within the individualist ethic: a party is free to pursue its self-interest under the contract in exactly the same way as it would without it, except that the party must accept that its self-interest is limited by what it agreed to do.44 The limit on the self-interest is determined by the ‘best interpretation’ of what the party agreed to,45 although it remains somewhat unclear what this means in practice, and how much context the court is to take into account when interpreting the contract.

B.  Purpose-based Model The purpose-based model of good faith emphasises the idea that rights conferred by contract perform a particular purpose, whether set out in the contract or recognised by the legal system as a social or economic purpose of a particular type of contract. It is bad faith to assert one’s contractual rights (or exercise one’s powers) in a way contrary to that purpose. This is captured by the doctrine of abuse of a 38 See section II.C. 39 D Markovits, ‘Good Faith as Contract’s Core Value’ in G Klass, G Letsas and P Saprai (eds), Philosophical Foundations of Contract Law (OUP 2014) 272. 40 ibid 284. 41 ibid 283. 42 ibid 279, citing the Official Comment to UCC 1-304 cmt [1], which states that good faith ‘does not create a separate duty of fairness and reasonableness which can be independently breached’. 43 See H Lücke, ‘Good Faith and Contractual Performance’ in PD Finn (ed), Essays on Contract (Law Book 1987) 155, 161. 44 Markovits, ‘Good Faith as Contract’s Core Value’ (n 39) 278, 284. 45 ibid 284.

Good Faiths and Contract Terms  71 right or a power46 (l’abus de droit),47 which exists in a number of jurisdictions.48 In French law, where it developed, the effect of this doctrine is sometimes thought to control the otherwise broad judicial discretion conferred by a good faith principle (to control good faith);49 the adjudication model that confers broad judicial discretion is described later in this chapter.50 Louis Josserand, a French scholar who developed much of the doctrine, argued that every right must have teleological or social limits, because human beings are only capable of holding rights by virtue of being members of an organised community (it makes no sense to say that an isolated human being has rights).51 Therefore, asserting a right contrary to its social function, ‘contrary to the spirit’ (or essence) of the right, or with the intention to harm is unlawful.52 The duty to act in good faith is intertwined here with the duty to act in accordance with the parties’ intentions and with a proper purpose.53 Working out whether a right has been abused involves a comparison between, on the one hand, the motive and purpose with which the right-holder asserts her right and, on the other hand, the ‘spirit’ and purpose of the right. The doctrine of abuse of right has attracted various criticisms. One is that abuse of a right simply amounts to a tort,54 so the doctrine is redundant and meaningless. In this vein, Marcel Planiol, a contemporary of Josserand, argued that ‘assertion of 46 The discussion of abuse of rights does not generally distinguish rights from powers in the Hohfeldian sense (see WN Hohfeld, ‘Some Fundamental Legal Conceptions as Applied in Judicial Reasoning’ (1913) 23 Yale Law Journal 16), but it is clear that both rights and powers can be abused. Reference to abuse of right in this chapter also refers to abuse of power. 47 See the French decision Colmar, 2 My 1855, Recueil Dalloz 1856.II.9, 10. For modern discussions of the doctrine in private law, see eg HC Gutteridge, ‘Abuse of Rights’ (1933–1935) 5 CLJ 22; A Gambaro, ‘Abuse of Rights in Civil Law Tradition’ (1995) 3 European Review of Private Law 561; J Gordley, ‘The Abuse of Rights in the Civil Law Tradition’ in R de la Feria and S Vogenauer (eds), Prohibition of Abuse of Law: A New General Principle of EU Law? (Hart Publishing 2011) 33; A di Robilant, ‘Abuse of Rights: The Continental Drug and the Common Law’ (2014) Boston University School of Law, Public Law Research Paper No 14–28, available at papers.ssrn.com; R Jabbour, La Bonne Foi dans l’Exécution du Contrat. Bibliothèque de Droit Privé (LGDJ 2016) paras 167–70; P Stoffel-Munck, L’Abus dans le Contrat, Essai d’une Théorie (LGDJ 2000). 48 See, eg, German Civil Code, §226; Swiss Civil Code, Art 2; Spanish Civil Code, Art 7; Polish Civil Code, Arts 5, 58 and 354; the Japanese Supreme Court in the 1972 case of Mitamura v Suzuki, 26 Saiko Saibansho minji hanreishu 1067 (Sup Ct, 27 June 1972), cited in K Sono and Y Fujioka, ‘The Role of the Abuse of Right Doctrine in Japan’ (1975) 35 Louisiana Law Review 1037, 1037. 49 See M Fabre-Magnan, ‘L’Obligation de Motivation en Droit des Contrats’ in Études Offertes à Jacques Ghestin. Le Contrat au Début du XXI siècle (LGDJ 2001) 301. 50 See section II.D. 51 In relation to the non-absolute nature of rights, Josserand was influenced by R von Jhering, Der Zweck im Rechts (Britkopf & Härtel 1877): see ch 1 in this volume. 52 L Josserand, De l’Esprit des Droits et de Leur Relativité: Théorie Dite de l’Abus de Droits, 2nd edn (1939, reprinted Dalloz 2006) 245; L Josserand, Cours de Droit Civil Positif Français, II: Théorie Générale des Obligations, 3rd edn (Sirey 1939) 246–47, para 428. 53 C Demolombe, Traité des Contrats et des Obligations Conventionnelles en Général (1868) para 393, available at http://bit.ly/1BWubaW; D Houtcieff, Droit des Contrats, 4th edn (Bruylant 2018) para 636. 54 This argument works in Civilian jurisdictions, like French law, where there is a general duty not do harm, and a duty to compensate any harm that has occurred as a result of one’s fault: Art 1240 (former Art 1382) of the French Civil Code. See also Gordley, ‘The Abuse of Rights in the Civil Law Tradition’ (n 47) 42.

72  Magda Raczynska a right ceases where abuse commences’.55 The criticism is unconvincing because, first, not every abuse of right causes harm. Second, even on Planiol’s view, it must remain true to say that abuse of a right presupposes that there once was a right, and in that sense the abuse of right doctrine is a mechanism that controls the exercise of the right.56 The more trenchant critique is that for the abuse of right doctrine to apply, one has to determine the purpose the right-holder had in mind when he or she acted, in order to compare it with the social purpose of the right, and establishing subjective belief in law is hard. Crucially, however, the difficulty with establishing the right-holder’s state of mind does not undermine the rationale for the abuse of rights doctrine; it merely impacts on how it applies. Given that the doctrine constitutes a court’s intervention in the assertion of rights, there are good reasons why the doctrine should not apply too broadly.57 It is interesting to note, however, that in modern practice in France, the consideration of whether a right has been abused appears to be a process of ­weighing a range of factors, rather than a search for the social purpose of the right. For ­example, in working out whether there has been an abuse of a power to unilaterally fix the price of goods, conferred under a master agreement,58 courts consider the behaviour of the power-holder, the contract terms and the actual amount of the increase.59 The power will not generally be abused just because the power-holder makes substantial profit,60 but it will likely be abused if an excessively disproportionate profit is made,61 or if the buyer was not given time to search for an alternative.62

C.  Implied Duty Model When English lawyers think of general good faith duty in other jurisdictions, they probably usually think of implied duties. On this model, a duty is added to a particular contract and necessarily overrides the freedom to pursue one’s self-interest. The main problem with this model is that, despite its supporters,63 there is generally little explanation of its basis, and why in particular circumstances one 55 M Planiol, Traité Élémentaire de Droit Civil, 11th edn (LGDJ 1928) 161, para 91. See also G Ripert, La Règle Morale dans les Obligations Civiles, 3rd edn (LGDJ 1935) 90. 56 Fibre-Magnan, ‘L’Obligation de Motivation en Droit des Contrats’ (n 49) para 323. 57 See Gutteridge, ‘Abuse of Rights’ (n 47) 35. 58 See French Civil Code, Arts 1164–1165. 59 Houtcieff, Droit des Contrats (n 53) para 391–4. 60 Cour d’appel de Limoges, chambre civile, 29 septembre 2015, No de RG 14/01430. 61 Cass (chamber civile 3) 1er octobre 2014, no 13-17114. 62 Cass (1ère) 30 juin 2004, 30 juin 2004, no 01-00.475 Bull I, no 190, RTD civ 2004, 749. 63 Tymshare Inv v Covell, 727 F.2d 1145, 1152 (DC Cir 1984); EA Farnsworth, ‘Good Faith in Contract Performance’ in J Beatson and D Friedmann (eds), Good Faith and Fault in Contract Law (OUP 1995) 153, 161; and T Rakoff, ‘The Implied Terms of Contracts: Of “Default Rules” and “Situation-Sense”’ in Beatson and Friedmann (eds), Good Faith and Fault 191.

Good Faiths and Contract Terms  73 party’s self-interest should be overridden. It seems to be generally thought that restricting a party’s self-interest is justified if one party’s interest ‘[weighs] much heavier’,64 that is, one party is more vulnerable than another. Such a general justification would be problematic in English law given its rejection of the ‘inequality of bargaining powers’ ethic of transacting.65 That said, there are some analyses that seek to justify the implication of terms. Todd Rakoff, for example, suggested a clearer and structured model, a ‘situation-sense’ approach, which centres on exchange justice.66 This model seems to be reflected in some Australian decisions.67 For example, the NSW Court of Appeal has expressly accepted the implied duty of good faith in the performance of contractual duties and powers in commercial contracts.68 As Sheller JA said in Alcatel Australia Ltd v Scarcella, ‘in New South Wales a duty of good faith, both in performing obligations and exercising rights, may by implication be imposed upon parties as part of a contract’.69 Implication of duties of good faith is clearly far-reaching, and it may be the reason why courts in Australia have been more cautious with implying the duty of good faith in all commercial contracts,70 and the High Court of Australia has yet to rule that good faith is necessarily an implied term of all contracts.71 It is less clear that the recognition of ‘an organizing principle of good faith’ in Bhasin v Hrynew72 fits in this model. The Supreme Court of Canada held that good 64 D Friedmann, ‘Good Faith and Remedies for Breach of Contract’ in Beatson and Friedmann (eds), Good Faith and Fault (n 63) 399, 400; see also J Beatson, ‘Public Law Influences in Contract Law’ in Beatson and Friedmann (eds), Good Faith and Fault (n 63) 263, 288. 65 National Westminster Bank plc v Morgan [1985] AC 686 (HL) 708 (Lord Scarman, with whom Lords Keith, Roskill, Bridge, Brandon agreed, questioning the need for developing a general principle of relief against inequality of bargaining power, set out by Lord Denning MR in Lloyds Bank Ltd v Bundy [1975] QB 326, 339, and noting (at 709) the court’s equitable jurisdiction to grant relief against undue influence when the transaction is unconscionable; Knatchbull-Hugessen v SISU Capital Ltd [2014] EWHC 1194 (QB)). 66 Rakoff, ‘The Implied Terms of Contracts’ (n 63). 67 But see the critique in Peden, Good Faith (n 35). 68 Renard Constructions (ME) Pty Ltd v Minister for Public Works (1992) 26 NSWLR 234, 261 [95] (Priestley LJ); United Group Rail Services Ltd v Rail Corporation New South Wales (2009) 74 NSWLR 61 [61] (Allsop P with whom Ipp JA and Macfarlan JA agreed). See also Paciocco v Australia and New Zealand Banking Group Ltd (2012) 236 FCR 199 (Full Federal Court) [287] (Allsop CJ) and Marmax Investments Pty Ltd v RPR Maintenance Pty Ltd (2015) 237 FCR 534, [2015] FCAFC 127 (Full Federal Court) [146]–[148]. 69 Alcatel Australia Ltd v Scarcella (1998) 44 NSWLR 349, 369 (Sheller JA with whom Powell JA and Beazley JA agreed). 70 CGU Workers Compensation (NSW) Ltd v Garcia (2007) 69 NSWLR 680, 704; [2007] NSWCA 193 [132] (Mason P with whom Hodgson and Santow JJA agreed) and [168] (Santow JA with whom Hodgson JA agreed). Similarly, see Esso Australia Resources Pty Ltd v Southern Pacific Petroleum NL [2005] VSCA 228 (Victoria Court of Appeal) [25] (Buchanan JA, with whom Warren CJ and Osborn AJA agreed); and Special Diagnostic Services Pty Ltd v Healthscope Ltd (2012) 41 VR 1 (Victoria Court of Appeal) [86]–[87]. 71 See n 33. 72 Bhasin v Hrynew (n 30) [33], [62]–[66], [71], [73], [92] (Cromwell J delivering judgment for the Court). See also J Robertson, ‘Good Faith as an Organizing Principle in Contract’ (2015) 93 Canadian Bar Review 809.

74  Magda Raczynska faith is the principle underpinning and informing various rules73 that contractual duties are to be performed ‘honestly and reasonably and not capriciously or arbitrarily’.74 Although the principle was said to constitute an independent and creative force, which resembles the basis for judicial power to imply terms discussed here, its thrust is the provision of a standard of conduct that is meant to help understand and develop the law in a coherent and principled way.75 This independent force of the organising principle of good faith was (rightly) not sufficient to justify the implication of a duty to cooperate, to renegotiate or to reallocate profits in a joint venture in Churchill Falls,76 after a long-term contract to sell electricity at fixed prices became unprofitable for the hydroelectric plant because the market prices of electricity increased considerably over time.

D.  Adjudication Model The last model involves the use of good faith as the basis for a judicial power to adjudicate in the absence of formal doctrinal basis for relief. The good faith duty is the direct basis for courts’ jurisdiction to control the parties’ behaviour, for example by rendering the assertion of rights or exercise of powers invalid, or by explicitly devising effects different from those that parties intended. The model captures the concepts of abuse of right77 and implication of terms. At its broadest, courts appear to have the power to ‘supplement’ the contract when they identify a gap, which they can do taking into account a wide range of factors: the contract’s purpose, content and context. A well-known example is the German doctrine of good faith, Treu und Glauben (meaning literally ‘fidelity and faith’78), encompassed by §157 and §242 BGB.79 The BGB generally reflects the prevailing view of the German jurists who wanted the judge to have a broad power to reject claims he or she considered inequitable.80 For example, in the 1954 decision of the German Federal Court in

73 ibid. Cromwell J referred to the following doctrines applicable to contracts that have traditionally involved aspects of good faith: unconscionability (at [42]), estoppel (at [88]), aspects of civil fraud (at [88]), implication of duties of good faith as a matter of law in contracts of employment (at [54]), insurance (at [55]) and construction of tenders (at [56]). 74 ibid [63]. Cromwell J did not explain the scope of the creative force of good faith, which is criticised by C Hunt, ‘Good Faith Performance in Canadian Contract Law. Case and Comment’ (2015) 74 CLJ 4, 7. 75 Bhasin v Hrynew (n 30) [62]. 76 Churchill Falls (Labrador) Corp v Hydro-Québec 2018 SCC 46, on an appeal from Québec. 77 See, eg, B Jalluzot, La Bonne Foi dans les Contrats, Étude Comparative de Droit Français, Allemand et Japonais (Dalloz 2001) 432, para 1503. 78 S Whittaker and R Zimmermann, ‘Good Faith in European Contract Law: Surveying the Landscape’ in Whittaker and Zimmermann (n 24) 7, 18. 79 It is not always clear whether the court bases its decision on §157 or §242 BGB, but for practical purposes this does not matter: ibid 29. 80 F Regelsberger, Pandekten (Duncker & Humblot 1893) and H Dernburg, System des Römischen Rechts, 8th edn (HW Müller 1911) vol 1, §118, both cited in Gordley, ‘The Abuse of Rights in the Civil Law Tradition’ (n 47) 35.

Good Faiths and Contract Terms  75 the case of two doctors who agreed to exchange their practices.81 Nine months later, the defendant wished to return to his old town and set up a new practice there, and the claimant wished to stop him. As the contract was silent on the question whether the defendant doctor could return, the Court found there was a gap, and it used the purpose of the contract to fill it (the purpose being the establishing of the doctors at their new practices, ie enabling each doctor to consolidate relations with the former doctor’s patients, which required time). Thus, the Court supplied a term that the right to return was qualified in light of the contract’s purpose and the objective criteria stated in §157 BGB. Another example is the use of good faith to adjust the amount owed to discharge an obligation in the Reichsmark case.82 There was massive hyperinflation in 1923 in Germany. The Supreme Court of the German Reich, relying on §242 BGB, refused to allow the debtor to discharge its obligation by paying the nominal value in paper marks, and refused to compel the mortgagee to consent to taking the mortgage securing the obligation off the register. At the heart of this model is the creative and active nature of good faith. To say that a jurisdiction has adopted this model of good faith in private law generally means that courts are vested with the power to control the exercise of legal rights, whether in contract, property law or legal process, for example to restrain reliance on rights that is harmful or which contradicts accepted principles. In so doing, courts can override a party’s self-interest. This model of good faith is reflected in Civilian legal systems whose law is, to a large extent at least, code-based (statutebased). In such systems, without the principle of good faith, it becomes difficult to go beyond what a statute says, or to apply it in a way that achieves justice in a particular case. The good faith principle helps prevent an argument, which otherwise could arise, that there should be no remedy because the code does not provide for it. A broad ‘good faith’ principle (itself provided by the code) can operate to legitimise courts’ power to soften the letter of the statute.83 Whether courts are quick to exercise this power, and in what circumstances they will do so, is another matter. It follows that the adjudication model of good faith is both broader and thinner than sometimes thought. It is broader because courts have the power to override the freedom to pursue a party’s self-interest; and thinner because it offers little detail on when the self-interest is subordinated. This means that whilst the courts in a given jurisdiction may have the power to override the freedom to pursue a party’s self-interest, it is hard to gather (from the face of the principle) whether and if it will be exercised. Courts could, therefore, in theory apply it in a way that is much closer to one of the narrower models discussed earlier.84 The question addressed in the sections that follow is whether the way in which courts approach contract terms under English law aligns with any of those models. 81 BGH, 18 December 1954. 82 Judgment of 28 November 1923, RGZ 107, 78.V Civil Senate. 83 There is some similarity between this model of good faith and the equitable doctrine of unconscionability at common law, see M Harding ‘Equity, and the Rule of Law’ (2016) 132 LQR 278. 84 See sections II.A–C.

76  Magda Raczynska

III.  Interpretation of Contract in Light of Reasonable Expectations of Honesty In a number of decisions, English courts have emphasised the need to uphold reasonable expectations of honesty. Lord Steyn in particular made this point frequently, both in his judicial capacity85 and writing extrajudicially,86 but the theme has appeared in more recent decisions too.87 Prima facie, reasonable expectation of honesty resembles the ‘interpretation with a standard’ model of good faith, at least in the sense of recognising a general requirement of honesty in contract, as described by Peden and Carter.88 However, a closer look at the judicial dicta suggests that there is little basis for such a conclusion. In First Energy v Hungarian International Bank, Steyn LJ (as he then was) said: A theme that runs through our law of contract is that the reasonable expectations of honest men must be protected. It is not a rule or principle of law. It is the objective which has been and still is the principal moulding force of our law of contract. It affords no licence to a judge to depart from binding precedent. On the other hand, if the prima facie solution to a problem runs counter to the reasonable expectations of honest men, this criterion sometimes requires a rigorous re-examination of the problem to ascertain whether the law does indeed compel demonstrable unfairness.89

The point Steyn LJ makes here is that the law protects reasonable expectations of honesty by directing the judge how to apply the existing rules and doctrines, if these rules and doctrines themselves allow for such an application, not by imposing a duty on the contract parties to behave in a particular way. Commenting on this passage, Bradgate compared protection of reasonable expectations to ‘a litmus test by which to judge the outcomes of individual cases’ as correct or not.90 If the application of rules and doctrines is not consistent with reasonable commercial expectations, the court is invited to consider distinguishing an apparent precedent to produce ‘a result more in accordance with commercial expectation’.91 If the precedent cannot be distinguished, it applies nevertheless. Inconsistency with

85 Associated Japanese Bank (International) Ltd v Credit du Nord SA [1988] 3 All ER 902, 903 (Steyn J); G Percy Trentham Ltd v Archital Luxfer Ltd [1993] 1 Lloyd’s Rep 25 (CA) 27 (Steyn LJ, with Ralph Gibson and Neill LJJ concurring); First Energy (UK) Ltd v Hungarian International Bank [1993] 2 Lloyd’s Rep 194 (CA) 196. 86 J Steyn, ‘The Role of Good Faith and Fair Dealing in Contract Law: a Hair-Shirt Philosophy?’ [1991] Denning Law Journal 131; J Steyn, ‘Contract Law: Fulfilling the Reasonable Expectations of Honest Men’ (1997) 113 LQR 433. 87 See Petromec Inc v Petroleo Brasileiro SA Petrobas [2005] EWCA Civ 891, [2006] 1 Lloyd’s Rep 161 [121] (Longmore LJ). 88 See text to nn 35–37. 89 First Energy (n 85) 196. 90 R Bradgate, ‘Contracts, Contract Law and Reasonable Expectations’ in S Worthington (ed), Commercial Law and Commercial Practice (Hart Publishing 2003) 651, 667–68 and 669. 91 ibid 670.

Good Faiths and Contract Terms  77 reasonable commercial expectations might then, at most, operate as an argument in favour of reform of the particular rule or doctrine.92 While English law does not require parties to behave in accordance with reasonable commercial expectations of ‘honest men’, in practice it sometimes looks as if it does. To illustrate this, let us consider the courts’ approach to contractual interpretation. If the language of contract is capable of more than one construction, ‘the court is entitled to prefer the construction which is consistent with business common sense [the commercial purpose of the agreement] and to reject the other’.93 If the language is unambiguous, there is no scope for the court to do so, even if the result is uncommercial.94 However, much turns on whether a term is considered ambiguous, and ‘language is a very flexible instrument’.95 For example, in HIH Casualty v Chase Manhattan Bank,96 a contract of insurance provided that the insured should have ‘no liability of any nature to the insurers for any information provided’. The literal interpretation suggested that the clause covered liability for deceit where the insured’s agent had dishonestly provided information known to be false. This interpretation was rejected, because parties must have contracted on an assumption of honesty and good faith,97 and so could not have intended to exclude their liability for deceit. Lord Bingham said that ‘[p]arties entering into a commercial contract … will assume the honesty and good faith of the other; absent such an assumption they would not deal’.98 This is said to reflect the basic principle that ‘the law does not expect people to arrange their affairs on the basis that others may commit fraud’.99 Contracts where parties do not expect honesty are likely to be very rare; indeed, an expectation of honesty is ‘[a] paradigm example of a general norm which underlies almost all contractual relationships’.100

92 ibid. 93 Rainy Sky v Kookmin Bank [2011] UKSC 50, [2011] 1 WLR 2900 [21] and see, too, [30] (Lord Clarke); Anzen Ltd v Hermes One Ltd [2016] UKPC 1, [2016] 1 WLR 4098 [32] (Lord Mance and Lord Clarke on behalf of the Board); Steyn, ‘Contract Law: Fulfilling the Reasonable Expectations of Honest Men’ (n 86) 441. 94 Rainy Sky (n 93) [23] (Lord Clarke); Cooperative Wholesale Society Ltd v National Westminster Bank plc [1994] 11 WLUK 256 (CA), [1995] 1 EGLR 97; First National Trustco (UK) Ltd v Page [2019] EWHC 1187 (Ch) [81] (Joanna Smith QC sitting as a Deputy Judge of the High Court). 95 Cooperative Wholesale Society (n 94) 99 (Hoffmann LJ). 96 HIH Casualty v Chase Manhattan Bank [2003] 1 All ER (Comm) 349 (HL), [2003] 2 Lloyd’s Rep 61. 97 ibid [15] (Lord Bingham) and see [68] (Lord Hoffmann): ‘in the absence of words which expressly refer to dishonesty, it goes without saying that underlying [the contract] … there will be a common assumption that the persons involved will behave honestly’. 98 ibid [15]. 99 See Takhar v Gracefield Developments Ltd [2019] UKSC 13, [2019] 2 WLR 984 [44] (Lord Kerr with whom Lord Hodge, Lord Lloyd-Jones and Lord Kitchin agreed). 100 Yam Seng (n 4) [135] and see also [137]–[139] (Leggatt J). See also Wales (t/a Selective Investment Services) v CBRE Managed Services Ltd [2020] EWHC 16 (Comm) [66] (HHJ Halliwell sitting as a Judge of the High Court) (referring to the claimant’s expectation of honesty as placing the defendant under ‘an implied obligation’ to act honestly, but, as just explained, the line of cases discussed in this section does not go quite so far).

78  Magda Raczynska The reasoning in HIH Casualty comes close to the interpretation model of good faith advocated by Peden and Carter, as it resembles the requirement of honesty in contract interpretation.101 But it also differs from that model, because it does allow the parties, at least in theory, not to expect honesty from one another so long as they say so expressly: words in a contract relieving a party of liability ‘will not extend to a case of fraud unless fraud is expressly mentioned’.102 As a matter of policy, it is appropriate that contracts are interpreted on the presumption that parties mean and expect honesty. But the broader point of principle is that parties are free not to intend or expect honesty. At best, this amounts to a weak and thin form of Peden and Carter’s version of the interpretation model of good faith. It is weak because the standard is only rebuttably presumed; and thin because its content is only that parties expect honesty, not that they, for example, perform reasonably or take another’s interests into account.

IV.  Implication of Good Faith The fact that courts have the power103 in English law to imply terms in contract, whether in fact or in law, functionally corresponds to the broad (fourth) conception of good faith in other jurisdictions.104 In that sense, a good faith doctrine already exists under English law. The difficult issue is to what extent a duty of good faith, understood as a standard of behaviour, can itself be implied in contracts. The law in this area has not yet stabilised. This section outlines the concerns with implying such duties, before considering which model English law aligns itself with.

A.  Implication in Fact In Yam Seng, Leggatt J (as he then was) observed that there is ‘no difficulty, following the established methodology of English law for the implication of terms in fact, in implying such a duty [of good faith] in any ordinary commercial contract’.105 He understood a duty of good faith to comprise honesty (the ‘core value’ of good 101 See text to n 36. 102 Bank of Beirut SAL v HRH Prince Adel El-Hashemite [2015] EWHC 1451 (Ch), [2016] Ch 1 [94] (Nugee J). 103 For dicta that implication of terms is a matter of the exercise of the courts’ power, see Liverpool City Council v Irwin [1977] AC 239, 253–54, 257, 262 and 265; Barratt Southampton Ltd v Fairclough Building Ltd [1988] 5 ELUK 215, (1988) 27 Con LR 62, 68–69 (Bowsher J) (that terms are implied ‘by way of judicial legislation’); Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472, 481 (Sir Thomas Bingham MR) (referring to implication of terms as ‘the exercise of … extraordinary power’ that is subject to ‘strict constraints’). 104 Section II.D. 105 Yam Seng (n 4) [132]. The issue of relational contract is discussed in section IV.B, in the context of implication in law.

Good Faiths and Contract Terms  79 faith), observance of other norms and standards of commercial dealing (whether generally accepted or specific to a particular trade or commercial activity),106 and fidelity to the parties’ bargain.107 It is therefore clear that good faith includes more than a duty to behave honestly.108 The key aspect of the test for implying terms in fact109 is whether the term is necessary to give business efficacy to the contract110 and/or so obvious that ‘it goes without saying’.111 It is not entirely clear why a requirement of honesty112 was obvious and necessary in Yam Seng. This may be because at the time the decision in Yam Seng was handed down, implication of terms (in fact) and interpretation of contract were seen as part of the same exercise,113 and the judge clearly treated them as such,114 taking time to look at honesty as part of background when interpreting the contract.115 It is now clear that implication of terms (in fact) and interpretation are logically distinct:116 interpretation is concerned with construing the words used in the contract; implication with inserting what has not been expressly agreed.117 Implying good faith in fact would now require an explanation of its necessity and/or its obviousness.118 It is difficult to envisage the circumstances in which a particular contract would be commercially unworkable unless a good faith duty were implied. Indeed, it is suggested that the process of implication of terms in fact (as currently understood) seems, as a matter of principle, incompatible with implying a duty of good faith. Both necessity and obviousness are judged with reference to a notional reasonable person in the position of the parties, knowing the surrounding circumstances119 at the time of contracting.120 At the very least, a reasonable person must be able to tell 106 Yam Seng (n 4) [135] and [139]. 107 ibid [140], although on the facts the judge’s comments are limited to honesty: ibid [138]. See also Wales v CBRE (n 100) [66] (HHJ Halliwell sitting as a Judge of the High Court). 108 Bates v Post Office Ltd (No 3: Common Issues) (Bates) [2019] EWHC 606 (QB) [710] (Fraser J). 109 See, eg, Marks & Spencer plc v BNP Paribas Securities Services (Marks & Spencer) [2015] UKSC 72, [2016] AC 742 [18]–[21] (Lord Neuberger, with whom Lord Sumption and Lord Hodge agreed). 110 The Moorcock [1886–90] All ER Rep 530, (1889) 14 PD 64 (CA); Reigate v Union Manufacturing Co [1918] 1 KB 592 (CA) 605 (Scrutton LJ); Arnold v Britton [2015] UKSC 36, [2015] AC 1619 [112] (Lord Carnwath). 111 See, eg, Shirlaw v Southern Foundaries (1926) Ltd [1939] 2 KB 206 (CA) 227 (MacKinnon LJ) (affirmed [1940] AC 701 (HL)); Arnold v Britton (n 110) [112] (Lord Carnwath). 112 See n 107. 113 Attorney-General of Belize v Belize Telecom Ltd [2009] UKPC 10, [2009] 1 WLR 1988 (PC) [16]–[21], [27] (Lord Hoffmann) (with whom Baroness Hale, Lord Rodger, Lord Carswell and Lord Brown agreed). 114 Yam Seng (n 4) [133]. 115 ibid [136]. 116 Marks & Spencer (n 109) [26]–[27], [29] (Lord Neuberger). See too Greenhouse v Paysafe Financial Services Ltd (Greenhouse) [2018] EWHC 3296 (Comm) [12] (Andrew Burrows QC sitting as a Judge of the High Court); Parker v Roberts [2019] EWCA Civ 121, [2019] 2 P&CR 16 [88] (Lewison LJ). 117 Greenhouse (n 116) [12] (Andrew Burrows QC sitting as a Judge of the High Court). 118 The tests can be alternative: Marks & Spencer (n 109) [21] (Lord Neuberger); Greenhouse (n 116) [12] (Andrew Burrows QC sitting as a Judge of the High Court). 119 In relation to obviousness, this is made clear in Equitable Life Assurance Society v Hyman [2002] 1 AC 408 (HL) 459 (Lord Steyn); Marks & Spencer (n 109) [21] and [23]. 120 Philips Electronique v British Sky (n 103) 482 (Bingham MR); Marks & Spencer (n 109) [23].

80  Magda Raczynska what the term means: it must be capable of being clearly expressed.121 The issue is not that ‘good faith’ might mean a standard of behaviour: there are examples where a term was implied in fact in order to supply a particular standard of behaviour (eg the duty of a surgeon undertaking to operate on a patient to exercise reasonable care and skill in doing so122). The issue is that ‘good faith’ could mean anything from bare honesty in disclosure when asked, to taking the other party’s interests into account.123 When implying a duty of good faith in fact, the court would need to specify what it means in a particular context; in which case, why not imply the more specific term instead? The case of Yam Seng itself is illuminating. On the facts, the distributorship agreement between International Trade Corporation (A) and Yam Seng (B) for Manchester United products required parties to communicate and cooperate with one another. A was to plan production, taking into account B’s expected demand. B was to market and sell the product. Leggatt J observed this meant that B could have argued that it (B) expected A would keep B informed about its (A’s) ‘best estimate’ of the supply levels and about any material change to its supply levels.124 B did not argue the case in this way, so the point was obiter, but we should notice that the judge made it reasonably clear that if a duty in this context were to be implied, the content of the implied duty would be to volunteer particular information, rather than a (broader) duty of good faith. Could, in theory, both duties (of good faith and, for example, duty to disclose) be implied? This is problematic for the process of implication of terms in fact, because a term is only implied in fact if it can be said, without doubt, that it is the preferred solution.125 If there are two overlapping but independent terms to be implied, each one is necessarily not preferred over the other. If, on the other hand, good faith is a ‘background’ duty, underpinning the specific duty of ­disclosure, good faith would be neither necessary nor obvious. This means that if the duty of good faith were to be implied in fact, it would have to be founded on an alternative basis to the tests of necessity/obviousness. There is no basis in law at present. Although as a matter of principle implying a good faith duty appears to be irreconcilable with the process of implying terms in fact, it remains possible as a matter of authority. It follows that the circumstances in which a court will consider a duty of good faith as obvious or necessary for the business efficacy of a particular 121 BP Refinery (Westernport) Proprietary Ltd v Shire of Hastings 180 CLR 266, (1977) 16 ALR 363 (PC). 122 Philips Electronique v British Sky (n 103) 481 (Sir Thomas Bingham MR). 123 See Russell v Cartwright (n 23) [90] (Falk J). 124 Yam Seng (n 4) [144]. 125 This means, eg, that if it is possible that a particular term was deliberately omitted, it cannot be implied: Atkins International HA v Islamic Republic of Iran Shipping Lines (The APJ Priti) [1987] 2 Lloyd’s Rep 37, 42 (Bingham MR); Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472 (CA) 482 (Bingham MR); and see Marks & Spencer (n 109) [38]–[40] (Lord Neuberger). It also means that a term should not be implied because it appears fair: Marks & Spencer (n 109) [21] (Lord Neuberger).

Good Faiths and Contract Terms  81 contract are likely to be very rare. This is supported by a spate of recent cases in which implication in fact of a duty of good faith was considered but was ultimately rejected.126 Courts are right to be circumspect about implying a term of good faith into a commercial contract’.127 To make matters worse, it does not seem easy to align implication of a good faith duty in fact with any of the good faith models identified in section II. The most likely ‘candidate’ is the ‘implied duty’ model of good faith, because we are talking about adding a duty. However, as has been seen,128 this model has shaky foundations, and those that emerge point towards the vulnerability of a party or the justice of the transaction, neither of which corresponds with the tests of business efficacy or obviousness required under current law.

B.  Implication in Law of Good Faith Implication of terms in law is based on the nature of the agreement, and courts have implied terms in law as an incident of certain common relationships.129 Good faith is readily implied in law in certain types of contract. Examples include insurance contracts130 and partnership agreements.131 Those contracts are well-defined, and implied duties of good faith address a particular issue in the circumstances, such as assessment of risk, for which disclosure of information is needed, or one’s commercial and legal status, which can be affected by dealings by a partner. Courts are also willing to find duties of good faith by analogy to those types of relationship. For example, in Nathan v Smilovitch (No 2), the court was willing to imply

126 See eg Hamsard 3147 Ltd v Boots UK Ltd [2013] EWHC 3251 (Pat) [84]–[85] (Norris J); UTB LLC v Sheffield United Ltd [2019] EWHC 2322 (Ch) [213] (Fancourt J); Wales v CBRE (n 100) [67] (HHJ Halliwell sitting as a Judge of the High Court); Russell v Cartwright (n 23) [89] (Falk J). See also Davies, ‘Excluding Good Faith and Restricting Discretion’ (n 23). 127 SDI Retail Services Ltd v The Rangers Football Club Ltd [2019] EWHC 1929 (Comm) [82] (Lionel Persey QC sitting as a Judge of the High Court): ‘considerable care needs to be taken before implying a term of good faith into a commercial contract’. 128 Section II.C. 129 Geys v Société Générale [2012] UKSC 63, [2013] 1 AC 523 [55] (Baroness Hale). 130 Carter v Boehm (1766) 3 Burr 1905, 1909–1910; 97 ER 1162, 1164 (Lord Mansfield); Wolff v Horncastle (1798) 1 Bos & P 316, 322; 126 ER 924, 928 (Buller J) (development of the term ‘utmost good faith’ (uberrimae fide)), codified in Marine Insurance Act 1906, s 17 and Insurance Act 2015, s 14. See generally Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co Ltd [1995] 1 AC 501 (HL); Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd (The Star Sea) [2001] UKHL, [2001] 2 WLR 170; Versloot Dredging BV v HDI Gerling Industrie Versicherung AG [2016] UKSC 45, [2017] AC 1. For the history of the concept of good faith in the context of insurance, see H Bennett, ‘Mapping the Doctrine of Utmost Good Faith in Insurance Contract Law’ [1999] LMCLQ 165. See also RA Hasson, ‘The Doctrine of Uberrima Fidei in Insurance Law – A Critical Evaluation’ (1969) 32 MLR 615. Contrast contracts of guarantee, which are not contracts of utmost good faith: the creditor owes no duty to disclose material facts to the surety: Seaton v Heath [1899] 1 QB 782, 792 (Romer LJ). 131 Blisset v Daniel (1853) 10 Hare 493, 68 ER 1022; R l’Anson Banks, Lindley & Banks on Partnership Law, 20th edn (Sweet & Maxwell 2019) paras 10–115 and 16-13 (duties owed by partners to one another).

82  Magda Raczynska duties of honesty and good faith on the basis that the relationship between the contract parties on the facts was closely akin to that of partnership and was in fact fiduciary.132 In Al Nehayan v Kent,133 Leggatt LJ (sitting as a first instance judge) held that there is now a new category of contract into which good faith can be implied in law: a relational contract.134 It is a contract ‘in which the parties are committed to collaborating with each other, typically on a long term basis, in ways which respect the spirit and objectives of their venture but which they have not tried to specify, and which it may be impossible to specify, exhaustively in a written contract’.135 This is an evolution from the previously recognised136 concept of relational contracts, into which good faith duties were generally implied in fact. More recent cases suggest the whole point of identifying a contract as ‘relational’ is said to be so that ‘[it] will as a matter of law include an obligation of good faith’.137 The area is problematic for a number of reasons. First, establishing whether a contract is relational is difficult in practice.138 It would be easier if relational contract were a context-specific category, like insurance or employment contract. The category is unusually broad, and even certain relationships typically described as relational, such as joint ventures or franchises,139 will not necessarily 132 Nathan v Smilovitch (No 2) [2002] EWHC 1629 (Ch) [9], [12] (Ferris J). 133 Sheikh Tahnoon Bin Saeed Bin Shakhboot Al Nehayan v Ioannis Kent [2018] EWHC 333 (Comm), [2018] 1 CLC 216 [174] (Leggatt LJ thought a duty of good faith could be implied in the contract before him, (which he held to be relational) either in law or in fact). 134 The concept of a ‘relational contract’, into which a duty of good faith and fair dealing is to be implied, has attracted considerable attention in the academic literature: see, eg, I Macneil, The New Social Contract: An Inquiry into Modern Contractual Relations (Yale University Press 1980); M Eisenberg, ‘Relational Contracts’ in Beatson and Friedmann (eds), Good Faith and Fault (n 63) 291; E McKendrick, ‘The Regulation of Long-Term Contracts in English Law’ in Beatson and Friedmann (eds), Good Faith and Fault (n 63) 305; I Macneil, ‘Relational Contract Theory: Challenges and Queries’ (2000) 94 Northwestern University Law Review 877; D Campbell (ed), The Relational Theory of Contract: Selected Works of Ian Macneil (Sweet & Maxwell 2001); H Beale, ‘Relational Values in English Contract Law’ in D Campbell, L Mulcahy and S Wheeler (eds), Changing Concepts of Contracts: Essays in Honour of Ian Maneil (Palgrave 2013); D Campbell, ‘Good Faith and the Ubiquity of the “Relational” Contract’ (2014) 77 MLR 475; H Collins, ‘Is a Relational Contract a Legal Concept?’ in S Degeling, J Edelman and J Goudcamp (eds), Contract in Commercial Law (Thomson Reuters Australia 2016) 37. 135 Yam Seng (n 4) 142; Al Nehayan v Kent (n 133) [167]. See also Bates (n 108) [705], [712]–[721], [725]–[726], [728] (Fraser J) (note the nine characteristics of relational contracts enumerated at [725]). 136 Cases in which courts found a relational contract include Bristol Groundschool Ltd v Intelligent Data Capture Ltd [2014] EWHC 2145 (Ch) [196] (Richard Spearman QC sitting as a Deputy Judge of the Chancery Division); D&G Cars Ltd v Essex Police Authority [2015] EWHC 226 (QB), [2015] 2 WLUK 452 [176] (Dove J, obiter); Amey Birmingham Highways Ltd v Birmingham City Council [2018] EWCA Civ 264 [92] (Jackson LJ) (a private finance initiative contract with a local authority for the rehabilitation, maintenance, management and operation of a road network). 137 UTB LLC v Sheffield United (n 126) [200] (Fancourt J) (emphasis added); and see Mid Essex Hospital v Compass (n 17) [105] (Jackson LJ). But see TAQA Bratani Ltd v Rockrose UKSC8 LLC [2020] EWHC 58 (Comm) [55]–[56] (HHJ Pelling QC sitting as a Judge of the High Court) (discussing a duty of good faith in a relational contract as based on necessity, ie implied in fact). 138 The uncertainty is reflected in the fact that parties litigate whether ordinary long-term agreements are relational: eg Oliver Morley v The Royal Bank of Scotland plc [2020] EWHC 88 (Ch) [143], [150], [159] (Kerr J) (ordinary loan agreement not relational). See also Davies (n 23). 139 Yam Seng (n 4) [142] (Leggatt J).

Good Faiths and Contract Terms  83 be relational: a duty of good faith will not be implied just because a contract is long-term140 or a long-term joint venture,141 a long-term franchising contract142 or a long-term partnership agreement conferring an exclusive right to market services.143 Second, the application of the relational contract criteria is unstable.144 Sometimes, courts are ‘dodging’ rigorous application of the test. For example, in Acer Investment v Mansion (clearly correctly decided), Laing J held that the agreement between distributors of financial products and independent financial advisers was not relational as a matter of ‘commercial reality’.145 In D&G Cars v Essex Police,146 on the other hand, a contract between a vehicle recovery business and a police authority to recover vehicles and to carry out the authority’s instructions was ‘relational par excellence’ because it was long-term with multiple individual transactions undertaken under it.147 The existence of a ‘substantial commitment’ or collaboration in the D&G Cars case was nowhere addressed (and it is hard to agree with the result). At other times, the criteria are unnecessarily (though unsurprisingly) multiplying, as the very recent decision of Fraser J in Bates v Post Office Ltd (No 3)148 illustrates.149 Third, even if a relational contract is established, the implied duty, although typically expressed as a good faith duty, is sometimes referred to as the duty of cooperation,150 and the difference between the two is not clear. It is hard to consider how the law measures up against the models of good faith, given that the law is in a state of flux. Nevertheless, an attempt is made. It is helpful to look at the following reasoning in SPI North v Swiss Post,151 in which Andrew Hochhauser, sitting as a Deputy Judge of the High Court, said that at its highest, the general nature of a duty of good faith … is to require honesty and fidelity to the parties’ bargain … having regard to the interests of the parties (which will, inevitably, at times conflict) and to the provisions, aims and purposes of the contract, objectively ascertained. This does not, however, require one party to subordinate its interests to that of the other … [I]n consequence, the content of any duty of good faith

140 Monde Petroleum v Westernzagros [2016] EWHC 1472 (Comm), [2017] 1 All ER 1009 [250] (Richard Salter QC sitting as a Deputy High Court judge). 141 Hamsard 3147 v Boots UK (n 126) [83]–[84], [86] (Norris J). 142 Carewatch Care Services Ltd v Focus Caring Services Ltd [2014] EWHC 2313 (Ch). 143 SPI North Ltd v Swiss Post International (UK) Ltd [2019] EWHC 2004 (Ch). 144 UTB LLC v Sheffield United (n 126) [195] (Fancourt J): ‘[the law] has not yet reached a stage of settled clarity, nor is there binding authority’. 145 Acer Investment Management Ltd v Mansion Group Ltd [2014] EWHC 3011 (QB) [107]. 146 D&G Cars v Essex Police (n 136). 147 ibid [176] (Dove J). This was obiter the parties accepted the existence of an implied term to act with honesty and integrity: [173]. 148 Bates (n 108) [725]. 149 See Davies (n 23). 150 Yam Seng (n 4) at [131], [142], [145] (Leggatt J); Globe Motors v TRW Lucas Varity Electric Steering [2016] EWCA Civ 396 [67]–[68] (Beatson LJ) (referring to implying a duty of good faith or co-operation); Al Nehayan (n 133) [170]. 151 SPI North Ltd v Swiss Post (n 144).

84  Magda Raczynska in a particular contractual context is heavily dependent on that context and has to be established through a process of construction of the contract … [T]he corollary is that whether a duty with some particular content can be implied depends on the terms of the particular contract.152

This suggests that, at a broad level, a duty of good faith in the context of relational contracts could be compared to the second (purpose-based) model of good faith.153 The emphasis on ‘fidelity to the bargain’ and the duty to act in accordance with the contract purpose suggests that parties are expected to subordinate their self-interest to the achievement of the joint purpose. This is a less serious limitation on one’s freedoms than is subordination of one’s self-interest to that of another; nevertheless, taken to its logical conclusion, it would mean that the parties must exercise contractual powers and rights for a particular purpose, whether or not it is stated expressly. However, at a more detailed level, the context-dependence of the content of good faith suggests that implied duty of good faith might not work in this way. Courts are, instead, identifying specific behaviour that reasonable and honest people would regard as dishonest and/or ‘commercially unacceptable’. For example, in Bristol Groundschool v Intelligent Data, which concerned a collaboration to produce training manuals for pilots, where A was to provide content and B was to convert it into digital format, a secret download by B in order to create his own version of the product was considered ‘commercially unacceptable’ and so in breach of an implied term of good faith.154 This is very similar to identifying and implying specific duties. In fact, in Al Nehayan v Kent, Leggatt LJ specified that in the relationship of joint venture on the facts, neither party was to covertly sell his share in the companies jointly owned to a third party, and was not to use his position as a shareholder of the companies to obtain a financial benefit for himself at the expense of the other.155 The identification of specific duties suggests that good faith operates as an implication of a standard of behaviour implied into certain types of contract. It resembles the third (‘implied duty’) model of good faith, as the parties’ behaviour is measured against what the law regards as acceptable. The approach emerging in English law is that in certain types of contract the behaviour standards are implied. This clearly goes against the individualist ethic of traditional contracting. However, the individualist ethic might not suit all commercial parties, and it is hard to require parties to contract into a non-individualist (relational) ethic where 152 ibid [62] (original emphasis), citing for the last proposition Carewatch Care Services Ltd (n 143) [101]–[112], Globe Motors Inc (n 150) [67]–[68]. See also D&G Cars v Essex Police (n 136) [175] (Dove J); Mid Essex Hospital v Compass (n 17) [150] (Beatson LJ) (‘what good faith requires is sensitive to context’); and Paciocco v Australia and New Zealand Banking Group Ltd [2015] FCAFC 50 [290] (Allsop CJ), cited with approval in Al Nehayan v Kent (n 133) [175] (Leggatt LJ). 153 Section II.B. 154 Bristol Groundschool v Intelligent Data (n 136) [196] (Richard Spearman QC sitting as a Deputy Judge of the Chancery Division). 155 Al Nehayan v Kent (n 133) [176].

Good Faiths and Contract Terms  85 they, for example, subordinate their interest to a common contractual purpose. If English law is to develop along this route, it is suggested that the criteria for a ‘­relational’ contract ought to be tightened (perhaps to encompass certain types of joint venture?) and more rigorously applied. It would be unfortunate if the development of a ‘relational’ ethic, even if desirable, subverted the traditional individualist ethic of contracting.

V.  The Doctrine of Proper Purpose and Good Faith Some contracts involve the exercise of ‘discretions’, understood as decisions making an assessment or a choice from a range of options, in contrast to absolute contractual rights, asserted by way of a ‘simple decision’,156 such as the choice whether or not to terminate the contract in response to a repudiatory breach. Under English law, it is generally thought that the exercise of discretions can be controlled by the court, whereas absolute contractual ‘rights’ (powers) are unfettered.157 This is sometimes justified on the basis that the controls applied to discretions relate to their content.158 Leggatt LJ, in Abu Dhabi National Tanker Co v Product Star Shipping (The Product Star) (No 2), held that a contractual discretion should be exercised ‘honestly and in good faith, but, having regard to the provisions of the contract by which it is conferred, it must not be exercised arbitrarily, capriciously or unreasonably’.159 More recent cases, such as Braganza v BP Shipping Ltd,160 tend to focus on controls that draw analogies with public controls, so that powers must not be exercised arbitrarily, capriciously or irrationally. The duty, which the powerholder is then subject to, is implied (and is referred to as ‘the Braganza Duty’161). However, analogies drawn with public decision making have attracted criticism as not particularly useful.162 A number of authorities163 also refer to the idea that 156 Mid Essex Hospital v Compass (n 17) [83] (Jackson LJ). 157 White & Carter (Councils) Ltd v McGregor [1962] AC 413 (HL) 430 (Lord Reid). 158 D Foxton, ‘A Good Faith Goodbye? Good Faith Obligations and Contractual Termination Rights’ [2017] LMCLQ 360. 159 Abu Dhabi National Tanker Co v Product Star Shipping (The Product Star) (No 2) [1993] 1 Lloyd’s Rep 397, 404 (Balcombe and Mann LJJ agreed with Leggatt LJ). 160 Braganza v BP Shipping (n 15); Socimer International Bank v Standard Bank London (n 15). 161 See, eg, Watson v Watchfinder.co.uk Ltd [2017] EWHC 1275 (Comm), [2017] Bus LR 1309 [104] (HHJ Waksman QC sitting as a Judge of the High Court). 162 See, eg, P Sales, ‘Use of Powers for Proper Purposes in Private Law’ (2020) 136 LQR (forthcoming); M Bridge, ‘The Exercise of Contractual Discretion’ (2019) 135 LQR 227; C Himsworth, ‘Transplanting irrationality from public to private law: Braganza v BP Shipping Ltd’ (2019) 23(1) Edinburgh Law Review 1; J Morgan, ‘Against Judicial Review of Discretionary Contractual Powers’ [2008] LMCLQ 230; J Morgan, ‘Resisting Judicial Review of Discretionary Contractual Powers’ [2015] LMCLQ 483. 163 Ludgate Insurance Company Ltd v Citibank NA [1998] Lloyd’s Rep IR 221 (CA) [35] (Brooke LJ); Equitable Life Assurance Society (n 119) 459 (Lord Steyn), 461 (Lord Cooke); Gan Insurance Company (n 15) [45], [61]–[64] (Mance LJ); Paragon Finance (n 15) [39]–[41] (Dyson LJ) (the mortgagee’s power (discretion) to set the interest rates from time to time was subject to an implied term not to exercise it

86  Magda Raczynska contractual powers should not be exercised for an improper purpose. Such references are non-distinctive in the sense that the duty not to exercise the power for an improper purpose is blended with, or becomes a variation of, the ‘public-law-like’ duty not to exercise the power ‘capriciously’, ‘arbitrarily’ or ‘unreasonably’.164 There is, however, a growing interest in the idea that powers are in fact controlled by the purposes for which they were given, as a mechanism separate from the control mechanism known from public law.165 This is referred to as the ‘proper purpose doctrine’. The idea of the proper purpose doctrine has its origin in the equitable doctrine of ‘fraud on a power’166 applicable in relation to the exercise of fiduciary powers.167 It has been applied primarily in relation to power given to company directors, but it has been extended to the exercise of non-fiduciary powers.168 While there is no direct authority that contractual discretions (or indeed all contractual powers) are controllable in this way, it is arguable that the doctrine applies to all powers.169 How does the doctrine operate? Lord Sumption, in Eclairs Group Ltd v JKX Oil & Gas plc,170 said that ‘the proper purpose rule is not concerned with excess of power by doing an act which is beyond the scope of the instrument creating it as a matter of construction or implication. It is concerned with abuse of power, by doing acts which are within its scope but done for an improper reason.’ There are reasonably strong parallels between the ‘proper purpose rule’ and the purpose-based model of good faith, specifically the abuse of right (power) doctrine that developed in French law. The French courts place less emphasis on how they determine the ‘spirit’ or purpose of the right (power) but it is clear that they do so. For example, in the context of the power to change the price, discussed in section II.B, it is clear that the French courts took into account the nature and scope of the power (Code-based power to change the price) and implicitly considered its purposes (eg market considerations).

dishonestly or for an improper purpose, although on the facts the power was held to have been exercised properly, not motivated by anything other than commercial considerations). 164 BT plc v Telefonica O2 UK Ltd [2014] UKSC 42, [2014] 4 All ER 907 [37] (Lord Sumption): ‘But it is well established that in the absence of very clear language to the contrary, a contractual discretion must be exercised in good faith and not arbitrarily or capriciously … This will normally mean that it must be exercised consistently with its contractual purpose’; Watson v Watchfinder.co.uk Ltd (n 161) [105]–[115] (HHJ Waksman QC sitting as a Judge of the High Court) (discussing the idea of a ‘target’ of a power’); BHL v Leumi ABL Ltd [2017] EWHC 1871 (QB), [2017] 2 Lloyd’s Rep 237 [39]–[40] (HHJ Waksman QC sitting as a Judge of the High Court); Multiplex Construction Europe Ltd v R&F One (UK) Ltd [2019] EWHC 3464 (TCC) [27] (O’Farrell J). 165 See also Sales, ‘Use of Powers for Proper Purposes in Private Law’ (n 162); Davies, ‘Excluding Good Faith and Restricting Discretion’ (n 23). 166 Eclairs Group Ltd v JKX Oil and Gas plc [2015] UKSC 71, [2016] 3 All ER 641 [15] (Lord Sumption). See also Vatcher v Paull [1915] AC 372, 378 (Lord Parker). 167 See, eg, Duke of Portland v Topham (1864) 11 HL Cas 32, 54 (Lord Westbury LC); Howard Smith Ltd v Ampol Petroleum Ltd [1974] AC 821, 834 (Lord Wilberforce). 168 IBM United Kingdom Ltd v Dalgleish [2017] EWCA Civ 1212. 169 Davies, ‘Excluding Good Faith and Restricting Discretion’ (n 23). 170 Eclairs Group Ltd v JKX Oil (n 166) [15] (Lord Sumption).

Good Faiths and Contract Terms  87 There is also a parallel between, on the one hand, determining (under English law) whether proper acts are done for an improper reason and, on the other hand, the French law approach of comparison of the purpose, with which the right (power) holder asserts the right (power), with the ‘spirit’ and purpose of the right. Under French law, where the right (power) holder acts with a purpose not consistent with the ‘spirit’ and purpose of the right, the right (power) holder abuses her right (power). Assuming that it is correct to interpret Lord Sumption’s determining of the purpose in Eclairs as not purely subjective, the parallel between English law and the French law doctrine of abuse of right is quite close. If this is correct, English law has its own form of the good faith doctrine, applicable in connection with contractual performance: the proper purposes doctrine. Given that it is concerned with the purpose for which it is granted, when applied to contracts, and where the purpose is a matter of contract, it is arguable that there would be little interference with the individualist ethic of contracting, at least less so than is the case with the implied duties based on controls established in public law.

VI. Conclusion When asked about the utility of good faith in carrying out contracts, many English commercial contract lawyers would likely say: ‘No Good Faith Please, We’re English!’171 This aversion, in the words of Lord Steyn, is not bred from any great familiarity with the way the principle works but it is profound.172 One very basic point made in this chapter is that whether English law is to swim with or against the tide of good faith,173 it is a good idea to know what this tide is. To help English lawyers make an informed choice, this chapter’s first main contribution is to the understanding of good faith generally, although it will be of relevance not only to those steeped in English law. It shows that there are various models (conceptions) of good faith (hence, the plural ‘Faiths’ in the title). The idea of classification of good faith is not new; the novelty here is the focus on how the concepts of good faith relate to the individualist ethic of freedom to pursue one’s self-interest. This categorisation is particularly useful when considering the development of English law because of the importance this jurisdiction attaches to the pursuit of one’s self-interest. This helps us see that the binary opposition of individualism (associated with contracts where parties pursue their own self-interest, without the law’s imposition of good faith) and cooperativism (associated with parties’ giving up the pursuit of their own self-interest, typically subject to some good faith) is simplistic. Second, this chapter contributes insights into the meaning of the individualist ethic underpinning the existing doctrines of English law that apply when courts

171 This

was in fact the original title of the paper when presented at the UCL Conference. ‘Contract Law: Fulfilling the Reasonable Expectations of Honest Men’ (n 86) 438. 173 See text to n 22. 172 Steyn,

88  Magda Raczynska are establishing the content of commercial contracts, that is, interpretation and implication of terms. The traditional view is that in commercial contracts, parties should not have to sacrifice their pursuit of self-interest, and therefore there is no room in English commercial contract law for a principle of good faith. This chapter shows that this traditional view is correct insofar as a general principle of good faith is concerned. Such a general principle is associated with the adjudication model of good faith, where courts have (methodologically) a considerable freedom to limit parties’ self-interest. English courts similarly, in principle, enjoy a substantial power to intervene, but they do so on the basis of established doctrines rather than broad principles. For example, the fact that English courts have the power to imply terms, particularly terms in law, resembles, functionally, the adjudication model of good faith, although it is worth remembering that in practice English courts rarely imply terms in law, and when they do, these generally tend to be (relational contracts aside) relatively specific terms in specific contexts.174 A general principle of good faith has no place in English law and English law should not try to adopt it. However, the application of specific doctrines in English law resembles some models of good faith, in a very limited way.175 The interpretation of contract in the light of reasonable expectations of honesty, and the implication of controls when exercising discretions, both limit parties’ pursuit of self-interest. Their comparison with the models of good faith suggests that English law already has two (subtle) conceptions of good faith. The interpretation of contracts with an expectation of honesty is a thin (and rebuttable) doctrine that resembles the ‘interpretation model’ of good faith. The proper purpose doctrine, assuming it is correct to see it as applicable to contracts, resembles functionally and in the way it operates the ‘purpose-based’ model of good faith. Both strengthen the point made by Lady Arden, that ‘the law is already slowly developing in a way which can accommodate the concept of good faith within contract law’.176 If English law is to move further towards a broader endorsement of good faith, this is best done through the development of specific doctrines such as the proper purpose doctrine, rather than by seeking to adopt a general principle of good faith (for which there is no room), or by implying good faith duties as a matter of necessity in contract (which, as argued in section IV.A, is very likely a contradiction in terms).

174 See section IV.B. 175 For example, outside contractual performance, it is worth remembering that parties negotiating a contract are not subject to any duty of good faith at the pre-contractual stage. 176 M Arden, ‘Coming to Terms with Good Faith’ (2013) 30 Journal of Contract Law 199, 200.

6 Excluding Good Faith and Restricting Discretion PAUL S DAVIES

Fairness has nothing to do with commercial contracts. The parties enter into them in a spirit of competitive co-operation, with a view to serving their own interest. Commercial parties can be most unfair and entirely unreasonable, if they can get away with it.

This robust view was recently expressed, extra-judicially, by Lord Sumption.1 Although controversial,2 it represents a traditional approach to commercial contracts that will resonate with many. Yet there appears to have been something of a shift in the commercial contracts that trouble the courts.3 Lord Goff (as he later became) once remarked that spot contracts on standard terms were the ‘staple diet’ of the Commercial Court.4 To a large extent that remains true, but there has been a marked increase in commercial litigation concerning what may loosely be termed ‘joint ventures’.5 Such contracts typically apply over a long period of time and require a degree of cooperation between the parties. In the context of such long-term agreements, can a party still act in an entirely self-serving manner? This question has given rise to divergent answers and some confusion. As highlighted in chapter 5, the ability of a party to act exclusively for its own best interests may be restricted by the terms of the contract. There has been much discussion both of implied duties of good faith in the performance of a contract and, relatedly, of implied terms controlling the exercise of contractual powers. But the basis of such terms remains unclear: should the focus be upon

1 Lord Sumption, ‘A Question of Taste: The Supreme Court and the Interpretation of Contracts’ [2017] OUCLJ 301, 310. 2 cp Sir George Leggatt, ‘Contractual Duties of Good Faith’, Lecture to the Commercial Bar Association (18 October 2016) [26]: ‘I believe it is a mistake to see contracting as an essentially ­adversarial activity’. 3 G Leggatt, ‘Negotiation in good faith: adapting to changing circumstances in contracts and English contract law’ [2019] Journal of Business Law 104, 105–06. 4 R Goff, ‘Commercial Contracts and the Commercial Court’ [1984] LMCLQ 382, 386–87. 5 This umbrella term covers a wide range of contracts: see generally I Hewitt, S Howley, J Parkes (eds), Hewitt on Joint Ventures, 6th edn (Sweet & Maxwell 2016).

90  Paul S Davies terms implied in fact, or terms implied at law? And are some restrictions based upon general doctrines of law rather than implied terms? This chapter will first examine express restrictions on self-interested behaviour, before analysing the potential basis of implied terms that restrict a party’s ability to act in a self-serving way.6 It is suggested that terms should generally be implied in fact rather than at law, and that the parties remain able to exclude ‘good faith’. Freedom of contract demands that parties can control the extent of the obligations to which they consent. However, it may be that some controls on the exercise of a power, for example, cannot be excluded. If so, the better justification is not that there are some mandatory ‘universal terms’7 in all contracts, but rather that there are some doctrines that cannot be excluded for reasons of public policy, just as liability for fraud or illegal conduct cannot be excluded. It is suggested that parties should not be able to exercise a power for an ‘improper purpose’, and that the doctrine of ‘fraud on a power’ should be recognised more clearly in the contractual context. Yet the range of ‘proper purposes’ may be very wide as a result of the parties’ agreement, with the effect that parties can still effectively act in a self-interested manner.

I.  Express Restrictions on Self-interested Behaviour Parties can expressly provide for restrictions on their ability to act in a selfinterested fashion. Sir George Leggatt has recently observed that it has become much more common to find in contracts that come before our courts clauses expressed in the language of ‘best endeavours’ or ‘reasonable endeavours’ or of obligations to act ‘in good faith’. Twenty or thirty years ago it would, in my experience, have been unusual to find such language in a commercial contract. Now it is an everyday occurrence.8

Fundamental concerns of party autonomy and freedom of contract mean that such terms should be enforced by a court. One potential barrier to enforcement may be that the term is insufficiently certain for a court to conclude that a binding agreement has been formed. The foundations of this obstacle lie in the well-known hostility of Lord Ackner to a duty of ‘good faith’ in Walford v Miles.9 However, Walford v Miles is a

6 Good faith is, of course, not necessarily inconsistent with acting in a self-interested manner, but its invocation in the commercial context generally constrains one party’s ability to do so: see section II. Ideas of good faith and proper purposes may also constrain a party’s ability to act out of reasons of spite, which are not necessarily self-interested: see section III. 7 M Furmston. ‘Universal Terms in Contract’ in L Gullifer and S Vogenauer (eds), English and European Perspectives on Contract and Commercial Law: Essays in Honour of Hugh Beale (Hart Publishing 2014); see too J Paterson, ‘Good Faith Duties in Contract Performance’ [2014] OUCLJ 283. 8 Leggatt, ‘Negotiation in good faith’ (n 3) 106. 9 Walford v Miles [1992] 2 AC 128.

Excluding Good Faith and Restricting Discretion  91 controversial decision,10 which concerned a potential duty to negotiate in good faith where a contract had not yet been concluded. Where a contract has already been agreed, courts have generally shown a willingness to enforce those terms. For instance, in Petromec Inc v Petroleo Brasileiro SA, Longmore LJ enforced a clause to negotiate certain extra costs in good faith, and remarked: It is not irrelevant that it is an express obligation which is part of a complex agreement drafted by City of London solicitors … It would be a strong thing to declare unenforceable a clause into which the parties have deliberately and expressly entered … To decide that it has ‘no legal content’ to use Lord Ackner’s phrase would be for the law deliberately to defeat the reasonable expectations of honest men …11

Similarly, in Emirates Trading Agency LLC v Prime Mineral Exports Private Ltd,12 Teare J held that a clause that disputes should be settled ‘by friendly discussion’ was enforceable (and had been complied with). Moreover, in certain circumstances, the ISDA Master Agreement13 requires one party to calculate sums due from the counterparty, with that calculation to be undertaken in good faith. The courts have been very clear that they will seek to give effect to this obligation.14 This is welcome. Courts should not readily find that contractual terms are too uncertain to be binding.15 As Leggatt J observed in Astor Management AG v Atalaya Mining plc, ‘[t]he role of the court in a commercial dispute is to give legal effect to what the parties have agreed, not to throw its hands in the air and refuse to do so because the parties have not made its task easy’.16 However, it should be noted that an express obligation to act in good faith is unlikely to be construed broadly. In Mid Essex Hospital Services NHS Trust v  Compass Group UK and Ireland Ltd (t/a Medirest),17 a clause of the contract provided that ‘The Trust and the Contractor will co-operate with each other in good faith.’ A question arose as to whether this applied to the general performance of the contract, or whether it applied more narrowly to two particular purposes, namely the ‘transmission of information and instructions’ and to enable the 10 See recently Leggatt, ‘Negotiation in good faith’ (n 3) 104. 11 Petromec Inc v Petroleo Brasileiro SA [2005] EWCA Civ 891, [2006] 1 Lloyd’s Rep 161 [121]. 12 Emirates Trading Agency LLC v Prime Mineral Exports Private Ltd [2014] EWHC 2104 (Comm), [2015] 1 WLR 1145. 13 Published by the International Swaps and Derivatives Association. On ‘reasonable endeavours’, see too Little v Courage (1995) 70 P & CR 469 (CA) 476. However, the object of the endeavours must be sufficiently certain and there must be sufficient objective criteria by which to evaluate the reasonableness of the endeavours: Dany Lions Ltd v Bristol Cars Ltd [2014] EWHC 817, [2014] 2 All ER (Comm) 403. See also Berkeley Community Villages Ltd v Pullen [2007] EWHC 1330 (Ch), (2007) 3 EGLR 101 (Morgan J); CPC Group Ltd v Qatari Diar Real Estate Investment Co [2010] EWHC 1535 (Ch), [2010] CILL 2908 (Vos J). 14 See, eg, Lehman Brothers International (Europe) v Lehman Brothers Finance SA [2013] EWCA Civ 188, [2014] 2 BCLC 451. 15 cf Wells v Devani [2019] UKSC 4, [2019] 2 WLR 617. 16 Astor Management AG v Atalaya Mining plc [2017] EWHC 425 (Comm), [2018] 1 All ER (Comm) 547 [64]. 17 Mid Essex Hospital Services NHS Trust v Compass Group UK and Ireland Ltd (t/a Medirest) [2013] EWCA Civ 200.

92  Paul S Davies contracting party ‘to derive the full benefit of the contract’. The latter interpretation was favoured.18 Beatson LJ insisted that a broad obligation to cooperate in good faith should not cut across other, more specific provisions,19 and that a narrower interpretation should be preferred.20

II.  Implied Restrictions on Self-interested Behaviour If courts will give effect to express terms imposing a duty of good faith, there is no reason why they should not give effect to implied terms with similar content.21 But the fact that well-advised commercial parties could have made express provision for good faith performance may mean that a court will be wary about implying such terms.22 It is not for the court to improve the contract made by the parties, or to rescue one party from its own bad bargain.23 Parties would therefore be well-advised to include express terms of good faith where desired.24 Problems can arise where the contract is silent. As Cromwell J observed in Bhasin v Hrynew: The jurisprudence is not always very clear about the source of the good faith obligations found in these cases. The categories of terms implied as a matter of law, terms implied as a matter of intention and terms arising as a matter of interpretation sometimes are blurred or even ignored, resulting in uncertainty and a lack of coherence at the level of principle.25

The suggestion that a contract can be ‘interpreted’ such that it must be performed in good faith, even where there is no express provision for good faith in the written contract, relies upon a broad approach towards interpretation. Although favoured by Lord Hoffmann,26 it seems inconsistent with the most recent, leading guidance from the Supreme Court in Arnold v Britton, which emphasised the importance of interpreting the words deliberately chosen by the parties, rather than departing from their plain meaning by reference to the factual matrix.27 This stricter approach should be favoured. 18 ibid [97]–[121], [148] and [149]. 19 ibid [154]. 20 See too Portsmouth City Council v Ensign Highways Ltd [2015] EWHC 1969 (TCC). This approach seems consistent with a ‘stricter’ approach to interpretation that now seems to be preferred to the more ‘liberal’ approach favoured by Lord Hoffmann: see section II. 21 Yam Seng Pte Ltd v ITC Ltd [2013] EWHC 111 (QB), [2013] 1 All ER (Comm) 1321. 22 cf Churchill Falls (Labrador) Corporation Limited v Hydro-Québec [2018] SCC 46. 23 See generally PS Davies, ‘Bad Bargains’ [2019] CLP 253. 24 eg E McKendrick, ‘The Regulation of Long-term Contracts in English Law’ in J Beatson and D Friedmann (eds), Good Faith and Fault in Contract Law (OUP 1995) 305. See too, eg, Lymington Marina Ltd v Macnamara [2007] EWCA Civ 151, [2007] 2 All ER (Comm) 825 (Arden LJ). 25 Bhasin v Hrynew [2014] SCC 71, [2014] 3 SCR 494 [52]. 26 eg Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896; Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] 1 AC 1101. See too Lord Hoffmann, ‘Language and Lawyers’ (2018) 134 LQR 553. 27 Arnold v Britton [2015] UKSC 36, [2015] AC 1619.

Excluding Good Faith and Restricting Discretion  93 Interpretation should be concerned with the express terms of the contract.28 Interpreting silence, on the other hand, is inherently ambiguous and prone to error. As a result, where the contract does not explicitly provide for a contract to be performed in good faith, or for a power to be exercised in a ‘reasonable’ manner, courts should not strain the boundaries of interpretation such that it engulfs implication. Where the contract is silent as to ‘good faith’ (or any other term), it can only be found to be part of the contract through implication. Strictly, that implied term would then need to be interpreted. As the Supreme Court insisted in Marks and Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd, the processes of interpretation and implication are best considered to be distinct.29 All terms have to be interpreted, but it is not inevitable that any terms have to be implied into an agreement. In a detailed commercial contract the usual inference of silence is that no term should be implied and the loss simply lies where it falls.30 A term should only be implied where it is necessary to do so.31 If duties of good faith respond to the parties’ intentions through the traditional, restrictive tests for implication in fact, they are not particularly intrusive.32 However, between commercial parties, at least, a term of good faith should not readily be implied. It will often be the case that a contract is efficacious even without such an implied term.33 Moreover, if an officious bystander were to ask both parties ‘Is there also a term that the contract be performed in good faith?’, the parties might not respond ‘Oh, of course!’ but rather ‘What do you mean by good faith?’ As a result, the ‘officious bystander’ test for implication would not be fulfilled.34 It is perhaps for this reason that in cases where a term has been implied in fact, judges have been more particular about the content of the implied term, rather than simply implying a general term of good faith. For example, in Yam Seng Pte Ltd v International Trade Corp Ltd,35 the terms implied into the bargain were

28 H Beale (ed), Chitty on Contracts, 33rd edn (Sweet & Maxwell 2018) para 13-041, accurately states that interpretation ‘denotes the process … by which a court arrives at the meaning to be given to the language used by the parties in the express terms of a written agreement’. 29 Marks and Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2016] UKSC 72, [2016] AC 742. The distinction appears to have been blurred a little in Wells v Devani (n 15), although, importantly, that case did not concern a written contract (and nor did Al Nehayan v  Kent [2018] EWHC 333 (Comm), [2018] 1 CLC 216): PS Davies, ‘Interpretation and Implication in the Supreme Court’ (2019) 78 CLJ 267. 30 See, eg, Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472 (Bingham MR). 31 Whether in fact (following Marks and Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd) (n 29)) or at law (Liverpool v Irwin [1977] AC 239). 32 See PS Davies, ‘The Basis of Contractual Duties of Good Faith’ [2019] Journal of Commonwealth Law 1. They are sometimes said to be ‘intrinsic’ to the bargain rather than imposed by law: Mid Essex Hospital Services (n 17) [82] (Jackson LJ). 33 Monde Petroleum SA v WesternZagros Ltd [2016] EWHC 1472 (Comm), [2017] 1 All ER (Comm) 1009 (this issue was not considered on appeal: [2018] EWCA Civ 25). 34 Shirlaw v Southern Foundries Ltd [1939] 2 KB 206, 227 (MacKinnon LJ). 35 Yam Seng (n 21).

94  Paul S Davies narrower and better-defined than a broad resort to ‘good faith’.36 Indeed, the focus on an individual contract means that the implied term can be tailored to the particular circumstances of the case. Thus in Al Nehayan v Kent, Leggatt LJ again implied more precise and narrow terms than ‘good faith’.37 In Equitas Insurance Ltd v Municipal Mutual Insurance Ltd, Males LJ described good faith as ‘merely a label’;38 the content of implied duties can be defined more precisely within the broader umbrella of ‘good faith’. In any event, a term should only be implied in fact if necessary to do so.39 The burden lies on the party seeking to imply a term. In British Telecommunications Plc v Telefónica O2 UK Ltd, Lord Sumption said that ‘it is well established that in the absence of very clear language to the contrary, a contractual discretion must be exercised in good faith and not arbitrarily or capriciously’.40 It is not clear whether Lord Sumption was considering an implied term to act in good faith, or whether he was considering a general doctrine such as ‘fraud on a power’, considered in section IV. If the latter, it is suggested that it should not be excludable.41 If the former, the initial presumption regarding implied terms should be that a term is not implied unless necessary; the burden is not placed on the other party to show that an implication is not necessary. If a term is to be implied without reference to the parties’ intentions then it is not a term implied in fact but rather a term implied at law. Courts are, generally, rightly wary of exercising their powers to imply terms at law.42 Imposing terms upon parties in such a manner is highly intrusive. That is why the requirement that such terms be necessary rather than just reasonable should be taken seriously. However, there can be a tendency to define the type of contract at issue narrowly, and thereby blur the boundary between terms implied in fact and terms implied at law:43 if the type of contract essentially covers barely more than the particular contract at issue, the distinction between a term implied at law and in fact is very thin indeed.

36 Namely, a duty to not ‘knowingly provide false information’ on which the counterparty was likely to rely, and a duty not to authorise the sale of relevant products in certain territories below the ‘­particular price’: see Davies, ‘The Basis of Good Faith’ (n 32) 16–18. 37 Namely, that a party to a joint venture would not enter into negotiations to sell his interest or part of his interest in the companies they jointly owned to a third party covertly and without informing the other beneficial owner, and a duty not to use his position as a shareholder of the companies to obtain a financial benefit for himself at the expense of the other: Al Nehayan (n 29) [176]; see Davies, ‘The Basis of Good Faith’ (n 32) 14–19. See too, eg, Bristol Groundschool Ltd v Whittingham [2014] EWHC 2145 (Ch); D&G Cars Ltd v Essex Police Authority [2015] EWHC 226 (QB). 38 Equitas Insurance Ltd v Municipal Mutual Insurance Ltd [2019] EWCA Civ 718 [116]. 39 eg Eastleigh BC v Town Quay Developments Ltd [2009] EWCA Civ 1391, [2010] 2 P&CR 2 [30]–[39] (Arden LJ); Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2) [2001] EWCA Civ 1047, [2001] 2 All ER (Comm) 299 [43]–[78] (Mance LJ), [97] (Staughton LJ). 40 British Telecommunications Plc v Telefónica O2 UK Ltd [2014] UKSC 42, [2014] 4 All ER 907 [37]. 41 See section IV. 42 After all, unelected judges do not have the democratic legitimacy of Parliament when deciding that all contracts of a certain type should include a particular implied term. 43 cf Scally v Southern Health and Social Services Board [1992] 1 AC 294.

Excluding Good Faith and Restricting Discretion  95 Examples of contracts where obligations of good faith have been implied at law often involve a power imbalance between the contracting parties, such as that arising in a contract of employment.44 Where there is no imbalance of power between commercial parties, it is less obvious that there is any need to imply a term of good faith. As regards spot contracts for the sale of commodities, it would be very surprising to see an implied term of good faith, and such a term would serve little purpose: any breach of a term of good faith would invariably constitute a breach of a different term of the contract anyway.45 The debate is now focused upon whether a term should be implied at law into ‘relational contracts’.46 In Monde Petroleum, Richard Salter QC, sitting as a Deputy High Court judge, was ‘clear that the mere fact that a contract is a long-term or relational one is not, of itself, enough to justify such an implication’.47 Such an approach should be supported. After all, some long-term contracts may be zero-sum rather than mutually beneficial,48 and it might be doubted whether the concept of ‘relational contracts’ is sufficiently definite to constitute a nominate category of contracts;49 some contracts will exhibit more ‘relationality’ than others. Nevertheless, there appears to have been a drift towards accepting a category of ‘relational contracts’ into which terms of good faith will be implied as a matter of law. This is very recent, and its utility doubtful. In Yam Seng, Leggatt J doubted whether English law had reached the stage where it would be ‘ready to recognise a requirement of good faith as a duty implied by law, even as a default rule, into all commercial contracts’.50 But in Al Nehayan v Kent, Leggatt LJ thought that ‘­relational contracts’ was a category of contract, observing that in Yam Seng: I drew attention to a category of contract in which the parties are committed to collaborating with each other, typically on a long term basis, in ways which respect the spirit

44 eg Malik v Bank of Credit and Commerce International SA (in liquidation) [1998] AC 20. 45 cf Bhasin (n 25) [60]. 46 However, it should be noted that if the duty of good faith consists only of the duty to act honestly and with fidelity to the terms of the contract then it may be difficult ultimately to justify its restriction to ‘relational’ contracts: see below. It has been said that terms can be implied into ‘commercial contracts’, but the boundaries of that notion seem even more problematic than ‘relational contracts’: Cavendish Square Holding BV v Makdessi [2015] UKSC 67, [2016] AC 1172 [168] (Lord Mance); cf JML Direct Ltd v Freesat UK Ltd [2010] EWCA Civ 34 [14] (Blair J). The restriction to types of contract seems to have been ignored in Australia, where there is evidence of the courts’ implying terms in law outside such common relationships: see, eg, Vodafone Pacific Ltd v Mobile Innovations Ltd [2004] NSWCA 15 [125], [189]; Burger King Corp v Hungry Jack’s Pty Ltd (2001) NSWCA 187 [159], [164]; Alcatel Australia Ltd v  Scarcella (1998) 44 NSWLR 349, 369; cf Renard Constructions (ME) Pty Ltd v Minister of Public Works (1992) 26 NSWLR 234, 263, where Priestley JA referred to a ‘hybrid’ between implied terms in fact and in law; and see generally E Peden, Good Faith in the Performance of Contracts (Butterworths 2003), ch 6. 47 Monde Petroleum SA v WesternZagros Ltd (n 33) [250] (not considered on appeal: [2018] EWCA Civ 25). 48 eg Property Alliance Group Ltd v The Royal Bank of Scotland Plc [2018] EWCA Civ 355, [2018] 1 WLR 3529. 49 For further discussion, see, eg, H Collins, ‘Is a Relational Contract a Legal Concept?’ in S Degeling, J Edelman and J Goudkamp (eds), Contract in Commercial Law (Lawbook Co 2016). 50 Yam Seng (n 21) [131]. cf Bhasin (n 25) [33].

96  Paul S Davies and objectives of their venture but which they have not tried to specify, and which it may be impossible to specify, exhaustively in a written contract. Such ‘relational’ contracts involve trust and confidence but of a different kind from that involved in fiduciary relationships. The trust is not in the loyal subordination by one party of its own interests to those of another. It is trust that the other party will act with integrity and in a spirit of cooperation. The legitimate expectations which the law should protect in relationships of this kind are embodied in the normative standard of good faith.51

Leggatt LJ went on to hold that ‘the nature of the contract as a relational contract implicitly requires (in the absence of a contrary indication) treating it as involving an obligation of good faith’.52 However, it was not necessary to rely upon a term implied at law in Al Nehayan v Kent: Leggatt LJ also held that a term could be implied in fact.53 Al Nehayan v Kent was an initial sign that the term implied in fact is ‘­hardening’54 into a term implied at law into relational contracts. This has been fortified by Bates v Post Office Ltd (No 3: Common Issues).55 The case involves claims brought by a large number of sub-postmasters against the Post Office regarding the operation of a computer system used for accounting. The litigation between the parties is extensive and ‘bitterly contested’.56 However, in this decision Fraser J did not decide whether the contracts between the Post Office and sub-postmasters had in fact been breached, but rather determined preliminary ‘Common Issues’ between the parties, mainly regarding the nature and content of the contracts. As Fraser J said, ‘The first Common Issue … is whether the contracts … are what is called “relational contracts” …. In my judgment, it is certainly one of the most important issues.’57 Fraser J went on to cite a number of cases to support his conclusion that ‘the concept of relational contracts is an established one in English law’.58 However, none of the cases cited in support of this conclusion makes it clear what the boundaries of a relational contract are. For example, in Amey Birmingham Highways Ltd v Birmingham City Council, Jackson LJ ‘refused to venture into those contentious issues’ about whether ‘relational contracts’ are subject to special rules.59 Although he recognised that the contract at issue in the case was ‘relational’, no consequences appear to flow from that conclusion. Fraser J also cited Globe Motors v TRW Lucas Varity Electric Steering, but in the passage cited Beatson LJ made no reference to

51 Al Nehayan (n 29) [167]. 52 ibid [174]. 53 ibid. 54 Leggatt, ‘Contractual Duties of Good Faith’ (n 2) [50]. 55 Bates v Post Office Ltd (No 3: Common Issues) [2019] EWHC 606 (QB). 56 ibid [10]; see too [2017] EWHC 2844 (QB), [2018] EWHC 2698 (QB), [2019] EWHC 871 (QB), [2019] EWHC 1373 (QB). 57 Bates (n 55) [31]. 58 ibid [705]. 59 Amey Birmingham Highways Ltd v Birmingham City Council [2018] EWCA Civ 264, [2018] BLR 225 [92].

Excluding Good Faith and Restricting Discretion  97 the language of ‘relational contract’ at all,60 which would be consistent with there being no ‘nominate category’61 of ‘relational contract’. Fraser J also relied upon the decision of the Court of Appeal in MSC Mediterranean Shipping Co v Cottonex Anstalt,62 despite the language of ‘relational’ appearing nowhere in the judgment and the court’s forceful rejection of a general duty of good faith in English law.63 Nevertheless, Fraser J said: It should be noted that Moore-Bick LJ did not say that there was no such concept [of good faith], but rather it was neither necessary nor desirable to resort to it in that case. I  consider this as consistent with, and further support for, the concept of relational contracts in English law, that is to say those that have an implied duty of good faith.64

This is surprising, and appears to go against the thrust of Moore-Bick LJ’s robust, traditional approach in Cottonex Anstalt. Admittedly, in D&G Cars Ltd v Essex Police Authority,65 Dove J thought that the contract under consideration was ‘a “relational” contract par excellence’. And the notion of a ‘relational’ contract has also been invoked by Leggatt LJ in both Yam Seng and Al Nehayan v Kent. But in none of those cases was the labelling of the contract as ‘relational’ crucial to the implication of a term or the outcome of the case. Describing the contract as ‘relational’ helped to bolster the conclusions already reached, but the same outcomes could have been reached without the language of ‘relational contracts’. However, the decision of Fraser J in the Post Office litigation goes further. The judge held that terms of good faith should be implied at law because the contracts were relational contracts,66 and said: I consider the following characteristics are relevant as to whether a contract is a relational one or not: 1. 2. 3. 4. 5.

There must be no specific express terms in the contract that prevents [sic] a duty of good faith being implied into the contract. The contract will be a long-term one, with the mutual intention of the parties being that there will be a long-term relationship. The parties must intend that their respective roles be performed with integrity, and with fidelity to their bargain. The parties will be committed to collaborating with one another in the performance of the contract. The spirits and objectives of their venture may not be capable of being expressed exhaustively in a written contract.

60 Globe Motors v TRW Lucas Varity Electric Steering [2016] EWCA Civ 396, [2017] 1 All ER (Comm) 601. 61 Collins, ‘Is a Relational Contract a Legal Concept?’ (n 49) 38. 62 MSC Mediterranean Shipping Co v Cottonex Anstalt [2016] EWCA Civ 789, [2017] 1 All ER (Comm) 483. 63 eg ibid [45]. 64 Bates (n 55) [705] (original emphasis). 65 D&G Cars Ltd v Essex Police Authority [2015] EWHC 226 (QB) [176]. 66 See eg Bates (n 55) [700].

98  Paul S Davies 6. 7. 8. 9.

They will each repose trust and confidence in one another, but of a different kind to that involved in fiduciary relationships. The contract in question will involve a high degree of communication, co-operation and predictable performance based on mutual trust and confidence, and expectations of loyalty. There may be a degree of significant investment by one party (or both) in the venture. This significant investment may be, in some cases, more accurately described as substantial financial commitment. Exclusivity of the relationship may also be present

I hesitate to describe this as an exhaustive list. No single one of the above list is determinative, with the exception of the first one. This is because if the express terms prevent the implication of a duty of good faith, then that will be the end of the matter. However, many of these characteristics will be found to be present where a contract is a relational one. In other cases on entirely different facts, it may be that there are other features which I have not identified above which are relevant to those cases.67

This is not a particularly helpful approach when deciding whether a nominate category of ‘relational contracts’ should be recognised. All these characteristics may be relevant, but do not clearly define the contours of a ‘relational contract’.68 After all, the nine ‘relevant characteristics’ of Fraser J are not necessary for there to be a relational contract: the ‘spirit and objectives … may not be capable’ of being expressed in written terms (5), there ‘may be’ significant investment or financial commitment (8) and exclusivity ‘may’ be present (9). These all appear optional. The other characteristics also lack clarity. How ‘long-term’ should a contract be (2)? How can it be determined whether the parties are ‘committed to collaborating’ (4)? What sort of ‘trust and confidence’ is necessary, and how does it differ from a fiduciary relationship (6)? What constitutes a ‘high degree of communication and predictable performance’ (7)? The boundaries are very unclear. It is perhaps telling that Fraser J thought it relevant that the ‘parties must intend that their respective roles be performed with integrity, and with fidelity to their bargain’ (3). If this is what the parties intend then the relevant term is likely to satisfy the officious bystander test and should be implied in fact. There is no need to resort to the more intrusive notion of implication at law. In any event, it would be misleading to say that (1) is determinative of there being a relational contract. If a duty of good faith is excluded, that would suggest, on the approach of Fraser J, that there is not a relational contract. It does not provide a positive reason to find that a contract is relational. In Interfoto, Bingham LJ noted that the law has ‘developed solutions in response to demonstrated problems of unfairness’.69 It is far from clear that there is 67 ibid [725]. 68 In UTB LLC v Sheffield United Ltd [2019] EWHC 2322 (Ch) [202], Fancourt J recently observed that ‘there is a danger in using the term “relational contract” that one is not clear about what exactly is meant by it’. 69 Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd [1989] QB 433, 439.

Excluding Good Faith and Restricting Discretion  99 a demonstrated problem of unfairness in this context. In any event, terms should only be implied at law into relational contracts once it is clear what a relational contract is. That type of contract must be readily discernible; otherwise there is a risk of inconsistency, incoherency and the courts’ intruding into the parties’ bargain where it is not necessary to do so. Attempts to create a discrete category of relational contracts seem invariably problematic.70 Moreover, the risk of selfseeking or opportunistic behaviour may simply be priced in to the contract from the outset, and implying terms at law may disturb a careful equilibrium crafted by the parties in the terms of the contract. Since it is now clear that express terms of good faith will be given effect by the court,71 the need to imply such terms at law into contracts between well-advised sophisticated commercial parties is reduced.72 Commercial parties should be aware of their ability to regulate their own agreements and relationships, and should be encouraged to do so. The absence of express duties of good faith may well be deliberate.73 Courts should be wary of implying a term of good faith that was not expressly included – especially through implication at law rather than implication in fact.

III.  Excluding Terms Restricting Self-interested Behaviour Bates v Post Office confirms that implied duties of good faith can be excluded, even if the term is implied at law rather than in fact.74 This possibility has recently been doubted,75 but Bates is consistent with a long line of cases in insisting that there is no scope for a court to imply a term that is contrary to the express terms of the contract.76 70 A good, but nevertheless uncertain, attempt to provide a test for a ‘relational contract’ is made by Collins, ‘Is a Relational Contract a Legal Concept?’ (n 49) 55: ‘First, there is a long-term business relationship that will provide sufficient pay-offs to both parties to continue with the relationship even through periods of considerable adversity. Second, obtaining the benefits of the business relationship will require adaptation, co-operation, and evolution of performance obligations, so that indeterminate implicit obligations of this kind must be central to the deal. Third, these implicit indeterminate obligations must be understood as arising not from general moral standards or norms of reciprocity such as honesty, but will be tailored to achieve what is necessary to secure the success of the venture. Business necessity in this context requires the acceptance of obligations derived from the general concepts of co-operation and loyalty or commitment to the project.’ However, even Professor Macneil ‘did not support the claim that there is a class of contracts that can properly be described as relational’: ibid 46. 71 See section I. 72 E McKendrick, ‘The Regulation of Long-term Contracts in English Law’ (n 24) 329–33. 73 Davies, ‘Bad Bargains’ (n 23) 262, 269–271. 74 See similarly, eg, Yam Seng (n 21) [149]. See too, eg, Vodaphone Pacific Ltd. V Mobile Innovations Ltd (NSWCA). cf R Hooley, ‘Controlling contractual discretion’ (2013) 72 CLJ 65, 81–82. 75 E Lim and C Chan, ‘Problems with Wednesbury Unreasonableness in Contract Law: Lessons from Public Law’ (2019) 135 LQR 88, 103. 76 See, eg, Assenagon Asset Management SA v Irish Bank Resolution Corp Ltd [2012] EWHC 2090 (Ch), [2013] 1 All ER 495, [46] (Briggs J); Dymoke v Association for Dance Movement Psychotherapy UK Ltd [2019] EWHC 94 (QB) [60] (Popplewell J).

100  Paul S Davies Bates also provides support for the view that good faith includes duties beyond an obligation to act honestly.77 This explicitly departs from the view taken in Chitty,78 but is consistent with earlier decisions such as Yam Seng and Al Nehayan v Kent. The uncertainty surrounding the content of an implied obligation of good faith is one reason why commercial parties are likely to be wary of it and may seek to exclude it. One irony of Bates may be that whilst it appears, on its face, to expand the reach of good faith, in practice it could have the opposite effect in commercial contracts: well-advised parties might act on the court’s recognition that express terms can prevent the implication of good faith, and insist upon clauses excluding good faith.79 Such clauses may even become incorporated into a party’s standard terms. The likelihood of this is perhaps increased if it is accepted that a term of good faith should be implied at law into ‘relational contracts’ and the parties are not sure whether their contract would be classified as ‘relational’. Parties may be inclined to limit the terms that could apply regardless of their intentions, and instead take control of their agreement by ensuring that only the express terms of the contract be enforced.80 However, it may be that courts will be unwilling to countenance that duties of good faith can readily be excluded.81 For example, a broad ‘entire agreement’ clause that purports to exclude all implied terms may be interpreted narrowly such that implied duties of good faith are not excluded. In GEC Marconi Systems Pty v. BHP Information Technology Pty, for instance, Finn J said ‘I find arresting the suggestion that an entire agreement clause is of itself sufficient to constitute an “express exclusion” of an implied duty of good faith and fair dealing where that implication would otherwise have been made by law.’82 But it is not clear why duties of ‘good faith’ should be treated differently from any other implied term. An entire agreement clause – like any other term of the contract – should be interpreted according to its natural meaning, and the courts should be slow to depart from that.83 Nevertheless, Hogg has observed that ‘[t]he preponderance of opinion is that entire agreement clauses will not exclude implied duties of good faith, this

77 ‘[T]here is implied an obligation of good faith (which is also termed “fair dealing” in some of the cases). This means that the parties must refrain from conduct which in the relevant context would be regarded as commercially unacceptable by reasonable and honest people. An implied duty of good faith does not mean solely that the parties must be honest’: Bates (n 55) [711]. See too, eg, Devon Commercial Property Ltd v Barnett [2019] EWHC 700 (Ch); New Balance Athletics Inc v Liverpool Football Club and Athletic Grounds Ltd [2019] EWHC 2837 (Comm) [44]. 78 Bates (n 55) [710], disapproving Beale (ed), Chitty on Contracts (n 28) para 1-058. 79 E Granger, ‘Sweating Over an Implied Duty of Good Faith’ [2013] LMCLQ 418, 424. 80 See R Calnan, ‘Controlling Contractual Interpretation’, ch 4 in this volume. 81 cf O Lando and H Beale (eds), Principles of European Contract Law (OUP 2000), who propose a mandatory duty of good faith and fair dealing (art 1:201), a default duty to ‘co-operate in order to give full effect to the contract’ (art 1:202) and the binding quality of ‘reasonable generally applicable usages’ (art 1:105). 82 EC Marconi Systems Pty v BHP Information Technology Pty [2003] FCA 50 at [922]. 83 See Arnold v Britton (n 27). cf Vodafone Pacific Ltd v Mobile Innovations Ltd (n 46).

Excluding Good Faith and Restricting Discretion  101 suggesting that a targeted exclusion might be successful’.84 Yet there is no reason why ‘targeted exclusions’ in a commercial contract should not be enforced. As Lord Bingham recognised in HIH Casualty and General Insurance Limited v Chase Manhattan Bank (regarding fraudulent misrepresentation), ‘if a party to a written contract seeks to exclude the ordinary consequences of fraudulent or dishonest misrepresentation … such intention must be expressed in clear and unmistakable terms on the face of the contract’.85 Some implied terms cannot be excluded as a result of legislation.86 But implied duties of good faith are not of the same nature. Indeed, if a commercial contract sets out the duties owed by the parties in a detailed manner then good faith may be impliedly excluded; express exclusion may be unnecessary.87

A.  Excluding a Duty to Perform the Contract Honestly The most difficult crunch point occurs when a party tries to exclude an implied term that a party perform the contract honestly. At first blush, this looks both unprincipled and impractical.88 After all, parties cannot exclude liability for fraud,89 and it may perhaps damage a party’s reputation if it seeks to exclude any term that it perform honestly.90 Nevertheless, given the expansive definition of dishonesty that has recently been favoured by the English courts, parties may try to exclude terms regarding honest performance. This is worth exploring further: if a promissory obligation to act honestly can be excluded, it follows that all duties found under the banner of ‘good faith’ can be excluded. Most objections to the exclusion of honesty rest upon a conception of honesty that is defined subjectively.91 But that is not how ‘dishonesty’ is generally understood in English law,92 and it now seems clear that a standard approach to dishonesty should be adopted throughout private law.93 Yet dishonesty remains a troubled concept.94 Honesty and dishonesty are essentially 84 M Hogg, ‘The Implication of Terms-in-Fact: Good Faith, Contextualism and Interpretation’ (2017) 85 George Washington Law Review 1660, 1688. 85 HIH Casualty and General Insurance Limited v Chase Manhattan Bank [2003] UKHL 6, [2003] 1 All ER (Comm) 349 [16]. 86 See, eg, Unfair Contract Terms Act 1977, s 2. 87 cf Mid Essex Hospital Services (n 17). 88 cf Plan B Trustees Ltd v Parker (No 2) [2013] WASC 216, (2013) 11 ASTLR 242 [232] (Edelman J). 89 Standard Chartered Bank v Pakistan National Shipping Corpn and Others (Nos 2 and 4) [2002] UKHL 43, [2003] 1 AC 959 [22]; HIH (n 85) [5], [16]. 90 Yam Seng (n 21) [135], [149]. 91 Hooley adopts a subjective definition of honesty in arguing that honesty should be non-excludable, but ‘parties retain the ability to contract out of any higher standard of behaviour’: see, eg, Hooley, ‘Controlling contractual discretion’ (n 74) 81–82, 88; SNCB Holding v UBS AG [2012] EWHC 2044 (Comm) [72] (Cooke J). 92 Ivey v Genting Casinos (UK) Ltd, trading as Crockfords [2017] UKSC 67, [2017] 3 WLR 1212. 93 ibid; Group Seven Limited v Notable Services LLP [2019] EWCA Civ 614 [57], [103]. 94 For recent difficulties in the context of dishonest assistance, see Group Seven (n 93).

102  Paul S Davies jury questions.95 In  the most recent Supreme Court decision dealing with this issue, Lord Hughes said that ‘dishonesty is by no means a defined concept. On the contrary, like the elephant, it is characterised more by recognition when encountered than by definition’.96 The inability clearly to define dishonesty may be unsettling to commercial parties. The best guidance seems to be that dishonesty incorporates an objective element:97 given what the defendant knew, would a reasonable person consider that defendant to be dishonest? A defendant may therefore be held to be dishonest even though he or she actually thought he or she was acting honestly.98 This could cause concern to commercial actors based in different countries who do not share a common cultural background. If, in good faith, they consider themselves to be acting honestly, they might be surprised nonetheless to be found to be in breach of an implied term to act honestly. In order to eliminate this risk, they may seek to exclude an implied term to act honestly, particularly after the decision of the Supreme Court in Ivey v Genting Casinos (UK) Ltd. Phil Ivey is a professional gambler. He played Punto Banco at a casino in London and won £7.7 million. The casino refused to pay because it thought Ivey had cheated: Ivey had noticed that the casino’s cards were marked and was able to exploit this through edge-sorting. The Supreme Court held that Ivey had breached an implied term not to cheat. But the Supreme Court further held, obiter, that Ivey had been dishonest, even though Ivey had plausible reasons for thinking that he was acting honestly: he did not touch the cards, the cards used were provided by the casino, and the pit manager agreed to all Ivey’s requests. But because a reasonable person would have considered Ivey to be dishonest, he was held to be dishonest. This is hard on Ivey on the facts, and may cause some concern amongst commercial parties. Dishonesty is a tricky concept for the civil law. Whilst it may have a useful role in the criminal law as a jury question, in private law it can tend to be circular.99 In Royal Brunei Airlines v Tan, Lord Nicholls said that dishonesty ‘means simply not acting as an honest person would in the circumstances’.100 This gives rise to the question ‘What would an honest person have done?’ The answer appears to be that he would not have acted dishonestly. But this is clearly circular and does not further our understanding of dishonesty. The next time a professional gambler wants to take advantage of any potential flaws in a casino’s operation, he or she may try to exclude an implied term not to act dishonestly. It is suggested that they should be able to do so. This enhances 95 Ivey (n 92) [48]; B Hale, ‘Dishonesty’ [2019] Common Law World Review 5, 12–14. Nevertheless, in Bhasin (n 25), Cromwell J said that a duty of honesty ‘is also clear and easy to apply’: [80]. This does not appear to reflect the English experience. 96 Ivey (n 92) [48]. 97 See too Yam Seng (n 21) [144]; Group Seven (n 93) ‘simplicity’ of two-stage test. 98 As happened in Ivey (n 92). The second limb of R v Ghosh [1982] QB 1053 no longer seems to apply: DPP v Patterson [2017] EWHC 2820 (Admin). 99 PS Davies, Accessory Liability (Hart Publishing 2015) ch 4. 100 Royal Brunei Airlines Sdn Bhd v Tan [1995] 2 AC 378 (PC) 389.

Excluding Good Faith and Restricting Discretion  103 the important principle of commercial certainty. It is unduly paternalistic not to allow the parties to agree to permit one party to do all they can to win. Of course, a casino may well not agree to such an exclusion, but if it does (in order to attract business, for instance) then the courts should be prepared to give effect to that clause. The same result might be reached by saying that because the parties have made it clear that ‘cheating’ is allowed, what the gambler does is no longer cheating, or dishonest, since the casino consents.101 But the outcome is the same. Indeed, it is not obvious that Lord Bingham was invariably correct to say that ‘[p]arties entering into a commercial contract … will assume the honesty and good faith of the other; absent such an assumption they would not deal’.102 Parties may be forced to contract with each other as a result of circumstances beyond their control. If both parties do not trust each other, but are nevertheless willing to enter into a contract because they decide that that is the best solution for each, then it is fictional to say that they are entitled to expect honest conduct from the other if in fact they do not.103 The parties may need to contract with each other out of necessity, but if both parties are happy to proceed with the contract in a spirit of mistrust and always checking the other party’s position, it is arguable that the law should not prohibit them from doing so.104 In any event, it is important to appreciate that allowing one party to exclude a term that the contract be performed honestly does not mean that a party could commit fraud with impunity. Rather, it means that any such fraud would not be a breach of contract.105 Instead, the innocent party would have to sue in the tort of deceit. This would affect the remedies available: the remedies for breach of contract and deceit are not the same.106 The remedies awarded for breach of a duty of good faith in Al Nehayan v Kent reflected the contractual expectation measure rather than the tortious measure. But because alternative avenues for redress exist through deceit, it is unnecessary to insist upon a mandatory contractual term of honesty. An English approach that favours a strong notion of freedom of contract is perhaps at odds with that recently adopted by the Supreme Court of Canada in Bhasin. Cromwell J thought that ‘[b]ecause the duty of honesty in contractual performance is a general doctrine of contract law that applies to all contracts, 101 cf Bhasin (n 25) [74], citing the Uniform Commercial Code (2012) § 1-302(b). 102 HIH (n 85). See too D Campbell, ‘Good Faith and the Ubiquity of the “Relational” Contract’ (2014) 77 MLR 460, 485. 103 For further discussion of the role of trust, see N McBride, Key Ideas in Contract Law (Hart Publishing 2017) ch 2. For consideration of further reasons why one party might consent to what appears to be an unbalanced contract, see H Collins, ‘Discretionary Powers in Contracts’ in D Campbell, H Collins and J Wightman (eds), Implicit Dimensions of Contracts: Discrete, Relational and Network Contracts (Hart Publishing 2003) 219, 226–31. 104 See too Hogg, ‘The Implication of Terms-in-Fact’ (n 84) 1667. 105 See too J Carter and W Courtney, ‘Good faith in contracts: is there an implied promise to act honestly?’ (2016) 75 CLJ 608. 106 S Whittaker, ‘Good Faith, Implied Terms and Commercial Contracts’ (2013) 129 LQR 463, 466–67.

104  Paul S Davies like unconscionability, the parties are not free to exclude it’.107 But it is not a term of the contract that a party will not act unconscionably. Rather, unconscionability operates as a general doctrine that may prevent a party from enforcing a contract, or allow the other party to rescind the agreement. It may be that honesty should operate in a similar manner. This is best explored in the context of restrictions on contractual powers.

IV.  The Exercise of Contractual Powers There has been a rise of cases insisting that contractual powers must be exercised honestly and not in an arbitrary, capricious or irrational way,108 which has sometimes been said to ‘boil down’ to a requirement of good faith.109 But the basis of such a restriction is again not clear. For example, it has been said that the controls may arise through construction or implication,110 without clarifying which. The focus has generally been on implied terms, although there is lively debate about whether such terms are implied in fact or at law.111 It is suggested that it may be better to have regard to well-known cases in equity on ‘fraud on a power’ in order to introduce a non-excludable restriction on parties’ ability to exercise contractual powers in a self-interested way.112 Any further restrictions should generally be based upon terms implied in fact. In any event, it is suggested that reliance on public law concepts and cases, such as Associated Provincial Picture Houses Ltd v Wednesbury Corporation,113 in the area of commercial contracts is unlikely to be helpful.114 The development of contract law in this area might mirror the experience of the equitable jurisdiction regarding control of discretionary powers. By the turn of the present century, English courts were flirting with the use of Wednesbury unreasonableness in the

107 Bhasin (n 25) [75]. 108 Most notably following Braganza v BP Shipping Ltd [2015] UKSC 17, [2015] 1 WLR 1661. 109 Hooley, ‘Controlling contractual discretion’ (n 74) 66. However, Hooley here uses good faith ‘in the sense that the party exercising the discretion must do so honestly’, which is narrower than the definition favoured in recent cases, such as Bates (n 55). cf M Bridge, ‘The Exercise of Contractual Discretion’ (2019) 135 LQR 227. 110 Watson v Watchfinder.co.uk Ltd [2017] EWHC 1275 (Comm), [2017] Bus LR 1309 [102]; BHL v Leumi ABL Ltd [2017] EWHC 1871 (QB), [2017] 2 Lloyd’s Rep 237 [40]. 111 cf Lim and Chan, ‘Problems with Wednesbury Unreasonableness’ (n 75) 102–03. 112 See similarly P Sales, ‘Use of Powers for Proper Purposes in Private Law’ (2020) 136 LQR (forthcoming). 113 Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1948] 1 KB 223. 114 See, eg, Sales, ‘Use of Powers’ (n 112); Bridge, ‘The Exercise of Contractual Discretion’ (n 109); C Himsworth, ‘Transplanting irrationality from public to private law: Braganza v BP Shipping Ltd’ (2019) 23(1) Edinburgh Law Review 1; J Morgan, ‘Against Judicial Review of Discretionary Contractual Powers’ [2008] LMCLQ 230; J Morgan, ‘Resisting Judicial Review of Discretionary Contractual Powers’ [2015] LMCLQ 483.

Excluding Good Faith and Restricting Discretion  105 equitable jurisdiction,115 but this has now been staunchly criticised and firmly sidelined. In the Court of Appeal in Pitt v Holt, Mummery LJ said: [A]nalogies with judicial review in public law are unhelpful and unnecessary. There is an elementary distinction between, on the one hand, the liability in private law of a fiduciary for breach of duty and, on the other hand, the availability of judicial review for the control of abuses of public power. There are surface similarities in the language of discretion and in the debates about the limits of discretionary power, but the contexts are so different that it is dangerous to develop the private law of fiduciaries by analogy with public law on curbing abuse of power. Judicial review in public law is concerned with the lawfulness of decisions and acts of public authorities to ensure that they are acting within the limits of a power usually set by statute. Breaches of duty in fiduciary law relate to discretionary dispositive powers privately entrusted to a fiduciary who has been selected to exercise the powers for the benefit of members within a designated class. The discretion of the fiduciary is not controlled by the court, which will not interfere with matters of judgment by the fiduciary. The only ground on which the court will review the exercise of the discretion is that of a breach of fiduciary duty. The underlying principles of fiduciary law and private property law are conceptually different from the public interest basis for reviewing the lawfulness of administrative action.116

These criticisms were approved by Lord Walker in the Supreme Court, who reiterated that any analogy with public law ‘cannot be pressed too far’.117 This rejection of the public law analogy at anything other than a high level of abstraction is sensible. Various methods have been well developed in equity for controlling discretions in an appropriate manner, and private law should properly focus its attention on these doctrines. Some could readily prove useful in the contractual sphere. In particular, the notion of ‘fraud on a power’, which ensures that powers are only validly exercised in pursuit of a proper purpose for which the power was granted, could profitably be exploited in contract law more generally. The doctrine of fraud on a power is of very long-standing.118 It is sometimes also known as the ‘proper purpose rule’.119 The nature of the rule was recently considered by Lord Sumption in Eclairs Group Ltd v JKX Oil & Gas Plc: In Duke of Portland v Topham (1864) 11 HLC 32, 54 Lord Westbury LC stated the rule in these terms: ‘that the donee, the appointor under the power, shall, at the time of the exercise of that power, and for any purpose for which it is used, act with good faith and

115 Scott v National Trust for Places of Historic Interest or Natural Beauty [1998] 2 All ER 705; Breadner v Granville-Grossman (2000) 2 ITELR 812, [2001] Ch 523; Edge v Pensions Ombudsman [1998] 2 All ER 547, [1998] Ch 512. For judicial discussion, see Seng v Tuang [2012] SGCA 41 [59]–[64]. 116 Pitt v Holt [2011] EWCA Civ 197, [2012] Ch 132 [235]. 117 Pitt v Holt [2013] UKSC 26, [2013] 2 AC 108 [11]. In the contractual context, see, eg, Lehman Brothers International (Europe) (In Administration) v Exxonmobil Financial Services BV [2016] EWHC 2699 (Comm), [2017] 2 All ER (Comm) 959 [287] (Blair J). 118 See, eg, Lane v Page (1754) Amb 233; Aleyn v Belchier (1758) 1 Eden 132. 119 Eclairs Group Ltd v JKX Oil & Gas Plc [2015] UKSC 71, [2016] 3 All ER 641 [15].

106  Paul S Davies sincerity, and with an entire and single view to the real purpose and object of the power, and not for the purpose of accomplishing or carrying into effect any bye or sinister object (I mean sinister in the sense of its being beyond the purpose and intent of the power) which he may desire to effect in the exercise of the power.’ The principle has nothing to do with fraud. As Lord Parker of Waddington observed in delivering the advice of the Privy Council in Vatcher v Paull [1915] AC 372, 378, it ‘does not necessarily denote any conduct on the part of the appointor amounting to fraud in the common law meaning of the term or any conduct which could be properly termed dishonest or immoral. It merely means that the power has been exercised for a purpose, or with an intention, beyond the scope of or not justified by the instrument creating the power.’ The important point for present purposes is that the proper purpose rule is not concerned with excess of power by doing an act which is beyond the scope of the instrument creating it as a matter of construction or implication. It is concerned with abuse of power, by doing acts which are within its scope but done for an improper reason.120

A number of points concerning the doctrine of fraud on a power should be noted that will be useful when considering how such a doctrine might apply in the contractual context. First, the doctrine of fraud on a power is distinct from requirements of good faith. A power may be exercised in good faith but for an improper purpose.121 Second, the doctrine applies to all powers. Admittedly, in Eclairs Lord Sumption said that ‘[t]he limitation of the power to its proper purpose derives from its fiduciary character’,122 but it would be a mistake to consider that the power must be fiduciary in nature for the proper purpose rule to apply. Eclairs concerned the powers given to company directors, and should be understood in that context. It is clear that the proper purpose rule is not limited to fiduciary powers.123 In Free Church of Scotland v Overtoun Lord Lindley said that it was ‘clear that there is a condition implied in … instruments which create powers, namely, that the powers shall be used bona fide for the purposes for which they are conferred’.124 In Wong v Burt, the New Zealand Court of Appeal sensibly observed that ‘[t]he notion of fraud on a power itself rests on the fundamental juristic principle that any form of authority may only be exercised for the purposes conferred, and in accordance with its terms. The principle is one of general application’.125 This opens up the possibility that it should apply to contractual powers too. 120 ibid. 121 See, eg, The Bell Group Ltd (in liquidation) v Westpac Banking Corporation (No 9) [2008] WASC 239 [4456] (Owen J). See too the separate provisions in the Companies Act 2006, s 171 (proper purpose rule) and s 172 (good faith). cf M Racynska, ‘Good Faiths and Contract Terms’, ch 5 in this volume. 122 Eclairs (n 119) [39]. 123 Aleyn v Belchier (n 118) 138–39; Pitt v Holt (SC) (n 117) [61]; Re Crawshay [1948] Ch 123; Re Brook’s Settlement [1968] 1 WLR 1661. 124 Free Church of Scotland v Overtoun [1904] AC 515, 695. See too G Farwell, A Concise Treaties on Powers, 3rd edn (Stevens and Sons 1916) 457: a ‘person having a limited power, must exercise it bona fide for the end designed’. 125 Wong v Burt [2005] 1 NZLR 91 [27].

Excluding Good Faith and Restricting Discretion  107 Third, the doctrine of fraud on a power is a mandatory rule that cannot be excluded. It applies even if a power is given to the donee in its ‘absolute ­discretion’ or ‘uncontrollable discretion’.126 Whilst it is possible to conceive of situations where a power may be exercised whimsically,127 it must still be exercised for a proper purpose.128 Fourth, the basis of the proper purpose rule is not universally accepted. It is sometimes said to be simply a rule of construction. Yet often the language used in the instrument conferring the relevant power will be entirely silent as to the purposes sought to be secured through the power. There will be no relevant words to be interpreted when undertaking the process of ascertaining what the proper purposes are. Instead, reference must be made to the particular ends intended to be achieved.129 Of course, there may well be a degree of overlap between construction and the proper purposes rule; in the equitable context, both refer to the intention of the donor of the power. The intention of the donor when conferring the power is obviously important when determining the relevant proper purposes. However, Lord Sumption was clear in Eclairs that an implied term ‘is not the basis of the proper purpose rule’.130 Instead, [t]he proper purpose rule is a principle by which equity controls the exercise of a ­fiduciary’s powers in respects which are not, or not necessarily, determined by the instrument. Ascertaining the purpose of a power where the instrument is silent depends on an inference from the mischief of the provision conferring it, which is itself deduced from its express terms, from an analysis of their effect, and from the court’s understanding of the business context.131

The context will naturally be significant when determining the proper purpose rule. Some powers are conferred for only one particular purpose, and some for a much wider range of purposes. In the latter situation, there may be virtually no ‘improper purposes’ for which the power could be exercised. Nevertheless, the rule is potentially applicable to all powers.132 There is no ‘gateway’ issue for the proper purpose rule. This might be contrasted with the discussion about whether a contractual power counts as a ‘discretion’ that is amenable to the application of ‘the Braganza duty’.133 Discretion is not a term of art, and trying to define it for the purposes of allowing in concepts from public law, for example, will invariably

126 See, eg, In Re Gulbenkian’s Settlement [1970] AC 508, 518. 127 Re Wright [1920] 1 Ch 108, 118; PJC v ADC [2009] EWHC 1491 (Fam) [15], [19]. 128 The scope of the proper purpose rule is broad. It may well be that a trustee’s duty to act in the best interests of his or her beneficiary, for example, is best considered to be one aspect of the proper purpose rule: Lord Nicholls, ‘Trustees and Their Broader Community: Where Duty, Morality and Ethics Converge’ (1995) 9 Trust Law International 71, 74; F&C Alternative Investments (Holdings) Ltd v Barthelemy [2011] EWHC 1731 (Ch), [2012] Ch 613 [229]. 129 R Nolan, ‘Controlling Fiduciary Power’ (2009) 68 CLJ 292. 130 Eclairs (n 119) [30]. cp Sales, ‘Use of Powers’ (n 112). 131 Eclairs (n 119) [30]. 132 cf G Thomas, Thomas on Powers, 2nd edn (OUP 2012) [11.76]. 133 Watson v Watchfinder.co.uk Ltd (n 110) [102] (HHJ David Waksman QC).

108  Paul S Davies prove problematic.134 The key question is often whether a right to act in a particular way is priced into the contract; determining whether something is a ‘discretion’ is irrelevant to this issue, which can readily be accommodated within the concept of proper purposes. Invoking the concept of ‘purpose’ in contract law is not novel.135 For example, in Hayes v Willoughby, Lord Sumption noted that in the law relating to contractual discretions it is important ‘to limit the decision-maker to some relevant contractual purpose’.136 And in Arnold v Britton, Lord Neuberger said that the ‘overall purpose of the clause and the lease’ should be borne in mind.137 This is consistent with the observation of Lord Wilberforce in Prenn v Simmons, that although pre-contractual negotiations should not be used in the interpretative process, ‘the factual background known to the parties at or before the date of the contract, including evidence of the “genesis” and objectively the “aim” of the transaction’ could be of assistance in deciding the meaning of a term.138 In Paragon Finance Plc v Nash,139 the lender had a discretion under the contract with a borrower to vary the interest rate. The Court of Appeal held that the power, if exercised, had to be for a proper purpose. Dyson LJ thought that ‘[a]n example of an improper purpose would be where the lender decided that the borrower was a nuisance (but had not been in breach of the terms of the agreement) and, wishing to get rid of him, raised the rate of interest to a level that it knew he could not afford to pay’.140 The discretion in that case was not given to the lender deliberately to force a borrower to default, but rather to reflect changes in the market, for instance. In Watson v Watchfinder.co.uk Ltd, HHJ David Waksman QC thought it was necessary to establish ‘what the “target” of that discretion is, in the sense of what the decision-maker is meant to be considering when deciding whether or not to 134 Dyson LJ has said that discretion ‘connotes the exercise of judgment in making choices’ (Carty v Croydon London Borough Council [2005] EWCA Civ 19, [2005] 1 WLR 2312 [25]). Such dicta can clearly apply very broadly indeed. A further difficulty is that the word ‘discretion’ often appears in contractual terms creating seemingly absolute rights: ‘absolute discretion’, ‘sole discretion’ and so on. Yet, as Foxton has pointed out, ‘[t]he term “absolute contractual right” is essentially conclusory, expressing, rather than explaining, the conclusion that the right is not subject to discretion obligations’: D Foxton, ‘A good faith goodbye? Good faith obligations and contractual termination rights’ [2017] LMCLQ 360, 372. See similarly Equitas Insurance Ltd v Municipal Mutual Insurance Ltd [2019] EWCA Civ 718 [111]–[113] (Males LJ). 135 See, eg, Weinberger v Inglis [1919] AC 606; Price v Bouch (1987) 53 P & CR 257; Gan Insurance Co Ltd (n 39) [67]; Paragon Finance Plc v Nash [2001] EWCA Civ 1466, [2002] 2 All ER 248, 262. 136 Hayes v Willoughby [2015] UKSC 36, [2015] AC 1619 [14], admittedly drawing an analogy with public law. See similarly the comments of Lord Sumption in British Telecommunications Plc (n 40) [37]. 137 Arnold v Britton (n 27) [15]. 138 Prenn v Simmons [1971] 1 WLR 1381, 1384–85. See too Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] 1 AC 1101; A Robertson, ‘Purposive Contractual Interpretation’ (2019) 39 Legal Studies 230. cf Excelsior Group Productions Ltd v Yorkshire Television Ltd [2009] EWHC 1751 (Comm) [25] (Flaux J). 139 Paragon Finance Plc v Nash [2001] EWCA Civ 1466, [2002] 2 All ER 248. 140 ibid [31].

Excluding Good Faith and Restricting Discretion  109 exercise it’.141 The ‘target’ of a discretion may be better expressed as a ‘purpose’. In any event, as the judge observed, ‘in many cases this is straightforward and stated as part of the discretion’.142 The context will also often be revealing. For example, a landlord who has a discretion whether or not to allow an assignment of a lease to a new tenant cannot refuse in order to inflict pain on the current tenant, which would be an improper purpose, but must instead consider the proper purpose of whether the assignee would be a suitable tenant.143 Of course, if the express language is silent as to purpose it may be more difficult to establish relevant proper purposes, but often this will be possible by reference to the factual matrix within which the term operates.144 An interesting recent example of the proper purposes rule being applied is the decision of the Court of Appeal in Property Alliance Group Ltd v Royal Bank of Scotland Plc.145 The case involved a swap contract, which included the following clause: The Borrower authorises the Bank from time to time to obtain an up to date Bank instructed and addressed professional valuation of all or any of the Charged Properties from a valuer/surveyor acceptable to the Bank and the Borrower shall meet the cost of any valuations obtained by the Bank provided that the Borrower shall not be liable for the cost of more than one valuation for each of the Charged Properties in any one calendar year other than a valuation obtained following the occurrence of an Event of Default.

At first instance, Asplin J emphasised that, even after Braganza, the claimant must still show that the implication of any controls over the Bank’s power to obtain a valuation must be necessary in order to make the contract work, and rejected any such limitations on the facts. This approach is appropriate when considering ‘duties’ of good faith, reasonableness and so on. But the proper purposes rule must still apply. Indeed, the Court of Appeal differed from the trial judge and held that some fetters on the power to ask for a valuation were inherent. The Court of Appeal rightly emphasised that the Bank was able to exert the power in its own interests and did not need to balance such interests with those of the Borrower, but nonetheless held that it should ‘be inferred that the parties intended the power … to be exercised in pursuit of legitimate commercial aims rather than, say, to vex PAG maliciously’.146 This is a sensible application of the

141 Watson v Watchfinder.co.uk Ltd (n 110) [105]. See too eg BHL v Leumi (n 110) [34]. 142 Watson v Watchfinder.co.uk Ltd (n 110) [105]. 143 See International Drilling Fluids Ltd v Louisville Investments (Uxbridge) Ltd [1986] Ch 513, 519–21; Bromley Park Garden Estates Ltd v Moss [1982] 1 WLR 1019. 144 eg BHL v Leumi (n 110) [34]–[36]. Of course, there will be difficult cases where views can reasonably differ: see, eg, recently British Airways Plc v Airways Pension Scheme Trustee Ltd [2018] EWCA Civ 1533. 145 Property Alliance Group Ltd v Royal Bank of Scotland Plc [2018] EWCA Civ 355, [2018] 1 WLR 3529. 146 ibid [169].

110  Paul S Davies proper purposes rule, without explicitly using such language. The purposes are gleaned from the parties’ intentions, and they may be broadly defined. Requiring a party to act in pursuit of legitimate commercial aims is not onerous and represents a suitably mild form of control. A difficult example is Mid Essex Hospital Services NHS Trust v Compass Group UK and Ireland Ltd (t/a Medirest).147 The contract contained a mechanism under clause 5.8 whereby an NHS trust could award ‘service failure points’ from which it could calculate appropriate payment deductions for a catering company’s performance failures. Jackson LJ said: There is no justification for implying into clause 5.8 a term that the Trust will not act in an arbitrary, irrational or capricious manner. If the Trust awards more than the correct number of service failure points or deducts more than the correct amount from any monthly payment, then that is a breach of the express provisions of clause 5.8. There is no need for any implied term to regulate the operation of clause 5.8.148

The power to award points and deductions was clearly given to the Trust for its own benefit in the event of defaults by the catering company. If the company failed to meet its obligations then it could not be improper for the Trust to award points and deductions. It is helpful to distinguish the narrow scope of the proper purpose rule from more broad and intrusive implied terms that may restrict the exercise of a contractual discretion. It may be that there is a small, mandatory, inherent restriction that applies to all contracts in the form of an obligation to exercise a power for a proper purpose. But that does not invariably extend to a contractual obligation to consider all relevant factors and no irrelevant factors, or to act in a reasonable way. Such terms should not invariably be inserted into all contracts. For instance, in Hamsard 3147 Ltd (t/a Mini Mode Childrenswear) v Boots UK Ltd, Norris J could ‘readily accept that there will generally be an implied term not to do anything to frustrate the purpose of the contract’ but thought that any restrictions beyond that would require extra justification.149 The proper purpose rule has been said very clearly not to rest upon an implied term,150 and should be considered mandatory. The proper purpose rule would then be similar to equitable doctrines such as unconscionability, in that it would apply generally without being a term of the contract. This will affect the available remedies. Breach of a term entitles the innocent party to damages, generally assessed by reference to its expectation loss.151 By contrast, breach of a mandatory rule renders the exercise of the power invalid in some way, presumably rendering it void or voidable.152 And of course a mandatory rule cannot, by definition, be

147 Mid

Essex Hospital Services (n 17). [92]. 149 Hamsard 3147 Ltd (t/a Mini Mode Childrenswear) v Boots UK Ltd [2013] EWHC 3251 (Pat) [86]. 150 Eclairs (n 119). 151 Morris-Garner v One Step [2018] UKSC 20, [2019] AC 649. 152 cf Pitt v Holt (SC) (n 117). 148 ibid

Excluding Good Faith and Restricting Discretion  111 excluded: if the parties expressly wanted to exclude the proper purpose rule, they would not be able to do so.153 In any event, it is possible for the parties to agree to controls beyond the narrow ambit of the proper purposes rule. But such controls will generally need to be established either as express terms of the contract, or as terms implied in fact. And, as already seen, implication is not automatic. For example, in Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2), Mance LJ observed that the ‘authorities do not justify any automatic implication, whenever a contractual provision exists putting one party at the mercy of another’s exercise of discretion’.154 Careful consideration of the particular contract is required if a term is to be implied in fact.

V. Conclusion The notion that commercial parties are under a duty to act in good faith deserves serious scrutiny.155 If commercial parties want such a duty to be incorporated into their contract then they should be encouraged to provide for it expressly.156 And the courts should enforce such terms. But where the contract is silent about the need to act in good faith, courts should generally only imply duties of good faith if they satisfy the stringent tests for implication of a term in fact. Terms should not be implied at law unless the category of contract at issue can be clearly defined. Whether there is room to imply a term will depend on the facts of a given case, but it is important to remember that ‘an implication of a duty of good faith will only be possible where the language of the contract, viewed against its context, permits it’.157 Duties of good faith – whether implied in fact or at law – can be excluded by the parties. This supports the parties’ ability to control the content of

153 cf Mid Essex Hospital Services (n 17) [83] (Jackson LJ): ‘a term [that a party not exercise its discretion in an arbitrary, capricious or irrational manner] is extremely difficult to exclude, although I would not say it is utterly impossible to do so’. 154 Gan Insurance Co Ltd (n 39) [62]. See too, eg, Eastleigh BC v Town Quay Developments Ltd [2009] EWCA Civ 1391 [30]; Watson v Watchfinder.co.uk Ltd (n 110) [102]; Myers v Kestrel [50]–[51]; Bridge, ‘The Exercise of Contractual Discretion’ (n 109). 155 Twenty years ago, Sir Thomas Bingham described good faith as ‘[t]he most important contractual issue of our time’: ‘Foreword’ in R Harrison, Good Faith in Sales (Sweet & Maxwell 1997) vi. cf White & Carter (Councils) Ltd v McGregor [1962] AC 413, 430 (Lord Reid): ‘[I]t never has been the law that a person is only entitled to enforce his contractual rights in a reasonable way and that a court will not support an attempt to enforce them in an unreasonable way.’ See too more recently SNCB Holding v UBS AG [2012] EWHC 2044 (Comm) [93] (Cooke J); also J Steyn, ‘The Role of Good Faith and Fair Dealing in Contract Law: A Hair-Shirt Philosophy?’ (1991) 1 Denning Law Journal 131, 140. 156 See too Gan Insurance Co Ltd (n 39) [99] (Staughton LJ): ‘If the parties believe that the market would welcome the term proposed in this case, or any other term on the same topic, they have only to persuade the market to adopt it for the future. I wonder whether they would succeed.’ 157 Globe Motors Inc (n 60) [68] (Beatson LJ), relying on Carewatch Care Services Ltd v Focus Caring Services Ltd [2014] EWHC 2313 (Ch).

112  Paul S Davies the agreements they voluntarily enter into. It also helps to enhance commercial certainty.158 However, the desire for some control of parties’ powers is understandable. The general, non-excludable doctrine of fraud on a power should provide limited controls on a party’s ability to exercise (contractual) powers: a power must be exercised for a proper purpose. But that does not mean that there is a term in every contract that the parties act in good faith. If well-advised commercial parties wish particular duties of good faith to govern their relationship, they should expressly provide for such duties in their contracts.

158 Morgan, ‘Against Judicial Review’ (n 114) 236–39, although cp Bhasin (n 25) [34]. See Davies ‘The Basis of Good Faith’ (n 32) 32–33.

7 The New Override of Bans on Assignment of Receivables HUGH BEALE

This chapter outlines the effects of non-assignment clauses (referred to in this chapter as ‘bans on assignment’ or ‘BoAs’) and the case that was made for overriding BoAs in receivables contracts. It compares the responses to the problem in other jurisdictions and asks why the Business Contract Terms (Assignment of Receivables) Regulations 2018 (SI 2018/1254) (‘the 2018 Regulations’ or ‘the Regulations’) are so complicated and contain so many ‘carve-outs’ compared to legislation elsewhere. It speculates on the role of the City lawyers, both in ‘signing off ’ deals and in law reform. Bans on assignment are found in various types of contract, for example in construction contracts and in contracts of loan. They are particularly frequent in contracts under which a supplier supplies goods or services to a customer on credit. The received wisdom is that BoAs in such contracts are a major hindrance to businesses’ obtaining receivables financing – and particularly to smaller businesses that are supplying larger firms. The larger firm will usually want to contract on its standard terms of purchase, and the small business may lack the bargaining power to get the BoA struck out or waived.

I.  Legislation in Other Jurisdictions The problems described in the preceding paragraph led to many jurisdictions adopting provisions that invalidate BoAs in contracts giving rise to trade receivables. The trend started in the US with the Uniform Commercial Code (UCC).1 Similar provisions have now been adopted widely,2 for example in the Personal

1 See now UCC rev § 9-406(d). 2 In many cases the reform was part of a wider reform of the law of personal property security that has taken place in many countries. Further information can be found on the website of the Secured Transactions Law Reform Project, available at https://securedtransactionslawreformproject.org.

114  Hugh Beale Property Security Acts (or PPSAs) in most provinces of Canada and in Australia;3 and also in the UN Convention on the Assignment of Receivables in International Trade of 2001.4 Some BoAs are also overridden in a number of civilian systems, for example in Germany5 and France.6 In the common law countries, the override takes one of two typical forms: either the BoA is simply invalidated, as in the UCC, or it is invalidated save to the extent that the debtor is liable in damages for any loss caused to the debtor. The Saskatchewan PPSA, section 41 is typical of the newer PPSAs in Canada:7 (9) A term in a contract between a debtor on an account or on chattel paper and an assignor that prohibits or restricts assignment of the whole of the account or chattel paper for money due or to become due: (a) is binding on the assignor, but only to the extent of making the assignor liable in damages for breach of contract; and (b) is unenforceable against third parties.

The Law Commission recommended a similar provision as part of its proposed reform of Company Security Interests.8 For reasons that had nothing to do with this provision, none of the Law Commission’s recommendations were adopted.9

II.  Summary of the 2018 Regulations English law has now also adopted an override of BoAs in contracts giving rise to many forms of trade receivables. Broadly speaking, the 2018 Regulations render ineffective any term in a contract that gives rise to a ‘receivable’ if the term prohibits or imposes conditions or restrictions on assignment of the claim.10 3 Personal Property Securities Act 2009 (Australian) 2009, s 81. 4 UN Convention on the Assignment of Receivables in International Trade (New York, 2001) Art 9. 5 §354a Handelsgesetzbuch makes an assignment of a claim arising from a reciprocal commercial contract, with the exception of loan agreements, effective despite a BoA. However, a debtor who pays the assignor even though it has been given notice of the assignment will get a good discharge. See H Beale and G Ringe, ‘Transfer of Rights and Obligations’ in G Dannemann and S Vogenauer (eds), The Common European Sales Law in Context: Interactions with English and German Law (OUP 2013) 521, 539–41. 6 On French law see 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, 228, which also lists other civil law countries that have an override at least in some cases. 7 The Personal Property Security Act, 1993 (Saskatchewan). The first PPSA, in Ontario, did not originally contain an override, but this was heavily criticised and an override (in similar terms to the other Canadian PPSAs) was introduced in 2006: Ontario Personal Property Security Act, s 40(4), as amended by 2006, c 34, sch E, s 11(4). 8 See Law Commission, Company Security Interests (Law Com No 296, 2005), para 6.73 and draft reg 35. 9 For the fate of the Law Commission recommendations, see H Beale, M Bridge, L Gullifer and E Lomnicka, The Law of Security and Title-based Finance, 3rd edn (OUP 2018), paras 23-161–23-173. 10 Business Contract Terms (Assignment of Receivables) Regulations 2018, reg 2(1) (unless indicated otherwise, references to regulations in this chapter are references to these Regulations). An issue

New Override of Bans on Assignment of Receivables  115 Imposing conditions or restrictions includes using a confidentiality clause, to the extent that the clause would prevent the assignee from obtaining listed items of information that are essential to determining the validity or value of the claim.11 The BoA may be in the contract that gives rise to the receivable or, to prevent evasion of the override by imposing a BoA in a secondary contract, in another contract between the same parties. However, the override is limited in two main ways. First, there are exceptions for many types of contract. The principal exception is of contracts for ‘prescribed financial services’, as listed in the enabling Act,12 with additions for various similar contracts.13 There are also exclusions for a number of other types of contract, including contracts concerning any interest in land, petrol licences, the acquisition, disposal or transfer of an ownership interest in a firm, and contracts entered by project companies, nuclear decommissioning contracts and operating leases.14 The exclusion of these contracts has a double operation: (i) sums due for goods, services or intangible assets under such contracts are outside the definition of ‘receivable’; and (ii) if the term is not in the contract that gives rise to the receivable but in a secondary contract between the same parties, and the secondary contract is on the list of exceptions, the prohibition in the secondary contract on assigning receivables due under the first contract is not affected by the override. The second way in which the override is limited is that it does not apply if, at the time of the assignment, the supplier was a large company or limited liability partnership (LLP) (within the meaning of the Companies Act 2006, and judged by the last accounts filed), or a special purpose vehicle (SPV) incurring a liability of more than £10 million.15 The Regulations apply to contracts governed by English law only, unless the choice of another law appears to be an attempt to avoid the effect of the Regulations;16 and they apply if any of the parties entered the contract in the course of carrying on a business in the UK.17

that is not addressed explicitly by the Regulations is whether the override of BoAs applies equally to prohibitions on creating security over the receivables by means of a charge, since arguably an equitable charge does not involve an assignment. The Small Business, Enterprise and Employment Act 2015 (‘the 2015 Act’) refers only to overriding bans on assignment and does not mention charges. The draftsman appears to have taken the tenable view that as s 136 of the Law of Property Act 1925 refers to assignments ‘by way of charge’, for the purpose of legislation, assignment includes charges. In any event, if a simple BoA does prevent the creation of a charge then it will no longer do so when the Regulations apply; if assignment does not include a charge, it would be necessary to use different wording in order to prevent the creation of charges, and it is not obvious that a customer would have any practical interest in seeking to prevent a charge when it cannot prevent an outright assignment. However, the legal effect of a BoA on charges remains uncertain. 11 See reg 2(2) and (3). 12 2015 Act, s 2. Reg 1 adds agreements regulated by the Consumer Credit Act 1974 to the definition. 13 Reg 4(j), on which see section XI.B. 14 See reg 4. 15 Nor do the Regulations apply if either party was a consumer: reg 4(c). On this see section VII.A. 16 Reg 1(4). 17 See reg 4(d), and further in section VIII.A.

116  Hugh Beale What is obvious is that the English Regulations are very much more complex than anything found in the other jurisdictions or the UN Convention. To be quite blunt, they look like a ‘dog’s breakfast’ of qualifications and exceptions, unnecessarily complex and hard to understand, an example of the kind of thing the Better Regulation agenda railed against.18 The main question I want to address in this chapter is why the Regulations became so complex and whether the complexity – in particular the many exclusions – is really needed. But first we should consider briefly the legal effect of a BoA and why an override is needed at all.

III.  The Legal Effect of a Ban on Assignment Space does not permit any detailed discussion of the legal effect of a BoA.19 Suffice it to say that a BoA does not affect either the contractual or, seemingly, the proprietary position as between the assignor and assignee.20 At any rate it does not prevent any monies received by the assignor being held in trust for the assignee. The assignor’s right to enforce the contract may also be held in trust for the assignee,21 but whether that gives the assignee the right to use the Vandepitte procedure22 to require the assignor to enforce the claim has been the subject of differing views in the Court of Appeal.23 What is clear from Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd24 is that the BoA prevents the assignee from acquiring any right to sue the debtor in its own name. Moreover, as the debtor is entitled to ignore any notice of assignment, a BoA enables the debtor to rely on any set-off it has against the assignor, even if the set-off arose only after notice of assignment has been given. 18 ‘Regulations should be clear and simple, and guidance, in plain language’: Better Regulation Task Force, Principles of Good Regulation (2003) 5. 19 For a discussion and further references, see Beale et al, The Law of Security and Title-based Finance (n 9) paras 7.84–7.90. 20 See M Bridge, ‘The Nature of Non-Assignment Clauses’ (2016) 132 LQR 47. Pace G Tolhurst and J Carter, ‘Prohibitions on Assignment: A Choice to Be Made’ (2014) 73 CLJ 405, the courts do not seem to accept that the parties can determine whether or not there is a property right (which I think the authors confuse with the parties’ ability to define what the contract rights are to be, eg that the right might be conditional or revocable; that is a different question). 21 Don King Productions Inc v Warren [1998] 2 Lloyd’s Rep 176 (Ch) (Lightman J), [2000] Ch 291 (CA). I shall not go into the question of whether a trust will be implied or imposed if there is no declaration of trust, merely a failed equitable assignment of the outstanding rights; nor of whether a clause purporting to forbid a declaration of trust as well as assignment would be contrary to public policy. 22 Vandepitte v Preferred Accident Insurance Corporation of New York [1933] AC 70, PC; see Beale et al, The Law of Security and Title-based Finance (n 9) para 7.88. 23 Barbados Trust Co Ltd v Bank of Zambia [2007] EWCA Civ 148, [2007] 2 All ER (Comm) 445. What was said was obiter, because the Court held that the Bank of America had no right to sue the Bank of Zambia (BoZ) because the assignment to the Bank of America had been made without the BoZ’s permission. This suggests that the Court thought that a trust would not arise automatically from a failed assignment, otherwise the question would have arisen whether the Bank of America could also have used (or have been forced to use) the Vandepitte procedure: see Bridge, ‘The Nature of Non-Assignment Clauses’ (n 20) 64. 24 Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85 (HL).

New Override of Bans on Assignment of Receivables  117

IV.  Why an Override is Desirable Space also does not permit any detailed discussion of the case for or against having a legislative override of at least some BoAs. The received wisdom was that BoAs in contracts giving rise to trade receivables were unnecessary on the one hand and a major obstacle to companies’ obtaining receivables financing on the other. Such BoAs are used to avoid: (a) the debtor’s having to change the destination of payments; (b) the risk of overlooking a notice of assignment; (c) the loss of the right to rely on set-offs of claims that are not related to the contract, and which arise only after the debtor has received notice of assignment; and (d) the risk that the rights will be assigned to a new creditor who has a less accommodating or more aggressive attitude than the assignor. It was argued that these reasons for BoAs have little relevance to trade receivables, where the goods or services will already have been delivered or performed and the question is simply one of whom the customer has to pay. With modern accounting and banking systems, changing the destination of the payment is hardly burdensome, and no well-run business should overlook notices of assignment. The customer will always have the right to rely on defences under the contract and claims under closely-related contracts;25 set-off of independent claims will very seldom be relevant as the supplier is unlikely to owe money to the customer. As to the last argument, it is recognised that the last reason is particularly strong in the case of financial services such as bank loans. A debtor that has taken a loan, particularly a loan that can be recalled on short notice, from Nice Bank might be justifiably aggrieved to find that it now has to deal with Nasty Bank. So BoAs are common in such contracts, and should not be interfered with. But because with trade receivables we are dealing with simple payment of a liquidated sum, it was argued that it really makes no difference whether the customer is liable to pay the supplier or a receivables financier. Bans on assignment were thought to be a major hindrance, because the receivables financier would not be willing to buy or take security over a debt that it could not enforce; and even if only some of the supply contracts contained BoAs, receivables financiers would find it impractical to examine each one to see if it contained a BoA, so they would simply refuse to finance the supplier. It was noted that BoAs are very seldom found in the terms and conditions used unless the debtor (whom I normally refer to as ‘the customer’) has significantly more bargaining power than the assignor (‘the supplier’).



25 See

further section VIII.B.

118  Hugh Beale

V.  Empirical Studies Whatever the law, it seems that receivables financiers would rather not rest on the chance that they might have a remedy. So the Asset Based Finance Association (ABFA) asked Louise Gullifer and me (as people who had been involved in the Law Commission’s work) to conduct a study for them. ABFA obviously hoped that this could be used to persuade the Government to introduce an override on BoAs in English law, but it was readily agreed that our conclusions should be wholly independent. The outcome of this study, conducted in 2011, and a further one that we conducted with Sarah Paterson in 2014, is described in detail elsewhere.26 I shall just outline the principal conclusions, which did not all support the case ABFA hoped to make. We concluded that BoAs do not cause a major problem for suppliers who can be relied on to operate a trust account and will be offered non-notification financing. The financier reserves the right to give notice to the customer only if and when the supplier gets into financial difficulty; and to overcome the lack of an action in their name, financiers have developed a workaround. The financier will take a floating charge over the supplying company’s assets, so that if the supplier goes into insolvency the financier can appoint a ‘friendly’ administrator who will be prepared to collect the outstanding receivable on the financier’s behalf.27 Bans on assignments do cause a real problem for receivables sold through websites, when the ‘charge workaround’ will not be feasible; and for small and start-up businesses, who are not trusted to be able to operate a trust account. Contrary to the received wisdom, it is not hard for the financier to check the supplier’s contracts – most small suppliers and start-ups will do most of their business with just a few customers, on the customers’ standard terms, which will be the same for each contract. So the financier will check, and if the contract does contain a BoA, it may approach the customer to see if the customer will waive the BoA in this case. Sometimes the customer will agree, but not always. If the customer will not agree, financing on the receivables concerned will have to be refused. It is possible that some small suppliers will be offered financing arranged by the customer, usually through a financial institution that will buy the debts. This is called ‘supply chain finance’. The customer will waive the BoA so that the financial institution can buy the debts and acquire direct rights, should they be needed, against the customer. But supply chain finance is not offered by many customers, and probably is not available to start-up businesses that will only be offered finance on a notification basis. 26 See Beale et al, ‘A case for interfering with freedom of contract?’ (n 6). A shorter version is H Beale, L Gullifer and S Paterson, ‘Ban on assignment clauses: views from the coalface’ (2015) 30 Butterworths Journal of International Banking and Financial Law 463. 27 The industry’s confidence in the ‘charge workaround’ may be misplaced, as the courts might hold that the BoA invalidates the charge too; see n 10.

New Override of Bans on Assignment of Receivables  119 On the other side of the equation, given that we are primarily concerned with a classic case of inequality of bargaining power in which the general argument for freedom of contract loses much of its force, the arguments for continuing to allow prohibitions on the assignment of trade receivables seemed weak. Customers told us that they do care about whom they have to deal with when a problem arises, for example if the goods have proved to be defective. They would prefer to deal with the original supplier, with whom they have a ‘commercial relationship’ (in other words, who hopes to get the customer’s business in future), than with a financier, who may not have the same attitude. But in most cases, as soon as the customer raises an issue about the debt, the financier will re-assign it to the supplier to sort out – so the customer ends up dealing with the supplier anyway. Some customers were concerned about losing the right to rely on set-offs that arise only after the assignment has been notified. However, customers who were seriously concerned appeared to be a minority and seemed pushed to produce examples of problems. This is particularly the case when the financing is done on a ‘non-notification’ basis. In the article we wrote about the two studies, however, we did note: We accept that the position on some points (such as set-off) may be more complicated for large and larger mid-cap companies with complex contractual relationships than for SMEs [small and medium-sized enterprises]. Consideration could, therefore, be given to restricting the reform to SMEs.

But we continued: However, such a restriction would almost inevitably turn out to be under- or overinclusive, and it may be unnecessary. We understand that suppliers who are large enough to be able to influence the terms on which they supply customers do not agree to a BoA in the first place; so in practice a general override of BoAs would not affect them.28

VI.  The Government’s Reaction How much influence our 2011 report or the later article had is impossible to tell, but the Department of Business, Innovation and Skills (BIS) was somehow sufficiently persuaded that there was a case, to ask, in its Discussion Paper, Building a Responsible Payment Culture, published in December 2013, whether removing contractual barriers to selling invoices (eg as a result of a ban on assignment) would be helpful to small businesses by increasing their access to services such as factoring and invoice finance. In December 2014 the Department reported that ‘a majority of respondents [had] agreed that removing contractual barriers to selling



28 Beale

et al, ‘A case for interfering with freedom of contract?’ (n 6) 231.

120  Hugh Beale invoices would help increase access to finance’.29 Whether a minority objected or simply expressed no opinion on the point, which was just one among many in the Consultation Paper, does not appear. I cannot see any law firms or City of London Law Society Committees in the list of Respondents. That is perhaps not surprising given the title of the Paper and the breadth of the topic. In any event, in May 2014 the Government announced its intention to ­legislate,30 and clauses 1 and 2 of the Small Business, Enterprise and Employment Bill provided the broad legislative power to do this. The clauses were identical to the sections of the 2015 Act. In sum, they empower the making of regulations to render ineffective any term in a contract for goods, services or intangible assets, other than an excluded financial services contract,31 if the term prohibits, or imposes a condition or other restriction on, the assignment by a party to the contract of the right to be paid under the contract or any other contract between the parties, provided that at least one of the parties has entered into it in connection with the carrying on of a business. Thus the intention from the start was that the assignment of rights to payment under financial services contracts would not be affected.32 The Department reported that it had no evidence to indicate that the problems encountered by SMEs in accessing invoice-based finance extend to contracts for financial services;33 and the Paper pointed out that with a bank loan agreement, the borrower will want to prevent assignment by the bank. Equally, when a bank takes a floating charge over a borrower’s receivables, the bank will want to prevent the borrower from assigning its receivables to third parties.34 Draft Regulations were published in December 2014 alongside a Consultation Paper35 that asked for views on the draft. As we shall see, the subsequent history was one of a steady accumulation of exceptions to and limits on the application of the Regulations. Although what follows is a broadly chronological account, in order to avoid too much fragmentation I shall explain what happened to most of the exceptions and limitations topic by topic. 29 Department for Business, Innovation and Skills, Small Business, Enterprise and Employment Bill: Nullification of Ban on Invoice Assignment Clauses (December 2014) available at https://assets. publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/392477/bis14-1232-nullification-of-ban-on-invoice-assignment-clauses-consultation.pdf, para 6.5. 30 Department for Business, Innovation and Skills, Building a Responsible Payment Culture: Government Response (May 2014) available at https://assets.publishing.service.gov.uk/government/ uploads/system/uploads/attachment_data/file/273436/bis-13-1234-building-a-responsible-paymentculture.pdf, para 102. 31 Section  2 comprises a list of financial services, which is based on s 40 of the Terrorist Asset Freezing, etc Act 2010. 32 A similar exception is found in the 2001 UN Convention on the Assignment of Receivables in International Trade, Art 9. 33 BIS, Nullification of Ban on Invoice Assignment Clauses (December 2014) (n 29) para 2.5; see also para 7.9. The Law Commission had excluded only insurance services and brokerage fees: see draft Bill, cl 2(3). 34 See BIS, Nullification of Ban on Invoice Assignment Clauses (December 2014) (n 29) para 7.7 35 See n 29.

New Override of Bans on Assignment of Receivables  121

VII.  Early Exceptions The draft Regulations of December 2014 were relatively short, but they contained some exclusions not pre-figured in the Bill, nor commonly found in other legislation.

A.  Consumer Contracts One exception is that whereas the Bill covered contracts where one party was a business, the Regulations excluded business-to-consumer contracts. This is also in the final version. In contrast, much of the legislation elsewhere, and the 2001 Convention too, also applies to payments due on credit cards, where the customer will often be a consumer. However, this difference has no practical meaning: what consumer will even try to insert a BoA into his or her contract with the credit card company, let alone succeed in doing so?

B.  Contracts Involving Interests in Land A second exclusion that has also survived into the final Regulations is for contracts that create an interest in land. The Government stated that it did not wish to interfere with, for example, ‘a landlord’s right to prevent a business tenant from assigning their property to another individual in a similar way to a subletting arrangement’.36 That seems sensible enough, but wholly irrelevant: the right to occupy is clearly not covered by the definition of receivables. Why a landlord should not be able to assign the rent to a receivables financier is far from obvious. However, ABFA did not seem bothered about this exclusion. Whether this is because they seldom fund landlords in this way, or because few tenants have the clout to require such a term in a lease, I do not know.37

C.  Supply Chain Finance A further exception was to allow supply chain finance providers to prevent the supplier from assigning to anyone else. As to the former, the Government had 36 See BIS, Nullification of Ban on Invoice Assignment Clauses, December 2014 (n 29) para 7.12. 37 During consultation, only two respondents argued against the exclusion, saying that landlords should be able to assign their right to receive rental income to a third party: Department for Business, Innovation and Skills, Nullification of Ban on Invoice Assignment: Summary of Response (March 2015) available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/ attachment_data/file/408130/bis-15-165-nullification-of-ban-on-invoice-assignment-clausessummary-of-responses.pdf (Responses) 6. Interestingly, Art 9 of the 2001 Convention (n 32) contains a similar exclusion.

122  Hugh Beale been told by some supply chain finance providers that it would be uneconomic to provide supply chain finance if their exclusivity clauses were overridden.38 However, in the consultation process, invoice financers, who in some cases also provided supply chain finance, thought that it would remain commercially viable without an exception, and voiced concerns that debtors would place exclusivity clauses into contracts for disingenuous reasons.39 The Government’s response was to drop this exclusion.40

D.  Confidentiality Clauses We were told that some customers use confidentiality clauses that might make it practically impossible for the supplier to sell its receivables. On confidentiality clauses, the Government had a change of heart. In 2014 it believed that in cases of commercial confidentiality, debtors should be able to maintain the contractual freedom to reserve confidentiality, even if that would prevent an assignment. However, the Government was persuaded that, on the one hand, this would enable customers to defeat the purpose of the legislation by forbidding their suppliers from revealing information without which receivables financiers would be unable make advances;41 and on the other, that normally the information that is essential to the financier will not include matters that must be kept confidential for commercial reasons.42 For example, it will suffice if the goods or services supplied are described only so far as is necessary to identify them. Although the Government response to the 2014 consultation stated that the legislation should ‘Permit debtors to take action against suppliers if they breach commercial confidentiality’, the revised draft Regulations of 25 September 2015 provided that a confidentiality clause would be caught to the extent that it prevented the assignee from determining the validity or value of the receivable. Even though this approach would not affect most if not all information that should be kept confidential, there were still concerns in some quarters, so in December 2017 a new approach was adopted: rather than rely just on a general prohibition, the Regulations should provide a list of information that must be available to the assignee. The final version of the Regulations provides that a term that prevents the assignee from determining the validity or value of the

38 BIS, Nullification of Ban on Invoice Assignment Clauses, December 2014 (n 29) para 7.13. 39 Responses (n 37) 7. 40 Department for Business, Innovation and Skills, Government response to invoice finance: nullifying the ban on invoice assignment contract clauses (August 2015) available at https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/451773/BIS-15-44 1-nullifying-the-ban-on-invoice-assignment-contract-clauses-government-response.pdf. 41 Responses (n 37) 7. 42 ‘None of the responses that agreed with the exemption were able to give examples of where their commercial confidentiality could be impinged by an assignment.’ It was also pointed out that for VAT purposes, normal invoice information is already exempt from confidentiality by HMRC and the supplier’s auditors, insolvency practitioners and legal advisers (ibid).

New Override of Bans on Assignment of Receivables  123 receivable is invalid as a ‘condition or other restriction’ on assignment, and lists the information that the assignee must not be prevented from obtaining.43

E.  Right to Damages Some versions of the legislative override of BoAs, though not the UCC, include a provision that the override does not affect the liability of the debtor in damages.44 It is not easy to see what loss the creditor would be able to prove was caused by a breach,45 but our survey revealed fears that suppliers might be put off from seeking receivables financing for fear of potential liability.46 Therefore we suggested a straightforward nullification without the possibility of a damages claim between the customer and supplier for breach of contract, following the US model.47 The 2014 Consultation Paper48 raised the question whether customers should be allowed to insert terms making the debtor who assigns liable in damages, but a majority appeared to have thought that a provision to this effect was either unnecessary or might be misused for ‘disingenuous reasons’.49 The Government response to the consultation does not mention the issue.

VIII.  The Legal Profession Begins to Respond By now the legal profession had been alerted to the proposed changes. In February 2015, the Financial Law Committee (FLC) of the City of London Law Society (CLLS) (chaired by Richard Calnan) published a response,50 which was supported (for the reasons given by the FLC) by the CLLS Litigation Committee and, it seems, by four law firms.51 It was argued that any override was wrong in principle as an unjustified interference with freedom of contract.52 It was preferable to leave the parties to solve the problem by agreeing suitable terms in their contracts.53

43 2018 Regulations, reg 2(2) and (3). 44 eg Saskatchewan PPSA, s 41(9). The phrase ‘only to the extent of making the assignor liable in damages’ is to prevent the debtor from exercising other remedies, especially treating a breach of the BoA as a ground for termination. cf the 2001 Convention (n32), Art 9(2). 45 See Beale et al, ‘A case for interfering with freedom of contract?’ (n 6) 217. 46 ibid 225. 47 ibid 229. 48 BIS, Nullification of Ban on Invoice Assignment Clauses, December 2014 (n 29) para 7.12. 49 Responses (n 37) 9. 50 City of London Law Society, Nullification of Ban on Invoice Assignment Clauses (2015) available at http://www.citysolicitors.org.uk/storage/2015/02/nullification-of-ban-on-invoice-assignmentclauses.pdf. 51 From the list of respondents, the firms appear to have been Allen & Overy, Clifford Chance, Herbert Smith Freehills and Travers Smith. 52 City of London Law Society, Nullification of Ban (n 50) paras 3–5. 53 ibid paras 6–7.

124  Hugh Beale If powerful counterparties are causing problems for small businesses, they should be dealt with by the competition authorities. It was pointed out that frequently the party due to pay will want ‘to control the identity of its payee’;54 while it was agreed that financial services contracts should be excluded, even though a list was bound to involve arbitrary and difficult-to-apply borderlines, ‘what is true of financial services contracts is doubtless true of many other business contracts where the parties should be free to agree the terms on which they contract’.55 However, the CLLS did not supply any details of what such contracts might be, and it was far from clear that this was any more than another form of the general argument for freedom of contract. The CLLS paper went on to argue that if reform were to go ahead, it should be limited to payees who are small businesses and should have limited territorial application.56 In addition, the customer’s right of set-off should be preserved. I shall return to the question of small businesses later; it is convenient to explain what happened on the other two issues first.

A.  Territorial Application The CLLS was quite right to raise the issue of territorial application and the conflict of laws. I was embarrassed that I had failed to raise such an obvious issue with the Department. It was helpful that the CLLS set out some of the questions that needed to be addressed. Agreeing a solution, however, was less straightforward. On the one hand, it was readily accepted that the legislative override should apply to contracts governed by English law57 if the parties are both based in the UK; and, at the other end of the spectrum, it was finally agreed that the override should not apply to contracts that have no connection with the jurisdiction save that the parties have chosen English law to govern them58 – though in the end it was decided to adopt the simpler approach of excluding contracts into which neither party had entered in connection with doing business in the UK.59 With more hesitation on the part of some practitioners, it was also agreed that parties should not be able

54 ibid para 15. 55 ibid paras 32–34. 56 ibid para 9; and see also para 19 (on small businesses) and paras 20–24 (territorial application). 57 The Explanatory Note states that the Regulations apply only to terms in contracts governed by the law of England and Wales or of Northern Ireland. I am not quite clear how contracts governed by Scots law are excluded, given that ss 1 and 2 of the 2015 Act also apply to Scotland (see s 163), and I do not see any express restriction in the Regulations themselves, though reg 1(4) (see n 60) clearly treats Scots law as ‘foreign’. 58 cf Unfair Contract Terms Act 1977, s 27(1). At one point ABFA argued that the Regulations should apply to all contracts governed by English law: see their submission to BIS, Nullification of Prohibitions Against Assignment, 9 November 2015. 59 See the 2018 Regulations, reg 4(d).

New Override of Bans on Assignment of Receivables  125 to choose the law of Scotland or another jurisdiction solely in order to avoid the Regulations.60 What about overseas parties? The initial Government response was that the Regulations should apply if at least one of the parties ‘carries on business within the UK’.61 It was strongly argued by some practitioners that even when the contract is governed by English law, the override should apply only where the supplier was acting in the course of business in the UK,62 as the purpose is to help UK suppliers. At the end of the day, the Government was swayed by a letter from the FCI, a ‘global association’ whose membership includes many major receivables financiers. The FCI argued that to limit the application of the Regulations to suppliers based in the UK would be seen by its members, and also by governments, as unfair discrimination against overseas exporters, at a time when the UK will shortly need to negotiate free trade agreements ‘in a spirit of fairness and evenhandedness’.63 Therefore the final version of the Regulations applies unless neither party entered the contract in the course of carrying on a business in the UK.64

B.  Preserving Rights of Set-off Although our survey had suggested that customers were not particularly concerned about set-offs that arise only after they have been given notice of assignment,65 practitioners pressed the Department hard to provide that any override of BoAs should not affect the customer’s right to set off unrelated claims that arise after the debtor has been give notice of an assignment.66 However, there are strong counterarguments. First, if the right to rely on new set-offs were to be left unaffected, the receivables financier (the assignee) would be left in a position where it could not protect itself against the build-up of set-offs by the debtor (which would make the debt much less valuable than when the financier bought it), even if it gave notice to the debtor. Second, the change is unlikely to be a problem for invoice financiers, 60 See reg 1(4), which is adapted from Unfair Contract Terms Act 1977, s 27(2). The CLLS (13 October 2017) argued that a provision of this kind would cause unacceptable uncertainty, but the Unfair Contract Terms Act, s 27(2) does not seem to have done so. 61 BIS, Government response to invoice finance (n 40). Curiously, the next draft of the Regulations (September 2015) did not address the issue. It was raised again by the CLLS on 8 October 2015. 62 See CLLS letter of 8 October 2015 and its Note of 5 December 2017, para 13. 63 Letter of 22 December 2017 to the Department for Business, Energy and Industrial Strategy (BEIS), as BIS was re-named in July 2016. The letter also claims that excluding import contracts would deny foreign exporters the benefit of the FCI’s ‘two-factor’ system. This is a procedure whereby a factor in the exporter’s country advances funds to the exporter after a second factor, in the importer’s country, has evaluated the importer and set a credit limit. The second factor also collects the debt and remits the funds to the export factor (see further at https://fci.nl/en/solutions/factoring/how-does-it-work). ‘Deny’ seems an exaggeration. 64 Reg 4(d). 65 Our surveys did not ask financiers specifically if the preservation of set-off rights after notice of assignment would be a significant problem for them. 66 See in particular the CLLS letter of 8 October 2015, 2–4.

126  Hugh Beale since notice to debtors is rarely given except immediately before enforcement. Factors, who give notice to debtors immediately after assignment and who then collect in the debts themselves, would be more severely affected by the suggested provision. Since the suppliers who factor debts are at present most harshly hit by BOAs, it would seem unfortunate if BoAs were partially overridden without affecting set-off, only for it to be difficult for such suppliers to obtain finance because of financiers’ concerns over the availability of set-off. I think much of practitioners’ concern was based on uncertainty as to which forms of set-off would be affected, particularly in the supplier’s insolvency. In sum, an assignee always takes ‘subject to equities’, and this is taken to mean that the debtor can raise against the assignee (i) any defence or claim that arises out of the same contract;67 (ii) claims arising under another contract that is so closely connected that it would be manifestly unjust to enforce payment without taking into account the cross-claim (this is sometimes called ‘equitable set-off ’); and (iii) also, even in the assignor’s insolvency, any other claim that the debtor may have against the assignor, provided that the claim arose before the debtor was given notice of the assignment.68 It seemed to be accepted that it is not wholly unfair to prevent a debtor who has been given notice of the assignment from relying on claims arising under fresh contracts to reduce the assignee’s recovery. The anxiety appeared to be over the extent to which rules (i) and (ii) would continue to apply: might they be treated as ‘a condition or restriction’ on the assignment of the receivable? When an assurance was given that, on the authorities, all the set-offs remain available (save, as noted, for unrelated claims that arose after the date of notice) and that the Regulations would not alter this in any way, the demand for ensuring that the override would not affect the debtor’s rights of set-off seemed to subside, and the final Regulations make no provision for it.69 But to make it absolutely clear, the Explanatory Note provides: A contractual right of set-off which the debtor could have exercised against the assignor prior to the assignment or but for the assignment is not a term that imposes a condition or other restriction on the assignment of a receivable for the purposes of these Regulations.

C.  Small Businesses From the outset it was recognised that the aim of the reform was to make it easier for small businesses to obtain financing. However, it was thought that it is only when the supplier has very little bargaining power with the customer that BoAs are found in receivables contracts; presumably because even if the proposed contract 67 eg that the goods supplied were defective and were worth less than they should have been, and perhaps have caused damage, both of which claims can be used to reduce the price the buyer must pay: Sale of Goods Act 1979, s 53(1)(a). 68 See Bibby Factors Northwest Ltd v HFD Ltd [2015] EWCA Civ 1908, [2016] 1 Lloyd’s Rep 517. 69 See note of Meeting on 20 December 2017, para 3.

New Override of Bans on Assignment of Receivables  127 is on the customer’s standard terms, the supplier will have any BoA deleted if it has enough influence.70 As it is difficult to find criteria by which to identify ‘small’ businesses, particularly criteria that can readily be verified by the parties involved, and as in any event any dividing line is bound to be somewhat arbitrary, the initial approach was to avoid the problem by applying the override to all receivables contracts, irrespective of the size of the supplier. This is the approach taken in all the other legislation. As noted earlier, the CLLS raised this issue in its response of February 2015; and at a meeting of the Department, the CLLS, ABFA and others in March of that year, ABFA agreed to produce evidence of what effect such a restriction of the override might have on its member’s activities. ABFA reported that if the override were limited to cases in which the supplier had a turnover of £25 million or less (a definition used for other purposes in the 2015 Act), this would include between 91 and 99  per  cent of the industry’s current clients in Great Britain. Therefore, though ABFA still had a ‘strong preference … for as broad a definition as possible and ideally no restriction at all’,71 it concluded that ‘it might be pragmatic for the measure – at least initially – to be focused for the benefit of SMEs in the implementing Regulations’.72 Despite this concession, the Department did not at this stage offer to restrict the override in this way. In part I may be to blame: I continued to emphasise the difficulties that would be involved. I now regret having done so; I think that if the proposal had been limited to SMEs earlier, a great deal of the subsequent debate would have been avoided and many of the exceptions would not have been demanded. But we were still thinking that the practical issue was confined to relatively simple contracts for receivables between parties of very unequal bargaining power. The next set of draft regulations73 had exemptions only for financial services contracts, contracts creating an interest in land or for a tenancy, and contracts where one party was not a business. We shall come back to small businesses in section XI.

IX.  Special Interest Exceptions: Other Government Departments? There followed a lengthy period of silence from BIS, broken in July 2016 by an email, attaching a fresh draft seemingly sent only to ABFA and us as consultants. 70 Beale et al, ‘A case for interfering with freedom of contract?’ (n 6) 207 (standard view) and 222 (seems to confirm). See also Asset Based Finance Association, Focusing nullification of effective bans on assignment for the benefit of SMEs (April 2015) para 2.3: ‘the onerous contractual practice of imposing bans on assignment is by far most frequently seen where there is a smaller supplier supplying to a larger customer’. 71 ABFA, Focusing nullification of effective bans (n 70) para 3.6. 72 ibid para 1.3. 73 18 September 2015. The CLLS letter of 8 October 2015 points out that there is no restriction to SMEs.

128  Hugh Beale The draft included a number of additional exclusions, for petroleum licences, electricity supply contracts involving contracts for difference and contracts with implications for national security. The email stated these resulted from consultation by the Department. This seems to have been an internal consultation with other departments, as no public consultation documents had been issued.

A.  Petroleum Licences and Electricity Supply Contracts Involving Contracts for Difference The justification for exceptions for petroleum licences and electricity supply contracts involving contracts for difference, which appear in the same form in the final Regulations, has never been explained.74 The exclusion for petroleum licences seems to be in order to preserve the effect of model clauses under existing regulations preventing a licensee that agrees to resell petrol that has not yet been ‘won’ from assigning its right to payment for the ‘unwon’ product without the Minister’s consent.75 Similarly, the exclusion for electricity supply contracts involving contracts for difference seems to be in order to preserve model clauses that forbid the transfer of such contracts.76 It is doubtful whether the electricity supply clauses were intended to forbid the transfer of receivables, rather than of other benefits under the contract;77 and we shall probably never know whether any thought was given to the need to preserve these forms of BoA. But in any event, ABFA was not bothered about the exclusions and the exceptions were left in without further debate.

B.  National Security This exclusion, which again found its way into the final version of the Regulations, did at least get some discussion. It seems to leave room for argument that BoAs in contracts for goods or services that have some distant connection to Government activities ‘concern national security interests’. Rather than provide that the 74 The CLLS asked for an explanation (Note of 5 December 2017, para 11); but it used the exemption for contracts for difference as a springboard to try to get an exemption for other bodies holding cash for a public purpose and for project finance generally, and I do not recall that the rationale for the original exemptions was discussed. 75 Reg 40 of the Petroleum Licensing (Exploration and Production) (Landward Areas) Regulations 2014 (SI 2014/1686). In this context, a ‘petroleum licence’ appears to mean not just a licence to store petroleum but a licence granted by the Crown to a company wishing to search for and extract petroleum, issued under the Petroleum Act 1998. The licensee would also have to pay rent to the Secretary of State, but it seems highly unlikely that the licence will contain a BoA of the rent, or that the Government would want to assign its rights. 76 See cl 49 of the model clauses, available at https://www.gov.uk/government/uploads/system/ uploads/attachment_data/file/348142/Generic_CfD_TCs__29_August_2014.pdf, made under Energy Act 2013, s 11. 77 cf contracts for interests in land in section VII.B.

New Override of Bans on Assignment of Receivables  129 Secretary of State’s certificate to that effect is conclusive, it might have been better to limit the exemption to cases in which the Secretary of State had actually issued a certificate; but the Department of Business persisted with its draft, perhaps under pressure from other Departments.78 After another period of silence, explained in part by a flurry of activity after the Brexit referendum and the re-organisation of the Department (now the BEIS, as noted earlier), there was a meeting between the Department, ABFA and us as consultants. This was followed by a fresh draft of 12 April 2017, again seemingly circulated only to ABFA and the consultants; but the only changes were to develop the provisions on confidentiality clauses79 and to introduce provisions dealing with private international law issues.80 Finally, in late September 2017, draft Regulations were laid before Parliament. This meant that in terms of wider consultation there had been a gap of almost two years.

X.  Taking Our Eyes Off the Ball: Contracts Between the Same Parties In the meanwhile, an unfortunate error in drafting had crept in. It will be recalled that the 2015 Act empowers the overriding of a term that prohibits or imposes a condition, or other restriction, on the assignment by a party to the contract of the right to be paid under the contract or any other contract between the parties (emphasis added). The point was to prevent parties using a second contract to forbid assignment when that could not be done by a term in the receivables contract. The drafts of the Regulations put out for consultation in December 2014 and September 2015 reflected this by defining a ‘receivable’ as ‘a right to be paid any amount under a contract or under any other contract between the same parties’. Meanwhile the Regulations were not to affect ‘excluded contracts’, which included contracts for the prescribed financial services. This drafting gave rise to some puzzlement and to a concern that a term in a master or netting agreement that prevented the parties from assigning receivables arising under other contracts for financial services might be invalidated: the master or netting agreement might not be treated as a contract for financial services.81 78 I understand that the Ministry of Defence agreed a protocol with ABFA that BoAs would not be used in defence supply contracts; and in 2008 there was a protocol permitting assignment of receivables under procurement contracts in general (Information Note 07/08 16 June 2008, Assignment of Debts Arising under Public Contracts). It is not clear if this protocol is still in force: the Government website listing current procurement policies does not mention it as either in force or out of date. In contrast, the US has a special restrictive provision on the assignment of claims against the US Government: 31 US Code § 3727. 79 See section VII.D. 80 See section VIII.A. 81 CLLS comment 8-10-15, para 22, which concluded ‘We suggest that it should be made clearer that the ban on a term that prohibits or restricts the assignment of a receivable that arises under an excluded

130  Hugh Beale Louise Gullifer and I recommended clarifying the draft by moving the reference to ‘other contracts’ to a later regulation. But, whether by accident or misguided design I no longer recall, we omitted the words ‘between the same parties’. We suggested: Reg 1(2): ‘“receivable” is a right to be paid any amount, whether or not earned by performance, under a contract for the supply of goods, services or intangible assets other than an excluded contract.’ Reg 3(1): ‘A term of a contract is of no effect to the extent that it prohibits or imposes a condition or other restriction on the assignment of a receivable arising under the same or any other contract.’

Read literally, that would cover negative pledge clauses that prohibit the creation of security over, or the assignment of, receivables owed to the borrower. A negative pledge clause cannot prevent an effective security interest from being created or an effective assignment: to that extent it is merely a contractual term. But breach of the term would normally be an event of default under the loan agreement, which would give the lender the right to accelerate the loan and terminate the agreement. In practice, these rights are very important. We realised our mistake a few months later and told BEIS that the reference should be to ‘other contracts between the same parties’, but the point was not picked up by the draftsman, or possibly he thought the words were unnecessary because the power under the 2015 Act to make regulations is limited to overriding BoAs in the receivables contract or any other contract between the parties. So the next drafts also omitted this phrase. Unfortunately we failed to notice this, and when in September 2017 the Department finally laid draft regulations before Parliament, the relevant regulations read: 2.

A term in a contract has no effect to the extent that it— (a) prohibits the assignment of a receivable arising under that contract or any other contract …

This wording produces results that would be most unfortunate and that no one intended. I think that single error explains some of the subsequent hostility to the draft among practitioners. But it also emerged that there were problems that no one had raised before.

XI.  The Profession’s Response to the 2017 Draft The principal response from the profession was a paper by the CLLS FLC, now chaired by Dorothy Livingston. This was supported, without such detailed

contract is outside of scope even if the term (prohibiting the assignment) is not itself contained in the excluded contract.’

New Override of Bans on Assignment of Receivables  131 analysis, by quite a number of bodies. The paper complained that in the lull in public consultation between 2015 and 2017, there had been no opportunity to consider the substantial changes in drafting. Given the unfortunate change that I have just mentioned in section X, the complaint was not wholly without justification. The CLLS pointed out that the wording would result in negative pledge clauses in contracts between the supplier and third-party lenders being affected; arguably it also rendered the Regulations ultra vires. The Department grasped the point very quickly, and decided that it was not feasible to rely on ‘reading down’ the wording in the light of the limits imposed by the Act. So, as it is not possible to amend Regulations once the have been laid, they withdrew them altogether. A  meeting in November 2017 to discuss them was cancelled. A new draft was circulated before the next meeting, in which the missing words were inserted; and it was also provided that terms in an excluded contract would not be affected. This was done in case there was a financing contract (for example, for supply chain finance) between the same parties and this financing contract contained a clause to prevent the supplier from assigning its receivables further. It was not the intention to override the clause in this situation either. This was by no means the only objection raised by the CLLS. Some could disposed of quickly: for example, that as the Regulations did not state that they applied only to contracts made after the commencement date, they appeared to act retrospectively. There is a well-known presumption that legislation is not retrospective.82 Nonetheless, the Department thought it would do no harm to make the position absolutely clear, and also to give a short period for those affected to adjust, so it was provided that the Regulations apply only to contracts made after a fixed date after their commencement.83 Most of the other objections, in particular to the provisions aimed at confidentiality clauses, the application of the Regulations to contracts under which the supplier was a foreign business and set-off, I have dealt with already.

A.  Transitional Services Agreements The CLLS did raise one new point, which is interesting in its own right but also because it was the first of a small flood of similar objections that had not been raised at any previous stage of consultation. The CLLS’s point was that ‘Transitional Services Agreements’ (agreements under which the vendor of a business supplies services, eg IT and record keeping, to the business being sold after the sale until the purchaser has time to set up alternative arrangements) should be exempted, as ‘they are very much tailor-made to the situation’. In subsequent discussion it became clear that the relationship between the vendor and purchaser of the



82 See

House of Commons Library Standard Note SN/PC/06454. 31 December 2018: 2018 Regulations, reg 1(2).

83 Ultimately

132  Hugh Beale business was considered to be particularly close, so that the purchaser should not have to deal with anyone else. Being ‘tailor-made’ does not strike me as a particularly strong reason, and it is not clear to me why this relationship is closer than any other between supplier and customer. But it could be argued that these transactions are likely to involve businesses of some size, and possibly of fairly equal bargaining power. In any event, the Department seems to have taken the view that it would be easier to agree to an exclusion than to argue, perhaps because these transactions are likely to be relatively uncommon and also temporary. And so the first additional exemption, generalised to cover contracts for the purposes of the acquisition, disposal or transfer of an ownership interest in a firm (and stated in the agreement to be such, so that it would be easy for potential assignees to identify such contracts84) was included.85

B.  Commodity Contracts Some further exclusions were relatively technical and uncontroversial. Thus the International Swaps and Derivatives Association worried that commodity contracts that are structured as contracts for physical delivery could fall within the general scope of the Regulations and might not be excluded under financial services. There was no policy objection to excluding these, or derivatives and similar contracts, where they are traded on a regulated market or trading facility,86 and so they are also exempt.87

C.  Construction Contracts Not all the arguments from practitioners were accepted. For example, the CLLS Construction Law Committee argued that the override should not apply to BoAs in construction contracts. However, the arguments seemed to be based on two premises. One was that if the BoA were overridden, the employer would lose it rights of set-off. However, as we have seen, the only set-offs that would be affected are those that arise after the date of notification, under separate transactions. How many employers will have claims against the building contractor under separate, unrelated agreements? Or that arise only at a late stage? I would have thought, very few. The second premise was of dishonesty: that somehow, because the contractor might receive part payment from the receivables financier earlier than from the 84 See Note of meeting on 20 December 2017, para 13. 85 See 2018 Regulations, reg 4(i). 86 This proviso was included to avoid catching simple contracts for the future delivery of commodities, such as grain to a flour miller. 87 See 2018 Regulations, reg 4(j). The final version was proposed by Allen & Overy in January 2018.

New Override of Bans on Assignment of Receivables  133 employer, the funds might not be used to pay the sub-contractors and suppliers but be diverted, and that the contractor might at that stage declare insolvency in order to divest itself of its outstanding obligations. Why this risk is greater in the construction industry than in any other is not apparent; nor why obtaining the funds earlier makes the risk greater. I was perhaps not the only one to suspect that the real objection was simply to any change ‘in our backyard’. In any event, the Department was not persuaded and there is no exception for construction contracts. These issues were discussed in two meetings between the Department, members of various CLLS committees, UK Finance (into which ABFA had by then merged) and us as consultants; and most of the issues were, I thought, dealt with. But when the Department circulated a revised draft in January there was a barrage of points that were new, or which at least had not been insisted on before – in particular from lawyers in a single firm, Allen & Overy LLP, that appeared to be taking up the mantle of the CLLS.

D.  Projects and for Decommissioning Contracts It was argued that there must be exclusions for projects and for decommissioning contracts. It was said: Projects transactions are highly structured and rely on the integrity of the cashflow structure. The cashflow structure is achieved through the whole suite of project and financing documents, and relies on prohibitions on assignment of receivables. It is therefore vital that projects contracts benefit from an exclusion. While the financing documents for a project should fall within the exclusion for financial services contracts …, there are a wide range of other projects contracts (including, but by no means limited to, construction and operating contracts) which may be caught by the regulations and do not currently benefit from an exclusion.88

In what way BoAs are essential to projects contracts is not explained; I do not see any relevant difference between projects and construction contracts. Likewise, it was argued that decommissioning funds have to be ring-fenced; but I do not see why that means that any ability to assign or otherwise dispose of them would be at odds with the requirement for the Secretary of State to approve the relevant decommissioning programme. Possibly good reasons were given in one of the meetings that subsequently took place between Allen & Overy and the Department, which I was not able to attend; at any rate, these exclusions found their way into the final Regulations.89 But equally, it is possible the Department was simply unwilling to argue further, for fear that the firm would cause such trouble that the whole reform might have to be abandoned. 88 Comments from Allen & Overy LLP on draft regulations circulated on 12 January 2018, sent to BEIS and others on 24 January 2019, at 3. 89 2018 Regulations, reg 4(k) and (l).

134  Hugh Beale But I do think it is legitimate to ask why was this brought up only at this very late stage? The firm had been involved in the consultation back in 2015 but had not raised any of these points.

E.  ‘Why the Proposed UK Regulation Prohibiting Non-assignment Clauses is Bad’ It soon became apparent why the Department was suddenly receiving so many critical comments, particularly from lawyers at Allen & Overy. The firm’s Global Law Intelligence Unit had just circulated a report entitled Why the proposed UK regulation prohibiting non-assignment clauses is bad. The first version that I have seen is dated 10 January 2018; the version still on the unit’s website is dated 16 January 2018. The later version begins with a summary, of which I quote the opening and closing paragraphs: The Government is proposing to pass a regulation within the next week or so which will nullify clauses in contracts which prohibit or restrict assignments of receivables arising under the contract. It will also nullify confidentiality and other clauses which prevent an assignee from getting detailed information about the contract. So you will not be able to stop assignments or disclosures to any third party coming into your contract, such as debt factors or competitors anywhere. … We believe that the initial and subsequent consultations have been inadequate. If you wish to object … you should act immediately.

Some points made in the report were undoubtedly correct: for example, that it might be wise to limit the reforms to helping suppliers that are SMEs. That point had already been made a number of times and, as I go on to explain, in the light of the new information that was coming out in this ‘new wave’ of consultation, it is unfortunate that this approach was not accepted earlier. Other points, such as that the Regulations as laid before Parliament in 2017 might affect negative pledge clauses in financing agreements, also had been true before the Regulations had been withdrawn and amended. But that had taken place well before January 2018. Likewise, any doubts as to whether or how far set-off would be affected had been settled some time before. And some of the complaints, for example about the effect on confidentiality clauses, seem considerably exaggerated – particularly in the light of the ‘list of information’ approach that the Department had adopted before Christmas 2017.

F.  Small Businesses Again Whatever its flaws, there is no doubt that the Global Law Intelligence Unit report was effective to galvanise the author’s colleagues into action. So at a very late stage

New Override of Bans on Assignment of Receivables  135 a plea was made to exempt operating leases, on which see more in section XI.G. More importantly, the report made the Department think again about the question of whether the Regulations should apply only when the supplier is a small business. It will be remembered that the objections to this were not objections of principle. Rather, on the one hand we thought it would be very difficult to avoid arbitrary distinctions and to find criteria that are readily verifiable, and on the other we thought that virtually all BoAs were found in contracts in which the customer was in a position to dictate terms to the supplier. It was not until this very late stage that we were given examples of BoAs in contracts that had been freely agreed between parties that were most probably not small by any measure. So it was decided that if suitable criteria could be devised, we should probably accept a restriction to small businesses. To use the criterion of the simple size of the business, for example whether it meets the EU’s definition of an SME, at the time of the assignment, would not be practical for financiers – they might have to carry out lengthy and expensive searches to find out the answer on the exact date. Moreover, there will be some ‘special purpose vehicle’ companies that have few if any employees, but which will be controlled by large parties and which will be vested with very valuable assets. So the Law Commissions, when they proposed that small business contracts should be subject to controls over unfair terms, had supplemented a size test (in that case, nine or fewer employees) with one of the value of the contract: contracts over the value of £500,000 would be exempt.90 The CLLS helpfully suggested a similar approach here, using either the value of the contract or the value of the invoice. But, as to the contract value, UK Finance pointed out that they deal in invoices not contracts and do not have the capacity to evaluate the value of contracts – and many contracts are not of a fixed value. Nor is the amount of an invoice a reliable guide: clearly an invoice might be for one small item supplied under a multi-million pound contract. Fortunately, UK Finance came up with a solution to the problem of how the size of the supplier can readily be determined: do not rely on the size in real time, as the Law Commissions had proposed, but base the criterion on the supplier’s status at the date it last filed accounts and the different accounting requirements for large, medium and small companies under the revisions to the Companies Act 2006. Coupled with an exclusion of SPVs incurring liability of £10 million or more, that became the approach the final Regulations adopted. As I said earlier, I wish that we had considered limiting the reform to SME suppliers more seriously and had come up with the ‘last filed accounts’ approach much earlier. I think that it might have headed off much of the ‘special pleading’ for the exclusion of various types of contract; if it had been proposed to limit the reform to suppliers who are small companies within the Companies Act regimes, I very much doubt that the law firms would have bothered to seek 90 Law Commission and Scottish Law Commission, Unfair Terms in Contracts (Law Com No 292, Scot Law Com No 199, 2005) paras 5.40 and 5.54.

136  Hugh Beale exclusions for particular types of contract, any more than they did on the initial consultation.

G.  No Rolling Back of Exclusions Had the Department not kept the SME limit ‘up its sleeve’ for so long, we might also have been able to persuade the practitioners that many of the exclusions were not required after all and should be taken out of the draft. The Department, UK Finance and we as consultants all agreed that this would hopefully be the outcome; but the pressure persisted. Let me take by way of example the latest exclusion sought, for receivables payable under operating leases. It was argued that airlines, for example, will object if they find that they are having to pay not the original company from which they have leased an aircraft but a financier.91 Why is that more of a problem for airlines than for any other customer who owes receivables for goods or services is not clear to me. But what I really cannot see is that this exclusion is required if the lessor is a small business – a plant hire company, for example. Why on earth should such a company not benefit from the new regime? The only answers I can come up with are, once again, fear that the law firms, who have considerable influence, might try to block the whole reform – or simple exhaustion. UK Finance seem to have concluded that they would just have to live with it.

XII.  The Process of Reform The resulting Regulations are, to my mind, profoundly unsatisfactory. I am of course glad that we have secured an override of BoAs for most trade receivables owed by small and medium-sized companies. But the Regulations are long and complex – the very things that the CLLS complained of initially, only worse. I think all this raises issues about the law reform process and the role played by the CLLS and other practitioners. There were mistakes on all sides. On the part of the Department and the consultants, it was very unfortunate that the draft circulated for consultation in 2017 did not limit is application to contracts between the same parties, and exclude terms in contracts for financial services. I have also said that I regret that, once it emerged that BoAs are used even in contracts between the same large parties, we did not act more quickly to limit the reforms to SME suppliers.

91 Some of the arguments raised seem based on the old confusion between assigning the receivable and transferring the lessor’s obligations to a different company (which can only be achieved by novation). Others, such as that the lessor might assign the rentals to a competitor airline, seem wholly fanciful.

New Override of Bans on Assignment of Receivables  137 On the other side, while it was correct for the CLLS to claim in 2017 that there had been no consultation over the (largely unintentional) changes in drafting since the first consultation in 2015, it and the law firms who supported the CLLS response had not provided any detailed arguments about the wider use of BoAs. The author of the CLLS report is of course not to blame; he or she may well not have been aware of the extent of the supposed problem. But I just wonder whether his or her colleagues gave enough thought to the issue, until the last moment, when draft regulations are actually laid or are being circulated with a view to laying them in a time slot secured by the Department. Then there is a risk that the project will lose its ‘slot’ and may never see the light of day – so people tend to make changes that are uneasy compromises or not wholly thought through. That does not make for good quality law reform. Nor am I convinced that the exclusions are right in substance. Why are so many needed when they do not appear in the many other legal systems that have adopted an override of BoAs? As far as we were able to discover in the limited time available, the contracts that were excluded only at a late stage do not give rise to problems in the other jurisdictions. And it is worth remembering that overrides of BoAs are in force in the United States of America (including in New York, so often said to be the law to which international business will turn if we in any way impose controls over contracts governed by English law), Australia, Germany and France – to name just major trading nations. Had the issues been raise earlier, we could have investigated. But my sense is that actually the excluded contracts do not cause problems in practice, or in the few cases where there are problems, easy workarounds have been developed. I have suggested that some of the demands for exclusions are a form of nimbyism. Perhaps that is unfair. But I would like to float for discussion the idea that the vehement objections from the profession result from a quest for certainty that is not really about what will happen if there is a dispute, but about lawyers feeling that they must be able to give their clients assurances of the position, even if it is very unlikely that there will ever be a problem. We have seen the same issue arising in the work of the Secured Transactions Law Reform Group on floating charges. We have been told several times that the issue over the difficult borderline between a fixed charge and a floating charge is not about cases that administrators, liquidators and courts actually have to deal with. The problem is that transaction lawyers are asked to ‘sign off ’ on the nature of the charge, and are reluctant to do so in case they get it wrong or their workaround fails. The fact that the client would have to prove that the lawyer was negligent is apparently not sufficient comfort: the lawyers want cast-iron protection. If the same applies here, what better than a nice exclusion for the particular type of contract that you advise on? I think such an attitude, if it exists, is regrettable. It gives the impression of certainty when that in fact that cannot be guaranteed. It is of course true that English law is more ‘certain’ than some of its competitors, in the sense that on the face of it judges have very little power to interfere with what the parties have expressly agreed – at least if they have done so in clear words. But lawyers are all

138  Hugh Beale aware of the ‘hidden weapon’ of interpretation, let alone other ways in which a court may be able to ‘re-characterise’ a point – a very simple example is to say that a statement of fact is not a representation if it has contractual effect92 – so as to reach a different result. And this leaves aside the power of the Supreme Court to change the law fairly fundamentally, and with retrospective effect unless the circumstances are very unusual – a power that is exercised not infrequently, and not always predictably. Perhaps the House of Lords decision in Re Spectrum Plus93 could have been anticipated from immediately after the decision in Siebe Gorman & Co Ltd v Barclays Bank,94 but who would have anticipated the ‘legitimate interest test’ for agreed damages laid down in Cavendish Square Holding BV v Makdessi and ParkingEye Ltd v Beavis?95 And if the reply is that their clients demand cast-iron certainty, then I think that lawyers need to manage their clients’ expectations better – or the clients may in reality be mislead. I also regret that the Global Law Intelligence Unit’s report appeared in the form and in the place it did. I do not deny that the report made some very good points and performed a useful function, even though it was partly out of date. But I wonder whether such a vehement condemnation, made and kept available publicly, helps to maintain the international reputation of English law. A legal system is not just the courts, the law firms and the Bar. The law reform agencies and government departments that can and do introduce changes are also part of the system, and their attitude – or rather potential clients’ impression of it – may well influence foreign lawyers and businessmen. Even granted that we had some (mildly) dirty linen, I think it was unwise to wash it in such a public manner.

92 See AXA Sun Life Services Plc v Campbell Martin Ltd [2011] EWCA Civ 133, [2011] 2 Lloyd’s Rep 1 (Rix LJ), applying the approach of Ramsay J in BSkyB Ltd v HP Enterprise Services UK Ltd [2010] EWHC 86 (TCC), (2010) 129 Con LR 147. Rix LJ pointed out that the word ‘representations’ was ‘completely sandwiched between words of contractual import’ (at [80]) and that ‘supersede’ is ‘a word of agreement rather than exclusion’ (at [81]). More complex examples of true re-characterisation are the decisions on Romalpa clauses and on fixed and floating charges. 93 Re Spectrum Plus [2005] UKHL 41. 94 Siebe Gorman & Co Ltd v Barclays Bank [1979] 2 Lloyd’s Rep 142 (ChD). 95 Cavendish Square Holding BV v Makdessi and ParkingEye Ltd v Beavis [2015] UKSC 67, [2016] AC 1172.

8 The Boundaries of a Borrower’s Freedom to Act: Negative Covenants in Loan Agreements LOUISE GULLIFER AND GRAHAM PENN

I. Introduction A simplistic view of contractual terms is as follows. A positive obligation is included in a contract because the obligee wants the obligor to perform, that is, to do what it promises to do. If the obligor fails to perform, the obligee is given remedies that seek to put it in the position it would have been had the contract been performed, including (sometimes) the right to force the obligor to perform. A negative obligation is included for the same reason: the obligee does not want the obligor to do what it has promised not to do. If it does it, the obligee is given remedies to put it in the position it would have been had the obligor not done that thing, including (sometimes) the right to stop the obligor doing that thing in the first place (if the obligee has notice). While crafting remedies relating to negative obligations raises more difficulties than remedies relating to positive obligations, the underlying premise is the same, namely, that the content of the term represents something that the obligee wants, either to happen or not to happen. We know from the growing jurisprudence on ‘negotiating damages’ that, at least in relation to negative obligations, there are exceptions to this world view, and that a negative obligation can be seen as creating an opportunity to negotiate a price payable to the obligee for its agreement to waive the breach rather than as embodying a fixed state of affairs the obligee does not want to happen in any circumstances. Even further along this spectrum are situations where negative obligations are included in contracts not for the purpose of opening up negotiations for a quid pro quo payment, but in order to enable the obligee to renegotiate the contract more broadly, or even to negotiate other outcomes that are, at least in law, unrelated to the contract terms.1 1 Although when included in commercial loan agreements, they are typically linked to the commercial objectives that underpin the inclusion of the relevant negative covenant; see the discussion in section II.

140  Louise Gullifer and Graham Penn This chapter argues that negative covenants in commercial loan agreements fall somewhere near the opposite end of the spectrum from the simplistic view set out in the first paragraph. They are part of a complex web of contractual techniques used by lenders to both manage and mitigate risk by imposing a range of restrictions on the business of the borrower. The loan agreements considered here are those entered into by business borrowers, most of whom will be companies. The control exerted by lenders is therefore a form of corporate governance, at least when the borrower is solvent. On the insolvency of the borrower, the lender will also want control rights, but these are likely to be obtained using different techniques, such as reliance on breaches of positive covenants or on events of default to give it a seat at the negotiating table, or reliance on its position as a creditor, for example a qualifying floating charge holder, to effect the appointment of an administrator or via its holding of share pledge security, which enables it to replace the directors. In making this argument, the chapter proceeds as follows. Section II examines briefly the risks a lender faces and the basic protection given by terms in a loan agreement. Section III outlines a map of loan terms, including the interaction between loan terms and contractual remedies. Section IV fleshes out the main argument in relation to negative covenants: that these are not primarily there to prevent activities by the borrower but to align the duties of the directors in deciding on the activities of the borrower with consideration of the lenders’ interests. Section V examines some specific negative covenants and, in particular, considers the need to strike a balance between strict restrictions and the freedom of the borrower to conduct its business. Section VI concludes. This chapter proceeds on the basis that there is a single lender, although if the quantum of the loan is significant it is likely to be syndicated, so that there will be many lenders. The principles are the same in relation to a syndicated loan, where the relevant covenants will be monitored and enforced by an agent bank acting on behalf of the lenders, and, to some extent, a bond issue, by a bond trustee, although the presence of multiple lenders or bondholders makes the processes more complicated.2

II.  The Purpose of Contractual Terms in a Loan Agreement A lender entering into a loan agreement faces a number of risks. By far the most significant is credit risk, namely, that the borrower will not be able to pay interest 2 See L Gullifer and J Payne, Corporate Finance Law: Principles and Policy, 2nd edn (Hart Publishing 2015) ch 8. Historically there was considerable difference in the type and scope of covenants contained in loan agreements and bond issues, but those differences have lessened in recent years as alternate credit providers (non-regulated banks) have increased their presence as debt providers in the international debt markets (in respect of both loans and bonds). The decision as to whether a loan or bond structure is used is increasingly influenced by the identity of the lenders and the tax and regulatory treatment that will apply to the proposed debt instrument.

Negative Covenants in Loan Agreements  141 or repay the capital lent. Other risks include that the borrower will repay early, or will not roll the loan over on maturity, thus depriving the lender of a source of profit, and the risk that the loan will become unprofitable compared to opportunities to deploy the lender’s capital elsewhere (for example, in relation to a fixed rate loan, if interest rates rise). Some terms will address these other risks, but most are designed to protect the lender against credit risk. Of course, all loans carry some risk, and if the market is functioning properly, risk and reward should correlate so that the margin component of the interest rate agreed reflects the risk/reward as perceived at the commencement of the loan transaction.3 One function of the loan covenants, then, is to enable the lender to maintain the position where the reward reflects the risk, by enabling adjustments to be made if the risk changes.4 They also play an important role in enabling the lender to maintain the margin component of the interest rate payable by the borrower for the term of the loan. This is typical in term loan agreements and is predicated on an assumption made by the lender at the outset that the borrower’s credit standing will be maintained, or at least not worsen, throughout the term of the loan.5 This perception of risk, however, includes the protection afforded to the lender by the terms of the agreement. In theory, at least, the more protection is given by the terms, the less risk the lender has and therefore the lower the reward. In practice, however, the correlation is unlikely to be exact. This is partly because of the difficulties of assessing risk, and partly because of the way in which the protection given by the terms will work in practice. As already indicated, it is also typical in the loan market for the risk/reward and associated contractual protections to be fixed for the tenor of the loan at the commencement of its availability, even in the case of a medium- or long-term facility. The restrictions included cannot operate purely to prevent the borrower from doing anything risky or from changing anything in relation to its business. This would have the effect of stopping the borrower from operating in the ordinary course of business, in which case the borrower would not be able to pay the interest or repay the capital. Instead, the terms have to achieve a balance between preventing behaviour that is truly harmful to the lender, and permitting the borrower to operate its business without undue constraint. This is

3 Borrowers typically pay interest at a margin above LIBOR, and the margin reflects the lender’s assessment of the borrower’s credit standing. 4 Typically in the current market, the margin component of the interest rate is agreed at the outset, and the adjustments that occur as a result of the bargaining discussed in this chapter arising from the existence of negative covenants do not give rise to an automatic change (by way of increase or reduction) in the amount of the margin. Any such adjustments are typically made following negotiations between the parties. This point is discussed in more detail in section IV.D. One exception is in the leveraged loan market, in which it is not uncommon to include deleveraging targets, which, if met, will trigger an automatic reduction in the margin. 5 As already mentioned in n 4, in the leveraged loan (‘Lev Fin’) market it is not uncommon for the margin to be reduced in circumstances where the risk profile is reduced as the borrower achieves certain deleveraging targets included in the agreements; however, it is very unusual for the margin to be increased in the event the risk profile of the borrower worsens by reference to the covenants typically included in a loan agreement.

142  Louise Gullifer and Graham Penn achieved in various ways, an important ingredient of which is a dialogue between the lender and borrower opened by various trigger points.

III.  Terms of a Typical Loan Agreement In order to assess the part played by negative covenants in this ‘risk-controlling’ exercise, it is necessary to map the typical terms of a loan agreement. While there is, of course, ample scope for negotiation between the parties, and market variations, the existence of standard-form contracts, such as those produced by the Loan Market Association, by professional support organisations and by law firms, means that there is considerable commonality in the kinds of terms included in such agreements. These agreements are long and complicated, and it is not the purpose of this section to describe all the terms in detail. Instead, it will draw a conceptual map of the main terms, with a view to positioning the negative covenants within this framework.

A.  Commercial Terms The first type of terms are commercial terms of the agreement, and are likely to have been negotiated in advance. They create positive obligations. They will, of course, depend on the type of loan made, for example whether it is a term loan or a revolving facility, and on market conditions. The lender is, typically, obliged to make the loan by advancing money on an agreed date or on demand, either in agreed amounts or up to a cap. The lender’s obligation to lend is typically conditional upon the borrower’s satisfying certain conditions at the time of drawdown, which include satisfaction of the covenants (typically by means of a ‘no default’ confirmation, which is linked to ongoing compliance with the covenants). The borrower is obliged to pay interest at the agreed rate,6 and to repay the loan at the time and in the manner agreed. The borrower may also be obliged to pay fees and indemnify expenses.

B.  Terms Relating to the Provision of Information A lender can only assess risk if it has the necessary information. It will perform due diligence before taking the final decision to lend, and at that stage it can ask for what information it wants, on the basis that it will not lend unless the information

6 As already indicated, this includes the margin that reflects the lender’s assessment of the borrower’s credit standing and is typically fixed for the tenor of the loan.

Negative Covenants in Loan Agreements  143 is provided.7 The decision to lend is made, and the ‘price’8 of the loan is set, on the basis of this information, which is then ‘fixed’ by the use of conditions precedent to the drawdown of the loan. If a document evidencing the information has not been provided, or a representation made by the borrower at the time the loan agreement was entered into becomes untrue, the lender is not obliged to advance funds.9 The lender also needs information to be provided throughout the life of the loan. Therefore, the contract will contain positive obligations on the borrower to provide information periodically. The lender does want these obligations to be performed, since it wants the result of performance, but only in a utilitarian way: the information it gets will enable it to monitor the company’s activities with a view to providing an early warning of potential problems, and also activating a trigger point for dialogue if it considers this necessary.

C.  Financial Covenants The financial covenants are the first type of actual triggers, the others being the negative covenants and the events of default. Although there is considerable variance in practice, the financial covenants are usually couched as positive obligations to maintain certain financial targets, such as a minimum net worth, a gearing ratio (borrowing expressed as a ratio of net worth) or an interest cover ratio (EBITDA10 to finance charges). This chapter is focused on negative covenants and not financial covenants.

D.  Negative Covenants The negative covenants, discussed in more detail in section V, are couched as obligations on the borrower not to do certain things. Typically, these will include the granting of security to anyone other than the lender, the incurring of debt to any other lender, the disposal of assets, the return of capital to shareholders, either in the form of dividends or otherwise, the entering into of any merger or other form of amalgamation, and any substantial change to the general nature of the 7 Of course, this depends on the market and also on the cost, which cannot be disproportionate to the amount at stake. It is common for the loan agreement to include extensive representations and warranties covering many of the issues a lender would expect to cover in its due diligence process. To the extent due diligence has been undertaken by the lender, that will typically be excluded from the representations and warranties given by the borrower. 8 This may include not only the interest rate, but also other benefits for the lender and the strictness of the terms of the contract. 9 Confirmation that the representations and warranties continue to be true is typically included in the drawdown notice that triggers the obligation to lend. See P Rawlings, ‘Avoiding the Obligation to Lend’ [2012] Journal of Business Law 89, 97. cf Bryan Court Limited v National Westminster Bank plc [2012] EWHC 2035 (QB). 10 Earnings Before Interest, Tax, Depreciation and Amortisation.

144  Louise Gullifer and Graham Penn business of the borrower. It will be clear from the nature of these matters that they cannot be completely prohibited or the borrower would not be able to function. As already mentioned, there will be exceptions, and these are likely to be quite heavily negotiated.

E.  Events of Default Breaches of any of the obligations mentioned will typically trigger an event of default under the loan agreement, which will also include other events of default that are not breaches of obligations but factual states. Examples of these include a change of control of the borrower,11 the enforcement of any security granted by the borrower, the execution of a judgment against the borrower, the commencement of insolvency proceedings in relation to the borrower, the borrower’s insolvency,12 a representation given by the borrower proving to be incorrect, cross-default (default by the borrower in relation to any other credit agreement) and (as a general ‘safety net’13) a material adverse change. The loan agreement will provide for ‘self-help’ remedies for the lender if an event of default (breach or otherwise) occurs. The lender will be entitled to accelerate repayment of the loan; cancel its lending commitment so that it is no longer obliged to make further advances; and/or declare that the loan is payable on demand. The effect of exercise of these remedies on the borrower is likely to be very severe. The lender has a choice whether to exercise these remedies or not.14 Thus, once an event of default has occurred, the lender is in a position to call for renegotiation of the loan agreement and seek to impose any condition it wishes on the borrower as the price of continuation of the loan (including an increase in the cost of borrowing), the only constraints being bargaining power, market

11 Although this is sometimes drafted as a trigger for mandatory pre-payment, to avoid triggering cross-default clauses in other credit agreements. 12 This is typically defined by using the test in s 123 of the Insolvency Act 1986. 13 See R Youard, ‘Default in International Loan Agreements: Part 2’ [1986] Journal of International Banking Law 378, 390. 14 The choice is probably an absolute one, without an implied term restricting the decision of the lender: see Sucden Financial Limited v Fluxo-Cane Overseas Limited, Manoel Fernando Garcia [2010] EWHC 2133 (Comm), [2010] 2 CLC 216 [50]; UBS AG v Rose Capital Ventures Ltd [2018] EWHC 2137 (Ch) [50]–[56]. Although see R Hooley, ‘Controlling Contractual Discretion’ (2013) 72 CLJ 65, who argues that the discretion is probably bounded by a requirement of subjective honesty. The ability to convert a term loan into an ‘on demand’ loan may temporarily ease some of the more adverse implications for the borrower of the loan’s being immediately accelerated, and give time for an appropriate standstill agreement to be put in place to give the lender and borrower ‘breathing space’ to renegotiate the terms of the loan. Absent such an agreement, the directors of the borrower may conclude the borrower is unable to continue as a going concern, and this may force them to file for insolvency. For discussion of controls of discretion generally, see M Raczynska, ‘Good Faiths and Contract Terms’, ch 5 in this volume; and PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6 in this volume.

Negative Covenants in Loan Agreements  145 pressure and concern for reputation. The ramifications of this conclusion will now be explored in relation to negative covenants.

IV.  How Do Negative Covenants Work? A.  General Remedies for Breach of Contract On the simplistic view set out in section III.D, negative covenants would be included in loan agreements in order to prevent the borrower from doing certain things that would adversely affect the borrower’s risk profile, and which, therefore, the lender would not want the borrower to do. On this view, the remedies should be either an ex ante remedy, that is, an order to stop the borrower doing that thing, such as an injunction, or an ex post remedy, that is, a remedy that put the lender in the position as though the breach had not occurred. However, were these the only remedies, the negative covenants would be relatively useless to lenders in most situations. An ex ante remedy will only be of use if the lender has sufficient notice of the threatened breach. An ex post remedy is unlikely to be of any use in the absence of either a proprietary right or a right against a third party who benefits from the borrower’s breach. First, it is unlikely to give the lender any more than the rights it has on breach of the positive covenant to repay the loan via a triggering of the events of default and, second, the only time the lender will have suffered any loss is when the borrower is insolvent, and since a lender with only a personal right would rank as an unsecured creditor, it would be unlikely to recover much if anything. The lender may, of course, also have a security interest in the borrower’s assets, which may well enable it to recover on the borrower’s insolvency, but this will be in addition to, and separate from, the covenants in the loan agreement. In some circumstances, the lender may also have bargained-for rights against third parties, such as under a guarantee. It is also possible that the company itself may be able to claw back payments made to shareholders in breach of the maintenance of capital rules in the Companies Act 2006,15 or that a liquidator or an administrator might be able to claw back a disposition at an undervalue,16 or proceed against directors for wrongful trading,17 but recoveries from such actions will go to the borrower and therefore be shared pari passu among all its unsecured creditors, rather than to the particular creditor who has the benefit of the negative covenant. There is also the possibility of suing a third party who has benefitted from the borrower’s breach on the basis of a tort claim for inducing breach of contract.18 15 Companies Act 2006, s 847. 16 Insolvency Act 1986, s 238 or s 423. 17 ibid, s 214. 18 The requisite elements of the tort will, however, be onerous for the lender to prove. See the review of the tort conducted by Lord Hoffmann in OBG Ltd v Allan; Douglas v Hello! Ltd; Mainstream Properties Ltd v Young [2007] UKHL 21, [2008] 1 AC 1.

146  Louise Gullifer and Graham Penn

B.  Liability of Directors for Inducing Breach of Contract As discussed in section IV.D, the negative covenants in loan agreements are normally breached only as a matter of conscious and deliberate decision, rather than by accident or even carelessness. The relevant decision will have been taken by the directors. The duties of the directors vis-à-vis the borrower are also considered in section IV.D, but it is relevant to consider at this stage whether the lender could sue a director for inducing breach of contract by the borrower. The question of when a director of a company can be liable for inducing the company to breach a contract with a third party has recently been examined in a context entirely different from the present one. In Antuzis v DJ Houghton Catching Services Ltd,19 two directors caused the company to breach its employment contracts with the claimants by causing them to work extremely long hours, by paying less than the statutory minimum for agricultural workers and, in many cases, by not paying them their wages at all. Lane J examined the case law on the liability of directors for inducing the company to breach a contract, which, in English law, is sparse and is based on the early-twentieth-century decision of Said v Butt.20 In that case, McCardie J said, obiter, that a director (or any agent or servant) who caused a company to breach a contract would not be liable for inducing breach of contract if he was acting bona fide within the scope of his authority. Despite some misgivings, this rule was accepted by the Court of Appeal in Welsh Development Agency v Export Finance Co Ltd21 and by Waller J in Ridgeway Maritime Inc v Beulah Wings Ltd (‘The Leon’).22 The main plank of the reasoning in The Leon is that, in doing the act that breaches the contract, the director is the alter ego of the company, and making him liable would undermine the whole doctrine of limited liability of companies.23 Lane J went on to discuss a recent Singaporean case,24 where Stephen Chong JA considered the Commonwealth jurisprudence25 as well as the English cases. The Singaporean judge’s conclusion was that the Said v Butt principle ‘should be interpreted to exempt directors from personal liability for the contractual breaches of their company … if their acts, in their capacity as directors, are not in themselves in breach of any fiduciary or other personal legal duties owed to the company’.26 Considering this conclusion, together with the views of Waller J in The Leon, Lane J took the view that ‘it is the officer’s conduct and intention in relation to his duties towards the company – not towards the third 19 Antuzis v DJ Houghton Catching Services Ltd [2019] EWHC 843 (QB). 20 Said v Butt [1920] 3 KB 497. 21 Welsh Development Agency v Export Finance Co Ltd [1992] BCC 270, 289–90, 296. 22 Ridgeway Maritime Inc v Beulah Wings Ltd (‘The Leon’) [1991] 2 Lloyd’s Rep 611, 624–25. 23 ibid 624. 24 Arthaputra v St Microelectronics Asia Pacific Pte Ltd [2018] SGCA 17. 25 O’Brien v Dawson (1942) 66 CLR 18 (High Court of Australia) 34; Imperial Oil Ltd v C&G Holdings Ltd (1989) 62 DLR (4th) 261, 266 (Newfoundland Court of Appeal); Knights Capital Group Ltd v Bajada and Associates Pty Ltd [2016] WASC 69 (Supreme Court of Western Australia). 26 Arthaputra v St Microelectronics Asia Pacific Pte Ltd (n 24) [62].

Negative Covenants in Loan Agreements  147 party – that provide the focus of the “bona fide” enquiry to be undertaken pursuant to the rule in Said v Butt’.27 In the context of a breach by a corporate borrower of a negative covenant in a loan agreement, this discussion not only makes it clear that it is very unlikely that the lender would have a cause of action against the directors of the borrower, but it also establishes that any possible liability is aligned with the breach by the directors of their duties to the borrower. This conclusion, therefore, informs and is consistent with the discussion in section IV.D about the way in which negative covenants align the directors’ decisions with the interests of the lender.

C.  Acceleration and Cancellation The discussion in section IV.B, though, does not really tell the full story, or even a significant part of the story. For a start, it ignores the self-help remedies discussed in section III.E. It could, therefore, be argued that negative covenants are there to provide lenders with a ‘get-out clause’. If the borrower does the prohibited thing, this affects the lender’s risk, and so the lender has the opportunity to cancel any outstanding obligation; to lend more money; demand immediate repayment of the amount outstanding; and/or convert the loan to an ‘on-demand’ facility.28 The lender’s decision as to which, if any, of these remedies it exercises29 will depend on a number of factors, including weighing the benefits of escaping from the relationship, which will depend on how seriously it perceives the breach has affected its risk, against the detriment of losing the profit from the loan. While this analysis seems more plausible, in that, at least, the lender obtains some appreciable benefit from the existence of the negative covenant, it is still very unsatisfactory. First, the choice of the lender whether to exercise any of the remedies and, if so, which ones is not merely dependent on the cost–benefit analysis suggested in the previous paragraph. Acceleration and cancellation of the obligation to lend may well deprive the borrower of its primary source of capital, and, depending on the market, it may not be able to replace this by a loan from another lender within the time period necessary for it to remain solvent. Moreover, if the borrower has other debt, cross-default clauses may be triggered, so that even if the original lender chooses not to accelerate and/or cancel, another creditor may do so, thus driving the borrower into insolvency.30 On insolvency, the lender is

27 Said v Butt (n 20) [114]. 28 This will include all principal and interest, plus in most cases additional amounts – break costs and, in the case of fixed rate loans, a ‘make-whole’ amount. As noted previously, the lender will also typically have the ability to convert the loan to one payable ‘on demand’. 29 It will typically be able to exercise them all. 30 This is why standstill agreements are typically entered into in such circumstances, to give the lender and borrower ‘breathing space’ to work out the optimal way forward; however, some crossdefault clauses will still be triggered notwithstanding their entering into a standstill.

148  Louise Gullifer and Graham Penn unlikely to recover the accelerated loan anyway, so the benefits of its exercising its contractual rights will be bleak, unless the lender is in the fortunate position of not having advanced much before the breach, in which case cancelling its outstanding obligation to lend does make sense. Second, if the breach of the negative covenant is serious enough for the lender to decide acceleration is the appropriate course, the breach may well have a serious effect on the borrower’s ability to repay in full the outstanding amount, for example because the borrower has disposed of assets or returned capital to shareholders that cannot be recovered by the lender, as explained in section IV.A. In the light of this discussion, the next subsection explores an alternative explanation of the significance and purpose of negative covenants in loan agreements.

D.  Duties of Directors: Aligning Interests The classic view of negative covenants is that they are focused on risks to lenders that arise from the ‘agency’ problem between the shareholders and the creditors. The shareholders benefit from a rise in the value of the company, and so they are likely to favour riskier projects, while the lender’s principal focus will be repayment, so it will favour low-risk projects that maintain the status quo.31 The company, though, is not run by the shareholders: the operational decisions are made by the directors. The ‘agency problem’ view relies on the alignment of the actions of the directors of the company with the interests of shareholders, and assumes that, unconstrained by any other considerations, the directors will cause the company to pursue the risky activities favoured by the shareholders. The creditors, at least adjusting creditors, will therefore protect themselves by contractually restricting the activities of the company so that the status quo can be maintained.32 One important point to emphasise is that the discussion in this chapter is concerned with contractual terms that restrict activities33 which are the result of conscious decisions, namely, negative covenants. The argument in the previous paragraph needs some unpicking. We should first consider the duties of directors. Under section 172(1) of the Companies Act 2006, a director is under a duty to act ‘in the way he considers, in good faith, would be most likely to promote the success of the company for the benefit of its members as a whole’, taking into account certain matters including the interests of certain stakeholders mentioned in that subsection. Creditors are not mentioned

31 MC Jensen and WH Meckling, ‘Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure’ (1976) 3 Journal of Financial Economics 305. 32 Protection is also given by financial covenants, which can operate in a similar way to negative covenants. 33 Such as disposition of assets, return of capital to shareholders and additional borrowing. Breaches of financial covenants can be the result of deliberate activity, but also can result from outside events.

Negative Covenants in Loan Agreements  149 until section 172(3), which provides that the duty ‘has effect subject to any enactment or rule of law requiring directors, in certain circumstances, to consider or act in the interests of creditors of the company’. That requirement does not arise in statute or common law until a point when the borrower is in the vicinity of insolvency,34 at which time the duties of the directors change significantly and the duty to the creditors (all creditors, not a particular lender) overrides all other duties. The duty is to minimise the loss to creditors and certainly not make the creditors’ situation worse; the steps that need to be taken will obviously vary on a case-by-case basis. What a lender wants to achieve by the inclusion of negative covenants in a loan agreement, then, is a redressing of the focus of the directors’ decisions, so that they no longer favour the shareholders but take the interests of that lender into account.35 How is this achieved? A contractual term in a loan agreement is binding on the borrower, but it is not contractually binding on the directors who are not parties to that agreement. Moreover, the directors are obliged, by company law, to act in the interests of the shareholders. The structure of the loan agreement, therefore, has to be such that it is in the interests of the shareholders that the directors cause the borrower to comply with the negative covenants. One might say that it is the duty of the directors at all times to ensure that the borrower complies with its legal obligations. This raises an interesting question about the nature of contractual obligations.36 One view is that they are normative, in the sense that breach is a wrong and that the obligee has a right to performance (’the rights-based view’). Another view is that the obligor has a choice whether to perform or to pay damages (‘the efficiency-based view’).37 This latter view supports the argument that there can be an efficient breach of a contract, in that even if the breaching party pays damages to the innocent party, everyone affected by the breach is better off, thus promoting efficiency. At least on this view, an efficient breach would be in the best interests of the obligor, that is, the borrower, and therefore (arguably) the directors would be under a duty to bring it about. The case law on the potential liability of directors for inducing breach of contract, discussed in section IV.B, also supports the view that causing the borrower to breach a contract can constitute performance, rather than breach, of the directors’ duties to the company. As already discussed, in the context of a breach of a negative covenant in a loan agreement, damages will not typically be a helpful remedy to the lender, and

34 The point has recently been defined by Lord Justice David Richards as ‘when the directors know or should know that the company is or is likely to become insolvent’: see BTI 2014 LLC v Sequana SA [2019] EWCA Civ 112, [2019] 2 All ER 784 [220]. 35 See P Wood, ‘Bondholders and Banks – Why the Difference in Protections?’ (2011) 6 Capital Markets Law Journal 188, 190: ‘The covenants are the obligations of management to ensure that the business of the borrower is run with the interests of the lender in mind’. 36 S Smith, Contract Theory (OUP 2004). 37 Expressed by Oliver Wendell Holmes as ‘the duty to keep a promise at common law means a prediction that you must pay damages if you do not keep it – and nothing else’: O Wendell Holmes, ‘The Path of Law’ (1897) 10 Harvard Law Review 457, 462.

150  Louise Gullifer and Graham Penn so the efficiency of a breach would relate to the choice whether to perform or to take the consequences of breach. It is therefore important to consider what, in ­reality, the consequences of a breach of a negative covenant would be. One possibility is that the lender would immediately accelerate the loan and cancel its obligation to make further advances. The practical consequences of this for the borrower would depend greatly on the circumstances. If the borrower were not in good financial shape, or the market for loans was tight so that replacement debt finance would be hard to find, the directors would know that acceleration and cancellation would be likely to be disastrous, perhaps even driving the borrower into insolvency.38 In these circumstances, it would clearly not be in the borrower’s (or the shareholders’) best interests to breach, so that the contractual obligation and the directors’ duties are aligned. There is one important caveat to this. As discussed previously, the lender will also be well aware of the serious implications that may follow in the event it accelerates the loan and/or cancels its obligation to lend. It will not want to drive the borrower into insolvency, certainly not until it has carefully explored all options, and in order to give the parties an appropriate ‘breathing space’ to consider those options, and also to protect the directors,39 it will typically enter into a standstill agreement, effectively suspending its right to take action in respect of the breach until those options have been explored and the loan potentially restructured. If the directors know, or strongly suspect, that the lender will not immediately seek to exercise its remedies, they could take the view that breach would not necessarily have the disastrous consequences previously suggested, and that it could, after all, be in the best interests of the borrower and the shareholders.40 The negative consequences to a borrower of breach are magnified if, as is usual, each of its debt finance agreements contains cross-default clauses. A breach under one loan agreement typically gives all lenders (and bondholders) the right to accelerate and terminate, whether or not the lender in respect of the loan that has been breached decides to exercise its remedies or not. The directors then have to consider not only how the counterparty will react, but also how all lenders will. If, however, the borrower was in good financial shape, and the market for loans was buoyant, the borrower could procure replacement debt finance easily, maybe even on better terms than the existing loan. In these circumstances, a breach could be in the best interests of the borrower, although there could be more efficient 38 Insolvency not only affects the creditors of the company; if a director allows the company to trade when insolvent, he or she can be liable for wrongful trading under the Insolvency Act 1986, s 214. In practice, concern about this potential liability often has an even greater effect in concentrating the minds of directors than the requirements of directors’ duties. Lenders, in their turn, are aware that directors might put the company into insolvency proceedings earlier than otherwise in order to avoid liability for wrongful trading: this concern can sometimes be used as a bargaining chip by directors when negotiating with lenders in relation to a breach or potential breach. 39 For example, from accusations of wrongful trading. 40 But see further below.

Negative Covenants in Loan Agreements  151 ways for the borrower to terminate the loan agreement, depending on whether the agreement contains appropriate prepayment provision.41 However, if this were not the case, the lender would be unlikely to want to trigger early repayment, since it would not want to lose its profitable loan asset. If all of this were known to the directors, arguably their duty would be to take the decision whether to breach on its merits: would the benefits of the breach outweigh the possible adverse consequences?42 It would be a question of predicting what the lender would do in these particular circumstances. There is a ‘halfway’ house that can change the dynamics somewhat. If the lender is still to make further advances (for example, in a revolving facility), it also has the option of refusing to allow further drawdowns.43 The legal basis for this will typically be a provision in the loan agreement that all advances are conditional on no event of default having occurred and being continuing at the time of the further drawdown. The consequences of not being able to draw down further advances is likely to be very serious to the borrower, but it does have the advantage of enabling the lender to decide, at least initially, to suspend the obligation to lend until the event of default is remedied (in which case the lender will not lose the benefit of its initial bargain) and/or to use the event of default as justification for renegotiating the terms of the loan, including pricing. The option, available to the lender, of cancelling the obligation to lend will put the lender in a strong negotiating ­position.44 Consequently, the borrower also does not necessarily lose the benefit of its initial bargain, although the terms may change. There are a number of refinements that need to be made to this analysis to capture properly the way in which negative covenants in a loan agreement shape the directors’ decision making. The analysis in the preceding paragraphs does not include a common reaction by a lender to a breach: it will not immediately accelerate and cancel, nor will it immediately decide not to do this. Instead, it will use the fact that it has the right to do this as a reason for opening negotiations with the borrower. The initial aim of these negotiations will be to restructure the loan on terms acceptable to the lender, and that will include a ‘price’ for waiving the breach and, therefore, not accelerating and/or cancelling the obligation to lend. This ‘price’ can be either a variation of the loan agreement, such as an increase in the margin (to reflect the increased risk of lending to the borrower) and/or stricter financial covenants (to reflect the increased risk resulting from the circumstances giving rise to the breach), or the obtaining of some other outcome, such as the grant of security, a change of direction of the business of the company or, very

41 In certain loans early prepayment will also trigger a ‘make-whole’ payment obligation, particularly in the case of fixed rate loans. The inclusion of such provisions would have been a matter of negotiation at the time the loan agreement was entered into. 42 Again, cross-default clauses would be relevant here. 43 DG Baird and RK Rasmussen, ‘Private Debt and the Missing Lever of Corporate Governance’ (2006) 154 University of Pennsylvania Law Review 1209. 44 The lender will have the right to exercise these remedies individually and/or cumulatively.

152  Louise Gullifer and Graham Penn occasionally, the replacement of directors.45 As long as the threat to accelerate and/ or cancel is real, the lender is in a strong bargaining position in these negotiations, and is likely to achieve the desired outcome. Although there is always a possibility that a lender will use its position to improve the bargain it struck at the time of the loan agreement, at least in theory the purpose of the negotiation will be to restore that bargain, in that the outcome negotiated for should reflect the increased risk that comes from the breach of covenant.46 As long as that is the case, it is difficult to see how agreeing to such an outcome as a result of negotiations could be inconsistent with the directors’ duties to the shareholders. This, then, could be seen as a method whereby the negative covenant has the effect of aligning the directors’ duties with the lender’s interests. There are still two drawbacks to this analysis of the utility of debt covenants. First, in order for the lender to react to the breach it has to know about it, which entails close monitoring of the company’s activities. As already mentioned, the loan agreement is likely to include positive obligations to provide information to the lender, but the information is almost always likely to be historical, and therefore will not typically enable the lender to react immediately to a breach. The lender may have other means of monitoring; for example, if it is the borrower’s relationship bank it will be able to monitor cash flow activity in the borrower’s bank accounts. However, these means will not always enable a lender to know about a breach of a negative covenant (such as a negative pledge clause), and a lender may not want to expend the costs required in such active monitoring.47 Second, and very importantly, most borrowers and their directors do not want to be in breach of a negative covenant. Apart from leaving the borrower vulnerable to harsh consequences or negotiations, it will also damage the borrower’s relationship with the lender, and make it less likely that the loan will be rolled over at maturity.48 Therefore, if the directors consider that, in the best interests of the shareholders, an action should be taken that will be a breach of a negative covenant, they are likely to approach the lender for a waiver49 before they 45 Baird and Rasmussen, ‘Private Debt and the Missing Lever’ (n 43); F Tung, ‘Leverage in the Board Room: The Unsung Influence of Private Lenders in Corporate Governance’ [2009] UCLA Law Review 115, 157. G Nini, D Smith and A Sufi, ‘Creditor Control Rights, Corporate Governance, and Firm Value’ (2012) 15 Journal of Financial Studies 1713 provide empirical evidence for this phenomenon (see esp 1716), although they are not able to document precisely how the change of CEO occurs (ibid 1758). 46 This approach can be contrasted with the approach typically taken in the bond market, where the ‘price’ will include an amount payable by the borrower to compensate the lender for changes in the prevailing market price of the debt. This is often paid by way of a consent solicitation fee. 47 If the loan is syndicated and the relationship bank is also appointed as the Agent Bank, it will be concerned about the potential liability that may arise towards the syndicate lenders in such circumstances, and as a consequence may legally separate the agency function from the lending and bank relationship function. 48 It may also damage the borrower’s reputation more generally, while if it asks for permission, it will be seen a ‘good’ borrower, which may improve its ability to obtain cheaper finance elsewhere or on more favourable terms: see CK Whitehead, ‘The Evolution of Debt: Covenants, the Credit Market, and Corporate Governance’ (2009) 34 Journal of Corporation Law 641, 666. 49 That is, permission to breach the negative covenant on this particular occasion.

Negative Covenants in Loan Agreements  153 undertake that action.50 The request for a waiver has two consequences that are useful to the lender. First, it puts the lender on notice of the potential breach, and thus overcomes the monitoring problem mentioned above. Second, it opens a dialogue between the lender and the directors of the borrower. This dialogue could take the form of the negotiations previously described, but additionally it could take the form of discussion about the actual action the directors wish to take, allowing the directors to explain the benefits to the lender, and the lender to assess more fully whether the action actually does damage its interests. After all, the action could be neutral, or even beneficial, to the interests of the lender.51 The fact that the directors are likely to ask the lender to waive the potential breach also acts as a form of ‘early warning system’, and gives the lender the ability to choose how to react to the request.52 The final refinement to the analysis is to consider how directors, consonant with their duties, should approach the decision to pursue an action likely to result in a breach of a negative covenant. Rather than reaching a decision purely taking into account the interests of the shareholders, and then asking the lender to waive the potential breach and having the dialogue with the lenders discussed above, sensible directors will anticipate the likely reaction of a lender to a request for a waiver. If the directors feel that a dialogue with the lender is likely to have a negative outcome for the borrower, or that they would rather not alert the lenders to the reasons for considering the course of action, they may take the view that they will not breach and will not, therefore, ask for a waiver. On the other hand, if they feel that the proposed course of action is aligned with the interests of the lenders, they will be happy to consult the lenders, who are likely not to object.53 Thus, the mere existence of the covenant has the effect of internalising the interests of the lenders within the directors’ decision-making process, even though the directors are not parties to the loan agreement, and neither, arguably, do their duties require them in every case to cause the company to comply with its negative obligations under the loan agreement. Despite its academic interest, it may be, then, that the debate about the normative nature of contractual obligations does not need to be resolved here. The preceding discussion shows that the influence of negative covenants in a loan agreement on the directors’ behaviour is more subtle than merely giving rise to a duty of the directors not to cause the borrower to breach its obligations. Instead, the existence of the covenants may increase the likelihood that the directors 50 In fact, the request for permission is often expressly included in the negative covenant, which is then drafted as prohibiting an action if undertaken without the consent of the lender. 51 If such a situation were to arise in the case of a bond issue, it is still likely that a ‘price’ (by way of a ‘consent solicitation fee’) would be demanded by the Trustee acting for the bondholders, particularly in circumstances where the market price of the bonds has reduced (ie the bonds are currently trading at a discount), in order to reflect the market perception of the increased risk profile of the borrower. 52 See R Rasmussen, ‘Taking Control Rights Seriously’ (2018) 166 University of Pennsylvania Law Review 1749, 1761. 53 Although a ‘price’ may still be payable; see n 51.

154  Louise Gullifer and Graham Penn will engage in a dialogue with the lender. Moreover, the interest of the lender is incorporated in a way that will influence the way in which the directors run the company, so that in some situations they will decide not to pursue a course of action necessitating that dialogue. This internalises a duty of the directors to act in the interests of the lender much more effectively than a general legal duty to act in the interests of creditors generally, such as arise in the vicinity of insolvency. Moreover, the influence that negative covenants give the lender is far-reaching, in that it is likely to give greater leverage, at least in some situations, to force changes of direction in the company’s business and achieve other outcomes that are outside the actual scope of the covenants themselves (such as the removal and replacement of directors).

E.  Financial Covenants Although this chapter concerns negative covenants, much the same analysis applies to some aspects of financial covenants. It is, of course, possible for financial covenants to be breached as a result of external factors and not as a result of conscious decision making by directors, in which case the analysis is different. Furthermore, financial covenants clearly play an important role in providing an early warning of financial distress, and give a lender the ability to take action that could lead to a financial restructuring of the loan or otherwise to better protect itself in the run-up to insolvency. However, the presence of financial covenants also can lead to the results discussed in section IV.D, since a course of action being considered by directors can have the foreseeable effect that a financial covenant would be breached. Moreover, if the current operation of the company is not going well, and a breach of a financial covenant looks likely, the existence of the covenant (and the consequent negotiating power of the lender) will incentivise the directors to take steps to prevent that breach, when in other circumstances they might not do so.

F.  Lender as Directors? There is one possible danger for a lender if it exercises the influence previously explained to too great an extent. It could become liable for the actions taken under its influence, either as a de facto director or as a shadow director. A de facto director is someone who undertakes the functions of a director even though not formally appointed as such.54 He or she is subject to the full range of directors’ duties. There is no one test as to whether a person is a 54 See Revenue and Customs Commissioners v Holland; Re Paycheck Services 3 Ltd [2010] UKSC 51, [2011] Bus LR 111; Green v Marston [2016] BPIR 1224 [14]–[20]; Instant Access Properties Ltd (In Liquidation) v Rosser [2018] EWHC 756 (Ch), [2018] BCC 751 [216].

Negative Covenants in Loan Agreements  155 de facto director; it will depend on the facts of the case. The test is high, however: one general formulation is ‘whether [the person] was part of the corporate governance system of the company and whether he assumed the status and function of a director so as to make himself responsible as if he were a director’.55 It would seem, then, that the risk of a lender’s becoming a de facto director is slight. A shadow director is a person in accordance with whose directions or instructions the directors are accustomed to act,56 and who has ‘real influence over the affairs of the company’.57 The case law, such as it is, indicates that a lender who is acting to protect its own interests is very unlikely to be held to be a shadow ­director.58 This is on the basis that although the lender is in a position to put pressure on the directors to take certain actions, as a condition of not calling in the loan, the directors are free to decide whether to accede to this pressure or not.59 This reasoning accords with the discussion in section IV.D, in that the existence of the covenants align the directors’ duties with the interests of the lender, but, however, the directors are still free, consonant with the compliance with those duties, to refuse to comply with the requests of the lender if they feel that it is in the best interests of the shareholders to do otherwise. It is precisely this alignment that prevents the lender from being a shadow director. Having said this, as suggested by counsel for the claimant in Ultraframe (and apparently not contradicted by the judge), there is likely to be a limit to the ‘protection’ of lenders from being shadow directors, and it is possible, in certain circumstances, for the line to be crossed.60 In the context of a lender reacting to an actual or potential breach of a negative covenant, this could be where the lender uses its power to interfere with the running of the borrower in a way that goes beyond the protection of its own interest, which it has as a result of the existence of the covenants. It is difficult to see how this would arise in practice, since the lender’s directors themselves would be bound to act in the best interests of the lender. It could only arise where there was some benefit unrelated to the loan itself, which the lender sought to achieve by its dialogue and negotiations with the borrower. It has been argued, however, that since a shareholder who falls within the statutory definition can be a shadow director,61 and a shareholder acts to protect its interest, a lender who takes sufficient steps to control the borrower can be a shadow director.62 Further, the argument suggests that a lender could be a shadow director, not only by insisting on conditions in order not to accelerate and/or cancel a loan,

55 Smithton Ltd v Naggar [2014] EWCA Civ 939, [2015] 1 WLR 189 [33] (Arden LJ). 56 Companies Act 2006, s 251 (as amended by Small Business, Enterprise and Employment Act 2015, s 90); Insolvency Act 1986, s 251. 57 Secretary of State for Trade and Industry v Deverell [2001] Ch 340, [2002] 2 All ER 365. 58 Ultraframe (UK) Ltd v Fielding [2005] EWHC 1638 (Ch) [1267]; Re PFTZM Ltd [1995] BCC 280, 290–292. 59 Re PFTZM Ltd (n 58) 292; Ultraframe (UK) Ltd v Fielding (n 58) [1268]. 60 Ultraframe (UK) Ltd v Fielding (n 58) [1268]. 61 For example, in Re Augustus Barnett & Son Ltd [1986] BCLC 170 62 E Hadjinestoros, ‘Fear of the Dark: Banks as Shadow Directors’ (2013) 34(6) Company Lawyer 169.

156  Louise Gullifer and Graham Penn but also merely because of the existence of covenants that encroach too greatly on the way the business of the borrower would otherwise be run.63 The possibility of covenants’ being over-restrictive, and the consequences of this, is considered in section V. The duties to which the lender would be subject were it to be a shadow director are now determined by section  89 of the Small Business, Enterprise and Employment Act 2015, which provides that general directors’ duties do apply to a shadow director of a company where and to the extent that they are capable of so applying. The Act does contain a power for the Secretary of State to make provision by regulation as to which general duties apply to shadow directors and which do not, but those regulations have not yet been made and so the precise scope of the duties is uncertain.64 However, if a lender does act as a shadow director, it is likely to be vulnerable to liability under section 214 of the Insolvency Act 1986 (wrongful trading) if the relevant conditions are met.

V.  The Content of Negative Covenants So far, we have treated all negative covenants equally. This section drills down into the actual content of negative covenants, in order to give more context to the arguments made in the previous section. The discussion also addresses a related point. Negative covenants are rarely absolute prohibitions: they are usually drafted as prohibitions with exceptions. This is because some forms of the actions prohibited are likely to take place in the ordinary course of business of the borrower. Were they to be drafted as absolute prohibitions, this would have adverse consequences for both the lender and the borrower. It would mean that every time the borrower wanted to perform an action in the ordinary course of its business, it would need to obtain a waiver from the lender.65 This would, from the point of view of the borrower, be time-consuming and costly, and also risky, since the lender could use the potential breach as a trigger for renegotiation and/ or justification for extracting an increased price,66 even if it did not actually affect the lender’s risk. Another consideration, in some situations, is that a request for a waiver may involve the disclosure of confidential or sensitive information, for example when additional borrowing is required for an acquisition bid.67 The time and costs considerations also apply to the lender, but it may be beneficial for the lender for the triggers for renegotiation to be as wide as possible. First, this would

63 ibid 177. 64 S Witney, ‘Duties Owed by Shadow Directors: Closing in on the Puppet Masters?’ [2016] Journal of Business Law 311; and see also Rosser (n 54) [259]. 65 Which, if given, would typically be in the form of a specific waiver for that particular breach. 66 See n 46. 67 See M Forbes, C Harley and J Wagstaff, ‘The Evolution of ‘Soft Cap’ Covenant Baskets in the European Loan Market’ (2016) Journal of International Banking and Financial Law 285.

Negative Covenants in Loan Agreements  157 give it maximum protection against changes in the risk profile of the borrower, since it is not necessarily easy to foresee in advance precisely what actions of the borrower will change the risk. Second, it would enable the lender to take advantage of a request for a waiver to change the terms of the loan if it felt it needed to do this because of an external factor, such as changes in the market or in the focus of its own business. Third, a waiver request will invariably provide the lender with useful information about the borrower and its business, which will reduce the need for the lender to undertake its own diligence and monitoring.68 There is, however, a danger for the lender if the terms require the borrower to ask for a waiver in a wide a set of circumstances, in that the lender could be more vulnerable to being classified as a shadow director, particularly in circumstances where, for example, covenants are extended to prescribe the adoption of restrictive business plans that are carefully monitored by the lender.69 If the lender is secured there is an additional danger that the exceptions to the prohibition on asset disposal may be too broad, and the security interest over those assets could thereby be classified as a floating charge. Fixing the exact balance between prohibitions and exemptions is therefore a product of negotiation between the borrower and the lender. Apart from the disadvantages that come from having to deal with requests for a waiver, it may also be in the interests of the lender for the borrower to have freedom to run its business, since, despite the agency conflict discussed in section IV.D, it is not necessarily in the interests of the lender for the borrower to refrain from all risky activity. First, if the company is unable to grow at all, shareholders are likely to exit, leading to a drop in the share price and a possible change of control.70 Second, if the company grows and expands, it is likely to need more debt finance, which could result in more profitable loans for the lender. Third, if the loan agreement is too restrictive, the borrower is likely to want to escape at the first possible moment at which it can find alternative finance. From the borrower’s point of view, while more freedom to operate is advantageous, it may be that this can only come at the price of a higher interest rate, while agreeing to more restrictive covenants involves less obvious cost implications. It is therefore arguable that the interrelationship between the duties of the directors and the interests of the lender comes into play at the stage of the negotiation of the loan agreement, and it could be that, in some circumstances, directors could be in breach of their duties by permitting, in the course of negotiation, covenants that are too harsh and not in the best interests of the borrower. Once directors, in compliance with their duties, have negotiated a loan agreement, albeit with strict restrictive covenants, can this be said to set the boundaries for what actions will comply with their duties during the course of that loan agreement?



68 See

section IV.D. section IV.F. 70 The ability of the shareholders to do this will depend on the liquidity of the equity. 69 See

158  Louise Gullifer and Graham Penn The main risks faced by a lender were identified in section II. The most typical negative covenants addressing those risks will now be considered. Certain typical aspects of covenants will be ignored in the interests of simplicity. First, the fact that the covenants usually apply not only to the borrower but also to any companies in the same group and any guarantors who do not fall within that category. Second, a typical agreement will use the drafting technique of exempting ‘permitted activity’ within the actual clause then defining ‘permitted activity’ within a set of definitions elsewhere in the agreement. Third, since exceptions are heavily negotiated, there is likely to be great variety in terms found in actual loan agreements. Moreover, as mentioned later, the context of the loan will make a considerable difference, for example many loans made available to finance a leveraged buyout are likely to be ‘covenant-lite’ in the current market. Fourth, there are likely to be other terms of the loan agreement that have a similar effect to negative covenants, such as certain specified events of default,71 which, though not expressed as negative obligations, still give the lender the right to accelerate the loan and/or cancel its obligation to make further advances.

A.  Restrictions on Asset Disposal This covenant is likely to be drafted as a prohibition on all types of disposal of the borrower’s assets, that is, not only outright dispositions such as sales and gifts, but also leases and licences. For most operating companies, it is essential that dispositions in the ordinary course of business and for full value are exempted. As already mentioned, this is likely to cause any charge over the assets disposed of to be a floating charge, and the ordinary course-of-business exemption is therefore likely to be limited to certain categories of asset.72 Even if charges over certain categories of assets are expressly created as floating charges, including restrictions on asset disposal can be useful to a lender, since the definition of ‘ordinary course of business’ is very wide in the context of a floating charge.73 A borrower with bargaining power may also be able to obtain more freedom of operation in the form of a ‘basket’, that is, an exemption for disposals of any sort up to a certain value (a ‘cap’). This, on the other hand, raises the danger that the entire security package could be recharacterised as a floating charge, and a lender would 71 For example, rescheduling of debt owed to other creditors, change of the nature of the business, change of control. Of these, the change of control clause is the most significant in the current market, and also illustrates the balance discussed in the text. A change of control can have a major effect on the lender’s risk, and so needs to be a trigger for renegotiation, but too many limits on change of control can have a chilling effect on corporate growth. 72 A particularly problematic question arises in relation to equipment, which the company will need to replace from time to time. A right to dispose without the consent of the lender as part of this replacement process may well lead to a charge over equipment being floating; see H Beale et al, The Law of Security and Title-Based Financing, 3rd edn (OUP 2018) 6.123–6.129. 73 For the scope of ‘ordinary course of business’ in a floating charge, see Ashborder BV v Green Gas Power Ltd [2004] EWHC 1517 (Ch).

Negative Covenants in Loan Agreements  159 typically insist on the disposal of certain types of fixed assets being excluded from the basket.

B.  Restrictions on Borrowing To some extent, financial covenants requiring a particular gearing ratio have the effect of restricting further borrowing additional to the loan that has the benefit of those covenants. However, loan agreements typically also include express restrictions on additional borrowing, or other indebtedness. Again, there will be exemptions: typically, credit incurred in the ordinary course of business will be excluded, plus a ‘basket’ of a certain amount of permitted additional indebtedness, and this will, again, typically be heavily negotiated, given that it will otherwise present difficulties for the borrower to engage in material business activities, particularly expansion or major investment, without additional borrowing, either from the existing lender or a new lender. One method of compromising the interests of both parties here is for the cap to be flexible, depending on EBITDA rather than being a fixed immutable amount, and therefore taking account of the growth of the borrower’s business.74

C.  Negative Pledge Clause A negative pledge clause prohibits the grant of any security to any person other than the lender. It performs a slightly different function depending on whether the lender is secured or unsecured. In relation to assets over which the lender has a fixed charge, which has the effect that any disposition (including the granting of security ranking above the charge) requires the consent of the chargee, the additional benefit of a negative pledge clause is to require the borrower to ask for permission to grant lower-ranked security. The senior secured creditor will want an opportunity to veto such security, or to insist on entering into an inter-creditor agreement with the junior-ranked creditor, since without contractual controls it could be disadvantaged by the junior creditor’s embarking upon enforcement, or by the presence of another secured creditor in restructuring or other insolvencyrelated negotiations. Moreover, the grant of security will only take place if the borrower is incurring additional debt, and, as with a restriction on borrowing, the fact that the borrower has to ask for permission to take this step provides useful monitoring information for the creditor. If the lender has a floating charge, a negative pledge clause is critical in enabling it to obtain priority over later charges. It is now mandatory to indicate in

74 F Booth and S Whitehead, ‘The Evolving Fiction of EBITDA in the European Leveraged Finance Loans Market’ (2018) 10 Journal of International Banking and Financial Law 603.

160  Louise Gullifer and Graham Penn the registered particulars whether the charge includes ‘a term which prohibits or restricts the company from creating further security that will rank equally with or ahead of the charge’.75 While it is not entirely clear that it is compulsory to include this information where the negative pledge clause is contained in a loan agreement separate to the charge document, it is clearly in the interests of the secured lender to do so, since those searching the register will then have notice of the negative pledge clause, and anyone taking a subsequent charge will take subject to the first charge.76 If the lender is unsecured, it will be looking to the unencumbered assets of the company for its repayment. It therefore wishes to prevent any encumbrances being granted over these assets, or, at least, to have the opportunity to demand grant of prior ranking or equal ranking security as a condition of waiving the breach. The existence of encumbrances not only reduces the assets available for repayment to the lender (if it remains unsecured), it will also reduce the lender’s influence on restructuring or insolvency; conversely, if there are no encumbrances, the lender, as a major creditor, is likely to be in a strong negotiating position vis-à-vis the borrower. The negative pledge clause is a classic example of the benefits of a negative covenant being the internalisation of the lender’s interests in the directors’ decision-making process, plus the fact that a borrower will seek the lender’s consent to waive a potential breach. If the borrower does grant security without permission, and in circumstances where the lender has no notice, the remedies for the lender are weak and uncertain. It can, of course, accelerate and/or cancel the loan, but this is likely to drive the borrower into insolvency, at which point the lender, as unsecured creditor, will rank behind the third party that took security in breach of the negative pledge. It could potentially bring a claim in tort against that secured creditor for inducing breach of contract, but this is unlikely to succeed unless it can be shown that the creditor not only had notice of the terms of the negative pledge, but also intended the breach either as an end or as a means to an end, and not just the foreseeable consequences of its actions.77 Some forms of negative pledge clause seek to address this problem by including a provision for the automatic grant of security to the lender if the negative pledge is breached and security is granted to a third party, but this will not produce the desired result unless the subject matter of the security interest is identifiable78 at the time the loan is entered into, and the security interest is perfected (typically this will be by means of registration) within the requisite period. This latter condition may be fatal, because the lender will have no notice of the breach and therefore will be

75 Companies Act 2006, s 859D(2)(c). 76 English & Scottish Mercantile Investment Co Ltd v Brunton [1892] 2 QB 700, 707 (Lord Esher MR). 77 See the review conducted by Lord Hoffmann, at n 18. 78 Which will not typically be a problem, since the ‘automatic’ security will be created over the same assets that were secured in favour of the third party in breach of the negative pledge in order to maintain parity of treatment with that creditor.

Negative Covenants in Loan Agreements  161 unaware of the ‘automatic’ creation of the security. Even if perfected, the security is likely to rank in priority after the security created in favour of the third party.79

D.  Prohibition of Return of Capital to Shareholder Although there are some statutory restrictions on return of capital in the form of dividends, redemption or otherwise, a loan agreement will typically include express restrictions. This is, first, because they can be more effective than the statutory restrictions,80 for example they can rely on a cash-flow test rather than a historic balance-sheet test, and also because they can be tailored to the circumstances of the parties more accurately.

VI. Conclusion This chapter has examined the way in which negative covenants in loan agreements not only constrain the activities of the borrower in order to limit the lender’s risk, but also incentivise directors to take the interests of the lender into account when making all decisions about the future conduct of the business. Moreover, they enable the lender to take part in a continuing dialogue with the directors about such conduct, as well being able to take steps to protect its own interests when the risk increases from that which it took on at the time of the loan agreement. While there are some checks on the lender’s ability to do this, namely, the bargaining power of the borrower and the possibility of adverse legal consequences (such as a lender being a shadow director, or a charge being floating rather than fixed), the utility of negative covenants, coupled with financial covenants, is very significant. In performing this role, the purpose and operation of negative covenants in loan agreements is somewhat different from many types of contractual obligations. This conclusion, however, does not undermine contractual norms but, rather, illustrates their flexibility.



79 See

Gullifer and Payne, Corporate Finance Law (n 2) 6.3.1.6.2. Act 2006, ss 830–833.

80 Companies

162

9 ‘Ethical Clauses’ in Global Value Chain Contracts: Exploring the Limits of Freedom of Contract LUCINDA MILLER*

I. Introduction As global production has transformed from a Fordist model, in which all stages of production – research, design, manufacture, distribution, marketing and retail – are owned and managed by the one capital unit located in a single state, towards a vertically disintegrated model, in which many of the components of production are outsourced offshore, global value chains (GVCs) have become the ‘world economy’s backbone and central nervous system’.1 These GVCs must rely on a legal infrastructure, and it is the institution of contract that has become key for the legal governance of the inter-firm linkages along the chain.2 The ‘openness’ of the contractual form to the multifarious ways in which business wishes to organise its activities means that it has become increasingly common for commercial parties to include obligations of a social and environmental nature in their transnational contracts. And it is such obligations, labelled as ‘ethical clauses’ in this chapter, that are the focus of this contribution. * I am grateful to Catharine Macmillan and Radosveta Vassileva for their insightful comments on an earlier draft, as well as to the participants at the UCL Conference, ‘The Contents of Commercial Contracts: Terms Affecting Freedoms’ (9 and 10 May 2019). 1 O Cattaneo, G Gereffi and C Staritz (eds), Global Value Chains in a Postcrisis World: A Development Perspective (WorldBank 2010) 7, available at https://openknowledge.worldbank.org/ handle/10986/2509. More forcefully still, it has been said that ‘the most important paradigm for understanding the global economy, and the political and social relationships that both guide it and stem from it, is no longer the template of the market but rather the role of global value chains’: K Sobel-Read, ‘Global Value Chains: A Framework for Analysis’ (2016) 5 Transnational Legal Theory 364, 367. 2 This account of the GVC is simplistic since GVCs rely on various mechanisms of transnational governance and a variety of different forms of inter-firm linkages – some traditional, such as contract law, others novel and relying on corporate law, property law or more informal mechanisms. Further, the governance mechanisms for each inter-firm linkage may vary at points along the chain and vertical integration of mid-tier suppliers may occur. For more detail see, Sobel-Read, ‘Global Value Chains’ (n 1).

164  Lucinda Miller Ethical clauses stipulate obligations in the commercial contracts between firms along the length of the value chain, in regard to such matters as human rights, labour conditions and environmental sustainability. More concretely, they may set standards for the working conditions in garment factories, they may stipulate that farmers are paid fair wages, that timber is legally harvested, or that products must not be manufactured with ozone-depleting substances or be created with engineering techniques such as genetic modification. These obligations are distinguishable from, and generally ancillary to, the core obligations of the commercial agreement, which concern matters such as the delivery, price, quality3 or quantity of the product. Ethical obligations broaden the traditional scope of the commercial contract, embracing within its architecture standards that would not traditionally form part of the commercial exchange or be dealt with by the contractual form. Their almost routine incorporation in commercial contracts,4 either explicitly or by reference,5 reveals the novel ways in which contracts are routinely being deployed by commercial parties. There is a range of factors that explain commercial parties’ decision6 to include these clauses in their contracts. Perhaps the most important7 relates to the pressure exerted on corporations by civil society, investors and by consumers. Whereas pressure was once exerted purely on public institutions, the increasing and visible market power of many corporations has shifted attention from public to private responsibility, and onto those corporations that society views as responsible for harms caused. Companies with a strong brand presence on the market are particularly exposed to a critical public opinion, and pressure especially mounts at times of disaster, such as the Rana Plaza disaster of 2013.8 In an age of social media and well-mobilised consumer groups and non-governmental organisations (NGOs), ‘best practice is increasingly being treated as setting up a contractual obligation on suppliers to meet specified CSR [corporate social responsibility] standards’.9 3 A distinction is made here between product quality and process quality, the latter being the type of obligation referred to here as an ethical obligation. 4 There is plenty of empirical evidence of this phenomenon. See, eg, MP Vandenberg, ‘The New Wal-Mart Effect: The Role of Private Contracting in Global Governance’ (2007) 54 UCLA Law Review 913; D McBarnet and M Kurchiyan, ‘Corporate Social Responsibility Through Contractual Control? Global Supply Chains and “Other-Regulation”’ in D McBarnet, A Voiculescu and T Campbell (eds), The New Corporate Accountability: Corporate Social Responsibility and the Law (CUP 2007) 59; F Cafaggi, ‘The Regulatory Functions of Transnational Commercial Contracts: New Architectures’ (2013) 36 Fordham International Law Journal 1557. 5 For fuller discussion on the different mechanisms through which ethical standards are incorporated into supply contracts, and thus become enforceable contractual terms, see A Rühmkorf, Corporate Social Responsibility, Private Law and Global Supply Chains (Edward Elgar Publishers 2015) 85–96. 6 Although, as it is the lead firm (buyer) who tends to have the strongest bargaining position and the most to gain from these clauses, it is this party that typically drives this choice in their contracts with their suppliers. 7 Another important factor is the desire to show a willingness to self-regulate in an attempt to reduce the need for public forms of regulation. 8 Reports state that 1,135 people died and approximately 2,500 people were injured in the collapse of a factory building, which housed a number of garment factories, in Dhaka, Bangladesh. For more on this tragedy, see www.bbc.co.uk/news/world-asia-22476774. 9 McBarnet and Kurchiyan, ‘Corporate Social Responsibility’ (n 4) 65.

‘Ethical Clauses’ in Global Value Chain Contracts  165 In light of the pervasiveness of ethical contracting along the global value chain, it seems timely for contract lawyers to look more closely at the characteristics of ethical clauses and consider whether they might have any transformative implications for the discourse and doctrine of contract law. We shall see that they are a distinctive kind of contractual obligation in two important, interrelated ways: in the public nature of the interest they seek to protect, and in their regulatory characteristics. Highlighting these features invites reflection on the demands that ethical clauses might make on the institutions of contract law and the extent to which they may challenge conventional understandings about the role that contract law should play. Of particular interest is the role that contract law plays in the enforcement of ethical clauses, and the extent to which its current framework might enable, or impede, the public and regulatory features of ethical contracting. Central to such an enquiry are important questions about the limits of freedom of contract as an organising principle when commercial parties contract in this way, and whether a less facilitative, more regulatory approach to contract law is desirable. Admittedly, to imagine the law of contract, an institution at the very heart of private law, as having any public-interest regulatory role seems odd and may be criticised for misunderstanding the very nature of private law.10 And yet it would be wrong to dismiss such a role too readily and consign these questions to the realm of public law. The public/private divide is becoming far less convincing in legal consciousness, and in many practical aspects the boundaries between the two realms are becoming increasingly blurred. More pertinently, since commercial parties are deploying contracts in this novel (public and regulatory) way, it is important to look more closely at the ways contract law can and should respond. This is not simply an argument based on the misalignment between contract law and commercial practices; a desirability that contract law should ‘catch up’ with the way that contracts are being deployed. Whilst it is indeed important that contract law should evolve in ways that respond to, and are compatible with, the environment it must regulate, the need to explore contract law’s role in responding to ethical contracting is more fundamental than that due to the nature of the interests that are involved. In essence, it seems crucial to reflect on whether contract law has any role in ensuring that public interests of the kind that are found in ethical clauses are not neglected or allowed to be instrumentalised though the commercial logic of private ordering. This chapter therefore invites reflection on contract law’s role in the governance of ethical clauses and the extent to which ethical contracting may have any implications for the contractual freedom of commercial parties. Difficult questions need to be asked about whether, and if so how far, we are prepared to recalibrate our current conceptions of contract law to deal with the regulatory functions and public interests that these clauses embody. With these aims in mind, section II first

10 Or

disparaged for being bluntly instrumentalist.

166  Lucinda Miller examines the distinctive public and regulatory characteristics of ethical clauses and, in doing so, highlights the tension they provoke with the autonomy-based understanding of contracts and of contract law. The focus of section III is on the enforcement of ethical clauses and the role that contract law can and should play in this respect. Section IV concludes.

II.  The Distinctiveness of Ethical Clauses This section directs its attention to the core features of ethical clauses that render them distinctive from the conventional commercial obligation – the public nature of the interests that these clauses seek to govern and their transnational regulatory potential. As suggested in section I, an appreciation of these features is important for the normative thrust of this chapter, for it invites scrutiny of where the limits of contractual autonomy might lie and may impel a reorientation in our thinking about the appropriate role that contract law should play. Before observing the distinctive features of ethical clauses, it is helpful to be reminded of the centrality of freedom of contract within our understanding of private ordering and of contract law. According to classical liberalism, justification for private ordering is found in the opportunities it gives to individuals to pursue their own conception of the ‘good’ with only minimal interference from the state. Similarly, from the viewpoint of neo-classical economics, justification for the private ordering paradigm is located in its promise of welfare maximisation. From both these perspectives, therefore, individual autonomy is viewed as a paramount value and a key organising concept for society. The principle of freedom of contract is the legal correlate of this logic, and is a principle that remains central to modern contract law. As such, contract law is primarily designed to support and protect this freedom by providing a set of rules and institutions that confer on parties the power to shape their own obligations, and by ensuring that freelychosen contractual commitments are upheld.11 As we examine the public and regulatory dimensions of ethical contracting, we start to see how ethical clauses may present a challenge to freedom of contract as an organising principle for contract law.

A.  The Public Dimension of Ethical Clauses There are two, closely-related aspects to the public dimension of ethical clauses that will be discussed here. First, the public (and deeply political) nature of the interest that the contracting parties are incorporating into their commercial contracts. And, second, the public nature of the role that private parties are performing

11 To

the extent that this is compatible with the moral standards of society.

‘Ethical Clauses’ in Global Value Chain Contracts  167 when contracting for interests of this kind. In highlighting both these aspects, the discussion examines the challenges that these clauses pose for our conventional understanding of contracts and of the centrality of individual autonomy. In turn, it also invites consideration of the role that contract law might play in enabling and enforcing interests of this kind. Insofar as the first element is concerned – the public nature of the ethical clause – whilst it is true that contracting parties will choose to incorporate ethical clauses into their contracts only if it is in their own self-interest to do so,12 at the same time the clauses concern interests that lie beyond those of the contractual parties. In contractual jargon we would describe ethical clauses as protecting ‘third-party interests’, but we should be careful not to let this language mask the real nature of these interests, for the third-party interests they invoke are collective, public interests of fundamental importance. These clauses concern such things as the labour conditions of workers in supplier factories, fundamental human rights, environmental standards in production processes or the sustainability of forestry practices. They are of global public concern, and thus of significance far beyond the purely private interests reflected in the bilateral contractual relationship. In depicting ethical contracting as serving the public interest, we should bear in mind that there is always a public dimension to contract. This is because private ordering has never been wholly detached from its broader institutional setting, even if its importance for the broader collectivity has often been lost within the powerful rhetoric of ‘freedom of contract’. The centrality of private autonomy for liberalism’s vision of society has already been alluded to above. In addition, liberal economic thinking has viewed private agreement and individual autonomy as related to the common good through the social institution of the market.13 Although the neat convergence between autonomy and social welfare has been contested,14 this thinking nevertheless illustrates how private normativity has always been embedded in social institutions and public values, and guaranteed by the state. And yet recognising that private agreements may serve the public interest in this way has posed little challenge to the individualist, bilateral rationality of contract law. Neither has it posed any threat to party autonomy. On the contrary, the public interest is wholly dependent on a broad sphere of party freedom and individual maximising behaviour. In contrast, what is challenging about ethical contracting is how the public interest that these clauses serve becomes uncoupled from the individualist,

12 As mentioned in section I, incorporation of this type of clause can generally be understood as a response to the ethical market preferences of consumers and investors, as well as to pressure exerted on them by civil society. There is clearly private gain in contracting for public interests. 13 B Lomfeld and D Wielsch, ‘Foreword. The Public Dimension of Contract: Contractual Pluralism, Beyond Privity’ (2013) 76 Law and Contemporary Problems i, iv. Of course, this economic, instrumental rationale is not the only conception of contract, but it shows how important it is for contract theory to take account of the institutional structure of society. 14 See, eg, MJ Trebilcock, The Limits of Freedom of Contract (Harvard University Press 1997).

168  Lucinda Miller welfare-maximising interests of the contracting parties. Although ethical clauses serve the self-regarding interests of the contractual parties, they also, and independently, serve public interests that are not necessarily synonymous with those of the contractual parties. In this way, freedom of contract becomes untethered from, and may conflict with, the collective interest, raising a unique set of questions about the appropriateness, and precise limits of, an autonomy-based conception of contract when confronted with clauses of this kind. To put it in economic language, the assumption that contractual allocations are welfare-enhancing and that it is therefore productive not to intervene into self-regulatory market processes is far less persuasive when contracts concern global public interests. Parties are not well placed to assess what is welfare-enhancing when they contract for interests that lie outside of the contractual bargain and, it could be argued, should not necessarily have presumptive authority over their contracts.15 On the one hand, contract theory seems to cope perfectly well with the parties contracting in this ‘public interest’ way – after all, provided that both parties consent,16 the parties are perfectly free to contract for altruistic interests, in the same way that they are free to contract for self-regarding interests.17 In this respect, therefore, the facilitative nature of contract law can be said to be perfectly compatible with ‘public-interest contracting’. On the other hand, however, there are some important tensions when contract law encounters interests that lie beyond the self-regarding interests of the parties. How, for example, should remedial structures be fashioned when the interests that the clause protects are those of the collectivity rather than private commercial interests? Alternatively, and perhaps more profoundly, should the intention of the contracting parties be central when the courts are tasked with interpreting commitments that pursue public goals?18 And what should be the role of contract law in the enforcement of clauses that secure interests of a public nature? We return to these points in section III of this chapter.

15 See A Bagchi, ‘At the Limits of Adjudication: Standard Terms in Consumer Contracts’ in LA DiMatteo and M Hogg (eds), Comparative Contract Law: British and American Perspectives (OUP 2016) 439, 450. 16 Although beyond the scope of this chapter, there are interesting issues concerning contract law’s limited ability to protect weaker participants in the chain, who may have one-sided and burdensome clauses imposed on them by lead firms and, at the same time, find value in the chain being increasingly shifted away from them. For a suggestion that the doctrine of unconscionability may play a role in policing these clauses, see V Ulfbeck, O Hansen and A Andhov, ‘Contractual enforcement of CSR clauses and the protection of weaker parties in the supply chain’ in V Ulfbeck, A Andhov and K Mitkidis (eds), Law and Responsible Supply Chain Management: Contract and Tort Interplay and Overlap (Routledge 2019) 46, 55–57. 17 ‘[T]he personal autonomy enshrined in contract law is abstract’, after all, and so commercial parties are free to contract for ethical interests if they so desire: J Wightman, ‘Private Law and Public Interests’ in T Wilhelmsson and S Hurri (eds), From Dissonance to Sense: Welfare State Expectation, Privatisation and Private Law (Ashgate Publishing 1999) 253, 266–67. 18 See A Beckers, ‘Corporate Codes of Conduct and Contract Law: a doctrinal and normative perspective’ in R Brownsword, R van Gestel and HW Micklitz (eds), Contract and Regulation: A Handbook on New Methods of Law Making in Private Law (Edward Elgar Publishing 2017) 89, 123.

‘Ethical Clauses’ in Global Value Chain Contracts  169 Turning to the second element of our discussion on the public dimension of ethical contracting – the public role that contracting parties are performing – what should be emphasised here is that it is the parties, rather than contract law (whether through mandatory rules, doctrine or through contract law litigation), who introduce the public interest into the contractual relationship. In other words, the public interest is present in the contract not because the law has made it a relevant factor in the contract, but because the parties themselves have chosen to introduce the public interest consideration to their contractual relationship. Although seemingly a trite observation, this latter point forces us to think more broadly about what is at stake when public interests are shaped by contractual parties. Traditionally, of course, we are accustomed to thinking about the regulation of these kinds of ethical issues in public, rather than private, law terms. But the neoliberal political and economic climate has tended to reify the market as a suitable locus for dealing with public issues, and it views private, rather than public, forms of ordering favourably. Such an ideological position fosters an environment in which private actors come to exercise significant authority as ‘new global rulers’19 and supports an increasing role for legal forms such as contract to organise global economic ordering. The deployment of contracts to respond to global social and environmental problems, therefore, denotes a critical shift in responsibility, rendering private actors (viz, contractual parties) as responsible for commitments that impact upon people and communities in a variety of life-changing ways. But when private actors wield public power in this way and define the general interest, there are at least two central issues that must be confronted. First, it raises the question as to how to construct an account of private ordering that can ‘legitimise’ such private authority. Such a task assumes particular importance when we appreciate that these clauses have distributional effects that may have huge consequences for the lives of individuals, and which may affect them in oppressive, ‘autonomylimiting’ ways.20 Yet whilst it seems entirely appropriate to examine ways in which new forms of governance should adhere to the democratic norms of the public sphere,21 at the same time this seems misplaced from the perspective of contract 19 T Büthe and W Mattli, The New Global Rulers: The Privatization of Regulation in the World Economy (Princeton University Press 2013). 20 T Macdonald and K Macdonald, ‘Non-Electoral Accountability in Global Politics: Strengthening Democratic Control within the Global Garment Industry’ (2006) 17 European Journal of International Law 89, 106. Once responsibility for these matters is shifted to the private domain, the capacity for regulatory self-determination is limited and, as Cafaggi and Pistor put it, individuals may be subjected to regimes of private ordering to which they have not consented and have not been able to contest through local forms of politics: F Cafaggi and K Pistor, ‘Regulatory Capabilities: A Normative Framework for Assessing the Distributional Effects of Regulation’ (2015) 9 Regulation and Governance 95. 21 J Freeman, ‘Extending Public law Norms Through Privatization’ (2003) 116 Harvard Law Review 1285 (discussing whether democratic norms of accountability, due process equality and rationality can be extended to private actors in the process of ‘contracting out’ public services – the ‘publicisation’ of the private sphere).

170  Lucinda Miller theory, since public law’s legitimising values (transparency, accountability, participation, equality, etc) are alien to the rationality and discourse of contract law. Contract law’s legitimising principle, freedom of contract, is in direct tension with the idea that the authority of contractual parties should be limited to take account of these public values, or that contracts should be scrutinised for their democratic credentials. And so here too we are confronted with the tension that arises when an autonomy-based institution is deployed to govern the public good. Second, and relatedly, we must be alert to the extent to which contractualisation may jeopardise fulfilment of the public interest. The danger is that when private actors regulate in areas that would be commonly perceived as public or governmental in nature, they tend to do so in ways that enable them to pursue their own interests and to further consolidate their economic power, and this comes at the expense of third parties or the collective interest. Anticipating the discussion in the next section on the regulatory role of these contracts, we can deploy the concept of ‘regulatory capture’ to depict this concern. Although developed within the context of regulatory decision-making by public bodies, the notion seems particularly apposite as a conceptual tool for our purposes, since it emphasises the risks involved when corporations contract for the type of fundamental public interest found within ethical clauses. Firms are essentially self-interested, partisan actors motivated by capital accumulation and primarily answerable to shareholders. They are thus programmed to enhance their economic interests, insofar as structural, institutional or legal limits allow, and are not constrained in the same way as public regulators by the requirement to regulate in ways that reflect publicly-defined social goals.22 In this respect, economic motivations will drive the lead firm’s decision to incorporate (or not to incorporate) ethical commitments in its contracts, and the same economic imperatives will also guide the manner in which the ethical clause is shaped. These economic considerations may undermine the way that the public interest is protected. For example, the corporation may set standards that meet the domestic laws of supplier states, but which are lower than the widely acknowledged standards of the international community. Or it may choose to articulate the ethical standard in deliberately vague or aspirational language, softening the obligational force of the ethical clause, and/or making it difficult to point to the standard required for performance or to determine whether breach has taken place. Finally, and this is a point that will be pursued later in the chapter, it may choose not to enforce the obligation on breach. Whilst all of these aspects may threaten the attainment of the collective interest, they are nevertheless permitted by contract law. Freedom of contract (and most pertinently here, ‘term freedom’) means that parties are at liberty to design ethical clauses in a manner that suits the economic logic of their commercial 22 For a more detailed discussion of the notion of ‘regulatory capture’, see BM Mitnick, ‘Capturing “Capture”: Definition and Mechanisms’ in D Levi-Faur (ed), Handbook on the Politics of Regulation (Edward Elgar Publishing 2011) 34.

‘Ethical Clauses’ in Global Value Chain Contracts  171 relationship, and contract law can do little to ensure parties design their contracts differently. As Wightman observes, the public interest is not an argument that tends to be deployed in favour of imposing obligations on parties23 but rather acts negatively, to deprive terms of effect where they are against public policy.24 It is a reminder of the limits associated with the use of the contractual instrument as a tool for public interests. Although the rationality of contract law is resistant to the idea that parties should be compelled to contract in ways that are sensitive to the collective stakes in their agreements, this does not mean that commercial parties cannot be influenced to do so through mechanisms outwith contract law. For example, we have already observed that private (market) pressure plays a role in channelling the transnational behaviour of corporations in ways that are responsive to ethical issues, and largely explains the increasing presence of ethical clauses in their contracts.25 But beyond this, it is easy to forget that the public sphere may also play a role in shaping contractual freedom and influencing the design and content of ethical clauses, even if such a role is more challenging in the transnational sphere than it is in the domestic setting. For example, although the regulatory power of the state is generally perceived to be weakened in a complex transnational economy,26 globalisation and the interconnectivity with which its processes are associated also bring some opportunities for states, allowing them to gain regulatory traction over events that take place abroad27 and, in some instances, to shape the content of transnational contracts along the value chain. Whilst more detailed research is required to ascertain the precise impact that domestic legislation might have on the design of contractual clauses in the value chain, there is some evidence to suggest that by imposing liability on firms in their jurisdiction for certain environmental or social harms, even when those harms occur beyond the regulating state’s borders, such legislation might indirectly control the substantive content of contracts. The UK Bribery Act 2010 provides a useful example of how this might function. This legislation makes it an offence for commercial organisations to fail to prevent bribery (section  7(1)) and imposes a fine if guilty of such offence (section  11(3)). A defence is available, however, if the commercial organisation can show that it had in place ‘adequate procedures’ designed to prevent persons ‘associated with’ the organisation from undertaking such conduct (section 7(2)). Under the Guidance

23 Although, as Wightman also notes, the imposition of obligations to protect the private interests of one party is common place: Wightman, ‘Private Law and Public Interest’ (n 17) 260. 24 ibid. 25 As Wai puts it, the market is a ‘social and regulatory order that deploys private power to control private power’: R Wai, ‘Private v Private: Transnational Private Law and Contestation in Global Economic Governance’ in H Muir Watt and D Fernández Arroyo (eds), Private International Law and Global Governance (OUP 2014) 34, 43. 26 This point is developed in section II.B. 27 See, eg, J Scott, ‘Extraterritoriality and Territorial Extension in EU Law’ (2014) 62 American Journal of Comparative Law 87.

172  Lucinda Miller published by the Ministry of Justice in 2011,28 it is recommended that organisations use anti-bribery terms and conditions in their supply contracts, and there is evidence that the potential liability under the Bribery Act has had an effect not only on the prevalence of bribery clauses in GVC contracts (at least, in circumstances where the lead firm is UK-based), but also on how strictly lead firms treat bribery within their GVC contracts.29 In brief, the threat of legal liability for lead firms seems to enhance the position within the GVC contract of the issue regulated by the legislation. It makes it less likely that the lead firm will draft the clause in vague, aspirational language, and may even encourage closer monitoring for compliance along the chain. Ultimately, domestic liability regimes of this type seem to offer a promising technique for ensuring that transnational contracts are shaped in ways that enhance the public interest, and certainly seems more effective when compared with the forced disclosure techniques that have been more widely deployed.30 Domestic regulators might do well to consider expanding such techniques, and may come under increasing pressure to do so. In addition, the public sphere may play a crucial role in setting the global ethical standards that later become incorporated into commercial contracts. It is increasingly common to find that the substantive content of the ethical clause has its origin in guidelines and standards that have been elaborated by public bodies or agencies, rather than drafted by the contractual parties themselves. For example, firms may incorporate into their contracts, either by an express term or by reference, the labour standards of the International Labour Organisation (ILO),31 the OECD’s Guidelines for Multinational Enterprises32 or the UN’s Guiding Principles for Business and Human Rights.33 These internationally recognised, non-binding (‘soft law’) global standards reflect a consensus amongst the international community that, at the very minimum, business needs to be compatible with cosmopolitan values, such as respect for human rights, labour standards and environmental sustainability. As such, these public standards, even though voluntary, establish the normative benchmarks34 for businesses acting both domestically 28 Ministry of Justice, The Bribery Act 2010: Guidance about Procedures which Relevant Commercial Organisations Can Put into Place to Prevent Persons Associated with Them from Bribing (section 9 of the Bribery Act 2010) (2010) available at www.justice.gov.uk/downloads/legislation/bribery-act2010-guidance.pdf (as the title indicates, the Guidance is required by s 9 of the Act). 29 One author has shown that anti-bribery duties are expressed in a particularly strict and verifiable contractual language, Rühmkorf, Corporate Social Responsibility (n 5) 113. 30 See, eg, s 54 of the UK’s Modern Slavery Act 2015; California Transparency in Supply Chains Act of 2010, SB 657, 2010 Reg Sess, 2010 Cal Legis Serv Ch 556 (West 2010) (codified at CAL. CIV. CODE § 1714.43). 31 For details on the eight fundamental Conventions, see at www.ilo.org/global/standards/ introduction-to-international-labour-standards/conventions-and-recommendations/lang--en/index. htm. 32 The text of the Guidelines can be found at http://mneguidelines.oecd.org/guidelines/. 33 The text of the Guidelines can be found at www.ohchr.org/Documents/Publications/Guiding​ PrinciplesBusinessHR_EN.pdf. 34 Their ‘social license to operate’ as Gunningham describes it: N Gunningham, ‘Corporate Environmental Responsibility: Law and The Limits of Voluntarism’ in McBarnet, Voiculescu and Campbell (eds), The New Corporate Accountability (n 4) 476.

‘Ethical Clauses’ in Global Value Chain Contracts  173 and transnationally, meaning that it becomes increasingly routine to incorporate them into commercial contracts.35 These brief examples illustrate how public law (soft and hard) may be able to govern and steer, albeit indirectly, the content of commercial contracts in the GVC and exert a disciplinary force on transnational economic power. To appreciate the importance of this interaction between public and private realms of activity, it might be helpful to conceptualise this as the public sphere’s providing a constitutional anchorage for transnational commercial contracting. This public dimension to these transnational contracts seems important. Although even the most advanced forms of transnational economic activity are not autonomous or functionally separate from state legal orders,36 nevertheless, private ordering is less obviously anchored to any public, constitutional framework when it takes place across and beyond jurisdictional borders, and so it is far more insulated from the public values and mechanisms of control that would usually be furnished by the domestic public sphere. When domestic contract lawyers speak of freedom of contract, it is implicitly acknowledged that that freedom is politically constructed, in the sense that it is the state that constitutes the (private) arena in which contractual freedom and the exercise of private economic power can take place.37 But when private ordering takes place in the transnational sphere, these ties to the public domain are far more tenuous, and we cannot readily assume that the basic structure of society in which these contracts operate is justified in the same way that we might when we consider domestic contracting. As Michaels and Jansen aptly observe, the ‘autonomy of private law takes place not in the presence and under the protection of the state but rather in, and due to, its absence’.38 The worry, therefore, is that contractual freedom and private power becomes unfettered, and market values and the self-interest of economic actors become overly privileged. This seems particularly concerning when private parties are contracting for interests of this public nature, making it even more important to ensure that market processes have some form of ‘touchdown’39 point in the public sphere, minimising the potential for the public interest to be undermined through the contractual practices of commercial parties.

35 The insourcing of these public standards into their commercial contracts is also attractive to a firm, since the public standard-setting bodies tend to have greater expertise in social and environmental matters. 36 There is not ‘lift-off ’ from state legal orders, since transnational GVC contracts interact with state regimes of private law at those sovereign venues that are identified in the contract through jurisdictional and choice-of-law clauses and at the moment that private actors deploy domestic contract law doctrines to contest and enforce transnational private rules; see further, R Wai, ‘Transnational Liftoff and Juridical Touchdown: The Regulatory Function of Private International Law in an Era of Globalization’ (2002) 40 Columbia Journal of Transnational law 209. 37 J Bakan, ‘The Invisible Hand of Law: Private Regulation and the Rule of Law’ (2015) 48 Cornell International Law Journal 279, noting the invisible hand of public law in constituting private power. 38 R Michaels and N Jansen, ‘Private Law Beyond the State? Europeanization, Globalization, Privatization’ (2006) 54 American Journal of Comparative Law 843, 872. 39 Again, adopting the language of Wai, ‘Transnational Liftoff ’ (n 36).

174  Lucinda Miller We return to this discussion in section III, when we consider the extent to which the institutions of contract law might serve as an additional disciplining force, or ‘touchdown’ point, through the enforcement of ethical clauses.

B.  The Regulatory Dimension of Ethical Clauses Although intimately connected with the discussion in section II.A, the regulatory dimension of ethical contracting is nevertheless worth setting apart here. The discussion will highlight the role that ethical clauses may play in the regulation of global production processes, as well as the tension that arises when viewing contracts through this regulatory lens. Non-state actors have been increasingly viewed as significant players in the governance and regulation of global production,40 and ethical clauses in GVC contracts can be analysed from this perspective. Transnational contracts that protect social and environmental interests can be depicted as a form of transnational private regulation41 and viewed favourably owing to their perceived value in responding to a ‘regulatory gap’ exposed by a diminishment in the regulatory power of the state.42 In a context of globally fragmented production processes, social risks have intensified, but at the same time have become dispersed and uncoupled from the domestic regulatory sphere.43 This means that traditional, state-based attempts at taming the pernicious effects of economic production grow in complexity, stretching the regulatory capabilities of even the most highly developed states. International instruments, whilst obviously valuable for dealing with global problems, are also fraught with difficulties, largely of a political nature.44 And so, as state power is diminished, ethical contracting steps in as a promising alternative or supplement for (in Polanyian terms) ‘re-embedding’ the market.45 Even when ethical clauses do no more than oblige suppliers to meet, 40 There is a plethora of private (as well as hybrid public-private) governance arrangements that set and implement standards in the areas of labour rights, human rights and the environment. For a detailed examination, see KW Abbott and D Snidal, ‘The Governance Triangle: Regulatory Standards Institutions and the Shadow of the State’ in W Mattli and N Woods (eds), The Politics of Global Regulation (Princeton University Press 2009) 44. 41 eg F Cafaggi, ‘New Foundations of Transnational Private Regulation’ (2011) 38 Journal of Law and Society 20. 42 Vandenberg ‘The New Wal-Mart’ (n 4). 43 What is more, vertical disintegration of production provides ‘a permissive environment for wrongful acts by companies of all kids without adequate sanctioning or reparation’: UN Secretary-General’s Special Representative for Business and Human Rights, ‘Report of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and Other Business Enterprises: Protect, Respect and Remedy: A Framework for Business and Human Rights’, UN Human Rights Council, 189 UN Doc A/HRC/8/5 (7 April 2008) (by John Ruggie). 44 The difficulties in reaching political consensus across a sufficiently broad number of sovereign states is an obvious hindrance to the value of these instruments. 45 K Polanyi, The Great Transformation: The Political and Economic Origins of Our Time, 2nd edn (first published 1944, Beacon Press, 2001); K Macdonald and S Marshall, ‘Transnational Business and Social Risk Re-Embedding Transnational Supply Chains Through Private Governance’ in B Lange,

‘Ethical Clauses’ in Global Value Chain Contracts  175 rather than go beyond, the substantive requirements of the supplier’s domestic law, there is still regulatory value in contractualisation. In such instances, it is in the alternative, private law method of enforcement that it may offer. This is especially useful in supplier states of the Global South, where public law enforcement of social and environmental standards may be lax. Similarly, when, as we saw previously, the ethical clauses are insourced from international ‘soft’ law guidelines and standards, contractualisation of these standards within the private law relationship transforms these principles from non-binding public law declarations into contractually binding obligations that have horizontal effect. The contract, therefore, offers promise as a mechanism for enhancing the effectiveness of public law principles and a useful regulatory tool.46 From the perspective of regulatory theory, conceptualising contracts as a form of ‘regulation’ does not seem too challenging, although it does require a wide definition of regulation to be embraced.47 This is because ‘regulation’ is traditionally understood to be a form of public or governmental activity,48 describing the way that the state and public authorities implement, monitor and enforce rules in order to organise and control economic, political and social activities, and secure the objectives and values of a particular polity. The notion of regulation through the instrument of contract, therefore, departs from the assumption that the state is the primary locus for articulating the collective goals for a community49 and embraces the idea that other social actors have a regulatory role to play. It is a definition of regulation that is becoming more widely accepted and aligns, for example, with Julia Black’s discussion on ‘decentred regulation’,50 in which regulation is depicted as being ‘decoupled’ from government and ‘diffused through society’.51 This broad definition is able to capture the notion of ethical contracting as a regulatory tool. From a contractual perspective, however, the depiction of commercial contracts as tools of global regulation might not seem very promising. Regulation and contract are very distinctive paradigms, and the integrity of the latter seems F Haines and D Thomas (eds), Regulatory Transformations: Rethinking Economy – Society Interactions (Hart Publishing 2015) 105. 46 For this point, see Cafaggi, ‘Regulatory Functions’ (n 4) 1565. 47 As Vandenberg ‘The New Wal-Mart’ (n 4) 941 puts it, ‘on the surface, [private contracting would] fall outside the scope of what most would consider regulation or governance’. 48 ‘[R]egulation is often spoken of as if an identifiable and discrete mode of governmental activity’, see R Baldwin, M Cave and M Lodge (eds), Understanding Regulation: Theory, Strategy, and Practice (OUP 2012) 2. 49 B Morgan and K Yeung, An Introduction to Law and Regulation: Text and Materials (CUP 2007) 4. 50 J Black, ‘Decentering Regulation: Understanding the Role of Regulation and Self-Regulation in a Post-Regulatory World’ (2001) 54 Current Legal Problems 103. Further, Baldwin, Cave and Lodge (eds), Understanding Regulation (n 48) 2–3, note that ‘[r]egulation is often spoken of as if an identifiable and discrete mode of governmental activity’, but they note the evolution from this position over the last decade and point out that regulation may now include activities carried out by a host of non-state bodies, including private actors such as corporations. 51 Black ‘Decentering Regulation’ (n 50); J Black, ‘Critical Reflections on Regulation’ (2002) 27 Australian Journal of Legal Philosophy 1.

176  Lucinda Miller at stake if one is to accept that it should be customised for these regulatory ends. The disjunction between the two seems insuperable: contracts serve the interests of the parties, regulation serves the broader public good; contract law is facilitative, regulation steers conduct to achieve compliance with some desired state of affairs. But despite this incompatibility, it is clear that contracts will often have regulatory effects, and many contractual obligations could be described as having a regulatory dimension. For example, terms might set standards of safety or of the quality of the goods that are the subject of the transaction. In this way, one could say that they regulate the behaviour of the contractual party by requiring them to behave in certain ways in respect of their obligations to the buyer.52 In addition, contracts often have regulatory effects that are felt beyond those that formally make the agreement. The aggregative use of standard form contracts, for example, may have a significant regulatory effect on market conditions,53 even though the terms themselves only seek to regulate the bilateral relationship of the contractual parties. Since bilateralism is a defining feature of contract law, the law finds it difficult to adequately articulate the variety of ways in which contracts generate effects beyond the parties.54 Moreover, the discourse of contract tends to be refracted through the principle of freedom of contract, and this may explain why any regulatory phenomenon, either inter pares or having consequences for third parties, is usually obscured.55 The regulatory dimension of ethical contracting is distinctive and far more conspicuous, however, since it is not merely incidental to the normal activities of contracting. Instead, commercial parties deploy the contractual form to consciously and purposefully protect (‘regulate’) public interests; ethical clauses ‘reach into the public realm deliberately, and from the outset’.56 From this perspective, contracting parties are transformed into ‘private regulators’57 and contracts into regulatory instruments. Indeed, freedom of contract itself becomes a regulatory tool.58 This is why ethical contracting might be described as distinctly regulatory. And since this private form of regulation does not suffer from the same territorial and jurisdictional challenges as public forms of regulation, the contract seems an especially valuable tool for dealing with the transnational harms caused by global production processes. But paradoxically, although private initiative and

52 P Verbruggen, ‘Regulatory Governance by Contract: The Rise of Regulatory Standards in Commercial Contracts’ (2014) 35 Recht der Werkelijkheid, 79, 82. 53 eg, in the way that these contracts regulate access to goods and services (especially for consumers) and the effect that they have on pricing. 54 See, A Bagchi, ‘Other People’s Contracts’ (2015) 32 Yale Journal on Regulation 211. 55 D Leczykiewicz, ‘Private Regulation, Compliance and Reviewability of Contracts’ in Brownsword, van Gestel and Micklitz (eds), Contract and Regulation (n 18) 330. 56 McBarnet and Kurchiyan, ‘Corporate Social Responsibility’ (n 4) 60. 57 More specifically, the lead firm becomes the ‘regulator’, the supplier (as well as the third party – see section III) the ‘regulatee’. 58 Leczykiewicz, ‘Private Regulation’ (n 55) 331.

‘Ethical Clauses’ in Global Value Chain Contracts  177 individual autonomy provide contractual instruments with their regulatory potential, we have seen that these same principles set important limits to the value of ethical contracting as a regulatory tool. Freedom of contract endows parties with the authority to contract in altruistic as well as regulatory ways, but it also means that they may legitimately choose to exercise their autonomy in ways that do not serve public values or global regulatory needs, if this would better serve their self-interest. Whilst it is true that the public sphere may play a role in steering private choice in ways that enable both public and regulatory objectives (see section  II.A),59 commercial parties are under no legal obligation to contract in ethically appropriate ways, since this would conflict with the autonomy-based understanding of contract. Ultimately, therefore, regulatory goals and the public interest rest somewhat precariously on a fortuitous ­coincidence – when they happen to correspond with exercise of individual autonomy and private interest. The discussion now turns to consider whether, once parties have chosen to exercise their autonomy and incorporate ethical clauses into their contract, contract law can and should play a role in enforcing the clause. In doing so, we consider whether the public and regulatory features described thus far might give normative weight to a reorientation in our thinking about the role that contract law should play in this regard.

III.  The Enforcement of Ethical Clauses This section considers the role of contract law in the enforcement of ethical clauses by means of discussion of enforcement both by the lead firm and by third parties. The enforcement of contractual terms is a key function of contract law, and it is therefore important to analyse whether the distinctive nature of these clauses makes any particular demands on contract law. Moreover, and adopting a regulatory lens once more, the logic of regulatory theory is based on the effectiveness of regulation, and for this reason ensuring compliance with regulation is critical. Whilst the language of ‘effectiveness’ seems anomalous within the traditional discourse of contract law – effective for what end(s)? – an understanding of ethical contracting as private regulation means that we can conceptualise ‘regulatory compliance’ as ‘performance of the contractual obligation’. For this reason, ensuring the enforcement of ethical clauses becomes a significant factor for reinforcing the regulatory role that these clauses might play. We shall also see that enforcement has a bearing on the extent to which the state 59 Through a regulatory lens, we might describe the public law steering as ‘meta-regulation’, in other words, ‘ways that outside regulators deliberately – rather than unintentionally – seek to induce targets to develop their own internal, self-regulatory responses to public problems’: C Coglianese and E Mendelson, ‘Meta-Regulation and Self-Regulation’ in R Baldwin, M Cave and M Lodge (eds), The Oxford Handbook of Regulation (OUP 2010) 146, 150.

178  Lucinda Miller (and contract law is understood as ‘a particular form of state legality’60) is able to discipline and govern the transnational economic behaviour of commercial parties.

A.  Enforcement by the Lead Firm One of the difficulties with imagining contracts as a potential mechanism for private regulation is that whilst freedom of contract gives governance authority to private parties to contract in this regulatory, public-interest way, private law’s facilitative framework also means that there is freedom as to whether to enforce the clause on breach. Contract law does not require that the victim of a breach should sue. It simply sets out the procedural mechanisms through which a claim can be brought and the substantive rules that apply to a dispute, but the victim of the breach can choose to do nothing in response to breach. In other words, it is entirely appropriate for contractual parties to waive the obligation if this is in their private interest. This is key to the very distinctive interpersonal nature of private law, which grants individuals the authority to decide whether they wish to invoke the coercive power of the state to vindicate their infringed rights. In other words, whilst private law is a state institution just like public law, it does not engage in ‘top-down’ regulation but is an enabling state institution that can be, but need not be, used by private claimants. It is therefore evident that should a ‘Tier 1’ firm breach one of the ethical obligations in its contract with the lead firm, the lead firm is under no duty to respond to that breach through a private law remedial action. Indeed, in the particular context of a breach of ethical commitments, it will often be in the interests of the lead firm not to pursue a contractual claim against its supplier.61 There are a number of reasons for this. To begin with, the public nature of a legal challenge may draw attention to the ethical irregularities in the lead firm’s supply chain and inadvertently highlight the firm’s non-compliance with generally accepted standards of corporate social responsibility.62 This is particularly true for branded firms, since a tarnish to their brand reputation may be difficult to displace once it has seeped into public consciousness. But even where there is no brand image to preserve, firms will often lack motivation for litigation. Ethical and social obligations in GVC contracts tend to operate at the periphery of the commercial exchange and are generally not the primary reason for contracting – to return to the observations previously made, they are interests that lie beyond those of the

60 Wai, ‘Transnational Liftoff ’ (n 36) 45. 61 Rühmkorf, Corporate Social Responsibility (n 5) 123, does not find a single case on breach of ‘CSR’ terms in his study. 62 Ethical standards are generally not easily visible to the western consumer, and they tend to rely on non-compliance being rooted out by NGOs, etc. Litigation provides an obvious opportunity for NGOs to unearth salient details about the lead firm’s (un)ethical practices.

‘Ethical Clauses’ in Global Value Chain Contracts  179 contracting parties. The contract can still be performed and the parties retain a transactional interest in the agreement. Furthermore, the lead firm may be deploying ethical clauses purely as a ‘greenwashing’ tool, using them either as a technique for positively improving its CSR image, or in a way that shields it from reputational damage should problems later emerge in their supply chain. In either case, there is little interest in enforcing the ethical standard on breach. If parties choose not to adopt contract law to enforce ethical clauses, the regulatory role of these clauses is likely to be diminished.63 And the nature of private law as an enabling institution means that it is difficult to imagine how the institutions of contract law can compel contractual parties to enforce terms in their contract if this does not correspond to their economic interest. Any attempt to suggest ways in which contract law could do so, would mean seriously undermining contract law’s commitment to contractual freedom, and would end up distorting the very nature of private law as a facilitative institution at the service of the contractual parties. One response to these difficulties is to explore the possibilities for third-party enforcement, and we turn to this in section III.B. Before that, it seems worth considering whether contract law might be able to facilitate, or encourage, the lead firm’s enforcement of these clauses. One way to do this might be in a reorientation of some of the rules on remedies, since the reluctance to sue may well be exacerbated by the inappropriateness of the traditional remedies for breach of contract when applied to obligations of this kind. Specific performance seems an unlikely candidate for any such reform. Not only is it an exceptional remedy for common law jurisdictions,64 but it will often be an unsuitable remedy for breach of an ethical obligation because of its process-related nature. For example, in the context of a supplier’s breach of an obligation to provide a certain standard of working conditions for its employees, an order of specific performance would involve the court’s getting embroiled in sensitive matters of governance. Similarly, in the context of an obligation not to degradate the local water supply in the process of production, an order for specific performance might require complex corrective measures (some of which might necessitate a fundamental overhaul of production processes) along a number of links in the chain, and involve a range of actors in the chain who are not necessarily bound by the particular contract that has been breached. Moreover, it is difficult to imagine how a court situated in one jurisdiction will be able to ensure compliance with its order of specific performance from a defendant situated in a different jurisdiction. Instead, one area of reform might be a reorientation in the damages rules to take better account of the type of non-pecuniary, and public, interests of relevance here. As the law currently stands, the rules on damages for breach of contract, with their focus on the economic value of the commercial party’s promise, have 63 Non-legal enforcement mechanisms are discussed in section III.C. 64 And whilst the remedy is more available in civil law jurisdictions, nevertheless, commercial parties in these jurisdictions tend to seek damages awards.

180  Lucinda Miller little purchase when applied, for example, to the breach of a supplier’s promise to supply goods that have not been manufactured with child labour. Provided that the goods are otherwise compliant, it is difficult to shoehorn the breach into a claim for economic loss by the lead firm. A claim for loss of reputation may offer one possibility, but this type of claim is not only difficult to quantify but may also not be within the reasonable contemplation of the parties at the time of contracting. And so, in the absence of specifically-tailored terms stipulating agreed damages for breach of ethical clauses, the rules on damages are ill-suited for these types of obligations and may exacerbate the reluctance of the lead firm to pursue a claim on breach. In English contract law, there has been some attempt to pursue claims for damages that reflect the interest the claimant has in performance of the contractual obligation – the ‘performance interest’ – rather than the economic value of performance. But such an approach has not fared particularly well in the courts,65 and in any case seems conceptually unsuitable for the type of interests reflected in ethical clauses, since it remains aligned with the interests of the contractual party – albeit its interest in the promised performance itself, rather than its economic interest in performance. Moreover, when we remember that commercial parties are generally perceived as having less interest in the performance of ethical clauses than they do in conventional commercial obligations, a claim based on the ‘performance interest’ seems especially inappropriate in this context. More appropriate is to enquire whether an exceptional category of damages might be developed, one that is able to account for the losses suffered beyond the contractual relationship. Such an award could not be perceived in any sense as a ‘compensatory’ award but rather, perhaps, as some form of a ‘regulatory sanction’. Such a proposal is contentious, not least because it rubs up against the private law logic of ‘bilaterality’, a concept that expresses the bipolar encounter between contracting parties, and the view that private law relationships should be governed purely by criteria originating in the relationship between the parties themselves, rather than by reference to any external goals.66 It seems far too much of a distortion of contract law’s rationality. In any case, any such reorientation in the rules on damages is unlikely to make much practical impact on the lead firm’s incentive to sue. If commercial parties have any real interest in pursuing a claim for breach, they are sophisticated enough to design their contracts, including their damages regime, in ways that reflect recovery of this interest themselves. The ‘network-like’ nature of the GVC also poses obstacles for contractual enforcement of ethical obligations by the lead firm. Breach of the obligation is likely to occur beyond the boundaries of the particular contractual relationship 65 The extent to which English law protects the performance interest was directly at issue in the House of Lords decision in Alfred McAlpine Construction Ltd v Panatown Ltd [2001] 1 AC 518, but the ruling was not conclusive in this respect. 66 This view is captured by the notion of corrective justice, see E Wienrib, The Idea of Private Law (OUP 1995).

‘Ethical Clauses’ in Global Value Chain Contracts  181 between the lead firm and its direct (Tier 1) supplier, meaning that if the lead firm does wish to sue for breach, the obligation would first need to have been cascaded up the chain. The doctrine of privity, or l’effet relatif, means that the lead firm is unable to impose obligations on parties outside of the contractual relationship and so must find other ways to ensure the obligation is cascaded up the chain. One way is for the lead firm to insist that its Tier 1 supplier imposes the same ethical obligation in its contract with the Tier 2 supplier. But empirical evidence reveals that it is rare that ethical obligations are actually cascaded beyond the lead firm and the Tier 1 supplier,67 which may confirm not only the difficulty in doing so,68 but also, and perhaps more cynically, that the lead firm has little commercial motivation to design its contracts in this way. Even when the obligation has been cascaded down the chain, if there is breach of the ethical obligation (eg by a Tier 2 supplier), the doctrine of privity may prevent the lead firm from enforcing the obligation, unless it can show that it is a third-party beneficiary of the contract between the Tier 1 and Tier 2 supplier. It is perfectly possible for contractual parties to contract for this, and the lead firm may attempt to govern the contractual chain to this end, but it is uncommon, and its rarity attests to the issues already raised – namely, the difficulties in the lead firm’s contractual governance of the chain beyond its own contractual relationship, combined with its lack of motivation to ensure the effective enforcement of these clauses. The criticism that the institutions of contract law are not appropriately designed for network constellations is not a new one, and there has been a vivid debate about whether, and in what way, contract law should play a role in relation to these ‘systems’ of interdependent, yet normatively and factually separate inter-firm relationships.69 To the extent that contract law does not perform its facilitative role of enabling the creation and enforcement of obligations along the chain, this may be a reason in itself for considering some sort of normative reorientation in contract law.70 Whether the regulatory and public-interest features of ethical contracting might provide additional and more powerful justification for such a reorientation may be worth exploring, although it seems difficult to imagine how the institutions of contract law might be stretched to accommodate the imposition of obligations and liabilities in the chain when the parties are outside of any direct contractual relationship. To do so seems to threaten contract law’s self-understanding as an autonomy-based institution and may be a transformation too far. 67 In the context of environmental clauses, see Vandenberg, ‘The New Wal-Mart’ (n 4) 945. 68 The power dynamics along the chain may make this far more difficult that one might first assume; see, eg, O Ben-Shahar and J White, ‘Boilerplate and Economic Power in Auto Manufacturing Contracts’ (2006) 104 Michigan Law Review 953, for an interesting discussion on the dynamics of power along chains in the automotive industry. 69 H Collins, ‘The Weakest Link: Legal Implications of the Network Architecture of Supply Chains’ in M Amstutz and G Teubner (eds), Networks: Legal Issues of Multilateral Co-operation (Hart Publishing 2009); C Mitchell, ‘Network Commercial Relationships: What Role for Contract Law?’ in Brownsword, van Gestel and Micklitz (eds), Contract and Regulation (n 18) 198. 70 See Mitchell, ‘Network Commercial Relationships’ (n 69) 201.

182  Lucinda Miller In light of these obstacles related to enforcement of ethical clauses by the lead firm, the possibility for enforcement by third-party beneficiaries becomes particularly salient. It is to this issue that we now turn.

B.  Enforcement by Third Parties If we are to accept that ethical contracting is an important vehicle for the regulation of global public interests (section II.B), then third-party enforcement seems to be significant in this respect. Further, and again from a regulatory perspective, we might find it worrying that third parties are subject to the regulatory decision making of private actors but are deprived of any participatory rights concerning the nature of the regulatory regime to which they have been subjected. Providing third parties with the opportunity to be able to enforce ethical obligations of which they are the beneficiaries, gives these regulatees a ‘voice’ – a voice that has been silenced in the political domain by the shift in regulatory responsibility from the public to private sphere.71 And so, third-party enforcement may go some way in allowing individuals to defend their democratic entitlements and to challenge the regulatory power of transnational corporations. With this in mind, third-party enforcement may also be valuable in the way that it responds to the accountability deficit in this form of private regulation. Even if one is not attracted by this regulatory take on third-party enforcement, one might instead view a third-party claim as a crucial mechanism for embedding transnational private ordering in a legal and institutional framework. Contractual enforcement ensures that private economic power is tethered to a ‘specific sovereign regime of private law’72 and, as we have seen, if lead firms are not motivated to pursue claims, third-party claims may be valuable for anchoring transnational contracts in a domestic constitutional setting within which they can be governed. As we shall see in section III.C, it is for this reason that non-legal methods of enforcement are not considered ideal alternatives to the contractual enforcement of ethical clauses. Despite these normative arguments in favour of third-party enforcement, there are significant practical and doctrinal limitations. As regards the former, perhaps the most obvious relates to the resource implications for individuals (usually in the Global South) who do not have the financial means or knowledge to pursue their contractual rights, even in those instances where they detect that the contract does indeed bestow rights on them. In addition, it is often the case that supplier jurisdictions, in which the third party tends to be based, offer less protection and have weaker legal systems, and so any financial value in the claim necessitates moving the dispute to the legal jurisdiction of the lead firm. Herein lies another obstacle. Not only does a transnational claim increase the financial burden for the third party, but the lead firm (most notably in buyer-driven

71 Cafaggi 72 Wai,

and Pistor, ‘Regulatory Capabilities’ (n 20). ‘Transnational Liftoff ’ (n 36) 46.

‘Ethical Clauses’ in Global Value Chain Contracts  183 chains) will generally avoid imposing contractual obligations on itself, preferring to delegate obligations further up the chain. Such contractual arrangements are embedded in the commercial logic of the chain. For this reason, the third party will struggle to successfully argue that it is a beneficiary of any contractual promise of the lead firm. This was one of the obstacles to a successful claim in the Wal-Mart litigation in California,73 in which employees of Wal-Mart’s suppliers located in such countries as Nicaragua, China and Bangladesh, argued, inter alia, that they were the thirdparty beneficiaries of a duty by Wal-Mart to inspect the supplier factories in which they worked. Since, they alleged, Wal-Mart had inadequately monitored its suppliers, it was in breach of this duty. This argument was rejected, both at first instance and on appeal.74 Interpreting the contract between the supplier and Wal-Mart, the Court of Appeal held that the ‘language and structure of the agreement show that Wal-Mart reserved the right to inspect the suppliers, but did not adopt a duty to inspect them’.75 This meant that there was no breach of any promise of which the employees could claim they were a beneficiary. In the absence of any ambiguity in the language of the contract, our understanding of the rules of contractual interpretation makes it difficult to argue that the Court should have found that the contract imposed a legal obligation on Wal-Mart to inspect the factory. Nevertheless, such an approach seems to ignore the power and control that corporations often yield across transnational production processes, and thus over their suppliers. One might argue that, regardless of the contractual language, or (adopting Macaulay’s terminology) regardless of the ‘paper deal’,76 firms often exercise day-to-day control over suppliers’ behaviour, and thus over the workplace standards of their suppliers, and that the nature of the ‘real deal’77 is often such that lead firms do undertake duties in relation to the activities of their suppliers in the chain. It could be argued, therefore, that these factual aspects of the lead firm’s control should be reflected in its legal obligations and responsibilities.78 Such an argument has resonance with observations made previously concerning the network features of the chain; if the focus is on the factual control of suppliers then it could be argued that liability for failings in the chain should be imposed on the lead firm, even when the parties are outside of any direct contractual relationship. And yet, even if we were to pointedly ignore the controversial nature of these arguments in relation to their fit with the doctrines and traditional 73 Jane Doe and others v Wal-Mart Stores Inc, No 05-CV-7307-AG (ND Cal March 30, 2007); Jane Doe and others v Wal-Mart Stores, United States Court of Appeal for the 9th Circuit, 572 F 3d 677. 74 ibid. 75 Jane Doe and others v Wal-Mart Stores, United States Court of Appeal (n 73) 681, A, paras 2 et seq. 76 S Macaulay, ‘The Real Deal and the Paper Deal: Empirical Pictures of Relationships, Complexity and the Urge for Transparent Simple Rules’ (2003) 66 MLR 44. 77 ibid. 78 For discussion on this idea of factual control serving as a valid argument to create obligations on the side of the controlling entity, A Beckers, Enforcing Corporate Social Responsibility Codes: On Global Self-Regulation and National Private Law (Hart Publishing, 2015) 61.

184  Lucinda Miller understanding of contract law,79 it asks the courts to perform an unenviable task, for they would have to ascertain whether the precise nature of the control that lead firms exercise over their suppliers is such as to justify the imposition of legal liability. In this respect, we should note that the GVC literature80 has highlighted that there is a range of possible forms of contractual governance structures. On a continuum from least to most hierarchical, four different forms have been identified – market, modular, relational and captive – and it is only in the more hierarchical of constellations that the appropriate levels of control are exercised by the lead firm to justify legal responsibility over suppliers. The danger is that legal liability could be imposed on the lead firm in instances where the supplier has more leverage and control than is actually the case. In addition to the significant obstacles already noted, even if the third party could establish that it is a beneficiary of a promise by the lead firm, or where it pursues a claim against the supplier, there are other doctrinal limitations. Of course, each jurisdiction approaches third-party rights differently, but there are some general observations that can be made as to what would be required for a successful claim. First, the third party would need to show that the contracting parties intended the term to benefit it.81 Such a position is justified by contract law’s autonomy-based underpinning and the need to distinguish intended thirdparty benefits from the incidental benefits that third parties can accrue from a whole range of contracts. In any case, it seems reasonable to argue that all ethical clauses are intended by the contracting parties, at least in part, to confer benefits on third parties.82 For example, in the context of terms prohibiting child labour, or terms whereby a supplier warrants that it will allow workers to unionise or that it will meet the strict environmental standards imposed by the lead firm – the contracting parties must have intended that there be benefits conferred on third parties.83 And whilst we have seen that ethical clauses would only be incorporated if they benefited the economic interests of the contracting parties, it is 79 Of course, we might instead turn to tort law and explore whether the firm owes a duty of care towards the supplier’s workers. In this respect, the criteria in the case of Chandler v Cape [2012] EWCA Civ 525, in which a parent company was found liable for the acts of its subsidiary on the basis that it had ‘assumed responsibility’ for those acts, might be applied to the supply chain context. Such an extension was accepted, in principle, in the recent Canadian case of Das v George Weston Limited [2017] ONSC 4129, and although the claims were dismissed (as indeed were the tort claims in the Wal-Mart case itself), tortious rather than contractual claims certainly seem worth exploring. For a more detailed discussion, see V Ulfbeck and A Ehlers, ‘Direct and Vicarious Liability in Supply Chains’ in Ulfbeck, Andhov and Mitkidis (eds), Law and Responsible Supply Chain Management (n 16) 91. 80 Most notably, G Gereffi, J Humphrey and T Sturgeon, ‘The Governance of Global Value Chains’ (2005) 12 Review of International Political Economy 78. 81 Under English law, the Contract (Rights of Third Parties) Act 1999 sets out the circumstances in which a third party is able to enforce a contract. If, as is routine in GVC contracts, the contract does not expressly confer the right to enforce the term (s 1(1)(a)), the third party would need to show that the term ‘purports to confer a benefit’ on it (s 1(1)(b)). The third party would need to show that, on an objective interpretation of the clause, the contracting parties intended the term to benefit it. 82 In some jurisdictions, such as England, the benefit to the third party need not be the ‘predominant purpose or intent’: Prudential Assurance Co Ltd v Ayres [2007] EWHC 775 (Ch), [2007] 3 All ER 946. 83 We consider the specificity of third-party identification in a moment.

‘Ethical Clauses’ in Global Value Chain Contracts  185 inconceivable that a firm (at least a lead firm) would risk its reputation by openly defending a third-party claim against it with the argument that the clause’s only intended purpose was as a greenwashing technique.84 For this reason, party intention to benefit the third party seems easy to satisfy. However, problems arise in those (very typical) instances where the lead firm has explicitly excluded the right for the third party to enforce the term.85 Here, provided that the contractual term is expressed in clear language, the intention of the parties is clear. The organising principle of contract law – freedom of contract – means that, from the perspective of contract law, the use of these terms is a ‘legitimate’ use of contractual power. And yet it is a stark illustration of how the autonomy-based rationality of contract law sits awkwardly alongside a rationality infused with the public good. Ultimately, it allows private parties to deploy the contractual instrument strategically, and in a manner that erodes the public and regulatory interests protected by ethical clauses. One suggestion is to prohibit the use of such clauses.86 But whilst the publicinterest regulatory nature of these clauses might justify such interventions into party autonomy, it is difficult to imagine how such an approach would work in practice, since it would create complex distinctions between those types of commercial contract in which such clauses are permitted and those in which they are not. Another danger is that it may well reduce the attractiveness of ethical contracting, and may therefore prove counterproductive. We touch again on this point in the discussion below. Alternatively, and perhaps slightly less controversially, one might argue that these clauses could be ‘disabled’ in certain situations. For example, lead firms will often publicly declare that they contractualise ethical commitments with their suppliers. Indeed, such public statements may stem from domestic legislation concerning non-financial reporting on supply chain governance.87 Here, since the lead firm does not also make public that it has excluded third-party enforcement, this imbalance in public information may render disablement of the clause ­justifiable.88 Of course, this might mean that companies choose to disclose the fact 84 Under English law, where the term purports to confer a benefit, if the parties are to rebut the presumption that they did not intend the term to be enforceable by the third party, it would need to be shown that both parties did not intend the term to be enforceable (s 1(2) of the Contracts (Rights of Third Parties) Act 1999 mentions the intention of the ‘parties’, rather than ‘party’). 85 A typical illustration of this type of clause is taken from GlaxoSmithKline’s terms and conditions of purchase. Under the title ‘Third Party Rights’, s 27 states, ‘No person who is not a party to the Agreement … shall have any rights to enforce any Term or Condition of the Agreement’: available at https:// supplier.gsk.com/irj/go/km/docs/pccshrcontent/procurement_terms/gsk-philippines-purchaseorder-terms-and-conditions.pdf. See also examples in Rühmkorf, Corporate Social Responsibility (n 5) 105. 86 Rühmkorf, Corporate Social Responsibility (n 5) 211, recommends that third parties who are expressly identified as intended beneficiaries should have a right to enforce the CSR obligation and that contractual parties cannot exclude applicability of the Contracts (Rights of Third Parties) Act 1999. 87 Section 54(1) of the Modern Slavery Act 2015 is one such example. 88 For this argument, see K Mitkidis, ‘Enforcement of Sustainability Contractual Clauses in Supply Chains by Third Parties’ in Ulfbeck, Andhov and Mitkidis (eds), Law and Responsible Supply Chain Management (n 16) 79.

186  Lucinda Miller that they exclude third-party enforcement of clauses made in their favour, but they would have to balance this against the reputational costs associated with putting such information in the public domain. Finally, we come to the third party’s difficulty in being able to show that it can be identified as a beneficiary of the ethical clause with sufficient specificity. English law, for example, requires that the third party be ‘expressly identified in the contract by name, as a member of a class or as answering a particular description but need not be in existence when the contract is entered into’.89 Terms requiring that suppliers meet labour standards for their ‘employees’ would satisfy this subsection – an employee in a supplier factory should be able to show that he or she is ‘a member of a class’. But a broad term requiring the supplier to ‘ensure respect for human rights’ or to meet certain environmental standards might prove more troublesome if there is no specifically identifiable group of beneficiaries to which these standards can attach.90 We have seen that there are significant limits to the role that contract law can play in the enforcement of ethical clauses. Innovations have been proposed that might enhance the opportunities for private law enforcement of ethical clauses, but we have seen that these rub up against the facilitative nature of contract law. There are undoubtedly bolder, more interventionist proposals of reform that could be suggested. But we should be clear about the inherent risks when such innovations demand that contract law take on a more muscular regulatory role. For each time that contract law is ‘employed to pursue heteronomous objectives, there seems to be an imminent danger of intrusions into freedom of contract and private autonomy’.91 And any intervention that reduces party autonomy and increases the potential liability of the lead firm may have a chilling effect on the frequency with which commercial parties decide to incorporate ethical obligations into their contracts. As recent empirical research has shown, a more regulatory, interventionist approach from private law will have consequences for corporate strategy and for the self-regulatory behaviour of firms. Firms will react to private law intervention in a variety of ways and, in some instances, may end up abandoning their CSR commitments completely.92 Similarly, any reorientation in domestic contract law that affects the contractual freedom of transnational actors may reduce the attractiveness of the legal system itself – ‘parties are not confined to using the autonomy granted to them by the legal order; rather they have the autonomy to choose the very legal order that grants this autonomy’.93 And so an additional underlying tension is that the more that a legal system’s rules 89 Section 1(3) of the Contracts (Rights of Third Parties) Act 1999. 90 The third party will also find it difficult to show that such a vaguely-worded clause imposes a sufficiently precise obligation on one of the contractual parties. 91 F Möslein and K Riesenhuber, ‘Contract Governance: A Research Agenda’ (2009) 3 European Review of Contract Law 248, 276. 92 A Beckers, ‘Towards a Regulatory Private Law Approach for CSR Self-Regulation? The Effect of Private Law on Corporate CSR Strategies’ (2019) 27 European Review of Private Law 221. 93 Michaels and Jansen, ‘Private Law Beyond the State?’ (n 38) 867.

‘Ethical Clauses’ in Global Value Chain Contracts  187 of contract law take on a regulatory role, the less attractive that legal system may be to commercial parties. Ultimately, both these elements may well undermine the role that domestic contract law could otherwise play in its oversight of ethical global production processes.

C.  Non-legal Mechanisms of Enforcement Before concluding this chapter, it is important to briefly consider whether nonlegal mechanisms of compliance may compensate for the limitations of contract law that have been described. Even if clauses are not enforced though the institutions of contract law, this does not necessarily mean that the clauses remain unenforced, for there are various non-legal or private compliance mechanisms through which performance can be secured. The question then is whether the regulatory potential of ethical contracting might instead be supported by private compliance mechanisms and, if so, whether contract law’s limits need trouble us in any way. Rather than resort to contract law, ethical commitments can be enforced by the lead firm through a range of non-legal mechanisms. The most powerful of these is the lead firm’s ability to suspend performance, or to refuse to accept nonconforming goods when faced with the supplier’s breach and/or to refuse future business with that supplier unless it complies. Furthermore, the proper functioning of a GVC tends to be reliant on highly coordinated inter-firm relationships, with coordination between actors being especially important where ethical, process-related obligations feature in the chain.94 Coordination along the chain will tend to require far more cooperative contractual relationships, with firms working closely to foster relationships in ways that support supplier compliance. As we have seen, the lead firm is often granted contractual rights to monitor suppliers’ business activities closely, and will guide suppliers where things go wrong. And so, rather than pursuing penalties and legal enforcement through the courts, ‘softer’ methods, such as ‘continuous improvement’ programmes and periodic audits,95 are generally preferred as a way to bring the supplier back into compliance.96 In some sectors, lead firms will require suppliers to comply with a certain certification scheme. This means that certification bodies will monitor compliance through inspections and audits, and act as ‘key proxies for buyers to achieve regulatory compliance among contracted suppliers and thus ensure outof-court contract performance’.97 Again, the objective is to restore compliance by the supplier.98 94 Sobel-Read, ‘Global Value Chains’ (n 1) 373. 95 Audits will often be conducted by a third party, rather than by the lead firm itself. 96 McBarnet and Kurchiyan, ‘Corporate Social Responsibility’ (n 4)75. 97 P Verbruggen, ‘Private Regulatory Standards in Commercial Contracts: Questions of Compliance’ in Brownsword, van Gestel and Micklitz (eds), Contract and Regulation (n 18) 304. 98 ibid.

188  Lucinda Miller If compliance is being ensured through alternative means, one might imagine that the inability of contract law to enforce ethical commitments is not disastrous to the regulatory potential of these clauses. Further, this more collaborative approach to pursuing enforcement seems quite in keeping with the actual practice of business identified by Macaulay some time ago.99 Commercial parties will often ignore the formal contract, in both its planning as well as its enforcement, preferring to preserve the business relationship by negotiating solutions to disputes, rather than resorting to litigation through the courts. Moreover, relational theorists have shown that ‘contractual’ relationships are often a combination of enforceable and unenforceable (formal and informal) elements that work together to achieve a range of goals, such as risk allocation and governance.100 And so, for relationalists, ‘un(der)-enforceability is not necessarily a bug but a feature of contract’.101 However, we lose something quite important through contract law’s inability to play a key role in the enforcement of ethical contracts. To begin with, the underlying threat of litigation through the courts serves an important function in our understanding of how informal mechanisms of enforcement function; informal, private governance of the business relationship is conducted in the shadow of the formal state enforcement mechanisms that are able to step in as a last resort. This important interplay between non-legal and legal enforcement is undermined by the doctrinal limitations to the enforcement of ethical commitments noted in section III. Further, private compliance mechanisms are not politically neutral. As one study shows,102 these private regimes disproportionately favour the vested interests of corporations and allow those with power to gain strategic influence over audit design, implementation and enforcement. In this way, they may further reinforce the economic interests of the lead firm and fail to improve firms’ environmental and social performance in any meaningful way. For this reason, they give cause for concern about their ability to adequately protect public interests and to enhance the regulatory potential of ethical contracting. Finally, and perhaps most saliently, non-legal mechanisms of enforcement, such as those noted above, seem problematic when applied to obligations that touch on global public interests, and which may impact on individuals in such fundamental ways. Whilst it might seem appropriate for commercial parties to negotiate informal, relationship-sustaining solutions to disputes, this approach is 99 S Macaulay, ‘Non-contractual relations in business: A preliminary study’ (1963) 28 American Sociological Review 1. 100 In the context of rapidly innovating industries, it has been shown how a mixture of informal and formal mechanisms are deployed (‘braiding’) in GVC contracts to counter the risk of opportunism; see RJ Gilson, CF Sabel and RE Scott, ‘Braiding: The Interaction of Formal and Informal Contracting in Theory, Practice, and Doctrine’ (2010) 110 Columbia Law Review 1377. 101 JC Lipson, ‘Promising Justice: Contracts (As) Social Responsibility’ (2019) Wisconsin Law Review 1109, 1146. 102 G LeBaron, J Lister and P Dauvergne, ‘The New Gatekeeper: Ethical audits as a mechanism of global value chain governance’ in C Cutler and T Dietz (eds), The Politics of Private Transnational Governance by Contract (Routledge 2017) 97.

‘Ethical Clauses’ in Global Value Chain Contracts  189 far less justifiable when public and regulatory interests are at stake. Put simply, relational responses to non-compliance may conflict profoundly with the public (and third-party) interest in legal enforcement. This point returns us to the observation raised earlier; enforcement in a sovereign venue embeds the transnational regulatory activities of private actors within an institutional and legal setting, and by enabling contestation of transnational economic power, exerts an important disciplining force.103 This means that if we are concerned to prevent contracts becoming tools through which large areas of the global economy can be shielded from the disciplining force of law, contract law’s doctrinal limits seem troubling.

IV. Conclusion This study on ethical clauses has highlighted how private autonomy is being increasingly exercised in a way that re-casts contracts as tools for regulating the public harms generated by the processes of global production. The suggestion that contracts can perform a public-interest, regulatory role seems provocative since, as we have seen, it is at variance with our conventional understanding of contracts as vehicles for the pursuit of individualist, self-regarding interests. More profoundly, conceptualising contracts in this way raises difficult questions about whether contract law should respond and, if so, how. If we are to imagine contract law as taking on a more regulatory role, thereby supporting the public-interest, regulatory characteristics of ethical clauses104 and disciplining transnational economic power, this would require a significant reorientation in the facilitative underpinning of its institutions. On the one hand, this chapter has emphasised the distinctiveness of ethical clauses and, in doing so, has suggested that there might be good reason to rethink the contractual architecture’s claim to individual autonomy as its guiding ­principle.105 Indeed, ethical clauses might force a reconfiguration of the boundaries within which private agreements are traditionally thought to lie, forcing us to revise our resistance to state interference into (allegedly) private relations. In addition, the regulatory perspective with which this chapter has flirted might also be valuable for encouraging us to view contract law’s role differently; if we consider contractual agreements as private regulation, we can reconceptualise state interference into private agreement as state control of private regulation,106 meaning that we might be receptive to a more robust, governing role for contract law. Indeed, we

103 Wai, ‘Transnational Liftoff ’ (n 36). 104 We might describe such a ‘bold’ regulatory role for contract law as ‘public values welfarism’ – regulation aimed at giving contract law protection to interests and values that are not related to the parties; see T Wilhelmsson, ‘Varieties of Welfarism in European Contract Law’ (2004) 10 European Law Journal 712. 105 Bagchi, ‘At the Limits’ (n 15) 450–51. 106 For this suggestion see Leczykiewicz, ‘Private Regulation’ (n 55) 323.

190  Lucinda Miller have seen that such a move is normatively attractive in the way that it provides the state with an opportunity for disciplining transnational economic power. On the other hand, however, the discussion has revealed how difficult it is for contract law to reformulate its autonomy-based institutions in ways that are able to enhance ethical interests, without fundamentally distorting its features in the process. Moreover, any regulatory reorientation may well prove counterproductive. The fear of liability, or of a diminishment of contractual freedom, may mean that private parties will either refrain from ethical contracting in the first place, or will turn to other, more autonomy-enhancing jurisdictions for their governing law. Ultimately, we may need to accept that contract law may be an ill-suited instrument for governing contracts in this way. And so, by acknowledging the regulatory limits to ethical contracting, it is hoped that we find the allure of transnational private power less enticing as a one-stop solution for dealing with the harms from global production, and instead think imaginatively about how to fashion a more productive interplay between public and private spheres.

10 Smart Contracts SARAH GREEN AND ADAM SANITT

The technology that makes smart contracts possible1 was developed with a view to enabling transactions to be made end-to-end without the intervention of third parties, intermediaries, adjudicators or courts. In this sense, it achieves in principle complete freedom of interaction. Whether this is the same thing as freedom of contract, however, remains to be seen. It is not yet clear, for example, which smart contracts will be legally enforceable, either because the parties do not want them to be, and/or because the courts do not recognise them as being so. What seems inevitable at this stage in the development of smart contract technology is that conventional contract law in its current form is unlikely to be the most effective way of adjudicating smart contract disputes. One reason for this is that securing performance will be far less of a problem under smart contracts than it is in relation to conventional contracts: the automated nature of the former means that actions are far more likely to be executed than those promised in the traditional way, albeit that their results might not accord with the parties’ expectations. Any issues are therefore far more likely to arise (or at least to be brought to a court’s attention) after a transaction has occurred. Automated execution means that parties are free to determine the contents of their agreements and that machines will abide by those agreed instructions. The way in which smart contracts operate, therefore, means that any adjudication of them is likely to need to emphasise restorative rather than enforcement remedies. The extent to which the law chooses to do this will effectively determine how free smart technology users are to make legally recognised contracts. In order to make clear how and why smart contracts differ so markedly from conventional contracts, it is worth setting out a brief description of the relevant technology before turning to its legal analysis.

I.  What is a Smart Contract? Smart contracts are self-executing transactions, written in computer code and performed in an automated way. They have been made possible by a particular

1 Which

is distributed ledger technology or DLT, explained in section I.A.

192  Sarah Green and Adam Sanitt series of technological developments that have happened over the past decade. These developments have enabled decentralisation and consensus to replace intermediaries in transactions between contracting parties. In other words, contracting parties can now deal directly with one another without banks, without offline registries and, potentially, without lawyers.

A.  Decentralisation and Distributed Consensus Decentralisation and distributed consensus are the components that distinguish smart transactions from those that use intermediaries and electronic bank money, such as Paypal, WorldPay and BACS. These components also explain why smart contracts are often described as ‘trustless’, meaning that transacting parties need not trust each other in the real world, so long as they trust the payment protocol (which, for reasons that will soon become apparent, they probably should). Decentralisation is probably easiest to explain in the context of Bitcoin and its Blockchain (Bitcoin being the cryptocurrency and Blockchain the distributed ledger), although there are other platforms, such as Ethereum, on which smart contracts could be made. Decentralisation is achieved through Distributed Ledger Technology (DLT). This means that everyone who wants to contract using a particular piece of software has, as an integral part of that software, a copy of a ledger. This ledger is a record of every transaction made using that protocol, and each computer operating the software (known as a node) has a copy of the entire thing: from the beginning (the ‘Genesis Block’) to today’s latest block. Blockchain, which was created to underpin Bitcoin, was the first distributed ledger, but there are now distributed ledgers of several different forms. Common to every one, however, is the idea that all participants have access to the full history of transactions made using that protocol. This is a novel way of dealing with the age-old double spend problem. Historically, the challenge of how to prevent double spending has been met in two ways: the first is by using physical tokens, whose corporeal form physically prevents their being spent more than once; and the second is by employing an independent third party, such as a bank, to keep a record of transactions and their effects on the subsequent spending power of the parties involved. The same thing is achieved by DLT, by sharing information with every user and by ensuring that the information so shared is perfectly synchronised. This way, ‘coins’ cannot be spent twice because everyone would know that this is what was being attempted, and the consensus necessary for validation and recording would not be reached. Security is thus achieved through complete transparency, and distributed ledgers have no need for any centralised record-keeping, nor for any third-party intermediary to verify the integrity of transactions. In other words, ‘total validation replaces central control’.2

2 H

Diedrich, Ethereum (Wildfire Publishing 2016) 113.

Smart Contracts  193 Such transparency is achieved through what is known as distributed consensus protocol, and this is characterised by two features: (a) All computers on the network (referred to as nodes) must agree on which transaction data are ultimately recorded on the ledger. (b) The transaction data must have been generated by an honest node3 The question remains how any of this can be achieved. One method, used by Bitcoin, is known as proof-of-work, and this allows nodes to reach a consensus on which transactions to record and in which order to do so. (The order is of course all-important, as it is with any spending pattern, since what you have already spent determines how much you can spend in the future.) Proof-of-work is the means by which nodes persuade other nodes that the block of transactions they wish to add to the chain is legitimate and should be trusted. The work involved here is cryptography – the solving of a mathematical puzzle – and this puzzle is of a very specific type; the optimal way of solving it is to work through very large numbers of trial and error iterations. In other words, lawyers might say that it is a difficult case, but not a hard one. It is clear what needs to be done, but it takes an immense amount of computational power4 to work through the many repetitions of the same calculation, each time trying a different input. The puzzle is known as a hash puzzle, and success requires finding the input necessary for a function to produce a specified output. The hash function itself, which is the crux of this whole process, could be described in non-mathematical terms as a function that takes an input and produces an output that will look nothing whatsoever like the input.5 In fact, it is practically impossible to discover the original input simply by looking at the output, unless you know the hash function. Diedrich has provided the following illustration of how hashes work:6 There are different hash algorithms. One hash algorithm that was once popular was called MD5. For example, this text: MD5’s designer Ron Rivest has stated ‘md5 and sha1 are both clearly broken (in terms of collision resistance)’. So MD5 should be avoided when creating new protocols, or implementing protocols with better options. SHA256 and SHA512 are better options as they have been more resilient to attacks (as of 2009). has the MD5 hash: ed7f56281b8d079ff101009105f75b44

3 For a more detailed and technical description of this, see A Narayanan et al, Bitcoin and Cryptocurrency Technologies: A Comprehensive Introduction (Princeton University Press 2016) 28–50. 4 Around $300 million a year on Bitcoin alone, which is comparable to the annual electricity bill for the whole of Ireland. See Diedrich, Ethereum (n 2) 150. 5 In the Bitcoin protocol, the hash function takes an input of any length and always produces an output of a fixed length. 6 For present purposes, the substantive content of the text being hashed is irrelevant.

194  Sarah Green and Adam Sanitt … The hash ed7f56281b8d079ff101009105f75b44 can be used as a unique id for the above text, for two reasons: 1. 2.

If any punctuation or word was changed, you’d get a different hash for it. You will not for your life be able to find another text that results into exactly this hash.7

Crucially, the same hash function, applied to identical inputs, will always produce the same output. Applying the same hash function, however, to an input that is different from the original in only a miniscule detail will nonetheless result in a completely different output. This explains why the only way to solve such puzzles is through repeated trial and error: there are no shortcuts or clues provided by previous work, and the results of previous attempts are not in any way cumulatively helpful. So, Sherlock Holmes could not take the approach he took to the Dancing Men. Instead, nodes who wish to add a block to the chain first take a set of transactions that have been broadcast8 across the network. They then add to this something called a blockhash. A blockhash is the output generated when the hash function is applied to the whole of the previous block in the chain. This blockhash serves therefore as a unique identifier of all the data in that previous block, and so is what links the blocks (hence the term ‘chain’) and at the same time makes the protocol so secure: embedding this unique summary of the previous block’s data in the following block, and therefore in all future blocks,9 makes the blockchain practically tamper-proof. If any node wants to alter or modify the content of any given block, it would also have to rehash all of the blocks that follow it, and compete to persuade all nodes (who would be working on the basis that the original data were correct) that the new reality was the right one. This is so difficult, particularly in terms of computational power, as to be well-nigh impossible in the current environment. So, to return to the proposing node, which has taken the set of transactions it wishes to verify, and added the blockhash.10 The node then starts the real work: it also adds a value as the mystery input, which is known as a ‘nonce’ in cryptography. It then hashes all of this together (which means applying a specific and consistent hash function, which in the case of Bitcoin is called SHA256) and keeps doing this,

7 Diedrich, Ethereum (n 2) 106–07. Although, as Diedrich makes clear, it is theoretically, if not practically, possible to find another text that results in exactly this hash. This practical impossibility is considered to provide sufficient security. 8 This means, for example, that Alice has purported to transfer a certain amount of bitcoin to Bob by entering the details of his public key (which functions rather like an email address) and issuing instructions for the network to transfer that amount from her stock of bitcoin to his. It would be foolhardy, however, for any payee to trust the integrity of any transaction at this stage, because it has not yet been validated or added to the blockchain. 9 And, by definition, the previous block will be similarly linked to all the blocks preceding it. 10 ‘Adding’ here takes the form of concatenation of data strings.

Smart Contracts  195 using a different nonce each time, until it gets an output with the ‘correct’ number of leading zeros.11 At the heart of this scheme is the fact that a nonce is hard to find but easy to verify. While it routinely takes trillions of trial-and-error calculations to find it, it requires but one calculation to verify it: to test that a nonce is in fact resulting into a hash with the required number of leading zeros when added to the block data that it was found for.12

This is known as ‘trapdoor’ maths because of its one-way structure; it is very difficult to find the answer, but once you have the answer, it is very easy to verify that it is the right one. A simple analogy can be made with the combination to a briefcase lock. Anyone wishing to open the briefcase without knowing the combination will have to try (probably) very many attempted combinations to find the one that opens the case. Once she has found it, however, it is very easy for someone else to verify that she has found the correct combination: the verifier simply needs to put that combination into the briefcase dial and see if it opens. The result is a binary one – pass or fail. Similarly, once a node has found what it believes to be the right nonce, the other nodes verify that it is correct by simply applying it to the publiclyavailable data. If this produces the same output (the one with the required number of leading zeros), verification is complete. Once this happens, the proposed block gets added to the chain and the transactions in it get confirmed.13 This, however, is not the only thing that happens when the block gets verified. Another of DLT’s revolutionary qualities is its alignment of self-interest with altruism. Verifying blocks is hard grind and very expensive in computational terms, and yet it is essential to the continuation and security of the system. So, when a node successfully adds a block to the chain, it gets rewarded with bitcoin. In the Bitcoin protocol, the verification process is known as mining and is simultaneously the means by which new coin is minted.14 This is a system, therefore, in which self-interest works in favour of the collective interest, and the two are mutually reinforcing. Not all DLT systems use proof-of-work, and not all reward nodes with cryptocurrency, but some sort of consensus is required to maintain the ledger. This consensus may be limited to approval by an entity or group of entities, known as notaries, nominated by the parties to a transaction, with the DLT system providing only a secure messaging protocol to communicate that consensus. A more distributed approach is achieved by DLT systems that dispense with a blockchain and instead require transactions to be approved by groups of later transactions forming a web carefully designed to avoid circularity (known as a

11 The ‘correct’ number is dictated by the protocol and it does not matter what it is – it is just a test for the nodes to prove the work they have done. 12 Diedrich, Ethereum (n 2) 149. 13 To a certain extent anyway. There is a possibility of forking. 14 There is, however, only a finite number of bitcoin available: 21 million.

196  Sarah Green and Adam Sanitt directed acyclic graph, or DAG). Other variants keep the blockchain but opt for consensus protocols other than proof-of-work. To the extent that DLT systems employ distributed consensus as the means of achieving decentralisation, they are all able to facilitate direct contracting between parties without needing to involve intermediaries. In practical terms, smart contracts differ from conventional contracts in the following ways: • The entire lifecycle of a smart contract, from formulation to execution, occurs online. • That lifecycle need not at any point involve any entity other than the contracting parties (since, for instance, payment can occur directly from bitcoin wallet to bitcoin wallet). • Performance is automated and is carried out by the machine, following its programmed instructions. • Smart contracts are immutably recorded and programmed on the distributed ledger and cannot be altered. If ‘rectification’ is required, a new contract must be written that reverses or modifies the effects of the previous contract. • A smart contract can only perform in a situation in which there is an ‘algorithmically determinable solution’. In other words, there is no room in a smart contract for discretion, reasonableness or judgement: potential outcomes will be binary in form. Smart contracts, therefore, will only be useful for transactions of a certain type. An obvious example of the type of transaction to which smart contracts would be well-suited is the transfer of title to assets. Such transactions often take the form of requiring title to be transferred on the receipt of a certain amount of funds into a particular account, perhaps with the added requirement of verifying the absence of any charges over the asset.15 According to the logic of machines, this is simply an action to be performed on the satisfaction of a given condition or set of conditions: ideal for a self-executing algorithm. Machines are arguably better placed than humans to perform such one-dimensional tasks, since, properly programmed, they are simply unable to refuse to act, to omit a condition or to fail to perform. Smart contracts offer, therefore, the potential for reducing transactions costs, defective performance and informational asymmetry between parties. (The potential for defective performance does not, of course, disappear; it shifts its locus to the coder responsible for programming the machine, but that is an issue for another time and place.)

15 Mischcon de Reya and HM Land Registry have very recently (April 2019) trialled the UK’s first end-to-end residential property transactions.

Smart Contracts  197

II.  Formation: The Spectrum of Smart Contracts in Practice A lawyer and a coder will not necessarily have the same conception of a smart contract. The term ‘smart contract’ was coined by Nick Szabo, and was defined by him as a computerized transaction protocol that executes the terms of a contract. The general objectives of smart contract design are to satisfy common contractual conditions (such as payment terms, liens, confidentiality, and even enforcement), minimize exceptions both malicious and accidental, and minimize the need for trusted intermediaries. Related economic goals include lowering fraud loss, arbitration and enforcement costs, and other transaction costs.16

This use of the ‘c’ word would probably not satisfy a lawyer. At the very least, however, there will be a subset of those transactions referred to by users as smart contracts that will also satisfy the legal criteria of enforceable contracts, such as those that adhere to the legal rules of formation. There may be more to a smart contract than an independent chunk of selfexecuting code. At one end of the spectrum, a conventional natural language contract may include a short section of code that implements a single clause. This is similar in effect to the inclusion by reference in a contract of a formula or a specific manufacturing design. The courts have interpreted these references strictly – requiring them to be followed precisely while simultaneously complying with all other obligations in the contract, even ones that are apparently contradictory: Where a contract contains terms which require an item (i) which is to be produced in accordance with a prescribed design, and (ii) which, when provided, will comply with prescribed criteria, and literal conformity with the prescribed design will inevitably result in the product falling short of one or more of the prescribed criteria, it by no means follows that the two terms are mutually inconsistent.17

The effect appears to be that a counterparty cannot satisfy its obligations merely by executing a piece of code, even where that code is explicitly agreed in the contract. Other contractual terms may be used to show that the counterparty had to do more. The code provides convenience but not necessarily certainty. The harshness of this approach is illustrated in Hojgaard v EON, from which the extract above is taken. In a contract to build an offshore wind farm, both parties agreed to use an international standard for the design of wind turbines. Unknown to them, one formula in the standard contained a constant with a misplaced decimal point. When the defect came to light, the builder was held liable for breach

16 N Szabo, ‘Smart Contracts’ (1994) available at http://www.fon.hum.uva.nl/rob/Courses/ InformationInSpeech/CDROM/Literature/LOTwinterschool2006/szabo.best.vwh.net/smart. contracts.html. 17 Hojgaard v EON [2017] UKSC 59, [2018] 2 All ER 22 [44] (Lord Neuberger).

198  Sarah Green and Adam Sanitt of warranties contained elsewhere in the documentation – its exact compliance with the design specification was not sufficient. Applied to smart contracts, this suggests that a piece of code included in a natural language contract will not be interpreted in isolation, and executing the code may not be sufficient to comply with the contract. Further along the spectrum is the combination of self-executing code and natural language contract known as the ‘Ricardian contract’.18 This consists of a piece of code contained in a DLT, which contains a hashed reference to a natural language contract. As described in section I.A, a hash is a cryptographically secure method of creating a reference to a particular document. Anybody with the natural language document can easily check that it corresponds to the hash in the code. And changing even a single comma in the document would lead to its producing a completely different hash. So this mechanism inextricably links the code and the written document into a single agreement. A Ricardian contract could be used simply to implement the single formula in code considered in section I.A. In practice, it is generally employed to supplement a contract that is primarily smart, by adding a few natural language provisions. Popular options for natural language clauses ‘bolted on’ to a smart contract in this way include governing-law and jurisdiction clauses and dispute resolution mechanisms. Furthest along the spectrum is a smart contract that consists only of code, and which expressly excludes any other language and all contractual oversight. This is achieved through an apparently paradoxical method. The smart contract includes a natural language provision stating expressly that it is not a legally enforceable contract: an explicit denial of an intention to create legal relations. This type of smart contract is for parties who wish to rely entirely on the DLT to carry out their intentions and to remove any possibility of the legal system’s stepping in to reverse their immutable transactions. The motivation for this may be ideological – parties might be transacting on a DLT in order to remove themselves from the sphere of influence of a particular legal system or to facilitate cross-border commerce when there is no single legal system to which they will agree to submit. Although it may seem far-fetched to lawyers and legal academics, this ideology is common in the communities from which DLT emerged. Indeed, Bitcoin is often portrayed as a means of payment that is beyond the reach of legal oversight. Whilst this has led in the past to negative publicity, centred on the technology’s involvement in criminal activity, it is a fact that many legal systems are expensive, inaccessible or corrupt, and many fiat currencies subject to bouts of inflation. There are, therefore, legitimate reasons why people seek to exclude the apparatus of national law. The concept of an intention to create legal relations in the context of smart contracting is examined further in section III.

18 I Grigg, ‘The Ricardian Contract’ (2004) WEC ‘04 Proceedings of the First IEEE International Workshop on Electronic Contracting 25–31.

Smart Contracts  199

III.  Code as Contract The technology that makes smart contracts possible, and perhaps more significantly the ethos underlying that technology, gives rise to a fundamental question: to what extent can and should the law govern smart, peer-to-peer interaction? The original Blockchain protocol was, after all, created with a view to eschewing conventional third-party intervention and regulation, explicitly substituting a model by which all users participate in complete self-governance. Although it is neither realistic nor practical to suggest that total internal governance should prevail, given the now extensive uses made of DLT, a question nonetheless remains about the extent to which those engaging in smart transactions could expect to have their interactions regulated by the protocol, rather than by external agencies such as courts, arbitrators and lawyers. Internal regulation and resolution is certainly possible as a result of DLT technology: this could take the form of coded responses to certain conditions, or a requirement for multi-signature approval for the transfer of funds. In the latter situation, a third party would need to verify that, for instance, payment conditions had been met before releasing funds to either party. Such an arrangement could perform the role played in conventional transactions by escrow or trust arrangements. In forensic terms, this amounts to a question of whether the parties have demonstrated the necessary intention to create legal relations. Such an intention is, of course, a prerequisite for an agreement to be the source of legal rights and obligations and to be enforceable by the courts. Most problems concerning an intention to create legal relations arise from an agreement that features a level of informality or vagueness that precludes the court from applying the machinery of contract law to it, disappointing one of the parties to the agreement who cannot enforce it. Smart contracts feature the converse situation: an agreement that is precise and intended to be carried out on commercial terms, but which the parties nevertheless do not wish to be enforceable through the courts. The self-executing nature of smart contracts mean that this is not a contradictory position, but it is one nonetheless that the courts have not previously addressed. Take a Ricardian contract, expressed partly in code and partly in prose, where the wording of the prose section includes a statement that the parties do not intend the agreement to create legal relations.19 The intuitive starting point is that this statement is inconsistent with an intention to create legal relations and so no contract is formed. That starting point is not necessarily correct. The courts will look at the entirety of the agreement to determine the objective intention of the 19 Where the smart contract does not contain an express denial of an intention to create legal relations, the courts could still consider whether a lack of intention is implied by the conduct of the parties. Here, the presumption that commercial parties intend to create legal relations will weigh heavily. Only if the transaction takes place in a market where lack of intention to create legal relations is a sustained and universal principle might this implication be satisfied. It is possible that one day DLT will constitute such a market.

200  Sarah Green and Adam Sanitt parties, and they will apply a strong presumption that a commercial arrangement is intended to be legally binding: Whether there is a binding contract between the parties and, if so, upon what terms depends upon what they have agreed. It depends not upon their subjective state of mind, but upon a consideration of what was communicated between them by words or conduct, and whether that leads objectively to a conclusion that they intended to create legal relations and had agreed upon all the terms which they regarded or the law requires as essential for the formation of legally binding relations.20

Applying this principle, the Supreme Court held that a statement that an agreement was ‘subject to contract’ did not prevent it from forming a contract when the subsequent conduct of the parties showed that they intended to treat the terms as binding. In particular, the fact that the operative terms of the agreement were actually carried out by the parties was a ‘very relevant factor’21 pointing towards enforceability. The analogy between a ‘subject to contract’ statement and conduct carrying out the terms of the agreement, and a ‘no intention to create legal relations’ statement and self-executing code that has or will carry out the agreement is a close one. It suggests that the courts, looking at the entirety of the relationship between the parties, could impose a legally binding contract on a Ricardian smart contract despite an express statement of lack of intention to create legal relations. The analogy is not, however, an exact one. ‘Subject to contract’ implies that the parties consider that at some future point a contract may come into existence, and the relevance of conduct is to determine whether that point has been reached. A straightforward denial of intention to create legal relations makes it harder to imply that intention from later conduct. In RTS Flexible Systems v Molkerei Alois Muller,22 this was a policy decision for the Court: should the apparatus of legal enforceability be available to this particular agreement? Where a commercial arrangement had actually been carried out and the price paid, the Supreme Court was reluctant to deprive the parties of the benefit of this apparatus. Other similar situations show that the courts tend to grant the parties the benefit of this legal apparatus in commercial situations. In its most recent consideration of this issue, Wells v Devani,23 the Supreme Court enforced an agreement to pay an estate agent commission on his successful sale of a block of flats. The express terms of the written agreement were too vague to be enforceable, because the trigger for payment of commission had not been specified. Acting with some innovation, the Court implied a term into the agreement to resolve this uncertainty and then applied the test of intention to create legal relations to the agreement including the implied term. There was also an explicit policy element: estate agent agreements 20 RTS Flexible Systems Ltd v Molkerei Alois Muller GmbH & Co KG [2010] UKSC 14, [2010] 2 All ER (Comm) 97. 21 ibid [54]. 22 ibid. 23 Wells v Devani [2019] UKSC 4, [2019] 2 WLR 617.

Smart Contracts  201 were subject to statutory control and could be unenforceable without proper formalities: the Court further exercised its discretion to allow the agreement to be enforced. Other policy-based examples can be provided. One is an express agreement by the parties to submit disputes to arbitration. This is an ouster only of enforcement by the courts rather than enforcement generally, yet it is subject to statutory limitations. Agreements for arbitration for remedies below a fixed sum are deemed unfair by the Arbitration Act 1996.24 Above this sum, an arbitration agreement may still be deemed unfair by the Unfair Terms in Consumer Contracts Regulations 1999 (SI 1999/2083)25 and now the Consumer Rights Act 2015. The unfairness lies in the consumer’s difficulty in obtaining effective and inexpensive enforcement of their agreement – a policy that would apply equally to smart contracts. Finally, where parties have used contractual estoppel to restrict their contractual liability, the courts have used statutory powers to strike this down as an unreasonable exclusion clause. For instance, parties may agree between themselves that a particular state of affairs exists – such as that neither party has made any representations or relied on the other for advice – thereby removing potential liability for misrepresentation or negligent advice. The Court of Appeal has carefully differentiated the ability of parties to do this from the power of the courts to interfere on the basis of policy. It is open to the parties to agree on a state of affairs and to be bound by it, but this must be subject to statutory control, which will be applied according to its policy, whatever legal formalities parties use to try and avoid it.26 These are no more than broad analogies with the smart contract issues dealt with in this chapter. An intention to avoid all legal consequences is not the same as an intention to avoid the courts by recourse to arbitration; an intention to exclude liability in a contract is not the same as an intention to exclude all liability by preventing an agreement from being a contract. All the policy arguments exhibit a certain circularity: if there is a contract, statutory controls apply and show that a contract exists; if there is not a contract then the agreement is outside the scope of these controls and cannot show that a contract exists. Nevertheless, it is clear from these examples that there are constraints on limiting judicial oversight, that those constraints are circumscribed by policy considerations and that attempts artificially to stay outside their perimeter are often unsuccessful. The conclusion on this point is that an express statement in the prose section of a Ricardian smart contract that it is not intended to create legal relations will not automatically prevent a legal contract from being recognised. Whether it will

24 Arbitration Act 1996, s 91. 25 As occurred in Mylcrist Builders v Buck [2008] EWHC 2172 (TCC), [2009] 2 All ER (Comm) 259. 26 See First Tower Trustees Ltd v CDS (Superstores International) Ltd [2018] EWCA Civ 1396, [2019] 1 WLR 637, considering the anti-exclusion provisions of s 3 of the Misrepresentation Act 1967 in relation to such a clause, commonly labelled a ‘basis clause’. ‘Section 3 of the 1967 Act must be interpreted so as to give effect to its evident policy’ (Leggatt LJ at [51]).

202  Sarah Green and Adam Sanitt do so ultimately comes down to fundamental questions of contract law policy. Just as DLT and smart contracts challenge the existing financial order, they also pose significant challenges to the legal order. Contract law is a centrally available means of enforcing agreements between parties, without which commercial relations would be either more costly or simply ineffective. The smart contract ecosystem proposes an alternative – enforcement by external parties made redundant through automatic execution. Assumption of the benefits and restrictions of contract law as a whole is meant to be voluntary, as shown by the need for an intention to create legal relations, so why should parties not be able to ignore it completely? The answer may be that they can. But when there is an imbalance in negotiating powers or one party is acting fraudulently, or the arrangement is tainted by illegality, it is hard to see the courts stepping back and pronouncing themselves powerless. Before the creation of DLT and smart contracts, the issues in this section would have been purely theoretical. It would have been nonsensical to enter into a commercial transaction and simultaneously remove all hope of legal enforcement. Smart contracts offer an alternative: a commercial transaction that enforces itself. Whether the law still has a role to play is a question that courts will at some point have to address.

IV. Interpretation For current purposes, what is important is that the smart contract, as programmed, will be performed on the satisfaction of a given condition or set of conditions. This means that the space between the formation and the performance of smart contracts is vanishingly small. Given that most conventional contract claims are about performance, this represents a major difference in the adjudicative requirements of each type of contract. Smart contracts are less like exchanged promises than they are like promised exchanges. The expectations of smart contracting parties are more likely to be frustrated by the wrong thing happening rather than by nothing happening at all. For those smart agreements that are deemed to be enforceable, therefore, the next legal question that will arise most often is not whether performance can be enforced but whether the outcome that has already occurred is a just one. It is almost as if smart contracts need a back-to-front legal contractual analysis, in which enforcement yields to correction. Smart contracts do not only make interpretation harder – they also call into question the very concept of interpretation, as it is currently understood. As discussed, interpretation is the process of divining the meaning the parties intended from the words they used. Meaning is used here in a specific way. It does not refer to all and any objective meaning that may be taken from any passage of prose. Only the part of the meaning that creates legal rights and obligations is relevant, and a large part of this meaning corresponds to the list of actions the

Smart Contracts  203 text imposes on each party. Specifically, the process of contractual interpretation determines the actions that each party will undertake. Legal rights and obligations are created that correspond to these actions. In other words, the parties are legally obliged to do what they promise they are going to do. As the Supreme Court put it, the task is to ‘ascertain the objective meaning of the language which the parties have used to express their agreement’.27 It is the meaning of ‘their agreement’, what they have promised to do, which is the focus. Contrast this with an agreement concluded by conduct: it can be as simple as the purchase of an item in a shop, an exchange of the item for cash. There is effectively no room for interpretation as to the key promises entered into by each party. The promise of the buyer is to hand over the amount of cash he actually hands over; the promise of the seller is to hand over the item she actually hands over. The contract attaches to the actions of the parties rather than to their words. The actions necessarily coincide with the obligations they undertake. Smart contracts created entirely in code behave more like conduct than words. The words of the code are, in themselves, actions. They are performative utterances, like the words ‘I do’ in a marriage ceremony. They actually have an effect on the real world, not through interpretation of their meaning but directly. In the same way, code, when it is part of a distributed ledger, acts directly on the world through its execution without any medium of interpretation. To work out what a piece of code means, you do not read it, you run it. There is no room for doubt that the legal rights and obligations intended to be created correspond to the actions undertaken by the code; the code is no more than those actions set down in frozen form. The obligations coincide with the code in the same way that the obligations coincide with the conduct in a contract entered into by conduct. Interpretation appears in this context to be superfluous. This chain of reasoning is part of the sentiment encapsulated in the slogan ‘the code is the contract’. Often heard in the smart contract community, this refers to a stance as much political as legal, in which all enforcement by the courts, not just the doctrine of interpretation, is rejected. Rendered into a legal argument, it has some force when limited to the potential superfluity of interpretation, but it is unconvincing as an overall attempt to oust legal oversight. It ignores vitiating factors in agreements, such as fraud and illegality, and it ignores statutory and general policy controls on voluntary agreements. Even restricted to interpretation, it ignores the fact that most agreements will contain both code and prose. They may be expressed as Ricardian contracts, or they may take effect within the context of an existing commercial relationship, with numerous other documents, emails and actions forming part of the factual-matrix within which to interpret the agreement. Additionally, a piece of code requires a platform on which to run. This platform includes software intrinsic to the particular blockchain, extending from interfaces seen by the user through to the plumbing that

27 Wood

v Capita [2017] UKSC 24, [2017] 4 All ER 615 [13] (Lord Hodge).

204  Sarah Green and Adam Sanitt underlies the system. And it extends to other software with which the platform interacts, such as third-party information providers and cloud-based storage. It is impossible to model and to predict all of these interactions, not least because many depend on real-time constraints. Characterising code as freestanding, selfexecuting pieces of frozen conduct ignores these interactions and dependencies. Despite these difficulties, ‘the code is the contract’ ethos illustrates how smart contracts can circumvent both interpretation and enforcement as forms of writing that act on the real world via DLT. It is a slogan that should serve as a warning to the legal community. Legal practitioners and courts will have to determine how, and whether, legal doctrines such as interpretation are relevant to smart contracts. In those instances in which judicial interpretation does become relevant to smart contracts, (where, for instance, the code’s operation is said by one of the parties not to have done what it was intended to do), it will be one of the thorniest problems for the legal treatment of smart contracts.28 It also has implications for parties’ freedom of contract. As established, the content of the code will dictate what outcome the computer achieves. The content of the code, however, as with the written form of a conventional agreement, might not mirror the substance of the parties’ agreement. Whilst this is a common problem for conventional contracts, it is even more complicated where smart contracts are concerned. When an agreement is put into computer code, an extra layer of interpretative difficulty is introduced: the person coding the contract might not be one of the parties to it, and so might misunderstand the instructions. Even if she does understand them, she might make an error transposing them into code. This means that the code will not execute in accordance with the parties’ agreement, and this is where a court will need to decide how to resolve the dispute. Conventional contracts produce fiendishly difficult interpretation problems because of the many ways in which human language can be used and understood. But these multiple possible meanings do not exist for artificial intelligence. As Susskind says, computers are ‘non-thinking, high-performing’ agents.29 From the perspective of the computer performing a smart contract, therefore, there is no room for interpretation, and the instructions presented to it have a single meaning. The potential meanings to both its authors and interested third parties, however, are several. The forensic process of interpretation in this sphere is therefore likely to be much more complicated than it is in the conventional context. The broad linguistic symmetry of conventional contracting, in which the authors of the agreement, along with any third parties, use the same basic lexicon,30 is absent

28 For a fuller analysis, see S Green, ‘Smart Contracts, Interpretation and Rectification’ [2018] LMCLQ 24. 29 R Susskind and D Susskind, The Future of the Professions: How Technology Will Transform the Work of Human Experts (OUP 2015) 276. 30 That is not to suggest that this is the only source of linguistic gulfs in the law of contract. Both standard form and industry-specific agreements can, particularly as against consumers, employ language

Smart Contracts  205 in this context: computer code is qualitatively different from human language. A human reading of code, therefore, will be of limited assistance in determining the intended effect of language directed towards an artificial intelligence. Whilst human-to-human interaction is open to conflicting interpretations on both sides, human-to-machine interaction will only have multiple meanings to those choosing the language to be used, and not to the recipient of that language. Whilst a computer will do what it is told, however, this does not of course mean that it will do what the parties intended. And as Lewison LJ has made clear, the task of the court is ‘not … to ascertain “what the parties intended to agree” but what the instrument means’.31 Since there is no such thing as a reasonable computer, there is no point in approaching the interpretative exercise in relation to smart contracts by asking what a reasonable machine, or a reasonable user of human language, would make of the language used. The task is made more difficult because, despite the growing numbers of individuals proficient in computer code, it remains the case that most people would not know what such code would mean to the machine for which it was intended. Outside of the specialist field, this is not (yet) a subject of general knowledge.32 Technical interpretation problems are not in themselves a novel issue. The common law has, for instance, developed a means of dealing with contracts whose terms have an industry-specific contextual meaning, and with those written in a human language unfamiliar to the court: it simply admits expert translation evidence: A judge is supposed to know the law, the English language and such facts as are common knowledge. If he refers to authorities or dictionaries or other works dealing with these matters he can safely do so because his general knowledge enables him to check and appreciate them. But if without assistance he attempts to handle highly technical matters he may easily go astray. Some fairly simple technical matters can nowadays be considered to be common knowledge, and I would not attempt to draw the line – it may alter from time to time. In this case some of the terms are simple enough but some are certainly not. So what is a court to do if on the face of a record there are technical terms which cannot be understood without expert assistance? There may be cases where a very little assistance is all that is necessary, and I think it would be unfortunate if such a strict rule were laid down that the mere presence of some technicality is enough to prevent a court from proceeding further in the absence of evidence to explain it.33

The same approach cannot, however, be transposed for use in relation to interpreting smart contracts (or smart elements of prose contracts), because code does that is drastically different from that with which the intended audience is familiar. It is certainly arguable that the courts require a bespoke interpretative approach in those circumstances as well. 31 Cherry Tree Ltd v Landmain Ltd [2012] EWCA Civ 736, [2013] Ch 305 [99] (emphasis added). 32 Although, with coding now a subject widely taught in schools, it is a problem that might disappear in a few decades’ time. 33 Lord Reid in Baldwin & Francis Ltd v Patents Appeal Tribunal [1959] AC 663, 684.

206  Sarah Green and Adam Sanitt not use a lexicon that is necessarily common to both parties. Whilst certainly the machine itself, and any humans familiar with coding, will understand the language used,34 those who agreed the substance of the contract may well not be able to follow it at all, were they ever even to see it.35 Interpreting the code through reference simply by referring to a code lexicon, could easily fail to address the issue of what the contracting parties agreed to (particularly if neither party is herself familiar with code because, say, a third party code wrote the text). The code lexicon would be a guide only to the coder’s intention, but does not get to the root of the interpretative issue: When the language used in an instrument gives rise to difficulties of construction, the process of interpretation does not require one to formulate some alternative form of words which approximates as closely as possible to that of the parties. It is to decide what a reasonable person would have understood the parties to have meant by using the language which they did.36

This highlights the tension between English law’s objective approach to contractual interpretation37 and its regard for the intentions of the contracting parties. The objective approach reduces the threat of uncertainty,38 and provides a means of dealing with a court’s obvious inability to read individuals’ minds. The benefit of such an objective approach stems from its universality; it promotes certainty because it provides a means external to the parties of deciding what their words mean but, whilst being external, remains accessible to them so that they have the opportunity to gauge how their words will be interpreted. Again, however, this cannot simply be carried across to the smart contract sphere because of the linguistic asymmetry just described. Coders, however, can bridge this gap with their ability to understand both human and machine language. Enlisting the services of such an expert in order to represent to the court what the machine in question had been required by the smart contract code to do is on one level little different from employing the services of a human language translator, as is standard practice where contracts are written in a foreign language.39 The difference as far as code is concerned lies in the court’s actions following that translation: ‘In principle, where a document has been translated, its proper interpretation is a matter for the court, and not a proper subject of expert evidence.’40 34 Although, with highly complex code, even proficient coders might, without practical testing of the code, have their own different predictions of precisely what it will achieve in practice. 35 And in practice, they often will not ask to see it. 36 Lord Hoffmann in Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] AC 1101 [21]. 37 J Steyn, ‘Contract Law: fulfilling the reasonable expectations of honest men’ (1997) 113 LQR 433, 433–34. 38 See, eg, Lord Hodge JSC in Wood v Capita Insurance Services Ltd [2017] 2 WLR 1095 [15]: ‘One of the attractions of English law as a legal system of choice in commercial matters is its stability and continuity, particularly in contractual interpretation’; and R Buxton, ‘“Construction” and Rectification after Chartbrook’ (2010) 69 CLJ 253, 261. 39 See K Lewison, The Interpretation of Contracts, 6th edn (Sweet & Maxwell 2015) 5-06. 40 ibid.

Smart Contracts  207 As explained, however, understanding machine language is not simply a question of translating directly the terms used. The syntax and logical architecture of code is such that, without knowing the context of its initial design objectives, a literal translation of code into English is unlikely to be amenable to effective interpretation by anyone unfamiliar with the conceptual organisation of machine instructions. This presents yet another challenge to current forensic practice: Although expert evidence may be necessary to explain technical terms to the court, it is not the function of an expert to interpret the contract. That remains the function of the judge.41

But a judge unfamiliar with code might not find it easy to fulfil this particular function. To illustrate the different cognitive responses of humans and machines to language, take as a very basic example the following instructions: Go to the shop and buy a newspaper. If there any eggs, get a dozen.

The most likely human response to this instruction is to buy a newspaper and, in the event that the shop has eggs, to buy a dozen eggs as well. A computer, on the other hand, presented with this instruction,42 would buy a newspaper and, in the event that eggs are also available, will buy 12 newspapers rather than one. In response to this instruction, it will not buy a dozen eggs because at no point has it been told to purchase eggs; the only object of the ‘buy’ mandate is a newspaper. Were a court to consider these instructions, and know that they have been issued to a human agent, there is certainly scope for considering more than one intended outcome43 (either the purchase of one newspaper plus one dozen eggs, or the purchase of a dozen newspapers) as being reasonable. Whilst the purchasing of a dozen newspapers by a human agent does not seem immediately reasonable to most adults with a standard social experience, there might be something in the context of the parties’ relationship, history or knowledge that renders it so on a particular set of facts.44 The point is that there is a valid question to be asked about the meaning conveyed by these words to human intelligence, but the question of interpretation simply does not arise when considering a computer’s response. Clearly, where computer code is concerned, therefore, it is not so much the meaning or use of individual words that is likely to give rise to interpretative ­difficulties45 but the way in which those words are combined to form instructions. 41 ibid. 42 This is, of course, a translation of a machine instruction into human language. The point is merely to illustrate how the same given instruction will have one meaning to a machine, and more than one meaning to humans. 43 In the event that eggs are available. If there are no eggs, both humans (most probably) and computers (definitely) would buy just a newspaper. 44 Say, for instance, there has been a salmonella scandal in relation to eggs and there has been an exclusive report in the local newspaper, urging people not to buy them. If the parties to this contract owned a holiday resort, it would make sense for them to distribute those newspapers to their residents. 45 Although it may well be that several technical terms, with meanings specific to context, are used.

208  Sarah Green and Adam Sanitt In order to be forensically effective, therefore, expert code evidence would have to do more than explain the meaning of individual terms: it would have to go further by explaining what effect certain combinations and juxtapositions of words will have on the artificial intelligence to which they are addressed. The extent to which any court engaging in such an exercise is able successfully to align its interpretation of a smart contract with the agreement actually made by the parties will therefore determine the scope of those parties’ freedom of contract. Whilst this is true even of conventional interpretation exercises, the additional layer of potential misalignment in smart contracting is bound to present greater challenges to the objective of giving effect to parties’ intentions.

V.  Concluding Remarks The issues dealt with in this chapter do not exhaust the forensic difficulties posed by smart contracts. Even where a court has deemed a smart contract to be the proper subject of judicial intervention (because, for instance, the requirements for contract formation have been met and a court is persuaded that the parties had the requisite intention to create legal relations), there remains the significant question of what the remedial response to the dispute should be. The particular difficulty here is that the objective of Nick Szabo’s smart contract was to create ‘immutable unstoppable and irrefutable computer code’ that can ‘self-enforce’.46 This has interesting implications for the ability of the court to fashion an appropriate remedy, particularly in situations in which ongoing performance is a feature of the contract. Where the obligations under the contract are of the one-off, discrete variety,47 and have produced an outcome with which at least one of the parties is disappointed, the situation can be remedied by the granting of a conventional remedy, external to the digital world, such as conventional compensatory damages or restitution.48 Where, however, the agreement is one under which obligations are continuous, something needs to be done to the contract itself in order to ensure that future performance conforms to expectations. Where, for instance, a court determines that the coding of a smart contract does not match the parties’ manifest agreement, this looks like a job for rectification. Rectification is not, however, possible in any real sense, because a smart contract, being immutable, cannot be 46 LD Wall, ‘“Smart Contracts” in a Complex World’ (Federal Reserve Bank of Atlanta, July 2016) available at https://www.frbatlanta.org:443/cenfis/publications/notesfromthevault/1607. 47 As explained at the start of this chapter, automated execution means that parties to smart contracts are very unlikely to need to bring claims for non-performance (except where any pre-performance conditions have been fulfilled and/or where one party has in some way prevented the code from executing), but are more likely to dispute the outcome of executed performance. 48 Or even specific performance. Whilst smart contracts will be performed as coded, their execution will still rely on the satisfaction of certain conditions or requirements. There remains scope, therefore, for the fulfilment of these conditions to be mandated, thereby achieving specific performance of the contract.

Smart Contracts  209 rectified. A court-ordered variation of such an outcome49 will therefore require the creation of a new smart contract to reverse or modify the effect of the first.50 This looks more like novation than it does rectification, and perhaps points to a future in which the former remedy plays a greater role in resolving contractual disputes than it has done in the past. Historically, and for reasons that are obvious in relation to conventional contracts,51 novation has been used primarily to substitute new for existing contracting parties. In the digital age, its remit may well expand so as to facilitate the modification of content currently dealt with by rectification. These are big questions and require fuller analysis than is possible here. The fact that they need to be asked, however, lends weight to the argument advanced in this chapter that the ability of smart contracts to meet commercial expectations will depend on the extent to which the common law is prepared to adapt some of its long-established methods and responses to an age in which promises yield to performance.

49 Presuming the parties wish to continue with automated performance and/or where the obligations in question can only be performed in this manner. 50 Such variation can of course occur simply as a result of the parties’ mutual decision making. 51 Since a conventional prose contract can be rectified, there is no need to create a new one in order to modify its content.

210

11 Disproportionate Penalties in Commercial Contracts WILLIAM DAY*

I. Introduction The rule by which contractual penalties become unenforceable attracts ­considerable controversy. It goes directly to the tension between the parties’ freedom of contract (or power to contract)1 and the extent to which that freedom is then affected by what has been agreed. Vast quantities of ink have been spilled on the rule: some have argued for its expansion;2 others for its abolition.3 In Cavendish Square Holdings BV v Makdessi,4 the Supreme Court had the opportunity to take either course but chose a third way: retaining the rule but significantly narrowing its scope and strength. Makdessi was the first time that the penalties rule had reached the highest appellate court for a century,5 and it may be as long again before it is considered at that level. The challenge now is to try to explain and justify the penalties rule through the prism of Makdessi and explore how its ambiguities might be resolved in future decisions in the lower courts. That is not an easy task. In Makdessi itself, Lord Neuberger and Lord Sumption doubted that ‘the courts would have invented the rule today if their predecessors had not done so three centuries ago’.6 * I am grateful to Joanna Smith QC for discussing the penalties rule with me at some length before and at the Contents of the Commercial Contracts conference at UCL, and to Edwin Peel for comments on an earlier draft. All errors are my own. 1 See R Stevens, ‘Binding Our Future Selves’, ch 2 in this volume. 2 See, eg, M Chen-Wishart, ‘Controlling the Power to Agree Damages’ in P Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press 1996) 271. This author was previously in that category: see W Day, ‘Penalty Clauses Reconsidered’ [2014] Journal of Business Law 512. 3 See, eg, J Morgan, Contract Law Minimalism: A Formalist Restatement of Commercial Contract Law (CUP 2013) 247–49; S Worthington, ‘Common Law Values: Party Autonomy’ in A Robertson and M Tilbury (eds), The Common Law of Obligations: Divergence and Unity (Hart Publishing 2016) 301. 4 Cavendish Square Holdings BV v Makdessi; ParkingEye Ltd v Beavis [2015] UKSC 67, [2016] AC 1172. 5 The last occasion was Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Co Ltd [1915] AC 79 (HL). 6 Makdessi (n 4) [36] (Lord Neuberger and Lord Sumption).

212  William Day This chapter’s central argument is that, after Makdessi, the penalties rule is best understood not as a rule against contractual punishment per se but rather as a rule against disproportionate contractual punishment. That is evident from the face of the test adopted by the Supreme Court in Makdessi: ‘[t]he true test is whether the impugned provision is a secondary obligation which imposes a detriment on the contract-breaker out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation’.7 One major difficulty with Makdessi is in understanding what is meant by ‘legitimate interest’. It is suggested that the answer to this lies in the Supreme Court’s affirmation of the requirement that the penalty be triggered by a breach of contract. The breach rule provides the target for the proportionality assessment, requiring a comparison to be made between the remedy the law would otherwise provide (the ‘default remedy’) and the agreed remedy. The default remedy reveals what the law considers to be a party’s legitimate interest in the enforcement of a primary obligation.8 Makdessi is to be welcomed because it strikes an appropriate balance between freedom on contracting and freedom after contracting. Parties are permitted to agree to and enforce penalties in the name of freedom of contract. Only disproportionate penalties are too great an intrusion on parties’ freedom after contracting, and so will not be enforced. This chapter – and this book – is concerned with commercial contracts. But non-commercial cases cannot be ignored entirely. The law of contract is not yet the law of contracts; the same basic common law principles underpin all contracts. Indeed, Makdessi itself was decided alongside a consumer appeal involving a car parking ticket, ParkingEye Ltd v Beavis, and the way that Beavis was determined will have an impact on the approach to penalties in commercial contracts in the future. However, there are specific statutory protections for consumers. In particular, the Consumer Rights Act 2015 renders unenforceable those terms that are not the main subject matter of the contract, and which cause ‘a significant imbalance in the parties’ rights and obligations under the contract to the detriment of the consumer’.9 A disproportionate penalty should always be caught by that statutory rule, so there should now be no need to turn to the common law rule on penalties in consumer cases. Thus it is only in the commercial context that the penalties rule still has true vitality.

II.  The Problem Before Makdessi, the penalties rule had lost doctrinal coherence and was riddled with artificialities, technicalities and exceptions. It was an ‘ancient, haphazard edifice’ that had ‘not weathered well’.10 The cause of this, according to the Supreme

7 ibid

[32] (Lord Neuberger and Lord Sumption), but see section IV.A of this chapter. the discussion of ParkingEye (n 4) in section IV.B.i. 9 Consumer Rights Act 2015, ss 62 and 64. 10 Makdessi (n 4) [3] (Lord Neuberger and Lord Sumption). 8 cf

Penalties  213 Court in Makdessi,11 was the status accorded Lord Dunedin’s speech in Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Co Ltd, which drew the following dichotomy:12 The essence of a penalty is a payment of money stipulated as in terrorem of the offending party; the essence of liquidated damages is a genuine covenanted pre-estimate of damage.

Lord Dunedin then proposed various tests that could be ‘helpful, or even conclusive’ in many cases in identifying the essence of the detriment. In particular, a detriment would, or may, be a penalty if: 1. 2. 3.

‘extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved to have followed from the breach’; ‘the breach consists only in not paying a sum of money, and the sum stipulated is a sum greater than the sum which ought to have been paid’; ‘a single lump sum is made payable by way of compensation, on the occurrence of one or more or all of several events, some of which may occasion serious and others but trifling damage’.13

In Makdessi, the Supreme Court blamed the state of the penalties rule on subsequent cases that had indulged in an ‘over-literal reading’ of Lord Dunedin’s speech, treating the tests as a ‘quasi-statutory code’.14 However, it is suggested that the true problem was in the substance of Lord Dunedin’s approach. The dichotomy he drew between penalties and reasonable pre-estimates of loss was a false one: there were plenty of detriments that fell between being a penalty and a reasonable preestimate of loss.15 Further, that dichotomy set the law against all deterrence-based detriments, no matter whether they were proportionate or disproportionate. This had the potential to disrupt many legitimate commercial arrangements. Most obviously, Lord Dunedin’s approach created a problem for default interest. The reality is that default interest is included by lenders in loan agreements to encourage timely repayment or, to put it another way, to deter late repayment. But legal acrobatics were required to ensure that default interest was enforceable. In Lordsvale Finance plc v Bank of Zambia,16 Colman J indicated his willingness to go as far as he could to enforce default interest: London is one of the greatest centres of international banking in the world. Here and in New York most of the world’s international syndicated loans are set up. Such loans

11 ibid [22] and [31] (Lord Neuberger and Lord Sumption) and [219]–[221] (Lord Hodge). 12 Dunlop (n 5) 86. 13 ibid 87–88. The third test is ‘no more’ than a presumption. A fourth test is that a sum will not be a penalty if ‘the consequences of breach are such as to make precise pre-estimation almost an impossibility’. 14 Makdessi (n 4) [22] and [31] (Lord Neuberger and Lord Sumption). 15 ‘Reasonable pre-estimate of loss’ rather than ‘genuine pre-estimate of loss’ is preferable because the test does not turn on the actual intentions of the parties at the time of contracting; it is determined by the usual objective principles governing the interpretation of contracts: Day, ‘Penalty Clauses Reconsidered’ (n 2) 513. 16 Lordsvale Finance plc v Bank of Zambia [1996] QB 752 (QBD) 766–67.

214  William Day almost invariably provide for enhanced rates of default interest to apply. It would be highly regrettable if the English courts were to refuse to give effect to such prevalent provisions while the courts of New York are prepared to enforce them.

The judge reconciled commercial practice with commercial law by conceptualising default interest as reflecting the ‘increased credit risk represented by a borrower in default’.17 That allowed standard default interest provisions to fall down the gap in the middle of Lord Dunedin’s dichotomy and remain enforceable.18 The result was right, but the reasoning was obviously artificial. A borrower’s credit risk on default is not modelled by lenders in advance. Indeed, many lenders will not lend to a borrower in default at all. Lordsvale became associated with a ‘new approach’ in which a detriment that was not a reasonable pre-estimate of loss was nonetheless enforceable if ‘commercially justifiable’.19 The problem was that this exception could, in principle, swallow up the rule.20 Punishment with the purpose of deterring breach of contract is commercially justifiable. The real issue is not the stipulated punishment but rather the extent of that punishment; in other words, whether a contractual deterrent is proportionate to the reasonably anticipated consequences of a breach of contract. But Lord Dunedin’s speech in Dunlop stood in the way of the law focusing on that issue. This was the legal context in which the disputes in Makdessi and Beavis arose.21 In Makdessi,22 the dispute arose out of the expansion of the WPP advertising empire in the Middle East. In 2008, Cavendish, a subsidiary in the WPP group, purchased the majority stake in the largest advertising and marketing communications business in the region. The sellers included Mr Makdessi, who was the founder of, and leading individual behind, the target group. The total consideration for the acquisition could have reached some $150 million, with the exact amount depending on the future profits of the target group (commonly known as an ‘earn-out’), but was staged, with the sellers initially receiving around $35 million. As is usual in such transactions, the share purchase agreement contained restrictive, non-compete covenants that bound the sellers.

17 ibid 767. 18 While leaving open the possibility that default interest that went beyond the changed credit profile of the borrower would be unenforceable as a penalty. 19 Makdessi v Cavendish Square Holdings BV [2013] EWCA Civ 1539, [2014] 2 All ER (Comm) 125 [84]–[104] (Christopher Clarke LJ). 20 Day, ‘Penalty Clauses Reconsidered’ (n 2) 515–16. 21 There has already been considerable commentary on the decision. See, eg, H Beale (ed), Chitty on Contracts, 33rd edn (Sweet & Maxwell 2018) paras 26.190–26.244; C Conte, ‘The Penalty Rule Revisited’ (2016) 132 LQR 382; W Day, ‘A Pyrrhic Victory for the Doctrine against Penalties’ [2016] Journal of Business Law 115; W Day, ‘Penalty Clauses following Makdessi: Postscript’ [2016] Journal of Business Law 251; J Morgan, ‘The Penalty Clause Doctrine: Unlovable but Untouchable’ (2016) 75 CLJ 11; A Summers, ‘Unresolved Issues in the Law on Penalties’ [2016] LMCLQ 95; and S Worthington, ‘Penalty Clauses’ in S Worthington and G Virgo (eds), Commercial Remedies: Resolving Controversies (CUP 2017) 366. 22 Makdessi (n 4) [44]–[55] (Lord Neuberger and Lord Sumption) and [121]–[121] (Lord Mance).

Penalties  215 The agreement provided for two consequences on breach of these covenants. First, it halted the payment of outstanding consideration. In addition to any outstanding fixed payments, that included the consideration which would otherwise be generated from the earn-out provisions. Second, it activated a call option, exercisable at the discretion of the purchaser, whereby the seller in default would be required to sell any remaining shares in the target group at a set price. Activation of the call option also ousted the defaulting seller’s put option under which the set price for the same shares was much higher. Taken as a whole, breach of any of the restrictive covenants resulted in the seller in question standing to lose millions of dollars otherwise contractually due. Mr Makdessi eventually accepted that he had an ongoing involvement with one of WPP’s competitors, Aegis. Cavendish sought both to halt payment of the outstanding consideration to Mr Makdessi and to exercise its call option to obtain Mr Makdessi’s outstanding shares. Mr Makdessi argued that both contractual provisions were unenforceable because they constituted penalty clauses,23 and the Court of Appeal agreed,24 but that decision was overturned by the Supreme Court. The facts could not have been more different in Beavis.25 This involved an incident with which many might have sympathy: Mr Beavis overstayed the two-hour limit on the use of the car park at the Riverside Retail Park in Chelmsford, Essex, and thereby incurred a ‘parking charge’ of £85. He, too, argued the charge was unenforceable as a penalty.26 Indeed, so much fellow feeling did Mr Beavis arouse that his counsel team worked on a pro bono basis, and over £6,000 was rapidly raised through ‘crowdfunding’ to cover other costs of appealing to the Supreme Court. However, the Supreme Court itself was less inclined to be sympathetic, with Lord Sumption remarking during judgment hand down that all Mr Beavis had really needed was a watch.27

III.  Breach of Contract and Secondary Obligations A.  Breach of Contract i.  Retention and Justification A persistent criticism of the penalty rule before Makdessi was that it operated only on breach of contract. Not only did this allow the parties to draft around it with relative ease, it was also thought not to fit with its underlying justification, whether 23 One oddity of the case is that relief from forfeiture was not also pleaded. 24 Makdessi (n 4). See J O’Sullivan, ‘Lost on Penalties’ (2014) 73 CLJ 480; and E Peel, ‘Unjustified Penalties or an Unjustified Rule against Penalties’ (2014) 130 LQR 365. 25 Makdessi (n 4) [89]–[92] (Lord Neuberger and Lord Sumption) and [123]–[125] (Lord Mance). 26 Also that it was unfair and therefore unenforceable under the Unfair Terms in Consumer Contracts Regulations 1999 (SI 1999/2083), which have since been superseded by the Consumer Rights Act 2015. 27 Day, ‘A Pyrrhic Victory’ (n 21) 116.

216  William Day that be substantive unfairness,28 procedural unfairness,29 or something else. It was said to give rise to the ‘absurd paradox’ that the law would ‘grant relief to a man who breaks his contract but [would] penalise the man who keeps it’.30 There had been some momentum towards abolishing the requirement for a breach of contract. The Law Commission had reached the provisional view that it should be abolished, primarily because of the latitude it gave to parties to draft around the rule and because of the difficulties of working out whether a detriment was triggered by breach or some other contractual right.31 The Law Commission suggested in its place that the ‘rules of penalties should be applied wherever the object of the disputed contractual obligation is to secure the act or result which is the true purpose of the contract’.32 More recently, in Andrews v Australia and New Zealand Banking Group Ltd, the High Court of Australia abandoned the breach rule on the basis that it was a requirement that had only ever applied as the rule had developed at common law and did not similarly restrict the rule in its older, equitable form.33 The High Court held that that the doctrine applied to all ‘collateral’ or ‘accessory’ obligations. While the historical analysis in Andrews is persuasive, the High Court has rightly come in for criticism for not also justifying its approach by reference to modern-day considerations.34 There was no dispute on the facts in both Makdessi and Beavis that the clauses in question operated on breach of contract, and that there had in fact been breaches of contract.35 Nonetheless, in obiter dicta, the Supreme Court declined to follow the High Court of Australia in expanding the rule in this manner.36 For Lord Neuberger and Lord Sumption, their primary motivation appears to have been to keep a functional check on the rule.37 Removing the breach requirement would ‘represent the expansion of the courts’ supervisory jurisdiction into a new territory of uncertain boundaries, with has hitherto been treated as wholly governed by mutual agreement’.38 But this fear of trespass upon freedom 28 Chen-Wishart ‘Controlling the Power’ (n 2) 286–87 and 298. 29 T Downes, ‘Rethinking Penalty Clauses’ in Birks (ed), Wrongs and Remedies in the Twenty-First Century (n 2) 259. 30 Bridge v Campbell Discount Co Ltd [1962] AC 600 (HL) 629 (Lord Denning). 31 Law Commission, Penalty Clauses and Forfeiture of Money Paid (Working Paper No 61, 1975) paras 17–26. 32 ibid para 26. 33 Andrews v Australia and New Zealand Banking Group Ltd [2012] HCA 30, (2012) 290 ALR 595. 34 JW Carter et al, ‘Contractual Penalties: Resurrecting the Equitable Jurisdiction’ (2013) 20 Journal of Contract Law 99, 108–09. 35 cf ParkingEye Ltd v Beavis [2015] EWCA Civ 402, [2012] RTR 27, [23] (Moore-Bick LJ). This issue in respect of Beavis occupied some time at the hearing, but there is almost no mention of it in the judgments. 36 Makdessi (n 4) [40]–[43] (Lord Neuberger and Lord Sumption) and [129]–[130] (Lord Mance). The other Justices did not engage with the issue. As Lord Hodge put it at [241], the point was really ‘peripheral to the main arguments in this appeal’. 37 They also, rather unpersuasively, took issue with the High Court’s historical analysis and its articulation of the rule without a breach requirement: Makdessi (n 4) [42]. See further Day, ‘A Pyrrhic Victory (n 21) 119–21. 38 Makdessi (n 4) [42].

Penalties  217 of contract is unfounded. The requirement that the detriment be a ‘collateral’ or ‘secondary obligation’ would perform the same check, save that it could not be circumvented with clever drafting with quite the same ease. In contrast, Lord Mance reasoned that there was ‘a real distinction, legal and psychological’ between clauses conditional on non-performance of another obligation and clauses stipulating a detriment on breach that would otherwise attract damages, an injunction or some other remedy.39 This is the preferable analysis. If the penalty rule is to be about disproportionate penalties, that begs the question ‘disproportionate to what?’ It is suggested that the breach rule provides the target for the proportionality rule: the comparison to be made is between the default remedy otherwise available on breach and the agreed remedy.40

ii.  Limits on Drafting Around the Breach Rule The Law Commission’s concern that the penalty rule might be evaded through careful drafting is better met through specific anti-avoidance techniques rather than cutting the anchor of the breach rule itself. In Makdessi, Lord Neuberger and Lord Sumption said that the ‘capricious consequences’ of the breach rule ‘are mitigated by the fact that, as the equitable jurisdiction shows, the classification of terms for the purpose of the penalty rule depends on the substance of the term and not on its form or on the label which the parties have chosen to attach to it’.41 They also said that ‘where it is clear that the parties have so circumvented the rule and that the substance of the contractual arrangement is the imposition of punishment for breach of contract, the concept of a disguised penalty may enable a court to intervene’.42 Given the authorities upon which Lord Neuberger and Lord Sumption (and Lord Hodge) relied for this point, it is likely that they had in mind the ordinary process of contractual interpretation.43 This appears also to have been the approach taken in the only case to consider the point since Makdessi. In Holyoake v Candy, fees charged on the extension of loans were structured so that, if paid on time, the fees would be set off against the outstanding balance, but, if paid late, the outstanding balance would not be reduced. Nugee J held that the fees were not triggered by breach on the basis that the extension fee was consideration for an extension of the time for repayment.44 39 ibid [129]. 40 See also P Saprai, Contract Law Without Foundations: Toward a Republican Theory of Contract Law (OUP 2019) 168–69; R Stevens, ‘Rights Restricting Remedies’ in A Robertson and M Tilbury (eds), Divergences in Private Law (Hart Publishing 2016) 159, 175–76. 41 Makdessi (n 4) [15]. There is not space in this chapter to consider whether, when or how the equitable jurisdiction ‘disappeared’ in England, but for this author’s previous scepticism that it has see Day, ‘A Pyrrhic Victory (n 21) 120–21. 42 Makdessi (n 4) [77] (Lord Neuberger and Lord Sumption); see too ibid [258] (Lord Hodge). 43 ibid [15] (Lord Neuberger and Lord Sumption), citing Street v Mountford [1985] AC 809 (HL) 819 (Lord Templeman), and [258] (Lord Hodge), citing American Law Institute, Restatement of the Law Second, Contracts (American Law Institute Publishers 1981) s 356. 44 Holyoake v Candy [2017] EWHC 3397 (Ch) [476]–[477]; also ibid [481]–[483].

218  William Day A substance over form approach based on interpretation is constrained by unambiguous language,45 but it has still some force. In particular, it is probably the basis on which the rule is engaged even when a detriment is, strictly speaking, triggered by termination of a contract for breach rather than directly by the breach itself.46 However, when the problem of unambiguous language is encountered, there may be some outlet in the sham doctrine.47 This operates as an anti-abuse check on the usual objective approach to transactions in private law,48 by looking at the parties’ actual intentions rather than as they would appear to a reasonable person.49 Its requirements are stringent, so clever drafting to avoid a breach trigger for a detriment will in many cases still succeed in ousting the rule against ­penalties.50 In the leading decision of Snook v London and West Riding Investments Ltd,51 Diplock LJ held that a sham meant acts done or documents executed by the parties to the ‘sham’ which are intended by them to give to third parties or to the court the appearance of creating between the parties’ legal rights and obligations different from the actual legal rights and obligations (if any) which the parties intend to create … [F]or acts or documents to be a ‘sham’, with whatever legal consequences follow from this, all the parties thereto must have a common intention that the acts or documents are not to create the legal rights and obligations which they give the appearance of creating.

Summers has argued that the sham doctrine is ‘less well suited’ to the penalties rule for two reasons. First, given the sham doctrine requires both parties to deliberately aim to avoid the penalties rule, it would intervene in ‘the very circumstances in where circumvention is least objectionable’. Second, the sham doctrine ‘serves to protect specific public policy objectives that override freedom of contract’, whereas the breach requirement demonstrates that the penalty is subservient to freedom of contract.52 Those arguments fall away once the penalty rule is conceptualised as a public policy rule that has nothing to do with the quality of the consent of the parties to the agreed stipulation, and the breach rule is no longer seen as simply a functional check on the rule in the interests of freedom of contract.53

45 Summers, ‘Unresolved Issues’ (n 21) 100. 46 Bridge v Campbell Discount Co Ltd; Cooden Engineering Co Ltd v Stanford [1953] 1 QB 86 (CA). See further L Gullifer, ‘Agreed Remedies’ in A Burrows and E Peel (eds), Commercial Remedies: Current Issues and Problems (OUP 2003) 191, 200–04. 47 Day, ‘A Pyrrhic Victory’ (n 21) 122. 48 Much of the commentary arises in relation to trusts, tax and self-employment; for a flavour of the literature see, eg, M Conaglen, ‘Sham Trusts’ (2008) 67 CLJ 176; J Vella, ‘Sham Transactions’ [2008] LMCLQ 488; and A Bogg, ‘Sham Self-Employment in the Supreme Court’ [2012] Industrial Law Journal 328. 49 Hitch v Stone (Inspector of Taxes) [2001] EWCA Civ 63 [66] (Arden LJ). 50 A wider approach was taken to shams by the Supreme Court in Autoclenz Ltd v Belcher [2011] UKSC 41, [2011] 4 All ER 745. But the court expressly distinguished self-employment cases from the ordinary commercial context: see ibid [21], [28] and [34] (Lord Clarke). 51 Snook v London and West Riding Investments Ltd [1967] 2 QB 786 (CA) 802. 52 Summers, ‘Unresolved Issues’ (n 21) 99. 53 See section IV of this chapter.

Penalties  219 Summers has suggested that there is a third possible explanation for the disguised penalty rule: a ‘substance over form’ rule that can ‘look past the parties’ unambiguous language’.54 This should be rejected. Contracts are understood by what a reasonable party understood the parties to mean by the words used in the relevant context or, exceptionally, by what the parties actually meant when using those words. There is no other standard by which the content of contractual obligations can be determined. Indeed, the dicta cited by Summers in support of this third possibility simply represent a contextual approach to interpretation.55

B.  Secondary Obligations A significant development in Makdessi was that a majority of the Supreme Court held that only ‘secondary obligations’ are subject to the penalty rule;56 a ‘primary obligation’ triggered by breach cannot be rendered unenforceable. This was surprising because the concept of ‘secondary obligations’ had been developed in this context as an alternative to the breach requirement.57 But the Supreme Court appears to have adopted it as a requirement additional to breach. The adoption of a secondary obligation limitation should not lead to a confusion as to the target of proportionality test. Secondary obligations are defined in Makdessi as ‘a contractual alternative to damages at law’,58 ‘mere security for the performance’,59 and as ‘penal remedies for breach of contract’.60 These definitions reinforce the conclusion that the comparison to be made is between default remedies available at common law and the agreed remedy. In other words, the two requirements of breach and secondary obligations are interlinked: the clause must not just respond to the breach of contract; it must also address the consequences of that breach. The language of ‘secondary obligations’ is conventionally used to describe damages at law. This has prompted some criticism in the commentary about its deployment in the penalties context, suggesting that ‘ancillary’ or ‘collateral’ obligations would be preferable, although this is really no more than a quibble about 54 Summers, ‘Unresolved Issues’ (n 21) 100–01. 55 Office of Fair Trading v Abbey National plc [2009] UKSC 6, [2010] 1 AC 696 [83] (Lord Phillips); Iman Sadeque v Bluebay Asset Management Services Ltd [2012] EWHC 3511 (QB), [2013] IRLR 344 [203] (Popplewell J). 56 Makdessi (n 4) [32] (Lord Neuberger and Lord Sumption) and [241] (Lord Hodge). Lord Mance was silent on this issue. This requirement was developed by the Justices; it was not part of any party’s case. 57 See section II.A.i. It is also perhaps surprising because this distinction was not ventilated at the hearing of the appeal, save in the very last question put to Joanna Smith QC in reply submissions by Lord Neuberger. 58 Makdessi (n 4) [77] (Lord Neuberger and Lord Sumption). 59 ibid [7] (Lord Neuberger and Lord Sumption). 60 ibid [241] (Lord Hodge).

220  William Day the appropriate label.61 In contrast, two issues merit greater exploration. First, how are primary obligations distinguished from secondary obligations? Second, even then, are all types of secondary obligations subject to the penalty rule?

i.  Distinguishing between Primary and Secondary Obligations While the distinction between primary and secondary obligations can be articulated in the abstract, it has already proved very difficult to apply that distinction in practice. Indeed, in Makdessi itself, there was no consensus on this point. Lord  Neuberger and Lord Sumption held that the clause in the share purchase agreement halting payment of the outstanding consideration was a primary obligation because it was a ‘price adjustment clause’.62 In contrast, Lord Clarke and Lord  Hodge kept an ‘open mind’ about this.63 There is some artificiality with construing the clause as doing no more than adjusting the consideration. The share purchase agreement did not contain any mechanism for recalculating the value of the target company following breach of the restrictive covenants in order to identify the precise damage inflicted on the company’s goodwill. The same ‘adjustment’ would thus always be made, regardless of any actual damage to goodwill as a result of Mr Makdessi’s activities. As Lord Mance was silent on the distinction between primary and secondary obligations, there was no clear majority support for Lord Neuberger and Lord Sumption’s analysis. It is suggested that their approach to this clause was wrong. Lord Neuberger and Lord Sumption also held that the call option in the share purchase agreement, allowing Cavendish to force Mr Makdessi to sell his remaining shares, was a primary obligation with the purpose of severing the shareholding connection between them. But this overlooks the fact that the forced transfer of the shares was at a substantial undervalue. There was no majority support for this application of the distinction. Lord Mance remained silent, and Lord Hodge supported by Lord Clarke concluded that this was a secondary obligation.64 Uncertainty has pervaded the post-Makdessi case law on the distinction between primary and secondary obligations. For example, there have already

61 Conte, ‘The Penalty Rule Revisited’ (n 21) 385; Summers, ‘Unresolved Issues’ (n 21) 102; Worthington, ‘Penalty Clauses’ (n 21) 376–78. cf D Foxton, ‘How useful is Lord Diplock’s distinction between primary and secondary obligations in contract?’ (2019) 135 LQR 249, 255–57. 62 Makdessi (n 4) [74] (Lord Neuberger and Lord Sumption). Summers, ‘Unresolved Issues’ (n 21) 103–06 suggests that detriments that are implemented by way of retrospective effect, ie by extinguishing or varying obligations already accrued, may now fall outside of the penalties rule. This is on the basis of comments made by Lord Neuberger and Lord Sumption when dealing with the question of deferred considerations. Those comments are best read not as a hard rule that retrospective detriments fall outside of the penalty rule but that they will fall outside the penalty rule only if, on construction, the detriment is a primary obligation rather than a secondary obligation. 63 Makdessi (n 4) [270] (Lord Hodge) and [291] (Lord Clarke). Lord Clarke changed his mind – or at least his judgment – after it was handed down: Day, ‘Penalty Clauses following Makdessi’ (n 21) 251–52. 64 Makdessi (n 4) [280] (Lord Hodge) and [291] (Lord Clarke).

Penalties  221 been a number of cases involving ‘bad leaver’ provisions, whereby an employee whose employment contract has been terminated has been forced to give up his or her shares in the company employing him or her at a substantial undervalue. In Richards v IP Solutions Ltd,65 May J indicated that this was a primary obligation, appearing to attach weight to the fact that the leaver provisions were in the company’s articles of association rather than the contract of employment. This appears to put form over substance. A highly formalistic approach was also taken in Signia Wealth Ltd v Vector Trustees Ltd,66 where Marcus Smith J held that obligation was a primary one not triggered by breach. That was because the compulsory surrender of shares under the articles of association was not triggered by breach but rather by an employee’s departure; breach only affected the valuation of the shares on departure.67 It is suggested that a preferable analysis was adopted in Gray v Braid Group (Holdings) Ltd,68 where the Inner House of the Court of Session held that bad leaver provisions in articles of association were secondary obligations, and it made no difference that only the price of the shares and not their compulsory surrender was triggered by breach. A difficult case on the distinction between primary and secondary obligations is Vivienne Westwood Ltd v Conduit Street Development Ltd,69 a rare successful invocation of the penalty doctrine after Makdessi. This case concerned rent due under a lease. The lease itself stipulated an annual rent of £110,000, payable quarterly, and subject to rent review. A side letter entered into at the same time as the lease stipulated that the landlord would accept rent at a lower rate, starting at £90,000 and with caps on the increases that could be imposed by rent review during the lifetime of the lease. The side letter could be terminated in the event of any breach of its terms, at which point the rent would not only revert to the level set out in the lease for future quarters, but would also retrospectively revert to that level for past quarters. The court held that the detriment was a secondary ­obligation.70 That was because the side letter was signed at the same time as the lease itself,71 so the lower rate contained in that letter was payable ‘at the outset’ and constituted, in substance, the real primary rent obligation between the parties. It is arguable that Vivienne Westwood is wrongly decided on this point, and that rent payable under a lease is a quintessential primary obligation. But on balance the characterisation of the increased rent as secondary in nature is probably correct. The side letter and the lease constituted a single transaction and were rightly read together. To treat the rent obligations in the lease as primary 65 Richards v IP Solutions Ltd [2016] EWHC 1835 (QB) [84]–[85]. 66 Signia Wealth Ltd v Vector Trustees Ltd [2018] EWHC 1040 (Ch) [653]. 67 See also Nosworth v Instinctif Partners Ltd (EAT, 28 February 2019) [65]–[71] (Slade J). 68 Gray v Braid Group (Holdings) Ltd [2016] CSIH 68, (2017) SC 409 [81]–[82] (Lord Menzies) and [106] (Lord Brodie). 69 Vivienne Westwood Ltd v Conduit Street Development Ltd [2017] EWHC 350 (Ch). 70 ibid [42]–[49] (Timothy Fancourt QC sitting as a Judge of the High Court). 71 cf Re SHB Realisations Ltd [2018] EWHC 402 (Ch), [2018] Bus LR 1173 [38] (Christopher Pymont QC sitting as a Judge of the High Court).

222  William Day obligations simply because they were contained in the lease would be to continue the overly formalistic approach seen in Richards and Signia. A striking feature of the case law that has followed Makdessi is that not once has a clause triggered by breach been construed as a primary rather than a secondary obligation. All the cases in which a penalties challenge has been dismissed on the basis that the detriment was a primary obligation have also been cases in which the detriment was not triggered by breach.72 Had Vivienne Westwood been decided differently, it would have been the sole exception to this trend. This obviously weighed on the mind of the judge in Vivienne Westwood itself, who emphasised the need to take ‘care’ before following the willingness of Lord Neuberger and Lord Sumption to treat obligations contingent on breach as primary obligations.73 Thus, what the post-Makdessi cases – Vivienne Westwood included – indicate is that first instance judges will be much more cautious than Lord Neuberger and Lord Sumption about characterising obligations triggered by breach as primary obligations.

ii.  Types of Secondary Obligations Subject to the Penalties Rule A variety of obligations triggered on breach are potentially subject to the rule against penalties. The ‘classic’ penalty clause stipulates for a sum of money to be paid by the contract-breaker to the innocent party.74 However, the Supreme Court in Makdessi equivocated over whether the doctrine also applied to clauses that, on breach, withhold a payment otherwise due. Lord Mance and Lord Hodge affirmed that the rule also applies to such clauses.75 Lord Neuberger and Lord Sumption were less sure: they were ‘prepared to assume, without deciding’ that such clauses were caught by the doctrine.76 Their difficulty was that such a clause might be characterised as a forfeiture clause and therefore subject to relief from forfeiture.77 The Supreme Court was similarly divided on non-monetary detriments. The Supreme Court had no difficulty in recognising that the penalties rule also applies to clauses requiring, on breach, a transfer of property either at an undervalue or at no value.78 But Lord Neuberger and Lord Sumption held that a clause providing for the reversal of a transfer of title in property on breach was also outside the purview of the penalties rule.79 In this, they were in a minority. Lord Mance and Lord Hodge carried majority support in their conclusion that the doctrine

72 See, eg, Edgeworth Capital (Luxembourg) SARL v Ramblas Investments BV [2016] EWCA Civ 412; BHL v Leumi ABL Ltd [2017] EWHC 1871 (QB). 73 Vivienne Westwood (n 69) [46] (Timothy Fancourt QC sitting as a Judge of the High Court). 74 Makdessi (n 4) [16] (Lord Neuberger and Lord Sumption). 75 ibid [170] (Lord Mance) and [228] (Lord Hodge). 76 ibid [73] (Lord Neuberger and Lord Sumption). 77 ibid [18] (Lord Neuberger and Lord Sumption). 78 ibid [16] (Lord Neuberger and Lord Sumption), [170] and [183] (Lord Mance) and [233] (Lord Hodge). 79 ibid [17] (Lord Neuberger and Lord Sumption).

Penalties  223 against penalty clauses and relief against forfeiture are not mutually exclusive. A court can first consider whether a clause is a penalty and therefore automatically unenforceable as a matter of law and, if not, go on to consider whether to exercise its discretion to offer relief from forfeiture.80 However, in practice, having two bites of the cherry is unlikely to make much of a difference to the contract breaker. It is difficult to envisage circumstances in which the detriment is held to be a proportionate penalty and then fall foul of the same proportionality exercise that applies in forfeiture cases.81 Relief from forfeiture may only really matter now in practice when the forfeiture is not triggered by breach and therefore beyond the reach of the penalties rule. Deposits, however, present a real problem. In obiter dicta, the Supreme Court held that the penalties rule is also applicable to them.82 The problem is that the nature of deposits, and the way in which they can be reviewed by the courts, differ fundamentally from liabilities triggered by breach. Deposits involve the imposition of a contractually agreed obligation to provide a benefit (usually money) necessarily antecedent to any breach of contract. When a breach occurs, no further liability is imposed because no right to recover the deposit exists that can be extinguished.83 Further, as Lord Hodge recognised in Makdessi, the test for the enforceability of a deposit is not the proportionality of the detriment versus the remedy offered at law but rather the reasonableness as earnest money compared to market practice.84 This is an entirely different standard of review.

IV. Proportionality A.  Formulating the Test In Makdessi, more than one formulation of the proportionality test was articulated. Lord Neuberger and Lord Sumption’s formulation attracted the support of four Justices: The true test is whether the impugned provision is a secondary obligation which imposes a detriment on the contract-breaker out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation.85 80 ibid [160]–[161] (Lord Mance), [227] (Lord Hodge), [291] (Lord Clarke) and [292] (Lord Toulson). 81 Stockloser v Johnson [1954] 1 QB 476 (CA) 490 (Denning LJ): ‘the forfeiture clause must be of a penal nature, in this sense, that the sum forfeited must be out of all proportion to the damage’. One possible substantive difference may be the ability to give relief on condition that the primary obligation is performed: see section IV.C.i. 82 Makdessi (n 4) [16] (Lord Neuberger and Lord Sumption), [156] (Lord Mance) and [234]–[238] (Lord Hodge). 83 C Conte, ‘Deposit Clauses’ in Worthington and Virgo (eds), Commercial Remedies: Resolving Controversies (n 21) 390, 400–06. 84 Makdessi (n 4) [234]–[238] (Lord Hodge). See further Workers Trust & Merchant Bank Ltd v Dojap Investments Ltd [1993] AC 573 (PC). 85 Makdessi (n 4) [32] (Lord Neuberger and Lord Sumption, Lord Carnwath agreeing) and [291] (Lord Clarke).

224  William Day Lord Hodge’s formulation attracted the support of three Justices: whether the sum or remedy stipulated as a consequence of a breach of contract is exorbitant or unconscionable when regard is had to the innocent party’s interest in the performance of the contract.86

Lord Mance’s formulation also attracted the support of two, or three, Justices: What is necessary in each case is to consider, first, whether any (and if so what) legitimate business interest is served and protected by the clause, and, second, whether, assuming such an interest to exist, the provision made for the interest is nevertheless in the circumstances extravagant, exorbitant or unconscionable.87

Cases after Makdessi have revealed uncertainty about which formulation to cite. Some judges have cited all three versions of the test.88 This is unnecessary and could cause confusion.89 Only the test as articulated by Lord  Neuberger and Lord Sumption need be, and should be, cited.90 It was the only formulation that attracted majority support in the Supreme Court. Further, there is nothing in Lord Hodge’s or Lord Mance’s judgments to suggest that they took issue with the substance of the formulation put forward by Lord Neuberger and Lord Sumption. Their judgments should be understood to be expressing the same test in a different way. Moreover, the formulation put forward by Lord Neuberger and Lord Sumption is the best expression of the test because it uses the clear, modern and more readily understandable language of proportionality. In this context, ‘extravagance’, ‘exorbitance’ and ‘unconscionability’ are best understood as old-fashioned ways of speaking about proportionality.91 Indeed, elsewhere in his judgment Lord Hodge used the concept of proportionality,92 and Lord Mance similarly contrasted extravagance and unconscionability on the 86 ibid [255] (Lord Hodge), [291] (Lord Clarke) and [293] (Lord Toulson). 87 ibid [152] (Lord Mance), [291] (Lord Clarke) and [292] (Lord Toulson), although cf [293] (Lord Toulson). 88 See, eg, Hayfin Opal Luxco 3 SARL v Windermere VII CMBS plc [2016] EWHC 782 (Ch) [132]–[135] (Snowden J); BHL (n 72) [45] (HHJ Waksman QC); Holyoake v Candy (n 44) [467] (Nugee J); SHB Realisations (n 71) [24] (Christopher Pymont QC sitting as a Judge of the High Court); GPP Big Field LLP v Solar Solutions SL [2018] EWHC 2866 (Comm) [61] (Richard Salter QC sitting as a Judge of the High Court); Cargill International Trading Pte Ltd v Uttam Galva Steels Ltd [2019] EWHC 476 (Comm) [37]–[40] (Bryan J). 89 Certainty in this context is particularly important for commercial parties negotiating contracts: J  Shelton, ‘Agreed Remedies: Comment’ in Burrows and Peel (eds), Commercial Remedies: Current Issues and Problems (n 46) 221, 222. 90 As in, eg, First Personnel Services Ltd v Halfords Ltd [2016] EWHC 3220 (Ch) [161] (Jeremy Cousins QC sitting as a Judge of the High Court); Signia Wealth (n 66) [652]–[653] (Marcus Smith J); ICICI Bank UK plc v Assam Oil Co Ltd [2019] EWHC 750 (Comm) [31] (Andrew Burrows QC sitting as a Judge of the High Court); CFL Finance Ltd v Bass [2019] EWHC 1839 (Ch) [46] (ICC Judge Briggs). 91 Beale (ed), Chitty (n 21) paras 26.228–26.229; cf Summers ‘Unresolved Issues’ (n 21) 110, arguing that ‘exorbitance’ and ‘unconscionability’ require a ‘supplementary test’, albeit then stating that ‘the assessment of exorbitance and unconscionability will depend on the proportionality of the stipulation compared with the innocent party’s legitimate interests’. 92 Makdessi (n 4) [226], [227], [247], [253], [254], [255], [265] and [287] (Lord Hodge).

Penalties  225 one hand with commensurability on the other.93 This is confirmed by Holyoake v Candy, where Nugee J concluded that there was not ‘any difference of substance between the three main judgments’ on this point;94 Lord Hodge and Lord Mance were simply ‘using traditional language’ to say the same thing as Lord Neuberger and Lord Sumption.95

B.  Understanding the Test In order to understand the proportionality test, it is necessary to unpack what is meant by a ‘legitimate interest of the innocent party in the enforcement of the primary obligation’. It is also important to appreciate the role that deterrence and the presumption in favour of freedom of contract play when determining whether the secondary obligation is ‘out of all proportion’ to that legitimate interest.

i.  Legitimate Interest Commentators have disagreed about the extent to which the Supreme Court’s reference to ‘legitimate interest’ should be understood a cross-reference to ‘legitimate interest’ elsewhere in the law on contractual remedies.96 Summers has suggested that the legitimate interest in the penalty rule is ‘essentially freestanding’.97 Rowan has considered whether an analogy can be drawn with the starting point in contract law that damages are an adequate remedy for breach, and the occasions on which the court will depart from that to award specific performance; but she argues that there are ‘limits’ to the analogy because of the particular concerns that arise when specific performance is sought.98 In any event, ‘legitimate interest’ does not do much (if any) work in the context of specific performance. Rowan proposes instead a number of other considerations that can inform the inquiry for legitimate interest, such as the importance of the obligation and the seriousness of the consequences of breach, the impact of

93 ibid [133], [143], [151] and [153] (Lord Mance). See also ibid [20] (Lord Neuberger and Lord Sumption), citing Clydebank Engineering & Shipbuilding Co Ltd v Yzquierdo y Castaneda [1905] AC 6 (HL) [20] (Lord Robertson). 94 Holyoake v Candy (n 44) [467] (Nugee J). 95 Vivienne Westwood (n 69) [38] (Timothy Fancourt QC sitting as a Judge of the High Court). 96 See also Scottish Law Commission, Discussion Paper on Penalty Clauses (Discussion Paper No 162, 2016) paras 5.36–5.53; Scottish Law Commission, Report on Review of Contract Law: Formation, Interpretation, Remedies for Breach and Penalty Clauses (Scot Law Com No 252) paras 20.20–20.26. 97 Summers ‘Unresolved Issues’ (n 21) 111. 98 S Rowan, ‘The “Legitimate Interest in Performance” in the Law on Penalties’ (2019) 78 CLJ 148, 151–54 and 159–61. Rowan also considers the role of legitimate interest in affirming a contract on breach and then suing for the agreed sum, and in the availability of gain-based damages: ibid 154–59 and 161–64.

226  William Day breach on the interests of third parties and the public interest, the protection of non-financial expectations and the characteristics of the contracting parties.99 Chitty has also proposed a range of considerations as to legitimate interests, some of which are consistent with the question of whether damages would be an adequate remedy – difficulty in proving loss,100 unavailability of substitute performance101 – but many of which are not.102 However, the wheel does not need to be reinvented. The law already has established principles to determine the extent of an innocent party’s legitimate interest in enforcing – or seeking a substitute for – primary obligations. These are the principles for the availability of relief by way of damages, debt, specific performance and injunctions. The innocent party’s legitimate interest in performance of the primary obligation for the purposes of the penalty rule is determined by which of these remedies would be available for breach of that primary obligation at law. The comparison to be made is between whichever of those remedies the law would provide absent an agreed remedy and the remedy in fact agreed between the parties. As we have seen, that is the best explanation for the existence of the breach rule.103 In practice, this requires distinguishing between positive primary obligations and negative primary obligations, because the remedies available at law differ between these two types of obligation.104 The way the Supreme Court determined the appeal in Beavis presents two difficulties as to what counts as a legitimate interest. First is the question of whose legitimate interest counts. Mr Beavis’s contract was with the car park operator, ParkingEye.105 ParkingEye in turn had a contract with the landlord of the retail park, which also owned the car park.106 In determining that the car-parking penalty was enforceable, the Supreme Court took into account the legitimate interests of the landlord in ensuring that there was available customer parking at the park.107 This does not fit with general test for disproportionate penalties, which is expressed to focus on the interest of the ‘innocent party’.108 However, Beavis should not be understood as broadening the proportionality test to encompass third parties who may be affected by breach, still less to the public interest at large.109 The better approach is that ParkingEye had assumed responsibility in

99 ibid 164–74. 100 Beale (ed), Chitty (n 21) para 26.215. 101 ibid para 26.216. 102 Beale (ibid) also suggests, inter alia, difficulties in detecting breach, third party losses and the insolvency risk placed on the claimant by the defendant’s breach: see paras 26.216 and 26.218–26.225. 103 See section III.A.i. 104 See further in section IV.C. 105 Makdessi (n 4) [94] (Lord Neuberger and Lord Sumption), [188] (Lord Mance) and [284] (Lord Hodge). 106 ibid [89] (Lord Neuberger and Lord Sumption), [193] and [197] (Lord Mance). 107 ibid [99] (Lord Neuberger and Lord Sumption), [194] (Lord Mance) and [286] (Lord Hodge). 108 See section IV.A. 109 cf Beale (ed), Chitty (n 21) para 26.218; Summers ‘Unresolved Issues’ (n 21) 112–13; Rowan, ‘The “Legitimate Interest in Performance”’ (n 98) 168–71.

Penalties  227 its contract with the landlord for ensuring there would be available customer parking, through imposing penalties on those who stayed in the car park beyond two hours. It thereby had an identity of interest in the performance of the primary obligation with the landlord, and it was in that context that ParkingEye in turn contracted with Mr Beavis.110 As Lord Hodge put it: ParkingEye had a legitimate interest to protect. It provided a service to its clients, the owners of the retail park which leased units to retailers. It undertook to manage the car park in a way which benefitted the owners and the retailers and also the public seeking to visit units within the retail park by encouraging the public to remain in the car park for no longer than two hours.111

Third-party or public interest in the enforcement of primary obligations should only be relevant where they have been assumed by the innocent party and would therefore affect the default relief to which the innocent party would be entitled at law.112 Second, the Supreme Court appeared to indicate that ParkingEye could have a legitimate interest not in enforcement of the primary obligation, but instead in enforcement of the secondary obligation. It had a legitimate interest in receiving the £85 penalty because it used the penalties to meet its operating costs and the costs of maintaining the car park.113 Again, this does not fit with general test for disproportionate penalties, which is expressed to focus on the interest in enforcement of the primary not the secondary obligation. The conclusion in Beavis suggests that the £85 ‘parking charge’ was not really a secondary obligation at all triggered by breach; it was in substance the price for staying in the car park for more than two hours.114

ii. Deterrence The most fundamental change in the law brought about by Makdessi was to overturn Lord Dunedin’s rejection of deterrence clauses per se.115 After Makdessi, it is accepted that it is legitimate to deter breach of contract, so long as the deterrence is proportionate to the innocent party’s legitimate interest in performance of the contract.116 This was central to the facts of both Makdessi and Beavis, where the detriment in each case had nothing to with the innocent party’s losses. The deferral 110 Makdessi (n 4) [197] (Lord Mance). 111 ibid [286]. 112 The two obvious but controversial examples are the relevance of third-party interests to damages (see, eg, Linden Garden Trusts v Lenesta Sludge Disposal Ltd [1994] 1 AC 85 (HL)) and the relevance of third-party interests to specific performance (Beswick v Beswick [1968] AC 58 (HL)). 113 Makdessi (n 4) [99] (Lord Neuberger and Lord Sumption), [193] (Lord Mance) and [286] (Lord Hodge). 114 ParkingEye (n 4) [23] (Moore-Bick LJ). cf Makdessi (n 4) [94] (Lord Neuberger and Lord Sumption) and [189] (Lord Mance). 115 See section II. 116 Makdessi (n 4) [28] and [31] (Lord Neuberger and Lord Sumption) and [248] (Lord Hodge).

228  William Day of consideration and the threat of compulsory transfer of shares at an undervalue in Makdessi, each contingent on compliance with restrictive covenants during that time, were obviously designed to incentivise compliance with those covenants.117 Similarly, the car-parking penalty in Beavis was a deterrent against staying in the car park beyond the two hours of free parking.118 However, this significant change to the law does not appear to have been appreciated by the courts yet. For ­example, the challenges to default interest that have followed Makdessi have tended still to follow the artificial analysis in Lordsvale,119 rather than recognise the reality that default interest provisions are designed to deter borrowers from breaching payment obligations.120 The legitimacy of deterrence clauses is significant in understanding the underlying justification for the penalty rule.121 It also plays an important part in understanding how the proportionality test operates. The remedy available at law does not act as a ‘cap’ on the agreed remedy but as an ‘anchor’. There is slack for the agreed remedy to move away from the default remedy, but if it strays too far the line will go taut. In other words, the court will enforce an agreed remedy that goes beyond the remedy that would be available at law. The difference between the two is what acts as the deterrent against breach. But the court will not enforce an agreed remedy that imposes a significantly greater detriment than the default remedy. What counts as disproportionate will depend on the primary obligation in question and the context in which the contract was concluded.

iii.  The Presumption in Favour of Freedom of Contract In Makdessi, Lord Neuberger and Lord Sumption held that there was a ‘strong initial presumption’ that ‘[i]n a negotiated contract between properly advised parties of comparable bargaining power … the parties themselves are the best judges of what is legitimate in a provision dealing with the consequences of breach’.122 This presumption was determinative in Makdessi itself. The parties were ‘sophisticated, successful and experienced commercial people bargaining on equal terms over a long period with expert legal advice’ and, in those circumstances, the parties at the time of contracting were the best judges of their commercial interests.123 It might be said that there is some tension between this approach and their Lordships’ insistence that the penalty rule ‘is substantive, not procedural. It does 117 ibid [82] (Lord Neuberger and Lord Sumption) and [280] (Lord Hodge). 118 ibid [97]–[99] (Lord Neuberger and Lord Sumption), [198]–[199] (Lord Mance) and [285] (Lord Hodge). 119 ZCCM Investments Holdings plc v Konkola Copper Mines plc [2017] EWHC 3288 (Comm) [37] (Lionel Persey QC sitting as a Judge of the High Court); Cargill International (n 88) [50] (Bryan J); ICICI Bank UK plc v Assam Oil Co Ltd (n 90) [32] (Andrew Burrows QC sitting as a Judge of the High Court). 120 See section II. 121 See section V. 122 Makdessi (n 4) [35] (Lord Neuberger and Lord Sumption). 123 ibid [75] and [82] (Lord Neuberger and Lord Sumption) and [282] (Lord Hodge).

Penalties  229 not normally depend for its operation on a finding that advantage was taken of one party’.124 The presumption should therefore be understood not as reflecting procedural considerations but rather as a starting point that freedom of contract should be respected.125 For that reason it adds extra slack to the line between the ‘anchor’ of the default remedy and the agreed remedy. Thus parties can stipulate detriments that go some way beyond what the law would provide as a default before the penalty rule is engaged. Put another way, penalties in commercial contracts are only unenforceable if they are not just disproportionate but ‘out of all’ proportion to what the law would provide as a remedy.126 Further, as will be seen, the ‘strong initial presumption’ is particularly apt in the context of negative primary obligations.127

C.  Applying the Test When applying the proportionality test, it is important to distinguish between positive and negative performance obligations, because the remedies at law differ between them.

i.  Positive Primary Obligations The default remedy for positive primary obligations such as the delivery of goods or the rendering of services is damages. Specific performance is only ordered exceptionally where it can be shown that damages would be inadequate,128 where there is a difficulty in assessing damages, or no substitute performance can be acquired from another source.129 Where damages would be the remedy at law, the comparison required by the proportionality test is between the highest possible quantum of damages contemplated for breach of the relevant primary obligation at the time of contracting and the quantum of the agreed detriment. This, of course, is the first of Lord Dunedin’s ‘tests’,130 and it was striking in Makdessi that the Supreme Court emphasised that in such ‘straightforward’ cases Lord Dunedin’s approach would be ‘perfectly adequate’.131 Taking their cue, the courts after Makdessi have continued to apply Lord Dunedin’s first test in cases

124 ibid [34] (Lord Neuberger and Lord Sumption). cf ibid [167] (Lord Mance) and [262] and [267] (Lord Hodge). 125 cf Summers ‘Unresolved Issues’ (n 21) 115–17, who sees the presumption as giving ‘some weight’ to procedural considerations. See also Rowan, ‘The “Legitimate Interest in Performance”’ (n 98) 172–74. 126 Makdessi (n 4) [32] (Lord Neuberger and Lord Sumption). 127 See section IV.C.ii. 128 Makdessi (n 4) [30] (Lord Neuberger and Lord Sumption). 129 Beale (ed), Chitty (n 21) paras 27.018 and 27.020. 130 See section II. 131 Makdessi (n 4) [32] (Lord Neuberger and Lord Sumption); see also ibid [221] and [255] (Lord Hodge).

230  William Day involving breaches of positive primary obligations such as the late delivery of a software and construction projects.132 What about claims where it is contemplated at the time of contracting that the law’s response to a breach would be an order for specific performance? Beyond contracts for the sale of land, specific performance is an exceptional remedy that is at the discretion of the court. It would be an unusual case in which this would be contemplated at the time of contracting as being the probable remedy awarded by the court on breach.133 Moreover, specific performance is seen as a significant intrusion on parties’ freedom after contracting, because it requires positive action by the defaulting party, backed up the threat of contempt of court that can lead to a fine or imprisonment.134 It follows that, in an exceptional case when specific performance would be the baseline against which an agreed penalty is to be compared, the proportionality test is likely to be satisfied for all but the most excessive of agreed penalties. It could be argued, by analogy with relief from forfeiture,135 that the defaulting party may be in a better position to challenge the penalty if it could cure the breach and undertakes to do so, together with an offer for compensation reflecting the losses arising from the lateness of performance. If the deterrence element of the secondary obligation is disproportionate, the penalty may then be unenforceable on terms that that undertaking to discharge the primary obligations is then fulfilled. That analogy has not yet been tested thoroughly, although in Jobson v Johnson, the Court of Appeal was prepared to give relief on terms under the penalties rule.136 It is suggested that post-Makdessi, the courts are unlikely to follow this approach because of a fear that it involves rewriting the parties’ bargain. Indeed, for that reason, the Supreme Court in Makdessi indicated that Jobson should be overruled to the extent that it purportedly ‘modified a penalty clause’.137 It is suggested that this fear is unfounded: properly applied, relief on terms does not change the contract, it simply means that its primary obligations are enforced. After Makdessi, the analysis is relatively straightforward when the primary obligation is a payment obligation. This sort of obligation is enforced by way of an action for the agreed sum. Until recently, no damages claim could lie at common 132 Triple Point Technology v PTT Public Co Ltd [2017] EWHC 2178 (TCC) [265]–[266] (Jefford J), affirmed [2019] EWCA Civ 230 [71]–[72] (Sir Rupert Jackson); GPP (n 88) [67] (Richard Salter QC sitting as a Judge of the High Court). 133 Co-Operative Insurance Society Ltd. Respondents v Argyll Stores (Holdings) Ltd [1998] AC 1 (HL) 11 (Lord Hoffmann). Although sometimes parties include statements in contracts to the effect that damages would be inadequate, in the (vain) hope of encouraging the court to exercise its discretion to order specific performance: cf Quadrant Visual Communications Ltd v Hutchison Telephone (UK) [1993] BCLC 442 (CA). 134 See, eg, Co-Operative Insurance Society (n 133) 12–13 (Lord Hoffmann). 135 Shiloh Spinners v Harding [1973] AC 691 (HL) 772 (Lord Wilberforce). cf Çukurova Finance International Ltd and another v Alfa Telecom Turkey Ltd (Nos 3 to 5) [2013] UKPC 20, [2016] AC 923. 136 Jobson v Johnson [1989] 1 WLR 1026 (CA). 137 Makdessi (n 4) [283] (Lord Hodge). See also ibid [84]–[87] (Lord Neuberger and Lord Sumption).

Penalties  231 law for the late payment of a debt, but such a claim is now in principle available,138 and in any event statutory interest may be available.139 It is now legitimate for an agreed secondary obligation to seek to deter breach by stipulating a sum exceeding the principal sum due plus interest that could be claimed at law, so long as the deterrence is not disproportionate. For that reason, even though it was not expressly overruled in Makdessi, Lord Dunedin’s second test now must be treated as having been jettisoned.140 In most cases, this will mean that default interest provisions will be upheld.141 However, there have been two cases after Makdessi in which default interest has been held to be disproportionate and therefore unenforceable. In Hayfin Opal Luxco 3 SARL v Windermere VII CMBS plc,142 default interest ranging between 850 per cent and 6,000 per cent was in obiter dicta rightly held to be unenforceable, although that extraordinary rate of default interest may have been a drafting error. A more difficult result is First Personnel Services Ltd v Halfords Ltd,143 where the court held that a 2 per cent monthly default interest rate applying to wages not paid within 7 days was a penalty, on the basis that the rate was ‘greatly in excess of rates of interest generally prevailing or awarded in respect of commercial debts at any material time in relation to the matters under consideration’. But there are many commercial debts that carry that level of interest. The case is best confined to its facts, given the judge went out of his way to note the deficiency in the evidence before the court.144 The third and only other case to date in which the penalty rule has applied after Makdessi is Vivienne Westwood.145 This involved an increase in the payment of rent on any non-trivial breach of the terms of a side letter.146 The court held that this was a penalty because its effect was to increase rent not just prospectively but also retrospectively back to the beginning of the lease. The contract also purported to make the tenant separately liable for damages at law, together with interest and costs.147 In other words, the increase in the rent was entirely for deterrence purposes. Given the significant extent of that increase, it is understandable that the judge considered this to be disproportionate. Nonetheless, there is a contrast in

138 Sempra Metals Ltd v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561 [74]–[100] (Lord Nicholls); also ibid [16]–[17] (Lord Hope), [132] (Lord Scott) [165] (Lord Walker) and [215]–[217] (Lord Mance). 139 Senior Courts Act 1981, s 35A; and Late Payment of Commercial Debts (Interest) Act 1998. 140 See section II. 141 See the cases cited at n 116. 142 Hayfin Opal Luxco 3 SARL v Windermere (n 88) [140] (Snowden J). 143 First Personnel Services Ltd v Halfords Ltd (n 90) [163] (Jeremy Cousins QC sitting as a Judge of the High Court). 144 ibid [163]: ‘The evidence before me simply does not address the issue.’ This may have put the onus on the wrong party: J Edelman (ed), McGregor on Damages, 20th edn (Sweet & Maxwell 2017) para 16.032. 145 Vivienne Westwood (n 69). 146 See section III.B.i. 147 Vivienne Westwood (n 69)[63]–[65] (Timothy Fancourt QC sitting as a Judge of the High Court).

232  William Day approach between Vivienne Westwood and Makdessi itself, where an even greater detriment was imposed on the defaulting party for deterrence purposes. The distinction may lie in the fact that Vivienne Westwood involved breach of a ­positive primary obligation, whereas Makdessi involved breach of a negative primary obligation, where (as will be seen in the next section) damages are not the default remedy and the comparative exercise makes it more likely that an agreed remedy will be enforced.148

ii.  Negative Primary Obligations Both Makdessi and Beavis were cases about negative primary obligations. Makdessi involved restrictive non-compete covenants; Beavis was concerned with an obligation not to leave a car in the car park beyond two hours. The remedial starting point for negative primary obligations is prohibitory injunctive relief,149 and that is the remedy the parties should reasonably contemplate at the time of contracting. That baseline for the proportionality test makes it much more likely that any agreed penalty will be enforced by the courts. While injunctions intrude less on parties’ freedom of contract because they do not compel positive action, they are still draconian measures backed up fines and loss of liberty for contempt of court. When damages are awarded in lieu of an injunction, they must constitute a substitute for the injunction, and it is often appropriate to calculate this by reference to the sum which reasonable people in the position of the parties would have agreed as the price after negotiating a release of the covenant.150 The very existence of the agreed secondary obligation provides a ready answer to that exercise: the parties have already negotiated a price for the release of the covenants. This provides a good explanation for the ‘strong initial presumption’ articulated by Lord Neuberger and Lord Sumption that detriments negotiated between parties of comparable bargaining power should be respected, which was determinative of the appeal in Makdessi.151 While the agreed detriment imposed on Mr Makdessi was very great indeed, the fact that that detriment had been agreed between the parties when contracting suggests that it was not disproportionate to the negotiating damages available at law that might reasonably have been contemplated at the time of contracting, not least given the parties’ difficulty at the time of contracting in speculating when a breach might occur.

148 And arguably ‘raises more difficulties’: see L Gullifer and G Penn, ‘The Boundaries of a Borrower’s Freedom to Act: Negative Covenants in Loan Agreements’, ch 8 in this volume. 149 Doherty v Allman (1878) 3 App Cas 709 (HL) 719–20 (Lord Cairns LC). See also Beale (ed), Chitty (n 21) para 27.077. 150 One Step (Support) Ltd v Morris-Garner [2019] UKSC 20, [2018] 2 WLR 1353 [41]–[63] and [95(3)–(5)] (Lord Reed). cf ibid [112]–[114] (Lord Sumption). 151 See section IV.B.iii.

Penalties  233 While injunctions are the primary remedy for breach of negative covenants, sometimes they will not be awarded due to a perception of oppression.152 In those circumstances, damages would be available at common law, but negotiating damages for breach of contract are not available unless the covenants seek to protect some pre-existing asset such as land, intellectual property or ­confidentiality.153 Instead, damages must be sought on the usual compensatory basis. If that risk of oppression is reasonably foreseeable at the time of contracting, such that the parties contemplated a damages award for breach, it may be appropriate to revert to Lord Dunedin’s first ‘test’ and compare the agreed remedy with highest possible quantum of damages contemplated for breach of the negative obligation at the time of contracting.154 Conversely, common law damages should not be treated as the default remedy for the purposes of the penalties rule simply because the breach has occurred in circumstances that create a fait accompli, denying in practical terms the availability of injunctive relief. The proportionality test is applied at the date that the parties enter into the contract, not at the date of breach. The parties cannot do more than speculate about the particular circumstances of breach at the date of contracting. The proportionality test should not focus on anything more than the nature of the primary obligation in question and the context in which the contract was concluded. In most cases, however, it should not be necessary to move away from the baseline of injunctive relief or damages in lieu of an injunction, and when that baseline is applied it is highly likely that the court will conclude that the detriment is not disproportionate.

V. Justification The Supreme Court in Makdessi did not provide a positive justification for the penalties rule. Indeed, Lord Neuberger and Lord Sumption appeared to suggest that there was no positive justification for the rule: We rather doubt that the courts would have invented the rule today if their predecessors had not done so three centuries ago. But this is not the way in which English law develops, and we do not consider that judicial abolition would be a proper course for this court to take.155

Instead, the Court provided a couple of reasons why the rule should not be abolished. First, variants of the rule against penalty clauses could be found in ‘almost 152 ‘Oppression’ being an umbrella term that is capable of capturing specific concerns about indirect specific performance and the issues that arise in personal service cases: Beale (ed), Chitty (n 21) paras 27.081, 27.0883–27.084 and 27.086. 153 One Step (Support) Ltd v Morris-Garner (n 150) [91] and [95(10)] (Lord Reed). See further W Day, ‘Restitution for Wrongs: One Step Forwards, Two Steps Back’ [2018] RLR 60, 67. 154 See section II. 155 Makdessi (n 4) [36] (Lord Neuberger and Lord Sumption).

234  William Day all major systems of law’ and had been incorporated into the various international attempts to codify the law of contract, such as UNIDROIT’s Principles of International Commercial Contracts.156 But this comparative exercise is not particularly persuasive unless the analysis goes on to consider why it is that a policy against enforcing contractual penalties exists across jurisdictions.157 The Supreme Court did not do that. Second, the Supreme Court persuasively dealt with the argument that the doctrine was made redundant by statutory intervention like the Consumer Rights Act 2015. That regulation only addressed the policy concerns in question for consumer, and not commercial, contracts.158 The Justices rejected the common argument that while legislation provides for consumer contracts, the absence of such legislation for commercial contracts suggests that there exists no similar concern for business-to-business transactions (or that, if the concern does exist, it should be addressed by way of statutory intervention rather than through the common law).159 Not only was this inconsistent with the conclusions reached by the Law Commission,160 ‘[t]he fact that Parliament has not sought to abolish or amend the doctrine, despite their [sic] existence, is just as capable of being invoked in its favour’.161 However, even this argument also only takes the case for the penalty rule so far: it still does not amount to a positive justification for the rule. The starting point from which the penalties rule is said to depart is freedom of contract.162 But freedom of contract is only one side of the coin; the other side is freedom after contracting. Acknowledging one freedom without the other leads to an imbalance in explanations of English contract law. This can be seen in Sir  George Jessel’s well-known dictum in Printing and Numerical Registering Company v Sampson: [M]en of full age and competent understanding shall have the utmost liberty of contracting, and that their contracts when entered into freely and voluntarily shall be held sacred and shall be enforced by Courts of justice. Therefore, you have this paramount public policy to consider – that you are not lightly to interfere with this freedom of contract.163

156 ibid [37] (Lord Neuberger and Lord Sumption), [164]–[166] (Lord Mance) and [263]–[265] (Lord Hodge). 157 See similarly Worthington, ‘Penalty Clauses’ (n 21) 371 on this point. 158 Makdessi (n 4) [38] (Lord Neuberger and Lord Sumption), [167] (Lord Mance) and [260] (Lord Hodge). 159 See Peel, ‘Unjustified Penalties’ (n 24) 369. Similarly E Peel, ‘The rule against penalties’ (2013) 129 LQR 152, 156. 160 Makdessi (n 4) [38] (Lord Neuberger and Lord Sumption), [163] (Lord Mance) and [263] (Lord Hodge). 161 ibid [167] (Lord Mance). 162 ibid [33] and [42] (Lord Neuberger and Lord Sumption). Also Makdessi (CA) (n 19) [44] (Christopher Clarke LJ). 163 Printing and Numerical Registering Company v Sampson (1874–75) LR Eq 462 (Ch) 465.

Penalties  235 The principle is still considered ‘fundamental to our commercial law’, requiring ‘the court to respect and give effect to the parties’ agreement’.164 However, while parties are given great freedom to contract in English law, the law does not invariably allow terms to determine the parties’ freedom after contracting. The courts decline to enforce validly constituted contracts for overriding public policy reasons,165 as Sir George Jessel himself recognised.166 Most obviously, contracts that require performance illegal by the governing law or by the law of the place of performance will not be enforced; similarly, contracts requiring immoral performance will also not be enforced.167 There are a number of other doctrines that lead to courts not enforcing the contractual obligations agreed by the parties. These are underpinned by a concern to promote personal autonomy of parties even after they are in a contractual relationship. That concern means, for example, that positive primary obligations are, more often than not, not specifically enforced,168 save in cases of payment obligations, and also that negative primary obligations are not specifically enforced when oppressive.169 Similarly, as a matter of public policy, restrictive covenants in restraint of trade are prima facie invalid unless reasonable and proportionate.170 (It is striking that while the restraint of trade doctrine is a creation of the common law like the penalties rule, and invoked much more successfully than the penalties rule in the courts, its existence has never been attacked in the same way.) The Supreme Court in Makdessi correctly identified that the penalty rule is a similar creature of public policy.171 It belongs, along with relief from forfeiture, in that same category of rules that limit, in the interest of personal autonomy, the extent to which freedom of contract can affect parties’ freedom after contracting. Placed in that context, it is suggested that the penalty rule appears to be less of an aberration than its critics suggest.172 But what exactly is the underlying public policy for the rule? Its usual name – ‘the penalty doctrine’ – suggests that private parties cannot agree to punish each other; punishment is a significant encroachment on a person’s liberty and, as such, should be an activity reserved to the state. This was probably the underlying public policy as understood by Lord Dunedin. It explains why, in Dunlop, he ruled out 164 Transocean Drilling UK Ltd v Providence Resources plc [2016] EWCA Civ 372, [2016] 1 CLC 585 [28] (Moore-Bick LJ). 165 There is a real and important difference between void contractual obligations and unenforceable contractual obligations: see Stevens (n 1). 166 Printing and Numerical Registering Company (n 163) 465. 167 Beale (ed), Chitty (n 21) paras 16.011–16.105. 168 See further, eg, M Chen-Wishart, ‘Specific Performance and Change of Mind’ in Worthington and Virgo (eds), Commercial Remedies: Resolving Controversies (n 21) 98. D Kimel, From Promise to Contract: Towards a Liberal Theory of Contract (Hart Publishing 2003) 95–109. 169 See section IV.C. 170 The principles are neatly summarised in Makdessi at first instance: Cavendish Square Holdings BV v El Makdessi [2012] EWHC 3582 (Comm) [15] (Burton J). 171 Makdessi (n 4) [7] (Lord Neuberger and Lord Sumption), [131] and [162] (Lord Mance) and [243] and [250] (Lord Hodge). 172 ibid [39] (Lord Neuberger and Lord Sumption).

236  William Day the enforceability of deterrence clauses.173 Deterrence and punishment cannot be separated. A deterrence clause is a penalty clause. That is because punishment is not an end in itself;174 the prospect of punishment functions to deter wrongdoing.175 However, in Makdessi, the rule was changed – or clarified – so that deterrence clauses are enforceable. That means penalty clauses are enforceable. The usual name for the rule is therefore a misnomer: the rule only renders some penalty clauses unenforceable. Its target are those clauses that impose a punishment disproportionate to the wrong of breaching contract. Put another way, English law will permit parties to agree to penalties in the name of freedom of contract. But disproportionate penalties are too great an intrusion on parties’ freedom after contracting, and so will not be enforced. A public policy against disproportionate punishment is evident in other branches of the law. For sentences in criminal law, the overarching public policy is that they must be proportionate to the crime. A custodial sentence must be for the ‘shortest term … commensurate with the serious of the offence’,176 and no custodial sentence should be passed unless the offence was so serious that neither a fine nor community service could be justified for the offence.177 Similar principles apply in respect of fines and community sentences.178 Given this public policy limitation in branches of the law by which the state can punish for wrongdoing, it would be incoherent for the state then to facilitate parties going beyond that limitation in their private arrangements. In tort law, punitive damages must be ‘moderate’ and ‘restrained’.179 What this means is that: the governing rule for quantum is proportionality. … Where they are awarded, punitive damages should be assessed in an amount reasonably proportionate to such factors as the harm caused, the degree of misconduct, the relative vulnerability of the plaintiff and any advantage or profit gained by the defendant …180

The caution that tort law adopts as to the extent of a party’s punishment for breach of duty (assuming they ought to be punished at all) reflects the underlying public policy against disproportionate punishment. It would be wrong to abandon that caution when the duty that has been breached is contractual rather than tortious, and the penalty is an agreed sum rather than damages. As we have seen, freedom of contract has never meant freedom to contract contrary to public policy. For the avoidance of doubt, however, the recognition that parties can agree to punish each

173 See section II. 174 Sadomasochism aside. 175 Punishment is recognised to also have other functions, such as retribution, rehabilitation and protection of the public, but these are irrelevant in the contractual context. 176 Criminal Justice Act 2003, s 153(2). 177 ibid, s 152(2). 178 ibid, ss 148(1) and 164. 179 Rookes v Barnard [1964] AC 1129 (HL) 1227–28 (Lord Devlin). 180 Whiten v Pilot Insurance Co [2002] SCC 18 [73] and [94] (LeBel J).

Penalties  237 other for breach of contract, so long as it is proportionate, does not open the door to punitive damages in contract law. The primary function of contract law is to give effect to parties’ agreements. When a penalty is enforced, because it is proportionate, the parties’ agreement is being enforced. Punitive damages, by definition, go beyond what the parties have agreed because they aim to put the defendant in a worse position and the claimant in a better position than ever envisaged by the contract itself.

VI. Conclusion After Makdessi, the penalties rule should be understood as a rule not against penalties per se but rather against disproportionate penalties. This strikes a good balance between freedom on contracting and freedom after contracting. It reflects a wider public policy in English law against disproportionate punishment. The question of proportionality for an agreed penalty is to be assessed against the remedy that would otherwise be available on breach as a matter of law, as reasonably contemplated at the time of contracting. It is suggested that this way of understanding the rule provides a substantive and robust justification for the breach and secondary obligation requirements articulated by the Supreme Court in Makdessi. It also provides a workable and coherent approach to the question of what constitutes an innocent party’s legitimate interest in the enforcement of primary obligations.

238

12 Opting for ‘Documentary Fundamentalism’: Respecting Party Choice for Entire Agreement and Non-Reliance Clauses JONATHAN MORGAN*

I. Introduction Entire Agreement Clauses (EACs)1 and Non-Reliance Clauses (NRCs) have ­generated considerable litigation and learned commentary in recent years. This should come as little surprise. It is but one engagement in the wider battle for the soul of English commercial law. This must be sketched before we consider the correct approach to EACs and NRCs in particular. ‘Textual formalism’ and ‘contextual relationism’ fight for supremacy. The first position conceives contracts as exclusively a matter of the contractual documents: the content of the contract, and its true meaning, is derived from the signed text and nowhere else. It is challenged by ‘relational contract’ scholarship. Its adherents argue that the narrow formalist tradition leaves out so much as to be quite misleading. Particularly in long-term contracts involving close cooperation, contractual documents have ever-decreasing relevance as the relationship evolves. Prior to a formal ‘contract’ (as epitomised by the signature), negotiation is a gradual process: more like turning a dimmer switch than flicking on a light. The relational critique has a common complaint. Much more emphasis needs to be placed on the real expectations and relations of contracting parties, and much less on the brittle facade of documentary ‘agreement’. It is no easy matter to resolve this grand battle of paradigms. In the inevitable cliché, there is truth in both. The frequent sociological accuracy of the gradual process leading up to ‘agreement’, or evolving relationships ‘post-agreement’, * I am grateful to Michael Godden (Norton Rose Fulbright) for his insightful commentary at the conference, and also to Stelios Tofaris. 1 Often called ‘merger’ clauses by American lawyers.

240  Jonathan Morgan is hard to deny. Its inconsistency with traditional textual contract law is obvious. But the law should not hasten to translate the ‘relational’ findings directly into doctrine. At the very least this would neglect the basic fact that many parties do, nonetheless, deliberately choose to enter into detailed written contracts, formally executed as their binding agreements. Is there disjunction between what contracting parties legally ‘say’ and how they actually behave? If so, to which register of parties’ expectations (the formal/textual, or the relational/contextual) should the law give priority? Compromise is not inevitably a virtue. As the national sadness of Brexit indicates, compromise may please absolutely no one. A ‘pluralist’ solution risks a chaotic multiplicity of approaches absenting rules about when precisely contract law should be ‘formalist’ and when ‘relational’. In commercial law it is typically better for the law to be clear than to be right. There is much to be said for making the default approach the clearest one available. If nothing else, the formalism of traditional doctrine makes fewer demands on the court and yields more predictable results. Its sheer familiarity gives it the value of stability too. The conservative’s credo is that unless it is necessary to change, it is necessary not to change. However misguided this might be as a political maxim, it should guide commercial law. This is a partial prima facie defence of the status quo. Alongside stability, however, commercial law rightly prizes party autonomy. Given this, formalism should be retained only as contract law’s (clear and predictable) starting point. It should be the ‘default’ approach, but no more. Parties that indicate a preference for a relational/contextual approach to their agreements should have their choice fully respected. The mistake made by some traditionallyminded contract lawyers is to insist that the formalist approach is not just a sensible (yet rebuttable) default but the compulsory framework for every contract. The former is defensible, the latter certainly is not. There is much to be said for clear and simple basic rules. But where contracting parties wish their agreement to be defined, interpreted and developed in a more flexible and relational way, the courts must do so. It does not lie beyond the courts’ capacity: arguments denying enforcement of express good faith clauses (etc) have been convincingly criticised by Sir George Leggatt.2 For a court to adopt a formal documentary approach despite its rejection by the parties would indeed be ‘fundamentalism’.3 And mistaken. The mirror-imagine mistake would be taking a flexible-contextual approach to a contract cast in a formal documentary register: ‘relational fundamentalism’ perhaps. The central point is that party choice must ultimately determine the correct approach. The law needs some sort of fall-back (‘backstop’?) position for parties that fail to register any clear choice. I have suggested that the starting point should be the traditional formal approach, but there are respectable arguments 2 G Leggatt, ‘Negotiation in good faith: adapting to changing circumstances in contracts and English contract law’ [2019] Journal of Business Law 104. 3 cf G McMeel, ‘Documentary Fundamentalism in the Senior Courts: The Myth of Contractual Estoppel’ [2011] LMCLQ 185.

Entire Agreement and Non-Reliance Clauses  241 against this. What is non-negotiable is that parties who do choose and expect a narrowly textual approach must be given it by the courts (no less than parties who opt for flexible relationism should be given that). In the end, subject to overriding public policy, party autonomy is paramount. This brings us back to EACs and NRCs. These should be upheld and enforced. In accordance with the general argument above, parties are opting for a formal approach to defining their contractual relations, that is, signalling that the documents are to supersede all informal statements in prior negotiations (whether ‘collateral warranties’ or ‘misrepresentations’).4 It is perfectly understandable why rational parties might sometimes wish to do this. Their choice should be respected. Only very strong public policy arguments can justify overriding it. But English law currently has an imperfect commitment to enforcing EACs and NRCs. Most insidious is the suggestion (commonly made) that because these clauses are commonly found in standard terms (‘boilerplate’), they cannot really represent the parties’ true intentions. If taken seriously, this argument would destroy the utility of virtually all commercial contracts – for it evidently ranges very much wider than EACs and NRCs in particular. Another contends that it is illogical, or question-begging, to give peremptory force to such a clause within a document whose validity, or exhaustiveness, is the entire question. But for the court to enforce the clauses is no less logical, and better respects parties’ choices. Third, we find the familiar common law techniques of restrictive attitudes to the incorporation and interpretation of EACs and NRCs. The former has little purchase, provided the contract has been signed. Contra proferentem construction is, I suggest, an unfortunate relic of the past. A mature system of commercial law should dispense with the hostility to exclusion clauses that underlies it and the artificial interpretations by which it is implemented. It is regrettable that contra proferentem retains support, although judicial and scholarly opinion is conspicuously divided. The final and most formidable stream of regulation is statutory. The need for protection of consumers is universally accepted. With sophisticated commercial parties the case is contestable. Yet since the Consumer Rights Act 2015, the sole remaining function of the rump of the Unfair Contract Terms Act 1977 (and the Misrepresentation Act 1967, section 3) is the regulation of commercial contracts. Courts have at times accepted drafting stratagems aimed at circumventing the statutory regulation in the present context. These courts’ instincts were correct. Reluctantly though, we have to accept that consistent with obedience to the will of Parliament, the permissive attitudes were wrong. In substance EACs and NRCs may be exclusion clauses, and so must face regulation accordingly. That this takes the form of judicial meditations upon ‘reasonableness’ rather than automatic invalidation is, naturally, of some comfort. But it is not the whole answer. 4 See C Mitchell, ‘Entire Agreement Clauses: Contracting Out of Contextualism’ (2006) 22 Journal of Contract Law 222; R Calnan, ‘Controlling Contractual Interpretation’, ch 4 in this volume (arguing that parties should (and should be permitted to) expressly choose how a contract should be interpreted).

242  Jonathan Morgan There is a strong case for repeal or curtailment of the legislation mentioned; it is regrettable that this received no attention from the Law Commission in its (admittedly consumer-focused) overhaul of the unfair terms legislation.

II.  Purpose and Use The amount of litigation suggests that both EACs and NRCs are in widespread use (a more rigorous empirical analysis is presented towards the end of the present section). Their common theme is limiting liability, whether contractual or non-contractual (ie for misrepresentation), to the undertakings and statements comprised in the contractual documents. This is commercially understandable. First, definitive written contracts are convenient. Whether for guiding subsequent performance or resolving any disputes that arise about it, one obvious source of friction is removed if there can be no disputes about what was actually agreed. Conversely there is clear potential for factual dispute about what was said in negotiations leading up to the contract. Even if all the negotiations were in writing, a large volume of draft clauses, collateral questions and discussions will inevitably pose evidential difficulties. Can a collateral warranty, or an actionable misrepresentation, be discerned from the parties’ correspondence? Effective EACs and NRCs make such inquiries irrelevant. Courts accept the simplifying advantages. Lightman J held that an EAC precludes statements in negotiations from having contractual effect unless incorporated into the final document, famously commenting that this prevents litigious parties ‘threshing through the undergrowth and finding in the course of negotiations some (chance) remark or statement (often long forgotten or difficult to recall or explain) on which to found a claim such as the present to the existence of a collateral warranty’.5 The advantages of the canonical document are heightened given the realities of contract negotiation. Contracting parties rarely now conduct negotiations ‘themselves’ and large enterprises never do – and indeed could not (short of direct involvement by the board of directors). The simple world of the Victorian cases has vanished, save in the merciful simplification of pedagogical examples. What actually happens, of course, is that deals are negotiated by agents on behalf of companies (principals). Such a negotiator (agent) may offer factual statements, advice, undertakings or other inducements to the counterparty (agents are frequently ‘paid by results’, ie on commission, and are accordingly keen to secure agreement). Usually, however, agents are not empowered to change the written text of the contract. Whether standard terms of business or a bespoke draft, the text will be the product of careful consideration and, often, professional legal advice. It is understood that changing the text requires authorisation

5 Inntrepreneur

Pub Co Ltd v East Crown Ltd [2000] 2 Lloyd’s Rep 611 [7].

Entire Agreement and Non-Reliance Clauses  243 at a senior level within the organisation (likely to be granted only on further legal advice). By limiting the contract (and auxiliary liabilities) to the officially sanctioned document, EACs and NRCs conduce to ‘intra-organisation’ control. The official document more closely reflects the ‘intentions’ of the ‘company itself ’ (ie authorisation at a senior level) than do remarks made in negotiations, given the reality of who is actually negotiating. Of course the evidential difficulties (who said precisely what), and those of control, are heightened in a large organisation with numerous agents. Agents obviously differ greatly over how far they are integrated within the firm in question (ie the putative defendant). An employed sales representative (typifying the examples given in the previous paragraph) is fairly closely integrated and controlled. At the other extreme is a party like the insurance broker in HIH Casualty Insurance.6 Given the particular features of the market for film finance insurance, the brokers in that case could easily have been viewed as agents for the insurer rather than agents for the assured. Hence the House of Lords unhesitatingly accepted how commercially sensible it was for the assured (Chase Manhattan Bank) to wish to ‘insulate’ itself from any misrepresentations or non-disclosure by the brokers (Heaths) to the insurers. The House unanimously agreed that this insulation extended to negligence on the brokers’ part. In dissent, Lord Scott would have accepted that in disclaiming ‘any liabilities’, Chase had even insulated itself (and its insurance) from fraud on Heaths’ part. Heaths (as specialist brokers) had much more knowledge about the underlying transaction and its risks than the assured bank (which was financing the films in question). It was understandable that the insurer would agree to exempt the assured bank over matters everyone knew lay outside the bank’s knowledge and control. The clause in HIH Casualty did not exempt the agents/brokers (Heaths) from their own direct liability for negligent or fraudulent misstatement. The court recognised that it may be commercially sensible to distance the principal (and the validity of the contract as a whole) from the misconduct of a negotiating agent. When the agent is itself a substantial commercial party (in HIH Casualty Heaths, the brokers, were a company at arm’s length from the financing banks), the preservation of claims against the agent is not a merely theoretical consolation, as it would be against a man of straw like the typical sales agent.7 These points are to some extent fact-specific; commercial contexts vary enormously. The general point is that there may be perfectly sensible reasons to agree to an EAC, NRC or (as in HIH Casualty) a related sui generis clause insulating parties from statements made in negotiations. Not all parties will agree to such limitations of liability. But many will, and do. It can be to the parties’ mutual benefit. 6 HIH Casualty and General Insurance Ltd v Chase Manhattan Bank [2003] UKHL 6, [2003] 1 CLC 358. 7 Hence, per Lord Bingham, ibid [13], ‘[t]he insurers for their part might reasonably accept this chink in their armour, recognising that their rights against Heaths in such an eventuality would remain unimpaired’.

244  Jonathan Morgan Sometimes they may be taken to derive equal advantage – at least ex ante (before a dispute arises) – from the evidential and agent-controlling features of such clauses. Even if the situation appears one-sided (eg only the assured bears duties of disclosure in the insurance context), the party agreeing to give up claims through an EAC or NRC may do so because it pays. That party can reasonably demand a price discount for accepting a contract with weaker rights. The price of a contract is not only a function of the quality of the goods or services obtained, but also a function of all of the (contingent) rights and claims conferred by the contract. In homely terms, horse-traders always paid premium prices for horses whose fitness was guaranteed compared to those sold ‘as is’ (without a warranty).8 Manufacturer’s guarantees are often sold to consumers as an additional optional entitlement in, for example, car and white goods sales – for an additional fee. The same is true for all the rights bundled within a set of written terms – or expressly excluded. As Chadwick LJ put it, ‘the price to be paid reflects the commercial risk which each party … is willing to accept’; and ‘[t]he tighter the warranties, the less the risk and (in principle, at least) the greater the price’.9 Sometimes the party excluding liability would not be willing to contract at all absent the disclaimers. In HIH Casualty, the judges accepted that unless the assured/bank was insulated from liability for the brokers’ misstatements and nondisclosures, its insurance would have been weakened to an ‘unacceptable’ degree.10 Some parties will refuse to accept the risk of un-redressed misrepresentations or of collateral promises ‘denuded of contractual effect’. Others may have a greater appetite for risk (when sufficiently compensated). After all, a party agreeing to a comprehensive set of EACs and NRCs can protect itself easily enough, by reading the contract carefully and insisting that any important statements in negotiations are incorporated into it. (The Court of Appeal upheld the ‘reasonableness’ of a NRC in the Law Society’s standard conditions for house sales, emphasising the express qualification permitting reliance on statements made in writing. A party could readily protect itself by requiring important representations to be put in writing.)11 In the landmark Peekay decision, an oral misrepresentation was in effect ‘corrected’ by the accurate information in the signed written contract. Moore-Bick LJ doubted whether it had been reasonable for a sophisticated investor to rely on the oral statement rather than the documentation.12 By signing without reading, the claimant’s representative had ‘taken the risk that the [final terms and conditions] contained subsidiary or ancillary terms that were not to his liking’; the written terms had been the ‘first and only opportunity he was given to satisfy himself that the nature of the investment and 8 See P Mitchell, ‘The development of quality obligations in sale of goods’ (2001) 117 LQR 645. 9 EA Grimstead & Son Ltd v McGarrigan (CA, 27 October 1999). 10 HIH Casualty (n 6) [19] (Lord Bingham). See further, eg, ibid [67] (Lord Hoffmann), [95] (Lord Hobhouse). 11 Lloyd v Browning [2013] EWCA Civ 1637, [2014] 1 P & CR 11. 12 Peekay Intermark Ltd v Australia and New Zealand Banking Group Ltd [2006] EWCA Civ 386, [2006] 1 CLC 582 [25], [27].

Entire Agreement and Non-Reliance Clauses  245 the terms relating to it were consistent with the broad description [the defendant’s representative] had given him’.13 Moore-Bick LJ evidently thought the claimant had only himself to blame for not taking that opportunity. As Chadwick LJ added, ‘it was necessary for [the claimant’s representative] to give careful consideration to the [written terms]’.14 Basic due diligence, surely, required it of a person described as ‘a man of substantial means who has considerable investment experience’.15 Another reason for these clauses’ frequency is commercial reaction to changes in the general law. The prevailing attitude might well be, echoing Lord Ashburton’s motion against the power of the Crown under George III, that pre-contractual liability ‘has increased, is increasing, and ought to be diminished’.16 Note first the weakened presumption that a written contract is an exhaustive statement of contractual obligations – the ‘parol evidence rule’. An EAC indicates party intention that the rule should indeed be applied – more stringently than now at common law.17 Judicial praise for the parol evidence rule’s commercial value at common law supports the case for respecting the autonomy of parties that choose such a strict regime.18 As for misrepresentation, English law’s attitude is stringent. Rescission is available even for innocent misrepresentation. While usually accepted without question, Michael Bridge argues convincingly that if this strict liability did not exist, we would surely not introduce it now.19 Lord Bingham observes that English insurance law’s disclosure obligations are ‘widely recognised to be very strict’, with the result that parties frequently seek to restrict them.20 These aspects of liability are longstanding. The Misrepresentation Act 1967 worsened the misrepresentor’s position further. Section 1 expanded the scope of rescission, removing the former bars of incorporation as a term and executed contract. Section  2(1)’s notorious ‘fiction of fraud’ has produced an inexplicably harsh damages remedy – in the deceit quantum unless the defendant can show that it took reasonable care when making the statement. Courts have so far not been willing to construe the section to avoid that questionable outcome.21 Yet the Act combined this expanded liability with restrictions on the parties’ freedom to exclude it (section  3), discussed in section VII. Taken together, the 1967 reforms were extremely onerous for defendants. The sense that EACs, NRCs and related clauses are growing in popularity is rather impressionistic – derived from the burgeoning litigation about them. One

13 ibid [52]. 14 ibid [68]. 15 ibid [2]. 16 House of Commons, 6 April 1780. 17 Scarcely a new point (at least in US): see, eg, JL Hartsfield, ‘Merger Clause and the Parol Evidence Rule’ (1949) 27 Texas Law Review 361. 18 eg Shogun Finance Ltd v Hudson [2003] UKHL 62, [2004] 1 AC 919 [49] (Lord Hobhouse). 19 M Bridge, ‘Innocent misrepresentation in contract’ (2004) 57 Current Legal Problems 277. 20 HIH Casualty (n 6) [19]. 21 Royscott Trust v Rogerson [1991] 2 QB 297.

246  Jonathan Morgan recent American study offers empirical support. Uri Benoliel examined 1,521 commercial contracts disclosed to the Securities and Exchange Commission by large publicly-traded corporations, as required by US regulations. He found that 75  per  cent contained an EAC (‘merger clause’).22 Moreover, he found a strong positive correlation between the presence of an EAC and other ‘textual’ drafting techniques, such as detailed recitals, definitions clauses and ‘No Oral Modification’ (NOM) clauses. (In another analysis of the same sample, Benoliel found that 80  per  cent contained a NOM clause.23) In both studies Benoliel infers a clear preference for textualist approaches to the formation, interpretation and variation of contracts by sophisticated commercial parties, with real-world evidence from the drafting of this large sample. (One cannot be sure that litigated contracts are typical – it is safer to assume the contrary.) To conclude, many compelling reasons exist why parties might propose, and accept, the inclusion of EACs and NRCs. Not all parties will agree, depending on: their respective appetites for bearing risk, the stringency of the proposed clauses, their respective bargaining power, and so forth. But when such a clause has been included, the choice should be respected. Where the clause fits with the commercial purpose of the deal (likely for the reasons discussed), ‘[t]he court has no right … to make a different contract for the parties’, that is, one making a different allocation of risk.24 One would have thought such an elementary proposition of commercial law uncontroversial. In fact, such clauses have received a hail of criticism from commentators and legal regulation. This chapter now considers the objections, and rejects them.

III.  Objection (1): Fraud and Public Policy Sufficiently pressing public policy may override party autonomy. However, ‘there is nothing inherently contrary to public policy in parties agreeing to contract on the basis that certain facts are to be treated as established for the purposes of their transaction, although they know the facts to be otherwise’.25 This was accepted by the Privy Council in a case of estoppel by deed. That doctrine’s rationale is to uphold the ‘formality and importance of the agreement’, which equally applies to factual declarations within a simple (non-deed) contract.26 It is, exceptionally, the force of agreement and not detrimental reliance 22 U Benoliel, ‘The Interpretation of Commercial Contracts: An Empirical Study’ (2017) 69 Alabama Law Review 469. 23 U Benoliel, ‘The Course of Performance Doctrine in Commercial Contracts: An Empirical Analysis’ (2018) 68 DePaul Law Review 1. 24 HIH Casualty (n 6) [126] (Lord Scott). 25 Prime Sight Ltd v Lavarello [2013] UKPC 22, [2014] AC 436 [47] (Lord Toulson). 26 N Goh, ‘Non-reliance clauses and contractual estoppel: Commercially sensible or anomalous?’ [2015] Journal of Business Law 511; J Braithwaite, ‘The origins and implications of contractual estoppel’ (2016) 132 LQR 120. cf Prime Sight Ltd v Lavarello (n 25) [30].

Entire Agreement and Non-Reliance Clauses  247 that prevents going back in these kinds of estoppel. It provides a sound conceptual basis for estopping a claimant from claiming misrepresentation in the face of a NRC, or raising a collateral warranty when the contract states that no contractual promises have been made other than those within the formal document (EAC). That the actual history of the transaction was different is not to the point – if the clauses were enforced if and only if they accorded with the conclusion that the court would otherwise reach on all the evidence, they literally would have no point. That the parties knowingly agree a basis contrary to the actual history is not of itself against public policy. Fraud is the one point at which ‘public policy’ bites. HIH Casualty unanimously approved the longstanding rule preventing a party from excluding liability for its own knowingly false statements. But excluding liability for one’s agent’s fraud was less clear (Lord Hoffmann thought counsel’s attempt to derive a majority opinion from the leading case had ‘more in common with reading tea leaves than with legal reasoning’).27 A majority of the House held that properly construed, the clauses did not extend to the agent’s (insurance broker’s) fraud. Lord Bingham (with Lord Steyn’s concurrence) and Lord Hoffmann therefore found it unnecessary to decide whether there was a ‘rule of law’ (not just one of construction) to this effect.28 Lord Hobhouse thought it would not be permitted even had clear words been used.29 Lord Scott, dissenting, thought the wording clear enough to exclude the agent’s fraud and rejected the supposed public policy rule.30 The point therefore remains open. There is much to commend Lord Scott’s approach. Lord Bingham cited New York authority enforcing such agent’s-fraud disclaimers, unconvinced that English law should concur.31 To New York may be added Delaware. In the leading case ABRY v F&W,32 Strine V-C accepted a ‘strong and venerable’ public policy against fraud, founded largely on ‘the societal consensus that lying is wrong’, bolstered by economic analysis. But the learned judge saw nothing immoral in a defendant broker, selling a company to a sophisticated investor, ‘creat[ing] exculpatory distance between itself and the company’, extending to ‘intentional lies by the company’s managers to the Buyer’.33 The buyer would still have redress against the traded company for its directors’ fraud. Strine V-C emphasised sophisticated investors’ autonomy, foundational to ‘Delaware law’s goal of promoting reliable and efficient corporate and commercial laws’.34 Such parties were quite capable of deciding ‘the risk they should bear and the due diligence they undertake … [and] to price factors such as limits on liability’.35

27 HIH

Casualty (n 6) [80]; S Pearson & Son Ltd v Dublin Corp [1907] AC 351. Casualty (n 6) [16]–[17], [76]–[82]. 29 ibid [98]. 30 ibid [119]–[126]. 31 ibid [17]. 32 ABRY Partners V LP v F & W Acquisition LLC 891 A2d 1032 (2006) (Del Ch). 33 ibid 1063. 34 ibid 1060. 35 ibid 1061. 28 HIH

248  Jonathan Morgan Moreover, for the plaintiff to advance a fraud claim against the defendant broker, having voluntarily accepted that the defendant would not be liable for the company’s fraud, was itself a kind of fraud. The plaintiff ‘proves itself … a liar in the most inexcusable of commercial circumstances: in a freely negotiated written contract’.36 That the two leading American commercial jurisdictions permit exclusion of liability for an agent’s fraud gives strong support for Lord Scott’s position, pace Lord Bingham. Kevin Davis points out that despite the widespread acceptance of principals’ legal liability for agents in contract law, it is of course strict liability.37 The principal/employer may bear no moral responsibility whatever for the agent’s wrongdoing. In particular, the moral condemnation of fraudsters does not properly extend to a principal, save when complicit in the fraud. Therefore parties should not be precluded on moral grounds from agreeing to exculpate the (nonfraudulent) principal. Exclusions may be positively advantageous if, for example, the counterparty is better able to detect the agent’s fraud than is the principal. Davis accordingly rejects the sweeping claim in §260 of the Restatement (Second) of Agency that a principal should never benefit from the fraud of his agent. Davis would also reject the simple invocation of the maxim fraus omnia corrumpit. Lord Bingham, in HIH Casualty, approved this ‘old legal rule’ of treating fraud as ‘a thing apart’, further citing Denning LJ’s statement that fraud ‘vitiates judgments, contracts and all transactions whatsoever’.38 But English law has not been quite so absolute. For example, in a case not cited by or to the House of Lords in HIH Casualty,39 Chitty J once accepted that parties could effectively make the certificate of a (third party) architect final and unchallengeable, even if fraud were alleged against the architect.40 Having referred to the ‘evils’ of the ‘innumerable’ challenges typically raised against such certificates, Chitty J thought that ‘those who deal in matters of this kind are wise in making the clauses more and more stringent’,41 to ensure finality. Even if the clause looked ‘terrific’ to a lawyer, businesses were entitled ‘as grown up men’ to contract in those terms. Fraud by one of the parties themselves would be ‘entirely different’. Although this decision has since been doubted on rather different grounds,42 it shows that a third party’s fraud need not ‘unravel all’. For the reasons set out by Chitty J, Strine V-C and Lord Scott, there are strong reasons to respect parties’ decision to insulate a principal from liability for an agent’s fraud. (We return later to the

36 ibid 1058. 37 K Davis, ‘Licensing Lies: Merger Clauses, the Parol Evidence Rule and Precontractual Misrepresentations’ (1999) 33 Valparaiso University Law Review 485. 38 HIH Casualty (n 7) [15]; Lazarus Estates Ltd v Beasley [1956] 1 QB 702, 712. 39 cf HIH Casualty and General Insurance Ltd v Chase Manhattan Bank [2001] EWCA Civ 1250, [2001] CLC 1853 [159] (Rix LJ). 40 Tullis v Jacson [1892] 3 Ch 441. 41 ibid 444. 42 eg Czarnikow v Roth, Schmidt & Co [1922] 2 KB 478, 488 (ousting the supervisory jurisdiction of the court).

Entire Agreement and Non-Reliance Clauses  249 ‘construction’  argument: the degree of explicitness necessary to exclude fraud, assuming that it is in principle allowed.)

IV.  Objection (2): ‘Mere Boilerplate’ Some would dismiss EACs and NRCs (and much else besides) as mere ‘­boilerplate’ – lawyer-speak buried in the standard terms and conditions that very few contracting parties read or understand. Legally they might be bound by these terms under the ‘signature rule’.43 But since the case for enforcing the clauses rests on party autonomy, as claimed in this chapter, what if the parties’ ‘consent’ is a legal fiction? Such points can be allied with the relational and contextual critique. What parties say and do in negotiations (and subsequent performance) should arguably have priority over the formal document. Those actions reveal the parties’ true expectations – a fortiori as formal contracts become longer, denser and everless comprehensible. Concerns about standard form contracts are scarcely new, and obviously not limited to EACs and NRCs. The criticisms have borne fruit in the scope of the UK unfair terms legislation, for example. This ranges far beyond our immediate subject matter. But EACs and NRCs do raise particular concerns. They aim to make the documentary contract watertight – to armour-plate the boilerplate. Let us accept that either boilerplate needs to be rejected as not truly consensual, or that contract lawyers need to re-think their paradigm of consent. The former is conceivable, and may even be feasible. The law could refuse to enforce any boilerplate terms on principle and replace them with the law’s own default rules.44 Whatever the virtues of this suggestion in a consumer context, I will say only that this displays a misplaced confidence in courts’ competence to lay down rules for the regulation of commercial relationships, compared to commercial parties’ capacity for self-regulation (and that of those who draft their contracts, including ‘boilerplate’ terms). That response assumes, again, that boilerplate (in the commercial context at least) has some relationship to the parties’ intentions. The assumption can be defended. Surely all commercial contractors (and perhaps a good proportion of consumers) understand that when they agree to buy goods or services (or whatever the subject matter might be), they also acquire rights and obligations relating to the contract? At a general level they must be aware that those terms can be laid down by contract terms (in addition to, or in derogation from, rights supplied by the background law). The rule in L’Estrange v Graucob is the central legal device for importing those terms. It is doubtless highly formalistic (a signature constitutes legal assent to all the terms in a signed document, whether or not they were

43 L’Estrange 44 J

v F Graucob Ltd [1934] 2 KB 394. Gibson, ‘Boilerplate’s False Dichotomy’ (2018) 106 Georgetown Law Journal 249.

250  Jonathan Morgan read or understood). But it reflects most people’s experience. They appreciate that signing a contractual document means they will be bound by its contents. The requirement of signature for the document to govern the relationship has to some degree a ‘cautioning’ function (to use Fuller’s terminology).45 For a sophisticated party, at least, to sign without reading is to take a risk that the document contains nothing to which it might take exception. In the view of the Court of Appeal in Peekay (noted in section II), such a party cannot complain later on that its expectations differed from what was clearly set out in the written contract. This is ‘consent’ of a rather attenuated kind. The signature estops the signer from denying that it consented – which is not the same thing as positive consent. But arguably it is enough. First, while most contracts are concluded on standard forms, these are ultimately the product of deliberate design choices by whoever drafted them. That might be a legal adviser to one of the parties, or frequently the drafting department of some trade organisation. Particularly when contracting commonly takes place on a popular industry standard form, it is not outlandish to postulate positive consent to all the detailed terms of that form, even in the absence of detailed knowledge. Parties expect that the industry-standard terms are sensible. This leads on to a second point. The more that EACs and NRCs are standard features of commercial contracts, the easier it becomes to imply agreement to them. The point that usual terms may become incorporated into contracts (even without the magic of the signature) is reflected in familiar common law rules. Industry-standard terms may be implied into a contract for just this reason. By contrast, an ‘unusual’ term could not be incorporated by anything short of equally unusual efforts to bring it to the other’s attention46 (or a signature) – although the court’s substantive judgment about whether the putative term is a reasonable one plays an equal if not greater role than the characterisation as atypical.47 Given the points already made regarding the widespread use of these clauses in commercial practice, it is plausible to attribute implicit consent to them. They are far from unusual. Many, perhaps, will object that all this talk of imputed or implied consent (or  consent-by-estoppel) is an evasive admission that often parties do not positively consent to standard forms, including EACs and NRCs within them. The average commercial contractor might not be as prosaic or bucolic as the celebrated McSporran, who stated that when consigning calves or cars on the MacBrayne ferry from Islay to the mainland, there was not much time to give to reading the requisite contract (estimating it would have taken ‘half a day to read and understand the conditions and then he would miss the boat’ – doubly pointless when

45 LL Fuller, ‘Consideration and Form’ (1941) 41 Columbia Law Review 799. 46 Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd [1989] QB 433. 47 ‘onerous’ (ibid); and see generally McCutcheon v David MacBrayne Ltd [1964] 1 WLR 125 (exclusion clause; previous course of dealing (contracts signed but not read) insufficient).

Entire Agreement and Non-Reliance Clauses  251 ‘there was no other way to send a car’).48 But many would admit similar ignorance. What is the law to do? If it insisted on terms’ actually being read and understood, it is evident that standard form contracts could no longer play any role in structuring contractual relationships. As Ian Macneil put it, the economy would come to a screeching halt (even truer today when, given Internet commerce, a yet higher proportion of transactions is governed by written standard terms).49 Reverting to conditions on Islay in 1960, had every consignor or passenger taken time actually to read the documents ‘MacBrayne’s office would be packed out the door’. Similar practicalities face the ‘nuanced’ solution to opportunistic invocation of an EAC or NRC: that there should be ‘realistic notice’ of written terms that contradict or disclaim prior representations.50 We have already noted the proposal to reject all such unread (incomprehensible?) standard terms and replace them with the law’s default approach. Surely this would be even further removed from any of the parties’ expectations? Nathan Oman argues that in commercial law, this debate about consent is ultimately irrelevant.51 The question is whether giving legal effect to standard form contracts is commercially useful, irrespective of consent. Does it support the market and facilitate transactions within it? Surely it does. If parties are to govern their own transactions, this can only be achieved by the long and detailed (and thus unreadable) sets of standard forms that so many contract theorists love to hate. There are many contingencies against which to provide. Standard forms permit self-regulation with, at the very least, the possibility of choice (unlike refusing to enforce them and imposing a rule upon the parties). The party that drafts the form, or which chooses an industry-standard contract, exercises that choice. The counterparty is able to acquaint itself with the terms, and negotiate for different ones if it will. ‘Boilerplate’ can then be defended pragmatically. It is the only realistic way to permit mass self-regulation of commercial relationships. Not every deal can be tailor-made. Consent, while attenuated, is not totally fictitious. The purist may be unconvinced. This could be dismissed as pragmatism on stilts – a pale imitation of actual contract. However, commercial lawyers are more realist than theorist. There is more than a glimmer of Oman’s pragmatism in Moore-Bick LJ’s praise of the signature rule (surely well aware of its artificiality) as ‘an important principle of English law which underpins the whole of commercial life; any erosion of it would have serious repercussions far beyond the business community’.52 There is no real alternative to giving contractual effect to formally attested documents, 48 McCutcheon v David MacBrayne Ltd (n 47) 132. 49 IR Macneil, ‘Bureaucracy and contracts of adhesion’ (1984) 22 Osgoode Hall Law Review 5. 50 R Korobkin, ‘The Borat Problem in Negotiation: Fraud, Assent, and the Behavioral Law and Economics of Standard Form Contracts’ (2013) 101 California Law Review 51 (admitting this imposes ‘a transaction cost … effectively a tax on contracting’). 51 N Oman, The Dignity of Commerce: Markets and the Moral Foundations of Contract Law (University of Chicago Press, 2016) ch 7. 52 Peekay Intermark Ltd (n 13) [43].

252  Jonathan Morgan as his Lordship recognised. The inclusion of EACs and NRCs shows the drafter’s intention that the documents should be paramount. Courts should respect it.

V.  Objection (3): Logically Unsustainable? It has sometimes been argued that as a matter of doctrinal logic, for rather different reasons, neither EACs nor NRCs can have the preclusive effect required if they are to serve their functions. However, we suggest that there is nothing strictly illogical in giving them conclusive effect. Commercial convenience, and respect for party autonomy, supports their enforcement. The courts have rightly taken that approach. On EACs, David McLauchlan supports the position taken by the Law Commission in its 1986 Report on the parol evidence rule.53 The Law Commission concluded that just as the ‘rule’ itself was actually just a presumption of documentary completeness, no more could an express EAC conclusively determine that all terms had been merged into the document containing it. Professor McLauchlan emphasises that the question of whether a document is the exhaustive source of the contract is a question of fact. To enforce the EAC per se would contradict this characterisation. The EAC may be strong evidence of the intention that the document be the entire contract, but it cannot decisively prove this fact in every case. What about an express promise that an undertaking in negotiations would be a binding collateral contract notwithstanding the document? In Professor McLauchlan’s opinion, the recital of fact in a typical EAC (‘This is the entire contract between the parties’) cannot render untrue the fact that such a collateral promise was made. Especially not when the EAC ‘is often part of the boilerplate that is not read and not expected to be read’.54 To the degree that Professor McLauchlan’s argument turns on the unreality of party agreement to standard terms (‘boilerplate’), I have suggested counterarguments. There is nothing ultimately illogical in holding parties to what they have apparently agreed in a signed contract. That would here mean treating the recital of fact in an EAC as conclusive of the question – precluding inquiries into collateral undertakings. This does not mean believing in ‘magical powers’ (pace McLauchlan). Rather it flows from a pragmatic judgment that more good than harm comes from permitting parties to impose on themselves a rigorous ‘parol evidence’ regime. Professor McLauchlan himself indirectly offers powerful reasons to do so – the ‘disarray’ of the common law rule. McLauchlan notes its contentious, treacherous and confused nature given the many exceptions to it – concluding (with Eric Posner) that across the common law world, one despairs of reconciling

53 D McLauchlan, ‘The Entire Agreement Clause: Conclusive or a Question of Weight?’ (2012) 128 LQR 521. 54 ibid 531.

Entire Agreement and Non-Reliance Clauses  253 the authorities. Little wonder, then, that parties (or drafters) who desire documentary exclusivity would say so expressly and not rely on the background legal position. And if they say so, the law should hold them to it. Professor McLauchlan claims that his approach largely preserves EACs’ utility, for they will still carry great evidential weight.55 But if the purpose of the EAC is to preclude evidential inquiries ranging outside the contractual document, that purpose is disappointed even if the EAC is accorded considerable respect when undertaking the inquiry. Recall the failed attempt to give NOM clauses strong but not conclusive status. This was rejected – there were no satisfactory degrees of evidential presumption.56 A straight choice between enforcing, or virtually ignoring, NOM clauses faced the Supreme Court.57 Its decision, to enforce, has not found universal favour.58 But that MWB v Rock permits parties to self-impose a formal modification regime gives significant, albeit indirect, support for the enforcement of EACs. Professor McLauchlan’s fall-back argument is that even if EACs are enforced in line with the majority of judicial opinion, they cannot rule out claims based on rectification or estoppel. Let us accept that if the contract is rectified to remove the EAC then of course it cannot have any effect whatever; but whether the mere presence of an alleged collateral contract is sufficient for rectification may be doubted. The exception would swallow the rule. On estoppel, Professor McLauchlan cites the clear (although obiter) statement by Gray J in Ryanair Ltd v SR Technics Ireland Ltd.59 Gray J would have enforced a collateral contract between the parties, undeterred by an EAC. His Lordship reasoned that because both parties proceeded on the basis of the collateral agreement (ie took action based on its legal effect), it would have been ‘unconscionable’ to rely on the EAC. Such reliance would accordingly have been estopped.60 Gray J cited Lord Denning MR’s judgment in Brikom Investments Ltd v Carr.61 In that case an EAC did not prevent the Court of Appeal’s giving effect to what was variously characterised as a collateral contract, promissory estoppel or sui generis equity. However the point seems to have been conceded by counsel.62 Only Lord Denning referred to the EAC. Notably (in a passage quoted by Gray J), Lord Denning said that the EAC was of no effect, ‘at any rate when the circumstances are such that it would

55 See also M Barber, ‘The limits of entire agreement clauses’ [2012] Journal of Business Law 487. 56 eg Virulite LLC v Virulite Distribution Ltd [2014] EWHC 366 (QB), [2015] 1 All ER (Comm) 204 [60] (Stuart-Smith J). 57 MWB Business Exchange Centres Ltd v Rock Advertising Ltd [2018] UKSC 24, [2019] AC 119. 58 See A Burrows, ‘Anti-Oral Variation Clauses: Rock-Solid or Rocky?’, ch 3 in this volume; and R Stevens, ‘Binding Our Future Selves’, ch 2 in this volume. cf J Morgan, ‘Contracting for self-denial: on enforcing “no oral modification” clauses’ (2017) 76 CLJ 589. 59 Ryanair Ltd v SR Technics Ireland Ltd [2007] EWHC 3089 (QB), [2007] 12 WLUK 597. 60 ibid [143]. 61 Brikom Investments Ltd v Carr [1979] QB 467, 480, cited in Ryanair (n 59) [142]. 62 Brikom Investments (n 61) 474 (‘that [EAC] clause may not present the same difficulty for the equitable defence’).

254  Jonathan Morgan not be fair or reasonable to allow the landlord to rely on it: see section 3(b) of the Misrepresentation Act 1967’.63 This cryptic comment suggests that the statutory regulation rather than estoppel prevented the EAC’s being raised as a defence. At  any rate, Brikom Investments is not overwhelming authority and, as noted, Gray J’s judgment on this point in the Ryanair case is obiter.64 What should be the approach? Taken at face value, Gray J’s judgment (supported by McLauchlin) would substantially undermine the value of EACs. Whenever a party is attempting to prove a collateral contract, reliance upon it is almost inevitable. Thus the estoppel ‘exception’ also threatens to turn the flank of the EAC entirely. One way to confine it would be to investigate whether the reliance on the collateral contract was reasonable (so as to render the defendant’s invocation of the EAC ‘unconscionable’). Naturally this requires a value-laden approach. A court might legitimately conclude that when the parties were sophisticated and well-advised, it was not reasonable to rely on collateral promises and so not unconscionable to raise the EAC. Of course this would require factual inquiry, again rather undermining the point of the EAC. Further, the facts of Ryanair could be characterised this way given the parties involved, and yet Gray J (as seen) would have held the defendants estopped from using the EAC. If the estoppel argument is to be confined, a bright-line rule would be superior. As was held in the different context of section 4 of the Statute of Frauds (requirement for guarantees to be in writing), while it would not be sufficient to make an oral promise of guarantee that was relied on by the lender, a specific promise by the guarantor not to raise the Statute of Frauds as a defence would raise an ­estoppel.65 A wider conception (any relied-upon oral guarantee) would effectively have repealed the statute. With EACs the concern is respect not for Parliament’s will but for that of the parties. Applied to the present context it entails that a specific oral undertaking not to use the EAC as a defence could raise an estoppel. Or a specific agreement that the written agreement did not need to be amended to reflect a common assumption between the parties in negotiations.66 This is a more precise doctrine than that envisaged in the Ryanair case and endorsed by Professor McLauchlin. The confinement is necessary if EACs are to retain their intended effect. The enforceability of NRCs has also been questioned using ‘doctrinal logic’. This critique has little support and should be rejected. The argument runs that a misrepresentation undermines a contract’s validity (rendering it voidable); that that invalidity extends to any NRC in the contract; that the NRC therefore cannot be used to defeat the misrepresentation. This was one of Lord Hobhouse’s arguments 63 ibid 480. 64 See, for criticism of Ryanair (n 59), Mileform Ltd v Interserve Security Ltd [2013] EWHC 3386 (QB) [104]–[105] (Gloster J). 65 Actionstrength Ltd v International Glass Engineering [2003] UKHL 17, [2003] 2 AC 541. 66 Mears Ltd v Shoreline Housing Partnership Ltd [2013] EWCA Civ 639.

Entire Agreement and Non-Reliance Clauses  255 against clauses purporting to exclude liability for an agent’s fraud. Since ‘the party deceived has not given a true consent to be bound’, the other party ‘cannot rely upon a term of a contract to which he has never validly obtained the [deceived party’s] agreement’.67 Although made with respect to fraud, if correct this would make excluding liability for misrepresentation of any kind impossible. Fraud has the same effect on the validity of an agreement (ie making it voidable) as innocent misrepresentation.68 So Lord Hobhouse’s point cannot be confined to fraud. His ‘logic’ therefore proves too much, for it purports to demonstrate that no clause restricting or excluding liability for misrepresentation ever has been, or could be, legally effective. No court has proceeded on that basis. Nor indeed did Parliament in 1967: were Lord Hobhouse correct, section  3 of the Misrepresentation Act would be literally pointless, addressed to a species of clause that, according to the ‘logic’, could never be enforceable. One exception would be a misrepresentation about the effect (or existence) of the NRC itself. In line with the approach to exclusion clauses generally,69 the NRC could not be relied upon in that specific situation. This doctrine does not rest on the invalidity of the entire contract– in the leading case of Curtis, Denning LJ held it was irrelevant that rescission of the contract as whole was unavailable.70 Denning LJ viewed it as a freestanding rule (whether of law or equity) not derived from estoppel.71

VI.  Common Law Regulation: Construction Contra Proferentem As every student knows, the courts have long taken a narrow approach to the incorporation and interpretation of clauses they dislike. Some points have already been made about the signature rule. An EAC or NRC will assuredly be incorporated by signature. (Its absence engages the vagaries of ‘reasonable notice’.) As for construction, an ongoing debate about the strength and propriety of the contra proferentem doctrine surrounds exclusion clauses. Some courts have accepted that past excesses of contra proferentem were, at best, an understandable yet ‘desperate’ expedient in the common law’s battle to regulate exclusions.72 Since Parliament stepped in with direct substantive regulation, the need for

67 HIH Casualty (n 6) [98]. 68 eg Shogun Finance (n 18) [8] (Lord Nicholls). 69 Curtis v Chemical Cleaning and Dyeing Co [1951] 1 KB 805. 70 ibid 810. 71 ibid 809: the statement ‘was probably not sufficiently precise and unambiguous to create an estoppel’. cf Springwell Navigation v JP Morgan Chase Bank [2010] EWCA Civ 1221, [2010] 2 CLC 705 [99]–[104] (Rix LJ). 72 BCCI v Ali (No 1) [2001] UKHL 8, [2002] 1 AC 251 [60] (Lord Hoffmann).

256  Jonathan Morgan ‘very strained construction’ to defeat them has lessened. This first point concerns the tools for the job. Assuming that exclusion clauses are to be regulated, it is better to do this directly (by statute), openly considering whether the clause should be enforceable, rather than rewriting it through ‘presumptions’ about what the parties’ ‘must have intended’. That is the point made in the House of Lords, and even by Lord Denning MR, in the wake of the transformative 1970s legislation.73 But second, do exclusion clauses always require regulating? Between commercial parties of equal bargaining power, is it implausible to suggest that they might have meant exactly what they said – to exclude the liability (‘any liability …’) defined? HIH Casualty Insurance is again illuminating. It shows a more receptive attitude towards commercial exclusion clauses – that is, a less stringent interpretive approach. However, the court declined to reject entirely one notorious artefact of contra proferentem. Indeed, as seen, the majority resorted to narrow interpretation to avoid answering whether, in principle, a party can exclude liability for its agent’s fraud. The House of Lords considered the ‘Canada Steamships doctrine’.74 This is the prime exemplar of the making-utter-nonsense school of artificial ­construction.75 In the Canada Steamships case, a clause excluding ‘any’ liability did not exclude liability for negligence. That is linguistically surprising. In HIH Casualty, Lord Bingham nevertheless accepted that the general approach was ‘sound’.76 Yet Canada Steamships was not a rigid invariable rule or ‘litmus test’. Their Lordships unanimously accepted on the facts that, given the commercial context (described above), the phrase ‘any liability’ had to include negligent misrepresentations. As Lord Hoffmann said, there was no ‘inherent improbability’ that the parties had intended ‘any liability’ to encompass the insurance broker’s negligence: on the contrary, if the clause did not have that effect, its ‘commercial objective … would be substantially undermined’.77 As seen, the court was divided when it came to the agent’s deliberate misstatement or non-disclosure. Lord Scott alone considered that the phrase ‘any liability’ logically had to extend equally to the agent’s fraud and to its negligence: ‘The language is certainly wide enough for the purpose. Why should it not be given its literal meaning?’78 For Lord Scott, the commercial purpose (insulating the 73 eg Photo Production Ltd v Securicor Transport Ltd [1980] UKHL 2, [1980] AC 827, 851, ‘any need for this kind of judicial distortion of the English language has been banished by Parliament’s having made these kinds of contracts subject to the Unfair Contract Terms Act 1977’ (Lord Diplock); George Mitchell (Chesterhall) Ltd v Finney Lock Seeds Ltd [1982] EWCA Civ 5, [1983] QB 284 (Lord Denning). 74 Canada Steamship Lines Ltd v The King [1952] AC 192. 75 A ‘specially exacting standard’ applied to exclusion clauses, compared with ordinary contra proferentem: Ailsa Craig Fishing Co Ltd v Malvern Fishing Co Ltd (The Strathallan) [1983] 1 WLR 964, 970 (Lord Fraser). 76 HIH Casualty (n 6) [11]. 77 ibid [67]. 78 ibid [124]. Note, wide but non-explicit exclusion clause encompassed agent’s wilful misconduct (theft) during performance: Frans Maas v Samsung Electronics [2005] 1 CLC 647.

Entire Agreement and Non-Reliance Clauses  257 assured bank from the insurance broker’s misconduct) was equally compromised by making the assured liable for the broker’s fraud as for its negligence.79 But for Lord Bingham it was essential that any attempt to exclude liability for an agent’s fraud must be ‘in clear and unmistakable terms on the face of the contract’ – and salutary.80 Lord Hoffmann agreed, predicting that it was ‘extraordinarily unlikely that parties to a contract will agree a term which excludes liability for fraud with sufficient clarity to raise squarely the question of whether it should be lawful to do so’.81 Lord Hobhouse also thought fraud different in kind from negligence.82 In his view, the bank should have insured itself against liability for its agent’s fraud (rather than seeking to exclude that liability).83 The HIH case prettily exemplifies the current state of contra proferentem. The courts have mostly backed away from its grosser excesses.84 They recognise commercial reality. But judgements about what commercial parties ‘must have intended’ can still sometimes be used to cut down clear, broad words. (None of the HIH judges suggested there was anything ambiguous in the words referring to ‘any liability’.) In quite a number of the EAC and NRC cases, courts have insisted on express exclusions. Jacob J refused to accept that an EAC excluded liability for misrepresentation – it simply did not say so, and a would-be excluder could not be ‘mealy-mouthed’.85 This call for clarity is hard to gainsay. (The tentative inference of consent to boilerplate EACs and NRCs is easier to sustain when they are drafted with unmistakable explicitness.) Standardisation of contractual clauses occurs for a number of reasons. One significant factor is judicial interpretation. Once a court has issued a definitive binding interpretation on the meaning of a clause, drafters can use it with considerable confidence about its legal effect. (Alternatively, they might seek to amend it, but the definitive effect of the court’s ruling will then be lost.) One irony of contra proferentem is that clauses are given judicial interpretations that diverge from the ordinary understanding those subject to the clauses might have. The meaning becomes entrenched, well known to the cognoscenti (lawyers and courts) but impenetrable to any lay person.86 Thus, as a device to encourage clear drafting (or to push clauses in a particular substantive direction) contra proferentem is not entirely successful – given that the clause may ‘stick’ once the court has interpreted it.

79 HIH Casualty (n 6) [126]. 80 ibid [17]. 81 ibid [81]. 82 ibid [97]. 83 ibid [98]. 84 cf The Mercini Lady [2011] 2 All ER (Comm) 522 [59]–[61]; cp Hirtenstein v Hill Dickinson LLP [2014] EWHC 2711 (Comm) [55]. 85 Thomas Witter Ltd v TBP Industries Ltd [1996] 2 All ER 573, 596. 86 M Boardman, ‘Contra Proferentem: The Allure of Ambiguous Boilerplate’ (2006) 104 Michigan Law Review 1105.

258  Jonathan Morgan Other commentators have addressed the merits of contra proferentem.87 Its place in commercial law has divided the Court of Appeal.88 Consistent with the views expressed here, the court should enforce what the parties have agreed, uninfluenced by the judges’ own preconceptions about what the parties ought to have agreed. In practical terms, the lessons for parties and drafters are fairly obvious. They should spell out precisely what liabilities are being excluded. In particular, exclusion of liability for an agent’s fraud must be explicit.89 Best of all, always use the wording of an EAC or NRC that has been judicially upheld as having the legal effect consonant with the parties’ commercial purpose. These steps, combined with the commercially sympathetic stance of most English courts, should usually bring about the desired effect. But contra proferentem is not dead – a doctrine with such historical roots is not so easily cut down.90 One study by experienced American commercial attorneys concluded that it was impossible to ensure that the contractual documents were completely watertight. It depended on the clarity of the drafting, but also on the courts’ attitudes to it.91 It is a reproach to English commercial law that similar doubts would be expressed by a cautious lawyer in this jurisdiction. Most of the uncertainty, however, stems from parliamentary regulation.

VII.  Statutory Regulation Parliament has legislated about exclusion clauses. There has been considerable litigation about the amenability of NRCs to section  3 of the Misrepresentation Act 1967. There is less authority about EACs, but it seems these are subject to section 3(2)(b)(i) of the Unfair Contract Terms Act 1977.92 The issues concerning the current law fall into two groups. First, whether the legislation is applicable to a certain clause at all. Second, the court’s determination of whether a clause to which the legislation applies is ‘reasonable’. These matters have been well rehearsed and will be considered only briefly.93 To the extent that some cases suggest unhappiness with the statutory regulation, we must also ask whether the basic regulatory 87 eg E Peel, ‘Contra proferentem revisited’ (2017) 133 LQR 6; S Tofaris, ‘Commercial Construction of Exemption Clauses’ [2019] LMCLQ 270. 88 eg Hut Group Ltd v Nobahar-Cookson [2016] EWCA Civ 128, [2016] 1 CLC 573; Transocean Drilling UK Ltd v Providence Resources Plc (The GSF Arctic III) [2016] EWCA Civ 372, [2016] 1 CLC 585; Persimmon Homes Ltd v Ove Arup & Partners Ltd [2017] EWCA Civ 373, [2017] 2 CLC 28. 89 But it is hard to imagine drafting a clause that could simultaneously preserve an agent’s liability while excluding the liability of an undisclosed principal. 90 J McCunn, ‘The Contra Proferentem Rule: Contract Law’s Great Survivor’ (2019) 39 OJLS 483. 91 GD West and WB Lewis, ‘Contracting to Avoid Extra-Contractual Liability: Can Your Contractual Deal Ever Really Be the Entire Deal?’ (2009) 64 The Business Lawyer 999. 92 Axa Sun Life Services plc v Campbell Martin Ltd [2011] EWCA Civ 133, [2011] 1 CLC 312 [49]–[50]. 93 See especially H Beale and G Palmer, ‘Non-reliance Clauses, Entire Agreement Clauses and Contractual Estoppel’ (forthcoming).

Entire Agreement and Non-Reliance Clauses  259 scheme is satisfactory. It has been in place for 40–50 years without serious review (despite considerable changes in the consumer contract regime). Yet it is questionable whether, in the light of those reforms, its residual application to commercial contracts (only) is justified. Arguably, the legislation should be repealed, or at least restricted to small and medium-sized enterprises (SMEs).

A.  Scope of Application of the Legislation Following considerable controversy, the application of section  3 of the Misrepresentation Act 1967 to NRCs now seems settled. While the story is well known, and might be of largely historical interest, it is revealing – and worth revisiting briefly. It is debateable whether even now the Misrepresentation Act applies automatically in every case. Mileage remains in the argument that, sometimes, parties are genuinely defining which statements, undertakings or advice have been made with legal effect (to be relied upon as such). It is inherent in the test for the actionability of either a warranty or misstatement that it was (objectively) intended to have legal effect, and reasonable to rely on it. This refinement reintroduces considerable uncertainty. It has been powerfully argued that the legislation should be given a broad scope, relying on the ‘reasonableness test’ to respect sensible commercial allocations of risk. While there is good reason to doubt this claim, it would apparently need legislation to displace it.

i.  Minimal Scope: NRC ‘Defining’ Liability? The very notion of ‘exclusion clauses’ is unstable.94 How, if at all, is a line to be drawn between excluding a contractual liability and declining to assume the underlying obligation? If regulation of ‘exclusion clauses’ catches all allocations of primary liability then the entire substance of every agreement seems open to judicial review. This would destroy freedom of contract. But there is an equal and opposite danger. It is straightforward enough for an advertent draftsman to cast any exclusion of liability as disclaimer of the underlying liability. With misrepresentation the favoured mode being the NRC’s statement that neither party has made, or relied on, any statements in negotiations. Since this binds factually (through ‘contractual estoppel’), it prevents liability from arising. Thus one could say (with Chadwick LJ) that a NRC is not an ‘exclusion’, for the parties were surely not ‘excluding’ a liability that ex hypothesi could never arise.95 Hence section  3 of the Misrepresentation Act 1967 does not apply. This supplies a ready path to 94 eg B Coote, Exception Clauses (Sweet & Maxwell 1964); N Palmer and D Yates, ‘The Future of the Unfair Contract Terms Act 1977’ (1981) 40 CLJ 108; J White, ‘Defining “exclusion” clauses and excluding “defining” clauses: the need to clarify the scope of the Unfair Contract Terms Act 1977’ [2016] Journal of Business Law 373. 95 Watford Electronics Ltd v Sanderson CFL Ltd [2001] EWCA Civ 317, [2002] FSR 19 [41].

260  Jonathan Morgan draft round section  3 entirely. Parliament cannot have intended such evasion. Moreover, it would apply in consumer cases as in the commercial context where Chadwick LJ made his decision. An earlier generation of authorities had (unsurprisingly) rejected such contentions – one declaring that ‘Humpty Dumpty would have fallen for this argument. If we were to fall for it [the 1967 Act] would be dashed to pieces’.96 Watford Electronics is quite inconsistent with the earlier cases.97 It has now been repudiated.

ii.  Maximal Scope: All NRCs ‘Exclusions’? The law now approaches a position where every NRC is an exclusion clause subject to review for ‘reasonableness’ under section 3 of the Misrepresentation Act 1967. Looking at the substance rather than the form, the NRC precludes a misrepresentation claim that would otherwise arise. What is this other than an exclusion clause?98 Of course that does not make every NRC invalid. Section 3 requires an assessment of reasonableness by the court. Nobody doubts the commercial sense and sensibility of the English judiciary. Still, many would agree with Christopher Clarke J’s observation that it is ‘obviously advantageous that commercial parties of equal bargaining power should be able to agree what responsibility they are taking (or not taking) towards each other without having to satisfy some reasonableness test’.99 The question is whether any stable compromise can be established consistent with the legislation.

iii.  A compromise? Some commentators, like James White, argue that courts should resist either of the extreme positions sketched so far. Such compromise sounds immediately attractive. But White’s own position (deciding on a case-by-case basis whether in substance a given clause excludes or defines liability) is open to the obvious objection of uncertainty. Is it not better for the law to be clear than just, as asserted at the start of this chapter?100 A similar objection can be made to one suggested compromise position on NRCs. Should a professional adviser not be entitled to clarify when it is speaking only informally (sales talk), and when it is formally tendering information and

96 Cremdean Properties Ltd v Nash (1977) 244 EG 547. 97 LK Ho and TB Mathias, ‘Basis clauses and the Unfair Contract Terms Act 1977’ (2014) 130 LQR 37. 98 eg Springwell [2010] EWCA Civ 1221 [182]; First Tower Trustees Ltd v CDS (Superstores International) Ltd [2018] EWCA Civ 1396, [2019] 1 WLR 637. 99 Raiffeisen Zentralbank Osterreich v RBS [2010] EWHC 1392 (Comm) [313] (emphasis added). 100 An argument said to have been overlooked in this section of the chapter, by the chair of the panel at the ‘Contents of Commercial Contracts: Terms Affecting Freedoms’ UCL Conference (9–10 May 2019), Lord Justice Leggatt.

Entire Agreement and Non-Reliance Clauses  261 advice on which the client can legally rely? In Carney v NM Rothschild & Sons Ltd, the defendants suggested how potential clients could minimise inheritance tax liabilities on occasions which included a luncheon and a cocktail party.101 Such relatively informal meetings are an unpromising basis for legal liability. The formal signed contract put this beyond doubt. Judge Wacksman stated such ‘basis’ clauses ‘serve the useful function of removing a grey area as to what might or might not be [actionable] representation’.102 He thought this ‘very apposite here where many of the alleged representations were given orally in a quasi-social setting and where differences of emphasis could make all the difference’.103 Therefore the NRC had defined the status of the words as informal (‘mere sales talk’); they were not exclusion clauses to which section 3 of the 1967 Act applied. Although Carney was not cited in the subsequent leading case First Tower Trustees, Lewison LJ accepted that ‘of course’ a supposed ‘exclusion’ may, on its proper interpretation, be defining the transactional basis.104 Leggatt LJ warned against treating the label ‘basis clause’ as a new means of formally evading the statutory controls.105 In Carney, Judge Wacksman explained that whether a clause truly defined the basis of the parties’ relationship is a ‘multifaceted’ question of substance, and set out a non-exhaustive list of four factors that would inform the inquiry.106 In principle this is correct. While the line will doubtless be elusive in practice, the alternative is a blanket rule that every NRC is a reviewable exclusion clause. As for ‘better clear law than correct law’, that would purchase certainty by drastically curtailing party autonomy. However, another attempt to limit the scope of section 3 was always dubious and has now been overruled. In Christopher Clarke J’s leading analysis in the Raiffeisen case there was a glimmer of a distinction based on the parties’ identity – their level of sophistication. Bankers might truly agree that their transaction legally proceeds on a (deemed) basis quite different from what actually took place; whereas to tell ‘the man in the street’ that a car was sound while agreeing that you had told him nothing of the sort would ‘rewrite history or [part] company with reality’.107 At first sight this distinction is again most attractive. It restores freedom of contract for sophisticated parties, while preserving the statutory regulation in consumer (and perhaps small business) contexts. Yet any sharp consumer-commercial dichotomy was never consistent with the statute. The Misrepresentation Act 1967, section 3 originally made no distinction between commercial and consumer contracts. To decant commercial exclusions from its scope was therefore inconsistent with its undifferentiated provisions. 101 Carney v NM Rothschild & Sons Ltd [2018] EWHC 958 (Comm). 102 ibid [354], citing Crestsign v NatWest [2014] EWHC 133 [115]–[117]. See also Carney v NM Rothschild & Sons Ltd (n 101) [80] (‘mere sales talk’). 103 Carney v NM Rothschild (n 101) [354]. 104 First Tower Trustees Ltd (n 98) [42]–[44]. 105 ibid [95]. 106 Carney v NM Rothschild (n 101) [94]. 107 Raiffeisen Zentralbank Osterreich v RBS (n 99) [314]–[315].

262  Jonathan Morgan The inconsistency is now greater still. Since the Consumer Rights Act 2015 came into force, section  3 of the 1967 Act applies only to commercial contracts (see section  3(2) (as inserted)). It would be absurd to elevate Christopher Clarke J’s dicta to dispositive status, excising from the 1967 Act the only NRCs (those in commercial contracts) to which the legislation still applies. (It must be stressed that the learned judge did not take his discussion so far – highlighting the priority of ‘substance not form’.108) All of these points have recently been made by Leggatt LJ with typical clarity and force.109 Parliament has legislated that clauses excluding misrepresentation liability are subject to a reasonableness test – even in commercial contracts. It is not for the courts to question the wisdom of this, even less to amend the statute by narrow construction. The result may be uncommercial, but it is the fault of the legislature. As Lord Holt CJ said, ‘an Act of Parliament can do no wrong, though it may do several things that look pretty odd’.110 Champions of freedom of contract must criticise the legislation, not its loyal interpreters. The vital question is what ‘reasonableness’ means in practice.

B.  Statutory Reasonableness Test The cases are replete with statements acknowledging that commercial parties are the best judges of their own interests. (Moreover, ‘[t]here remains much harshness in the world, and [sophisticated business] entities are unlikely candidates to place at the head of the line for judicial protection’.)111 Sophisticated contractors are not (as seen) exempt from the legislation, but their sophistication is most relevant to the fairness of exclusion clauses.112 However, that is not a complete answer. The court must always assess the clause’s substantive fairness. It cannot simply defer to the commercial parties’ judgement in every case. That would effectively disclaim the jurisdiction. And there are indeed commercial cases in which NRCs have been found ‘unreasonable’ – so the jurisdiction is not theoretical. The risk is heightened by the familiar doctrine that appellate courts do not reassess reasonableness here de novo.113 This has inevitably led to senior courts upholding the ‘unreasonableness’ of clauses they would otherwise have approved. In Cleaver v Schyde Investments Ltd, the trial judge had struck down as unreasonable the Law Society’s Standard Conditions of Sale excluding liability for vendors’ errors and omissions (unless fraudulent or reckless). The Court of Appeal held the judge entitled to this view – even though Etherton LJ gave five powerful



108 ibid

[310]. Tower Trustees Ltd (n 98)[103]. 110 City of London v Wood (1706) 12 Mod 669, 687–88. 111 Abry Partners V LP (n 32) at 1062. 112 First Tower Trustees Ltd (n 98)[105] (Leggatt LJ). 113 George Mitchell (Chesterhall) Ltd (n 73) 816 (Lord Bridge). 109 First

Entire Agreement and Non-Reliance Clauses  263 reasons why the exclusion had been ‘a perfectly rational and commercially justifiable apportionment of risk in the interests of certainty and the avoidance of litigation’.114 Longmore LJ (concurring) said it was ‘a good example of the width of the interference with freedom of contract generated by s 11 of [the Unfair Contract Terms Act 1977]’.115 In First Tower Trustees, the Court of Appeal not only declined to interfere with but positively approved the trial judge’s finding of unreasonableness. A landlord allegedly failed to disclose the presence of asbestos in a building, contrary to the requirements of the Commercial Property Standard Enquiries form. There were two NRCs. One, in the agreement preceding the lease, stated that the tenant had not relied on any statement made on behalf of the landlord, subject to two important qualifications: first, for statements ‘made in writing by the landlord’s solicitors’; second, for ‘fraudulent misrepresentation’. Both courts found this qualified NRC reasonable. But the second NRC, in the lease itself, went further. It simply stated that the tenant had not relied on any statements made by or for the landlord (ie without any ‘carve outs’). The trial judge found this unreasonable. The Court of Appeal agreed. Were the clause to be upheld, ‘the important function of replies to enquiries before contract [in conveyancing] becomes worthless’.116 The lesson seems clear. Parties can validly limit their liability to ‘replies to inquiries’ embodied in formal fashion (such as a solicitor’s letter). Attempting to exclude liability altogether, however, will be unreasonable. But the robust believer in freedom of contract will ask: If the parties wanted a qualified NRC why not say so? If they have not said so, why should the court impose on them what they did not want (or give the claimant for free what it could have obtained – and paid for – in the negotiations)? In vain did the defendant landlord in First Tower Trustees argue that the tenant, as a ‘substantial concern’ represented by solicitors, could and should have negotiated for a carve-out in the lease NRC similar to the one in the preceding agreement, had it wanted such protection.117 Even if commercial drafters were confident that courts rarely deem exclusions unreasonable118 – that is, intervene if not never then ‘hardly ever’ – this is less reassuring than it sounds. As Rex Adhar argues, the ‘never say never’ approach to contract doctrines generally (leaving room for an ‘exception’ where the merits demand) creates disproportionate uncertainty.119 The force of this point was accepted by Lord Hoffmann in Union Eagle Ltd v Golden Achievement Ltd.120 The jurisdiction to assess NRCs’ (and similar clauses’) ‘reasonableness’ under section 3 114 Cleaver v Schyde Investments Ltd [2011] EWCA Civ 929, [2011] 2 P & CR 21 [38]. 115 ibid [54]. 116 First Tower Trustees Ltd (n 98)[75] (Lewison LJ). 117 ibid [74] (cf [75], ‘Although there is some force in the landlords’ argument …’). 118 eg ibid [67], ‘in cases involving the sale of complex financial products to sophisticated investors [a NRC] may well be [reasonable]’. 119 R Adhar ‘Contract doctrine, predictability and the nebulous exception’ (2014) 73 CLJ 39. 120 Union Eagle Ltd v Golden Achievement Ltd [1997] AC 514, 519 (not cited by Adhar, ‘Contract doctrine’ (n 119)).

264  Jonathan Morgan of the Misrepresentation Act 1967 does significant harm to commercial predictability. The strictly limited appellate oversight deepens the problem. None of this is to criticise the courts (although Lewison LJ’s blithe claim that section 3 ‘should not cause consternation’ to commercial contractors is surely debateable).121 The problems are inherent in the legislation. Should it therefore be reviewed, amended or even repealed?

C.  Legislative Reform There is a strong case for removing at least some commercial contracts from the scope of the Unfair Contract Terms Act 1977 and section 3 of the Misrepresentation Act 1967 altogether (protection under the Consumer Rights Act 2015 is outside our present concern – and clearly necessary). For sophisticated parties, the jurisdiction to review NRCs or EACs for reasonableness may well do more harm than good. If this is accepted, a difficult question arises over SMEs. As Gerard McMeel puts it, defending the legislation’s wide application to commercial NRCs, contract law must cater for the car dealer as well as the bond dealer.122 The legislation could be amended to protect only businesses of moderate size (as defined by their turnover, or number of employees or in some other way). Naturally, any sharp distinction like that causes new legal problems (even as it would desirably exempt large businesses from inappropriate regulation). Some would defend the current approach as a superior legal strategy – viz, to retain all contracts (and contractors) within the scope of the legislation and leave it to the courts to tailor the appropriate level of protection on a case-by-case basis. Recall, however, that the Law Commission’s tentative 2002 proposal that consumer legislation should be extended to all business-to-business contracts, with the appropriate level of protection ultimately left to the courts (presumably a less demanding standard than for consumers),123 was dropped after negative views in consultation. The concerns expressed were not sufficiently met by trusting the courts to regulate in a fact-sensitive manner. The Law Commission in its 2005 Report did still recommend the extension of protection to SMEs.124 While this was apparently accepted by the Government at the time, the Report was not implemented, and when consumer contract reform returned to the legislative agenda the SME proposals were not taken forward.125 This is a story of continually diminished regulatory ambitions. However, the existing regulation (1967 and 1977 Acts) was 121 First Tower Trustees Ltd (n 98)[67]. 122 McMeel, ‘Documentary Fundamentalism’ (n 3). 123 Law Commission, Unfair Terms in Contracts (Law Com CP No 166, 2002), Part V (NB 5.36, ‘complexity’ of separate legal regime for SMEs). 124 Law Commission, Unfair Contract Terms in Contracts (Law Com No 292, 2005), 4.8–4.17, Part V. 125 Law Commission, Unfair Terms in Consumer Contracts: a new approach? (Issues Paper, 2012) 1.2, 1.26.

Entire Agreement and Non-Reliance Clauses  265 left in place, without serious reappraisal – still applicable to all commercial parties (‘car dealers’ and ‘bond dealers’ alike). This protection could instead be confined to SMEs or, alternatively, excluded from particular fields of economic activity. Many challenges to NRCs involve financial services provided to business investors. The clauses here have invariably been upheld as reasonable.126 It would prevent much costly litigation if such contracts were removed from the legislation altogether. Notably, the Unfair Contract Terms Act already has limited scope. It is inapplicable to ‘international supply contracts’ (as defined by section  26). This limitation does not, however, appear in the Misrepresentation Act 1967.127 It is not clear why. At minimum, it is suggested that an exclusion analogous to section 26 of the 1977 Act should be inserted into the 1967 Act. These proposals are offered tentatively. Obviously, there would be costs and benefits for different groups of winners and losers in any reform. (Equally, one could say this about the status quo – which is hardly a neutral starting point.) Rigorous empirical research is needed. Probably only a full public consultation of the kind routinely conducted by the Law Commission could elicit the relevant information, and stakeholder support, for a significant contraction of the scope of the legislation. Such an exercise is desirable. The hostile opposition to the Law Commission’s proposed extension of consumer protection to (all) businesses, in its last major review of the unfair terms legislation, has been noted. However, the Commission recommended retention of the existing business-to-business ­protection.128 Yet the points made against the Law Commission’s suggested extension also counted against that retention – especially since these encompass commercial ‘big fish’ as well as SMEs. A final note of caution. The present legislation satisfies the judicial concern to regulate unfair exclusion clauses (ie those perceived by the courts as unreasonable). The common law controls, prior to the 1977 Act, were notoriously unsatisfactory. Leggatt LJ recently said that Parliament’s policy of regulating waiver of claims for misrepresentation (‘a paradigm “vitiating factor” which undermines the validity of a contract’) was ‘readily understandable’.129 If the courts no longer had the legislative weapons to mount a direct attack on a NRC (etc) perceived to be unfair, they might revert to tools such as tortured construction to ‘stab the clauses in the back’.130 This would be still less satisfactory in terms of predictability. Of course it might not happen. If Parliament consciously scaled back the current policy of regulation, it would become ‘the duty of the courts to uphold and not to subvert that policy choice’ too.131 But when the courts are strongly committed to a given



126 eg

First Tower Trustees Ltd (n 98) [67]. by Bridge, ‘Innocent misrepresentation’ (n 19) 295. 128 Law Com No 292 (n 124), pt IV. 129 First Tower Trustees Ltd (n 98)[104] (Leggatt LJ). 130 cf George Mitchell (Chesterhall) Ltd (n 73) 297 (‘The secret weapon’) (Lord Denning MR). 131 First Tower Trustees Ltd (n 98)[104] (Leggatt LJ). 127 Noted

266  Jonathan Morgan policy, Windscheid’s aphorism that what is thrown out of the door will re-enter via the window frequently holds good.132 Deregulation will only be as successful as the courts allow it to be. All the more reason for any amendment to the legislation to follow the full consultation process of a Law Commission investigation.

VIII. Conclusion This chapter has been too long. Yet its main argument is simple. Courts face a choice of how ‘formally’ (textually) or how ‘relationally’ (contextually) to understand a given contract. There is no one-size-fits-all answer. This has been recognised in the much-litigated field of interpretation;133 it holds good generally, including in the present context. When the parties (or the drafters of the contract to which they have signed up) have expressly made this choice, the court’s task is simple. As Nye Bevan said, you do not have to gaze into a crystal ball when you can just read the book. The point necessarily cuts both ways. Where parties oblige themselves, for example, to renegotiate a contract in good faith when obstacles emerge, the courts must enforce this.134 The characteristic default formalism of English law must give way when the parties expressly signal that their contract should be understood relationally. Party autonomy, freedom of contract, requires no less. Equally though, when parties expressly choose a highly formalistic regime as epitomised by EACs and NRCs, the courts must enforce that too. It enlarges autonomy and reduces costs to give businesses ‘a cheap and certain means of effectively expressing their intent’ to be governed exclusively by the documentary terms.135 We may speculate that such clauses (and NOM clauses and others yet to be litigated, such as ‘textual interpretation’ or ‘no good faith’ clauses)136 will only become more common as the English flirtation with broad notions of purposive interpretation, economic duress, good faith, etc is increasingly consummated. Not all parties ‘feel the love’. Some may well prefer ‘documentary fundamentalism’. If they signal that preference, it ought to be respected. The legislation that ordains regulation of these choices should be confined in scope, or even repealed entirely. Until that happens, the courts must apply it with greater respect for the commercial realities outlined here.

132 cf R Zimmermann, The Law of Obligations: Roman Foundations of the Civilian Tradition (OUP 1996) 581. 133 Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173 [13] (Lord Hodge). 134 Consider Associated British Ports v Tata Steel UK Ltd [2017] EWHC 694 (Ch), [2017] 1 CLC 826. 135 G Klass, ‘Parol Evidence Rules and the Mechanics of Choice’ (2019) 20 Theoretical Inquiries in Law 457, 475. 136 See Calnan, ‘Controlling Contractual Interpretation’ (n 4); and PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6 in this volume.

13 Planning for Failure: Contract Design, Ineffective Bargains and Restitution NIAMH CONNOLLY

I. Introduction Lawyers drafting contracts to facilitate complex commercial transactions have enough to think about without worrying that the contract itself might be ineffective. Nonetheless, there is a real possibility that the parties’ dealings will not be governed from beginning to end by their carefully crafted contractual documents. A contract might cease to be effective due to termination for breach or frustration. Worse, the contract could be invalid from the start. What would this mean for parties who had performed their obligations and transferred value to each other? The good news is that the default legal rules governing restitution of unjust enrichment will offer a remedy in these situations. Across these scenarios, the doctrine of failure of consideration can apply to enable a party to recoup value it has conferred under the bargain. In the case of frustration, English statute law provides for readjustment and loss sharing. These remedies are generally adequate to prevent one party from being left entirely out of pocket. However, the solutions they offer may not be very fine-tuned, notably because of the requirement of total failure of consideration. The better news for contract drafters is that they can control the potential operation of the law of restitution to their dealings through their initial contract design. Even though restitution of unjust enrichment arises by operation of law, they have scope to influence it. They can tailor outcomes that respond better to their preferences and the specificities of their deal. They can achieve greater certainty and avoid potential pitfalls. In those cases where the contract is valid but later ceases to be effective, the parties can directly dictate mechanisms for restitution and protection of their reliance interest. Even where the contract is invalid, its contents are evidence of the parties’ intentions. It will often affect the valuation of benefits, and will inevitably shape how the doctrine of failure of consideration is applied to the bargain.

268  Niamh Connolly This chapter aims to assist contract drafters by showing how the default rules governing restitution can interact with choices they make in constructing the transaction. This will enable them purposefully to exert the capacity for control that they necessarily possess. The chapter will focus on three principal ways in which contract drafters can control the operation of restitution law to their deal. First, they can exclude restitution, where the contract justifies the act of performance. This also applies to multi-party situations. Second, they can provide expressly for the practical consequences of termination or force majeure. Third, they can structure their bargain in ways that ensure that the doctrine of failure of consideration will apply in a predictable, well-calibrated way that responds to the parties’ own understanding of what is essential in their bargain.

II.  The Case for Planning for Unforeseen Eventualities Nobody expects the Spanish Inquisition. It might seem paradoxical to suppose that people can plan for the unforeseen and unforeseeable. If people suspected their contract was going to be invalid, surely they could avoid the defect? In fact, it is entirely possible to provide for the three categories of situation with which we are concerned: force majeure, termination and invalidity. There is no need to predict the specific chain of events that will produce these conditions.1 Contract drafters should employ foresight to answer a different question: if the contract ceases to be operative, how do the parties wish to provide for restitution and possible loss sharing? This will take careful planning where contracts are performed over time, and where performing parties incur costs ahead of delivery. Transaction lawyers must drive this planning process. Their clients focus on the substance of a transaction. Sometimes significant transactions are planned with little input from lawyers, and this shows in contracts that do not provide for contingencies.2 Commercial actors are alert to market risks and the risk of non-performance, but less so to possible termination or invalidity. Professional drafters routinely provide for foreseeable risks related to the nature of a transaction or the trading environment.3 Those who regularly draft commercial contracts have the ability further to insulate their clients from risk by addressing legal risks like contractual invalidity. The main reason not to make detailed provision for remote contingencies is the cost of planning. There is a trade-off between the costs of trying to provide in

1 VP Goldberg, ‘Impossibility and Related Excuses’ (1988) 144 Journal of Institutional and Theoretical Economics 100, 114. 2 T Daintith, ‘The Design and Performance of Long-Term Contracts’ in T Daintith and G Teubner (eds), Contract and Organisation: Legal Analysis in the Light of Economic and Social Theory (De Gruyter 1986) 164, 178. 3 A Choi and G Triantis, ‘Strategic Vagueness in Contract Design: The Case of Corporate Acquisitions’ (2010) 119 Yale Law Journal 848, 871.

Planning for Failure  269 advance for every remote contingency and waiting until the unexpected happens to find solutions, which might then involve litigation.4 The costs of contingency planning are likely to be dissuasive in a one-off or low-value transaction.5 However, in a more valuable or longer-term exchange, or for repeat players, the incentives look different.6 Moreover, the cost of planning depends on the techniques used. Some planning for legal risk can be achieved through standard form contracts, though in many cases tailored plans are better.7 Perhaps counterintuitively, vagueness can be an efficient choice when providing for the unexpected.8 Vague contractual standards leave room for judicial discretion in interpreting them, but within agreed boundaries.9 A strategic mix of precise and vague provisions can provide efficiently for the risks of contractual ineffectiveness.10 Force majeure clauses typically illustrate this balance. Another argument against dealing with these risks in the contract is that business people often respond collaboratively to disruptive events. Long-term contracts often build in flexibility and allow for contract variation to resolve ­difficulties.11 Commercial actors with ongoing business relationships do not always stand on their legal rights. Even from this perspective, providing guidance in the contract for the practical resolution of unforeseen events is useful. It can help create a shared understanding to aid renegotiation or amicable variation.12 While planning for legal risks when designing contracts will benefit parties to significant transactions, it remains important that the law offers suitable remedies where the parties have not made active choices. Commercial entities are not equally empowered with legal expertise. Many of those engaged in commerce are unsophisticated legal actors, and not all commercial contracts are negotiated or reviewed by lawyers.13 4 H Beale and T Dugdale, ‘Contracts between Businessmen: Planning and the Use of Contractual Remedies’ (1975) 2 British Journal of Law & Society 45; K Crocker and KJ Reynolds, ‘The Efficiency of Incomplete Contracts: An Empirical Analysis of Air Force Engine Procurement’ (1993) 24 Rand Journal of Economics 126, 126; RE Scott and GG Triantis, ‘Incomplete Contracts and the Theory of Contract Design’ (2005) 56 Case Western Reserve Law Review 187, 197. 5 RA Hillman and JJ Rachlinski, ‘Standard-Form Contracting in the Electronic Age’ (2002) 77 New York University Law Review 429, 437. 6 MR Miller, ‘Contract Law, Party Sophistication and the New Formalism’ (2010) 75 MLR 493, 532. 7 MA Eisenberg, ‘The Limits of Cognition and the Limits of Contract’ (1995) 47 Stanford Law Review 211, 243. 8 Choi and Triantis, ‘Strategic Vagueness’ (n 3) 854. 9 RE Scott and GG Triantis, ‘Anticipating Litigation in Contract Design’ (2006) 115 Yale Law Journal 814, 856. 10 Choi and Triantis, ‘Strategic Vagueness’ (n 3) 881. 11 IR Macneil, ‘Contracts: Adjustment of Long-Term Economic Relations under Classical, Neoclassical, and Relational Contract Law’ (1977–1978) 72 Northwestern University Law Review 854, 865; A Al Faruque, ‘The Rationale and Instrumentalities for Stability in Long-Term State Contracts’ (2006) 7 Journal of World Investment & Trade 85, 96. 12 CP Gillette, ‘Commercial Rationality and the Duty to Adjust Long-Term Contracts’ (1985) 69 Minnesota Law Review 521, 556; Daintith, ‘Long-Term Contracts’ (n 2) 187. 13 Miller, ‘Contract Law, Party Sophistication and the New Formalism’ (n 6) 522; Beale and Dugdale, ‘Contracts between Businessman’ (n 4) 47; LT Garvin, ‘Small Business and the False Dichotomies of Contract Law’ (2005) 40 Wake Forest Law Review 295, 312.

270  Niamh Connolly The focus of this chapter is on the interaction between restitution law and contract drafting. A preliminary question concerns choice of law. On its face, unjust enrichment falls under the Rome II Regulation.14 However, the Rome I Regulation governs all the situations where restitution arises from a contractual relationship.15 Its scope includes excuses from performance such as force majeure,16 the consequences of breach17 and the consequences of nullity.18 The regulation cuts the Gordian knot by treating choice of law clauses in invalid contracts as if the contract were valid.19 This means that when parties choose the law of their contract, they implicitly choose the same law for restitutionary questions arising from the relationship.20 English law is a popular choice because of its certainty and commitment to upholding bargains.21 As a result, English courts have decided many swaps cases involving international parties who chose English law and jurisdiction.22 There is no compelling reason for parties to decouple the choice of law for contract and related restitution claims. Rather, coherence and simplicity favour keeping adjudications of restitutionary consequences together with the underlying contractual issues. Those constructing commercial deals can also employ other strategies that fall beyond the scope of this chapter to guard against risks such as invalidity.23 Warranties in a valid overarching agreement might create a contractual estoppel or entail damages for breach of warranty if a party lacks capacity for a specific ­transaction.24 In addition, insurance or guarantees may provide necessary protection from contractual risks.

III.  Valid Contracts as Justifying Grounds The first, highly significant way in which contracting parties control the application of restitution law to their interactions is by making a contract. When a person 14 Art 10, Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II); SGA Pitel, ‘Rome II and Choice of Law for Unjust Enrichment’ in J Ahern and W Binchy (eds), The Rome II Regulation on the Law Applicable to Non-Contractual Obligations (Martinus Nijhoff Publishers 2009) 231. 15 L Collins (ed), Dicey, Morris & Collins: The Conflict of Laws, 15th edn (Sweet & Maxwell 2015) [36-023]; Pitel, ‘Rome II and Choice Of Law’ (n 14) 240. 16 Art 12(1)(b), 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). 17 Art 12(1)(c) Rome I. 18 Art 12(1)(e) Rome I; G Schulze, ‘Article  12 Rome I’ in G-P Calliess (ed), Rome Regulations. Commentary on the European Rules of the Conflict of Laws, 2nd edn (Kluwer 2015) 307, 317–18. 19 Art 10(1) Rome I. 20 Haugesund Kommune v Depfa ACS Bank [2010] EWCA Civ 579, [2012] Bus LR 230 [60]. 21 PS Davies, ‘Bad Bargains’ (2019) Current Legal Problems 1, 1–2. 22 J Braithwaite, ‘Thirty Years of Ultra Vires: Local Authorities, National Courts and the Global Derivatives Markets’ (2018) 71 Current Legal Problems 369, 390, 393. 23 ibid, 400. 24 Credit Suisse International v Stichting Vestia Groep [2014] EWHC 3103 (Comm), [2015] Bus LR D5.

Planning for Failure  271 performs an obligation required by contract, that performance is not ‘unjust’.25 The value transferred cannot be recovered in restitution unless and until the contract is avoided.26 Furthermore, there may be a defence to unjust enrichment where a ‘payment is made for good consideration’, in particular to discharge a debt.27 This putative defence seems to overlap with the orthodox proposition that a contract is a justifying ground that excludes unjust enrichment. On the other hand, restitutionary remedies may be available in a contractual setting where a benefit is conferred outside the scope of the bargain, the contract is terminated for breach or frustrated, or the contract is invalid. Moreover, the boundary between contract and restitution is less clearly delineated than it might at first appear.

A.  Primacy of the Contractual Bargain The law is shaped by a strong policy against subverting contractual bargains. In The Trident Beauty, Lord Goff stated that where a contract between the parties provides for a certain eventuality, it is both unnecessary and inappropriate for the law of restitution to intervene.28 Where a contract stipulates the price due, the person who confers a benefit cannot seek an alternative measure of payment outside the agreement.29 Judicial dicta state categorically that restitution is not available when contract damages might be awarded.30 Although the case law on termination for breach suggests a more nuanced proposition, these dicta reflect the law’s commitment to holding parties to their contractual bargains and preventing them from suing in restitution where this would be more advantageous.31 This policy prioritises respect for contracting parties’ choices. There are other reasons for judges to hesitate to intervene in contractual bargains. Commercial actors can make a sensible allocation of risk better than courts,32 and the bargain as a whole may represent a subtle, multi-stranded allocation of risks.33 Many commentators consider that the onus is on contracting parties to provide in advance for

25 A Burrows, ‘Is There a Defence of Good Consideration?’ in C Mitchell and W Swadling (eds), The Restatement Third: Restitution and Unjust Enrichment Critical and Comparative Essays (Hart Publishing 2013) 165, 174. 26 Dimskal Shipping Co SA v International Transport Workers Federation (The Evia Luck) (No 2) [1992] 2 AC 152, 165; Kleinwort Benson Ltd v Lincoln CC [1999] 2 AC 349, 407–08. 27 Barclays Bank Ltd v WJ Simms, Son and Cooke (Southern) Ltd [1980] QB 677, 695; Burrows, ‘Defence of Good Consideration’ (n 25) 167. 28 Pan Ocean Shipping Co Ltd v Creditcorp Ltd (The Trident Beauty) [1994] 1 WLR 161, 164. 29 Re Richmond Gate Property Co Ltd [1965] 1 WLR 335; Pavey & Matthews Pty Ltd v Paul (1987) 162 CLR 221, 265. 30 Taylor v Motability Finance Ltd [2004] EWHC 2619 (Comm). 31 P Birks, An Introduction to the Law of Restitution (Clarendon, 1985) 47. 32 Hillman and Rachlinski, ‘Standard Form Contracting’ (n 5) 432. 33 LE Trakman, ‘Frustrated Contracts and Legal Fictions’ (1983) 46 MLR 39, 47.

272  Niamh Connolly arrangements such as loss sharing, rather than to expect a judge to interfere in their relationship.34

B.  The Scope of the Bargain The first important point to note is that the fact that parties are contracting partners is not enough to prevent any restitution claims between them.35 If there is no contractual obligation to make a particular payment, it might be ­recovered.36 The same applies where a party does additional work that falls outside the scope of its contractual duty.37 The key question is whether the scope of the bargain covers the enrichment conferred. This depends on the terms of the agreement. A surveyor cannot claim extra payment for supervising a bigger job than planned, if the contract required him to supervise all the work that would be done.38 Similarly, negotiating an insurance claim may fall within a finance director’s contractual duties.39 By defining a party’s obligation narrowly or broadly, drafters can make it more or less likely that some benefit that party confers might fall outside its contractual duties and may ground a claim for additional remuneration. Contracts for services, in particular, should define clearly what work is required under the contract.

C.  The Boundary Between Contract and Restitution Second, the boundary between contract and restitution is not absolute.40 Beatson has influentially argued that the reason we usually do not allow restitution where a contract is performed is to avoid ‘inconsistency and circularity’.41 This recognises space for restitution alongside a contract if it is not inconsistent with the terms of the contract and does not contradict the agreed allocation of risks.42 The termination of a contract due to breach or its frustration clears the way for restitution of value conferred before termination, for which the contractual

34 VP Goldberg, ‘After Frustration: Three Cheers for Chandler v Webster’ 68 (2011) Washington and Lee Law Review 1133, 1136. 35 S Waddams, Sanctity of Contracts in a Secular Age: Equity, Fairness and Enrichment (CUP 2019) 143. 36 P Birks, Unjust Enrichment (Clarendon 2005) 121–22. 37 MP Gergen, ‘Restitution as a Bridge over Troubled Contractual Waters’ (2002) 71 Fordham Law Review 709, 713. 38 Gilbert & Partners v Knight [1968] 2 All ER 248, 250. 39 Taylor v Motability Finance Ltd (n 30) [21]. 40 Roxborough v Rothmans of Pall Mall Australia Limited [2001] HCA 68; P Birks, ‘Failure of Consideration and its Place on the Map’ (2002) 2 Oxford University Commonwealth Law Journal 1, 5; Waddams, Sanctity of Contracts (n 35) 145ff. 41 J Beatson, ‘Restitution and Contract: Non-cumul?’ (2000) 1 Theoretical Inquiries in Law 1, 4. 42 ibid; Birks, ‘Failure of Consideration’ (n 40) 5.

Planning for Failure  273 right had not accrued. This remedy on termination is long established in the case law. It seems to be an exception to the proposition that unjust enrichment is not appropriate where a person has performed a valid contractual obligation. Indeed, restitution may here enable a person to avoid a bad bargain. However, restitution is a straightforward, intuitively appropriate response to fundamental breach, allowing the innocent party to walk away from the deal.43 Beatson’s analysis offers an explanation for why this is permissible. The incompatibility criterion also has wider implications. It means that the relationship between contract and unjust enrichment depends on the terms of each specific bargain rather than general propositions. We must carefully interpret each contract to assess whether the agreement is incompatible with restitution in the circumstance that has eventuated.44 Often a contract is silent about a contingency. There may be a general obligation to perform, with no express exemption for some unforeseen eventuality. There are divergent views about what silence may mean about the allocation of risks. One view is that a contract implicitly distributes all risks to one side or the other. Then a loss lies where it falls.45 Parties can make express provision for the unexpected, so their choice not to do so merits respect.46 Certainly, silence sometimes represents a rational choice by the parties to assume a known risk, or even unknown risks.47 The alternative approach recognises that there may be contractual gaps, especially in long-term contracts.48 Known risks will be tacitly or expressly allocated, but the bargain may not represent any sort of agreement about truly unforeseen contingencies.49 There ‘just is no agreement’ about something that turns out to be important.50 If contracts do not allocate all risks consensually between the parties, common law restitution rules might legitimately address genuine gaps. Where we cannot solve the problem by respecting the parties’ agreement, Fried argues that we have no choice but to resort to external principles.51 Where the parties have made no choice at all, we may legitimately apply default rules that give effect to justifiable values and principles, just as the default rules of contract law do.52 Further, the silence of the parties might be understood as consent to such default rules.53 43 C Fried, Contract as Promise (Harvard University Press, 1981) 114. 44 S Waddams, ‘Contract and Unjust Enrichment: Competing Categories, or Complementary Concepts?’ in C Rickett and R Grantham (eds), Structure and Justification in Private Law: Essays for Peter Birks (Hart Publishing 2008) 175. 45 Gillette, ‘Commercial Rationality’ (n 12) 534. 46 Davies, ‘Bad Bargains’ (n 21) 6. 47 Trakman, ‘Frustrated Contracts’ (n 33) 47; Gillette, ‘Commercial Rationality’ (n 12) 536. 48 Beatson, ‘Non-cumul?’ (n 41) 12, 14. 49 Barton v Gwyn-Jones [2019] EWCA Civ 1999 [62]; Waddams, Sanctity of Contracts (n 35). 50 Fried, Contract as Promise (n 43) 59. 51 ibid 60. 52 RE Barnett, ‘The Sound of Silence: Default Rules and Contractual Consent’ (1992) 78 Virginia Law Review 821, 822. 53 ibid 861–62.

274  Niamh Connolly Whenever we are required to determine if a specific contract implicitly allocates the risk in question to a party, we should be careful.54 It can be tempting to interpret contractual silence as reflecting an agreement that no adjustment will occur in unexpected circumstances. However, saying that the contract assigns a risk is the conclusion of our reasoning process, not the beginning.55 We can only decide that silence imposes a risk on one party or the other after we know what non-contractual remedies the law offers; and that is the question we are trying to answer.56 Further, the meaning of silence depends on the parties’ assumptions, including about background legal rules.57 One business community may assume that losses due to unforeseeable events lie where they fall, and another that default legal rules will provide a fairer allocation of losses. These assumptions may change over time. Some older authorities may have refused unnecessarily to consider restitution of overpayments because there was a contract in place.58 If an element of the contract is externally imposed rather than negotiated by the parties, it is more likely not to form part of their agreed risk allocation.59 English courts may now be becoming more willing to hold that restitution will not undermine a valid contract. In Barton v Gwyn-Jones, the contract stipulated a fee if a property sold at a certain price but was silent about payment if the selling price was lower. The Court of Appeal construed the contract carefully and allowed a claim in unjust enrichment because it did not subvert a contractual allocation of risk.60 Roxborough v Rothmans of Pall Mall Australia is a puzzling authority that shows the possible limits of the proposition that there cannot be restitution of money paid under a binding contract.61 A buyer successfully recovered money paid under a valid contract to cover a licensing fee that turned out not to be due. Virgo condemns the decision as unjustifiable, because there was a continuing basis for the payment.62 He considers that an exception to the rule would only be warranted

54 Beatson ‘Non-cumul?’ (n 41) 4; Waddams, ‘Competing Categories’ (n 44) 173; Waddams, Sanctity of Contracts (n 35) 148. 55 J Edelman, ‘Liability in Unjust Enrichment where a Contract Fails to Materialize’ in A Burrows and E Peel (eds), Contract Formation and Parties (OUP 2010) 159, 160. 56 Fried, Contract as Promise (n 43) 64; F Wilmot-Smith, ‘Replacing Risk-taking Reasoning’ (2011) 127 LQR 610, 613. 57 Barnett, ‘The Sound of Silence’ (n 52) 862. 58 Green v Portsmouth Stadium [1953] 2 QB 190 (CA); Orphanos v Queen Mary College [1985] AC 761 (HL); Birks, ‘Failure of Consideration’ (n 40) 5. 59 Birks, Unjust Enrichment (n 36) 124. 60 Barton v Gwyn-Jones (n 49). 61 Roxborough v Rothmans of Pall Mall (n 40). 62 G Virgo, ‘Demolishing the Pyramid – the Presence of Basis and Risk-Taking in the Law of Unjust Enrichment’ in A Robertson and Tang HW (eds), The Goals of Private Law (Hart Publishing 2009) 477, 491; J Beatson and GJ Virgo, ‘Contract, Unjust Enrichment and Unconscionability’ (2002) 118 LQR 352, 355, 356.

Planning for Failure  275 if the bargain demonstrably placed the risk of invalidity on the seller.63 In contrast, Burrows approves the outcome. He proposes that ‘the rule that one cannot have restitution for mistake where the mistaken payment was legally owed is not an absolute one, but permits limited exceptions’.64 It is a heuristic; we can evaluate in each case whether the payment is unjust notwithstanding that it was owed. The failure of the condition in Roxborough outweighed the continued existence of the obligation. Besides, restitution did not interfere with the contractual allocation of risks because the levy was not part of the parties’ negotiated bargain.65 In effect, the eventuality of the tax not actually being due was a contractual gap. That risk was not consensually allocated to either party.66 Implied terms might sometimes do the work of restitution when the unexpected happens in a contractual relationship. In The Trident Beauty, Lord Goff was willing to imply a contractual term for the repayment of advance hire paid for a period for which it was not due.67 In Barton, the Court of Appeal accepted an implied term as an alternative avenue to remuneration for work done.68 However, the standard for implying a term is hard to fulfil in cases where the parties did not envisage the eventuality that has occurred.69 It may often be more difficult to establish an implied term than to show that the contractual obligation is not an obstacle to restitution.

D.  Multi-party Transactions The law’s policy against allowing claims in unjust enrichment where there is a valid contract has distinct ramifications in multi-party transactions. In MacDonald Dickens & Macklin v Costello, a builder contracted with an intermediary company to erect houses on land owned by another individual.70 When the company did not pay, the builder sued the landowners in unjust enrichment. Even though the

63 Virgo, ‘Demolishing the Pyramid’ (n 62) 491; Beatson and Virgo, ‘Contract, Unjust Enrichment and Unconscionability’ (n 62) 356. 64 Burrows, ‘Defence of Good Consideration’ (n 25) 177. 65 A Burrows, ‘Restitution of Mistaken Enrichments’ (2012) 92 Boston University Law Review 767, 767–77; Burrows, ‘Defence of Good Consideration’ (n 25) 177. 66 See also M McInnes, ‘Unjust Factors, Juristic Reasons and Contracts in Anglo-Canadian Law’ in P Giliker (ed), Re-examining Contract and Unjust Enrichment: Anglo-Canadian Perspectives (Martinus Nijhoff 2007) 23, 33. 67 Waddams, Sanctity of Contracts (n 35) 153. 68 Barton v Gwyn-Jones (n 49). 69 Pluczenik Diamond Co NV v W Nagel (A Firm) [2018] EWCA Civ 2640 [34]; Waddams, ‘Competing Categories’ (n 44) 179. 70 MacDonald Dickens & Macklin v Costello [2011] EWCA Civ 930; HLE Verhagen, ‘The Policies against Leapfrogging in Unjust Enrichment: A Critical Assessment’ (2018) 22 Edinburgh Law Review 55.

276  Niamh Connolly landowners were the owners of the company, and had interposed the company as the contracting party, the law treated them as fully separate. The Court of Appeal held that by knowingly contracting with the company, the builder had taken the risk of that company’s not paying or becoming insolvent. The structure the parties chose for the transaction imposed those risks on the builder.71 It would subvert the contractual allocation of risk to bypass the contractual counterparty and impose a restitutionary obligation on the recipient.72 This judgment endorses a ‘general policy of refusing restitutionary relief for unjust enrichment against a defendant who has benefited from the plaintiff ’s services rendered pursuant to a contract to which the defendant was not a party’.73 This dispute was essentially one about contract performance, for which contract law’s rules and remedies were best suited. This does not mean that there might not also be a concurrent remedy in unjust enrichment. Undoubtedly, applying unjust enrichment law to a multi-party scenario like this is complex. One concern is that the transfer from the builder to the landowner might not be sufficiently direct because of the interposition of the contractual relationship. Factually, though, the builder’s services directly increased the value of the defendant’s land, which should count as a sufficiently direct transfer.74 In addition, failure of consideration requires a shared understanding between the builder and landowner about the builder’s expectation of payment. Even though the builder performed on condition that the intermediary paid, this should suffice, as the defendant was aware of this condition and that it was not fulfilled. Some multiparty scenarios present additional reasons not to allow a claim in restitution that were not present here.75 A defendant who paid (or was liable to pay) the intermediate contracting party might not be enriched. If the defendant were contractually entitled to receive the benefit from the intermediary, that would likely justify his enrichment. Each of these issues might be addressed individually, if the claimant’s choice of contracting partner were not viewed as imposing on him the risk of another party’s receiving the benefit and nobody paying for it. Perhaps the law’s policy concern not to subvert the contract would be diluted if a court considered there was collusion between the intermediary and the person who received the benefit. Costello shows that in a multi-party scenario, the parties’ choices of contracting partners can be vital. It will not be possible to obtain remuneration from a third party on whom a benefit is conferred in performance of a contract with somebody else. Parties should take care to avoid this trap, and seek guarantees from others involved in multi-party transactions.



71 MacDonald

Dickens & Macklin v Costello (n 70) [21]. [23]. 73 ibid [30]; Lumbers v W Cook Builders Pty Ltd (in liq) [2008] HCA 27. 74 Verhagen, ‘Leapfrogging in Unjust Enrichment’ (n 70) 63. 75 ibid 62. 72 ibid

Planning for Failure  277

IV.  Termination and Frustration: Scope for Provision in Valid Contracts A.  Terminated Contracts Some serious breaches of contract entitle the innocent party to terminate the contract.76 Termination operates prospectively, releasing both parties from their obligations to perform.77 The terms of a valid contract remain binding and continue to govern those rights that have accrued under the contract.78 The innocent party is entitled to damages for breach. Where a right has accrued before termination, the promisee is entitled to its performance as a matter of contract law. However, if a party has paid or performed in advance and not received a benefit in return, she can claim restitution.79 Even a contract-breaker can recover her advance payment, though she will be required to pay contract damages.80 The unjust factor is total failure of consideration: her performance was conditional on counter-performance that will not take place. Restitution of the value of a benefit conferred depends on the other party’s contractual right to that benefit not yet having accrued.81 The questions of whether a contractual right has unconditionally accrued and whether there has been failure of consideration are interconnected. The law avoids the circularity of ordering a contractual payment that could immediately be reclaimed for failure of ­consideration.82 Consequently, an innocent party cannot claim payment of instalments that fell due before termination if it has not performed the corresponding work. It will instead have its contractual remedies, including damages for loss of bargain. Because termination affects valid contracts, the parties prospectively enjoy the power directly to determine what repayments and adjustments will be made. Constructing the transaction, they control the central question of when rights accrue. They can also devise their own plan for the consequences of termination.

B.  Termination Clauses Termination clauses are common in modern contracts.83 They enable the parties to specify different, perhaps more permissive, triggering conditions than the

76 Hong Kong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd [1962] 2 QB 26, 70. 77 McDonald v Dennys Lascelles Ltd (1933) 48 CLR 457, 476–77; Johnson v Agnew [1980] AC 367, 396; Lombard North Central Plc v Butterworth [1987] QB 527, 535. 78 Beatson, ‘Non-cumul?’ (n 41) 16. 79 McDonald v Dennys Lascelles (n 77). 80 Dies v British and International Mining and Finance Corporation Ltd [1939] 1 KB 724. 81 Mann v Paterson Constructions Pty Ltd [2019] HCA 32 [176]. 82 Stocznia Gdanska SA v Latvian Shipping Co [1998] 1 WLR 574. 83 Choi and Triantis, ‘Strategic Vagueness’ (n 3) 858.

278  Niamh Connolly default law. They may authorise termination as a response to breach or to protect against future commercial impracticability. They permit parties to allocate the burden of proof as they choose.84 Termination clauses also enable parties to stipulate remedial consequences. Express contractual provision will displace the default rules on the consequences of discharge.85 The contractual provision then justifies the retention of a benefit. Contract drafters can provide for any adjustment that suits the needs of the parties and the nature of the transaction. This may mean a mix of restitution, protection of reliance expenditure and compensation. The clause may set out the basis on which repayments will be calculated.86 If the parties provide for the retention of instalments already paid, this may be interpreted as requiring the payments of unpaid sums that had fallen due.87 Where the cancellation is due to breach, they may choose to stipulate a compensatory element.88 A clause may enable a buyer to take over partially-completed works.89 A termination clause authorising the retention of payments does not prevent that party from availing itself of the common law rules to claim other payments that had accrued before termination.90 It would require clear words in the contract to rebut the general presumption that a party does not wish to abandon the remedies offered by contract law. However, if the contract provides a ‘complete code’ of remedies, it may implicitly exclude recourse to common law restitution rules.

C.  Structuring Performance Obligations i.  Entire Obligations Those drafting a contract may choose to make one party’s obligation an entire obligation, and this matters if the contract is interrupted. The contractual right to payment for an entire obligation does not accrue until that person’s performance is complete.91 Nor would the person who partially performs an entire obligation generally have a claim in restitution for the work done. There is an exception where the other party prevents the claimant from completing the work; then there can be termination and restitution.92 McFarlane and Stevens explain entire obligations cases on the basis that the parties’ shared understanding is that the performing party is only to be paid for his services when performance was 84 Scott and Triantis, ‘Anticipating Litigation in Contract Design’ (n 9) 874. 85 Beatson, ‘Non-cumul?’ (n 41) 18. 86 M Fontaine and F de Ly, Drafting International Contracts: an Analysis of Contract Clauses (Martinus Nijhoff 2009) 335. 87 Hyundai Heavy Industries Co Ltd v Papadopolous [1980] 1 WLR 1129. 88 Beale and Dugdale, ‘Contracts between Businessman’ (n 4) 52. 89 Meritz Fire & Marine Insurance Co Ltd v Jan de Nul NV [2011] EWCA Civ 827 [11]. 90 Stocznia Gdanska SA v Latvian Shipping Co (n 82). 91 Cutter v Powell [1795] 101 ER 573; Sumpter v Hedges [1898] 1 QB 673, 674; Cleveland Bridge UK Ltd v Multiplex Constructions (UK) Ltd [2010] EWCA Civ 139 [137]–[138]. 92 Giles v Edwards (1797) 101 ER 920.

Planning for Failure  279 complete.93 If the work is partially done, the condition of payment is not fulfilled.94 However, the entire obligation clause is not understood as an agreement that there will be no remuneration if the customer repudiates the contract before work is complete.95 The terms of each specific bargain will indicate whether the parties share the understanding that partial performance will not be paid for.96 A lump sum payment is not enough to draw this inference.97 Those drafting contracts should choose and clearly express whether they intend that partial performance must be remunerated.

ii.  The Timing of Obligations Contracting parties have freedom to structure the timing of performance as they choose.98 How drafters structure payments and phases of performance is very important in determining which rights accrue when. There are many commercial reasons to phase performance by one or both parties over a period of time.99 It may finance stages of the work, reduce risk100 or strike a compromise between ensuring a supplier receives payment while not excessively inconveniencing its customer. A contract may require instalments to be paid into an escrow account and used only for construction expenses.101 If a contract is terminated or frustrated, the phasing will determine when rights to certain performances accrue. Accordingly, it is a very important way to ensure that parties’ expenditure in performance of the contract will be compensated. Manufacturers of unique or customised goods have most acute need of this protection.102 It makes sense to match reciprocal units of performance on each side within a larger transaction that is performed over time. In shipbuilding cases, the courts have deemed that each instalment corresponded to work done by the shipbuilder.103 By pairing instalments with stages of performance, drafters can ensure that the contractual entitlement to payments accrues when each distinct phase of work is done.

iii.  Advance Payments and Deposits Similarly, commercial buyers often make advance payments to assist their suppliers’ cash flow or guarantee their commitment.104 Normally even contract-breakers 93 B McFarlane and R Stevens, ‘In Defence of Sumpter v Hedges’ (2002) 118 LQR 569. 94 J Goodwin, ‘Failure of Basis in the Contractual Context’ (2013) 21 Restitution Law Review 24, 27. 95 Mann v Paterson Constructions Pty Ltd (n 81). 96 Waddams, ‘Competing Categories’ (n 45) 176. 97 Beatson, ‘Non-cumul?’ (n 41) 23. 98 Goldberg, ‘After Frustration’ (n 34) 1135. 99 ibid 1146. 100 Beale and Dugdale, ‘Contracts between Businessman’ (n 4) 52. 101 Meritz Fire & Marine Insurance Co Ltd v Jan de Nul NV [2010] EWHC 3362 (Comm) [20]. 102 Goldberg, ‘Impossibility and Related Excuses’ (n 1) 113–14. 103 Stocznia Gdanska SA v Latvian Shipping Co (n 82). 104 S Mateut, ‘Reverse trade credit or default risk? Explaining the use of prepayments by firms’ (2014) 29 Journal of Corporate Finance 303, 304.

280  Niamh Connolly can recover advance payments, subject to their liability to pay damages.105 To avoid this, parties can require deposits as a guarantee of performance, or make some payments unconditional by inserting a forfeiture clause.106 Deposits cannot be recovered on the ground of failure of consideration when the payor does not complete his performance. They strengthen the position of the innocent party in a number of ways. It has the money in its pocket and the buyer will bear the burden of proof if there is a dispute.107 In addition, a permissible deposit may provide greater compensation than contract damages. The terms of the contract indicate whether a payment is a deposit or a prepayment.108 A seller who wishes a payment to be a deposit must make that clear, for example by labelling it ‘non-refundable’.109 The law regulates deposits and forfeiture clauses. The courts will not recognise an advance payment as a deposit unless the sum is ‘reasonable as earnest money’.110 Where the purported ‘deposit’ is excessive and therefore a penalty, it may be recovered in full, and the purchaser will compensate the vendor for actual loss due to the breach. However, respect for the contract means that genuine deposits are upheld, even where they work harshly, as where 10 per cent of the price is forfeit for a 10-minute delay.111 Like part payments, instalments can be subject to forfeiture clauses. The court can grant equitable relief if retention would be unconscionable, as where it is ‘out of all proportion to the damage’.112 However, courts will be slow to exercise this equitable power in a commercial case involving a freely-negotiated contract.113

V.  Frustration or Force Majeure Historically, the common law allowed the loss to lie where it fell when a partperformed contract was frustrated.114 In Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd, the House of Lords established that advance payments could be returned on the ground of failure of consideration.115 However, the case highlighted a problem that restitution of unjust enrichment did not address. The manufacturer of machinery had incurred costs in connection with performing the 105 Dies v British and International Mining and Finance Corporation Ltd (n 80); Stockloser v Johnson [1954] 1 QB 476, 489. 106 Howe v Smith (1884) 27 Ch D 89; Dies v British and International Mining and Finance Corporation Ltd (n 80). 107 Scott and Triantis, ‘Anticipating Litigation’ (n 11) 867. 108 Howe v Smith (n 106). 109 Dies v British and International Mining and Finance Corporation Ltd (n 80). 110 Workers Trust & Merchant Bank Ltd v Dojap Investments Ltd [1993] AC 573. 111 Union Eagle Ltd v Golden Achievement Ltd [1997] AC 514. 112 Stockloser v Johnson (n 105) 485; Jobson v Johnson [1989] 1 WLR 1026, [1989] 1 AER 621. 113 Cadogan Petroleum Holdings Ltd v Global Process Systems LLC [2013] EWHC 214 (Comm) [38]. 114 Chandler v Webster [1904] 1 KB 493. 115 Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd [1943] AC 32.

Planning for Failure  281 contract. When it returned the advance payment, it was left at a loss. Where parties incur expenditure prior to frustration, loss sharing may be needed.116 In response to this concern, the common law rule was superseded in England by the Law Reform (Frustrated Contracts) Act 1943. Judges now have a discretionary power to make allowance for a party’s reliance losses when awarding restitution, if they consider it just.117 The judicial power under the Act to limit restitution is highly discretionary. When a concert was cancelled and the organisers sought restitution of the advance fee from the band, the court took account of both parties’ wasted expenditure. There is vanishingly little case law applying the Act.118 Commercial parties may prefer to use force majeure clauses in order to control the outcome themselves and avoid such broad judicial discretion.119

A.  Force Majeure Clauses Like termination, frustration affects valid, effective contracts, and this enables drafters expressly to provide for its consequences. Force majeure clauses usually define what constitutes force majeure, provide for notice and stipulate consequences.120 Parties tend to relax the requirements for force majeure compared to national laws, especially the narrow default rule in English law, by allowing a response even when events do not make performance impossible.121 That response might involve suspending or cancelling performance obligations, or renegotiation or arbitration.122 A force majeure clause offers scope to control the financial effects when a contract is permanently frustrated. Where a contract is clearly intended to govern what happens in the event of frustration, it displaces the statutory adjustment regime.123 Parties can avoid the judicial discretion concerning loss sharing authorised by the English Frustrated Contracts Act. In other jurisdictions, where there is no statute and unjust enrichment law still applies on frustration, a clause allows parties to fashion more flexible solutions reflecting reliance expenditure. The range of options available is similar to termination clauses, and some contracts cross-refer to the termination clause. However, it is increasingly common to provide specifically for the practical consequences of termination resulting from 116 Fried, Contract as Promise (n 43) 70. 117 Law Reform (Frustrated Contracts) Act 1943, s 1(2) and (3). 118 BP Exploration Co (Libya) Ltd v Hunt (No 2) [1979] 1 WLR 783, [1983] 2 AC 352; Gamerco v ICM/ Fair Warning (Agency) Ltd [1995] 1 WLR 1226. 119 Goldberg, ‘After Frustration’ (n 34) 1147. 120 M Polkinghorne and C Rosenberg, ‘Expecting the Unexpected: The Force Majeure Clause’ (2015) 16 Business Law International 49. 121 H Konarski, ‘Force Majeure and Hardship Clauses in International Contractual Practice’ (2003) 4 International Business Law Journal 405, 408; Fontaine and de Ly, Drafting International Contracts (n 86) 405; Polkinghorne and Rosenberg, ‘Expecting the Unexpected’ (n 120) 52. 122 Fontaine and de Ly, Drafting International Contracts (n 86) 432. 123 Law Reform (Frustrated Contracts) Act 1943, s 2(3).

282  Niamh Connolly force majeure.124 One difference is that termination clauses may include some form of compensation for a wrongful breach, whereas force majeure clauses address situations where nobody is in the wrong. The focus is therefore on restitution, reliance and possible compensation for costs directly connected with the frustrating event. A simple clause may provide for a refund of advance payments.125 If both sides have performed some of their obligations, the clause may authorise the retention of the benefits received and payment for them. Alternatively, the parties might permit the retention of some payments even if the payor has not actually received a benefit.126 Some parties might prefer to exclude any settlement of accounts and require one party to obtain insurance.127 Contract drafters should tailor their plan to the specificities of the transaction rather than using boilerplate. The adjustment mechanisms should reflect the nature of the obligations concerned.128 Contracts that are performed over time will require relatively complex provision. It is not just about reliance. The different reasons why the parties sought phased performance will affect what they want to happen if the contract is frustrated. Contract drafters should think ahead to how performance on both sides is scheduled to occur, and envisage whether significant losses would result if the contract were to be frustrated at various junctures in performance.129 Sometimes, even if the parties agree a unit price as part of a large transaction, it would not be commercially viable for a divisible part of the overall performance. Ideally, the parties should negotiate the plan for possible adjustments. They understand their deal and are better able than a judge to determine what will be a fair outcome.130 Their interests may shift over the course of the contract’s scheduled performance.131 Some parties may be especially concerned about fixed costs at the outset. Involving both parties in planning can improve the prospects of an amicable resolution later. An alternative strategy is to recognise an unquantified obligation for one party to pay the other a sum that is reasonable in the circumstances of the case.132 The vagueness of the provision is economically efficient, as it saves the expense of detailed planning in advance for a remote contingency. The parties can make more informed decisions when they know what the effects of the frustrating event have 124 Fontaine and de Ly, Drafting International Contracts (n 86) 435. 125 P Thompson, ‘A Practical Guide to Force Majeure and Frustration of Contract’ (1997) 18 Credit Control 9, 12. 126 Dies v British and International Mining and Finance Corporation Ltd (n 80). 127 BP v Hunt (n 118) 806–07. 128 Polkinghorne and Rosenberg, ‘Expecting the Unexpected’ (n 120) 50; Fontaine and de Ly, Drafting International Contracts (n 86) 450. 129 Goldberg, ‘Impossibility and Related Excuses’ (n 1) 114. 130 H Hunter, ‘Commentary on Pitfalls of Force Majeure Clauses’ (1991) 3 Journal of Contract Law 214, 217. 131 Goldberg, ‘After Frustration’ (n 34) 1158. 132 R Christou, Drafting Commercial Agreements, 6th edn (Sweet & Maxwell 2016) Precedent 2.1, cl 10.1.

Planning for Failure  283 been. Although it does not grant certainty and there will be room for disagreement in implementing it, many commercial actors in long-term relationships prefer to resolve difficulties collaboratively by modifying their obligations.133

VI.  Contractual Bargains and Unjust Enrichment Claims The preceding sections considered how contract drafters can provide expressly for the consequences of termination and force majeure. More broadly, questions of restitution arising because a contract is ineffective will be governed by the common law of unjust enrichment. In order to establish a claim in unjust enrichment, a claimant must show that the defendant has been enriched at his expense, and that the transfer is unjust.134 Defences are then considered. The unjust factor that is most relevant to claims arising from contract is failure of consideration. This is usually the ground of recovery in both termination and invalidity cases. Unlike cases involving the termination of valid contracts, invalid contracts are not binding and cannot directly stipulate the consequences of unforeseen events. Yet they do affect how the common law rules apply to each case. The crucial point is that the default rules on restitution necessarily reflect the parties’ own bargain even when that bargain is invalid. First, the parties’ understanding constitutes evidence of the benefit’s value. Moreover, the contractual bargain is central to the doctrine of failure of consideration in each case. This means that the drafting of the contract usually influences common law restitution, whether the parties realise it or not.

A.  Valuing the Enrichment i.  Contract Price and Value The valuation of the enrichment transferred is central to an unjust enrichment claim. The contract price provides evidence of value in the circumstances of the case. It cannot be directly determinative because unjust enrichment is distinct from enforcing the contract.135 Benedetti v Sawiris establishes the objective market value of the benefit as the starting point for valuing enrichment.136 An arm’slength bargain between the parties is presumptive evidence of market value.137 133 Macneil, ‘Adjustment of Long-Term Economic Relations’ (n 11) 865; Al Faruque, ‘Stability in Long-Term State Contracts’ (n 11) 96. 134 Banque Financière de la Cité v Parc (Battersea) Ltd [1999] 1 AC 221, 227. 135 Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd (n 115) 64. 136 Benedetti v Sawiris [2013] UKSC 50. 137 ibid [168]; Way v Latilla [1937] 3 AER 759; Taylor v Motability Finance Limited (n 30) [26].

284  Niamh Connolly However, there are a few qualifications to using the parties’ bargain as evidence of value. The contract price may not be a useful benchmark, if it reflects altruistic ­considerations.138 If an invalid contract is vitiated by duress or mistake, it is unlikely to offer persuasive evidence of the parties’ preferences.139 Furthermore, commercial contracts are usually more complex than simple exchanges of a specific item for an agreed sum. All the terms that the parties agreed as part of the overall transaction may have affected the agreed price, so that once some terms fall away, it cannot be inferred that the stated price equates to the parties’ valuation of the item received.140 Once the market value of the enrichment is known, the defendant is entitled subjectively to devalue goods or services.141 He can show that they are worth less to him than the market rate. This mechanism protects his freedom of choice: he will not be forced to pay more than he ever would agree to pay for this benefit. The defendant may point to the contract price as evidence of how much he believed the benefit was worth, if that is lower than the market value. The law does not recognise the converse proposition, that the value of the enrichment should increase if the defendant would be willing to pay more than market price.142

ii.  Policy Concern: Contract Price as a Cap? Apart from the valuation question, there is a widespread policy concern that parties should not be allowed to use unjust enrichment to get more than they are entitled to in a contractual bargain.143 This concern is most acute for services. It has been proposed that the contractual price must be imposed as a ceiling where a services contract is terminated for breach, and that restitution for partial performance should be proportioned to the overall contract rate.144 Older authorities allowed non-breaching parties to recover above the contract price for their performance.145 However, contemporary English and Australian law is oriented towards limiting recovery to the contract price, or a proportion of it, in cases where a contract is terminated for breach. In Taylor v Motability Finance, Cooke J stated that even if restitution were possible where there was a valid contract, there could be ‘no justification … for recovery in excess of the contract limit’.146 This question was central in Mann v Paterson Constructions Pty

138 Barton v Gwyn-Jones (n 49) [50]. 139 S Arrowsmith, ‘Ineffective Transactions and Unjust Enrichment: A Framework for Analysis’ (1989) 9(2) Legal Studies 121, 129. 140 ibid 131. 141 Benedetti v Sawiris (n 136) [18]. 142 ibid. 143 A Kull, ‘Restitution as a Remedy for Breach of Contract’ (1994) 67 Southern California Law Review 1465, 1472; Birks, Introduction to the Law of Restitution (n 31) 46–47. 144 J Beatson, ‘Benefit, Reliance and the Structure of Unjust Enrichment’ (1987) Current Legal Problems 71, 84. 145 Lodder v Slowey [1904] AC 442. 146 Taylor v Motability Finance Limited (n 30) [26].

Planning for Failure  285 Ltd. There, builders had terminated a contract due to the other party’s breach, and sought payment in restitution for the work for which rights to contractual payment had not accrued.147 The contract price was less than the market value of the work done. The majority considered that the appropriate solution is to cap restitution at the contract price.148 The work was done on the basis of the agreed price, and there is nothing inherent in termination that requires departure from that.149 The dissenting judgment rejected the claim in restitution because of the same concern: the contract fixed a price for the work and the customer was entitled to rely on that bargain.150 The proper remedy for the work done for which the right to payment had not yet accrued was damages for loss of bargain, reflecting the contract price.151 It seems likely now that the contract price will constitute a ceiling in cases of terminated contracts. The need to keep the parties to their bargain is perhaps less compelling for invalid contracts. In Rover v Cannon, Kerr LJ deemed it incoherent to limit a claim to the amounts due under a contract whose invalidity was the reason for the claim.152 Besides, if the contract is not valid then direct arguments about evading a contractual bargain lose their force.

VII.  Failure of Consideration and the Parties’ Bargain Failure of consideration is usually the unjust factor that justifies restitution of unjust enrichment following termination or invalidity. It operates whatever form the enrichment takes.153 The underlying logic is that payment was made conditionally, and both parties knew the condition on which it was paid. If that condition is not met, it is recoverable.154 To apply this test, we must identify what the claimant’s intention was conditional upon. That condition must have been known by the other party, but this requirement is generally fulfilled where there is a contract between the parties. The terms of the contract will evidence what the parties understood each to be bargaining for, whether or not the contract is legally binding.

A.  The Contract and Conditions for Performance There is no single definition of ‘consideration’ in this context. Usually, where there is a valid contract, we are concerned with the actual performance of the 147 Mann v Paterson Constructions Pty Ltd (n 81). 148 ibid [101], [215]. 149 ibid [205]. 150 ibid [37]. 151 ibid [20]. 152 Rover International Ltd v Cannon Film Sales Ltd [1989] 1 WLR 912, 927. 153 Cobbe v Yeoman’s Row Management Ltd [2008] UKHL 55; Wilmot-Smith, ‘Replacing Risk-taking Reasoning’ (n 56) 617–18. 154 Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd (n 115) 65.

286  Niamh Connolly contractually-promised consideration.155 The question is not whether the claimant received any benefit. The focus is on the other party’s ‘essential obligation under the contract’.156 Sometimes, a person’s performance may be conditional on a state of affairs.157 It may implicitly be conditional on more than one thing, so that if either condition is not met, the performing party does not intend the recipient to have the benefit.158 It is the parties who decide what the consideration for each person’s performance will be. Judges will identify the core of the bargain objectively by interpreting the contract.159 Ancillary benefits do not count.160 A car buyer contracts for title, not mere possession.161 For a film distributor, receiving copies of films to distribute is merely facilitative.162 A racing driver bargains for test drives, not travel, pit passes and the right to sponsorship.163 Every transaction has its essential purpose. Though the core of the deal might seem inherent in the substance of the transaction, drafters have the power to frame and structure the deal as they wish and to make any element a condition. Stocznia Gdanska v Latvian Shipping Co indicates the scope for contracting parties to choose how to frame their transaction.164 Consider a manufacturer who constructs and sells a unique product. The transaction could simply be a sale, or it could be a bargain to design, construct and deliver it. If it is simply a sale, there will be total failure of consideration unless the finished product is delivered.165 In contrast, if the deal is to design and construct the product, performance of those stages means there will not be total failure of consideration. In Stocznia Gdanska, payment of the price in instalments supported the inference that the bargain encompassed design and construction. A choice to frame the bargain as for manufacture as well as delivery can assist the supplier if the deal is derailed before the product is delivered, because it will be able to retain some or all of the payments according to the work done. Contracting parties can communicate in their bargain that their performance will be conditional on elements that they subjectively consider essential. Performance could be conditional on counter-performance being of a certain quality.166 The lesson is to ensure that anybody reading the contract will know if there are particular elements or aspects of performance that are critical for

155 ibid 48; Stocznia Gdanska SA v Latvian Shipping Co (n 82) 588; Guinness Mahon & Co Ltd v Kensington and Chelsea Royal LBC [1998] EWCA Civ 294, [1999] QB 215, 240. 156 Guinness Mahon & Co Ltd v Kensington and Chelsea Royal LBC (n 155) 240. 157 Roxborough v Rothmans of Pall Mall (n 40). 158 F Wilmot-Smith, ‘Reconsidering “Total” Failure’ (2013) 72 CLJ 414, 434. 159 Giedo van der Garde BV v Force India Formula One Team [2010] EWHC 2373 (QB) [285]–[286]. 160 ibid [285]. 161 Rowland v Divall [1923] 2 KB 500. 162 Rover International Ltd v Cannon Film Sales Ltd (n 152) 924–25. 163 Giedo van der Garde BV v Force India Formula One Team (n 159). 164 Stocznia Gdanska SA v Latvian Shipping Co (n 82). 165 Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd (n 115); Stocznia Gdanska SA v Latvian Shipping Co (n 82) 590. 166 Gartell & Son v Yeovil Town Football & Athletic Club Ltd [2016] EWCA Civ 62.

Planning for Failure  287 these  parties. Drafters can further express what each individual obligation is conditional upon. For example, they might specify that payments intended to meet government levies are for that purpose and conditional on the charge’s being due.167 They can avoid obligations being characterised as entire obligations by indicating that payment is conditional on completion of components of performance. Payment in instalments is evidence of this intention.168 Equally, parties may provide that certain obligations are unconditional, or not conditional on counter-performance, as is the case with non-refundable deposits.169

i.  Consideration in Invalid Contracts Historically, English law resisted granting unjust enrichment when a party performed an invalid contract.170 The House of Lords later renounced this view because restitution does not indirectly enforce an ultra vires contract.171 A person who performs an invalid contract might invoke the unjust factors of either mistake or failure of consideration. However, mistake is not always available: a person who pays money under protest, thinking it is probably not due, does not pay because he is mistaken.172 Consideration is understood in a distinctive way for invalid contracts. A person’s performance is understood to be conditional not just on receiving counter-performance from the other party, but also on having a legal entitlement to that counter-performance.173 In Westdeutsche Landesbank Girozentrale v Islington London Borough Council, Lord Browne-Wilkinson stated that ‘the consideration for one party making a payment is an obligation on the other party to make counter-payments’.174 Even though the parties received the intended benefits under the contract, the fact that the contract was invalid meant there was no consideration. This might seem counterintuitive. It is possible because a person’s performance can be conditional on both legal entitlement and receipt of a benefit. If one of these conditions fails, the performing party does not intend the recipient to have the benefit. This seems to establish a principle of general application: payments made under an invalid contract are ‘necessarily made for a consideration which has totally failed and are therefore recoverable’.175 People generally perform their 167 Roxborough v Rothmans of Pall Mall (n 40). 168 Mann v Paterson Constructions Pty Ltd (n 81); P Butler, ‘Advance Contractual Payments’ in Rickett and R Grantham (eds), Structure and Justification in Private Law (n 44) 225, 231. 169 Beatson, ‘Non-cumul?’ (n 41) 20. 170 Sinclair v Brougham [1914] AC 398. 171 Westdeutsche Landesbank Girozentrale v Islington LBC [1996] AC 669, 687; Haugesund Kommune v Depfa ACS Bank (n 20) [87]. 172 Marine Trade SA v Pioneer Freight Futures [2009] EWHC 2656. 173 Westdeutsche Landesbank Girozentrale v Islington LBC [1994] 1 WLR 938, 953; Westdeutsche (n 171) 711; Guinness Mahon (n 155). 174 Westdeutsche (n 172) 711. 175 Guinness Mahon (n 155) 230.

288  Niamh Connolly contractual obligations on condition that both parties are actually obliged. However, it may be that some people do not perform conditionally on there being binding contractual rights.176 If so, the Westdeutsche solution might not apply to certain contracts. In Haugesund Kommune v Depfa ACS Bank, the court considered and rejected the argument that the bank took the risk of invalidity.177 Commercial parties could protect themselves in case of invalidity by providing that each party makes its payment on condition both that the contract is valid and that the other side performs. Conversely, they could indicate that each side performs solely in order to receive counter-performance and that their performance is not conditional on the contract’s being valid. The question is whether they would want there to be restitution if a fully-performed contract were discovered to be void.

B.  Total Failure of Consideration The common law requires that consideration must fail totally to ground recovery. There must be ‘a total failure of consideration, or a total failure of a severable part of the consideration’.178 Whincup v Hughes indicates why this rule may have seemed appropriate.179 When a master died near the start of an apprenticeship, the court found it impossible to value each stage of the apprenticeship and refused restitution of the premium. The relative value of the apprentice’s labour and master’s instruction would shift over time. However, the restriction is now almost universally criticised.180 An all-or-nothing rule does not produce just results. In many cases, there is no reason why there cannot be mutual counter-restitution. The courts long admitted restitution for partial failure where goods can easily be divided into units, of which the total price ‘is merely the arithmetic sum’.181 These days, they are increasingly willing to divide the overall considerations into parts, and allow restitution where some elements have failed.182 In Goss v Chilcott, Lord Goff differentiated between a borrower’s obligations to pay interest and repay the capital sum. Despite some interest payments, there was still a total failure of consideration for the loan capital.183 In a striking example, a builder had been paid most of the total contract price, but the court allowed restitution for the excess

176 C Mitchell, P Mitchell and S Watterson (eds), Goff & Jones: The Law of Unjust Enrichment, 9th edn (Sweet & Maxwell 2016) [13–32]. 177 Haugesund Kommune v Depfa ACS Bank (n 20) [143]–[153]. 178 Mann v Paterson Constructions Pty Ltd (n 81) [168]. 179 Whincup v Hughes (1870–71) LR 6 CP 78. 180 Wilmot-Smith, ‘Reconsidering “Total” Failure’ (n 158) 417. 181 SA Smith, ‘Concurrent Liability in Contract and Unjust Enrichment: the Fundamental Breach Requirement’ (1999) 115 LQR 245, 257. 182 David Securities Pty Ltd v Commonwealth Bank of Australia (1992) 175 CLR 353, 383. 183 Goss v Chilcott [1996] AC 788, 797–98.

Planning for Failure  289 work done.184 Judges are willing to calculate net enrichments where it is easy to do so and order mutual counter-restitution.185 The easiest case is where there have been monetary payments in two directions.186 Plenty of case law allows restitution despite the claimant’s having received part payments of the contract price. Birks understood these authorities to mean that total failure is not required when mutual counter-restitution is easy.187 We already value non-monetary benefits as enrichments, and can do the same for counter-restitution.188 Wilmot-Smith argues that the better interpretation of the law is that a claimant may only recover ‘if a substantial part of the condition is unsatisfied’.189 The total failure rule is now highly attenuated, and not likely to survive.190

i.  Avoiding the Total Failure Restriction Business actors are unlikely to have an appetite for arbitrary legal risks, such as the all-or-nothing rule that total failure of consideration creates. We have a cognitive bias that feels the effect of losses more than gains.191 This suggests that people would generally prefer to avoid a situation involving an equal chance of an arbitrary loss or windfall. A ‘maximin’ decision strategy, which focuses on minimising the risks of a worst-case scenario, favours avoiding an all-or-nothing rule like this.192 It follows that commercial actors would generally prefer to draft their contract in a way that facilitates more finely calibrated solutions. Despite the judicial trend towards allowing apportionment where possible, parties cannot depend on this happening. We have seen that how the parties frame their deal interacts with the question of whether consideration has failed totally.193 Drafters should also construct their deals in ways that facilitate apportionment by making it easier to divide them into parts. In Giedo van der Garde BV v Force India Formula One Team, the complex bundle of entitlements made it impossible to determine how much of the contract price corresponded to specific test drives.194 Where there are multiple elements of consideration within a transaction, the parties should itemise each element of the bargain and value them by attributing parts of the overall consideration to specific benefits. In transactions such as construction projects, parties can facilitate apportionment and the accrual of 184 DO Ferguson v Sohl [1992] 11 WLUK 211. 185 David Securities (n 182) 383; Goss v Chilcott (n 183) 798. 186 Westdeutsche (n 173) 945. 187 Birks, Introduction to the Law of Restitution (n 31) 242. 188 J Edelman, ‘Restitution for a Total Failure of Consideration: When a Total Failure is Not a Total Failure’ (1996) 1 Newcastle Law Review 57, 64. 189 Wilmot-Smith, ‘Reconsidering “Total” Failure’ (n 158) 422. 190 Smith, ‘Concurrent Liability’ (n 181) 245. 191 D Kahneman and A Tversky, ‘Prospect Theory: an Analysis of Decision under Risk’ (1979) 47 Econometrica 263, 279. 192 JL Harrison, ‘A Case for Loss Sharing’ (1983) 56 Southern California Law Review 573, 597–98. 193 Stocznia Gdanska SA v Latvian Shipping Co (n 82) 600. 194 Giedo van der Garde BV v Force India Formula One Team (n 159).

290  Niamh Connolly contractual rights by dividing work into phases involving mutual performance.195 Staged payments are likely to be interpreted as meaning that the corresponding performance obligations contain ‘divisible obligations of performance’.196 This will allow accurate, controlled restitution depending on how many milestones have been reached. Commonly, part of the contract price is attributable to transaction taxes. Parties can protect themselves against the risk that these are invalid. The person who pays taxes indirectly may need to reclaim them from his contracting partner because he cannot recover them directly from the state.197 The obstacles he will face are the continuing validity of the contract and the need to show total failure of consideration. The High Court of Australia decision in Roxborough indicates a solution.198 Because the tobacco licence fee was not negotiated and was listed as a distinct part of the contract price, the Court found that there was total failure of consideration for it. The structure of the deal indicated the parties’ shared understanding that this sum was specifically intended to pay the tax. This part of the price could be apportioned and the consideration for it had failed. It is important to identify taxes and external levies separately, distinct from the bargain price, so that restitution might be possible if they turn out to be invalid.199

VIII. Conclusion The default common law and statutory rules provide satisfactory remedies in cases where a contractual bargain goes awry. There is no gap in the law that requires parties to fend for themselves. However, they do have scope for direct and indirect control of restitution. Indeed, choices about contract design necessarily determine how failure of consideration will apply to every transaction. It follows that lawyers should consciously exercise this control. Lawyers can assist their commercial clients by considering these issues when structuring bargains and drafting contracts. Alongside the risks they routinely consider, they can usefully take account of the legal risk of the contract’s becoming ineffective. It is easier than one might assume to plan for contractual ineffectiveness, because drafters do not need to predict the specific events that might lead to termination, impracticability or invalidity.200 The real thinking that is required is

195 See Stocznia Gdanska SA v Latvian Shipping Co (n 82); Butler, ‘Advance Contractual Payments’ (n 168) 232. 196 Mann v Paterson Constructions Pty Ltd (n 81) [26], [176]. 197 Investment Trust Companies (In Liquidation) v Revenue and Customs Commissioners [2018] AC 275. 198 Roxborough v Rothmans of Pall Mall (n 40). 199 Orphanos v Queen Mary College (n 58). 200 Choi and Triantis, ‘Strategic Vagueness’ (n 3) 865.

Planning for Failure  291 to identify the remedies that the parties will want depending on what performance has occurred when the contract becomes ineffective. Termination clauses and force majeure clauses are extremely common; they should address consequential financial readjustment, including restitution and loss sharing. The interaction of the doctrine of failure of consideration with the structure of people’s bargains may be less widely considered in practice. If those who draft contracts exert the control that the law allows them when designing contracts, they can achieve greater choice, control and certainty when restitutionary scenarios arise. Parties can calibrate when payments accrue and what adjustments will be required according to when performance is interrupted. They can make it clear what each party considers the core of the bargain and what each act of performance is conditional upon. They can ensure more nuanced solutions than the requirement for total failure of consideration might produce. Some general provisions are suited to routine inclusion. For example, parties could stipulate that each party performs its contractual obligations on the condition that this contract is valid. This would copper-fasten the likely inference that there is a failure of consideration where the contract is invalid. The parties could state that they consider that performance on both sides is divisible, so that part of the overall price corresponds directly to smaller components of each side’s consideration. Similarly, they may indicate that they agree that there should be mutual counter-restitution for incomplete performance. To assist a judge further in valuing separate components, contract drafters should indicate which portions of the consideration on each side mutually correspond. This will require careful thought about how much money a party would need to receive to protect it from loss if the deal were interrupted at various milestones. Engaging with these issues at the contracting stage can enhance certainty by forestalling doubts about how the law might apply to the parties’ deal. Ultimately, this may enable them to reduce or eliminate litigation costs if things go wrong.

292

14 ‘All Watched Over by Machines of Loving Grace’? The Inevitable Conflict between Contract Law and Free Speech in Cyberspace NICHOLAS J McBRIDE*

‘[W]hen differences are sufficiently far reaching, we try to kill the other man rather than let him have his way.’ OW Holmes, ‘Natural Law’ (1918) 32 Harvard Law Review 40, 41 ‘Persecution for the expression of opinions seems to me perfectly logical. If you have no doubt of your premises or your power, and want a certain result with all your heart, you naturally … sweep away all opposition.’ Holmes J, Abrams v United States, 250 US 616, 630 (1919)

This chapter seeks to make out three theses: (a) It is inevitable that contract law will be relied on to censor conservative views that are expressed on the Internet. (b) Private law will provide very little protection to conservatives whose views are subject to such censorship. (c) We should not be surprised by (b) – private law cares very little about freedom of speech. It is not, therefore, a particularly edifying or inspiring story that I wish to tell today. It is instead one that suggests there is something wrong with private law, and that it is in need of fundamental reform.

* All references to online resources are up-to-date as of 31 August 2019.

294  Nicholas J McBride

I.  Welcome to the Machine It seems undeniable that conservative voices are subject to increasing censorship on the Internet.1 The first high-profile victim of such censorship was the rightwing conspiracy theorist and provocateur Alex Jones. Following a campaign against Jones’s presciently titled ‘InfoWars’ channel by CNN, Apple removed Jones’s InfoWars podcasts from its iTunes and Podcast apps on 5 August 2018. Facebook and YouTube followed suit the next day, deleting all InfoWars videos that had been uploaded to those platforms over many years, and – at least before 2016 – with no complaints or concerns from either Facebook or YouTube. The following day, the InfoWars company page was removed from LinkedIn, the email messaging service MailChimp deleted InfoWars’ and Jones’s email accounts, and InfoWars’ pages on Pinterest were also removed. Twitter initially declined to follow suit and delete Jones’s or InfoWars’ Twitter accounts. In a tweet on 8 August, company head Jack Dorsey explained that ‘We didn’t suspend Alex Jones or InfoWars yesterday. We know that’s hard for many but the reason is simple: he hasn’t violated our rules. We’ll enforce if he does.’ The decision was condemned on the Vox news website as ‘breathtakingly immoral, as well as regressive, terrible decision-making’.2 Jones’s and InfoWars’ Twitter accounts were subsequently permanently suspended on 6 September for violating Twitter’s ‘abusive behaviour policy’, one day after Jones posted up on Twitter videos of his confronting various individuals at the US Senate (including Dorsey) about the limits that had been placed on his freedom of speech. The particular video Twitter seems to have taken exception to was one where Jones berated CNN journalist Oliver Darcy for his personal involvement in the campaign to shut down InfoWars. My first thesis is that attempts to censor individuals like Jones on the Internet – and others who do not go anywhere near as far as Jones does in the extremity of his views3 – are inevitable, and that contract law (‘the rules’ referred to by Jack Dorsey) will be the preferred vehicle for censoring right-wing voices online. In order to understand why this is, we have to understand a particular worldview that is prevalent both in Silicon Valley and – for reasons I shall explore below – within journalism, the entertainment industry and academia. I call this worldview liberal gnosticism. In order to understand liberal gnosticism, we first need to talk about gnosticism. 1 Any doubts on this count will be completely dispelled by watching episode #1258 of The Joe Rogan Experience, at www.youtube.com/watch?v=DZCBRHOg3PQ, where left-liberal journalist Tim Pool takes on Twitter’s CEO Jack Dorsey and lawyer Vijaya Gadde over Twitter’s record of censoring conservative voices. 2 A Romano, ‘Twitter’s stance on InfoWars’ Alex Jones should be a moment of reckoning for users’ (Vox 8 August 2018) available at www.vox.com/2018/8/8/17662774/twitter-alex-jones-jack-dorsey. 3 Other well-known right-wing subjects of online censorship include: Owen Benjamin, Joe Biggs, Steven Crowder, ‘Carpe Donktum’, ‘Diamond and Silk’, Mark Dice, ‘Godfrey Elfwick’, ‘Harold Finch’, Pamela Geller, ‘Imperator Rex’, Laura Loomer, Michelle Malkin, Gavin McInnes, Stefan Molyneux, Candace Owens, Larry Schweikart, ‘soph’, Lauren Southern, Roger Stone, Brandon Straka, Paul Joseph Watson, Thomas Wictor, James Woods and Milo Yiannopoulos. (Speech marks indicate online pseudonyms.)

Limiting Free Speech on the Internet  295

A. Gnosticism In discussing gnosticism, it is customary and important to distinguish ancient (or religious) and modern (or secular) forms of gnosticism. The term ‘gnosticism’ is derived from the Greek word for ‘knowledge’, gnosis. The particular kind of knowledge that was shared by ancient (or religious) gnostics was that the universe is evil, and the earth is nothing but a miserable prison for human beings.4 Salvation from the grim plight in which human beings find themselves can only come from a benevolent God, who is not responsible for the evil state of the universe but has the power to intervene to create a new universe and, within it, a new earth that will provide a happier home for human beings. Ancient gnosticism was killed off by the rise of Christianity, but has been replaced by a modern (or secular) form of gnosticism. Like their ancient counterparts, modern gnostics believe that there is something wrong with the world, but unlike ancient gnostics, modern gnostics believe that it is possible for human beings to recreate the world in a way that will eliminate the evils that currently afflict it. So for modern gnostics, the salvational role played by God in ancient gnosticism is instead played by human beings – or at least the select group of human beings who are ‘in the know’ as to what is wrong with the world and what can be done to set things right. The political theorist Eric Voegelin – who played the leading role in identifying modern gnostic trends in political thought5 – traced the development of modern gnosticism back to the ninth century, but argued that the writings of the twelfth-century theologian, Joachim of Fiore, gave the most significant impetus to modern gnosticism. Joachim argued that history was divided into three different epochs, the final one of which would commence in 1260 and would see complete freedom and universal love reign over the earth. All forms of modern gnosticism share this feature of Joachim’s thought: of looking forward to a time when history – in the sense of a succession of social arrangements – will culminate and terminate in a social order that amounts to ‘heaven on earth’.

B.  Liberal Gnosticism Not all forms of modern gnosticism are liberal in nature. For example, Friedrich Nietzsche’s thought was both gnostic and illiberal. What is wrong with the world, according to Nietzsche, is the fact that it is reigned over by the ‘last men’ – stultifyingly boring mediocrities who uncomprehendingly blink at questions like ‘What is love? What is creation? What is longing? What is a star?’ and lead small lives

4 For discussion, see K Rudolph, Gnosis: The Nature and History of Gnosticism (Harper & Row 1983); H Jonas, The Gnostic Religion, 3rd edn (Beacon Press 2001). 5 See, in particular, E Voegelin, The New Science of Politics (University of Chicago Press 1952); and E Voegelin, Science, Politics, and Gnosticism (Regnery Gateway 1968).

296  Nicholas J McBride concerned with seeking out warmth and avoiding pains and burdens.6 Under the reign of the last men, ‘Everyone wants the same thing, everyone is equal, and anyone who thinks differently goes voluntarily into the madhouse.’7 This hell on earth (in Nietzsche’s eyes) is destined, Nietzsche contended, to implode by virtue of both its vacuity and the removal of its religious underpinnings. The implosion will be marked by a ‘long … sequence of breakdown, ruin, and cataclysm … [and] a gloom and an eclipse of the sun whose like has probably never yet occurred on earth’.8 This time of ruin will be succeeded by an age when certain men will find the courage to throw off the shackles that they formerly allowed to be placed on their basic human ‘will to power’ (both over themselves and other people), and they will be transformed into noble ‘beast[s] of prey avidly prowling around for spoil and victory’.9 A man who transforms himself into such a lion-like ‘blonde beast’ will be someone ‘who justifies man himself … who makes up for and redeems man, and enables us to retain our faith in mankind!’10 Nietzsche’s gnosticism had as its target liberal gnosticism, which sees the ‘end of history’ as culminating in a social order made up of the ‘last men’ that Nietzsche despised.11 For liberal gnostics, the evil of which the world needs to be cleansed – and will be cleansed if liberal nostrums are observed – is not, pace Nietzsche, mediocrity and smallness of spirit but cruelty. As Judith Shklar observes, with the agreement of other liberals such as Annette Baier12 and Richard Rorty,13 ‘liberal and humane people … would choose cruelty as the worst thing we do’14 and liberalism begins ‘with a summum malum, which all of us know and would avoid if only we could. That evil is cruelty and the fear it inspires, and the very fear of fear itself.’15 The kind of cruelty that Shklar, Baier and Rorty – and their predecessors16 – seemed to have had in mind as the greatest evil is what we might call the cruelty of brutality.17 That is, the deliberate infliction of violence on another for no other sake than the enjoyment of seeing someone suffer, or the deliberate infliction of 6 F Nietzsche, Thus Spoke Zarathustra (1883) Prologue. 7 ibid §5. 8 F Nietzsche, The Gay Science (1882) §343. cf F Nitezsche, Ecce Homo (1888), ‘Why I am a destiny’ §1: ‘One day my name will be connected with the recollection of something enormous – with a crisis like never before on earth, with the deepest clash of conscience, with a decision solely invoked against all that had until then been believed, demanded, hallowed.’ 9 F Nietzsche, On the Genealogy of Morality (1887) First essay, §11. 10 ibid §12 (original emphasis). 11 See F Fukuyama, The End of History and the Last Man (Free Press 1992), proclaiming – from a liberal gnostic perspective – that the end of history has arrived. 12 A Baier, ‘Moralism and Cruelty: Reflections on Hume and Kant’ (1993) 103 Ethics 436. 13 R Rorty, Contingency, Irony, and Solidarity (CUP 1989) ch 4. 14 J Shklar, Ordinary Vices (Harvard University Press 1984) 44. 15 J Shklar, ‘The liberalism of fear’ in NL Rosenblum (ed), Liberalism and the Moral Life (Harvard University Press 1989) 29. 16 Shklar invokes Michel de Montaigne and Charles Montesquieu; Baier, David Hume; and Rorty, George Orwell and Vladimir Nabokov. 17 See also J Glover, Humanity (Jonathan Cape 1999), which is explicitly designed to reinvigorate the moral resources required to ensure that ‘we find [atrocities] intolerable and will do what we can to eradicate them’ (44).

Limiting Free Speech on the Internet  297 terrible physical suffering on another for any reason whatsoever.18 However, even at the time Shklar, Baier and Rorty were writing on this topic – in the late 1980s and early 1990s – the summum malum that liberal politics was designed to target had also come to cover other kinds of cruelty. One such cruelty is the cruelty of humiliation – of making some feel that they are second-rate. Another kind, and one which is particularly important in our politics today, is what I shall call the cruelty of exclusion.19 Someone experiences the cruelty of exclusion when they are denied the ability to take advantage of a valuable opportunity or lifestyle because of who they are. That the liberal summum malum covers the cruelty of exclusion seems obvious when one surveys the overriding emphasis that is placed by liberals nowadays on the politics of identity and inclusion. The eradication of the bar on same-sex couples getting married, debates around the issue of who should be allowed to use which restrooms and which changing rooms, the ‘open borders’ movement, the agitation to allow severely disabled people the same opportunities of ending their lives (through euthanasia or assisted suicide) that able-bodied people have, campaigns in favour of women priests, and the equivalent trend in the theatre of encouraging women to play traditionally male roles – all these developments seem targeted at bringing to an end the cruelty of exclusion, and they are all developments that liberals would pride themselves on having instigated. The pride is justified, in that cruelty is (presumptively) a bad thing and therefore the elimination of cruelties around what roles one can play in the theatre, etc, is (presumptively) a good thing. However, what makes liberal gnosticism distinctive are the beliefs that: (i) cruelty is the worst thing in the world, so that almost any measures are justified in order to eliminate it and no good reasons can ever be given for not eliminating it where it exists; and (ii) it is possible to live in a cruelty-free world.

C.  Potential Objections to Liberal Gnosticism Both conservatives and what we might call liberal realists20 would reject the tenets of liberal gnosticism, as set out in (i) and (ii) in section I.B. 18 This is not a definition offered by any of Shklar, Baier or Rorty, but it does not seem to be one from which any of them would have seriously dissented. For discussion of Shklar’s (not very satisfactory) positions on the meaning of cruelty, see J Kekes, ‘Cruelty and Liberalism’ (1996) 106 Ethics 834, 836. 19 The desire to extirpate the world of cruelties of exclusion had become sufficiently familiar and widespread by the start of the 1970s that it could be expressed in song form by John Lennon in his 1971 hit Imagine. Going much further back, the spread of Protestantism and the French Revolution can be seen as attempts to rid the world of hierarchies of knowledge and privilege that, by virtue of their hierarchical nature, necessarily gave rise to cruelties of exclusion. See also J Heller, Picture This (Putnam 1988) 120: ‘In 1646 … Amsterdam bakers of fancy cakes were prohibited from displaying overdecorated wares in their windows “lest they bring sadness to people too poor to buy them …”’. 20 A phrase suggested to me by Oliver Butler. Modern-day examples of liberal realists would be Dave Rubin and the already-mentioned (n 1) Tim Pool. Both Rubin and Pool have also been subject to online censorship.

298  Nicholas J McBride A liberal realist will tend to reject (i) and (ii) because they will argue that some cruelties, bad though they are, have to be accepted as inevitable. In a world characterised by the scarcity of goods, it is inevitable that some people will lose out when those goods are allocated. Such people will experience the cruelty of exclusion – it happens every year when thousands of students are told that their university applications have been unsuccessful – but such cruelty is, sadly, inescapable. Where it is inescapable, the best we can hope for is to convince those on whom the cruelty of exclusion has been visited that there was some good reason why it was visited on them and not someone else. But that becomes very difficult, if not impossible, within the liberal gnostic view that ‘cruelty is the worst thing we do’. A conservative will tend to reject (i) and (ii) because they will reject the belief that lies at the root of all forms of gnosticism. This is the belief that the evils that afflict humanity are not humanity’s fault. In the case of ancient gnosticism, they are the fault of the structure of the universe; in the case of modern gnosticism, they are society’s fault. The evils will disappear when the universe/society is reordered. However, conservatives will agree with Aleksandr Solzhenitsyn’s view that the line separating good and evil passes not through states, nor between classes, nor between political parties either – but right through every human heart – and through all human hearts. This line shifts. Inside us it oscillates with the years. And even within the hearts overwhelmed with evil, one small bridgehead of good is retained. And even in the best of all hearts, there remains … an unuprooted small corner of evil.21

If this is right, then no form of gnosticism can possibly succeed – the evils from which it seeks to free humanity are ineliminable because they are internal to humanity. They can only be destroyed by destroying humanity. If the previous point is right then not only will liberal gnosticism prove futile, it may actually end up replicating the exact evil that it is supposed to root out. As reality proves resistant to being remade in a way that will purge cruelty from society, more and more extreme measures may be resorted to in order to achieve this end. Communism is the ultimate form of liberal gnosticism, when it is pushed to its very limits, and within communist societies the gulag is the ultimate weapon for achieving liberal gnostic goals.22 And it may well be that some liberal gnostics will feel no compunction about implementing such extreme methods to achieve their goals.23 As Charles Taylor observes: The very moment when we feel purest ourselves is when we’re hammering the really bad guys, making war on Milosevic, or fighting the ‘axis of evil.’ We at last know who are 21 A Solzhenitsyn, The Gulag Archipelago (1958–68) vol II, pt IV, ch 1 (ellipsis in the original). 22 This explains why a typical liberal gnostic does not feel any horror at communism, when Nazism does (quite rightly) fill them with horror – and this is despite the fact that communist regimes have (so far) killed about 100 million people, four times as many as the Nazis managed. Liberal gnosticism is fervently opposed to Nietzschean gnosticism, and Nazism is a form of Nietzschean gnosticism, while communism is a form of liberal gnosticism. 23 Witness: (i) how many liberals have recently found themselves willing to consider whether democracy has outlived its usefulness; (ii) the rise of the paramilitary wing of liberal gnosticism in the shape of Antifa;

Limiting Free Speech on the Internet  299 making the world worse, and who thus needs to be conquered or eliminated. We make a feast of our righteous anger. This is the moment we’re readiest to allow ourselves the worst atrocities. And not even notice it, at least at the time.24

But even this, as bad as it is, is not actually the worst outcome that might result from attempts to implement liberal gnosticism. The worst outcome will follow on the moment when a liberal gnostic, having gone to extremes to create a world free from cruelty, realises that Solzhenitsyn was right after all, and that the liberal gnostic dream of living in a cruelty-free world can only ever be exactly that – a dream, not a reality. At such a moment, Bernard Williams points out, ‘the result is likely to be … self-hatred and self-contempt … The self-hatred, in this case, is a hatred of humanity.’25 And there are no limits to the cruelties that might be inflicted by someone whose disillusionment with liberal gnosticism has inspired within them a genuine hatred of humanity. The points that liberal realists and conservatives make against liberal gnosticism seem to have something going for them, and are worth discussing. However, liberal gnostics seem incapable of considering that their worldview is incorrect, and instead dismiss criticisms of liberal gnosticism as ‘fake news’ and ‘hate speech’.26 What is it about the worldview represented by liberal gnosticism that accounts for (i) why it is adopted in the first place, in the face of fairly obvious objections to it such as those made above, and (ii) why adherents to this worldview find it impossible to consider that their worldview is incorrect, when they would find it very easy and interesting to consider whether or not other aspects of their worldview (for example, that the world we live in actually exists and is not just a computer simulation) are incorrect?

D.  The Prevalence and Fervour of Liberal Gnosticism Histories of the computer revolution make it clear that liberal gnosticism has been hard-wired into the worldview of Silicon Valley since its origins in the (iii) increasing attempts to deny critics of liberal gnosticism access to money by (a) preventing them from using platforms like Patreon and PayPal to obtain donations from supporters and (b) discouraging banks from dealing with them – something which, if taken to its logical conclusion, would result in people having to pass a political test in order to be allowed to buy food. Shklar herself, despite being a childhood refugee from communism, was willing to make an ‘exception to the rule of avoidance [of cruelty]’ where cruelty is required for ‘the prevention of greater cruelties’ (Shklar ‘The liberalism of fear’ (n 15) 30). 24 C Taylor, ‘Perils of Moralism’ in C Taylor, Dilemmas and Connections: Selected Essays (Harvard University Press 2011) 347, 363. See also C Lasch, The Revolt of the Elites and the Betrayal of Democracy (WW Norton 1995) 28: ‘Opposition makes humanitarians forget the liberal virtues they claim to uphold. They become petulant, self-righteous, intolerant. In the heat of political controversy, they find it impossible to conceal their contempt for those who stubbornly refuse to see the light – those who “just don’t get it”, in the self-satisfied jargon of political rectitude.’ 25 B Williams, ‘The Human Prejudice’ in B Williams, Philosophy as a Humanistic Discipline (Princeton University Press 2006) 152. 26 An amusing recent example is Facebook’s banning as ‘hate speech’ St Augustine’s statement that ‘Let us never assume that if we live good lives we will be without sin … men are hopeless creatures’: Catholic Herald, 16 July 2019.

300  Nicholas J McBride hippie communes of the 1960s.27 Marshall McLuhan’s 1964 statement that ‘The computer … promises by technology a Pentecostal condition of universal understanding and unity’28 is representative of the kind of liberal gnostic worldview that inspired the creation of personal computers and the Internet, and continues to inspire tech companies such as Apple and Google to this day. (The same is true of the poem that supplies this chapter with its title.29) However, this does not account for why liberal gnosticism should also have made serious inroads into the worlds of journalism, entertainment and academia. Some may argue that liberal gnosticism is so popular in these particular fields of activity because that is where all the smart people are.30 But such an observation does not solve but rather magnifies our puzzle. Given that there are so many smart people employed in these fields of activity, why are so few of them alert to the possible weaknesses in liberal gnosticism pointed out here, and why are so few of them willing even to consider that the case against liberal gnosticism might be stronger than the case for? I think a possible answer to this question might lie in the following proposition – the more plastic your profession, the more likely you are to be a liberal gnostic. To explain: the plasticity of a profession can be measured by the variety of the paid work you do, and the number of uses that can be made of what you produce through that work. The more plastic your profession, the harder it will be for you to gain a sense of self-worth and self-respect from the paid work you do, because the paid work you do will not provide you with anything solidly worthwhile with which you can identify yourself. The result is that the more plastic your profession, the more likely it is that you will have to identify yourself with something outside the paid work you do in order to gain a sense of self-worth and self-respect. For some, this may be the unpaid work they put into raising a family. But for many others it will be identifying themselves with a gnostic creed such as liberal gnosticism. Doing so allows them to see themselves as being on the ‘right side of history’, on the side of the ‘good guys’ fighting evil, thereby giving them the sense of selfworth and self-respect that they cannot get from their work. If this is right then it would also follow that someone who has been led to become a liberal gnostic by virtue of the plasticity of their profession will be highly resistant to suggestions that liberal gnosticism is flawed: entertaining such suggestions would trigger an identity crisis by endangering the foundations of their sense of self-respect and self-worth. 27 See, eg, F Foer, World Without Mind (Jonathan Cape 2017) 11–30; F Turner, From Counterculture to Cyberculture: Stewart Brand, the Whole Earth Network, and the Rise of Digital Utopianism (University of Chicago Press 2006). See also E Morozov, To Save Everything, Click Here (Penguin 2014). 28 M McLuhan, Understanding Media (Routledge & Kegan Paul 1964) 80. 29 R Brautigan, ‘All watched over by machines of loving grace’ in All Watched Over by Machines of Loving Grace (Communication Company 1967). 30 See, eg, Leslie Green’s post, L Green, ‘Why it is hard to be a campus conservative’ (Semper Viridis, 18 September 2017) available at https://ljmgreen.com/2017/09/18/why-it-is-hard-to-be-a-campusconservative/.

Limiting Free Speech on the Internet  301 It should be obvious that the kinds of employment where liberal gnosticism has made most headway are characterised by a high degree of plasticity. The products produced by Silicon Valley tend to be what Ivan Illich called ‘tools for conviviality’: ‘they can be easily used, by anybody, as often or as seldom as desired, for the accomplishment of a purpose chosen by the user’.31 They are therefore highly plastic, with corresponding costs for those that supply those products in terms of their being able to get any sense of self-respect or self-worth from the work they do supplying those products. The kind of work done by journalists, actors and academics is plastic in another way, in that the nature of the work they are paid to do is relatively heteronomous – changing circumstances repeatedly force them to hop from one topic/issue/production to another, with the result that there is usually no continuing through-line in the work that they do that they can fasten onto and identify with for personal validation. Instead, and in line with the prediction in the previous paragraph, validation tends to come from outside their work, and comes from the causes they personally identify with; those causes tend to be liberal gnostic in nature.

E.  Enter Contract Law It follows that those who use the Internet to express views that question liberal gnosticism are using platforms that cannot but be tilted against them. Those who maintain those platforms will very likely be liberal gnostics, and are just as likely to resist the use of these platforms – that they have worked hard to create and maintain – to attack views that are central to their sense of self-respect and self-worth. My first thesis – and it is one that is borne out both by Jack Dorsey’s initial response to the co-ordinated attempt to ‘unplatform’ Alex Jones and by the basis on which he later permanently suspended Jones’s account – is that liberal gnostics who seek to control speech on the Internet will typically prefer to do so by acting against those who can be colourably alleged to have violated the terms of service governing someone’s ability to use a particular platform on the Internet. This is for a variety of reasons. First, attempts to control speech that cannot be said to be in violation of the terms and conditions on which a company is willing to provide a platform for speech on the Internet will inevitably appear to be arbitrary and illegitimate. Second, and as we shall see, the control of speech that does not violate a platform provider’s terms and conditions always incurs the risk of a lawsuit by the person whose speech has been controlled, on the basis that the platform provider is in breach of their contract with the claimant. Third, the control of speech that can be said to be in violation of those terms and conditions will not be regarded as objectionable by those on the right who take the view that owners of private property

31 I

Illich, Tools for Conviviality (Calder & Boyars 1973) 22.

302  Nicholas J McBride should be allowed virtually unlimited freedom to determine the terms on which others are allowed to take advantage of that property. Hence contract law will, for the time being,32 be the preferred vehicle for controlling Internet-based speech that challenges the tenets of liberal gnosticism. Unsurprisingly, the terms and conditions that govern the use of platforms such as Facebook, Twitter and YouTube provide plenty of scope for controlling such speech. Users of any of these platforms are taken to have entered into a contract with the company managing the platform. The terms of this contract are defined by the company’s ‘Terms of Service’.33 These terms of service require the user to observe further rules as to what content can be posted on the platform – in the case of Facebook, these rules are set out in its ‘Community Standards’; for Twitter, the ‘Twitter Rules’; and for YouTube, its ‘Community Guidelines’.

i. Facebook Facebook’s Community Standards ‘do not allow hate speech … because it creates an environment of intimidation and exclusion’.34 Hate speech is defined as a ‘direct attack’ on people based on a variety of protected characteristics, including ‘national origin, religious affiliation, sexual orientation … gender, gender identity and serious disease or disability’, and in some cases ‘immigration status’. The concept of an ‘attack’ includes ‘dehumanising speech’ or ‘calls for exclusion’. The Community Standards also allow Facebook to reduce ‘the spread of false news on Facebook … by showing it lower in the News Feed’.35 ‘False news’ is identified either by ‘third-party fact-checkers’, or by ‘a machine learning model that predicts which stories may be false’ by relying on ‘various signals, including feedback from our community’.

ii. Twitter Under the Twitter Rules,36 users are forbidden from engaging in ‘abusive behaviour’, which is defined as ‘an attempt to harass, intimidate, or silence someone else’s voice’. ‘Hateful conduct’ is also barred. This is defined as threatening or harassing other people on the basis of a list of protected characteristics very similar to Facebook’s. Twitter’s ‘Hateful Conduct Policy’ explains further that Twitter prohibits ‘targeting individuals with content intended to incite fear or spread fearful 32 Note that Twitter has recently expressed a willingness to censor tweets that do not technically violate its rules: ‘The challenge for us has been: how can we proactively address … disruptive behaviors that do not violate our policies but negatively impact the health of our conversation?’ (See at https:// blog.twitter.com/official/en_us/topics/product/2018/Serving_Healthy_Conversation.html.) 33 Facebook, at www.facebook.com/terms.php; Twitter: https://twitter.com/en/tos; YouTube, at www. youtube.com/static?template=terms. 34 www.facebook.com/communitystandards/hate_speech. 35 www.facebook.com/communitystandards/false_news. 36 https://twitter.com/rules.

Limiting Free Speech on the Internet  303 stereotypes about a protected category’ and ‘targeting individuals with repeated slurs, tropes or other content that intends to dehumanize, degrade or reinforce negative or harmful stereotypes about a protected category’. Twitter’s policy on ‘Abusive behaviour’ permits Twitter to ‘take action against excessively aggressive insults that target an individual, including content that contains slurs or similar language’.

iii. YouTube YouTube’s Community Guidelines do not allow content to be posted on YouTube that promotes ‘violence or hatred against individuals or groups based’ on the same kind of list of protected characteristics used by Facebook and Twitter, but also including ‘age … [and] veteran status’. Under YouTube’s ‘Harassment and cyberbullying policy’, ‘Content or behaviour intended to maliciously harass, threaten or bully others is not allowed’. The policy further instructs YouTube content creators not to post content that ‘is deliberately posted in order to humiliate someone’, or which ‘makes hurtful and negative personal comments … about another person’ or claims ‘that specific victims of public violent incidents or their next of kin are actors, or that their experiences are false’. YouTube content providers who enjoy a large number of followers can apply for their videos to be ‘monetised’ under YouTube’s Partner Program. Under this program, advertisements will be placed on the content provider’s videos and YouTube will split the fee for the advertisement with the content provider. However, a video posted by a member of the Partner Program will be ‘demonetised’ (so the member makes no money from the video) if it violates the ‘Content policies’ section of YouTube’s ‘AdSense program policies’.37 Under these ‘Content policies’, among many other things, videos that include ‘Shocking content’ or ‘Content that harasses, intimidates or bullies an individual’ (which is said to include ‘Content that suggests a tragic event did not happen’) will be demonetised.

II.  Gimme Shelter It is clear how terms such as those described in section I.E might be used to suppress and control the speech of those sceptical of the liberal gnostic agenda. Such speech might well (and probably will in the case of conservative voices) include claims that: (i) the liberal gnostic worldview is radically detached from reality, with the result that liberal gnostics are incapable of thinking critically or intelligently about the state of the world, and are positively dangerous when they attempt to influence government policy; (ii) journalists who have signed up to a liberal gnostic agenda are, as a result, highly selective about what news they report and what

37 https://support.google.com/adsense/answer/48182.

304  Nicholas J McBride they do not report, and some such journalists manufacture news to advance their agenda; (iii)  various government agencies have engaged, and may well engage in the future, in staging ‘false flag’ incidents for the purpose of advancing liberal gnostic goals; (iv) the government of a particular country should focus exclusively on protecting the interests of that country’s population and not concern itself with the fate of people living in other countries; (v) the quality of certain TV series and film franchises have plummeted because they have been taken over by creators and showrunners who are concerned to use those series and franchises to push a liberal gnostic agenda. All such speech could be colourably characterised as violating the rules, standards and guidelines set out in the previous section. Obviously, liberal gnostics will not, and do not, care if speech like this is suppressed on the Internet.38 However, those (including me) who value freedom of speech – on the basis that allowing ‘a free and open encounter’ between Truth and Falsehood is the best and only way of ensuring that we do not fall into Falsehood39 – will care a great deal that we do not end up living in a world where the only speech that is allowed online is speech that is consistent with the liberal gnostic worldview. Concerns at the inroads that have already been made on free speech on the Internet through invocation of the rules and standards employed by Facebook, YouTube and Twitter have resulted in calls for a variety of regulatory reforms, such as (i) a digital Bill of Rights, which would have the effect of extending First Amendment protection to speech on the Internet;40 (ii) amending the immunity from civil liability enjoyed by providers of interactive computer services under section 230 of the (American) Communications Decency Act, so that such immunity will be withdrawn from those who provide their services in a way that evidences demonstrable political bias; or (iii) opening up companies such as YouTube and Twitter to competition from rivals that are avowedly in favour of unrestricted free speech, such as (in the case of Twitter) Gab.ai or (in the case of YouTube) BitChute.41 However, my interest in this part of my chapter is in exploring the extent to which the rules and doctrines of private law can be used to protect freedom of speech on the Internet. My second thesis is that such rules and doctrines provide very limited protection for such speech. Let us begin making out this thesis by looking at the contractual arguments that might be made by a right-winger whose speech has been censored online under the rules set out in the previous section. 38 Laying the intellectual foundations for suppressing freedom of speech online, see P Napoli, ‘What if More Speech is No Longer the Solution? First Amendment Theory Meets Fake News and the Filter Bubble’ (2018) 70 Federal Communications Law Journal 55; AE Waldman, ‘The Marketplace of Fake News’ (2018) 20 University of Pennsylvania Journal of Constitutional Law 845; T Wu, ‘Is the First Amendment Obsolete?’ (2018) 117 Michigan Law Review 547. 39 J Milton, Areopagitica (1644) 58. 40 Notably, ‘hate speech’ counts as protected speech under the First Amendment and cannot be the subject of adverse action by the US Government: Snyder v Phelps, 562 US 443 (2011). 41 To this day, Apple refuses to approve and make available on its App Store apps for either Gab.ai or BitChute, thus effectively preventing those services from being provided in an easily accessible way on phones or tablets.

Limiting Free Speech on the Internet  305

A.  Contractual Arguments Let us assume we are dealing with one C, whose ability to use an online forum maintained by D has been suspended (permanently or temporarily) because of politically-inspired views C has expressed on that forum. C wants advice on whether she can bring a claim for breach of contract against D. Given the international reach of online services, the first issue we would have to deal with in advising C is what law governs C’s contract with D. Table 1 sums up the position, as laid out in the terms of service of the service providers that are the focus of this chapter. Table 1  The law governing the contract between C and D Identity of D

Law of the C–D contract

YouTube

If C resides in EEA or Switzerland, the laws of C’s country; otherwise, California.

Facebook

If C is a business: California; if C is a consumer: law of the country in which C resides.

Twitter

If C resides in USA: California; if C resides outside USA, no express provision other than that D will be an Irish company – so the law of the contract will depend on what the law of C’s country says as to what law will govern a contract between C and D where D is an Irish company.42

With that clarified, we can now turn to the main issue, which is when (if ever) C can claim that D has breached its contract with her by suspending her from D’s online service. D’s first line of argument might be that it is entitled to suspend its customers from using its service for any reason whatsoever.43 On this view, any contract between D and C is unilateral, with C undertaking obligations to D to observe the terms of service in consideration of D’s providing that service, but D undertakes no obligations to C in the opposite direction. While all of Facebook’s, YouTube’s and Twitter’s terms of service fiercely insist that they are assuming no obligations to their users,44 it seems unlikely that any court would accept

42 So if C were residing in England, C’s contract with D would be governed by English law if C were a consumer, and Irish law if C were not: 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) [2008] OJ L177/6. 43 This line of argument was accepted in Twitter v Superior Court ex rel Taylor, A154973 (Cal App Ct, Aug 17 2018) but on the non-contractual (and highly dubious) ground that s 230 of the Communications Decency Act of 1996 (USA) gives a company like Twitter immunity from being sued under Californian law (whatever the cause of action) whenever it exercises a publisher-like function, such as deciding who can use its service. The decision will be disregarded in the discussion that follows, which is concerned to illuminate and illustrate private law’s values, and is therefore only concerned with what private law would make of C’s claim if it were honestly applied to C’s case. 44 See, eg, cl 4.3 of Facebook’s terms of service: ‘To the extent permitted by law, we … Disclaim All Warranties, Whether Express or Implied, Including the Implied Warranties of Merchantability, Fitness for a Particular Purpose, Title and Non-Infringement’ (original emphasis).

306  Nicholas J McBride this analysis of the C–D contract. Given those service providers’ vital commercial interest in attracting users, and the substantial investment of time and effort that such users make (and are encouraged to make) in building up a profile on those providers’ services, a court would likely find that services like Facebook, YouTube and Twitter assume certain obligations to their users in consideration of those users’ using their services. But what obligations? As has just been observed, no obligations will be spelled out in D’s terms of service, so any obligations that D is subject to would have to be implied into the C–D contract.45 Some suggested obligations – all beginning ‘D will not suspend C from using D’s service unless …’ – might be: (1) ‘… C is in breach of the rules governing the supply of that service.’ (2) ‘… C has breached the C–D contract.’ (3) ‘… C is in serious breach of the C–D contract.’ Some such basic obligation might be supplemented by further, procedural obligations, such as: (4) ‘D will not suspend C from using D’s service without first giving C a hearing to determine whether suspension is justified.’ (5) ‘When suspending C from using D’s service, D will supply C with a proper explanation as to whether suspension is justified in C’s case.’ (6) ‘D will act in good faith in determining whether or not to suspend C from using D’s service.’ The courts are most likely to imply term (1) into the C–D contract, as that is the only term that could possibly be implied in fact into the C–D contract, as reflecting C and D’s apparent intentions in entering into that contract. However, such a term is least likely to be of any help to C because – as we have seen – the rules governing the supply of D’s services to C are couched in such a way as to make it very easy for D to say that C is in breach of them if C strays into expressing opinions that are unwelcome within the liberal gnostic worldview. Far more helpful to C is term (2). This is because C can argue that under the general rules of contract law she will not be contractually bound by any rules as to what she can say on D’s service that are vague, unconscionable or contrary to public policy.46 A violation of those rules by C will not therefore amount to a breach of contract and cannot, under (2), justify D’s suspending C. And, obviously, term (3) is helpful to C in the same way that term (2) is, with the added benefit that C could only be suspended if C was in repeated breach of the C–D contract, or C breached the C–D contract in a way that had serious effects for D. 45 Facebook’s terms of service include an entire agreement clause (cl 4.5.1), but the clause (‘These Terms … make up the entire agreement between you and Facebook Inc’) does not seem strong enough to exclude terms from being implied into the contract between Facebook and its users. 46 In the case where the C–D contract is governed by California law, C would invoke California Civil Code, §§1649 (vagueness), 1667 (public policy), 1670.5 (unconscionability).

Limiting Free Speech on the Internet  307 If the C–D contract is governed by Californian law, it seems unlikely that either terms (2) or (3) could be implied into the C–D contract. Under §1655 of the California Civil Code, a term will only be implied into a contract if it is ‘necessary to make a contract reasonable, or conformable to usage … in respect to matters concerning which the contract manifests no contrary intention’. It seems unlikely that a court would find that it is necessary to imply terms (2) or (3) into the C–D contract to make it ‘reasonable’, if we assume that term (1) can already be implied into the C–D contract. Likewise, if the C–D contract is governed by English or Irish law, (2) or (3) would not be implied by law into the C–D contract, as a term implied in law amounts to a term ‘which the law will imply as a necessary incident of a definable category of contractual relationship’.47 While C–D’s relationship might amount to a definable (though relatively novel) category of contractual relationship, it seems unlikely48 that the courts would find there was compelling reason for making terms (2) or (3) (as opposed to term (1)) a necessary incident of a contractual relationship like C and D’s. The courts of any jurisdiction might be more willing to find it ‘reasonable’ to find that procedural terms such as (4), (5) or (6) are necessary incidents of a relationship such as C and D’s. In California, the courts will readily imply in law into any contract a term requiring the parties to the contract to act in good faith, which term ‘imposes upon each contracting party the duty to refrain from doing anything which would render performance of the contract impossible by any act of his own, but also the duty to do everything that the contract presupposes that he will do to accomplish its purpose’.49 While the English courts have been more cautious about imposing on contracting parties obligations to act in good faith, Leggatt J suggested in Yam Seng Pte Ltd v International Trade Corp Ltd50 that the courts should be ready to imply into any contract a term ‘requiring honesty in its performance’,51 and into ‘relational contracts’, which ‘involve a longer term relationship between the parties [to] which they make a substantial commitment’, a term requiring the parties to ‘share information relevant to the performance of the contract’.52 The UK Supreme Court has also held in Braganza v BP Shipping Ltd53 that where a contracting party has been ‘given the power to exercise a discretion,

47 Scally v Southern Health and Social Services Board [1992] 1 AC 294 (HL) 307 (Lord Bridge). 48 Unlikely but not impossible: the courts could take the view that the obligation placed on them by s 6 of the Human Rights Act 1998 to protect individuals’ rights of freedom of speech under Art 10 of the European Convention on Human Rights gave them such compelling reason, particularly where D held effective monopoly control over C’s ability to engage in a certain form of speech. For dicta supporting this position, see Appleby v United Kingdom (2003) 37 EHRR 38 (ECtHR) [47]. However, this point is of limited interest for the current exercise, which is an inquiry into private law’s values. 49 Harm v Frasher, 181 Cal App 2d 405, 417 (Cal App Ct, 1960). 50 Yam Seng Pte Ltd v International Trade Corp Ltd [2013] EWHC 111 (QB). 51 ibid [137]. For discussion of implied duties of good faith, see M Raczynska, ‘Good Faiths and Contract Terms’, ch 5 in this volume; and for exclusion of duties of good faith, see PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6 in this volume. 52 Yam Seng (n 50) [142]. 53 Braganza v BP Shipping Ltd [2015] UKSC 17, [2015] 1 WLR 1661.

308  Nicholas J McBride or to form an opinion as to relevant facts’, the courts will readily imply a term requiring the power-holder to act reasonably in exercising that power, especially where the exercise of that power involves a ‘conflict of interest’ or where there is a ‘significant imbalance of power between the contracting parties’.54 At first sight, the implication of such terms into the C–D contract might be thought to be of considerable assistance to C, as common complaints among people who have suffered the same treatment as C are that: (i) no explanation (other than that ‘You have violated our rules’) is ever given to them as to why they have been suspended, and (ii) service providers like D are patently acting in bad faith in suspending people like C for violating D’s terms of service, as they never suspend liberal gnostics for identical violations.55 However, where C is in admitted breach of D’s terms of service, it is hard to see that the existence of procedural terms such as (4)–(6) would be of any assistance to C. The courts will not find that the C–D contract requires D to hold a hearing or give an explanation in a case where the facts of what C did are clear;56 and C can have no contractual grounds for complaint merely because someone in the same position as C has been treated more generously by D, any more than a debtor can claim to be relieved of a debt just because his creditor has forgiven the debts of everyone else who owed him money.

B.  Non-contractual Arguments In the case where D has suspended C from using an online service maintained by D because of views C expressed on that service, it is unlikely that the law of tort will afford C any more recourse than C will be able to obtain under the law of contract. There are three principal torts C might complain that D has committed in relation to her, in suspending her from using D’s service: (i) conspiracy to cause C loss; (ii) defamation; and (iii) negligence. Whatever State or national law C relied on in order to make out her claim that D committed one or more of these torts in relation to C by suspending her from using its service,57 C would face formidable difficulties in making out her claim.58 54 ibid [18]. 55 An obvious and oft-cited example is that of the (now deceased) actor Peter Fonda, whose Twitter account was not suspended even after tweeting on 20 June 2018 that he hoped Barron Trump would be ripped from Melania Trump’s arms and ‘put in a cage with pedophiles’. More recently, Hollywood producer Jack Morrissey’s putting out a tweet on 19 January 2019 calling for children who wear ‘MAGA’ hats to ‘go screaming, heads first, into the woodchipper’ did not result in his account being suspended. 56 cf Lackey v CBS Radio Inc, 2008 WL 283801 (Cal DC), holding (at 6) that the Supreme Court of California’s decision in Cotran v Rollins Hudig Hall Int’l, 948 P 2d 412 (1998), that an employer is required to give his employee a hearing before firing him for just cause, ‘applies only when the employee disputes the occurrence of the underlying conduct leading to the termination’. 57 If conspiracy, C would likely have to rely on the law of the State/nation in which D (and D’s fellow co-conspirators) were domiciled – presumably California. If defamation or negligence, C would probably be able to rely on the law of the State/nation in which C suffered the harm for which she seeks recourse, which would presumably be the State/nation in which C is domiciled. 58 Should C seek to rely on Californian law, a claim against D for conspiracy to cause C harm would require evidence that D (and/or his co-conspirators) intended to use unlawful means to silence C (see,

Limiting Free Speech on the Internet  309 The most fundamental difficulty is that the kind of loss C has suffered – a limitation on her freedom of speech – is not the kind of loss that these torts, or tort law in general, are designed to protect C from suffering. The tort of conspiracy has only ever been found to apply in cases where a claimant has suffered economic loss at the defendant’s hands,59 and is for that reason categorised as an ‘economic tort’.60 The tort of defamation is obviously only concerned to protect a claimant’s reputational interests – so much so that in England, a claimant whose sole concern is to vindicate those interests must rely on the tort of defamation or drop her claim.61 And while the law of negligence is theoretically open-ended in the range of harms for which it might allow a claimant to recover – as it focuses on the manner in which the claimant was harmed, rather than on the kind of harm suffered by the claimant – the law of negligence has tended not to be applied in England outside the fourfold categorisation of harms to the person, mind, property and bank balance.62 A claimant who has been unreasonably declined service by a defendant who practises a common calling – such as a common innkeeper – will be able to bring a non-contractual63 claim against the defendant for denying her service.64 However, it seems highly unlikely that the courts would designate D as being engaged in a common calling. The law on common callings seems to have developed as a way of

eg, Swipe & Bite, Inc v Chow, 147 F Supp 3d 924 (ND Cal 2015)) – something that would be impossible to make out where D (and its co-conspirators) simply refused to allow C to use its services and did not commit a breach of contract in doing so. A defamation claim would be subject to D’s rights under the First Amendment (as it would not be if D were being sued for breach of contract: Cohen v Cowles Media Co 501 US 663, 668 (1991)). 59 An attempt to extend it to a case where the defendant’s conspiracy resulted in the claimant’s suffering severe emotional distress was rejected in Mbasogo v Logo Ltd [2006] EWCA Civ 1370, [2007] QB 846. The Employment Relations Act 1999 (Blacklists) Regulations 2010 (SI 2010/493) – adopted after the uncovering of a blacklist of over 3,000 trade union members that was maintained among over 40 contractors in the building industry – provide that a defendant who maintains an unlawful blacklist may be required to pay compensation to a blacklisted worker, where the compensation ‘shall be assessed on the same basis as damages for breach of statutory duty and may include compensation for injury to feelings’ (reg 8(2)). The fact that the Government felt the need to pass these Regulations may provide implicit support for the view that the law of conspiracy does not protect non-economic interests. My thanks to Virginia Mantouvalou for bringing the 2010 Regulations to my attention. 60 A categorisation that was endorsed by the UK Supreme Court in JSC BTA Bank v Ablyazov (No 14) [2018] UKSC 19 [6]. 61 Gulf Oil (Great Britain) Ltd v Page [1987] Ch 327 (CA) 334; Lonrho plc v Fayed (No 5) [1993] 1 WLR 1489 (HL). 62 cf J Plunkett, The Duty of Care in Negligence (Hart Publishing 2018) 139–48. 63 In the leading case of Constantine v Imperial Hotels Ltd [1944] KB 693 (KB), a claimant who was unjustifiably turned away from a hotel was simply said by Birkett J to have an ‘action on the case’ against the hotel owner. In P Winfield, The Law of Quasi-Contracts (Sweet & Maxwell 1952), Percy Winfield said that an action under the law on common callings was an action ‘in pseudo-quasi-contract’ (26, 30–34), which shows how difficult it is to categorise. My thanks to Robert Stevens for bringing this point to my attention. 64 See Blackstone, 3 Commentaries on the Laws of England 165 (1753): ‘[I]f an inn-keeper … hangs out a sign and opens his house for travellers, it is an implied engagement to entertain all persons who travel upon that way; and upon this universal assumpsit an action on the case will lie against him for damages if he, without good reason, refuses to admit a traveller.’

310  Nicholas J McBride quelling people’s anxieties that if they travelled to an unfamiliar part of the country, they might be denied service; it was therefore rooted in a public policy in favour of encouraging travel and trade.65 That public policy consideration does not apply in C and D’s case. Moreover, the limits that the law on common callings places on people’s abilities to do what they like with their own property has resulted in that area of law’s being called into question, with Oliver Wendell Holmes observing that ‘an attempt to apply [the law on common callings] generally at the present day would be thought monstrous’.66 As a result, it has been held that under English law,67 the owner of a privately-owned shopping mall is perfectly free to exclude from the mall a campaigner seeking to collect signatures protesting against a development concreting over playing fields,68 or a group of youths who have been hanging around the mall because they have nothing better to do.69 Given this, it is highly unlikely that any court would extend the law on common callings to someone like D.

III.  The Sound of Silence It follows that if C is looking to private law for protection against D’s seeking to prevent C using his services to express her views, she is likely to be disappointed. The reason is not hard to find: private law is simply not interested in protecting freedom of expression. If yet more evidence of this brutal fact were required, it is provided by the history of the First Amendment, which as drafted ‘by James Madison … generated almost no public debate when ratified in 1791’.70 It was assumed that the rule that ‘Congress shall make no law … abridging the freedom of speech, or of the press’ simply gave effect to Blackstone’s understanding of what the ‘liberty of the press’ entails. According to Blackstone, ‘this consists in laying no previous restraints upon publications, and not in freedom from censure for criminal matter when published’.71 It followed for Blackstone, then, that ‘where blasphemous, immoral, treasonable, schismatical, seditious, or scandalous libels, are punished by the English law … the liberty of the press, properly understood, is by no means infringed or violated’.72 65 See NJ McBride, The Humanity of Private Law, Part I: Explanation (Hart Publishing 2018) 79; also DS Bogen, ‘The Innkeeper’s Tale: the Legal Development of a Public Calling’ [1996] Utah Law Review 51. 66 OW Holmes, The Common Law (Macmillan & Co 1882) 203. 67 For discussion, see T Allen, Property and the Human Rights Act 1998 (Hart Publishing 2005) 219–23; K Gray and SF Gray, ‘Civil Rights, Civil Wrongs and Quasi-Public Space’ [1999] European Human Rights Law Review 46. 68 Appleby (n 48). 69 CIN Properties Ltd v Rawlins [1995] 2 EGLR 130 (CA); Anderson v United Kingdom (1998) 25 EHRR CD172 (ECommHR). 70 T Healy, The Great Dissent: How Oliver Wendell Holmes Changed His Mind – And Changed the History of Free Speech in America (Metropolitan Books 2013) 54. 71 Blackstone, 4 Commentaries 152 (1753) (original emphasis). 72 ibid 151 (original emphasis).

Limiting Free Speech on the Internet  311 This understanding of the First Amendment meant that the constitutionality of the Sedition Act 1798 – which criminalised ‘any false, scandalous, and malicious writing … against the [federal] government … or either house of the Congress or the President,73 with intent to defame … or to bring … either of them into contempt or disrepute’ – was not questioned by the courts. Indeed, the view that the First Amendment only invalidated prior restraints on the freedom of speech and not punishments for speech that had been uttered survived until the start of the twentieth century.74 Change only came, first, with Holmes J’s judgment in Schenck v United States, where he tip-toed away from his earlier view that the First Amendment only governed prior restraints on speech75 and formulated a ‘clear and present danger’ standard for what sort of speech could be constitutionally restrained under the First Amendment; and then with Holmes J’s dissent in Abrams v United States, where he made his judgment in Schenck count by ruling that the ‘clear and present’ danger standard was not satisfied in Abrams,76 and provided a clear rationale for the First Amendment’s limiting the state’s ability to interfere with freedom of speech.77 In other words, the precedence given to freedom of speech under the First Amendment in the twentieth century owed absolutely nothing to the American Republic’s inheritance of English law and English legal thinking. It is a thing apart: a flower that first grew in American soil and was not transplanted from England. But why is this? Why does English private law – and the private law of common law jurisdictions generally – not give greater priority to the importance of freedom of speech? The answer will be obvious to those who think that private law is concerned simply to maximise wealth, or to preserve our mutual independence as persons – there is simply no need to protect freedom of speech in order to achieve these goals. However, those of us who adopt a more expansive view of private law have some explaining to do when it comes to the relationship between private law and freedom of speech. Certainly, the lack of importance that private law places on 73 Derogatory statements about the Vice-President were not targeted by the Sedition Act, as the primary objective of the Act was to allow the then President John Adams to lock up supporters of his Vice-President, Thomas Jefferson, in the run-up to the 1800 Presidential election, which was expected to be contested between Adams and Jefferson. See A Lewis, Freedom for the Thought That We Hate (Basic Books 2007) 11–15; and, in much more detail, PI Blumberg, Repressive Jurisprudence in the Early American Republic (CUP 2010) 1–50, 72–145. 74 Patterson v Colorado, 205 US 454, 462 (1907) (Holmes J). 75 Schenck v United States, 249 US 47 (1919) 51–52. 76 Abrams v United States, 250 US 616 (1919) 628. This was something Holmes J was unwilling to do in Schenck itself, even though the ‘clear and present danger’ test was not satisfied in that case either: see Healy, Great Dissent (n 70) 112–14. 77 Essentially the same as that advanced by Milton in his Areopagitica (n 39). See Abrams (n 76) 630: ‘[W]hen men have realized that time has upset many fighting faiths, they may come to believe even more than they believe the very foundations of their own conduct that the ultimate good desired is better reached by free trade in ideas – that the best test of truth is the power of the thought to get itself accepted in the competition of the market, and that truth is the only ground upon which their wishes safely can be carried out.’

312  Nicholas J McBride freedom of speech seems to falsify John Gardner’s thesis that private law is an institution that is responsive to value in general,78 as we have in freedom of speech a value to which private law seems relatively unresponsive. Does the explanation of private law that I offered in The Humanity of Private Law, Part I79 – and which I call ‘F’ – fare any better? Unsurprisingly, my answer is ‘yes’ – in fact, F explains the occasions when certain forms of speech are protected by private law, as well as explaining why private law attaches little importance to freedom of speech of and in itself. And at the same time, if F is correct, then it provides us with a principled basis for thinking that private law should attach far more importance to freedom of speech than it has done in the past. F argues that private law seeks to promote the flourishing of its subjects while maintaining its legitimacy. Private law adopts a particular view of what human flourishing involves, which I call ‘the RP’ – the picture of human flourishing that most people in Western liberal democracies would adopt if they reflected seriously on the issue of what human flourishing involves, not least because it is the picture of human flourishing that such people receive from the culture in which they live through films, advertising, education and so on. RP-flourishing (flourishing according to the RP’s vision of what amounts to human flourishing) involves leading a life in which one enjoys goods such as: (1) health; (2) knowledge; (3) friendship; (4) skilful accomplishment at work and play; (5) marriage; (6) practical reasonableness; (7) a sense of aesthetic appreciation, and harmony with, the universe; (8) identifying with the life one leads; (9) creativity; (10) a ‘desire of the heart’ to see that some significant goal or project is achieved; (11) living in a political community that seeks to foster the flourishing of all of its members.80 It is this kind of life that private law seeks to help us live – and this accounts for those occasions when private life seems to be concerned to protect freedom of expression. On those occasions, it is not freedom of expression that private law is concerned to protect but instead the enjoyment of one of the goods listed above. To take a few examples: (i) communications between married partners are privileged81 because marriage is important to RP-flourishing;82 (ii) the form in which original ideas are expressed is protected by copyright because creativity is important to RP-flourishing;83 (iii) a contractual term that allows A to dictate to B how B leads his life will be invalid84 because such a term is likely to lead B not to identify with the way his life is going; (iv) an employee may be protected (within limits) against 78 See J Gardner, ‘Tort Law and Its Theory’ in J Tasioulas (ed), The Cambridge Companion to Philosophy of Law (forthcoming, CUP 2020); and J Gardner, From Personal Life to Private Law (OUP 2018). 79 McBride, Humanity of Private Law (n 65). 80 The first seven goods are inherited from John Finnis’s account of human flourishing. The final four are my additions: ibid 86–107. 81 Wennhak v Morgan (1888) 20 QBD 635 (QB). 82 Other forms of communication are privileged because giving them such privileged status is important to securing various secondary goods on which someone’s RP-flourishing depends: see McBride, Humanity of Private Law (n 65) 138–40. 83 ibid 148–49. 84 Horwood v Millar’s Timber and Trading Co [1917] 1 KB 305 (CA).

Limiting Free Speech on the Internet  313 being dismissed for expressing dubious opinions85 because having a decent job is important to RP-flourishing. In contrast, freedom of expression per se is not protected by private law, and this is because if we accept the picture of human flourishing that the RP provides us, we shall conclude that ‘With few exceptions people’s interest in … free expression is rather small’86 – one can enjoy an RP-flourishing life without enjoying a large degree of freedom of expression. However, while F allows us to see why private law places a low priority on protecting freedom of expression, F also allows us to see that private law is mistaken to do so. If private law is concerned to promote the human flourishing of its subjects, it has to be sure that the particular conception of human flourishing that it adopts in pursuing that agenda is correct; and each of us has an interest in private law’s getting it right as to what human flourishing involves. Otherwise, there is a danger that we shall be trapped in the equivalent of Robert Nozick’s ‘experience machine’87 – that is, of being subjected to influences that lead us to believe that our flourishing consists in a, b, c, when actually our flourishing consists in x, y, z.88 The only way to be sure that we are not living in the moral equivalent of an experience machine is to allow unrestricted freedom of expression as to what human flourishing involves. So each of us has an interest in living in society where expression relating to the nature of human flourishing – including discussions of the truth or otherwise of liberal gnosticism – suffers no sanction whatsoever. Private law could usefully contribute to our living in such a society by saying that: (i) A will commit a tort in relation to B if A denies, or threatens to deny, B essential goods or services because B adopts (or questions) a particular conception of what human flourishing involves; and (ii) any contractual term that allows A to impose a sanction on B because B adopts (or questions) a particular conception of what human flourishing involves will be invalid on grounds of public policy. Private law does not currently say this, or anything close to it, but our future flourishing depends on its doing so as soon as possible.

85 See V Mantouvalou, ‘“I Lost My Job Over a Facebook Post – Was That Fair?” Discipline and Dismissal for Social Media Activity’ (2019) 35 International Journal of Comparative Labour Law and Industrial Relations 101. 86 J Raz, ‘Free Expression and Personal Identification’ (1991) 11 OJLS 303, 303. 87 R Nozick, Anarchy, State and Utopia (Basic Books 1974) 42–43. 88 In the case of the actual experience machine, ‘a, b, c’ is ‘making good choices within the situations presented to us by the machine’, and ‘x, y, z’ is ‘unplug from the machine’.

314

15 Private Law and Public Concerns: Non-Disclosure Agreements in English Contract Law CATHARINE MacMILLAN*

Non-disclosure agreements are much in the news at present. In England the legal focus has been Sir Philip Green’s action to prevent publication of complaints about his conduct where the complainants had signed non-disclosure agreements.1 Other high-profile cases involving non-disclosure agreements have included Harvey Weinstein and Zelda Perkins,2 and those entered into by young women serving as hostesses at the infamous President’s Club dinner.3 These instances have prompted parliamentary actions; the Women and Equalities Committee’s report on the use of non-disclosure agreements in discrimination cases4 and the Department for Business, Energy and Industrial Strategy’s published response on confidentiality clauses.5 Plans for government legislation to regulate non-disclosure agreements have also been proposed.6 There has, however, been very little consideration of to the contractual force of non-disclosure agreements in English law. This is surprising, given the widespread use of these clauses, and also unsurprising, given that their fundamental * I would like to thank the symposium participants for their comments, and Professor Keith Ewing and Mr Robin Brooks who commented on an earlier draft of this paper. Any remaining errors and omissions are my own. 1 ABC v Telegraph Media Group Ltd [2018] EWCA Civ 2329, [2019] 2 All ER 684. 2 Z Perkins, ‘Harvey Weinstein’s ex-PA: law on sexual harassment at work overdue’ Financial Times (19 June 2019). 3 M Marriage, ‘Men Only: Inside the charity fundraiser where hostesses are put on show’ Financial Times (23 January 2018); M Marriage and Poppy Wood, ‘The Presidents Club investigation: one year on’ Financial Times (25 January 2019). 4 Women and Equalities Committee, The use of non-disclosure agreements in discrimination cases (HC 2017-19, 1720). 5 The Department for Business, Energy & Industrial Strategy, ‘Confidentiality Clauses – Response to the Government consultation on proposals to prevent misuse in situations of workplace harassment or discrimination’ (July 2019), available at https://assets.publishing.service.gov.uk/government/uploads/ system/uploads/attachment_data/file/818324/confidentiality-clause-consultation-govt-response.pdf. 6 The Department for Business, Energy and Industrial Strategy, ‘Crack down on misuse of NonDisclosure Agreements in the workplace’ (21 July 2019), available at www.gov.uk/government/news/ crack-down-on-misuse-of-non-disclosure-agreements-in-the-workplace.

316  Catharine MacMillan purpose is to create silence. This chapter considers the regulation and contractual enforceability of non-disclosure agreements. It addresses the fundamental issue of how private law has an impact upon public values. The argument is made that many non-disclosure agreements have a negative effect not only upon the parties who enter into them, but also upon wider public interests. English contract law, focused as it is upon the rights of individual contracting parties, struggles not only to regulate these negative public interest effects but also to recognise these effects. While many might think the wider public interests outlined in this chapter are best addressed in other arenas, most notably through existing legislation, it is clear that this has proved to be an inadequate response. It is also appropriate, given that the silence brought about by non-disclosure agreements is a silence created by contract law, to examine the ways in which contract law can ameliorate this silence. The chapter begins with an overview of the use and abuse of non-disclosure agreements, before considering the limited extent to which there is statutory regulation over some non-disclosure agreements. This is followed by an examination of how non-disclosure agreements are, and sometimes are not, enforceable by reason of general contractual doctrines. The chapter concludes with an assessment of what can be drawn from these examinations.

I.  Use and Abuse British parties have long included confidentiality clauses in various types of contracts. ‘Non-disclosure’, as a term, has crossed the Atlantic from America. It denotes a wider range of information that is to be kept silent, and has become particularly associated with concealing misdeeds. Non-disclosure agreements do have legitimate and necessary functions in many transactions,7 and they are present in most corporate interactions. They protect, for example, the disclosure of sensitive information necessarily provided to undertake mergers and acquisitions. It will often constitute due diligence to ensure that professional advisers involved in commercial arrangements do not disclose the information they have acquired. Internal investigations seeking out wrongdoing within an organisation may also be subject to non-disclosure agreements. In an employment contract, such agreements articulate what would be implied in any case, and provide clarity and transparency to the employment relationship.8 In settlement agreements these clauses can provide a quick and cheap end to litigation. They may well preserve matters both parties wish not to disclose. They can act to prevent further vexatious litigation in a particular area.9 7 G Stevens and L Subar, ‘Confidentiality in Settlement Agreements Point/Counterpoint’ (2012) 29 GPSolo 24. 8 T Aplin et al Gurry on Breach of Confidence: The protection of confidential information, 2nd edn (OUP 2012) ch 11. 9 G Stevens and L Subar, ‘Confidentiality is a Virtual Necessity’ (2012) 29 GPSolo 28.

Non-Disclosure Agreements  317 The use of non-disclosure agreements has risen rapidly in recent years, and they are now present in almost all commercial contracts. The ability of individuals to rapidly disseminate information over the Internet creates an incentive to control what is disseminated. Non-disclosure agreements provide this control. And it seems probable that once their utility became apparent, their use proliferated. In some instances management styles have developed around the utility of this contractual clause. The suppression of information that might otherwise come into the public domain is cause for concern in a variety of instances. While the recent concern about non-disclosure agreements has focused upon cases of sexual and racist harassment in the workplace, there are other areas where important information has been similarly suppressed. The clauses have been used widely throughout the National Health Service (NHS), to the detriment not only of individuals but also the public.10 The Health Secretary’s plan to discontinue their use11 created cause for alarm rather than comfort, given that his predecessor made exactly the same announcement six years previously.12 Another such use has been by governmental departments in relation to Brexit. It appears that departments have been entering into contractual arrangements under the cloak of non-disclosure agreements, a secrecy imposed by the Government.13 The Public Accounts Committee expressed concern over what it found as the complacency of the Department for Transport in its preparation for a no-deal Brexit.14 The situation was exacerbated by the ‘cloak of non-disclosure agreements, and this secrecy hampers the ability of the business community at large to prepare’.15 Businesses that cooperated with the Department had no option but to sign non-disclosure agreements, about the exact ambit of which the

10 The Mid Staffordshire NHS Foundation Trust Public Inquiry, chaired by Robert Francis QC, ‘Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry’ (HC 947, 2013) available at https://webarchive.nationalarchives.gov.uk/20150407084231/http://www.midstaffspublicinquiry.com/ report. 11 Press Association, ‘NHS non-disclosure agreements to end, vows Matt Hancock’ The Guardian (22 April 2019). 12 Press Association ‘Ban on NHS gagging orders’ The Guardian (14 March 2013). 13 Committee of Public Accounts, Defra’s progress towards Brexit (HC 2017-19 1514) 13, available at https:// publications.parliament.uk/pa/cm201719/cmselect/cmpubacc/1514/1514.pdf. The point is reaffirmed in HL Deb 14 March 2019, vol 796, col 1133, available at https://hansard.parliament.uk/lords/2019-03-14/ debates/CD1AA883-C4E5-45AC-A704-3C7A9305D94A/BrexitNon-DisclosureAgreements. 14 Committee of Public Accounts, Department for Transport’s implementation of Brexit (HC 2017–19, 1657) 6, available at https://publications.parliament.uk/pa/cm201719/cmselect/cmpubacc/1657/1657. pdf. The Treasury provided a similar account of the use of non-disclosure agreements: Treasury, Brexit: Written question – 190947, 13 November 2018, available at www.parliament.uk/business/publications/written-questions-answers-statements/written-question/Commons/2018-11-13/190947/. The Department for Transport was unable to indicate the extent of its use of non-disclosure agreements: Department for Transport, Brexit: Written question – 205939, 7 January 2019, available at www.parliament.uk/business/publications/written-questions-answers-statements/written-question/ Commons/2019-01-07/205939/. 15 Department for Transport, Brexit: Written question (n 14). The Department for Transport could not provide the agreements.

318  Catharine MacMillan Department could not inform the Committee.16 It was clear to the Public Accounts Committee that the use of non-disclosure agreements, coupled with ‘caution’ in public communication about exiting the European Union (EU), meant that the public and businesses could not adequately prepare themselves for the possibility of a no-deal Brexit.17 Similar concerns were raised with regard to the use of nondisclosure clauses by the Department for Food, Environment and Rural Affairs (Defra) in its preparations for Brexit.18 The secrecy set by the Government has continued,19 and has encompassed even the extent to which non-disclosure agreements were employed.20 As Defra and the Department for Transport struggle to prepare for Brexit, it is appropriate to allow individuals sufficient information to prepare themselves privately. The stifling of information about a post-Brexit Britain extends to the panel of experts set up to provide the Department for International Trade with independent perspectives on Britain’s free-trade agreements. These experts have been required to sign nondisclosure agreements, the terms and scope of which are unknown.21 In this context non-disclosure clauses are highly questionable, if not dangerous. They operate to prevent informed public discussions about the future of the United Kingdom. Viewed more broadly, there are indications that non-disclosure agreements are now in widespread use in both the private and public sectors. This includes the House of Commons,22 universities,23 schools,24 local councils,25 the police,26 railway companies27 and the Church of England.28 Prominent organisations that 16 ibid 13. 17 ibid 6. 18 Committee of Public Accounts, Defra’s progress towards Brexit (n 13), 13. The extent of the usage of non-disclosure agreements and their ambit was unclear: 6. 19 Committee of Public Accounts, Brexit and the UK border: further progress review (HC 2017–19, 1942) 12, available at https://publications.parliament.uk/pa/cm201719/cmselect/cmpubacc/1942/1942.pdf. 20 HM Treasury, ‘Treasury: Disclosure of Information: Written question – 203844’ (19 December 2018) https://www.parliament.uk/business/publications/written-questions-answers-statements/writtenquestion/Commons/2018-12-19/203844/. 21 A Isaac and A Tovey, ‘Shhh! Ministers hide trade secrets’ The Daily Telegraph (London, 4 April 2019) 1. 22 R Syal, ‘Gagging clauses cost House of Commons £2.4m in five years’ The Guardian (London, 21 June 2018). 23 H Dixon, ‘MPs hear how universities use NDAs to silence assault victims’, The Daily Telegraph (London, 27 March 27 2019) 11; S Murphy, ‘UK Universities pay out £90m on staff ‘gagging orders’ in past two years’ The Guardian (London, 17 April 2019). 24 J Staufenberg, ‘This is oppression’ Non-disclosure agreements used to muzzle teachers’ The Guardian (London, 6 August 2019) 30. 25 Press Association, ‘Councils use gagging orders in most staff settlements, finds investigation’, The Guardian (London, 3 April 2016). 26 R Brown, ‘This is how much Cambridgeshire police paid out to gag staff ’, Cambridgeshire Live (Cambridge, 17 June 2018) www.cambridge-news.co.uk/news/cambridge-news/police-paid-out-gagstaff-14790774; S Swinford, ‘Police pay departing officers up to £250,000 to sign gagging orders’ The Daily Telegraph (London, 18 May 2013) 19. 27 S Knapton, ‘Woodland Trust condemns rail chiefs for secrecy over new line’ The Daily Telegraph (London, 7 March 2019) 13. 28 J Hardy, ‘Church ‘used NDA to hush up report into sex case vicar’ The Daily Telegraph (London, 6 December 2018) 2.

Non-Disclosure Agreements  319 have used them, and many which may no longer be using them,29 are the civil service30 and the BBC.31 Disturbingly, it appears that charities cooperating with the Government are subject not only to non-disclosure terms, but also to terms prohibiting them from criticising government ministers.32 There are so many non-disclosure terms used in relation to the public sector that it has become difficult to judge how well taxpayers’ money is spent.33 These reports indicate that rather a lot of information is being concealed through the contractual silencing of individuals. In many cases this concealment is adverse to public interests. It has allowed wrongful and dangerous conduct to continue: ‘confidentiality prevents the public from knowing about systemic wrongful conduct’.34 The prevention of public knowledge removes the ability to effect change: one cannot change what one does not know about.

II.  Contract Law and Non-disclosure Agreements There are two legal approaches to curtailing these abusive non-disclosure agreements. The first is through specific legislation; the second is through the application of existing principles and doctrines of contract law. Many might think it best to deal with these issues by the first approach rather than the application of contractual doctrines. While there is force in such an approach, this is removed when one realises that these other means have been ineffectual in regulating non-disclosure agreements. And if these agreements were unenforceable or voidable by reason of contract law, there are two important effects. First, they enable a party to the contract to take action on the contract independent of any legislative restrictions. Second, the contractual unenforceability of these agreements is a disincentive for stronger parties to enter into them.35

29 Evidence to the Women’s Equality Committee’ NDA Inquiry: www.parliament.uk/business/committees/committees-a-z/commons-select/women-and-equalities-committee/inquiries/ parliament-2017/nda-inquiry-17-19/. 30 S Swinford, ‘Huge cost of gagging axed civil servants’ The Daily Telegraph (London, 3 April 2013). 31 J Martinson, ‘Journalist Olenka Frenkiel says BBC sexism and ageism still an issue’ The Guardian (London, 7 November 2014). 32 P Morgan-Bentley, ‘Gagging clauses: Criticism of Theresa May banned in Grenfell safety deal’ The Times (London, 6 November 2018) and ‘Secret State: The government’s use of NDAs exposes the state’s obsession with secrecy’ The Times (London, 7 November 2018). 33 BBC News, ‘Concern over gagging clauses in public sector’ (London, 19 March 2019) www.bbc. co.uk/news/uk-wales-politics-47617728. 34 RL Burdge, ‘Bad for Clients, Bad for Lawyers, Bad for Justice’ (2012) 29 GPSolo 25. 35 Andrew Burrows (A Burrows, ‘Anti-Oral Variation Clauses: Rock-Solid or Rocky?’, ch 3 in this volume, section VI) has, however, observed that parties have found it commercially useful to include exemption clauses that are unenforceable by reason of the Unfair Contract Terms Act 1977 and the Consumer Rights Act 2015. From another perspective, though, if the inclusion of a clearly unenforceable non-disclosure agreement by a lawyer was a breach of professional ethics, as Richard Moorhead has argued (R Moorhead, ‘Professional Ethics and NDAs: Contracts as Lies and Abuse’, ch 16 of this volume, section III), this would act as a disincentive where the contract was drafted by lawyers.

320  Catharine MacMillan

A.  Regulation of Specific Contractual Terms There is specific regulation of non-disclosure agreements in relation to both whistle-blowing and the settlement of employment claims. With regard to whistle-blowing, the Public Interest Disclosure Act 1998 provides a legislative scheme to protect employees who make certain kinds of disclosures. The application of the Act is not without problem. It amends the Employment Rights Act 1996 with section 43J, which provides that a contractual term that purports to preclude a worker from making a protected disclosure is void. The section applies to both employment contracts and, inter alia, any proceedings for breach of contract. Quite apart from the statutory provision, it is unlikely that a court would issue an injunction preventing disclosure to a prescribed regulator.36 While section 43J would seem to preclude the silencing of whistle-blowers with non-disclosure agreements, the evidence indicates that this has not always been the practical effect. First, the provision has not prevented many employers from putting these clauses into contracts.37 Second, some of the settlement agreements with non-disclosure clauses include further clauses in which the worker warrants that they are unaware of any matters that are protected disclosures under the Public Interest Disclosure Act. Richard Moorhead carefully considers the dilemma created for a potential whistleblower by such warranties.38 Whistle-blowers may also be required to agree to non-disparagement clauses. Both these devices work to diminish the efficacy of the legislation. Regulation also occurs in employment situations. Where a non-disclosure clause appears in a settlement agreement in an employment case, for the settlement to be valid the employee must have received independent advice. The independent advice must contain information on the effect of the settlement agreement upon the person’s ability to pursue their claim before the employment tribunal.39 The independent adviser need not be a lawyer,40 and while the advice must be as to the ‘terms and effect of the proposed agreement’,41 this does not require great knowledge of the effect of non-disclosure agreements. Lastly, the Equality Act 2010, section 77, provides that clauses that prevent the voluntary discussion of one’s pay and relevant working conditions are ‘unenforceable’.42



36 In

re A Company [1989] 1 Ch 477. Bowers et al, Whistleblowing Law and Practice, 2nd edn (OUP 2012) 233. 38 Moorhead, ‘Professional Ethics and NDAs’ (n 35) section VI. 39 Employment Rights Act 1996, s 203(3). 40 ibid s 203(3A). 41 ibid s 203(3)(c). 42 T Aplin et al, Gurry on Breach of Confidence, 2nd edn (OUP 2012) 477–78. 37 J

Non-Disclosure Agreements  321

B.  General Contractual Principles i.  Formative Elements General contractual principles of formation present few barriers to effecting nondisclosure agreements. English contract law is largely concerned with enforcing agreements because the obligations were created by the consent of the contracting parties. Little heed is paid in this context to the substantive reality of the consent. In most instances, all of the formative elements necessary to constitute a binding contract will be present. Usually the terms will be entirely drafted by the party seeking to prevent disclosures. Even where this contract is presented with little or no opportunity to scrutinise its terms, the signature of the other party binds them to the terms, regardless of whether they have read or understood them.43 While there has been judicial regret regarding this rule, it has been applied consistently.44 Many might find it odd that the young women engaged as hostesses at the Presidents Club dinner, who were presented with a five-page non-disclosure agreement, had less protection than if they were buying a toaster.45 In this instance consideration may have been a problem, for if the hostesses were already engaged, the consideration was past.46 Consideration is unlikely to be an insurmountable difficulty, though, given that it can be overcome by use of a deed.

ii.  Vitiating Elements a.  Misrepresentation and Mistake It is possible that while the formative elements necessary for a contract are present, an apparent consent is removed by a vitiating element. Actionable misrepresentations might well be established. Where the profferor misrepresents the contract’s terms or their effect, the full extent of the contract cannot be relied upon.47 Even ‘[i]f it conveys a false impression, that is enough’.48 If the situation arises in which the profferee is mistaken about the terms the profferor presents and the profferror, although not acting fraudulently or negligently, is aware of the profferee’s mistake, the contract may be void for a unilateral mistake.49 If the profferee is mistaken as to the commercial or legal effect of the terms, it is unlikely that the contract is void.50 43 L’Estrange v Graucob [1934] 2 KB 394. 44 ibid 405 (Maugham LJ). 45 Consumer Rights Act 2015. 46 Roscorla v Thomas (1842) 3 QB 234; Re McArdle [1951] Ch 669. 47 Curtis v Chemical Cleaning and Dyeing Co [1951] 1 KB 805. 48 ibid 808 (Denning LJ). 49 Smith v Hughes (1870-71) LR 6 QB 597. 50 Clarion Ltd v National Provident Institution [2000] 1 WLR 1888; Brennan v Bolt Burdon [2004] EWCA Civ 1017, [2005] QB 303.

322  Catharine MacMillan b.  Unconscionability and Undue Influence Unconscionability and undue influence are much stronger possible grounds for avoiding a non-disclosure agreement. The existing cases reveal that the weaker party entered into these agreements under a sense of pressure. A consideration of these vitiating elements in the context of non-disclosure agreements reveals, however, the general weaknesses of these doctrines in English law. Equity has the capacity to relieve against harsh and oppressive bargains, but this is a capacity little realised in the modern law of contract. It is a rather curious situation that while many common law jurisdictions have established a form of relief under a doctrine of unconscionability,51 based upon the English case of Aylesford v Morris,52 English law itself has resisted this development. While Lord Denning famously observed that English law gave relief to one who, without independent advice, entered into a contract when his bargaining power was impaired,53 the rejection of this ‘inequality of bargaining power’ by Lord Scarman as an independent head of relief54 worked to largely remove unconscionability as a ground of relief. While Nourse LJ, in Credit Lyonnais Bank Nederland NV v Burch,55 thought it ‘very well arguable’ that the legal charge could have been set aside as an unconscionable bargain,56 the point was made in obiter dicta. At the heart of the decisions was the view that conscience was shocked by a transaction in which a young woman without advice was prevailed upon to charge her only asset without consideration. Burch’s case is distinguishable from the non-disclosure agreements, for these provide consideration and so remove the element of shock. The case, though, points to another possible ground of relief in some nondisclosure agreements: undue influence. We begin with presumed undue influence, or those situations where the evidence is not of an actual undue influence but where the transaction calls for explanation.57 The doctrine is clearly aimed at situations where, in an existing relationship of trust and confidence between the parties, the stronger party has procured the consent of the weaker party by unacceptable means. It is concentrated ‘upon the unfair exploitation by one party of a relationship which gives him ascendancy or influence over another’.58 There are inherent difficulties here in seeking relief for non-disclosure agreements. One is that the complainant needs to establish that they reposed trust and confidence in the other in the management of their financial affairs.59 Another underlying weakness is that 51 D Capper, ‘Undue influence and unconscionability: a rationalisation’ (1998) 114 LQR 479, 479. 52 Aylesford v Morris (1872–73) LR 8 Ch App 484. 53 Lloyds Bank Ltd v Bundy [1975] QB 326, 329. It was hoped English law was entering a period of flexibility: SM Waddams, ‘Unconscionability in Contracts’ (1976) 39 MLR 369. 54 National Westminster Bank plc v Morgan [1985] AC 686 (HL), 707–08. 55 Credit Lyonnais Bank Nederland NV v Burch [1997] 1 All ER 144, (1997) 74 P & CR 384. 56 ibid 151. 57 Noting Lord Clyde’s reservations with this approach: Royal Bank of Scotland v Etridge (No 2) [2001] UKHL 44, [2002] 2 AC 77 [92]. 58 R v A-G for England and Wales [2003] UKPC 22 [22] (Lord Hoffmann). 59 Etridge (No 2) (n 57) [14]. The necessary criteria are set out by Lord Nicholls in [11], [12], [14], [21], [22], [24] and [25].

Non-Disclosure Agreements  323 in instances of presumed undue influence, the presumption is rebutted once the weaker party has received independent legal advice. While it has been said that whether the ‘outside advice had an emancipating effect’ upon any undue influence ‘is a question of fact to be decided having regard to all the evidence’,60 in virtually all instances in which the complainant received independent legal advice the transaction stood.61 It is also the case that, despite statements to the contrary,62 undue influence in modern English law has largely focused upon the those in a spousal or similar situation,63 and in relation to the transfer of land or other property. Non-disclosure agreement cases indicate that there will not be presumed undue influence because: (i) the necessary trust and confidence is not present; and (ii) the weaker party generally has some independent advice. c.  Actual Undue Influence and Duress The equitable doctrine of actual undue influence now overlaps with common law duress.64 Actual undue influence requires some relationship of trust and confidence, and for the reasons given, this is unlikely to exist between parties to non-disclosure agreements.65 Because it appears that many parties sign nondisclosure agreements because pressure has been brought to bear upon them, duress offers a more promising possibility of avoiding the agreement. If a reason one party entered into the non-disclosure agreement was a threat to their physical well-being by the other party, the contract is voidable for duress.66 It is unlikely that duress to the person would be employed, especially where the dominant party has legal representation. But something akin to economic duress is often present. Reports indicate that weaker parties feel compelled to enter into the contract: ‘I was placed in a position with not a great deal of choice but to sign an agreement to silence me.’67 Economic duress is ‘a coercion of will, which vitiates consent. It must be shown that … the contract entered into was not a voluntary act’.68 Duress need only be a reason for contracting and is not removed where the weaker party had other reasons to contract.69 Different causation tests apply to different forms of duress.70

60 ibid [20] (Lord Nicholls). 61 Indeed, courts focus upon the solicitor’s duties: ibid [69]–[74]. 62 ibid [86], relying upon Credit Lyonnais Bank Nederland NV v Burch (n 55). 63 Pesticcio v Huet [2004] EWCA Civ 372, [2004] NPC 55. 64 Etridge (No 2) (n 57) [8]; Holyoake v Candy [2017] EWHC 3397 (Ch) [407]. 65 It is not impossible: AVB v TDD [2014] EWHC 1442 [204]. 66 Barton v Armstrong [1976] AC 104 (PC); North Ocean Shipping Co Ltd v Hyundai Construction Co Ltd [1979] QB 705. 67 ‘Lancaster radiographer wins “first hurdle” of tribunal case against bullying gagging order’, Lancaster Guardian (3 April 2019) available at www.lancasterguardian.co.uk/news/lancaster-radiographerwins-first-hurdle-of-tribunal-case-against-bullying-gagging-order-1-9688951. 68 Pao On v Lau Yiu Long [1980] AC 614, 636 per Lord Scarman. 69 DSND Subsea v Petroleum Geo-Services [2000] BLR 530 (QB); Barton v Armstrong (n 66). 70 Al Nehayan v Kent [2018] EWHC 333 (Comm) [190].

324  Catharine MacMillan The test is higher for economic duress because there is, frequently, a certain amount of pressure in commercial relationships. To establish economic duress, the complainant must establish, first, pressure amounting to compulsion of the will of the victim and, second, that the pressure was illegitimate.71 In establishing this first requirement it is said that the complainant has little choice but to submit.72 The pressure applied need not overbear the will but must be such as to remove practical alternative actions from the complainant.73 Whether or not the complainant protested at the time of contracting is relevant to this determination.74 A challenge to establishing this pressure in relation to a non-disclosure agreement, particularly those in a settlement, is that an alternative course of action is available. A complainant need not settle the agreement but could litigate their claim. What deters them from this alternative is cost, a point considered by Richard Moorhead.75 Litigation is costly, especially when the defendant has deep pockets. A further deterrent, in employment situations, is the fear that a complainant may leave their employment without a reference. In many circumstances there may be no practical alternative but to enter the non-­ disclosure agreement. Courts, however, have often indicated that they cannot rectify the imbalances brought about by external factors.76 With regard to the second requirement, an illegitimate pressure, it is not clear that a refusal to contract at all could ever be an illegitimate pressure.77 Indeed there are Commonwealth authorities that, absent a duty to contract, a party is entitled to set his own terms, even if these appear harsh.78 Where parties have a contractual relationship, the Court of Appeal, in Times Travel v Pakistan International Airlines, has provided new guidance. Illegitimate pressure consists of two elements: the nature of the threat and the nature of the demand.79 The Court held, in relation to the first element, that the threat or commission of a tort or similar wrong, or an offence, would necessarily be treated as illegitimate. The Court left open the possibility that some threats of a breach of contract would not be treated as illegitimate.80 In some instances non-disclosure agreements are proffered on the basis that a pre-existing contract will not be performed unless a non-disclosure 71 R v A-G for England and Wales [2003] UKPC 22 [15] (Lord Hoffmann) (based upon Universe Tankships Inc of Monrovia v International Transport Workers Federation [1982] 2 All ER 67, 88. 72 DSND v Petroleum Geo-Services (n 69) [131]; Times Travel (UK) Ltd v Pakistan International Airlines Corporation [2019] EWCA Civ 828, [2019] 3 WLR 445 [29]. 73 Huyton SA v Peter Cremer GmbH & Co Inc [1998] 1 Lloyd’s Rep 620, 638. 74 DSND v Petroleum Geo-Services (n 69). 75 Moorhead, ‘Professional Ethics and NDAs’ (n 35) section VII. 76 Alec Lobb Ltd v Total Oil [1983] 1 WLR 87; Times Travel (n 72). 77 ‘It is not clear whether a threat not to enter into a contract unless the threatener’s terms are met could ever amount to improper pressure’: H Beale (ed), Chitty on Contracts, 33rd edn (Sweet & Maxwell 2018) 8-048. 78 Morton Construction v City of Hamilton (1961) 31 DLR (2d) 323 (Ont CA); and Smith v William Charlick (1924) 34 CLR 38 (HCA). 79 Times Travel (n 72) [51], based upon Lord Scarman’s decision in Dimskal Shipping Co SA v International Transport Workers Federation (The Evia Luck) (No 2) [1992] 2 AC 152. 80 Times Travel (n 72) [51].

Non-Disclosure Agreements  325 agreement is entered into. In such cases of a threatened breach of contract, the second element of illegitimate pressure, the nature of the demand, must then be considered. There is authority that courts will view a threat of contractual nonperformance as illegitimate if the threat is made in bad faith.81 It should constitute bad faith where the purpose of the threatened breach of contract is an attempt to conceal poor behaviour either at or somewhat below a criminal standard. Times Travel addresses whether or not the threat of a lawful act can constitute illegitimate pressure. In most instances involving non-disclosure agreements the threat will be of a lawful act. So, for example, the dominant party might threaten not to contract at all without a non-disclosure clause. Or the dominant party might threaten not to settle a claim without a non-disclosure clause. Where the act, or threatened act, is lawful, the court turns directly to the second element of illegitimacy, the nature of the demand.82 If the lawful pressure exerted by one party (A) upon another (B) is used to compel B to concede to a demand to which A ‘does not bona fide believe itself to be entitled’, B’s agreement is voidable for economic duress.83 But if A genuinely believes himself to be entitled to the result he seeks, this will not constitute duress even where there are no reasonable grounds, determined on an objective basis, for the belief.84 This leaves a very narrow ambit to lawful act duress. The decision is based on a concern to allow a party to use ‘lawful commercial pressure in support of a purely commercial demand’.85 A mere inequality of bargaining power does not change the situation. An underlying premise in the decision, though, is that weaker parties, such as consumers, will be specifically protected by legislation.86 The result in the case of non-disclosure agreements is that unless there is a specific legislative protection, the lawful demand will not constitute duress unless the stronger party does not genuinely believe himself to be entitled to achieve his result. While this may work well in a commercial agreement, the requirement of bad faith will tend to prevent duress as an effective defence in most non-disclosure agreements.

iii. Illegality The doctrine of illegality presents a more formidable challenge to a non-disclosure agreement than attempts to avoid it by reason of a vitiating element. Illegality renders a contract or a contractual term unenforceable but the doctrine involves

81 ibid [131]; Adam Opel GmbH v Mitras Automotive UK Ltd [2007] EWHC 3205 (QB). P Birks, An Introduction to the Law of Restitution (Clarendon 1985) 183; and P Birks, ‘The Travails of Duress’ [1990] LMCLQ 342, 347. 82 Times Travel (n 72) [52]. 83 ibid [62] (David Richards LJ). 84 ibid [70], [105]. Reasonability has a wide ambit: R v A-G (n 71). 85 Times Travel (n 72) [105] (David Richards LJ). 86 ibid [41].

326  Catharine MacMillan many variables that make the formulation of rules problematic. These difficulties arise because the values involved are those of both private and public law.87 In finding a contract tainted by illegality, courts seek to protect the integrity of the legal system. Lord Toulson, in Patel v Mirza, explained this point and provided new multi-factorial guidance: [O]ne cannot judge whether allowing a claim which is in some way tainted by illegality would be contrary to the public interest, because it would be harmful to the integrity of the legal system, without (a) considering the underlying purpose of the prohibition which has been transgressed, (b) considering conversely any other relevant public policies which may be rendered ineffective or less effective by denial of the claim, and (c) keeping in mind the possibility of overkill unless the law is applied with a due sense of proportionality. We are, after all, in the area of public policy. That trio of necessary considerations can be found in the case law.88

O’Sullivan, commenting upon this approach, observed that contractual enforcement will be rarer than remedial responses to illegality.89 Illegality could render a non-disclosure agreement unenforceable in certain circumstances. First, the non-disclosure agreement itself may be illegal as formed, or an illegal performance may be intended. In such clear cases there is little need for a detailed scrutiny of Lord Toulson’s factors.90 This situation might arise in a number of ways. The agreement itself might involve the common law offence of perverting the course of justice or a conspiracy to pervert the course of justice.91 The offence requires a positive act and is committed when one acts or embarks on a course of conduct that tends, and is intended, to pervert the course of justice.92 An agreement might attempt, for example, to persuade or intimidate a witness not to give evidence, or to alter her evidence or give false evidence. The agreement might otherwise seek to hide from the police the commission of a serious crime, such as sexual assault, or to delay or prevent the arrest of a wanted person for a serious crime. This is most likely to occur when the non-disclosure agreement purports to prevent an individual from speaking to the police or giving evidence at a trial. If the non-disclosure agreement did pervert or otherwise obstruct the course of justice it would be illegal – indeed the ‘paradigm case … [of] … a criminal offence’.93 The result is that the contract is void and any rights derived from it non-existent.94 The same result would prevail even if only one party intended to

87 A Burrows, ‘Illegality after Patel v Mirza’ (2017) 70 CLP 55, 56. 88 Patel v Mirza [2016] UKSC 42, [2016] 3 WLR 399 [101]. 89 J O’Sullivan, ‘Illegality and Contractual Enforcement after Patel v Mirza’ in S Green and A Bogg (eds), Illegality after Patel v Mirza (Hart Publishing 2018) 172. 90 Beale (ed), Chitty on Contracts (n 77) 16-015. 91 Considered in detail by Moorhead, ‘Professional Ethics and NDAs’ (n 35) section II. 92 R v Vreones [1891] 1 QB 360. It does not matter that the conduct was performed before the matter was investigated or discovered: R v Rafique [1993] QB 843. 93 Les Laboratoires Servier v Apotex Inc [2014] UKSC 55, [2015] AC 430 [23] (Lord Sumption). 94 ibid. As Lord Sumption explained, ‘the ex turpi causa principle precludes the judge from performing his ordinary adjudicative function in a case where that would lend the authority of the state to the

Non-Disclosure Agreements  327 perform the contract in a way that perverted the course of justice.95 These would be very egregious non-disclosure agreements, and it is unlikely any professional lawyer would be involved with such an agreement. It is also an indictable offence under section 5 of the Criminal Law Act 1967 for one who knows or believes that a relevant offence has been committed, and who has information that might be of material assistance in securing the prosecution or conviction of any offender for it, to accept consideration for not disclosing the information. Any such agreement, which is criminal by virtue of section 5, is also unenforceable as a civil contract. This statutory section presents complications. One is that it appears on the wording of the section96 that it is acceptable for a victim to settle their civil claim for compensation.97 However, as Buckley observes, it is improbable that the section allows the stifling of a prosecution in exchange for such compensation.98 The section probably does not remove the bar on contractual enforceability.99 The question arises, though, as to when a compromise effected for compensation is enforceable and when it is not. It has been argued, in light of the historical development of the criminal law, that such a contract should be enforceable unless there is some ground of public interest for avoiding it.100 Factors that might tell against the enforceability of the agreement are matters such as an offence that is of a ‘public character’ rather than a ‘private character’,101 or that the agreement has been procured by improper pressure.102 As is argued in the following subsection, in many instances there will be public policy reasons for avoiding them.

iv.  Contrary to Public Policy Some non-disclosure agreements may well be contrary to public policy so that a court will refuse their enforcement. We begin with those grounds that are well recognised in English law as contrary to public policy,103 and then proceed to consider whether or not these can be expanded, particularly in light of the new approach advanced in Patel v Mirza.

enforcement of an illegal transaction or to the determination of the consequences of an illegal act’: ibid. See also Bilta (UK) Ltd v Nazir (No 2) [2015] UKSC 23, [2016] AC 1. 95 ParkingEye Ltd v Somerfield Stores Ltd [2012] EWCA 1338, [2013] QB 840 [33]. 96 Criminal Law Act 1967, s 5, ‘any consideration other than the making good of loss or injury caused by the offence, or the making of reasonable compensation for that loss or injury’. 97 Beale (ed), Chitty on Contracts (n 77) 16-065. 98 RA Buckley, Illegality and Public Policy, 3rd edn (Sweet & Maxwell, 2013) 131. 99 ibid. 100 AH Hudson, ‘Contractual Compromises of Criminal Liability’ (1980) 43 MLR 532, 542. See, also, E Peel (ed), Treitel The Law of Contract, 14th edn (Sweet & Maxwell 2015) 557; and Beale (ed), Chitty on Contracts (n 77) 16-065. 101 Hudson, ‘Contractual Compromises’ (n 100) 540. 102 ibid 541. 103 Beale (ed), Chitty on Contracts (n 77) 16-008 identifies five groups.

328  Catharine MacMillan Contracts prejudicial to the administration of justice have long been recognised as illegal, being contrary to public policy.104 In addition, English law will not prevent the publication of material of such a nature that public interest requires its disclosure, for confidence does not prevent the disclosure of iniquity.105 Where non-disclosure agreements are designed to conceal harassment falling short of a crime, or activities that are of a whistle-blowing nature, it can be argued that these seek to cloak with confidence an iniquity, and that there are good public reasons to disclose the information.106 So, for example, contracts that prevent or seek to prevent a party from giving information to third parties to secure convictions for fraud or prevent the commission of frauds, have been held to be unenforceable.107 Applying the multi-factorial approach of Lord Toulson, the disclosure acts to further the public interest in the information, a common underlying factor that allowed these courts to overcome a confidentiality prohibition. That the information concealed a form of wrong strengthens the reason for the disclosure. Patel v Mirza also requires a further balancing with the second of Lord Toulson’s criteria, namely, whether there are other public policies rendered less effective or ineffective as a result. This may well act to leave some non-disclosure agreements enforceable. The rights of the defendant under the Human Rights Act 1998 establish the countervailing public policy in many of these cases. Should public policy grounds be expanded in relation to non-disclosure agreements to include the contravention of fundamental human rights? The argument made here is that this extension is now appropriate. English contract law tends to a form of conservatism, sometimes an extreme form, and public policy arguments are viewed with scepticism. Courts, we are told, cannot ‘invent a new head of public policy’,108 and public policy has been described as ‘a very unruly horse, and when once you get astride it you never know where it will carry you. It may lead you from the sound law.’109 Courts have also been concerned not to usurp the role of the legislature in rendering agreements unenforceable by reason of human rights.110 This scepticism and detachment is no longer justified in the twenty-first century. It is accepted that in relation to illegality novel applications exist,111 and that ‘public policy is a variable notion, depending on changing manners, morals and economic conditions’.112 While earlier generations were concerned with 104 Buckley, Illegality and Public Policy (n 98) ch 8. 105 Lion Laboratories v Evans [1985] QB 526, 550; Fraser v Evans [1968] 1 QB 349, 362; Gartside v Outram (1856) 26 LJ Ch 113, 114. 106 Hubbard v Vosper [1972] 2 QB 84; Lion Laboratories Ltd v Evans (n 105); Attorney-General v Guardian Newspapers Ltd [1992] 1 WLR 874. 107 Howard v Odhams Press, Limited [1938] 1 KB 1, 42 (Greene LJ), ‘[t]he promise of secrecy is void as being against public policy’. 108 Janson v Driefontein Consolidated Mines Ltd [1902] AC 484, 491 (Lord Halsbury). 109 Richardson v Mellish (1824) 2 Bing 229, 252 (Burroughs J). 110 Cheall v APEX [1983] 2 AC 180, 191–92. See also Buckley, Illegality and Public Policy (n 98) 87–93, 100–04. 111 Peel (ed), Treitel (n 100) 550. 112 ibid 549; Maxim Nordenfelt Guns and Ammunition Co v Nordenfelt [1893] 1 Ch 630, 666.

Non-Disclosure Agreements  329 sexual immorality as contravening public policy, a more modern concern is the protection of human rights. English judges are now used to the consideration of public policy issues in matters such as tort law, the application of human rights and EU law. It is important to remember that a major reason courts have been reluctant to find contracts unenforceable on public policy grounds is that to do so interferes with freedom of contract: [I]f there is one thing which more than another public policy requires, it is that men of full age and competent understanding shall have the utmost liberty of contracting, and that their contracts when entered into freely and voluntarily shall be held sacred and shall be enforced by courts of justice.113

This reason ought not to be a barrier to most non-disclosure agreements, given that these are not really voluntarily agreements freely entered into by the weaker party. The real challenge is how to incorporate human rights issues into private law. There are justifiable concerns in apparently imposing liberal standards intended for states upon individuals. It can also be argued that the consideration of human rights in private law takes judges dangerously out of the judicial function and into a legislative one, a change of function they may not be best placed to make. Human rights, though, are not incompatible with private law. Courts have now begun to fashion a right to privacy through breach of confidence actions decided with reference to Article 8 of the European Convention on Human Rights (ECHR).114 As Professor Collins has observed: Private law is surely not opposed to the values and principles embodied in the terse statements of human or fundamental rights found in constitutions, conventions and charters. On the contrary, many of those values influenced the doctrines of private law. The individual right to liberty may be discovered at the core of such key ideas of private law as freedom of contract …115

In this context it is reasonable to accept that certain human rights form essential norms within our society: contractual terms that erode these essential norms are contrary to public policy. In determining whether or not such contractual terms are contrary to public policy courts are not usurping a legislative function, neither do they act with an unrestricted discretion. Lord Toulson’s approach in Patel v Mirza establishes a judicial process to be undertaken. This approach is adopted here in considering the possible impact of public policy arguments in relation to three sample forms of non-disclosure agreements. These are: (i) the non-disclosure 113 Printing and Numerical Registering Co v Sampson (1875) LR 19 Eq 462, 465 (Jessel MR). 114 P Giliker, ‘A Common Law Tort of Privacy – The Challenges of Developing a Human Rights Tort’ (2015) 27 Singapore Academy of Law Journal 761; and G Phillipson, ‘Transforming Breach of Confidence? Toward a Common Law Right of Privacy under the Human Rights Act’ (2003) 66 MLR 726. 115 H Collins, ‘On the (In)compatibility of Human Rights Discourse and Private Law’ in HW Micklitz (ed), Constitutionalization of European Private Law (OUP 2014) 27.

330  Catharine MacMillan agreement that seeks to silence an employee from making misconduct complaints; (ii) whistle-blowers; and (iii) government operations and policy formation. The non-disclosure agreement that seeks to silence an employee who has been subjected to sexist or racist harassment clearly works against basic human rights. The preamble to the Universal Declaration of Human Rights provides, ‘whereas recognition of the inherent dignity and of the equal and inalienable rights of all members of the human family is the foundation of freedom, justice and peace in the world’.116 By Article 1, ‘all human beings are born free and equal in dignity and rights’. It is not the argument here that these elements form a part of English law but that they constitute an underlying public policy of equality, dignity and inalienable rights. To these we can add provisions that do have the force of law, namely Articles  10 (freedom of expression), 14 (prohibition of discrimination) and 17 (prohibition of abuse of rights) of the ECHR. There are also the policies underlying the Equality Act 2010. These instruments, and their acceptance within British society, leads to the reasonable formulation of a public policy that accepts that people are equal and hold inalienable rights. These include: the right of expression through the free communication of information; the right not to be discriminated against on the basis of grounds such as sex or race; and that abuse of these rights is prohibited. If we consider this public policy in relation to the three considerations set out by Lord Toulson, the reason for the prohibition inherent in these public policy statements is that these are the rights associated with human dignity that underpin a democratic society. The rights are not absolute, and the second relevant consideration requires an assessment of ‘any relevant public policies which may be rendered ineffective or less effective’117 by denying the claim. It is entirely appropriate for the court to examine the rights accorded to the profferor of the non-disclosure agreement. In considering whether or not to issue an injunction prohibiting a third party to the non-disclosure agreements from publishing details of the ‘silenced’ behaviour, the Court of Appeal in ABC v Telegraph Media identified section 12 of the Human Rights Act 1998, Article  10 of the ECHR and a body of case law in which restrictions upon publication had been imposed.118 This evaluative process is equally applicable to establishing the countervailing claims of refusing to enforce an agreement on public policy grounds. The final consideration, of proportionality, is also to be undertaken.119 This approach allows an incremental and evolutionary change to English contract law. It is not a sudden departure upon an unruly horse to an unknown destination. Indeed, it supports broader considerations of the importance of justice

116 The Declaration is enshrined within EU law by the Court of Justice and underpins the Charter of Fundamental Rights of the European Union. 117 Patel v Mirza (n 88). 118 ABC v Telegraph Media (n 1) [7]–[21]. 119 Arguably the proportionality argument is not necessary as it has been adequately addressed in the first two considerations: E Lim, ‘Ex Turpi Causa: Reformation not Revolution’ (2017) 80 MLR 927, 937.

Non-Disclosure Agreements  331 in society. If we allow the rich and powerful to silence those they have abused, it is publicly destructive of justice and of the legal system entrusted with its administration. The words of one of Green’s alleged victims appeared on the front page of a national newspaper: I would like nothing more than to speak freely about what Philip Green did to me, but I can’t tell you what happened. I was paid for my silence and I have kept it, not just because of the NDA that I was forced to sign, but because of the fear … Green is like a general hiding behind an entire army. His armour is made of wealth, power, and influence. It’s how men like him get away with behaviour which is unacceptable in normal society. … Their sphere of influence extends far and wide. Those who have stood up to these people can be left feeling utterly victimised and helpless. … Some like Green have the power and influence to stop you working ever again if they choose.120

These front-page allegations extended to racism: ‘Sir Philip is also alleged to have racially abused a senior black employee, telling him that his “problem” was that he was still “throwing spears in the jungle”.’121 These well-reported allegations give the public reason to believe that fundamental laws concerned with equality are seriously eroded and that the legal system allows wealthy and powerful individuals to ‘buy’ a different treatment beyond the law. This also presents considerations beyond those related to the parties to a particular non-disclosure agreement. Allowing an abuser to silence his victims allows him undertake a repeated and spiralising cycle of abuse.122 Later victims need not be victims if their potential abuser is recognised as such. In many organisations the operational efficiency and the reputation of the organisation will be endangered by the presence of such an abuser. In the second example, there is an obvious benefit to allowing whistle-blowers to speak up, particularly in public organisations. There is an oft-repeated pattern of severe failures within the NHS in situations where whistle-blowers have been overlooked or harassed.123 While whistle-blowers do receive some protection under the law, it is clear that it is not always sufficient in the face of a non-­disclosure agreement. Gary Walker, the former chief executive of United Lincolnshire Hospitals Trust, is one example. Walker was fired in 2010, allegedly for gross misconduct but as he recounted, for failing to meet government targets. 120 ‘The terror of being pursued is still enough to confine me to silence’, The Daily Telegraph (London, 9 February 2019) 1. 121 ibid. See also the Women and Equalities Committee, The use of non-disclosure agreements report (n 4) [68]. 122 DA Hoffman and E Lampmann, ‘Hushing Contracts’, (2019) Research Paper No 19-07, University of Pennsylvania Institute for Law and Economics, 5–6, available at https://ssrn.com/abstract=3328569. 123 R Mannion et al, ‘Understanding the knowledge gaps in whistleblowing and speaking up in health care: narrative reviews of the research literature and formal inquiries, a legal analysis and stakeholder interviews’ (2018) Health Services and Delivery Research 6 (30) 33–51, available at http://nrl.northumbria.ac.uk/38183/1/Mannion%20et%20al%20-%20Understanding%20the%20knowledge%20gaps%20 in%20whistleblowing%20and%20speaking%20up%20in%20health%20care.pdf. See also Alexander’s Excavations, ‘Mostly whistleblowing, NHS underbelly but other stuff too!’, available at https://minhalexander.com/.

332  Catharine MacMillan His claim was settled by a contract with a non-disclosure agreement.124 His case, and the non-disclosure agreement itself, became public when he broke its terms to speak to the BBC following the Mid Staffordshire enquiry. He then gave evidence to the Health Select Committee. Walker’s story makes for uncomfortable reading, as he reveals a pattern of harassment and bullying from a more senior NHS figure as he attempted to express concerns about patient safety in light of unrealistic government targets. Walker’s settlement contained non-disclosure clauses broad enough to attempt to prevent him from whistle-blowing. He stated that he knew of hundreds of other such settlements, many made on the same basis. In the resulting furore over this disclosure, the Health Secretary announced that such ‘gagging’ clauses would no longer be used within the NHS.125 More than half a decade on, though, it is apparent that there has never been any legal ban within the NHS and that the clauses still appear to be used.126 There are good reasons for finding such non-disclosure agreements unenforceable on public policy grounds. A challenge in this instance, though, is that they should already be void as a result of the Public Interest Disclosure Act 1998. Can such an agreement be unenforceable at common law on the basis of public policy if it is already invalid by statute? The statutory prohibition, however, reinforces the public policy grounds rather than diminishing or supplanting them. This is particularly true where there are noted weaknesses in the Public Interest Disclosure Act127 and where it is possible, using other contractual clauses such as those preventing derogatory remarks or penalties, to evade the Act altogether. The purpose of the prohibition in the Act was to prevent a party from stifling both the information the whistle-blower released and the whistle-blower themself. The information available indicates that there are NHS managers who act in a bullying and harassing manner, confident that their actions will not be uncovered and criticised because they are able to effectively silence those employees who have been harassed. That these employees, like Walker, are attempting to uphold patient safety makes the managers’ actions all the more reprehensible. The availability to these managers of a binding non-disclosure agreement effectively condones their actions. This causes a second result, namely, that the stifling 124 Walker’s written evidence, available at www.parliament.uk/documents/commons-committees/ Health/FRA03GaryWalkerA-L.pdf. Walker was blacklisted and never worked in the NHS again: see at www. pressgazette.co.uk/hospital-chief-who-blew-whistle-patient-safety-bbc-blacklisted-working-nhs/. 125 Press Association ‘Ban on NHS Gagging Orders’ The Guardian (14 March 2013), available at www. theguardian.com/society/2013/mar/14/ban-on-nhs-gagging-orders. 126 S Knapton, ‘NHS uses gagging orders to silence staff who raise bullying and harassment complaints’ The Telegraph (24 February 2019) available at www.telegraph.co.uk/news/2019/02/24/ nhs-uses-gagging-orders-silence-staff-raise-bullying-harassment/; Health Estates & Facilities Management Association ‘Hunt Bans NHS Gagging clauses (14 March 2019); www.hefma.co.uk/about-us/ news/item/hunt-bans-nhs-gagging-clauses. 127 Women and Equalities Committee, The use of non-disclosure agreements (n 4) [79]–[86]; J Ashton, ‘15 Years of Whistleblowing Protection under the Public Interest Disclosure Act 1998: Are We Still Shooting the Messenger?’ (2015) 44 Industrial Law Journal 29; D Lewis, ‘Ten Years of Public Interest Disclosure Act 1998 Claims: What Can We Learn from the Statistics and Recent Research?’ (2010) 39 Industrial Law Journal 325; Mid Staffordshire NHS Public Inquiry, ‘Report’ (n 10) vol 1, 234–42.

Non-Disclosure Agreements  333 of whistle-blowers prevents the massive and widespread ill-treatment of NHS patients from being exposed. Inquiries into the Mid Staffordshire case established that hundreds of people died as a result of ill-treatment not exposed by putative whistleblowers. In considering whether public policy renders non-disclosure agreements unenforceable, courts do need to balance countervailing interests such as the necessary confidentiality of the organisation or patient information. In establishing whether or not it is a proportionate response, a court could consider that disclosure of the whistle-blower’s information, particularly to the press, allows the development of public political pressure to change the management conditions. It is also proportionate because, in the most egregious situations, the public have a right to know of matters in a public health service that put patient safety at serious risk. The last of our sample non-disclosure agreements, those used by the Government when contracting with private companies for the Government’s Brexit preparations, is more difficult. What would form the public policy basis for finding these clauses unenforceable? The objectionable element to these non-disclosure clauses is that they prevent the operation of transparent and open government. People have a right to know what their government does, particularly in areas where it appears that the government is concealing from citizens the serious consequences of government actions. In the case of Brexit, there have been concerted attempts by the Government to minimise the disadvantages from this (avoidable) political action. By suppressing all information in its contractual actions, the Government removes from the public critical information necessary to determine the consequences of Brexit at a point where this course of action could be avoided or where individuals could prepare themselves for these consequences. A public policy supporting this goal of open and transparent government is found in a number of different places. One is the Ministerial Code, which requires holders of public office to act and take decisions in an open and transparent manner. Information is not to be withheld from the public, ‘unless there are clear and lawful reasons for doing so’.128 Ministers should be as open with the public as possible, refusing to provide information only when to do so would not be in the public interest in accordance with statutes and the Freedom of Information Act 2000.129 A second ground to support this public policy lies in the Freedom of Information Act itself. While it might be thought that the Act is the best means of achieving open government, consistent failures to provide requested information greatly diminishes its efficacy. The Institute for Government has stated that during this Brexit period, fewer Freedom of Information Act requests are being met than ever before,130 and there are indications that government departments 128 The Seven Principles of Public Life, Annex A, Ministerial Code (January 2018), available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/ file/672633/2018-01-08_MINISTERIAL_CODE_JANUARY_2018__FINAL___3_.pdf. 129 ibid 1, 1.3 d. 130 Institute for Government, ‘Communication and Transparency’ (Whitehall Monitor, January 2019), available at www.instituteforgovernment.org.uk/publication/whitehall-monitor-2019/communication-andtransparency; O Wright, ‘Ministers “Abuse powers to hide Brexit truths”’ The Times (28 December 2018).

334  Catharine MacMillan have a coordinated policy to refuse to meet these requests regarding Brexit.131 The result is that a profound constitutional and economic change is being undertaken in a manner that deprives the public of critical information.132 If the non-disclosure clauses in supply and other contracts were unenforceable, information about this contracting could be made public. In balancing such a decision, factors such as national security could be considered.133 The arguments as to proportionality are more difficult to consider than in the NHS whistle-blowing scenario, because it is harder to evaluate whether public information and resulting pressure are the best means of assessing government action. The real difficulty in tackling these sorts of non-disclosure agreements in government contracts is a combination of the legal and the practical. Many private parties contracting with the Government may have no particular objections to a non-disclosure term; many may well welcome such terms, as they prevent disclosure to competitors. The offensiveness of the non-disclosure arises in relation to a public harm or a perceived public harm. The concern arises with members of the public who are, of course, third parties to the contract and to whom privity denies any standing to challenge such a contract. The result is that while these nondisclosure terms may be unenforceable, litigation will never arise. In concluding this consideration of how non-disclosure agreements are unenforceable, by reason of illegality or public policy, a formidable obstacle to this process must be realised. If a non-disclosure agreement were unenforceable, one would expect restitution of the property that would otherwise have passed. As Lord Toulson observed, ‘a person who satisfied the ordinary requirements of a claim in unjust enrichment will not prima facie be debarred from recovering money paid or property transferred by reason of the fact that the consideration which has failed was an unlawful consideration’.134 This is a considerable problem, because few parties will come forward to challenge their non-disclosure agreement if they believe it will cause them to lose the monetary advantage they have gained. How could such a problem be overcome? It may well be that the party who received payment has so altered their position as to provide a defence to an unjust enrichment claim. It may also be that the party has provided ample consideration beyond that within the unenforceable non-disclosure clause. In such a case a substantial contract remains.135 Careful drafting of a non-disclosure agreement could, however, prevent the preservation of a partial consideration.

131 T Barnes, ‘Brexit: Government advised public bodies how to refuse FOI requests around plans for no-deal’ The Independent (London, 25 January 2019), available at www.independent.co.uk/news/uk/ politics/brexit-no-deal-plans-government-foi-requests-public-bodies-a8746681.html. 132 A Cheung, ‘Freedom of Information: Secrecy continues at DExEU and Cabinet Office’ (21 September 2017), available at www.instituteforgovernment.org.uk/blog/freedom-of-informationdexeu-cabinet-office-foi. 133 Exemptions within the Freedom of Information Act in ss 21–40 are relevant in this evaluative process. 134 Patel v Mirza (n 88) [116]. 135 Goodinson v Goodinson [1954] 2 QB 118.

Non-Disclosure Agreements  335 Lord Neuberger, writing in the minority in Patel v Mirza, recognised that in some instances the restitution would not always follow on the unenforceability of the contract for illegality.136 Two such instances137 were: (i) where the legislation was designed to protect one of the parties to the contract; and (ii) where the defendant was unaware of the facts giving rise to the illegality. A relevant factor is that the weaker party was probably subjected to some pressure in entering into the nondisclosure agreement. Nevertheless, the uncertainty surrounding the question of restitution where the contract is unenforceable presents a barrier to a challenge on the basis of illegality or public policy.

v. Penalties One final point about non-disclosure agreements must be considered. The force of many non-disclosure agreements, particularly those within a settlement, lies in the contractual consequences that arise if the agreement is breached. These agreements typically require the repayment of all of the sums paid along with certain other amounts.138 This raises the question of whether the specified sum is recoverable as liquidated damages or irrecoverable as an unenforceable penalty.139 This question, one of law, has been clarified by the Supreme Court in Square Holdings BV v Makdessi; ParkingEye Ltd v Beavis.140 It is a matter of construction as to whether the provisions are a penalty at the time the contract is agreed.141 In this construction, ‘the real question when a contractual provision is challenged as a penalty is whether it is penal, not whether it is a pre-estimate of loss’.142 The clear effect of the repayment terms for breach of the non-disclosure clauses is to compel continued silence. A part of the test for enforceability in Cavendish v Makdessi is whether the means by which the contracting party’s conduct is to be influenced is unconscionable.143 In this determination the circumstances in which the contract was made are not entirely irrelevant.144 The Supreme Court found that in circumstances where two parties of equal bargaining power had legal advice, this indicated that the clause was more likely not to be penal. In contrast, the clauses considered in this chapter involve a weaker party who enters a contract that might survive a scrutiny for duress but which clearly exhibits a strong compulsion. 136 Patel v Mirza (n 88) [161]. 137 ibid [162]. 138 See Walker’s non-disclosure agreement at cl 5, 1.5, Walker’s written evidence (n 124) Annex A and that between Green and a former employee: Arcadia Group Ltd v Telegraph Media Group Ltd [2019] EWHC 223 (QB) Appendix. 139 For discussion of such clauses generally, see W Day, ‘Disproportionate Penalties in Commercial Contracts’, ch 11 in this volume. 140 Cavendish Square Holdings BV v Makdessi; ParkingEye Ltd v Beavis [2015] UKSC 67, [2016] AC 1172. 141 ibid [9] (Lord Neuberger and Lord Sumption). 142 ibid [31] (Lord Neuberger and Lord Sumption). 143 ibid. 144 ibid [35] (Lord Neuberger and Lord Sumption).

336  Catharine MacMillan To determine if the clause is a penalty, the following test is applied: The true test is whether the impugned provision is a secondary obligation which imposes a detriment on the contract-breaker out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation. The innocent party can have no proper interest in simply punishing the defaulter. His interest is in performance or in some appropriate alternative to performance.145

Chitty on Contracts identifies a number of instances that have constituted legitimate interests in performance deserving of protection.146 It is hard to argue that any of these instances are applicable in the situation of the non-disclosure agreements considered here. Indeed, it would be better argued that since the protection purports to extend to a matter that offends public policy or is illegal, it can never be a legitimate interest. Even if a legitimate interest in obtaining performance instead of damages is made out, an agreed damages clause must not be extravagant or unconscionable. In the words of Lord Hodge, affirmed by Lord Toulson,147 ‘the correct test for a penalty is whether the sum or remedy stipulated as a consequence of a breach of contract is exorbitant or unconscionable when regard is had to the innocent party’s interest in the performance of the contract’.148 The meaning of ‘unconscionable’ in this context is uncertain. In particular, it is not clear if it was intended to have any independent effect.149 There are indications, though, that it extends to situations in which the clause was originally agreed.150 It should therefore be unconscionable when the parties are not of equal bargaining power, where one has acted under compulsion (and possibly without legal advice) and where even a minor breach151 of the confidentiality terms triggers the repayment of the entire consideration. There is one final problem associated with penalty clauses. The preceding discussion has been predicated on the assumption that the relevant clause is a secondary obligation, triggered by a breach of contract. If, in contrast, the clause is drafted as a primary obligation of the contract, it falls outside the rule against penalties and the considerations discussed are irrelevant. While this difference between primary and secondary obligations is critical to the penalty rule, it is a distinction that can be hard to draw in particular cases.152 This difference enables a clever draftsperson to express what might be considered a secondary obligation as a primary obligation and so evade the scrutiny set out in this chapter. It might be 145 ibid [32] (Lord Neuberger and Lord Sumption). 146 Beale (ed), Chitty on Contracts (n 77) 26-225. 147 Cavendish Square Holdings BV v Makdessi (n 140) [255], endorsed by Lord Clarke at [293]. 148 ibid [255]. See also Halson’s consideration of unconscionability: R Halson, Liquidated Damages and Penalty Clauses (OUP 2018) 111–16. 149 Beale (ed), Chitty on Contracts (n 77) 26-229. 150 ibid, relying upon the judgments of Lords Neuberger, Sumption and Mance in Cavendish v Makdessi (n 140) and ZCCM Investments Holdings Plc v Konkola Copper Mines Plc [2017] EWHC 3288 (Comm). 151 Vivienne Westwood Ltd v Conduit Street Development Ltd [2017] EWHC 350 (Ch). 152 See Halson, Liquidated Damages and Penalty Clauses (n 148) 60–65.

Non-Disclosure Agreements  337 possible to argue, by analogy with employment cases,153 that the contract should be interpreted in accordance with the reality of the situation and in light of the fundamental inequality between the parties. The success of such an argument remains to be seen, although there are strong public policy grounds for allowing it.

III. Conclusions It is clear from the case law and media reports that there is an ever-increasing use of non-disclosure agreements in a manner that is abusive. The abuse affects not only the individuals compelled to enter these non-disclosure agreements, but also society as a whole. Non-disclosure agreements can, and do, work against broader public interests by effectively silencing information that would otherwise lead to broader public debate. A free and democratic society requires information of public harms to be known in order to respond to them. Non-disclosure agreements also work to undermine the rule of law, for one result can be seemingly to place individuals and institutions above the law. There is little statutory regulation in this area; legal practice has moved more quickly than Parliament has been able to respond. While these non-disclosure agreements depend upon contract law for their existence there are few counterbalances within existing contractual doctrines to challenge their enforceability. It does the law little credit if ordinary people – often trying to work for public good themselves – are forced into silence by powerful abusers. A number of reasons give rise to this situation. The first is that the very nature of these agreements is to create silence, and there is a comparative absence of litigation in this area. A second and more fundamental reason is that, as we have seen, English law struggles to protect weaker parties. While contracts are based on the consent of the parties, consent is really established through procedural mechanisms. English contract law is now not only conservative in its outlook but also tends to require individuals to be fairly robust in looking to their own affairs. Those doctrines that ought to assist the weak, such as unconscionability or duress, are currently formulated in such a way as to provide little protection. The underlying reason for this can be seen in cases such as Times Travel, where the assumption is made that weaker parties receive fundamental protection through legislation. In some situations, such as those involving consumers, this protection is there. Non-disclosure agreements occur where there is little legislative protection. A third reason for this situation is that the wider public interest problems created by this secrecy arise in the context of private rights. The focus of contract law is upon a particular contract and particular parties, making it very difficult to address these wider public concerns.



153 Autoclenz

v Belcher [2011] UKSC 41, [2011] 4 All ER 745.

338  Catharine MacMillan It is, however, appropriate for contract law to address these wider public concerns. In assessing the various contractual doctrines that might render a nondisclosure agreement unenforceable, illegality and public policy seem best able to respond. The continuing inability of Parliament to address these wider problems through effective regulation means that it is for courts to develop mechanisms to ensure that non-disclosure agreements are not used in the oppressive and abusive ways they currently can be.

16 Professional Ethics and NDAs: Contracts as Lies and Abuse? RICHARD MOORHEAD*

Catharine MacMillan’s chapter in this volume takes us comprehensively and critically though contractual doctrine relevant to non-disclosure agreements (NDAs).1 She points to the potential for serious abuse; the weakness in relying on the ‘emancipating effect’ of outside advice; and especially the potential for challenge because of duress or illegality.2 Her analysis presents a challenge from within contract law to judges and lawyers who see these agreements primarily in terms of freedom of contract.3 I wish to present a challenge more external to contract law: a challenge posed by professional ethics. The paucity of challenge to NDAs from within contract is partly a function of the complexity of the law, but more likely the result of structural inequalities presented by rich, well-resourced individuals deploying canny drafters and vigorous ‘reputation management’ teams. Professional regulation poses challenges that lawyers, but also judges, as interpreters of contracts, need to take more seriously. As MacMillan also reminds us, NDAs enable cloaks of secrecy to be thrown over sometimes serious wrongdoing in areas of public interest as wide-ranging as health, education, policing and the Church.4 It is in the silencing of women harassed and attacked by men that they have gained their most telling notoriety. The movie producer Harvey Weinstein and the retailer Philip Green are now synonymous with NDAs and the #metoo movement. At a charitable President’s Dinner event, employed hostesses were, ‘groped, sexually harassed

* I have benefited significantly from the thoughts of Prince Saprai, Philippa Collins, Paul Davies and Lucinda Miller in developing this chapter. All errors are my own. 1 C MacMillan, ‘Private Law and Public Concerns: Non-disclosure Agreements in English Contract Law’, ch 15 in this volume. 2 ibid. 3 J Cohen, ‘Non-Disclosure Agreements: The Truth behind the Headlines’ Counsel Magazine (London, February 2019) available at www.counselmagazine.co.uk/articles/non-disclosureagreements-the-truth-behind-the-headlines. 4 MacMillan, ‘Private Law and Public Concerns: Non-disclosure Agreements’ (n 1) section I.

340  Richard Moorhead and propositioned’.5 In apparent anticipation, they were asked to sign NDAs, which they did not have time to read and were not allowed to keep.6 I shall focus, in particular, on the Green and Weinstein cases, showing with regard to the former that whilst balancing key rights, privacy and the rights of the press in the public interest, upholding contracts has so far trumped in law but may not in professional ethics.7 I shall explain how the judges in that case failed to consider a key and, I would argue, inappropriate clause in giving Philip Green’s NDA a clean bill of health. In outline, the arguments are that the contract asserted rights that did not exist, asked for warranties that could not be enforced and sought indemnities that should not be paid (as penalties). Its terms were, in professionally punishable ways, likely to be lies and abuse. I use it as an example of how, when negotiating contracts that misstate the law or take unfair advantage of their client’s opponents, lawyers are in professionally punishable ways cheating on their clients’ behalf. I highlight some of the professional ethics dimensions of NDAs and how those agreements might, on the facts as known, instantiate criminal offences. I am conscious that such arguments are challenging ones. Some will recoil with an emotional intensity. It is unseemly in the extreme to suggest that reputable lawyers might be liars, cheats, even criminals. But those arguments stand up on the facts as known. Equally, because the facts are not all known, I do not assert that criminal offences have been committed here. That would require a rounded consideration of all the facts by a jury. And such facts are not available to us. There is, however, a colourable claim that offences have been committed on the face of those agreements. My interest in this illegality is not to seek prosecution, or to shame individuals, but to show how far into serious professional misconduct the drafting of bad contracts may take us. I consider along the way some of the other potential professional misconduct problems highlighted by the two case studies: those of Zelda Perkins and Philip Green.

I.  Zelda Perkins Zelda Perkins left her job with Miramax in 1998, following allegations that Harvey Weinstein had attempted to rape a colleague of hers and repeatedly sexually harassed her. Weinstein denied allegations of non-consensual sex or acts of retaliation against any women for refusing his advances.8 Having spoken out publicly,9 5 M Marriage, ‘Men Only: Inside the Charity Fundraiser Where Hostesses Are Put on Show’ Financial Times (23 January 2018) available at https://www.ft.com/content/075d679e-0033-11e89650-9c0ad2d7c5b5. 6 ibid. 7 Arcadia Group Limited and others v Telegraph Media Group Ltd [2019] EWHC 223 (QB) [40]. 8 M Garrahan, ‘Harvey Weinstein: How Lawyers Kept a Lid on Sexual Harassment Claims’ Financial Times (23 October 2017) available at https://www.ft.com/content/1dc8a8ae-b7e0-11e7-8c12-5661783e5589. 9 ibid.

Professional Ethics and NDAs  341 Perkins gave written and oral evidence to the House of Commons Women and Equalities Committee (WESC) as part of their investigation into sexual harassment in the workplace.10 Simons Muirhead and Burton (SMB) represented Perkins. A recently qualified solicitor conducted the negotiations with Miramax and Weinstein. Those negotiations were hosted and conducted by Allen and Overy, with a lawyer from Miramax also present. Advised against seeking Weinstein’s prosecution, Perkins and her colleague understood that they ‘had no option’ other than settlement, with stringent NDAs.11 At Perkins’ insistence, SMB requested £250,000 in compensation (her salary was about £20,000 at the time and her colleague’s was £16,000, so well in excess of one year’s income).12 Although SMB thought this unrealistic, the request was accepted subject to negotiation of terms. Those negotiations apparently concentrated on the Weinstein team pushing Perkins to name each person she had spoken to about her claims, and Perkins seeking undertakings that Weinstein would attend psychological therapy and that Miramax would institute policies to protect future complaints within the firm. Perkins describes the assistant solicitor who conducted the negotiations – albeit sometimes with a more experienced barrister on hand for advice – as, ‘utterly out of their depth’,13 and the negotiating sessions as gruelling, intimidating and frightening. One session ran ‘from 5pm until 5am’. A week after the negotiation, Perkins and her colleague were brought back to a meeting at which Weinstein was present, to sign the agreement. She said: He had a long conversation with us, trying to bring us back to the company and apologising for his behaviour. In fact, it was almost a full admission, which my lawyer noted. He was then not allowed to leave the room with that piece of paper unless it was destroyed.14

She was not allowed to keep a copy of the agreement.

A.  The Weinstein-Perkins Agreement An extract from the NDA has been published.15 Beyond that, most of what we have heard about the agreement comes from Zelda Perkins. Mark Mansell, 10 Z Perkins, ‘Written submission from Zelda Perkins (SHW0052)’ (March 2018) available at http://data.parliament.uk/writtenevidence/committeeevidence.svc/evidencedocument/women-andequalities-committee/sexual-harassment-in-the-workplace/written/80725.html; Z Perkins, ‘Oral evidence: Sexual Harassment in the Workplace’ (HC 725, 28 March 2018). 11 Perkins, ‘Written submission’ (2018) (n 10). 12 Perkins, ‘Oral evidence’ (n 10). 13 ibid. 14 ibid. 15 Excerpt NDA Agreement, Z Perkins, ‘Supplementary Written Evidence from Zelda Perkins (SWH0058)’ available at www.parliament.uk/documents/commons-committees/women-andequalities/Correspondence/Zelda-Perkins-SHW0058.pdf.

342  Richard Moorhead the partner from Allen & Overy (A&O) responsible for drafting it, declined to comment on the agreement specifically because of client confidentiality. He did provide comments that tell us quite a bit about the agreement, though, because he commented generically on agreements ‘like it’ whilst also indicating the agreement was unusual, in some ways exceptionally so.16 Thus, we learn that refusing a copy for one of the parties ‘would be extremely rare – very, very rare’. He could think of no other case where non-disclosure required a medical practitioner to sign an NDA before treating the woman. And a condition that a signatory cannot get further legal advice without having another NDA with that lawyer ‘would not be normal’. Long overnight negotiations, whilst sometimes justified, he indicated, would not be at all usual. Similarly, opponents would never be present in the same room in a negotiation, only in a mediation. That background given, let us turn first to looking at some of the actual clauses. Of crucial significance is clause 6(a). It requires Perkins to keep any information she has confidential unless she has ‘the prior written consent of Harvey Weinstein or Bob Weinstein’. And any legal advice on the agreement she receives must be given under an NDA approved by Miramax. Even then, disclosure is expressly prohibited except to any entity if required by legal process … but you will first, in the case of any civil legal process and where reasonably practicable in the case of any criminal legal process, give not less than forty eight (48) hours prior written notice to the Company through Mark Mansell at Allen & Overy before making any such disclosure and if any disclosure is made you will use all reasonable endeavours to limit the scope of the disclosure as far as possible. You agree to provide reasonable assistance to the Company and its legal advisers if it elects to contest such legal process. In the event that the Company does not contest such legal process or the challenge is not successful, you may make disclosure to your legal advisors (who must first agree in writing to execute a confidentiality agreement in a form satisfactory to the Company in the form of paragraph 6) but you will use all reasonable endeavours to limit the scope of the disclosure to your legal advisors as far as possible.

So, written consent is needed to make any disclosures voluntarily. Disclosure must only be given in legal proceedings if Perkins is required to do so. And any attempt to compel Perkins through legal process has to be notified to Miramax. Any disclosure has to be as limited in scope as possible (presumably in discussion with Miramax and/or Weinstein, or Perkins would be required to guess for herself what is necessary), but prior to that, she must provide any assistance asked for in Miramax’s contesting her being so compelled. Notably, the clause envisages notification and contestation taking place prior to her seeking legal advice. The clause thus prevents Perkins from making a criminal complaint. If Perkins wished unilaterally to disclose information to the police, under the agreement she



16 M

Mansell, ‘Oral evidence: Sexual Harassment in the Workplace’ (HC 725, 28 March 2018).

Professional Ethics and NDAs  343 would need the Weinsteins’ permission in writing. Similarly, Perkins cannot instigate civil legal process or respond voluntarily to others who do so. It might be argued that she could ask for consent and the Weinsteins would have to grant it because to do otherwise might pervert the course of justice. To know this was possible, she would presumably need legal advice.17 This would be rendered difficult because she was refused a full copy of the agreement. Perkin’s new solicitor might also object to advising under an NDA on the basis that it would compromise their independence and ability to represent the client’s best interests. Even when cooperating with lawful process, the clause permits disclosure by Perkins only if she is required to cooperate. Ordinarily, a witness is not required to give evidence to the police. Preventing Perkins from giving evidence voluntarily thus rules out the dominant means by which the police would investigate such matters. If a witness refuses to cooperate having been approached, there is a strong likelihood the police would not proceed to investigate the lead – they cannot compel a witness to give evidence, short of arresting them for the offence under investigation and seeking to interview them (and then the arrestee can remain silent).18 Similarly, in civil matters, such as a claim by a fellow employee that Weinstein had harassed them, civil litigants could issue a witness summons to require Perkins to give evidence but would not know what her evidence would be, and might have to contend with the potential for well-resourced objections from Miramax. To take the matter further, how would a potential litigant or prosecutor know to compel Perkins as a witness? They would want to know, in broad terms, what Perkins would be likely to say before considering compelling her to give evidence at trial. If they summoned her to give evidence, that would raise a suspicion that Perkins had already breached the NDA. In this way, Perkins would likely feel at risk of breaching the NDA if she had even intimated willingness to talk if compelled. Even if these hurdles are overcome, the requirements to limit disclosure suggest significant opportunities to inhibit and shape Perkins’ giving of evidence. There is potential, for instance, for a requirement not to disclose more than is necessary to have a significant chilling effect on any evidence given. In this way, the dominant, likely impact of the clause is to effectively preclude Perkins from cooperating with either civil or criminal process. Even in the unlikely event that cooperation were to be compelled, that cooperation is inhibited and the agreement buys time, and opportunities for intervention, for Miramax/Weinstein in the process.

17 Her evidence does not suggest that any doubt about the enforceability of the clauses was conveyed through advice at the time. 18 It is possible for potential witnesses to be compelled to give evidence by a magistrates’ court, under the Crime and Disorder Act 1998, sch 3, para 4, but this power is granted on application to a magistrate after a defendant has been charged and is to be sent for trial. A prosecutor would have to have sufficient evidence to charge before making this application.

344  Richard Moorhead Mark Mansell only hints at an explanation for the onerous agreement, saying agreements may be more extensive where high-profile public figures have particular sensitivities around reputation. Mark Mansell also concedes, ‘it would not be either reasonable or lawful to prevent somebody from participating in a criminal process’. As we have seen, the agreement seems plainly intended to influence whether, when and how any such participation takes place. In particular, he suggests that a clause could legitimately be used to prevent certain kinds of confidential information being given to the police: ‘information being disclosed that is not necessary for that process, [such that] the individual who is seeking to protect those interests has an opportunity to be involved’.19 In seeking an opportunity for a suspect to be involved in guarding the evidence of a witness, the opportunities to apply pressure to Perkins are clear. Unfortunately, it is not implausible that leading lawyers would engage in such pressure. Indeed, other partners in Mansell’s firm in a different case have been so implicated.20 Any time or information gained during this opportunity might enable Weinstein’s team to spread their efforts to understand who the other complainants against Weinstein might be – apparently a key element of the original settlement negotiations.21 This might suggest a wider attempt to control evidence as well as adverse publicity. They may, although this is speculation, also have NDAs against some of the other witnesses to whom the prosecution have spoken, which they may similarly seek to enforce.

II.  Might the Use of This Kind of Clause Amount to Perverting the Course of Justice? The offence of perverting the course of justice (PTCoJ) is committed where someone (i) acts or embarks upon a course of conduct, (ii) which has a tendency to and (iii) is intended to pervert (iv) the course of public justice.22 Does the Zelda Perkins clause, given evidence on the context, give rise to a concern that the offence may be made out? An NDA clearly constitutes a positive act. A tendency to pervert the course of justice requires, ‘a possibility that what the accused has done “without more” might lead to injustice’.23 There is an obvious risk that an NDA in the above terms

19 ibid. 20 A serious bribery trial was adjourned at significant cost because of allegations that A&O lawyers had pressured prosecution witnesses in the week leading up to a trial. C Binham, ‘Two Allen & Overy Lawyers at Risk of Probe over Dahdaleh Case’ Financial Times (26 March 2014) available at www. ft.com/content/b0e47460-b4dd-11e3-af92-00144feabdc0. 21 Perkins, ‘Oral evidence’ (n 10). 22 R v Vreones [1891] 1 QB 360. 23 R v Murray (GE) [1982] 1 WLR 475 (CA), (1982) 75 Cr App R 58; and see R v Firetto [1991] Crim LR 208 (CA).

Professional Ethics and NDAs  345 ‘without more’ leads to an injustice. Perkins is likely prevented from reporting and inhibited from cooperating with police enquiries. Indeed, if my construction of the clause is accepted, it operates to preclude disclosure of evidence to the police pre-charge. And it is likely to shape the nature of any disclosure. A potential suspect is empowered to limit or shape criminal proceedings. Plainly, PTCoJ covers concealing or destroying evidence with intent to influence criminal investigations and preventing a witness giving evidence. Agreeing to conceal a crime prior to an investigation is also capable of being PTCoJ. In many ways, the NDA’s practical impact is similar. As for intention, it is not necessary to show dishonesty. What has to be shown is that the act complained of has a tendency to pervert the course of justice and that the defendant intended it to do so. It is not necessary to show the wrongdoer believed they were acting through unlawful means but that they had engaged in ‘the intentional doing of an act having a tendency, when objectively considered, to pervert the course of justice’.24 The test would be, then, whether the inhibiting of Zelda Perkin’s disclosures to the police was intended to affect the course of justice. There are a number of reasons for thinking that it was. The agreement was intended to reduce substantially the likelihood of information from Zelda Perkins, or from any of the other people she has spoken to, coming to the attention of the police or third parties. Whilst this intention might be sufficiently evidenced by the agreement itself, the broader attempts to identify who Perkins has spoken to, and the wider history of Weinstein’s conduct being ‘widely known’, would likely be relevant.25 The more the lawyers involved can distance themselves from such knowledge, and the more plausibly they can mount the case that the NDA was solely aimed at protecting reputation rather than inhibiting the flow of information to the police, the less likely the intention is made out. We should not dismiss out of hand the possibility that controlling information that goes to the police might be legitimate. After all, the police sometimes use information inappropriately during investigations.26 However, it is not enough to claim that protection of reputation was one of the aims. If a substantial aim was also to inhibit or shape a police investigation, this is enough, potentially, for an offence to be made out. Similarly, whilst it would be argued that the NDA was part of a normal and legitimate strategy for managing sexual harassment allegations, it may not be 24 Archbold, citing A-G’s Application; Attorney-General v Butterworth [1963] 1 QB 696 (CA) 726 (Donovan LJ); Connolly v Dale [1996] QB 120, [1996] 1 Cr App R 200 (DC) 205 (Balcombe LJ); R v Meissener (1995) 130 ALR 547 (High Court of Australia) (cited in Lalani [1999]1 Cr App R 481, 491–92); and I Cram (ed), Borrie and Lowe’s The Law of Contempt, 3rd edn (LexisNexis 2010) 64ff, 410ff. 25 Perkins gave this evidence and it was supported by the evidence of R McGowan, ‘Submission to the Women and Equalities Committee on Sexual Harrassment and the Abuse of Non-Disclosure Agreements (NDAS)’ (19 April 2018) [sic] available at http://data.parliament.uk/WrittenEvidence/ CommitteeEvidence.svc/EvidenceDocument/Women%20and%20Equalities/Sexual%20harassment%20in%20the%20workplace/written/81746.html. 26 The disclosure of confidential information from the police investigation of the former minister Damian Green and the televised investigation of Cliff Richard come to mind.

346  Richard Moorhead enough that the strategy was potentially legitimate. Even if the agreement is lawful, this is not in and of itself a protection against a criminal charge. By way of example, in contempt cases, a solicitor’s employing lawful threats improperly has been found to amount to contempt.27 A lawful threat, or exercise of a legal right, can constitute an act intended to pervert the course of justice, ‘if the end in view is improper’.28 What matters is the intent behind the NDA, not the lawfulness of the agreement (and in any event the lawfulness of this agreement is in doubt). Influencing justice through a respectable solicitor is not a prophylactic against prosecution. Wrongful interference with a witness can include ‘improper pressure’.29 It is even possible for an intention to pervert to be found when improperly pressuring a dishonest witness to tell the truth.30 It is not certain, but there is sufficient evidence to suggest that what was done here was not just dubious, it was criminal. Under Crown Prosecution Service guidance, prosecutors bringing such case also have to satisfy themselves that there are serious aggravating features.31 And any defence might seek to argue that applying PTCoJ in such circumstances is an extension of the offence into new territory, which should only be done cautiously, step by step.32 The latter argument is weak: the analogy between what was done here and the simple concealing of evidence is strong, but a judge faced with any prosecution of reputable solicitors might be persuaded.33 Only a trial, with fuller evidence, would be likely to decide. But my principal interest in making this argument is not prosecution but showing how far from professional principles the use of NDAs can lead solicitors to stray.

III.  Professional Ethics Perspectives What, then, of professional rules applicable to this situation? The Solicitors’ Code of Conduct requires solicitors to uphold the rule of law and the proper administration of justice. They are also required to: act with integrity; not allow their independence to be compromised; act in the best interests of each client; provide a proper standard of service to their clients; and behave in a way that maintains the 27 Re Martin (P), The Times (DC, 23 April 1986). 28 R v Toney; R v Ali [1993] 1 WLR 364, (1993) 97 Cr App R 176 (CA). 29 R v Kellett [1976] QB 372, (1975) 61 Cr App R 240 (CA). 30 ibid. 31 R v Sookoo [2002] EWCA Crim 800. 32 R v Clark (Mark) [2003] EWCA Crim 991, [2003] 2 Cr App R 23 (CA). 33 As Proops’ and others’ submissions to the WESC made plain, other offences might also be considered, including that under s 4(1) of the Criminal Law Act 1967 (knowing or believing another to be guilty of a ‘relevant’ (arrestable) offence, to act with intent to do any act with intend to impede his apprehension or prosecution without lawful authority or reasonable excuse): A Proops et al, ‘Written Submission from Anya Proops QC, Aileen McColgan, Natalie Connor and Jennifer Robinson (SHW0059)’ (18 April 2018) available at http://data.parliament.uk/writtenevidence/committeeevidence.svc/ evidencedocument/women-and-equalities-committee/sexual-harassment-in-the-workplace/ written/80878.html.

Professional Ethics and NDAs  347 trust the public places in them and in the provision of legal services.34 The 10 principles are mandatory and ‘all-pervasive’.35 And where36 ‘two or more Principles come into conflict the one which takes precedence is the one which best serves the public interest in the particular circumstances, especially the public interest in the proper administration of justice’. An agreement found to have strayed into PTCoJ territory (as set out in section II) would clearly be a breach of professional conduct rules: at the very least, the integrity, administration of justice and trust obligations would be breached. There are other ways of thinking about the problem, though. What if the agreement’s errant clauses were merely unenforceable rather than criminal? MacMillan’s chapter37 reminds us that identifying unenforceability may not be straightforward, but for now I want to ask the question whether putting a plainly unenforceable clause into a contract is a professional breach. Lawyers thinking about their obligations to the administration of justice tend to concentrate on not attempting ‘to deceive or knowingly or recklessly mislead the court’ (O(5.1) in the Code). This includes not being complicit in another person’s deceiving or misleading the court (O(5.2)). The Code’s indicative behaviours (IBs, ie examples of when the Code is probably breached) suggest that one example is ‘constructing facts supporting your client’s case or drafting any documents relating to any proceedings containing: (a) any contention which you do not consider to be properly arguable’ (IB(5.7)). If a settlement agreement relates to any proceedings it might fall within this example, although a court is unlikely ever to see it and be misled by it. If we continue to assume that the agreement is unenforceable and that a lawyer could not stand up and properly argue that the inhibitions on Ms Perkins’ cooperation with the police were enforceable, can they include that term in the contract if it does not risk misleading the court? What might the professional rules say about that situation? The Bar’s Code of Conduct is clearer: it recognises an obligation not to knowingly or recklessly mislead or attempt to mislead anyone (rc9.1). The new Solicitors’ Code of Conduct now contains similar obligations.38 Even before that, solicitors could not knowingly or recklessly mislead people more generally because of their duty to act with integrity. As the Court of Appeal recently opined: [A] solicitor conducting negotiations or a barrister making submissions to a judge or arbitrator will take particular care not to mislead. Such a professional person is expected to be even more scrupulous about accuracy than a member of the general public in daily discourse.39 34 Solicitors Regulation Authority, SRA Handbook (Version 21, 2018). 35 ibid. 36 ibid. 37 See MacMillan, ‘Private law and Public Concerns: Non-disclosure Agreements’ (n 1). 38 Solicitors Regulation Authority, SRA Standards and Regulations (2019) para 1.4, available at https://www.sra.org.uk/solicitors/standards-regulations/code-conduct-solicitors/. 39 Wingate & Evans v The Solicitors Regulation Authority [2018] EWCA Civ 366, [2018] 1 WLR 3969.

348  Richard Moorhead Putting an onerous, but unenforceable, clause into an agreement risks misleading opposing parties, either during the negotiation or afterwards. It might be implied that the lawyer is contending that the enforceability of such a clause is properly arguable by the fact of its inclusion. If they know it is not then they are seeking to create a belief in another and something which is deliberately misleading. Scrupulousness would demand that it is not included. It might even be said that they are lying if they know the clause not to be enforceable. The need to consider one’s duty to act with integrity is strengthened by O(11.1), which prohibits taking ‘unfair advantage of third parties in either your professional or personal capacity’. This is supplemented by examples: IB(11.7), taking unfair advantage of an opposing party’s lack of legal knowledge where they have not instructed a lawyer; and IB(11.8), demanding anything that is not legally recoverable. The argument here would be that demanding unenforceable elements in an NDA is akin to claiming something that is not legally recoverable and so taking unfair advantage. It could be argued that IB(11.7) indicates that unfair advantage problems are confined to unrepresented parties. This is probably a misreading. Indicative behaviours are just examples, they do not limit the rules. The Code does not confine the obligation to acting against unrepresented parties, and the SRA’s Walk the Line guidance talks of being ‘careful not to take unfair advantage of the opponent or other third parties … Special care is needed where the opponent is unrepresented.’40 Similarly, the SRA’s Warning Notice on NDAs implies that unfair advantage can be taken of represented parties, by suggesting that ‘Where the employee is not represented, your obligations will be heightened, to ensure that there is no abuse of position, or unfair advantage taken.’41 If we assume for a moment that Weinstein’s lawyers managed to exploit a lack of knowledge or understanding in Perkins’ lawyer (again, this is speculation not fact), they are not necessarily protected simply by the fact that Perkins was represented. As the SRA notes, ‘solicitors involved in litigation [need not generally] … ensure that their opponents do not fall into traps of their own making’.42 But any misinformation around the clauses in the NDA might well be sufficient to lead to a ‘taking advantage’ finding. It would be difficult, if not impossible, for the SRA to legislate in advance for all the possible meanings of unfair advantage. Equally, practitioners struggle with their ethical obligations when acting against an unrepresented party, worrying about an inherent unfairness to their client if they are seen to be helping their opponent. They are likely to find doubly problematic the risk of being seen to be ‘soft-pedalling’ with a represented opponent. The A&O–SMB negotiation is a case in point. Was it for A&O or SMB to ensure the agreement was not unfair to Zelda Perkins? How far should a judgment

40 Solicitors 41 ibid

42 ibid.

Regulation Authority, Walking the Line: The Balancing of Duties in Litigation (2015). (emphasis added).

Professional Ethics and NDAs  349 on this reflect concerns about SMB’s decision, as alleged, to send in a two-yearqualified solicitor to negotiate with a sizeable team from A&O and Miramax? If the evidence is correct, agreeing to turn over some of one’s attendance notes, the way in which the client, not the lawyer, seemed to be leading the negotiation, agreeing to Perkins’ not having a copy of the agreement, and the (apparent) failure to challenge terms which are manifestly excessive, all give rise to concerns about the quality of Zelda Perkins’s representation during the negotiation. Mark Mansell suggested when giving evidence that his professional ethics were not impugned because any situation like that, where you have an individual who is legally advised, there is a negotiation, seeking to reconcile the interests of the two parties. I think, in doing that, I am compliant with my obligations.

I do not agree. Proving unfair advantage may be harder but not impossible in such circumstances; he is not relieved of responsibility by dint of a represented opponent. After all, one does not escape responsibility for throwing a punch by arguing one’s opponent should have moved their head. The key question is whether a punch was thrown, not whether the opponent should have moved. Asking for an unenforceable clause to be included is highly likely to be unfair if one accepts that it is misleading, or that it seeks to make a legal claim which has no substance. Additionally, it imposes a significant practical detriment in that it forces the opponent to adhere to or challenge enforcement of such clauses. Those opponents are typically without the resources to mount such challenges. What is more, representation of the more vulnerable party is not continuous; the unfairness of the clauses continues after the agreement is executed. Furthermore, the NDA has a life beyond its creation. It can be designed to take advantage of the counter-party after they are represented. Most ex-employees are not going to retain lawyers to help them interpret their obligations under an NDA. Unfair advantage must be a particular concern given that Perkins was required to give warnings and assistance when compelled to give evidence, before she could seek legal advice. The SRA document sensibly concludes with this reminder: There will always be complex situations where maintaining the correct balance between duties is not simple and all matters must of course be decided on the facts. It is important for solicitors to recognise their wider duties and not to rationalise misconduct on the mistaken basis that their only duty is to their client.43

These arguments become especially important when thinking about one string to Mr Mansell’s defence of himself before the WESC, which leads us towards situations where the tactics employed in a case, or clauses employed in a contract, are arguably legal. For Mr Mansell that was the argument that it might be proper to limit the disclosure of some material to the police.



43 ibid.

350  Richard Moorhead

IV.  The Balancing Exercise How to interpret a clause shaping disclosure to the police evidence in professional ethics terms? Shaping disclosures to the police plainly can constitute perverting the course of justice. Mr Mansell may not have thought about that at the time; or he might have taken the view that the potential risks of that amounting to a perversion were not sufficiently strong or clearly spelled out in the law and that is sufficient for the clause to be permissible. So, for example, because there is not a case of a solicitor’s drafting a similar NDA and being successfully prosecuted for it, he might feel that the law is sufficiently uncertain for him to disregard potential restrictions in favour of his client. Or he might take the view that because one can articulate the view that a clause is potentially justifiable, because the law does not prohibit parties from exerting any influence on potential witnesses, it is in fact justified in the instant case. Such an approach is consistent with the idea of professional minimalism I have set out elsewhere.44 Under professional minimalism, only unarguably illegal acts are restrained and the client gets all the benefit of uncertainty; a mere risk of perverting the course of justice is sanctified as legitimate by dint of private bargaining and representation. Professional minimalism legitimates decisions about, and therefore distances responsibility for, unsavoury tactics. Those decisions are often shielded by client confidentiality and legal professional privilege. The client can say ‘I was acting on advice’; and the lawyer can say ‘I was only following instructions.’ There are a number of reasons for giving clients the benefit of uncertainty. First, and importantly, it prevents lawyers from having to apply legal uncertainty against their own clients, bolstering loyalty. It is also usually in their commercial interests to align as fully as they can with clients. If they neglect their clients’ interests, there is the risk of being sued; whereas if they neglect the public interest in the administration of justice, the risk is lower. Indeed, it is rare for this neglect to be revealed, because lawyer and client interests are usually aligned, confidentiality protects the lawyer and the client from scrutiny, and enforcement is rare. Similarly, ethical and tactical dilemmas, and the psychological burdens of practising law, are simplified considerably when it is only if there is a clear breach of law or professional ethics that lawyers must restrain the imperative to act in their clients’ interests. Lawyers also see themselves as bound to accept their clients’ instructions on how to handle a case. This is wrong. Under their professional principles, they have an obligation to act with independence, and to consider their broader obligations to protect the rule of law and the administration of justice. They must take account of clients’ best interests when thinking about these issues, and the factors in the previous paragraph will generally encourage them to do so, but they are not to see themselves as absolved of responsibility for their decisions on the basis that 44 R Moorhead and V Hinchly, ‘Professional Minimalism? The Ethical Consciousness of Commercial Lawyers’ (2015) 42 Journal of Law and Society 387.

Professional Ethics and NDAs  351 they were just following the clients’ instructions. Responsibility and judgement are axiomatic to professionalism, and that judgement requires a rounded consideration of all the relevant professional rules and principles, not just the clients’ best interests. So lawyers have to decide for themselves whether deploying a tactic in settlement discussion is misleading or taking unfair advantage or not. As Lord Chief Justice Judge remarked: Something of a myth about the meaning of the client’s ‘instructions’ has developed. As we have said, the client does not conduct the case. The advocate is not the client’s mouthpiece, obliged to conduct the case in accordance with whatever the client, or when the advocate is a barrister, the solicitor ‘instructs’ him …. That is the foundation for the right to appear as an advocate, with the privileges and responsibilities of advocates and as an advocate, burdened with twin responsibilities, both to the client and to the court.45

Farooqi was about criminal advocacy, but the words apply doubly in civil contexts, especially in cases resolved away from the courts, where judges cannot exercise supervisory restraint. As a result, the balancing of ethical principles requires a more nuanced approach than saying ‘something is theoretically permissible, therefore I can do it’. The lawyer may be bound, if we come back to the PTCoJ frame, to think about what the likely and intended effects of the agreement were, not just what legitimate purposes the agreement could be put to.

V.  Common Law and the Freedom of Contract The preceding analysis shows how professional ethics considerations turn partly on the enforceability of clauses, but also on concepts such as unfair advantage and protecting the rule of law and administration of justice. It also suggests the need for balance and judgement. An alternative view is that NDAs promote the rule of law through their facilitation of settlement agreements. Freedom of contract and the enforceability of contracts are important public interest concerns in their own right. Freedom of contract is, of course fundamental to English commercial law.46 In essence, this means parties are free to make agreements, and the courts should enforce the terms of those agreements unperturbed by, even showing an ugly reverence for, sharp practice: ‘fairness has nothing to do with commercial contracts’ since ‘[c]ommercial parties can be most unfair and entirely unreasonable, if they can get away with it’.47 This view ignores the professional obligation

45 R v Farooqi [2013] EWCA Crim 1649, [2014] 1 Cr App R 9 [108]. 46 PS Davies, ‘Bad Bargains’ (2019) 72(1) Current Legal Problems 253. I am indebted to this piece for the cursory summary of the law that follows. 47 Lord Sumption, ‘A Question of Taste: The Supreme Court and the Interpretation of Contracts’, (Harris Society Annual Lecture, Oxford, 8 May 2017) available at www.supremecourt.uk/docs/ speech-170508.pdf.

352  Richard Moorhead not to take unfair advantage, and could be used to support a professionally minimalistic judgement about NDAs. As MacMillan’s chapter48 shows us, contract law has a number of ways of suggesting that ‘getting away with it’ is subject to proper restraint (duress, illegality and so on). Non-disclosure agreements may be contracting out of the Equality Act 2010, even though section 144 renders any such term unenforceable unless it is a ‘qualifying settlement agreement’ (section 147). Section  43J of the Employment Rights Act 1996 renders void any provision in an agreement that ‘purports to preclude the worker from making a protected disclosure’ (in very broad terms protecting whistle-blowers upon disclosure of crimes or other wrongs in the public interest to prescribed organisations). Freedom of contract depends on what the courts talk of as ‘real choice’49 and the criticality of the assumption that ‘consent to the terms of the contract has been obtained fairly’.50 MacMillan’s plea for more critical focus on NDAs,51 and a broader recognition that the law on illegality is complex and voluminous,52 suggests courts may not take these restraints as seriously as they might. Only in the most unusual of cases will the resource constraints on employees be overcome. Persuading a court to believe that consent has been vitiated is difficult: being under pressure, or having a weak negotiating position, is not enough on its own.53 And, as we shall see in section VI, the courts seem willing to regard independent representation as enough to suggest that a real choice was properly exercised.

VI.  The Court Consideration of the Green Cases Freedom of contract was at the forefront of the decisions in the Philip Green/ Arcadia/Topshop/Topman’s (the Group) injuncting of the Daily Telegraph.54 The newspaper sought to expose the Group’s use of NDAs to deal with alleged impropriety by Green. As the Court of Appeal noted in its ABC decision,55 five Group employees made allegations of ‘discreditable conduct’ against Green. All five cases

48 See MacMillan, ‘Private Law and Public Concerns: Non-disclosure Agreements’ (n 1). 49 G Leggatt, ‘Making sense of contracts: the rational choice theory’ (2015) 131 LQR 454, 474. In the commercial context, a major exception arises regarding exclusion clauses given the Unfair Contract Terms Act 1977, although even here a court is unlikely to interfere where a contract was made between well-advised parties of similar bargaining power: Watford Electronics Ltd v Sanderson CFL Ltd [2001] EWCA Civ 317, [2001] 1 All ER (Comm) 696; Unfair Contract Terms Act 1977, sch 2(a). 50 First Tower Trustees Ltd v CDS (Superstores International) Ltd [2018] EWCA Civ 1396, [2019] 1 WLR 637 [104]. 51 MacMillan, ‘Private Law and Public Concerns: Non-disclosure Agreements’ (n 1). 52 E McKendrick, Contract Law: Cases Text and Materials, 8th edn (OUP 2018) ch 20B. 53 Wood v Sureterm Direct Ltd [2015] EWCA Civ 839. 54 See Arcadia Group Ltd v Telegraph Media Group Ltd [2019] EWHC 96 (QB), [2019] 1 WLUK 173. 55 ABC & Others v Telegraph Media Group Ltd [2018] EWCA Civ 2329, [2019] 2 All ER 684.

Professional Ethics and NDAs  353 were settled, with substantial payments made to the complainants. Confidentiality provisions were included in the agreements. All the complainants were reportedly independently advised. And the Court of Appeal felt that ‘[t]he Agreements safeguarded the complainants’ rights to make legitimate disclosures (including reporting any criminal offences) if they chose’.56 That is, the Court of Appeal thought the agreements did not fall foul of whistle-blowing legislation. In July 2018, a Daily Telegraph journalist contacted the claimants for comments on the allegation and the NDAs. This suggested to the Group that the information in question had been disclosed to the newspaper by one or more of the complainants or by other employees who were aware of the information and of the NDAs, and they immediately commenced the present proceedings.57

And: At an early stage in the proceedings Nicklin J directed that attempts be made to ascertain the attitudes of the five complainants to whether information about their complaints should be published, even if they were not named. One complainant said that they were happy for their complaint, and the settlement, to be disclosed, provided they were not named. Two said that they supported the Claimants’ application for an injunction. One said they did not support the application.58

We do not know how that information was forthcoming. Presumably, it was garnered by the lawyers for the Group, as only they would know the identities of the complainants. The court does not express any doubt about the depth and reliability of any of the complainants’ reported views on this matter. The interim injunction application was unsuccessful at first instance, but successful in the Court of Appeal. There the Court weighed Article 10(1) of the European Convention on Human Rights (ECHR) (‘Freedom of expression’) and ‘the protection of the reputation or rights of others, for preventing the disclosure of information received in confidence (Article 10(2) ECHR); respect for private and family life (Article 8(1) ECHR); and whether the court would be sufficiently likely to prohibit publication at a full trial. ‘Sufficiently likely’ does not mean ‘more likely than not’ if the adverse consequences of disclosure seem sufficiently grave to the judge. One such adverse consequence was the impact on Green’s reputation. Various elements of public interest were in play too: the principle that there is no confidence in an iniquity; whether it was vital in the public interest to publish confidential information; and ‘a public interest that confidences should be preserved and protected by the law’. I note in passing that protection of reputation appears three times in the justifications offered by the Court: in Article 10(2) ECHR; in the softening of the likelihood test; and in the protection of confidences.



56 ibid

[4]. [5]. 58 ibid [6]. 57 ibid

354  Richard Moorhead Balancing the competing interests required a test of proportionality, having regard to the nature of the information and all the relevant circumstances, [whether] it is legitimate for the owner of the information to seek to keep it confidential or whether it is in the public interest that the information should be made public.59

What seemed to weigh heavily in the balancing was that duties in such cases arose out of contract, especially where the obligation in question is contained in an agreement to compromise, or avoid the need for, litigation, whether actual or threatened. Provided that the agreement is freely entered into, without improper pressure or any other vitiating factor, and with the benefit (where appropriate) of independent legal advice, and (again, where appropriate) with due allowance for disclosure of any wrongdoing to the police or appropriate regulatory or statutory body, the public policy reasons in favour of upholding the obligation are likely to tell with particular force, and may well outweigh the article 10 rights of the party who wishes to publish the confidential information.60

In this way, the Court of Appeal suggests that the public interest can be compromised by the creation of private rights at the end of a dispute. Putting aside the question of whether this is right in principle, central to the argument is that whistleblowing provisions were protected by the agreement. I disagree. And at the very least, even at the interim stage, this merits a more searching inquiry into the propriety of the agreement than the Court itself attempted in its published judgment. What did the Court pay attention to in reaching its decision? The judges said they paid close attention to ‘reasonably credible’ evidence of wrongdoing (the harassment presumably), but there were ‘factors going the other way that need to be weighed in the balance at this, interlocutory, stage’, including ‘that the most serious allegations made by the complainants had been denied and that the settlement of the ET [Employment Tribunal] claims meant that the opportunity to have their truth determined by an independent tribunal had been lost’.61 They negate ‘reasonably credible’ evidence on the basis of a denial and a settlement that supports no factual finding either. The first instance judge was criticised because he had ‘left entirely out of account the important and legitimate role played by [NDAs] in the consensual settlement of disputes, both generally but in particular in the employment field’.62 There is no evidence that any of the Settlement Agreements were procured by bullying, harassment or undue pressure by the Claimants. Each Settlement Agreement records that the employee was independently advised by a named legal adviser. Each

59 ibid [21], citing HRH the Prince of Wales v Associated Newspapers Ltd [2006] EWCA Civ 1776, [2008] Ch 57 [68]. 60 ABC & Ors (n 55) [22], here citing HRH the Prince of Wales v Associated Newspapers Ltd (n 59) [69] approvingly. 61 ABC & Ors (n 55) [33]. 62 ibid [41].

Professional Ethics and NDAs  355 Settlement Agreement contained provisions authorising disclosure to third parties in a range of cases, including to regulatory and statutory bodies. They did not in the present case, therefore, on the face of the evidence at this interlocutory stage, have any of the unethical vices criticised by the WESC Report.63

Here the Court of Appeal appears to be relying on the agreement itself and the existence of independent advice as evidence of fairness and propriety. To rebut the presumption that the agreements were freely entered into, the defendant newspaper would have had to have evidence of the negotiations, which, presumably, would very likely reveal the confidentiality breaches of the employees themselves (or likely others subject to general duties of confidentiality as former or existing employees or advisers of the claimants). The agreement itself closes off this risk that such evidence would come to light: it shapes the course of justice in a way not recognised by the Court. Second, and more fundamentally, it seems plausible that the Court of Appeal did not pay close enough attention to the terms of the NDA itself, because the agreements certainly did in my view create the very mischief with which the WESC was concerned. We can see this by looking at a now published NDA relating to one of the ex-employees of Arcadia given the pseudonym ‘Alex’. It was set out in an appendix to the last Arcadia judgment, given when the original injunction was discharged.64 The agreement indicates three types of compensation to be paid to Alex: a lump sum ‘for injury to feelings and aggravated damages’ in settling a tribunal claim (clause 4.1.1); a further lump sum in respect of compensation for the termination of Alex’s employment and any other claims against their employer and Sir Philip (or associates) (clause 4.1.2); and a third set of monthly compensation sums (paid until November 2018) for similar reasons (clause 4.1.3). As part of the bargain, Alex is prohibited from disclosing information about the grievance, the termination of their employment and their claim (including the settlement of it), ‘save to immediate family/professional advisers’, or ‘where required by any governmental, regulatory or other competent authority or by a Court of law or Her Majesty’s Revenue and Customs’. Under clause 13 there are some further exceptions to the agreement: making a ‘protected disclosure’ (under whistleblowing legislation) is allowed for instance; as is ‘reporting a criminal offence to any law enforcement agency; and/or co-operating with any law enforcement agency regarding a criminal investigation or prosecution’. It follows that the agreement does appear to allow cooperation with, and reporting to, at least some law enforcement agencies. This explains the Court’s view that Each Settlement Agreement contained provisions authorising disclosure to third parties in a range of cases, including to regulatory and statutory bodies. They did not in the

63 ibid

[43]. Group Limited (n 54).

64 Arcadia

356  Richard Moorhead present case, therefore, on the face of the evidence at this interlocutory stage, have any of the unethical vices criticised by the WESC Report.65

However, a very interesting question arises as to whether that apparently positive protection of whistle-blower rights is likely to be undone by clause 13.3, which states: You warrant, however, that you do not know of any circumstances which would lead you to making a disclosure in the form of or in the circumstances referred to in this clause.

On breaching this warranty, Alex would be obliged to pay two out of the three heads of compensation received, recoverable as a debt, ‘together with our costs, including legal fees, in doing so’. At least one of the NDAs to which this case relates involves over £1million in compensation payments, so the repayment penalty might be high. Leaving to one side whether this was an unenforceable penalty clause, one more fundamental and very interesting question raised by this clause is its likely and intended effect. Let us imagine I have been assaulted, possibly sexually, and yet promise that I do not know of any circumstances in which I would make a report of that conduct to the police. One interpretation is that I am being given rights to disclose, but I am promising never to exercise, or perhaps save in unforeseen circumstances, those rights. If I do, I may risk paying the indemnity and being placed in the compromising situation of having promised something that is contrary to what I know. That might be used in an attempt to discredit me later should I give evidence. Analysing it under the PTCoJ framework, critical questions are raised about what is intended or likely as a result of the clause. Am I more likely not to disclose wrongdoing to an investigation of my own volition or if approached as a result of this clause? Is that its intended effect? It seems to me that a likely and foreseeable effect of the clause is that the former employee can report to or cooperate with legal investigations in theory, but would do so fearing the risks of breaching the warranty. This would be particularly true if the employee could have envisaged, prior to signing the agreement, that they might be approached and asked to make a disclosure to a prosecutor, or if they could envisage changing their mind and disclosing to a law enforcement agency of their own volition, in which case any such disclosure might be said to indicate a breach of the warranty. If that analysis of the clause is correct, it appears to be drafted with the intention of making it significantly less likely that Alex will disclose wrongdoing to, or cooperate with, a law enforcement agency or to an employee thinking of making a claim. It does not preclude the exercise of whistleblowing rights, but it does make their exercise illusory or close to it.

65 ABC

& Others (n 55) [43].

Professional Ethics and NDAs  357 Now, we do not know what the parties say about the reasons for drafting that clause, and so we must remain circumspect, but on its face the clause gives rise to significant concerns. Whether this would fall within the ambit of PTCoJ depends a great deal on the persuasiveness of alternative explanations for the clause. One suggestion is that the agreement may be aimed at circumstances where there is an employee who has engaged in poor behaviour but there is no evidence of any criminal act. The employer is engaged in buying the silence of those with legitimate but non-criminal complaints. The clause might be intended to ensure that the signatory has disclosed all of the issues that may have occurred. Where a settlement is paid on the basis that everything is now dealt with, it gives the company a mechanism for clawing this back where the signatory comes out later with further issues that they had not disclosed.66 In this way it provides the employer with sufficient certainty they can settle ‘difficult’ cases. Given the multiple complaints of harassment and other allegations made in the Green case about the probity of their handling (such as allegedly compromised internal investigations), it is hard to see how the lawyers involved could claim such purity of purpose. But one problem of principle with this interpretation is that rather than protecting the employer in other ways (eg mandating full disclosure of the allegations to the employer, and documenting those as part of the settlement), it seeks to warrant something that is likely to interfere with someone’s willingness to report a crime. It allows the private ordering of a public function that should not be interfered with: there is a public interest in investigating allegations of iniquity, proportionately protected by allowing the subjects of NDAs to talk to the police unrestrained by cash for silence. The counter-parties to NDAs are not being allowed to trash the employer’s reputation, they are being permitted to make a complaint to, or cooperate with, the police. The main risk to the employer is that the police take that evidence seriously. This is not a risk they should be able to button down. It goes to the heart of a process that the offence of PTCoJ seeks to protect. Use of warranties to ‘work round’ what would otherwise be seen as clear requirements under the Public Interest Disclosure Act 1998 may not be uncommon. The whistle-blowing charity Protect, for instance, reports that whistle-blowers can be required to warrant that the matter they have raised has been ‘satisfactorily concluded’, risking, on the face of such agreements, claims for repayment of compensation and costs should they disclose in a way that suggests they were not in fact satisfied.67 And they risk attacks on their credibility. Warranties can include the former employees being asked to promise that there are no circumstances of which they are aware that would amount to a breach of the regulatory 66 I am grateful to Rob Stevens and Robin Brooks for these points. 67 See Protect, ‘Written Submission from Protect (NDA0038)’ (January 2019) available at http://data. parliament.uk/writtenevidence/committeeevidence.svc/evidencedocument/women-and-equalitiescommittee/the-use-of-nondisclosure-agreements-in-discrimination-cases/written/95054.html. Private communication with the author.

358  Richard Moorhead requirements applicable to the company, even where the complaints they raised might or do in fact raise such concerns. Or that they withdraw all appeals/grievances, data protection requests and any complaints to any ombudsman or similar authority. And similarly agreements that provide for payment of automatic indemnities if the individual exercises or attempts to exercise any of the statutory rights referred to in the agreement for Public Interest Disclosures (a less subtle version of Alex’s clause). On their face, and again subject to any proper explanations that can be offered, these seem to be deliberate attempts to frustrate whistle-blower protections.68 At least in circumstances where criminal and/or civil proceedings are in train, and probably where such proceedings are in contemplation, there is a significant potential that these agreements are likely to and intended to pervert the course of justice. If drafted by a solicitor, on the face of the agreements alone, significant questions about breaches of the professional code are also raised for similar reasons to those discussed previously. Even if one accepted that such clauses might, in certain circumstances, be legitimate, at the very least a judgment balancing the public interest suggests that the possibility that this clause would lead to interference should be canvassed by the court. It was not. The Court of Appeal judges showed themselves insensitive to the significant public interest challenges posed by the agreements instead protesting the public interest in enforcing them. They relied on a theoretical, decontextualised and inadequate understanding of the contract in the face of highly significant public interest concerns: it was an unbalanced balancing. A fair response in the instant case might be that this was an interim hearing, where the judges would not be expected to engage fully with the merits but seek simply to protect the status quo.69 I would have more sympathy with this argument had the Court not indicated that the agreement does not appear to give rise to the mischief complained of by the WESC. This incautious statement suggests a premature belief that the agreement was probably sound, when there were problems on its face. There is, though, a third point of concern that was not considered, which relates to arguments about costs.

VII.  The Costs Arguments The negotiation of NDAs takes place in a system of usually significant structural inequality: a soon-to-be or actually unemployed person presented with a compensation payment and an NDA on a take-it-or-leave-it basis. Where there is doubt

68 Interestingly, the Financial Conduct Authority prohibits settlement agreements that include warranties related to protected disclosures: www.handbook.fca.org.uk/handbook/SYSC/18.pdf. 69 I am grateful to Lord Sales for emphasising this point.

Professional Ethics and NDAs  359 about the appropriateness of an NDA in general, or in relation to specific terms, those concerns are likely either ignored or underplayed because: • there is a widespread practice of accepting NDAs, which means that any negotiations take place against an industry norm that accepts widespread and widely drawn non-disclosure agreements; • they are, or have been, seen as a useful way of managing reputational and legal risk – so even if not legally enforceable, widely-drawn clauses help restrain most employees from discussing allegations of wrongdoing; • employees are often either unrepresented or under-represented (eg on limited retainers, where a lawyer advises on whether they can sign an NDA) and may, at the point of settlement, feel they have little alternative or interest in challenging the breadth of an NDA that they do not really understand; and • are operating in a system that concentrates on compensation on exit/dismissal as the main ex-post response to sexual harassment and discrimination. But I want to concentrate on one more specific element that calls into question the idea of a bargain freely chosen: costs.70 It is relevant to MacMillan’s arguments about duress.71 This relates to another element of the Green/Arcadia/TopShop agreements, which the court mentioned but did not reflect on: the size of the agreed compensation. Ordinarily, employment cases would be pursued in the employment tribunal, although harassment cases might potentially be pursued through the courts. Sex discrimination claims arising out of employment can only be pursued in the Employment Tribunal.72 Ordinarily, each party bears its own costs, win or lose, although a tribunal can make an exceptional award for unreasonable conduct. Employers have been known to argue that failure to accept a settlement is unreasonable conduct, including failure to accept an NDA (although such arguments are generally, if perhaps not always, unsuccessful).73 As Regan has noted, a respondent wishing to secure an NDA can make ‘an enhanced offer to settle if an NDA were executed’.74 That is, they offer more than the maximum award to which a claimant is entitled,75 or is likely to be awarded (possibly including a payment of unrecoverable costs), on condition that an NDA 70 I am indebted to communications with Dominic Regan, ‘Written Submission from Professor Dominic Regan (NDA0073)’ (March 2019) available at http://data.parliament.uk/writtenevidence/ committeeevidence.svc/evidencedocument/women-and-equalities-committee/the-use-ofnondisclosure-agreements-in-discrimination-cases/written/98704.html. 71 MacMillan, ‘Private Law and Public Concerns: Non-disclosure Agreements’ (n 1) section II.B.ii.c. 72 Sheriff v Klyne Tugs (Lowestoft) Ltd [1999] IRLR 481 (CA). 73 See, eg, Anderson v Cheltenham & Gloucester Plc [2013] UKEAT/0221/13/BA, where a claimant was ordered at first instance to pay a defendant’s costs having turned down a Calderbank offer (an offer made without prejudice save as to costs), but this was overturned on appeal. 74 Regan, ‘Written Submissions’ (n 70). 75 For example, unfair dismissal cases are ordinarily subject to a maximum of £83,682 or 52 weeks’ net pay, whichever is the lower.

360  Richard Moorhead is signed. Let us call these ‘over-par offers’, being offers that cannot realistically be beaten in any final hearing. Over-par offers put the claimant wishing to resist an NDA in a difficult situation. Financially, the offer may come at a time of great financial vulnerability. Moreover, if they are funded under a no-win-no-fee agreement, the solicitor may decide to withdraw from representing them unless they accept the offer; if they are insured, the insurer may similarly withdraw on the basis that a (more than reasonable) offer of compensation has been made; the same is true for union funding.76 If the claimant were funding themselves they would be able to proceed, but only if they were willing to risk substantial accumulation of legal costs and a potential application to the tribunal that they had been unreasonable. Such an award of costs might be very unlikely but nonetheless used to put pressure on the claimant. The situation may be even worse for claims taken to a court, not a tribunal. Given that costs ordinarily follow the event, a claimant declining any reasonable offer can suffer costs penalties, for example by reason of offers without prejudice save as to costs that may wipe out or exceed their compensation. Any over-par offer is likely to be unbeatable, putting the claimant at significant risk. Regan reports phone-hacking litigation as an area where over-par settlements were offered to ensure settlement with NDAs. The size of the settlements reported in the Green case raises a suggestion that such may have been the tactic here. Such costs penalties might well exceed any compensation awarded, even assuming the claims were successful. Regan puts it thus: [A] defendant has the capacity to throw money at a claim and seek to buy it off. I have no personal knowledge of recent claims but, if press reports are accurate, it would appear that victims of sexual and racial discrimination have accepted sums which would never be remotely recoverable in a Tribunal or Court case. They cannot be compelled to keep quiet but the financial threat they face is overwhelming and there are law firms which boast of their ability to ‘protect reputations’.77

VIII. Conclusions The problems seen in the Green and Weinstein–Perkins agreements are suggestive of an interesting set of tensions. Contract law purports to protect freedom but here instantiates the silencing of women in ways that, I have argued, are professionally problematic. In showing that they may sometimes be potentially criminal, I have concentrated on showing how contractual freedom can be taken to extremes: de facto, through the professional failings of the lawyers involved; and, given decisions in Green, de jure, shielding the powerful from scrutiny without that same 76 ‘No Help for Doctor Who Refused to be Gagged’ The Times (30 January 2012) available at www. thetimes.co.uk/article/no-help-for-doctor-who-refused-to-be-gagged-s3g5073phq3. 77 Regan, ‘Written Submissions’ (n 70).

Professional Ethics and NDAs  361 contract law’s being effectively engaged because the contract ‘looks okay on its face’ when in my opinion it looks highly questionable. The importance of professionalism in the drafting of contracts has been bolstered by the Solicitors Regulation Authority’s Warning Notice on NDAs. It provides guidance against the abuse of NDAs, covering some, but not all, of the points raised in this chapter.78 In particular it warns against the use of NDAs as a means of improperly threatening litigation or other adverse consequences, or otherwise exerting inappropriate influence over people not to make disclosures which are protected by statute, or reportable to regulators or law enforcement agencies.79

Whether contract law pays sufficient attention to the kinds of issues I have raised here is moot. What is striking is how irrelevant duress and other attacks on the NDAs were to a consideration of their interim enforceability. The possibility that the indemnities in the Green case were unenforceable penalties got only the slightest of mentions. Professional enforcement now beckons against Mr Mansell,80 and a similar fate may await the drafter of Mr Green’s NDAs. We shall see. But whilst nice arguments about freedom turn the heads of contract lawyers, the regulators are showing signs of being more concerned with the inappropriate application of professional power.

78 SRA, ‘Use of Non-Disclosure Agreements (NDAs): Solicitors Regulation Authority’ (12 March 2018) available at www.sra.org.uk/solicitors/guidance/warning-notices/use-of-non-disclosure-agreementsndas--warning-notice/. 79 ibid. 80 ‘SRA Attempts to Prosecute A&O Lawyer over Controversial Weinstein Gagging Deal’ Legal Business (3 April 2019) available at www.legalbusiness.co.uk/blogs/sra-attempts-to-prosecute-ao-lawyerover-controversial-weinstein-gagging-deal/.

362

17 Choice of Court and Choice of Law Agreements: Freedom of Freedom of Contract ALEX MILLS*

I.  Freedom of Contract Justifications for Party Autonomy in Private International Law Party autonomy is now very widely accepted in private international law, which is to say, parties to cross-border contracts are frequently given the power to determine for themselves which courts have jurisdiction over their disputes and what law governs their private legal relationship. Choice of court1 and choice of law2 agreements are very commonly given strong and even decisive effect in determining the court in which disputes can be heard or the applicable law for contractual and sometimes even non-contractual3 disputes arising between the parties.4 What is less well understood is that this is a radical departure from the uncertain legal position of party autonomy at the beginning of the twentieth century. At that time, party autonomy was frequently rejected by academics, and to a lesser extent (but still commonly) by courts. In the context of choice of law, the most prominent among the academic critics was Joseph Beale, an influential Harvard * This chapter draws on research previously published in Party Autonomy in Private International Law (CUP 2018). 1 See, eg, Art 25 of Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (recast) [2012] OJ L351/1 (‘Brussels I Regulation’), which (outside of certain insurance, consumer and employment disputes) is trumped only by the rules of exclusive subject matter jurisdiction in Art 24 and submission to another forum in Art 26. 2 See, eg, 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) [2008] OJ L177/6 (‘Rome I Regulation’). 3 See, eg, Art 14 of Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II) [2007] OJ L199/40 (‘Rome II Regulation’). 4 See generally A Mills, Party Autonomy in Private International Law (CUP 2018).

364  Alex Mills professor and the author of the First Restatement of Conflict of Laws in 1934, who rejected party autonomy5 on the basis that it purported to attribute to individuals a power over states, arguing that ‘The fundamental objection to [party autonomy] in point of theory is that it involves permission to the parties to do a legislative act.’6 Party autonomy in private international law has come to appear an obvious feature of modern contracting practice, but it is striking that to many scholars a century ago it was equally obvious that it was impossible. The principal focus in this chapter is on the choice of law aspects of party autonomy, rather than on choice of court, although both are covered to some extent and (as discussed further in section III) it should be noted that these are importantly related. An effective choice of forum determines, for example: (a) which procedural law applies (as each court applies its own procedural rules, such as its own rules of evidence);7 (b) which choice of law rules apply (as each court applies its own choice of law rules, which then determine among other things the effectiveness of any choice of law agreement entered into by the parties); and (c) which rules of mandatory law and conception of public policy are applicable, as restrictions on the applicable law designated by choice of law rules (as these are again, at least principally, matters in which each forum applies its own law). The choice of forum is thus also, among other things, a choice of law – although not ordinarily of the main rules that are to be applied to govern substantive legal questions arising in a dispute or relationship. In the context of choice of law, the development of party autonomy has often been justified on the basis that it reflects the parties’ freedom of contract. Peter Nygh, for example, author of the seminal work Autonomy in International Contracts (1999), advocated party autonomy in private international law as an automatic implication of freedom of contract. In his 1995 Hague courses, Nygh argued that ‘[a]s part of freedom to determine the terms and conditions of their contract, the parties also have the freedom to choose the applicable law’.8 Symeon Symeonides, 5 See JH Beale, ‘What Law Governs the Validity of a Contract. I’ (1909) 23 Harvard Law Review 1 and JH Beale, ‘What Law Governs the Validity of a Contract. III. Theoretical and Practical Criticisms of the Authorities’ (1910) 23 Harvard Law Review 260; American Law Institute, Restatement of the Law: Conflict of Laws (American Law Institute Publishers 1934) ss 332 and 358 (applying a combination of the laws of the place of contracting and place of performance). 6 See, eg, JH Beale, A Treatise on the Conflict of Laws (Baker Voorhis and Co 1935) 1079; see further discussion in A Mills, ‘The Identities of Private International Law – Lessons from the US and EU Revolutions’ (2013) 23 Duke Journal of Comparative and International Law 445; HE Yntema, ‘Contract and Conflict of Laws: “Autonomy” in Choice of Law in the United States’ (1955) 1 New York Law Forum 46, 54ff. 7 Although different views may be adopted as to what is procedural and what is substantive for these purposes: see R Garnett, Substance and Procedure in Private International Law (OUP 2012). 8 PE Nygh, ‘The Reasonable Expectations of the Parties as a Guide to the Choice of Law in Contract and in Tort’ (1995) 251 Recueil des Cours 1, 297; see similarly PE Nygh, Autonomy in International Contracts (OUP 1999) 8.

Choice of Court and Choice of Law Agreements  365 the leading international comparativist in choice of law, similarly argued in a recent work that ‘[t]he principle of party autonomy is simply the “external” side of a domestic law principle, usually referred to as “freedom of contract”’.9 It has thus often been argued that choice of law (and choice of court) agreements should be given effect simply as a matter of contract law – that such clauses are ordinary contractual terms, and enforcing them follows simply and directly from the policy of enforcing contractual bargains. Historically, the adoption of party autonomy often appeared to follow as a ‘natural’ consequence of freedom of contract – perhaps as its ‘internationalisation’.10 As discussed further in section II.B, to the extent that state regulation of contract law was viewed as taking the form of ‘default rules’ from which the parties could opt out through the adoption of specific terms in their contract (a view that became highly influential in the nineteenth century common law),11 it became more difficult to resist the argument that the selection of a foreign legal system with different default rules should be given similar effect. A choice of foreign law would, in this conception, just be a shorthand way of drafting the parties’ preferred terms. The comments to the Second Restatement rule giving effect to choice of law clauses, for example, explain that ‘Giving parties this power of choice is also consistent with the fact that, in contrast to other areas of the law, persons are free within broad limits to determine the nature of their contractual obligations.’12

II.  The Distinction between Private International Law Party Autonomy and Contractual Party Autonomy The essential argument of this chapter is that the conflation of contractual freedom, which we can call contractual party autonomy, and freedom to choose the applicable law, which we can call private international law party autonomy, is a mistake. This section outlines the distinction between these concepts, and the following section explores the consequences that follow.

9 SC Symeonides, Choice of Law (OUP 2016) 361. See similarly M Keyes, ‘Jurisdiction under the Hague Choice of Courts Convention: Its Likely Impact on Australian Practice’ (2009) 5 Journal of Private International Law 181, 182 (‘[t]he fundamental justification [for party autonomy in private international law] is that the parties’ freedom to contract should be respected by enforcing their agreements’). 10 H Muir Watt, ‘“Party Autonomy” in international contracts: from the makings of a myth to the requirements of global governance’ (2010) 6 European Review of Contract Law 250, 257 (‘according to traditional discourse, the empowerment of private actors to choose the law governing their relationship is a natural consequence, and indeed the mirror image, of freedom of contract in the domestic sphere’). 11 See generally, eg, PS Atiyah, The Rise and Fall of Freedom of Contract (OUP 1985). 12 American Law Institute, Restatement of the Law Second: Conflict of Laws 2d (American Law Institute Publishers, 1971) s 187 comment (e).

366  Alex Mills

A.  Distinguishing Two ‘Party Autonomies’ The starting point here is apologetically but necessarily basic. Freedom of contract (referred to here as contractual party autonomy) means the legal freedom of the parties to enter into their preferred forms of agreement. It governs the limits of legally effective private agreements within a legal order, even for purely domestic contracts – it is what contract lawyers generally mean when they refer to the limits of party autonomy.13 Those legal limits have to come from a system of private law, and they are different in different legal systems. The degree of contractual party autonomy is, as a consequence, determined by the law applicable to the contract. The law applicable to a contract is not necessarily determined by an exercise of party autonomy – sometimes parties do not expressly or implicitly choose a governing law, and the applicable law is determined through an ‘objective’ choice of law rule based on various connecting factors.14 The law thereby selected determines the scope of contractual party autonomy, because it governs the validity of the contract and its terms. Even in this scenario (without a party choice of law), there is something special about private international law rules, because they claim a higher status than other rules of law. They determine the applicability of those other rules – to put this another way, private international law is about the regulation of regulation.15 Where there is a choice of law agreement and it is given effect, that is an exercise of private international law party autonomy, which then (instead of objective choice of law rules) determines the law that sets out the limits of contractual party autonomy. It therefore cannot be that party freedom to choose the applicable law is justified by or a matter of party freedom of contract. The choice of law question determines the extent of freedom of contract, it is not determined by it. To put this another way, there is something special about the private international law terms in contracts (just as there is something special about private international law rules in general) – they are not ordinary contractual terms, because they claim a higher status than those terms or, to put this another way, they operate at a higher level of analysis. Party autonomy in private international law is a freedom to choose between different freedoms – the diverse degrees of contractual autonomy available in different national systems of private law. It is not a matter of freedom of contract but of freedom of freedom of contract. The distinction between party autonomy in private international law and party autonomy in contract law is a fundamentally important one. The claim that ‘[a]s part of freedom to determine the terms and conditions of their contract, the parties also have the

13 For a comparative study, see GA Berman (ed), Party Autonomy: Constitutional and International Law Limits in Comparative Perspective (Juris 2005). 14 See, eg, Rome I Regulation, Art 4. 15 A Mills, The Confluence of Public and Private International Law (CUP 2009) 19ff.

Choice of Court and Choice of Law Agreements  367 freedom to choose the applicable law’16 misleadingly conflates these two distinctive forms of autonomy.17 One prominent advocate of party autonomy has implicitly recognised these concerns, in justifying party autonomy on the further basis that there is ‘a general principle of freedom of contract which allows the parties to choose the applicable legal system and which precedes national law’.18 But this justification is somewhat difficult to unpack. Public international law recognises a conception of ‘general principles of law’ common to the major legal systems of the world, but this normally serves a gap-filling function in international law rather than acting as an independent source of legal obligation, let alone as an obligation on private parties.19 Since party autonomy was long rejected by many legal systems (and is still rejected by some), it is also difficult to claim that it is a general principle of contract law ‘which precedes national law’, whatever that might mean. Other scholars have suggested that party autonomy has gained the status of a ‘rule of customary law’,20 although again this is challenging to reconcile with the historical (and some continuing) inconsistency of actual state practice. While these claims are unconvincing they are nevertheless illuminating, as they show the conceptual distinctiveness of party autonomy in private international law from rules of contract law, and the consequential difficulties that justifying party autonomy presents. In these arguments, the illogicality of justifying party autonomy as an aspect of freedom of contract is recognised – private international law party autonomy cannot be justified on the basis of a national doctrine of contract law, because it is a higher-order question – which leads to claims that it is justified by a higher-order rule (as some kind of ‘general principle’, perhaps as part of international or natural law, although the claims are not necessarily that precise). The question of the justifications for private international law party autonomy is explored further in section III.A.

B.  Relationship between the Two ‘Party Autonomies’ The need to draw a careful distinction between the doctrines of contractual party autonomy and private international law party autonomy does not imply that they

16 Nygh, ‘Reasonable Expectations’ (n 8) 297; see similarly Nygh, Autonomy (n 8) 8. 17 Note also similarly A Briggs, Agreements on Jurisdiction and Choice of Law (OUP 2008) 12; M Zhang, ‘Party Autonomy and Beyond: An International Perspective of Contractual Choice of Law’ (2006) 20 Emory International Law Review 511, 552. 18 M Lehmann, ‘Liberating the Individual from Battles between States: Justifying Party Autonomy in Conflict of Laws’ (2008) 41 Vanderbilt Journal of Transnational Law 381, 390. 19 See generally, eg, R Yotova, ‘Challenges in the Identification of the ‘General Principles of Law Recognized by Civilized Nations: The Approach of the International Court’ (2017) 3 Canadian Journal of Comparative and Contemporary Law 269. 20 AF Lowenfeld, ‘International Litigation and the Quest for Reasonableness’ (1994) 245 Recueil des Cours 1, 256 (‘support of party autonomy is by now so widespread that it can fairly be called a rule of customary law’). See also Nygh, Autonomy (n 8) 45.

368  Alex Mills are unrelated. As noted, the greater the degree of contractual autonomy in a legal system, the more the function of contract law may be considered to be merely providing default rules to govern the relations between the parties. If this were the only function of contract law (if parties had complete freedom of contract, and the function of contract law were only to fill gaps in their agreement) then choosing a foreign legal order to govern those relations could be viewed as merely an efficient means of opting out of one set of default rules in favour of a different set of default rules. There would appear to be no difference in principle between an exercise of contractual autonomy and private international law party autonomy, because both would merely involve selecting the terms of the contract, albeit through different drafting methods. Thus, the broader the degree of party autonomy allowed by a legal system, the more difficult it is to argue that the parties should be precluded from choosing a foreign legal order to govern their legal relations. There are two reasons why this is not, however, a complete account of the issues. The first is that this analysis only works to the extent that each potentially applicable legal system takes the same view of freedom of contract. If ‘local’ law provided for extremely broad freedom of contract, and the parties chose a foreign system of law which did not, this would not (or not only) have the effect of importing default terms from that system of law but (also) potentially of invalidating the contract or some of its terms. The consequences of a choice by the parties of a system of law that (apparently) renders their contract invalid is a rather complex issue, and a full analysis of this is beyond the scope of this chapter.21 The second and more decisive reason is that providing default rules is not the sole function of contract law in any legal system – rules of contract law limit the exercise of contractual autonomy, as well as support it, through invalidating certain types of contracts, through striking out unfair, unconscionable or undesirable terms, and sometimes through imposing mandatory terms. A choice of foreign law thus not only fills gaps in the contract, equivalently to an exercise in drafting, but also selects a law under which the obligations articulated in the contract may be shaped or constrained. Although functionally equivalent effects could at least generally be achieved through drafting (the parties could, for example, simply leave out clauses that would be invalidated under their chosen applicable law), the submission of a contract to a legal system is different because it involves accepting a higher authority over the terms of the contract than the parties themselves – an authority whose position might even change after the terms of the contract have been settled between the parties. Contract law of course also regulates some ‘unfair’ (potential) contracts through finding that they lack genuine consent, because of concerns such as mistake, duress and undue influence. This is a distinct issue because it relates to the question of formation of the contract – it is not concerned with freedom of contract but with freedom to contract. It thus raises a further set of complex questions that are also

21 See

Mills, Party Autonomy (n 4) 387ff.

Choice of Court and Choice of Law Agreements  369 beyond the scope of this chapter, including the particularly difficult question of what law should govern the validity of a (potential) contract that includes a choice of law clause, including the validity of the choice of law clause itself.22 Specialised rules are generally required to deal with these issues – these rules view the question of consent as still ordinarily governed by the law chosen by the parties, but this is subject to an exception to avoid a party’s becoming subject to a contract if the application of a chosen foreign law to govern the issue of consent would be unfair.23 The basic point for present purposes is that the conceptual links between contractual party autonomy and private international law party autonomy do not justify collapsing the distinction between them. This is not to deny that a choice of law agreement or choice of court agreement is or can be analysed as a contract (although the point is not entirely without contention),24 and that as a contract it is bound by a system of contract law. It is rather to emphasise that the choice involved in a choice of law or choice of court agreement is distinctive from the normal choices of substantive terms involved in contract law, which take place within the scope of contractual party autonomy.25

III.  Consequences of the Distinction This section sets out three important consequences that follow from the distinction between private international law party autonomy and contractual party autonomy. The first is that private international law party autonomy needs a different justification than contractual party autonomy. The second is that the limitations on private international law party autonomy are conceptually distinct from the limitations on contractual party autonomy. The third is that it is equally important to draw a clear and related distinction between a choice of law and an act of incorporation by reference of rules of foreign (or non-state) law.

A.  Justifications for Party Autonomy The preceding analysis has argued that private international law party autonomy is distinct from contractual party autonomy – as a consequence, freedom of contract cannot provide the justification for giving effect to private international law party autonomy. Indeed, the two types of freedom could conceivably be viewed as in 22 See, eg, Mills, Party Autonomy (n 4) 379ff. 23 See, eg, Rome I Regulation, Art 10. 24 See discussion in Mills, Party Autonomy (n 4) chs 3 and 7. 25 See M Hook, The Choice of Law Contract (Hart Publishing 2016) for an exploration of the consequences of this for choice of law agreements, viewing them as subject to a ‘framework that fuses principles of choice of law and contract’ (ibid 12).

370  Alex Mills tension with one another. If a legal system were strongly committed to the particular degree of freedom of contract provided for under its contract law, there is even perhaps an argument that it should not allow the parties to choose another system of contract law with a different (higher or lower) degree of freedom of contract. A choice, at least, would need to be made between the commitment to that particular degree of freedom of contract and the special exercise of freedom in the form of a choice of law agreement – the latter might, for example, invalidate some of the terms of the contract that would be valid under the legal system’s own law, or allow for an escape from the limitations on freedom of contract within that legal system. So why, then, allow the parties to ‘contract out’ of the rules governing contract law itself – what are the justifications for private international law party autonomy? It might be argued that choice of law and choice of forum agreements are or should be enforced simply because they are contractual terms to which the parties have agreed, an approach that has perhaps been particularly associated with the common law.26 But this is question-begging – an agreement is (at least conventionally) only a contract if a state has decided to make it legally enforceable. Contracts exist not only to serve the parties, but also to serve broader public interests, such as promoting the efficient functioning of economic relations by putting state enforcement resources behind private promises. States choose whether to view expressions of party autonomy (choice of forum or choice of law clause) as enforceable contracts – indeed, such agreements were historically often not enforced, and some states still treat at least some types of these agreements as unenforceable (although the evident trend in private international law is toward giving more rather than less effectiveness to exercises of party autonomy).27 When such agreements were not enforceable, any argument that they should be enforced simply because they were contractual terms would have been rapidly dismissed – on the contrary, they were evidently not viewed as valid contractual terms because they were not enforceable. Choice of law and forum agreements are, in any case, something more than ordinary contractual terms – they are special, because they purport to take the parties outside (or bring the parties inside) the institutions and governing law of the legal system evaluating them and which confers on them the status of contract.28 They are private legal acts that seem, almost paradoxically, to claim a status higher than the laws of the states that give such private acts their legal character. This distinctiveness might be (and historically was at times) viewed as a good reason to treat such agreements differently from other contracts (or contractual terms) – to refuse to enforce them. So the question remains, why choose, in principle, to make choice of law and forum agreements enforceable?

26 See generally, eg, Briggs, Agreements on Jurisdiction (n 17); Yntema, ‘Contract and Conflict of Laws’ (n 6) 47 (‘[i]n cases of conflict of laws, it is thus a natural inference from the principle of contract that the law contemplated by the parties should apply in situations where their rights are to be measured by their agreement’). 27 See Mills, Party Autonomy (n 4) ch 2 for a detailed historical analysis. 28 See, eg, Northwestern National Insurance Co v Donovan, 916 F 2d 372 (7th Cir 1990).

Choice of Court and Choice of Law Agreements  371 Under the modern common law, private international law has been primarily viewed as concerned with private rights – as a flexible and frequently discretionary set of rules focused on doing justice in individual cases and meeting the needs of the parties. Under this framework, it is usually argued that party autonomy is respected because it corresponds with party expectations. It is important to recognise that agreements on choice of law and jurisdiction may indeed form part of the bargain between the parties – that a preferred court or governing law may be selected as part of a trade-off involving substantive contractual terms, and refusing to enforce part of that trade-off would risk unbalancing the contract. Such considerations have again been particularly prominent under the common law approach to party autonomy.29 There is, however, something circular about this reasoning. If well-advised parties knew that choice of court and choice of law clauses were unenforceable, they would not negotiate for them as terms of a contract – they would not trade away other potential contractual rights in exchange for their supposed benefits. It is only because such clauses are generally enforceable that the parties view them as desirable. Obviously what the law should not do is to hold out that such clauses will be enforceable and then fail to enforce them, but of course no one suggests such an approach. The argument that choice of court and choice of law agreements should be enforced simply because they are understood by the parties to be part of their bargain thus risks circularity – it justifies giving effect to such agreements, because the law provides and thus parties expect that such agreements will indeed be given effect. Part of the answer to this issue is that when party expectations are referred to in this context, this may not mean the actual expectations of the parties, which are likely to differ in a dispute, and which in any case are or should be shaped by existing rules of private international law. If private international law party autonomy were clearly not accommodated in the law, well-informed parties would not expect their wishes as to jurisdiction or applicable law to be taken into consideration, and their expectations would be met by failing to give effect to any such wishes. As argued previously elsewhere,30 party expectations in this context might instead mean one of two possible things. First, it might mean the mere (subjective) wishes of the parties – viewing choice of law or forum agreements as an expression of those wishes. But this leaves us with a further secondary question as to why such wishes should be fulfilled, and a general sense that the law should be in the business of wish-fulfilment is not a satisfactory justification because the law will often not enforce the wishes of parties, such as where there are opposing public policies or interests. Giving effect to private international law party autonomy at least puts such public policies and 29 See, eg, The Bremen v Zapata Off-Shore Co., 407 US 1, 14 (1972) (‘There is strong evidence that the forum clause was a vital part of the agreement, and it would be unrealistic to think that the parties did not conduct their negotiations, including fixing the monetary terms, with the consequences of the forum clause figuring prominently in their calculations.’). 30 See further Mills, Confluence (n 15) 8ff.

372  Alex Mills interests at risk, if it leads to the evasion of rules of law that would otherwise apply, as discussed further in section III.B. Second, ‘party expectations’ might mean the (more objective) reasonable expectations of a reasonable party, if there were no rules of private international law – viewing choice of law or forum agreements as giving rise to a legitimate and reasonable expectation, when considered from behind a ‘veil of ignorance’ as to the present legal position. Such an argument may leave us little the wiser as to why such agreements do give rise to expectations that are legitimate and reasonable, except perhaps based on the idea that expectations created by agreements should generally be considered legitimate, because agreements should generally be enforceable – an intuition commonly expressed through the Roman (and international) law maxim ‘pacta sunt servanda’. But once again this comes up against the public policy concerns discussed in section III.B – some agreements are not enforceable, and for good reason. Under either of these approaches, the appeal to party expectations is revealed to be really little more than an appeal to an intuition that, in the absence of compelling reasons, the state should generally enforce rather than obstruct the wishes or freely-made agreements of private parties. But this proposition is in tension with the idea that each legal system identifies particular limits on contractual autonomy that it normally imposes as part of its contract law. Even if choice of court and choice of law clauses are understood to be intuitively ‘binding’ on the parties, they do not bind the court or courts that must determine their effectiveness, or at least do not do so for the same reasons. Why, then, should the courts give effect to the agreement between the parties, instead of the rules of law (including choice of law rules) that form part of their own legal order? Providing a full account of the justifications for party autonomy in private international law is beyond the scope of this chapter,31 but a range of types of justifications may be identified. A first approach might be described as justifying party autonomy based on a form of deontological libertarianism. This is because it starts from the position that the autonomy of the individual is inherently valuable, and it is an interference with this autonomy, not the autonomy itself, that requires justification. A choice of law or forum may then be viewed as ‘a necessary expression of individual autonomy’,32 or as a product of ‘the natural will of the individual along with the corollary right to craft such will by virtue of the individual’s innate freedom’.33 There is a long philosophical tradition that can be drawn on in support of this perspective, in which the work of Immanuel Kant may be singled out as particularly influential in positing autonomy as a moral good in itself in facilitating individual self-fulfilment.34 31 For further detail, see Mills, Party Autonomy (n 4) ch 2. 32 Lehmann, ‘Liberating the Individual’ (n 18) 417. 33 J Basedow, ‘The Law of Open Societies – Private Ordering and Public Regulation of International Relations’ (2013) 360 Recueil des Cours 9, 196. 34 See further generally ibid 196ff; JB Schneewind, The Invention of Autonomy: A History of Modern Moral Philosophy (CUP, 1997). For a variation based on the work of John Rawls, see JA Pontier, ‘The Justification of Choice of Law: A Liberal-Political Theory as a Critical and Explanatory Model, and the Field of International Consumer Transactions as an Example’ (1998) 45 Netherlands International Law Review 388.

Choice of Court and Choice of Law Agreements  373 Peter Nygh argued, for example, that ‘[f]or idealists [party autonomy] is an expression of the human right of individuals to arrange their personal and economic lives as they see fit, subject only to such constraints as are necessary to maintain public order and prevent the exploitation of the weak’.35 Where a legal relation crosses borders, it may particularly be argued that no state has an overriding claim to regulate that relationship, and that party autonomy emerges as a residual claim: From a legal perspective, too many contradictory rules with equal right to be followed create a void. Thus, we are back in a state of nature, or more precisely, in a ‘modern state of nature’, where there is no objectively applicable law – paradoxically because of an abundance of law. In this context, the parties regain their residual power to regulate their relationships. Party autonomy means nothing more than that people can take care of their own affairs.36

In essence, this approach arguably involves a recognition of a form of individual sovereignty alongside, or perhaps even prior to, the sovereignty of the state.37 An alternative justification for party autonomy may be found in arguments focusing on the beneficial effects of recognising individual choice (rather than the inherent value of individual freedom), which might be characterised as a form of private consequentialist libertarianism. It may be argued that individuals should be free to choose the forum or law for their disputes not because of any inherent autonomy they possess, but because giving them that freedom has useful consequences for them. For example, one often cited reason why choice of forum and choice of law agreements might be enforced is the complexity and difficulty of the rules that apply in the absence of any party choice.38 An exclusive choice of court agreement in particular may also have the benefit of replacing multiple potential courts with a single forum, making jurisdiction more predictable. In both choice of court and choice of law, divergences in private international law rules between states can multiply the uncertainties. As the Commentary to the Hague Principles on Choice of Law in International Commercial Contracts 2015 notes, ‘Determination of the law applicable to a contract without taking into account the expressed will of the parties to the contract can lead to unhelpful uncertainty because of differences 35 Nygh, Autonomy (n 8) 258. 36 Lehmann ‘Liberating the Individual’ (n 18) 414 (footnote omitted – but strikingly, this is to the libertarian philosopher Robert Nozick, Anarchy, State, and Utopia (1974)); note also the citations to Kant, Hayek and Nozick, ibid 418). See also similarly Basedow, ‘The Law of Open Societies’ (n 33) 200. 37 See, eg, Zhang, ‘Party Autonomy’ (n 17) 553 (‘the importance of granting autonomy to the parties lies with the belief that the will of the parties is sovereign in the field of contract’). 38 See, eg, F Maultzsch, ‘Party autonomy in European private international law: uniform principle or context-dependent instrument?’ (2016) 12 Journal of Private International Law 466, 476; Basedow, ‘The Law of Open Societies’ (n 33) 193; CI Nagy, ‘What Functions May Party Autonomy Have in International Family and Succession Law? An EU Perspective’ (2012) 30 Nederlands Internationaal Privaatrecht 576, 579 (‘[o]ne of the merits of party autonomy is that the question of the applicable law is not left to the intricacies of conflicts law’); Lehmann, ‘Liberating the Individual’ (n 18) 385 (‘More and more states allow parties to cut the “Gordian knot” of conflict of laws by choosing the applicable law themselves.’); Nygh, Autonomy (n 8) 3.

374  Alex Mills between solutions from State to State.’39 If the rules on choice of law or choice of court are otherwise difficult to predict, giving effect to party autonomy allows the parties to simplify them and thus to order their relationship. A further potential consequence of party autonomy in choice of forum and law is that it allows parties to properly cost and indeed reduce their litigation risks, which is in turn likely to lead to more efficient dispute settlement, as commercial parties are less likely to litigate rather than settle a dispute if their legal positions are clear.40 Giving effect to party autonomy may also allow parties to choose a law or court that is very familiar to them, or which is most effectively or efficiently adapted to their particular relationship or to the types of disputes that they anticipate might arise between them, again reducing the costs and risks of dispute settlement.41 Giving the parties a choice over the law or forum that governs their relationship may be further justified not only as a practical matter of efficiency, but also as a matter of legitimacy. It may be argued that for a cross-border relationship or dispute, party autonomy provides the most legitimate way of determining which court or system of law should have regulatory power, because its legitimacy is derived from the consent of the parties themselves, bypassing the competing claims of potentially applicable legal orders.42 Party autonomy in private international law may also be justified through arguments that focus more on the systemic effects of recognising party autonomy, rather than its effects on individuals – a distinct form of public consequentialist libertarianism. It may be argued that individuals should be free to choose the forum or law for their disputes because this has benefits not necessarily for them as individuals, but rather for the society or legal system as a whole (this might be framed as a benefit for a state, a region or for the global economy at large). For example, it may be argued that enforcing choice of forum or choice of law agreements will encourage cross-border activity by reducing litigation risk, with the increase in economic activity leading to greater public welfare.43 Peter Nygh argued, for example, that ‘[f]or pragmatists [party autonomy] may be seen as the necessary accompaniment of the globalisation of international trade and commerce’.44 This distinct form of instrumental or ‘utilitarian’ justification45 is particularly prevalent 39 Hague Conference on Private Intentional Law, Principles on Choice of Law in Intentional Commercial Contracts (The Hague Conference on Private Intentional Law Permanent Bureau, 2015) [1.2]. 40 See generally, eg, G Rühl, ‘Party Autonomy in the Private International Law of Contracts: Transatlantic Convergence and Economic Efficiency’ in E Gottschalk et al (eds), Conflict of Laws in a Globalized World (CUP 2007) 153; MJ Whincop and M Keyes, Policy and Pragmatism in the Conflict of Laws (Ashgate 2001). 41 See, eg, Rühl, ‘Party Autonomy’ (n 40); AT Guzman, ‘Choice of Law: New Foundations’ (2002) 90 Georgetown Law Journal 883; EA O’Hara and LE Ribstein, ‘From Politics to Efficiency in Choice of Law’ (2000) 67 University of Chicago Law Review 1151. 42 L Brilmayer, ‘Rights, Fairness, and Choice of Law’ (1989) 98 Yale Law Journal 1277, 1298ff. 43 Basedow, ‘The Law of Open Societies’ (n 33) 194. 44 Nygh, Autonomy (n 8) 258. 45 Watt, ‘Party Autonomy’ (n 10) 256 (‘its dominant justification is essentially utilitarian, linked to the needs of international trade’); see further H Muir Watt, ‘Conflicts of Laws Unbounded: The Case for a Legal-Pluralist Revival’ (2016) 7 Transnational Legal Theory 313; R Wai, ‘Transnational Liftoff and Juridical Touchdown: The Regulatory Function of Private International Law in an Era of Globalization’ (2002) 40 Columbia Journal of Transnational Law 209.

Choice of Court and Choice of Law Agreements  375 in the context of European Union (EU) private international law rules, which have been developed in a context focused on how the rules can benefit the European internal market by promoting cross-border economic activity, rather than on any benefit to the parties themselves. Some recent expressions of this type of justification have articulated it as an argument that party autonomy should be recognised in order to create a ‘law market’46 regarding both institutions and procedural law (choice of forum) and substantive law (choice of law). At least in the commercial context, party autonomy may thus be inherently linked to both the globalisation of private markets in practice, and the internationalisation of the theory of market regulation, as domestic freedom of contract is transposed to the higher-level question of a choice between the institutions and substantive rules of different national legal orders.47 Whether and to what extent private international law should be put at the service of such interests is at the heart of disputes over the foundations and limits of party autonomy. In the context of choice of law, the public-systemic justification for party autonomy advocates a form of ‘law shopping’. The theory is that the ability for parties to select among a variety of applicable laws should lead to ‘regulatory competition’48 – the existence of a law market creates incentives for legal systems to improve their effectiveness and thus their attractiveness to parties, in order to increase their scope of application. (It is less clear whether this theory is supported by practice.49) As already noted, a choice of law will also allow the parties to choose a law with which they are familiar (perhaps they might even have litigated similar issues under it previously) and which is appropriate for their relationship, reducing transaction costs and again leading to market efficiencies. These different theoretical justifications for party autonomy in private international law will, to a great extent, support one another – each offers a justification for basic rules of party autonomy. This is not least because the deontological and consequentialist forms of libertarianism may be viewed as sharing some common foundations. However, the different justifications for party autonomy also each require a slightly distinct analytical focus, looking variously at the sovereignty of

46 Watt, Party Autonomy’ (n 10) 258 (‘[b]y allowing parties to cross jurisdictional barriers unhindered, the principle of free choice generates a competitive market for legal products and judicial services’). See further, eg, H Eidenmüller, ‘The Transnational Law Market, Regulatory Competition, and Transnational Corporations’ (2011) 18 Indiana Journal of Global Legal Studies 707; EA O’Hara and LE Ribstein, The Law Market (OUP, 2009); Rühl, ‘Party Autonomy’ (n 40). 47 Watt, Party Autonomy’ (n 10). 48 See further references in, eg, Basedow, ‘The Law of Open Societies’ (n 33) 107ff; Eidenmüller, ‘Transnational Law Market’ (n 46); Rühl, ‘Party Autonomy’ (n 40); Mills, Confluence (n 15) 202ff. 49 S Vogenauer, ‘Regulatory Competition through Choice of Contract Law and Choice of Forum in Europe: Theory and Evidence’ (2013) 21 European Review of Private Law 13 (concluding at 77 that ‘the emergence of regulatory competition in the relevant fields cannot be observed anywhere in the real world. There is no empirical evidence that parties choose the applicable contract law or the forum on the basis of the quality of specific legal rules and that lawmakers improve the quality of these rules to attract more users.’). For further empirical analysis, see G Cuniberti, ‘The International Market for Contracts: The Most Attractive Contract Laws’ (2014) 34 Northwestern Journal of International Law and Business 455.

376  Alex Mills the individual, whether recognising individual choice has efficiency benefits for the parties and the resolution of their disputes, and the systemic benefits of regulatory competition. They may therefore be taken to provide slightly distinct answers to the question of exactly when party autonomy should be given effect – under what circumstances party agreements on a forum or applicable law should or should not be enforced.

B.  Limitations on Party Autonomy A second important consequence that follows from the distinction between contractual party autonomy and private international law party autonomy is that the limitations on each form of party autonomy are different. Within each system of contract law, there are constraints on freedom of contract, which may be either positive or negative in form. The positive form generally consists of mandatory terms out of which the parties cannot contract, while the negative form generally consists of rules of public policy that invalidate certain prohibited contractual terms.50 But these are rules of contract law, dependent on the law that is applicable to the contract. The application of these constraints to private international law issues (such as the question of the validity or effectiveness of a choice of law agreement) would thus pre-judge the applicable law question. In private international law, a distinct and narrower category of mandatory rules and public policy is thus recognised.51 Mandatory rules in private international law are rules that apply regardless of the law applicable to the contract, and the private international law conception of public policy is used to disapply rules of foreign law that might impose or invalidate contractual terms.52 These constraints are an important recognition of countervailing public interests that may not be served through private international law party autonomy, but they need to be understood in light of the underlying justifications for the power they are constraining – which, as argued in section III.A, are not the same for private international law party autonomy as they are for contractual party autonomy. They must be interpreted restrictively, otherwise they would risk undermining the essential purpose(s) of private international law party autonomy – if foreign law were disapplied just because it is different from local law, the choice of law by the parties would be neutralised. If a court were to apply its own contractual mandatory rules to invalidate contractual terms that were valid under the law chosen by the parties, private international law party autonomy would effectively be reduced to contractual

50 See generally GA Bermann (ed), Party Autonomy: Constitutional and International Law Limits in Perspective (Juris 2005); Nygh, Autonomy (n 8) 28ff. 51 See generally Mills, Party Autonomy (n 4) 476ff. 52 See generally A Mills, ‘The Dimensions of Public Policy in Private International Law’ (2008) 4 Journal of Private International Law 201.

Choice of Court and Choice of Law Agreements  377 party autonomy, and no more. The parties’ freedom to choose foreign law would be constrained by the degree of contractual party autonomy available in local law. If private international law party autonomy is truly to be adopted then even some non-derogable rules of contract law should not be applicable if the parties have chosen a foreign system of law, as they should be viewed as having chosen a different set of limits on their contractual party autonomy. However, a distinct and narrower set of mandatory rules or public policy interests may nevertheless still play a role in policing the limits of private international law party autonomy, as limitations on that distinctive form of freedom. A possible qualification to this arises where the contractual relationship is internal to a single legal system – in that context the argument may be made that the public interests of that system (in terms of its contractual mandatory rules) should be fully respected, even if the parties have chosen foreign law. This is indeed the position adopted as a matter of EU law.53 But outside that context, the conception of mandatory rules and public policy as more narrowly defined in the context of private international law reflects the fact that in a cross-border situation, the regulatory claim of any connected state is itself attenuated.54 The source of both mandatory rules and public policy in the private international law sense is nevertheless not the applicable law governing the contract (which would equally pre-judge the question), but rather the law of the forum in which the validity or effectiveness of the choice of law agreement is being determined. This establishes an important link between party autonomy in choice of court and party autonomy in choice of law. Because the constraints on party autonomy in choice of law depend on the forum in which the dispute is being heard, which may be determined by the effectiveness of a choice of court agreement, it is a choice of court agreement that may determine which state’s public policy or mandatory rules limit the exercise of party autonomy in choice of law. This raises the important and difficult question of whether a choice of court agreement should sometimes be invalidated because it would have the effect of disapplying the mandatory rules of the forum – the case would be litigated in a foreign court that would not apply those rules. This approach has been exceptionally adopted as part of the common law, particularly in Australian courts, but it does not seem possible under EU law because there is no provision in the Brussels I Regulation for a jurisdiction agreement to be invalidated as an ‘evasion of law’.55

53 See Rome I Regulation, Art 3(3) and (4). 54 A similar constraint functions in relation to public policy – see Mills, ‘Dimensions of Public Policy’ (n 52). 55 See further Mills, Party Autonomy (n 4) 481ff. As noted, however, if the choice of forum is within the EU then in some cases the application of foreign mandatory rules may be required under the Rome I Regulation, and thus the jurisdiction agreement may be ineffective to evade the mandatory rules of a foreign forum.

378  Alex Mills

C.  Distinction between Choice of Law and Incorporation by Reference A third consequence of the important distinction between contractual party autonomy and private international law party autonomy is the need to maintain a careful conceptual distinction between the process of choice of law and the incorporation by reference of rules of law. Again, if the function of contract law were solely to provide ‘default rules’ to fill gaps in the parties’ agreement, this distinction would collapse. The choice of foreign law would simply be an efficient way of adopting a different set of default rules, which could alternatively be adopted by including the text of those rules as contractual terms – private international law party autonomy would be merely another contractual drafting technique. However, as noted, the function of contract law is not limited in this way, and the selection of a foreign law to govern a contract may lead to terms being mandatorily included, or invalidated. This distinction opens up the space for the practice known as ‘incorporation by reference’, whereby terms of the contract are defined by reference to some external source, to be applied also to foreign legal sources. Incorporation by reference is perhaps most commonly used to incorporate a set of general terms and conditions, usually belonging to one of the parties, as terms of a particular contract. But it may equally be used to incorporate terms from other sources that have not been prepared by the parties, which might include foreign rules of law, without this necessarily involving a selection of that system of law as a governing law for the contract. The parties might agree, for example, to comply with the terms of a statute that forms part of the ‘home’ legal system of one of the parties – that party may even be required by the statute to impose that contractual obligation. But ultimately a clause of this type is merely a shorthand way of drafting the contract, equivalent in legal effect to cutting and pasting the text of those rules into the pages of the contract. The effectiveness of an effort to incorporate terms by reference will thus be determined by the applicable law, under which some of the terms that are (purported) to be incorporated may be invalidated or overridden. Even if the source of the terms incorporated by reference is a foreign system of law, that system of foreign law does not necessarily determine their validity. On its own, an incorporation by reference of rules of foreign law may determine the content of the contract – it is an exercise of contractual autonomy – but not (necessarily) the law that governs the contract – it is not an exercise of private international law party autonomy. This is, once again, not to deny that these concepts are related, nor to claim that they are always clearly distinguished in practice. For example, parties who incorporate by reference terms from a particular legal order may well have in mind that those terms will be interpreted and applied consistently with the practice in that order. The inclusion of statutory provisions as part of a contract is thus a strong indicator of a possible implied choice of law agreement, and it is treated as such by the courts.56

56 See

Mills, Party Autonomy (n 4) 314ff.

Choice of Court and Choice of Law Agreements  379 But the link between the two is evidentiary rather than logical – the incorporation by reference of statutory terms from a national legal order does not necessarily imply a choice of that system of law to govern the contract, and there is nothing incoherent about an express choice of a different law, or the courts finding that no implied choice can be identified in the circumstances. An illustration of the confusion between these concepts in practice may be taken from an old US Supreme Court decision, Pritchard v Norton.57 The Court was concerned with the validity of an appeal bond entered into in New York, relating to litigation in Louisiana. In analysing the choice of law question, the Court noted that in case of contract, the foreign law may, by the act and will of the parties, have become part of their agreement, and in enforcing this, the law of the forum may find it necessary to give effect to a foreign law which, without such adoption, would have no force beyond its own territory.58

Thus far, this appears to be recognition of an exercise of contractual autonomy through incorporation by reference. Terms of foreign law have become part of the agreement between the parties, and so must be given effect in order to enforce the agreement. The Court, however, went on to argue that: This, upon the principle of comity for the purpose of promoting and facilitating international intercourse and within limits fixed by its own public policy, a civilized state is accustomed and considers itself bound to do, but, in doing so, nevertheless adheres to its own system of formal judicial procedure and remedies. And thus the distinction is at once established between the law of the contract, which may be foreign, and the law of the procedure and remedy, which must be domestic and local.59

This analyses the situation as an example of private international law party autonomy – foreign law is recognised as a matter of comity, not simply as a matter of giving effect to the contractual terms. Foreign contract law is viewed as a separate legal order, and if validly selected applies as the substantive governing law of the contract (not just as a source of its terms), while the procedural rules remain a matter of forum law. The distinction between a choice of law and the incorporation by reference of rules of law is particularly important in the context of non-state law, because choice of law rules generally prohibit the selection of ‘non-state law’ to govern a contract.60 As noted, the incorporation by reference of rules of law is, essentially, a short-hand technique for drafting a contract – instead of setting out all its terms in a single document, those terms may be included by reference from another source, which may be state law, non-state law or indeed any other source. There are few



57 Pritchard 58 ibid

59 ibid. 60 See

129.

v Norton 106 US 124 (1882).

generally Mills, Party Autonomy (n 4) ch 10.

380  Alex Mills limitations on the origins of contractual provisions that might be incorporated by reference, provided the terms of the contract are clear and readily ascertainable to both parties. Thus, even where choice of law rules do not permit a choice of non-state law as the governing law for a contract, the parties may be able to give effect to those rules through their incorporation as terms of the contract, if those rules are valid contractual terms under the national law that governs the contract. Courts have, for example, endeavoured in at least some cases to give effect to the intentions of parties who (invalidly) purport to choose non-state law (such as, for example, religious law) as the governing law for their contract, by taking into consideration the rules of their chosen system of law in the interpretation of the contract, on the basis that those rules of interpretation have been incorporated as contractual terms.61 However, this secondary role for the chosen law can only be of limited effect – it cannot, for example, affect questions of the validity of the terms of the contract.

IV. Conclusions Choice of court and choice of law clauses are common features of modern contracting practice – indeed so common that their significance may be overlooked. In analysing these clauses, it is important to recognise their distinctive character – they are clauses that seek to affect the legal system (institutional and substantive) which has power over the contract. These clauses cannot be understood as ordinary contractual terms, which should be given effect as a matter of freedom of contract, because they determine which legal system’s conception of freedom of contract governs the contract. They are clauses concerned with ‘freedom of freedom of contract’ – the freedom to choose between the different degrees of contractual freedom available in different legal systems. The distinctive character of these clauses means that they require special justification and also limitation, and necessitates a careful distinction between choice of law and the incorporation by reference of rules of law.



61 See,

eg, Halpern v Halpern [2007] EWCA Civ 291, [2007] 2 Lloyd’s Rep 56.

18 Illegality in English Arbitration Law after Patel v Mirza UGLJEŠA GRUŠIĆ AND MANUEL PENADES FONS

I. Introduction A pays £620,000 to B under a contract, whereby B agrees to use the money to bet on the price of shares using inside information. Insider dealing is a crime in England. The inside information fails to materialise and B keeps the money. A brings a claim for restitution. These well-known facts form the factual background of Patel v Mirza,1 the leading case on illegality in English private law, in which the Supreme Court clarified the test to address allegations of illegality. A lot of ink has been spilt on describing and predicting the effects of Patel on various branches of substantive private law in England.2 By contrast, Patel has so far escaped scrutiny by arbitration specialists. Arbitration law bolsters freedom of contract by allowing parties to create their own dispute resolution mechanism. An arbitration clause adds a procedural layer to a contractual relationship, which limits the ability of courts to hear disputes, including allegations of illegality, related to the contract. Returning to the abovementioned example, suppose there is an international element in the contract between A and B, that the contract contains an arbitration clause and that A has commenced arbitral proceedings. What, if any, is the relevance of Patel on these facts? Difficulties posed by illegality are amplified in arbitration law. First, illegality can be raised before either the arbitral tribunal or a domestic court. Second, illegality can be raised before different domestic courts that exercise different supervisory and supporting functions in relation to arbitration and at different



1 Patel 2 See

v Mirza [2016] UKSC 42, [2017] AC 467. A Bogg and S Green (eds), Illegality after Patel v Mirza (Hart Publishing 2018).

382  Uglješa Grušić and Manuel Penades Fons stages of the arbitral process. An English court, for example, may deal with illegality when it is asked to enforce the arbitration agreement,3 prevent the enforcement of the arbitration agreement,4 determine a preliminary point of jurisdiction5 or law,6 set aside the award7 or enforce the award.8 Third, an English court may deal with illegality in relation to arbitration seated in England or abroad.9 Fourth, illegality can be raised before the tribunal as an issue concerning either jurisdiction or the merits. Similarly, illegality can be raised before an English court as an issue concerning the jurisdiction of the tribunal, the merits,10 public policy11 or arbitrability.12 Fifth, the matter is further complicated by the fact that different aspects of arbitration (eg the arbitration agreement, the procedure or the merits) may be governed by different laws and that the source of alleged illegality may be yet another law. Finally, if illegality is raised before an English court as an issue of public policy at the post-award stage, the matter is even further complicated by the fact that the arbitral award adds a procedural layer of separation between the court and the illegality allegedly affecting the arbitration agreement or the underlying contract. It therefore comes as no surprise that the topic of illegality in arbitration law is complex and highly controversial. We aim to contribute to the debate by examining whether the Supreme Court judgment in Patel is relevant for English arbitration law and, if so, what the relevance of Patel may be. Section II deals with the first of these two questions. After showing that Patel is relevant for English arbitration law, in section III we turn to the Supreme Court judgment in Patel to explore what questions are raised by this case and what parts of the judgment may be relevant for arbitration law. We then focus, in sections IV and V, on the potential relevance of Patel to pre-award and post-award litigation brought in England. Section VI concludes.

3 For example, by staying proceedings (Arbitration Act 1996 (AA 1996), s 9), appointing an arbitrator (AA 1996, s 18), filling a vacancy (AA 1996, s 27), issuing an anti-suit injunction (Senior Courts Act 1981, s 37), awarding damages for breach of the arbitration agreement (Senior Courts Act 1981, s 50) or refusing to give effect to a foreign judgment given in violation of the arbitration agreement (Civil Jurisdiction and Judgments Act 1982, s 32; see also Ecobank Transnational Inc v Tanoh [2015] EWCA Civ 1309, [2016] 1 WLR 2231 on anti-enforcement injunctions). 4 For example, by issuing a declaration that the arbitration agreement is invalid, that the tribunal is not properly constituted or that the dispute falls outside the scope of the arbitration agreement and an anti-arbitration injunction (AA 1996, s 72(1); Senior Courts Act 1981, s 37). 5 AA 1996, s 32. 6 ibid, s 45. 7 ibid, ss 31(5), 67–69, 72(2). 8 ibid, ss 66, 100–103. 9 Pursuant to AA 1996, s 2, ss 9 and 66 of the Act apply to both English and foreign arbitrations; ss 100–103 apply to awards made in the territory of a foreign state that is a party to the New York Convention. An English court will not give effect to a foreign judgment given in violation of an arbitration agreement, regardless of the seat of arbitration. 10 AA 1996, ss 45, 69. 11 ibid, ss 66, 81(1)(c), 103(3). 12 ibid, ss 66, 81(1)(a), 103(3).

Illegality in English Arbitration Law  383

II.  The Relevance of Patel for English Arbitration Law Patel concerned unjust enrichment. Commentators, however, agree that the Supreme Court judgment is relevant for English private law more generally.13 But when commentators state that the ‘approach [in Patel] was clearly being put forward as applicable across civil law’14 or that ‘Patel was a pivotal moment in English private law’,15 they actually mean that Patel is relevant for English substantive private law, that is, ‘generally across tort, contract, trusts and employment law’.16 Little, if any, attention has been paid to the question whether Patel is relevant for other branches of English private law such as arbitration law or private international law. As we shall see, this question has started to surface in practice. Our argument is that Patel is indeed relevant for English arbitration law for the following reasons: (a) the clarification of the test to address allegations of illegality brought by Patel is relevant whenever English law governs the arbitration agreement or the merits; (b) Patel, read together with a preceding Supreme Court judgment on illegality in Les Laboratoires,17 might have changed the treatment of foreign illegality under the common law of conflict of laws: courts might now be able to take into consideration illegality other than that provided by the proper law or the law of the place of performance of the contract. The treatment of foreign illegality under the common law of conflict of laws is inextricably connected with the treatment of illegality in arbitration law; (c) the methodology of analysis consecrated by Patel confirms the appropriateness of the approach traditionally adopted by English courts concerning allegations of illegality in arbitration-related litigation; (d) the reasoning in Patel might add a layer of sophistication and nuance to the approach developed by English courts specifically with respect to illegality in post-award litigation. This concerns the relevance of glosses, the use of the integrity of the legal system as a relevant factor and the possibility of applying slightly different degrees of review in setting-aside and enforcement proceedings. These four effects have both substantive and methodological implications. If English law governs the arbitration agreement and a party argues that the agreement is invalid, inoperative or incapable of being performed because of illegality, the existence and effects of the illegality are assessed under English law,18 13 Bogg and Green (eds), Illegality (n 2). 14 A Burrows, ‘A New Dawn for the Law of Illegality’ in Bogg and Green (eds), Illegality (n 2) 23, 27. 15 J Goudkamp, ‘The Law of Illegality: Identifying the Issues’ in Bogg and Green (eds), Illegality (n 2) 39, 40. 16 A Bogg and S Green, ‘Introduction’ in Bogg and Green (eds), Illegality (n 2) 1. 17 Les Laboratoires Servier v Apotex Inc [2014] UKSC 55, [2015] 1 AC 430. 18 Dalmia Dairy Industries v National Bank of Pakistan [1978] 2 Lloyd’s Rep 223 (CA).

384  Uglješa Grušić and Manuel Penades Fons which includes Patel. If English law governs the merits and there is an allegation of illegality concerning the underlying contract, the test to address allegations of illegality brought by Patel applies to decide on the substantive rights and obligations of the parties.19 If English law does not govern the arbitration agreement or the merits, it is not obvious that Patel holds any relevance for English arbitration law. Our argument is that it may be relevant because Patel and Les Laboratoires have arguably affected the common law conflict-of-laws rules on foreign illegality, which are inextricably connected with the treatment of illegality in arbitration law. Illegality in the English private international law of contracts is now typically addressed under Article 9 of the Rome I Regulation,20 which concerns overriding mandatory provisions. Rome I, however, does not apply to arbitration, because arbitration agreements are expressly excluded from its scope21 and because the addressees of the Regulation are Member State courts, not arbitral tribunals.22 This means that for the present discussion it only matters that illegality in arbitration law is inextricably connected with foreign illegality in the common law of conflict of laws.

A.  Patel and Illegality in the Common Law of Conflict of Laws At common law, illegality is governed by the proper law of the contract.23 There are some exceptions to this principle, the most important of which is public policy. Public policy operates in negative and positive ways.24 English courts can refuse to apply a rule on illegality of the foreign applicable law if that rule or its application is contrary to English public policy. Even if a contract is legal under its proper law, English courts can in some circumstances apply an English rule on illegality. 19 The tribunal can decide whether an allegedly illegal contract is invalid or unenforceable: Harbour Assurance Co (UK) Ltd v Kansa General International Insurance Co Ltd [1993] QB 701 (CA); Westacre Investments Inc v Jugoimport SDPR Holding Co Ltd [1999] QB 740, 768–71; Fiona Trust & Holding Corp v Privalov [2007] UKHL 40, [2007] 4 All ER 951. 20 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) [2008] OJ L177/6. See also Art 16 of Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II) [2007] OJ L199/40. Law Applicable to Contractual Obligations and Non-Contractual Obligations (Amendment etc) (EU Exit) Regulations 2019 (SI 2019/834) will transpose the provisions of Rome I and Rome II into UK law after the end of the Brexit implementation period. 21 Rome I, Art 1(2)(e). 22 Arbitrators may resort, however, to Rome I to solve choice-of-law questions in the absence of choice by the parties. 23 Kahler v Midland Bank Ltd [1950] AC 24 (HL); Zivnostenska Banka National Corp v Frankman [1950] AC 57 (HL); Re Helbert Wagg & Co Ltd [1956] 1 Ch 323. 24 T Hartley, ‘Mandatory Rules in International Contracts: The Common Law Approach’ (1997) 266 Hague Recueil 341, 350–54, 396–402.

Illegality in English Arbitration Law  385 Another exception is laid down in cases in which English courts gave effect to foreign illegality even if the contract in question was governed by English law in which there was no equivalent illegality. An English court will not enforce an English contract to the extent to which its performance is illegal under the law of the place of performance.25 Although the matter is controversial, the prevailing view is that this rule forms part of English contract law only, and not of English private international law.26 This means that illegality in the place of performance might not be a decisive factor to render the contract invalid or unenforceable if the contract is subject to foreign law other than the law of the place of performance and it is legal under its proper law. An English court will, however, hold any contract invalid if the real object and intention of the parties necessitates their joining in an endeavour to perform in a foreign and friendly country an act that is illegal under the law of that country, even if that act is not necessarily required by the terms of the contract.27 This is a conflicts rule, not merely a rule of English contract law.28 In cases falling within this exception, English courts must refuse to give effect to the contract; there is no room for discretion.29 This rule-based approach to foreign illegality has been brought into question by the recent Supreme Court judgments in Les Laboratoires30 and Patel. Les Laboratoires concerned a claim for damages on a cross-undertaking given by the claimant that it would comply with any order that the court might make if it should later find that an interim injunction the claimant had obtained against the defendant had caused loss to the defendant. The claimant raised as a defence that the defendants’ lost profits would have accrued from sales in England of a product whose manufacture in Canada would have infringed a Canadian patent. It was conceded in the Court of Appeal that the illegality defence could apply where the source of the profits from sales in England was illegal under foreign law.31 The Supreme Court omitted to discuss English cases on foreign illegality and apparently proceeded on the basis that violations of foreign laws were to be treated in the same way as violations of domestic laws for the purposes of applying English law rules founded on the maxim ex turpi causa non oritur actio.32 The Supreme Court held that the infringement of the Canadian patent did not constitute a relevant illegality (‘turpitude’) for the illegality defence to operate. Had the Supreme Court considered English cases on foreign illegality, it should have assessed whether the 25 Ralli Bros v Cia Naviera Sota y Aznar [1920] 2 KB 287 (CA). See also Lemenda Trading Co v African Middle East Petroleum Co [1988] QB 448. 26 L Collins (gen ed), Dicey, Morris and Collins on the Conflict of Laws, 15th edn (Sweet & Maxwell 2012) [32-094], [32-096]–[32-103]; Ryder Industries Ltd v Chan [2015] HKCFA 32 [43] (Lord Collins). 27 Foster v Driscoll [1929] 1 KB 470 (CA); Regazzoni v KC Sethia (1944) Ltd [1958] AC 301 (HL). 28 Collins (gen ed), Dicey, Morris and Collins (n 26) [32-193]. 29 Barros Mattos v MacDaniels Ltd [2004] EWHC 1188 (Ch), [2005] 1 WLR 247. 30 Les Laboratoires Servier (SC) (n 17). See also Lilly Icos LLC v 8PM Chemists Ltd [2009] EWHC 1905 (Ch), [2009] Bus LR D43. 31 Les Laboratoires Servier v Apotex Inc [2012] EWCA Civ 593, [2013] Bus LR 80 [69]. 32 L Collins (gen ed), Dicey, Morris and Collins on the Conflict of Laws: Fifth Supplement to the 15th edn (Sweet & Maxwell 2018) [32-102].

386  Uglješa Grušić and Manuel Penades Fons claim in Les Laboratoires should have been refused on the basis that Canadian law could simply not be taken into account because it was neither the proper law nor the law of the place of performance of the relevant obligations. Since Patel did not concern foreign illegality, the Supreme Court neither approved nor disapproved this aspect of Les Laboratoires. Nevertheless, it is important to note that whilst the majority in Patel disagreed and the minority agreed with the rule-based approach to illegality adopted in Les Laboratoires, none of the judges in Patel questioned the assumption in Les Laboratoires that the infringement of a Canadian patent would have been relevant for the illegality defence to operate had it constituted a relevant illegality (‘turpitude’). We cannot draw any solid conclusions about the correctness of the omission of a conflicts analysis in Les Laboratoires from the mere fact that such omission was not questioned in Patel. But, as the editors of Dicey, Morris and Collins conclude, there may be scope to extend the flexible, policy-based approach to illegality of the majority in Patel to foreign illegality and reconsider the rule-based approach to foreign illegality of the preceding English case law.33 At most, this could entail giving English courts discretion to give effect to illegality under the law of a foreign country even if that law is not the proper law of the contract and if the place of performance of the contract is not in that country. The consequence of this happening would be a tectonic shift in the common law of conflict of laws. On the one hand, foreign illegality would become the concern of both contract law and conflict of laws. On the other hand, it would harmonise the methodological approach to illegality in English substantive law and the common law of conflicts of laws.

B.  Patel, Illegality in Arbitration Law and the Common Law of Conflict of Laws Illegality in English arbitration law is inextricably connected with foreign illegality in the English common law of conflict of laws. In Soleimany34 and Westacre,35 the leading cases on illegality in English arbitration law, the courts discussed illegality by reference to some of the leading cases on foreign illegality in the common law of conflict of laws. In Soleimany, the court started its discussion of the ‘illegality issue’ by reproducing Stuart-Smith LJ’s summary of the law relating to illegality in Royal Boskalis Westminster NV v Mountain,36 which prominently featured Foster v Driscol37 and Regazzoni.38 Westacre was largely about distinguishing Soleimany and Lemenda.39

33 ibid.

34 Soleimany

v Soleimany [1999] QB 785 (CA). Investments Inc v Jugoimport SDPR Holding Co Ltd [2000] QB 288 (CA). 36 Royal Boskalis Westminster NV v Mountain [1999] QB 674, 691–62 (CA). 37 Foster v Driscoll (n 27). 38 Regazzoni (n 27). 39 Lemenda (n 25). 35 Westacre

Illegality in English Arbitration Law  387 It is for these reasons that Patel may be relevant for English arbitration law even if English law does not govern the arbitration agreement or the merits. If Les Laboratoires and Patel have affected the common law conflict-of-laws rules on foreign illegality, the arbitration jurisprudence based on those rules might have to be redefined. There is only one reported post-Patel case concerning illegality in arbitration law. In Sinocore,40 the court dealt with the enforcement of a New York Convention award. The dispute arose out of a contract for the sale of goods. The seller had presented forged bills of lading. The tribunal decided in favour of the seller. The High Court ordered the enforcement of the award because the contract was not unlawful or otherwise contrary to public policy and because the award, on its face, was not based on the fraudulent presentation of the forged bills.41 A basis of appeal was that the judge had applied the wrong test for assessing the consequences of the seller’s illegality: He applied the overly narrow (and now discarded) test, set out in Tinsley v Milligan [1994] 1 AC 340, of whether [the seller’s] pleaded claim relied on its own fraud. He should have applied the more flexible approach laid down by the Supreme Court in Patel v Mirza.42

Hamblen LJ disagreed: Patel v Mirza does not affect the principles to be applied when considering recognition and enforcement under section 103 [of the Arbitration Act 1996]. Patel v Mirza will be relevant to whether, on the facts found, there is any illegality as a matter of English law. The policy considerations taken into account for the purpose of that determination may overlap with those to be considered under section 103, but Patel v Mirza neither purports to nor does it consider or decide the proper approach to be taken in section 103(3) cases. In particular: (1) Prior to Patel v Mirza it is clear that a distinct approach applied in the context of a challenge to enforcement of arbitration award under section 103(3), compared to the enforcement of a substantive claim. (2) In Patel v Mirza the Supreme Court did not consider any of the authorities on illegality and public policy in the context of section 103(3). These authorities were not cited in argument, nor were they referred to in the judgment. There is nothing in the judgment to suggest that the Supreme Court contemplated that the approach it set out might also be applicable in the context of section 103(3). (3) There are sound justifications for taking a different approach to substantive claims and enforcement claims, reflecting the different role performed by the court in each circumstance … (4) It may be that Patel v Mirza has moved the jurisprudence on illegality as a defence to a substantive claim rather closer to the multifactorial approach that has always

40 RBRG Trading (UK) Ltd v Sinocore International Co Ltd [2018] EWCA Civ 838, [2019] 1 All ER (Comm) 810. 41 Sinocore International Co Ltd v RBRG Trading (UK) Ltd [2017] EWHC 251 (Comm), [2018] 1 All ER (Comm) 576. 42 RBRG Trading (UK) Ltd (CA) (n 40) [21](i).

388  Uglješa Grušić and Manuel Penades Fons applied in the context of illegality as a public policy defence to enforcement, but the context and the relevant factors remain different. In particular, as the authorities make clear, there is always a strong public policy in support of enforcement.43

One should not read too much into this extract. Hamblen LJ did not say that Patel was irrelevant for arbitration law. He said that the principles to be applied when considering recognition and enforcement under section 103 had not been affected by Patel. We agree with this. Our argument is that Patel may be relevant for English arbitration law even if English law does not govern the arbitration agreement or the merits either, at least in some cases, as a matter of substance or as a matter of methodology. In order to test this hypothesis, we need to take a closer look at the Supreme Court judgment in Patel and the principles and rules applicable in preaward and post-award litigation brought in England.

III.  Patel: A Closer Look Lord Toulson’s judgment in Patel, which sets out the majority’s flexible, policybased approach to illegality, is well known. Lord Toulson summarised this approach as follows: The essential rationale of the illegality doctrine is that it would be contrary to the public interest to enforce a claim if to do so would be harmful to the integrity of the legal system (or, possibly, certain aspects of public morality …). In assessing whether the public interest would be harmed in that way, it is necessary (a) to consider the underlying purpose of the prohibition which has been transgressed and whether that purpose will be enhanced by denial of the claim, (b) to consider any other relevant public policy on which the denial of the claim may have an impact and (c) to consider whether denial of the claim would be a proportionate response to the illegality, bearing in mind that punishment is a matter for the criminal courts. Within that framework, various factors may be relevant, but it would be a mistake to suggest that the court is free to decide a case in an undisciplined way. The public interest is best served by a principled and transparent assessment of the considerations identified, rather by than the application of a formal approach capable of producing results which may appear arbitrary, unjust or disproportionate.44

Patel was allowed to enforce his claim for unjust enrichment. The fact that the money he sought to recover was paid for an unlawful purpose did not bar the claim. Patel was not one of the ‘rare cases where for some particular reason the enforcement of such a claim might be regarded as undermining the integrity of the justice system’.45



43 ibid

[26]. v Mirza (n 1) [120]. 45 ibid [121]. 44 Patel

Illegality in English Arbitration Law  389 Patel has been criticised despite the welcome clarification of the test to address allegations of illegality, largely because of the number of questions it left open.46 Asking some of these questions from the perspective of arbitration law helps us to inform the discussion that follows. One question left open by Patel is whether the majority’s approach to illegality, assuming that it extends across private law, superseded other rules that had been developed in specific types of case.47 Assuming that Patel is of some relevance for arbitration law either directly because of the change it and Les Laboratoires have arguably made to the treatment of foreign illegality in the common law of conflict of laws, or indirectly through the methodology of analysis it consecrated, this question can be rephrased as whether Patel superseded the rules laid down in Soleimany, Westacre and similar cases or whether it constitutes a ‘mere crosscheck’48 on those rules. Another question left open by Patel is how the majority’s approach to illegality is to be applied, in particular which factors can or should be taken into account in applying this approach and how much weight is or should be given to each factor.49 From the perspective of arbitration law, some of the relevant factors are mentioned in section 1 of the Arbitration Act 1996. Other relevant factors and the weight to be given to them have to be uncovered by looking at the relevant case law and arbitration theory. Patel does not clarify whether the illegality doctrine is a single rule or encompasses two rules, one of which prevents liability from arising and the other of which affects only the remedy.50 The importance of this for substantive law is that ‘if there are two rules … the liability rule might be controlled by one test (the “range of factors” approach enunciated in Patel) and the quantum rule by another’.51 The importance of this for arbitration law is that the doctrine of illegality might operate differently in situations that raise the issue of validity of the arbitration agreement or the award and in situations that concern their enforcement. It is not clear what does or should count as ‘turpitude’ for the purposes of Patel,52 including whether Patel affects statutory illegality.53 From the perspective of arbitration law, it can be asked not only whether criminal conduct, ‘quasicriminal’ conduct and civil wrongs count as turpitude and what relationship the turpitude must have with the claim,54 but also how the typically international character of arbitration affects the analysis of what counts as turpitude. This includes

46 See Goudkamp, ‘The Law of Illegality’ (n 15). 47 ibid 42–44. 48 ibid 44. 49 ibid 44–48. 50 ibid 48–49. 51 ibid 49. 52 ibid 52–54. 53 Bogg and Green, ‘Introduction’ (n 16) 19; Burrows, ‘A New Dawn’ (n 14) 24–25; J O’Sullivan, ‘Illegality and Contractual Enforcement after Patel v Mirza’ in Bogg and Green (eds), Illegality (n 2) 165, 179. 54 See Les Laboratoires (SC) (n 17).

390  Uglješa Grušić and Manuel Penades Fons the fact that an English court may be required to deal with illegality in relation to arbitration whose seat is in England or a foreign arbitration, that different aspects of arbitration may be governed by different laws and that the source of alleged illegality may be yet another law. It can further be asked whether different approaches exist or should be adopted to different kinds of illegality. Violation of European Union (EU) law and statutory illegality are obvious candidates for a separate treatment. The following discussion addresses some of these questions from the perspective of arbitration law. It is divided in two sections concerning pre-award and post-award litigation brought in England.

IV.  Pre-award Litigation Pre-award litigation in which illegality is raised typically concerns the validity and enforcement of the arbitration agreement. The source of the legal rules whose violation constitutes the illegality in question is crucial here. If the source of illegality is the violation of EU law, a unique set of considerations applies, at least until the end of the Brexit implementation period. The best example is Accentuate Ltd v Asigra Inc,55 where the High Court refused to stay its proceedings and refer the parties to arbitration in Canada because the claim before the court was based on certain mandatory provisions of EU law origin contained in the UK implementation of the Commercial Agents Directive,56 and the Canadian arbitrators had already decided the dispute by applying the law of Ontario. In cases that trigger the application of mandatory provisions of EU law, the EU public policy of the creation of the internal market prevails over conflicting public policies, including the promotion of party autonomy and arbitration. The dominance of the EU public policy of market building, however, does not necessarily eliminate arbitration in cases that trigger the application of EU law. Accentuate was an unusual case in that a party applied for a stay of court proceedings under section 9 of the Arbitration Act 1996 after the arbitral proceedings had been concluded and the arbitrators had refused to apply the UK implementation of the Commercial Agents Directive. In a typical case in which a party applies for a stay of court proceedings under section 9 either before or during the arbitral proceedings, there is a way to balance competing public policies. If the tribunal has or is given a mandate by the parties to apply EU law, it would be a disproportionate interference with the public policies of promoting party autonomy and 55 Accentuate Ltd v Asigra Inc [2009] EWHC 2655 (QB), [2010] 2 All ER (Comm) 738. 56 Commercial Agents (Council Directive) Regulations 1993 (SI 1993/3053), implementing Council Directive 86/653/EEC of 18 December 1986 on the coordination of the laws of the Member States relating to self-employed commercial agents [1986] OJ L382/17. See also Case C-381/98 Ingmar GB Ltd v Eaton Leonard Technologies Ltd [2000] ECR I-9305; Case C-126/97 Eco Swiss China Time Ltd v Benetton International NV [1999] ECR I-3055.

Illegality in English Arbitration Law  391 arbitration for English courts to refuse to enforce the arbitration agreement by staying proceedings under section 9.57 Even if Patel is not directly applicable to cases that trigger the application of EU law, it is worth noting that the methodology of balancing competing public policies and applying the test of proportionality is precisely the kind of methodology laid down in the majority judgment in Patel. If the source of illegality is the violation of a statute forming part of English law, the statute may expressly or impliedly determine the effect of its violation on the validity and effectiveness of the arbitration agreement. There is nothing in Patel to suggest that it affects the process of statutory construction.58 If the source of illegality is not the violation of EU law or an English statute, an additional crucial factor becomes the law governing the arbitration agreement. If English law governs the arbitration agreement and the source of illegality is the violation of an English common law rule, the effects of the common law illegality are assessed under English law. English law includes not only the principles laid down in Patel, but also a pro-arbitration interpretation of arbitration agreements and a very strong doctrine of separability.59 If Patel applies as part of the governing English law, the tribunal and domestic courts should apply the methodology of balancing the public policy promoted by the violated English common law rule, that is, the integrity of the legal system, and competing public policies60 and the test of proportionality. If English law governs the arbitration agreement and the source of illegality is the violation of a rule of foreign law, the following principles and rules apply. The Court of Appeal in Soleimany stated obiter that an arbitration agreement is invalid if it submits to the tribunal the question of how to split the proceeds of a joint venture that has as its object the commission of illegal acts in a foreign and friendly state.61 This rule is the transposition into arbitration law of the common law rule that an English court will hold a contract invalid if the real object and intention of the parties necessitates their joining in an endeavour to perform in a foreign friendly country an act that is illegal under the law of that country.62 A related common law rule on foreign illegality that has an impact on arbitral jurisdiction

57 Fern Computer Consultancy Ltd v Intergraph Cadworx & Analysis Solutions Inc [2014] EWHC 2908 (Ch), [2014] Bus LR 1397, [126]–[128]; J Kleinheisterkamp, ‘Overriding Mandatory Rules in International Arbitration’ (2018) 67 International and Comparative Law Quarterly 903; cf Accentuate Ltd v Asigra Inc (n 55) [89]. 58 Bogg and Green, ‘Introduction’ (n 16) 19; Burrows, ‘A New Dawn’ (n 14) 24–25; O’Sullivan, ‘Illegality and Contractual Enforcement’ (n 53) 179. 59 AA 1996, ss 2(5), 7; Fiona Trust (n 19); see also Harbour Assurance (n 19); Westacre (QBD) (n 19) 755–67; National Iranian Oil Co v Crescent Petroleum Co International ltd [2016] EWHC 510 (Comm), [2016] 2 Lloyd’s Rep 146 [7]–[17]. 60 Competing public policies include the promotion of party autonomy and arbitration: S  Brekoulakis, ‘The Historical Treatment of Arbitration under English Law and the Development of the Policy Favouring Arbitration’ (2019) 1 OJLS 124. 61 Soleimany (n 34) 797–98. Compare Beijing Jianlong Heavy Industry Group v Golden Ocean Group Ltd [2013] EWHC 1063 (Comm), [2013] 2 All ER (Comm) 436. 62 Foster v Driscoll (n 27); Regazzoni (n 27).

392  Uglješa Grušić and Manuel Penades Fons is that an English court will not enforce an English contract to the extent to which its performance is illegal under the law of the place of performance.63 Transposed into arbitration law, this rule states that an English arbitration agreement will not be enforced to the extent to which it provides for an arbitration that is illegal under the law of the country where the arbitration agreement has to be performed, that is, the law of the seat of arbitration.64 The relevance of Patel in these scenarios might be the softening of the transposition into arbitration law of the common law rules on foreign illegality. In other words, Patel (read together with Les Laboratoires) would entitle the tribunal and domestic courts to apply the test of proportionality and the methodology of balancing the public policy promoted by the violated rule and competing public policies.65 If foreign law governs the arbitration agreement, the existence and effects of illegality will typically be assessed under that foreign law.66 There are some recognized exceptions to this principle. An exception is derived from the following statement of Waller LJ in Soleimany: There may be illegal or immoral dealings which are, from an English law perspective, incapable of being arbitrated because an agreement to arbitrate them would itself be illegal or contrary to public policy under English law. The English court would not recognise an agreement between the highwaymen to arbitrate their differences any more than it would recognise the original agreement to split the proceeds.67

The language used in this judgment suggests that English courts would never recognise an arbitration agreement between highwaymen, regardless of where they operated, the seat of the arbitration and the law governing the arbitration agreement. There are two ways to conceptualise the operation of this illegality. The first is to regard it as a positive operation of English public policy, that is, as an overriding application of an English rule on illegality. The other is to regard it as a consequence of the application of English rules on arbitrability. Waller LJ’s judgment is inconclusive. The extract above provides support for the second conceptualisation,68 but some other parts of the judgment provide support for the first.69 63 Ralli Bros (n 25). 64 Tamil Nadu Electricity Board v ST-CMS Electric Company Private Ltd [2007] EWHC 1713 (Comm), [2007] 2 All ER (Comm) 701 [46]–[49]; Abuja International Hotels Limited v Meridien SAS [2012] EWHC 87 (Comm), [2012] 1 Lloyd’s Rep 461 [27]. 65 In addition to the public policies mentioned in n 60, the competing public policies include comity if the violated rule belongs to the law of a state that is foreign and friendly to the UK. 66 Westacre (CA) (n 35); Omnium de Traitement et de Valorisation SA v Hilmarton Ltd [1999] 2 All ER (Comm) 146. 67 Soleimany (n 34) 797. See also ibid 799, referring to contracts for trading with the enemy and between robbers. 68 See also ibid 799 (‘it may be that in the case of palpable illegality (to use the expression adopted by Colman J in Westacre Investments Inc v Jugoimport-SPDR Holding Co Ltd [1999] QB 740, 767F), an English court would declare that there was no arbitrable dispute’) and 802 (‘the arbitrator has no jurisdiction to make an award’). 69 Waller LJ refers, in Soleimany (n 34) 797, to an arbitration agreement between highwaymen being ‘void for illegality’ and ‘invalid’ See also ibid 799, where Waller LJ refers to contracts for trading with the enemy and between robbers, where the court would hold the arbitration clause ‘void’.

Illegality in English Arbitration Law  393 The High Court judgment in Westacre supports the second conceptualisation.70 This judgment further clarifies that courts should balance the competing policies promoted by the violated rule and sustaining international arbitration agreements when deciding whether a dispute concerning illegality is arbitrable.71 Proportionality is also relevant: consideration should be given to the weight that ought to be attached to the risk that arbitrators might reach the wrong decision in a way which could not be challenged and thereby give effect to an underlying contract the courts would have declined to enforce,72 and ‘whether the case involves consideration of complex principles of English law capable only of being safely determined by an English court’.73 This methodology of balancing competing public policies and applying the test of proportionality is precisely the kind of methodology laid down in the majority judgment in Patel. Patel and Les Laboratiores might be seen as lending support to another exception to the principle that the existence and effects of illegality will be assessed under the foreign law governing the arbitration agreement. If foreign law governs the arbitration agreement, and a party argues that the agreement is invalid, inoperative or incapable of being performed because of the violation of a rule that belongs to another foreign law, the illegality will typically be irrelevant if the governing law attaches no relevance to it. The prevailing view is that English common law rules on foreign illegality are irrelevant in this situation because they are part of English contract law only.74 But if Patel and Les Laboratiores have indeed affected the common law conflict-of-laws rules on foreign illegality, this might have had an effect on the treatment of foreign illegality in arbitration law.

V.  Post-award Litigation Issues of illegality can arise in a number of ways in the context of post-award litigation. A party might challenge or resist the recognition and enforcement of the award for lack of jurisdiction.75 What has been said in section IV about Patel and arbitral jurisdiction is also relevant in this context. Illegality can found a challenge on a point of English law.76 Patel is central to this analysis, as it establishes the approach to illegality in English private law. Finally, courts might be asked

70 Westacre (QBD) (n 19) 768–69 (Colman J): ‘The approach to the question whether as a matter of English public policy an agreement to arbitrate that issue should be treated as enforceable must be determined by considerations similar to those deployed by the United States Supreme Court in the context of statutory illegality in relation to the anti-trust legislation in Mitsubishi Motors Corporation v Soler Chrysler-Plymouth Inc, 105A SCt 3346.’ 71 Westacre (QBD) (n 19) 769. 72 ibid. See also ibid 767. 73 ibid 769. 74 See references cited at n 26. 75 AA 1996, ss 31(5), 66(3), 67, 72(2), 103(2). 76 ibid, s 69.

394  Uglješa Grušić and Manuel Penades Fons to consider issues of illegality when a party argues that the award is contrary to public policy or concerns an inarbitrable matter.77 This section focuses on the relevance of Patel for situations in which illegality is invoked under the guise of public policy. Public policy constitutes a ground to set aside an English award and to refuse its recognition and enforcement. The relevant public policy is English public policy. In these contexts, courts are not dealing with the underlying contract but with the award that resolves a dispute derived from that contract, and the existence of illegality in the contract is just one of the relevant factors. Considerations that operate in the sphere of illegality, such as the integrity of the legal system and comity (in cases of foreign illegality), must be balanced against considerations favouring the finality of awards, deference toward arbitrators (and, if appropriate, foreign supervisory courts) and a general policy in support of arbitration. As a result, public policy is applied only in exceptional circumstances. In Patel, illegality was not analysed merely as one relevant factor in the balancing exercise but occupied centre stage. One might argue, then, especially in light of Sinocore,78 that Patel is irrelevant to the operation of public policy in post-award litigation where pro-arbitration considerations traditionally take centre stage. On this view, Patel would inhabit an earlier phase of the analysis, aimed at identifying the existence of illegality and its effects on the jurisdiction of the tribunal and the merits, whereas its influence would vanish in the second stage, when the existence of illegality and the policies around it would be balanced against other relevant considerations pertaining to the arbitral context. The reality, however, is that Patel cannot be confined to that first phase. By rejecting a rule-based approach and adopting a multifactorial approach, Patel unifies the method with which issues of illegality are examined in the various areas of private law, including arbitration. The resulting methodological consistency increases the coherence of English private law in the field of illegality and confirms the appropriateness of the approach traditionally adopted by English courts in arbitration-related litigation concerning allegations of illegality. Before we proceed, we should note that, as in the context of pre-award litigation, illegality based on the violation of EU law currently receives a separate treatment at the post-award stage. The duty to protect the primacy of EU law and to guarantee its effectiveness reduces the room for manoeuvre allowed by the multifactorial approach. The result is that English courts might be bound to set aside and refuse the recognition and enforcement of awards that breach mandatory EU law. The following text therefore focuses on the effect of illegality other than the violation of EU law on the setting-aside and recognition and enforcement proceedings in England. Section V.A describes the current law. Section V.B explores



77 ibid,

ss 68(2)(g), 81(1)(a) and (c), 103(3). Trading (UK) Ltd (CA) (n 40).

78 RBRG

Illegality in English Arbitration Law  395 substantive and methodological effects that Patel might have on the treatment of illegality at the post-award stage.

A.  The Flexible, Policy-based Approach in Post-award Litigation As in English substantive private law, the balancing of competing policies in post-award litigation is unavoidably case-specific. This is not an impediment to attempting to make some general categorisations. English law makes a fundamental distinction between awards that give effect to illegality in the underlying contract and awards that are ‘tainted’ by illegality. Courts may scrutinise the awards falling into the first category, although the level of scrutiny depends on whether illegality is apparent from the face of the award or requires going beyond the facts determined by the arbitrators. In Soleimany,79 the parties and the arbitrator recognised that the underlying contract was an illicit enterprise under which it was the joint intention of the parties to smuggle carpets illegally out of Iran. The arbitrator assumed jurisdiction and ordered the defendant to pay the claimant his share of the profits of the enterprise. As the award expressly noted the illegality, the court refused to enforce the award. The same treatment will be given to an award that gives effect to an illegal practice, such as payment or recovery of a bribe, or which upholds a fraudulent claim to payment based on the presentation of documents admitted by the parties, or found by the tribunal, to be forgeries.80 When the finding of illegality requires going beyond the facts determined by the arbitrators, the balancing of competing policies becomes evident. If the alleged illegality refers to a narrow category of ‘universally condemned activities’,81 English courts will probably entertain the argument, regardless of whether or not the evidence supporting the argument was considered by the tribunal.82 If English courts find illegality of that exceptional nature, they will not enforce the award, ‘for it would be contrary to public policy that the arbitrator should be entitled to ignore palpable and indisputable illegality’.83 As mentioned in section IV, this rule might be better characterised as one of arbitrability. According to the Court of Appeal in Soleimany, in the case of palpable illegality there is ‘no arbitrable dispute’84 and ‘an arbitrator has no jurisdiction to make an award’.85 When the nature of the

79 Soleimany (n 34). 80 Sinocore International Co Ltd (HC) (n 41) [30], referring to National Iranian Oil Co (n 59) [41]–[42]. 81 Westacre (CA) (n 35) 302 (Waller LJ). 82 ibid 311 (Waller LJ), 316–17 (Mantell LJ). 83 Westacre (QBD) (n 19) 767 (Colman J), confirmed in Soleimany (n 34) 803 (Waller LJ). 84 Soleimany (n 34) 799 (Waller LJ). 85 ibid 802 (Waller LJ). See also Westacre (QBD) (n 19) 768–69 (Colman J).

396  Uglješa Grušić and Manuel Penades Fons countervailing illegality is less severe, deference towards the adjudicative function of arbitrators reduces the possibility of examining issues of illegality at the postaward stage. If the illegality question was subject to debate during the arbitration and the tribunal concluded that the conduct was legal, courts will generally uphold the resulting award.86 The only exception concerns situations in which the party opposing the award is able to adduce evidence that was unavailable during the arbitration.87 In these scenarios, the court will ‘have to consider whether the public policy against the enforcement of illegal contracts outweighs the countervailing public policy in support of the finality of awards’.88 In none of the reported cases did the balancing exercise result in English courts interfering with the arbitrators’ decision on illegality. The second category comprises awards that are ‘tainted’ by illegality but do not give effect to it. The lesser level of opprobrium associated with these cases justifies the generally positive attitude of English courts toward the arbitrators’ decision. For instance, in Honeywell89 and National Iranian Oil,90 the High Court ordered the enforcement of two awards notwithstanding the allegations that the underlying contracts had been induced by bribery. The reason was that whilst bribery is clearly contrary to English public policy and contracts to bribe are unenforceable, contracts that have been procured by bribes are not unenforceable but voidable.91 The same conclusion applies where illegality is not central to the claim that motivated the arbitration or to the underlying contract, even if such illegality is apparent from the face of the award. In Sinocore,92 the Court of Appeal upheld the enforcement of the award in favour of the seller notwithstanding the fact that the seller had forged bills of lading to obtain payment. The illegality was not taken into account because the arbitrators found that the buyer had breached the contract before the illegality occurred.

B.  The Silent Ramifications of Patel in Post-award Litigation The influence of Patel in post-award litigation is not confined to confirming the appropriateness of the multifactorial approach traditionally adopted in arbitration law. While ‘the context and the relevant factors remain different’,93 it is possible to 86 R v V [2008] EWHC 1531 (Comm), [2009] 1 Lloyd’s Rep 97 [13]–[16]. 87 This also means that where a party has failed to furnish available evidence to the tribunal, estoppel will operate to prevent the reopening of the matter: Westacre (CA) (n 35), 305–10 (Waller LJ), 316 (Mantell LJ); Honeywell International Middle East Ltd v Meydan Group LLC [2014] EWHC 1344 (TCC), [2014] 2 Lloyd’s Rep 133 [89]. 88 Westacre (QBD) (n 19) 767–68 (Colman J). 89 Honeywell International Middle East Ltd (n 87) [178]–[185]. 90 National Iranian Oil Co (n 59) [31]–[51]. Here the tribunal found that there was no bribery but just a failed attempt to bribe. 91 Wilson v Hurstanger Ltd [2007] EWCA Civ 299, [2007] 1 WLR 2351 [38] (Tuckey LJ). 92 RBRG Trading (UK) Ltd (CA) (n 40). 93 ibid [26].

Illegality in English Arbitration Law  397 identify some aspects of Patel that could affect the scrutiny of arbitral awards by English courts in matters of illegality.

i.  Enforcement Proceedings v Annulment Proceedings: A Measured Approach? The first aspect of Patel that could affect the scrutiny of arbitral awards by English courts in matters of illegality concerns the potential to distinguish between enforcement proceedings and setting-aside proceedings. While the illegal conduct might coincide in both scenarios, the procedural position of the party seeking to benefit from it is fundamentally different. In enforcement actions, that party requests a judicial decision ordering the other party to comply with the award, against which illegality operates as a defence. In setting-aside proceedings, the aggrieved party in the arbitration requests the court to annul the award because of illegality. At the moment, the case law concerning illegality in post-award litigation does not disclose any difference in treatment based on the type of proceeding. There are, however, reasons to doubt this unifying logic. Success in an enforcement action translates into a positive result. The action seeks to turn the arbitral order into tangible action by the losing party at the request of the executory power of the judiciary. Conversely, success in a setting-aside action typically results in a negative outcome, a mere declaration. The award becomes a nullity and, as a matter of English law, ceases to exist and produce effects between the parties. This fundamental distinction of concepts is not alien to the English substantive law on illegality. In Patel, Lords Toulson and Clarke distinguished between the effects that illegality could have on the validity and enforceability of a contract, that is, ‘whether the contract should be regarded as tainted by illegality [or] whether the relief claimed should be granted’.94 Similar conceptual scission was already present in Walker v Chapman,95 where Lord Mansfield distinguished between a claim to overturn an illegal contract and a claim to obtain benefit. It was also supported by the Law Commission in its 2009 Report, The Illegality Defence, where it concluded that the policies that underlie the illegality defence are less likely to come into play where parties are attempting to undo, rather than carry out, an illegal contract.96 This entrenched distinction in the English substantive law of illegality gives rise to the question whether English courts should adopt a different approach to illegality in setting-aside proceedings and in enforcement proceedings. It could be the case that considerations of illegality should feature more prominently in enforcement actions, where the consequence of ignoring them is the issuance of an order

94 Patel v Mirza (n 1) [219] (Lord Clarke), referring to [115] (Lord Toulson). 95 Walker v Chapman (1773) Lofft 342, 98 ER 684. 96 Law Commission, The Illegality Defence: A Consultative Report (Consultation Paper No 189, 2009) [4.37].

398  Uglješa Grušić and Manuel Penades Fons requiring a party to act. In setting-aside proceedings, on the other side, the balancing test could be more inclined in favour of upholding the award (and tolerating higher levels of illegality), leaving the stricter scrutiny to an eventual enforcement stage, in England or abroad. This measured approach would be consistent with the considerations of purposiveness and proportionality required by Lord Toulson in Patel: In assessing whether the public interest would be harmed in that way, it is necessary (a) to consider the underlying purpose of the prohibition which has been transgressed and whether that purpose will be enhanced by denial of the claim … (c) to consider whether denial of the claim would be a proportionate response to the illegality, bearing in mind that punishment is a matter for the criminal courts.97

The suggested dichotomy would be consistent with the national origin of the definition of public policy and with the cross-border effects of annulment decisions. If, according to the prevailing interpretation of Article V(1)(e) of the New York Convention, the setting aside of an award should bar its enforceability at an international level,98 the application of national standards such as illegality to annul an award should be subject to special caution. Otherwise, a country would risk exporting its internal notions of substantive public policy and restricting the effectiveness of an award that may never be intended to be enforced in the territory of the seat. This is also the logic that underpins Article IX(2) of the European Convention on International Commercial Arbitration of 1961, which excludes annulment decisions based on arbitrability and public policy from the circumstances that can be relevant under Article V(1)(e) of the New York Convention. This might be one of the factors why, to the knowledge of the authors, in England there has not been any successful challenge of an award on the basis of the violation of substantive public policy. The suggested dichotomy would not be contrary to the doctrine of comity. Enforcing an award that violates the law of a foreign and friendly state is not the same thing as merely refusing to annul such an award. In cases of enforcement, English public policy might require courts to refuse to give effect to the award in order to avoid the disregard of relevant foreign laws (like those of the country of performance). Such level of deference, however, is not required in setting-aside actions for, even if confirmed, the content of the award is not legally implemented. The element of comity in English public policy cannot be so wide as to internalise the protection of relevant foreign laws when the court is not being asked to order an act in breach of them. Ultimately, every country should be responsible for the enforcement and protection of its own public policies.

97 Patel v Mirza (n 1) [120] (Lord Toulson) (emphasis added). 98 Yukos Capital SarL v OJSC Oil Co Rosneft [2014] EWHC 2188 (Comm), [2014] 2 Lloyd’s Rep 435 [11].

Illegality in English Arbitration Law  399

ii.  The Integrity of the Legal System Despite Arbitration The balancing exercise required by post-award litigation includes a heavy component of arbitration-friendly considerations (finality of awards, deference toward arbitrators, promotion of London as an arbitral seat, etc). These considerations eliminate the possibility of reviewing the merits of the award and shield errors of law from judicial scrutiny. In exchange, they increase the expediency, effectiveness and commerciality of arbitration as a dispute-resolution mechanism. These are laudable benefits. The result is that courts will uphold or enforce an award in some situations in which illegality would prevent them from enforcing the underlying contract. That is, arbitration operates as a procedural layer that has the potential to hinder the correct application of the law and the achievement of the ideals of justice that give content to the illegality defence under English substantive law. Case law demonstrates that, unlike in pre-award litigation where courts give weight to the consideration that ‘arbitrators might reach the wrong decision in a way which could not be challenged and thereby give effect to an underlying contract which the courts would have declined to enforce’,99 at the post-award stage the fact that upholding or enforcing an award would harm the integrity of the legal system has never played a decisive role.100 Such tendency is somehow paradoxical in the context of illegality, since Lord Toulson insisted in Patel that [t]he essential rationale of the illegality doctrine is that it would be contrary to the public interest to enforce a claim if to do so would be harmful to the integrity of the legal system (or, possibly, certain aspects of public morality …).101

This view was further reinforced by Lord Kerr, who concluded that ‘the underpinning of the preservation of that integrity … is par excellence a public policy consideration’ and wondered ‘[W]hat is the preservation of the integrity of the legal system, if not a public policy consideration?’102 In light of these statements, one could ask whether in future decisions courts should include the integrity of the legal system as an autonomous consideration when illegality is considered in post-award litigation.103

iii.  The Relevance of Legal Certainty and the Risk of Ossification Patel is finally relevant because of what it tells us about rules and glosses in the law of illegality. While adopting an essentially multifactorial approach, the increasing 99 Westacre (QBD) (n 19) 769 (Colman J). 100 The integrity of the legal system, however, was mentioned by Waller LJ in Soleimany (n 34) 800 and Westacre (CA) (n 35) 315–16. Recent cases have not included such references. 101 Patel v Mirza (n 1) [120]. 102 ibid [125]. 103 It must be noted, however, that the protection of the integrity of the legal system is not an absolute aim. In fact, Patel was allowed to enforce his claim for unjust enrichment.

400  Uglješa Grušić and Manuel Penades Fons sophistication of English courts in their treatment of illegality at the post-award stage has led to elaborate attempts to capture the body of case law in judge-made glosses104 and doctrinal pseudo-codifications.105 While these efforts contribute to the clarity and understanding of the law and help to increase legal certainty for the benefit of parties involved in arbitration as well as arbitrators, they do not come risk-free. In an area where the Supreme Court has expressly abandoned a rule-based approach to embrace a flexible and policy-based approach, the successive repetition and direct application of the principles distilled in those glosses and pseudo-codifications could prove dangerous. Patel reminds us of the limited, albeit unquestionably useful, role of jurisprudential distillations, and warns us of the risk of ossification, that is turning principles into rules and structured discretion into disciplined application of precedent. For instance, it was mentioned earlier that in the absence of fresh evidence or save in exceptional circumstances, English courts will not interfere with the arbitrators’ decision on illegality. Despite the use of this approach in various cases,106 it is dangerous to perceive it as a fixed rule in light of Patel. An alternative flexible approach to this issue was suggested obiter by Waller LJ in Soleimany and later in Westacre. According to Waller LJ, regardless of whether there is fresh evidence, the court should entertain the illegality argument at least to examine whether there is prima facie evidence that the award is based on an illegal contract: In our view, an enforcement judge, if there is prima facie evidence from one side that the award is based on an illegal contract, should inquire further to some extent. Is there evidence on the other side to the contrary? Has the arbitrator expressly found that the underlying contract was not illegal? Or is it a fair inference that he did reach that conclusion? Is there anything to suggest that the arbitrator was incompetent to conduct such an inquiry? May there have been collusion or bad faith, so as to procure an award despite illegality? Arbitrations are, after all, conducted in a wide variety of situations; not just before high-powered tribunals in international trade but in many other circumstances. We do not for one moment suggest that the judge should conduct a full-scale trial of those matters in the first instance. That would create the mischief which the arbitration was designed to avoid. The judge has to decide whether it is proper to give full faith and credit to the arbitrator’s award. Only if he decides at the preliminary stage that he should not take that course does he need to embark on a more elaborate inquiry into the issue of illegality.107

Mantell LJ and Sir David Hirst in Westacre found ‘some difficulty with the concept [of some kind of preliminary inquiry short of a full scale trial] and even greater concerns about its application in practice’ but decided to accept the guidelines

104 Westacre (QBD) (n 19) 767–68, and RBRG Trading (UK) Ltd (CA) (n 40) [25] are examples of this. 105 For instance, Collins (gen ed), Dicey, Morris and Collins (n 26) [16-150]. 106 Westacre (QBD) (n 19) 768–71, 773, 777–84 (Colman J); National Iranian Oil Co (n 59) [46]; RBRG Trading (UK) Ltd (CA) (n 40) [25(2)]. 107 Soleimany (n 34) 800. See also ibid 803–04.

Illegality in English Arbitration Law  401 ‘uncritically’.108 Notwithstanding those difficulties, Patel can be seen as providing some support for Waller LJ’s approach.

VI. Conclusion Patel is one of the most important decisions of the last decade in the field of English private law, but its effects in the areas of private international law and arbitration law are very unclear. This chapter has examined Patel through the lenses of arbitration law to identify possible substantive and methodological implications. The importance of the question lies in the fact that arbitration operates primarily in the field of private law and is, in and of itself, the result of freedom of contract. The overarching conclusion is that the adoption of the factor-based approach by the Supreme Court confirms the appropriateness of the similar approach used traditionally in arbitration-related litigation to examine issues of illegality. This methodological consistency increases the coherence in the treatment given to illegality under English private law. From a more substantive perspective, Patel (read together with Les Laboratoires) might have changed the treatment of foreign illegality under the common law of conflict of laws. Illegality other than that provided by the proper law or the law of the place of performance might now be taken into consideration. This, in turn, might have changed the approach to foreign illegality in arbitration law. Finally, Patel might produce some specific effects in post-award litigation. It could be used to sustain the argument that the treatment of illegality in enforcement actions should be stricter than in annulment proceedings; it confirms the relevance of the integrity of the legal system as a policy aim in its own right in the treatment of illegality; and it reminds us of the need to apply caution when relying on inherited rules and glosses concerning illegality in arbitration.



108 Westacre

(CA) (n 35) 316–17.

402

19 The Reform of Insurance Warranties: Looking Beyond the Past JOHN LOWRY AND ROD EDMUNDS

I. Introduction Warranties have long occupied a pivotal place in the development of insurance contract law. Yet, as is acknowledged by the leading insurance law works,1 defining the term has proved somewhat convoluted.2 For sure, the term ‘warranty’ carries a meaning different from that used elsewhere in the general law of contract.3 Otherwise the linguistic contortions have not been helped by the distinction that once was drawn between pre-contractual undertakings that went to the root of the insurance contract4 and the more familiar notion of fundamental terms as risk-defining clauses in the concluded contract. Broadly speaking, definitional questions were settled in jurisprudence attributable, in no small measure, to the judicial ingenuity of eighteenth-century judges who were dealing with the burgeoning number of marine insurance claims,5 most notably Lord Mansfield CJ. In Pawson v Watson he stated that [t]here is no distinction better known to those who are at all conversant in the law of insurance, than that which exists, between a warranty or condition which makes part of 1 See, eg, R Merkin, Colinvaux’s Law of Insurance, 11th edn (Sweet & Maxwell 2017) ch 7; MA Clarke, The Law of Insurance Contracts, Looseleaf (Informa 2018), ch 20; and J Lowry, P Rawlings and R Merkin, Insurance Law: Doctrines and Principles, 3rd edn (Hart Publishing 2011) ch 7. 2 Admittedly, this is not to deny the influence of the legislative definition found in s 33(1) of the Marine Insurance Act 1906, which is now judicially accepted as applicable to all types of insurance: ‘A warranty … means a promissory warranty, that is to say, a warranty by which the assured undertakes that some particular thing shall or shall not be done, or that some condition shall be fulfilled, or whereby he affirms or negatives the existence of a particular state of facts.’ 3 See, eg, E Peel, Treitel on The Law of Contract, 14th edn (Sweet and Maxwell 2015); and E ­McKendrick, Contract Law: Text, Cases and Materials, 8th edn (OUP 2018) chs 8 and 22. 4 Typically incorporated into the agreement by a ‘basis of contract’ clause, a device first prohibited for consumer insurance by the Statements of Insurance Practice 1986 and now generally prohibited by the modern insurance legislation in 2012 and 2015, discussed in section V. 5 Traces of this approach to warranties are discernible in earlier case law. See, eg, Lethulier’s Case (1692) 2 Salk 443 and Woolmer v Muilman (1763) 1 W Bl 427. See further, WR Vance, ‘The Early History of Insurance Law’ (1908) 8 Columbia Law Review 1, 16.

404  John Lowry and Rod Edmunds a written policy, and a representation … Where it is part of the written policy, it must be performed … nothing tantamount will do, or answer the purpose; it must be strictly performed …6

Such judicial observations formulate what amounts to a warranty by reference to the need for strict compliance, a feature that finds an echo both in the codifying statute7 and in the denial of a need to demonstrate a causal link between breach and loss,8 or, indeed, any loss whatsoever.9 Arguably, the way insurance practice developed in later years marked a shift away from regarding warranties as purely indicative of the risk to be underwritten in favour of encompassing terms that had less connection to the insured risk – doubtless, in a bid to attract the harsh consequences attendant upon non-compliance. The judicial consideration of warranties received renewed impetus during the period spanning the latter part of the nineteenth century, continuing well into the twentieth century, partly in response to the growing sophistication of commercial practice, which was no longer confined to the business of shipping in far-distant oceans.10 It resulted, first, in judicial fashioning of rules aimed at avoiding a determination that a term was a warranty and, second, in the all-ornothing consequences of its breach that, in effect, privileged the interests of the insurer over those of the insured party. Our concern goes beyond questions of mere characterisation and lies squarely with the legal response that follows a breach of warranty. Unsurprisingly, the law’s direction of travel on these matters has not escaped controversy, which in the second half of the twentieth century prompted a number of calls for reform11 and intervention by law reform committees.12 Eventually legislative change came, but only midway through the second decade 6 Pawson v Watson (1778) 2 Cowp 785, 787–88. Here, written instructions described a vessel as having 12 guns and 20 men. Lord Mansfield CJ observed, at 790, ‘If a man warrants that a ship shall depart with twelve guns, and it departs with ten only, it is contrary to the condition of the policy.’ 7 The strict nature of the warranty in insurance law is provided for in the statutory definition of the term contained in s 33(3) of the Marine Insurance Act 1906: ‘A warranty … is a condition which must be exactly complied with, whether it is material to the risk or not …’ What has been changed by the Insurance Act 2015 is the repeal of s 33(3)’s second sentence: ‘If it be not so complied with … the insurer is discharged from liability …’ 8 See Yorkshire Insurance Co v Campbell [1917] AC 218; and Forsikringsaktieselskapet Vesta v Butcher [1989] AC 852. 9 This is not to deny the possibility of waiver by the insurers: see s 34(3) of the Marine Insurance Act 1906. See further Lord Goff ’s speech in The Good Luck [1992] 1 AC 233, 262–63, who explained that the terminology of estoppel is more appropriate. 10 This can be better understood when one brings to mind that by the 20th century, the need for insurance applied in a much wider range of commercial and non-business settings than had been the case in the maritime-dominated world of the 17th and 18th centuries. 11 For notable examples, see A Longmore, ‘An Insurance Contracts Act for a New Century?’ [2001] LMCLQ 356; A Longmore, ‘Good Faith and breach of warranty: are we moving forwards or backwards?’ [2004] LMCLQ 158; M Clarke, ‘Insurance Warranties: The Absolute End? [2007] LMCLQ 474; J Birds, ‘Warranties in Insurance Proposal Forms’ (1977) 21 Journal of Business Law 231; and RA Hasson, ‘The “basis of the contract clause” in insurance law’ (1971) 34 MLR 29. 12 Law Reform Committee, Fifth Report Condition and Exceptions in insurance policies (Cmnd 62, 1957); Law Commission, Insurance Law – Non-Disclosure and Breach of Warranty (Law Com No 104,1980).

The Reform of Insurance Warranties  405 of the twenty-first century. Here the law has shifted its position dramatically, from a resolute insistence on discharging the insurers from all liability to the modernday notion of responding proportionately to the insured’s breach as provided for by the Insurance Act 2015, sections 10 and 11.13 It is tempting to suppose that this long campaign resulted in significant alteration of the pre-existing order. However, it is arguable that such an assessment, if not premature, is superficial and that the position is far more nuanced. All of this is, of course, currently open to conjecture until we get judicial consideration of provisions that only entered into force in August 2016. It is instructive to start with the recent decision in Bluebon Ltd v Ageas (UK) Ltd (formerly Fortis Insurance Ltd).14 Although the case pre-dates the 2015 reforms, it neatly illustrates the length to which the courts had to go to avoid the severe consequences that attached to a finding that a policy term amounted to a warranty.

II.  Bluebon Ltd In Bluebon, the insured arranged insurance covering a wide range of risks, including fire damage, in respect of its hotel, the ‘Star and Garter’ in Linlithgow, West Lothian. The policy contained a term, the Electrical Installation Inspection Warranty (EIIW), which provided as follows: It is warranted that the electrical installation be inspected and tested every five years by a contractor approved by the National Inspection Council for Electrical Installation Contracting and that any defects be remedied forthwith in accordance with the Regulations of the Institute of Electrical Engineers.

The only inspection that had taken place was in September 2003. The policy commenced in December 2009. In October 2010 the hotel was destroyed by fire. The insurers declined the claim contending breach of warranty rendering the contract void ab initio, conceding, however, that they were liable to refund the premium only. Bryan J disagreed. He found that the EIIW’s five-year requirement was, as a matter of law, a suspensive condition rather than a warranty. This meant that the insurers were not on risk because the breach had not been remedied prior to the occurrence of the fire. The judge’s conclusion therefore favoured the 13 For the proximate antecedence of these immediate reforms, see Law Commission and Scottish Law Commission, Insurance Contract Law Issues Paper 2 Warranties (Law Com No 353, Scot Law Com No 238, 2006); Law Commission and Scottish Law Commission, Insurance Contract Law: The Business Insured’s Duty of Disclosure and the Law of Warranties (Law Com CP No 204/Scot Law DP No 155, 2012); Law Commission and Scottish Law Commission, Insurance Contract Law: Business Disclosure; Warranties; Insurers’ Remedies for Fraudulent Claims; and Late Payment (Law Comm No 353, Scot Law Com No 23, 2014). Admittedly, this was not the first time reforms of this ilk were proposed – see pt VI of Law Commission, Non-Disclosure and Breach of Warranty (ibid). See generally P MacDonald Eggers, ‘The Past and Future of English Insurance Law: Good Faith and Warranties’ [2012] UCL Journal of Law and Jurisprudence 211, 240. 14 Bluebon Ltd v Ageas (UK) Ltd (formerly Fortis Insurance Ltd) [2017] EWHC 3301 (Comm), [2018] Lloyd’s Rep IR 503.

406  John Lowry and Rod Edmunds insurers’ argument that in a one-year policy, the only commercially sensible interpretation of the term, and one which corresponded with the natural and ordinary meaning of its wording, is that the inspection should have taken place at fiveyearly intervals beginning with the last inspection rather than at the inception of the policy.15 The facts, of course, arose before the implementation of the Insurance Act 2015. This prompts questions about the impact the new law would have on the reasoning and the outcome. By classifying the EIIW as a suspensive condition the decision seems to contain shades of the new approach heralded by the Act.16 Equally, the reasoning adopted by the judge in Bluebon highlights the length to which the courts were driven to mitigate the draconian consequences that flowed from a finding that a term was, in fact, a warranty, and which, in large measure, inspired the Law Commissions’ reforms. In Bluebon, as with the line of insurance cases stretching back to the beginning of the last century, the Commercial Court saw its initial task as one of contractual interpretation. What was the nature of the EIIW? There were three possibilities canvassed in argument by the parties. For the insurers, the term was either, as described in the policy document, a warranty, simpliciter, relieving them from all liability, or, in the alternative, a suspensive condition resulting in cover being suspended from the inception of the policy.17 By way of response, the insured submitted that the ‘warranty’ was merely a term of the policy requiring compliance as a condition precedent to the insurers’ liability to provide cover (termed a ‘risk specific condition precedent’), meaning that liability could only be denied if there was non-compliance with the stipulation. So, unless the fire was caused by the electrical installation, the insured was on cover. That making this determination exercised much of the judicial reasoning is indicative of the complexities that bedevil the interpretation of insurance contracts. These distinctions are rarely easy to discern. The courts have fallen back on the rules of construction as one means of avoiding the unwanted consequences that necessarily follow a finding that a contractual term is a warranty. They have distilled interpretative aids to assist with the determination of the nature of a term. In recent times, Rix LJ’s trio of tests, laid down in HIH Casualty & General Insurance v New Hampshire Insurance Co, have held sway (as they did in Bluebon): One test is whether it is a term which goes to the root of the transaction; a second, whether it is descriptive of or bears materially on the risk of loss; a third, whether damages would be an unsatisfactory or inadequate remedy.18 15 In reaching his conclusion here, Bryan J drew upon a range of authorities harnessed by the courts in approaching the question of construction, notably Lord Neuberger JSC’s five principles in Arnold v Britton [2015] UKSC 36, [2015] AC 1619 [14]–[23] (with whom Lord Sumption and Lord Hughes JJSC agreed). 16 Notably, the Insurance Act 2015, s 11. 17 The classic case being Provincial Insurance Co v Morgan [1933] AC 240, discussed in section IV. 18 HIH Casualty & General Insurance v New Hampshire Insurance Co [2001] Lloyd’s Rep IR 702, [101]. See further Toomey (of Syndicate 2021) v Banco Vitalicio de Espana SA de Seguros y Reasseguros [2004] EWCA Civ 622, [2005] Lloyd’s Rep IR 423, where the Court of Appeal held that a term may be a warranty even though it is not expressed as such, provided it satisfies Rix LJ’s trio of tests.

The Reform of Insurance Warranties  407 Any notion that this injects certainty into the interpretative process is somewhat illusory. Indeed, Bryan J found that the EIIW arguably satisfied all three tests and yet was not a warranty.19 This then touches a central line of enquiry: what shift in direction, if any, has insurance contract law taken? Certainly, the decision in Bluebon is far from novel. It represents what had become a fairly settled response to an insurers’ plea before the Insurance Act 2015.

III. Antecedence There has been considerable discourse over how the warranty as a fundamental term first came to appear in marine insurance contracts. The mechanics of concluding a policy cannot be overlooked here insofar as the nature of the risk being presented by the broker to the underwriter at the Lloyd’s Coffee House was by the standards of the day highly speculative, necessitating precise delineation. As a matter of custom, certain terms, such as ‘sea-worthiness’ and ‘no-deviation’, entered into common usage not as a result of the creativity of the lawyers of the day or the courts, but rather by virtue of the insurance professionals seeking to protect themselves from events beyond their control.20 Their anxiety and ambition was to impose a straitjacket on the insured, guaranteeing strict compliance with the policy terms. This objective undoubtedly contributed to the stance adopted by the courts. It took judicial willingness to translate these interests of the underwriter into strictly applied legal principles invariably applied to the disadvantage of the insured. Although the origins of the law on warranties is marked by a certain tentative approach towards embedding the idea of strict compliance,21 by 1786 the opportunity was taken to settle the law on a strict footing. In De Hahn v Hartley,22 the policy for a voyage between Africa and the West Indies warranted that the insured ship would have ‘fifty hands or upwards’ on its departure from Liverpool. However, there were 46 hands on board for the short journey between Liverpool and Anglesey, where six additional crew were taken on board. That number remained with the vessel until it was lost in the Atlantic. The underwriters successfully pleaded breach of warranty even though the breach had no bearing on the risk. Questions of materiality were adjudged to be beside the point. Ashurst J stated that ‘the very meaning of a warranty is to preclude all questions whether

19 Bluebon Ltd (n 14) [158]. 20 See, eg, VR Vance, ‘The History of the Development of the Warranty in Insurance Law’ (1911) 20 Yale Law Journal 523; and EW Patterson, ‘Warranties in Insurance Law’ (1934) 34 Columbia Law Review 595. 21 For evidence of judicial incrementalism towards the development of warranties, see Pawson v Watson and Bean v Stupart (1778) 1 Doug 11. 22 De Hahn v Hartley (1786) 1 TR 343.

408  John Lowry and Rod Edmunds it has been substantially complied with; it must be literally so’.23 In a similar vein, Lord Mansfield observed that ‘It is perfectly immaterial for what purpose a warranty is introduced; but being inserted, the contract does not exist unless it be literally complied with.’24 Admittedly here the judicial focus is bifurcated. One gives emphasis to the anterior enquiry requiring identification of the nature of the term and the second, which is closely aligned, considers the consequences of breach. On this latter point, Lord Mansfield was moved to note in De Hahn that the policy was void.25 Over the next quarter of a century or so, Lord Mansfield’s stance on literal compliance with a warranty became accepted as the orthodoxy, not solely for shipping insurance but also in relation to non-marine indemnity policies. Take, for example, fire insurance. In Newcastle Fire Insurance Co v Macmorran & Co,26 cotton and wool spinners insured their premises, warranting them to be ‘first class’ with stoves not more than one foot from the wall, with pipes or flues not more than two feet in length. On 7 December 1805 the mill was destroyed by fire, for which the insurers successfully repudiated liability on the basis of breach of warranty. Finding that the insured premises were ‘second class’ at the time the policy was concluded, Lord Eldon LC remarked: It is a first principle of the law of insurance that, in the case of a warranty, the thing must be exactly as it is represented to be. The only question in such cases is merely as to the fact whether the thing was or was not as represented.27

Once again questions of materiality were rendered irrelevant. Moreover, the insurers’ defence was upheld, notwithstanding that at the time of the fire the breach had been remedied.28 This strikingly illustrates the harshness of the requirement of literal compliance. It proved representative of much of the law and practice at the time. Which is not to suggest that this approach was all-pervasive. Even Lord Mansfield, it might be argued, could, on occasion, be relied upon to temper the absolute nature of the prevailing view. A vivid illustration can be found in

23 ibid 346. 24 ibid. 25 ibid. The reference here to the breach rendering the contract ‘void’ reflected the judicial thinking at that time to the effect that this was consistent with the intention of the contracting parties: see, eg, Dawsons Ltd v Bonnin [1922] 2 AC 413. This position was not reflected in the Marine Insurance Act 1906, s 33(3), which stated that in the event of non-compliance, ‘the insurer is discharged from liability as from the date of breach of warranty’ (this has been substantially reformed by the Insurance Act 2015, s 10 (discussed in section V)). For an insightful analysis of this point in its broader historical context, see M Lobban ‘The Law of Insurance’ in W Cornish et al (eds), The Oxford History of the Laws of England 1820–1914 (OUP 2010) 674. 26 Newcastle Fire Insurance Co v Macmorran & Co (1815) 3 Dow 255. 27 ibid 262. 28 The position here was codified by s 34(2) of the Marine Insurance Act 1906, which provided ‘Where a warranty is broken, the assured cannot avail himself of the defence that the breach has been remedied, and the warranty complied with, before loss.’

The Reform of Insurance Warranties  409 contingency insurance. In Ross v Bradshaw,29 the policy on the life of the insured, Sir James Ross, commenced in October 1759 warranting that he was then ‘a good life’. In fact, some 12 years earlier he had been wounded in the battle of La Feldt (Lauffeld). His injuries triggered palsy and incontinence. Yet nothing of these facts had been disclosed to the insurers. Eleven months into the policy, the insured died, having succumbed to malignant fever. Finding there was no breach of the duty of disclosure in the light of the facts warranted, Lord Mansfield turned to the warranty and told the jury: But it must, in general, be proved, if litigated, that the life was in fact a good one: and so it may be, though he had no particular infirmity. The only question is, whether he was in a reasonably good state of health, and such a life as ought to be insured on common terms.30

What is particularly noteworthy in this passage, is how Lord Mansfield’s recourse to reasonableness in terms of assessing ‘good state of health’ injects a note of equivocation, if not leniency, into the test for determining compliance that is at odds with his more robust utterances seen in Pawson and De Hahn. Arguably, this less absolutist approach presages developments seen throughout the twentieth century and, at the very least, as Bluebon Ltd illustrates, continues to cast its shadow right up until the statutory reforms in 2015.

IV.  Contractual Construction: An Exercise in Judicial Contortions? With the advent of what can be compendiously described as modern indemnity and contingency insurance in the early years of the twentieth century, the judges became ever more creative in overcoming the approach doggedly adhered to in the early shipping cases towards the inviolability of the warranty. Their creativity manifested in a number of forms. One involved the deployment of diverse canons of construction: the particular focus being to resolve apparent ambiguities.31 A  second involved the courts in frustrating the ambition behind the ‘basis of contract’ clause that had become increasingly common place in proposal forms. These had afforded insurers an easy and furtive means of creating warranties by incorporating pre-contractual questions and the answers to them into the

29 Ross v Bradshaw (1761) 1 Black W 312, 96 ER 175. More fully reported elsewhere, see, eg, Marsh Ins 770 (ed 1808); Park’s Ins 649 (ed 1817). 30 ibid 313. 31 See n 15. Both marine and non-marine insurance attract the same rules of construction that operate in the general law of contract: Thomson v Weems (1884) 9 App Cas 671. For more recent authority, see. Rainy Sky SA v Kookmin Bank [2011] UKSC 50, [2012] 1 Lloyd’s Rep 34; Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173.

410  John Lowry and Rod Edmunds insurance contract.32 Many of the early twentieth-century cases should be viewed in this context, because the practice of cross-incorporation of terms attracted particularly heavy criticism. What is clear from these endeavours is that the courts were able to fashion the idea of the suspensory condition.33 Here, breach of such a term merely results in cover being suspended until such time, if at all, as it is remedied; the judicial objective being to craft what may be termed an ‘equitable’ resolution of the dispute, one which leans in favour of the insured.34 Before the court was prepared to accept a term as a warranty, it was insistent that the language was framed unequivocally, demonstrating the clear intention to create a fundamental term (with all the consequences that entailed). Labelling was not material. The absence or adoption of the word ‘warranty’ was not, therefore, decisive. This is captured in HIH Casualty and General Insurance Ltd.35 Here the question was whether a policy term to which no label had been given, stating that six films were to be made, was a warranty. HIH had provided insurance against the event that the filmmakers would not generate sufficient income to repay a funding loan it had received from Chase Manhattan Bank. Rix LJ decided that it was a fundamental term, so that there was a breach if only five films were produced because there would be a loss of revenue. Laying down the modern approach applicable to the determination of an insurance term,36 he observed that ‘It is a question of construction, and the presence or absence of the word “warranty” or “warranted” is not conclusive.’37 The decision is hardly surprising and may be defended on the basis that the term encapsulated the whole risk that the insurer was underwriting. As such, it represents the epitome of a warranty in its pure and least objectionable form. It has been observed that ‘nothing turns upon the use of the word “warranted” [which is] always used with the greatest ambiguity in a proposal’.38 Other factors may assist in classifying and even mitigating the effects of the interpretation of contractual terminology used in the policy. One such guiding principle can be

32 See Anderson v Fitzgerald (1853) 4 HL Cas 484. 33 It is worth sounding a note of caution, because the judges are not uniform in the term used in this connection. References include variations such as: ‘suspensive warranty’; ‘description of risk’ clause; and, particularly in some of the older cases, ‘risk delimiting clause’ or ‘exceptions’. See further, Morland J in Kler Knitwear Ltd v Lombard General Insurance Company Limited [2000] Lloyd’s Rep IR 47, 49–50, citing with approval the words of Mr Justice Lambert in the Canadian case of Case Existological Laboratories Ltd v Century Insurance Co of Canada (‘The Bamcell II’) [1986] 2 Lloyd’s Rep 528, 533. 34 Patterson, ‘Warranties in Insurance Law’ (n 20) 623. 35 HIH Casualty & General Insurance (n 18). 36 See Rix LJ’s trio of tests (ibid). 37 ibid [101]. See further, Roberts v Anglo-Saxon Insurance Ltd (1927) Ll L Rep 313 (CA). 38 Roberts v Anglo-Saxon Insurance Ltd (1926) 26 Ll L Rep 154 (QB) 157 (MacKinnon J). In Virk v Gan Life Holdings plc [2000] Lloyd’s Rep IR 159, 165, Potter LJ observed that ‘In relation to the exercise of construction to be undertaken by the court, the question of “labelling” is influential rather than decisive, particularly if the label … has been applied to a number of terms of differing type and practical importance. On the other hand, if care and logic appear to have been applied in the attachment of the label to one term but not to another, the label (or absence of it) is likely to be decisive.’

The Reform of Insurance Warranties  411 found in AC Ward & Sons Ltd v Caitlin (Five) Ltd,39 where the Court of Appeal noted that the more unreasonable or draconian the effects of an insurance term, the clearer and more specific its wording had to be. This is evident in the approach of the House of Lords in Provincial Insurance Co Ltd v Morgan.40 Here the warranty, which derived from answers to questions contained in the proposal form, stated that the insured vehicle was to be used for the delivery of coal. However, in fact, it was also used for the haulage of timber, as was the case at the time of the accident. The insurers had repudiated liability. Finding in favour of the insured, Lord Buckmaster said: I wish again to repeat that it is perfectly open to people to make such a bargain, and when made it is useless to complain that the bargain is harsh. But it is at least essential that the bargain should be plain in order that it may be clear that a man has contracted on the faith of something which may rob the insurance of the greater part of its value … To state in full the purposes for which the vehicle is to be used is not the same thing as to state in full the purposes for which the vehicle will be exclusively used, and as a general description of the use of the vehicle it is not suggested that the answer was inaccurate.41

This view is echoed in Lord Russell’s assessment that the insured’s answers could be read as nothing more than mere statements of ‘intentions as to the use of the vehicle and the goods to be carried in it’.42 Therefore, what was stated in the proposal form did not amount to a warranty but was merely descriptive of the risk insured. This typifies the extent to which the courts were driven to identify and apply nice distinctions to sidestep the unwanted effects that attach to warranties sensu stricto. Yet, perhaps paradoxically, it reveals the true nature and significance that a warranty should assume: that is, it should go to the root of the risk that is being underwritten. This response of finding a term ‘descriptive of risk’43 adds complexity to the legal landscape. However, it has a considerable lineage. Farr v Motor Traders’ Mutual Insurance Society44 aptly illustrates the point. The disputed term of a motor policy provided that each of the insured’s two taxis was only to be driven in one shift each 24 hours. Subsequently, one of the taxis was driven for two shifts during the day while the other was under repair. At the time of the claim there was, however, due compliance with the term. The insurers effectively argued that

39 AC Ward & Sons Ltd v Caitlin (Five) Ltd [2009] EWCA Civ 1098, [2010] Lloyd’s Rep IR 301. 40 Provincial Insurance Co v Morgan (n 17). 41 ibid 247 (approving the reasoning of Bankes LJ in Roberts v Anglo-Saxon Insurance Ltd (CA) (n 37) 314). Echoing Lord Buckmaster’s sentiment, Lord Wright said, at 255, ‘A policy ought to be so framed, that he who runs can read. It ought to be framed with such deliberate care, that no form of expression by which, on the one hand, the party assured can be caught, or by which, on the other, the company can be cheated, shall be found upon the face of it.’ 42 Provincial Insurance v Morgan (n 17) 278. 43 See n 33. 44 Farr v Motor Traders’ Mutual Insurance Society [1920] 3 KB 669.

412  John Lowry and Rod Edmunds as a warranty this was irrelevant so that their liability was discharged. The Court of Appeal, upholding the trial judge’s finding, disagreed. It was held that the term was merely descriptive of risk, so that cover was suspended during the time when the vehicle was driven in two shifts but had resumed at the time of the claim.45 The judgment offers little assistance in establishing the nature of the distinction and how it is to be drawn. As Professor Birds has pointed out,46 the result in Farr might have been different had the wording of the disputed term been slightly more nuanced. In sum, the approach taken here makes the law less than predictable. On occasion the courts have taken the opportunity to broaden the line of attack on attempts by insurers to elevate all statements to the status of warranties by criticising the practice of cross-incorporation of terms via basis of contract clauses. For example, in Provincial Insurance, Lord Wright went on to say: Though this general scheme of policy has been, as it were, sanctified by long usage, it has often been pointed out by judges that it must be very puzzling to the assured, who may find it difficult to fit the disjointed parts together in such a way as to get a true and complete conspectus of what their rights and duties are and what acts on their part may involve a forfeiture of the insurance. An assured may easily find himself deprived of the benefits of the policy because he has done something quite innocently but in breach of a condition, ascertainable only by the dovetailing of scattered portions.47

This practice, to use Lord Wright’s words, of ‘dovetailing … scattered portions’ came to the fore in Printpak v AGF Insurance Ltd,48 in a different guise that involved a multi-sectioned policy, that is, one divided into separate policy schedules. The issue was whether the contract should be construed as a whole rather than treating its schedules as distinct, so that a breach of warranty contained in one schedule would enable the insurers to repudiate the contract even though the particular risk in question was covered by another schedule in the policy. Hirst LJ explained that it does not follow from the fact that the policy is a single contract that it is to be treated as a seamless contractual instrument. On the contrary, in the present case, its whole structure is based on its division into sections … The commercial inclusive endorsements are all stated in terms to be ‘operative only as stated in the policy schedules.’ In my judgment those words explicitly write the warranty into the relevant section in which it appears and not into the others …49

Throughout these judicial contortions there has, of course, been reliance upon traditional rules of construction such as the contra proferentem, strictly construing 45 Such a term would now fall within the scope of the Insurance Act 2015, s 11. 46 J Birds, ‘Warranties in Insurance Proposal Forms’ [1977] Journal of Business Law 231, 232–33. 47 Provincial Insurance Co Ltd v Morgan (n 17) 252. 48 Printpak v AGF Insurance Ltd [1999] Lloyd’s Rep IR 542. See J Davey, ‘Printpak v AGF Insurance: no need to be alarmed?’ [1999] Journal of Business Law 580; cf James v CGU Insurance plc [2002] Lloyd’s Rep IR 206. 49 Printpak v AGF Insurance Ltd (n 48) 546.

The Reform of Insurance Warranties  413 any ambiguity in the wording of a policy against the party, typically the insurer, who is claiming its benefit. The onus is therefore squarely on the insurers to express a warranty in plain language or otherwise suffer the consequence of its being interpreted as a lesser term, one that the insured might reasonably have understood it to be. On occasion we see the judges employing less than temperate language in their assault upon ‘slovenly’ contract drafting designed to ‘entrap’ the insured.50 In Re Bradley and Essex and Suffolk Accident Indemnity Society,51 Farwell LJ, commenting on warranties, albeit using the terminology of the general law of contract, emphasised: It is especially incumbent on insurance companies to make clear, both in their proposal forms and in their policies, the conditions which are precedent to their liability to pay, for such conditions have the same effect as forfeiture clauses, and may inflict loss and injury to the assured and those claiming under him out of all proportion to any damage that could possibly accrue to the company from non-observance or non-performance of the conditions. Accordingly, it has been established that the doctrine that policies are to be construed ‘contra proferentes’ applies strongly against the company.52

The point is graphically illustrated in Kennedy v Smith and Ansvar Insurance Co Ltd,53 where the insured warranted that he was ‘a total abstainer from alcoholic drinks and have been since birth’. After a bowling match he visited a pub. He was then involved in an accident in which his two passengers were killed. The underwriters refused to indemnify him on the basis that at the time of the accident his ‘warranty’ that he had been a total abstainer from birth was untrue, and that when the claim arose he was under the influence of intoxicating liquor. The Lord President rejected the underwriters’ claim to repudiate liability in robust terms: [I]f insurers seek to limit their liability under a policy by relying upon an alleged ­undertaking as to the future prepared by them and accepted by the insured, the language they use must be such that the terms of the alleged undertaking and its scope are clearly and unambiguously expressed or plainly implied and that any such alleged undertaking will be construed, in dubio, contra proferentem.54

The myriad judicial contortions adopted to avoid finding a warranty, which, as seen in section II, was extensively explored in Bluebon, underline the preference for seeing policy terms as suspensive conditions. The high-water mark for

50 See Zurich General Accident and Liability Insurance Co Ltd v Morrison [1942] 2 KB 53, 57–58. 51 Re Bradley and Essex and Suffolk Accident Indemnity Society [1912] 1 KB 415. 52 ibid 430. See further Notman v Anchor Assurance Co (1858) 4 CB NS 476, 481 (Lord Cockburn CJ); Fowkes v Manchester and London Life Assurance and Loan Association (1863) 3 B & S 917, 929 (Blackburn J); Cook v Financial Insurance Co Ltd [1998] 1 WLR 1765, 1771 (Lord Lloyd); John A Pike (Butchers) Ltd v Independent Insurance Co Ltd [1998] Lloyd’s Report IR 410. 53 Kennedy v Smith and Ansvar Insurance Co Ltd [1976] SLT 110 (First Div). 54 ibid 116–17. For further illustrations of the adoption of strict construction working against the interests of the insurers, see Bonney v Cornhill Insurance Co Ltd (1931) 40 Ll LR 39; and Pratt v Aigaion Insurance Co SA [2008] EWHC 489 (Admlty), [2008] Lloyd’s Rep IR 610.

414  John Lowry and Rod Edmunds this tendency is the decision in Kler Knitwear Ltd v Lombard General Insurance Co Ltd.55 There, a policy term required the insured’s sprinkler system to be inspected by an approved engineer within 30 days of contract renewal. In fact, it was inspected after 60 days. A claim was made to recover for losses resulting from storm damage. Had it been construed as a warranty, the insurers would have escaped liability. However, Morland J came to the ‘clear and unhesitating conclusion’ that it was merely a suspensive condition,56 so that liability re-attached after compliance and before the event that gave rise to the claim. This unequivocal stance is open to question. The reasoning that led to the conclusion here lacks transparency, and the judge’s categorisation is puzzling. The commercial parties had agreed to tightly drafted terms that included General Condition 2: [E]very Warranty to which this Insurance or any Section thereof is or may be made subject shall from the time the Warranty attaches apply and continue to be in force during the whole currency of this Insurance and non-compliance with any such Warranty; whether it increases the risk or not, or whether it be material or not to a claim, shall be a bar to any claim in respect of such property or item, provided that whenever this Insurance is renewed a claim in respect of Damage occurring during the renewal period shall not be barred by reason of a Warranty not having been complied with at any time before the commencement of such period.

It is difficult to see how the insurers could have achieved greater specificity in defining their objective. The insurer clearly identifies the nature of the term by stating what would result from its breach. Morland J, however, preferred to emphasise that General Condition 2 focused on the consequences of non-compliance, the absence of a definition of a warranty in the policy and the dictates of ‘rational business sense’.57 As Bluebon makes clear, by the early twenty-first century the judicial willingness to mitigate and avoid the harsh consequences attendant upon finding a warranty had traversed the dichotomy between consumer and commercial insurance. Admittedly, this willingness has not always led to an ‘equitable’ outcome to the advantage of the insured. However, much of the case law preceding the recent legislative reforms is indicative of the lengths to which the courts were prepared to go when striking a nuanced construction of policy terms. This judicial insistence on forensic deconstruction of policy language is also evident in Sugar Hut Group v Great Lakes Reinsurance.58 Here, the insureds operated four nightclubs in and around London, one of which was substantially damaged by fire in September 2009. The insurers denied the claim on a number of grounds, including breach

55 Kler Knitwear Ltd v Lombard General Insurance Co Ltd [2000] Lloyds Law Rep 47. See also De Maurier (Jewels) Ltd v Bastion Insurance Co [1967] 2 Lloyd’s Rep 550; and CTN Cash & Carry v General Accident [1989] 1 Lloyd’s Rep 299. 56 Kler Knitwear Ltd v Lombard General Insurance Co Ltd (n 55) 55. 57 ibid. 58 Sugar Hut Group v Great Lakes Reinsurance [2010] EWHC 2636 (Comm).

The Reform of Insurance Warranties  415 of warranty.59 The court was confronted with a range of contractual obligations that required close examination. For present purposes, two of the policy terms are pertinent. One, described in the policy as a ‘warranty’, required, first, that kitchen ducting be kept free of contact with combustible materials and, second, its inspection ‘at least once every 6 months’. Having sub-divided the ‘warranty’ into two, Burton J had no difficulty in finding a breach of the combustible materials element, irrespective of the fact that this was not causative of the fire. Moreover, he resisted counsel’s submission that the breach should be overlooked on the application of the de minimis principle.60 As to the second part, requiring inspection of the duct, the judge found that on its true construction it was a suspensive condition, not a warranty. Ultimately, the need to draw such a distinction was found to be inconsequential because the fire took place when cover would have been suspended. By contrast, a second ‘warranty’ in the policy, which required the installation and operation of a specific Central Monitoring Station Alarm, was held to be a ‘true’ warranty that, on the facts, had been breached.61 This involved a textbook application of Rix LJ’s three tests in HIH Casualty & General Insurance v New Hampshire Insurance Co,62 because the need for the alarm ‘was of fundamental importance’ to the policy and ‘was significantly material to the risk of loss’, and, once again, it was irrelevant that the absence of such an alarm did not cause the loss. All of this demonstrates the futility of underwriters’ declaring, by default, each and every term to be fundamental to the risk, and the consequential judicial contortions required to achieve an ‘equitable’ balance between the competing interests of the parties. It also provides an insight into the consensus that the law was short on transparency and inaccessible. While the urgent need for reform was widely accepted, deciding the best way forward proved more contentious. Fast-forwarding to the 2015 Act, it neither abolishes nor defines ‘warranty’. Abrogating the all-ornothing approach to breach, it provides a proportionate response. In addition, the new legislation is designed to prevent the reliance by insurers upon breaches that are not connected to the insured risk. Even if these reforms reduce the need for techniques of construction that occupied such extensive consideration in the case law, they will therefore continue to be relevant in some measure or other, particularly in relation to determining whether a policy term is a warranty. The statute, to which we now turn, introduces a fresh source of potential contention in relation to terms that do not define the risk as a whole. Inevitably, this will require the exercise of judicial interpretation.

59 The other grounds were non-disclosure and reliance on a condition precedent. 60 Burton J accepted (Sugar Hut (n 58) [44]) that there was no decision applying de minimis to excuse the breach of an insurance warranty, but he did acknowledge the existence of obiter references to that end (citing McNair J in Overseas Commodities Ltd v Style [1958] 1 Lloyd’s List Law Rep 546, 557; and Tomlinson J in Bennett v AXA Insurance plc [2004] Lloyd’s Rep IR 615, 620). 61 Kler Knitwear Ltd v Lombard General Insurance Co Ltd (n 55) [49]. 62 HIH Casualty & General Insurance (n 18).

416  John Lowry and Rod Edmunds

V.  Insurance Act 2015: A Brave New World? The Insurance Act 2015 implements most of the recommendations put forward by the Law Commissions.63 As noted in section I, sections 10 and 11 address the mischief of the common law,64 by rendering the consequences of a breach of warranty less severe.65 In particular, warranties will have the effect of ‘suspensive conditions’. This is captured in section 10(2): An insurer has no liability under a contract of insurance in respect of any loss occurring, or attributable to something happening, after a warranty (express or implied) in the contract has been breached but before the breach has been remedied.

Consequently, once the breach has been remedied the insurer’s liability for losses will resume: a breach of warranty will no longer automatically discharge the insurers’ liability.66 As such, this appears to sweep away the contortions that had so long exercised the judges when they harnessed rules of construction to neutralise claims that a term was a warranty in the various ways explored in our previous section and which had so dominated the judgment in Bluebon. Nonetheless, on closer inspection this area of judicial activity may become supplanted by new lines of enquiry that part 3 (‘Warranties and other terms’) of the 2015 Act have opened up. For instance, while section 10(5)(b) defines the date when a breach of warranty is deemed to have been remedied, the appearance of certainty that this seems to establish will, on occasion, prove to be illusory and problematic. The Law Commissions recognised as much in their 2014 Report, noting that the courts will continue to utilise a functional approach when assessing whether a breach has come to an end.67 More fundamentally, there are going to be instances when the need for a judicial diagnosis of the nature of a term, including whether it is a warranty in the true sense or some other lesser term, will still be required. For one thing, this is because in the case of non-consumer insurance contracts, it is now open to the parties to contract out of section 10.68 In effect this means that they may provide for automatic discharge for breach of a policy term, and therefore import the consequences of breach of warranty that the courts were at such pains to avoid. Arguably this is entirely unobjectionable

63 Law Commission, Insurance Contract Law Issues Paper 2 Warranties (n 13). 64 s 9(2) of the Act abolishes ‘basis of the contract’ clauses in non-consumer contracts. Such clauses, or any other attempt to give pre-contractual representations the status of warranties by way of a provision in the insurance contract, will not be valid and it will not be possible to contract out of this prohibition. For consumer insurance policies, such clauses are prohibited by the Consumer Insurance (Disclosure and Representations) Act 2012, s 6. 65 Such a term must be clear and unambiguous and brought to the attention of the insured (ss 16–17). 66 s 10(3) prevents an insurer from avoiding an insurance contract if a warranty ceases to be applicable to the circumstances of the contract due to a change of circumstances, or if it is rendered unlawful by a subsequent change of law or is waived by the insurer. 67 See the Law Commission, Insurance Contract Law Issues Paper 2 Warranties (n 13) paras 17.38–17.39. 68 And also s 11 – but not s 9 (see s 16(1)).

The Reform of Insurance Warranties  417 in so far as it places the matter within the realm of freedom of contract and any power imbalance between the parties being addressed, at least in part, by the statutory requirement that ‘the insurer ‘must take sufficient steps to draw the disadvantageous term to the insured’s attention before the contract is entered’.69 Although the provision refers to the ‘insured’, typically in the context of business insurance the insured is represented by a specialist broker in its dealings with underwriters, which goes to make the facility for contracting out seem entirely comprehensible. Even so, it remains to be seen whether the prerequisites will trigger a new sphere of legal challenge around issues such as determining (i) when a term is disadvantageous; and (ii) the sufficiency of the steps taken to alert the insured to the presence of the term. Doubtless there is yet more fruitful scope for other avenues of judicial activity elsewhere in this part of the legislation, most notably section 11, which provides: (1) This section applies to a term (express or implied) of a contract of insurance, other than a term defining the risk as a whole, if compliance with it would tend to reduce the risk of one or more of the following: (a) loss of a particular kind; (b) loss at a particular location; (c) loss at a particular time. (2) If a loss occurs, and the term has not been complied with, the insurer may not rely on the non-compliance to exclude, limit or discharge its liability under the contract for the loss if the insured satisfies subsection (3). (3) The insured satisfies this subsection if it shows that the non-compliance with the term could not have increased the risk of the loss which actually occurred in the circumstances in which it occurred. (4) This section may apply in addition to section 10.

The focus here, therefore, is on ‘increase of risk clauses’, including terms designed to reduce the risk of a particular type of loss, or of loss at a particular time or in a particular location.70 Such terminology might suggest that warranties, as terms that define the risk as a whole, are not within its scope.71 This, however, is not necessarily the case: and an interrelationship between sections 10 and 11 is explicitly indicated in section 11(4). So, where a warranty is suspended by virtue

69 Insurance Act 2015, s 16(2). 70 s 11(4) provides this provision may apply in addition to s 10. Thus, a broken warranty that has not been remedied at the time of the loss will be within s 11. Although it is not beyond doubt (see Law Commission, Insurance Contract Law Issues Paper 2 Warranties (n 13) para 18.35), it appears that as a matter of principle, s 11(1) does not apply to terms drafted as exclusion clauses, as by delimiting the scope of cover such clauses may be taken to be defining the risk as a whole. Besides, they are not terms with which the insured must comply, and therefore they do not fall within the ambit of s 11(1). 71 Take a policy with a warranty requiring a burglar alarm to be maintained while the premises are left unattended. If loss is suffered as a result of fire,71 there would be no recovery at common law. By contrast, s 11 necessitates interpreting whether the alarm requirement defines the risk as a whole, in which case it is outwith the scope of the provision. See Milton v Brit [2015] EWCA Civ 671, [2016] Lloyd’s Rep IR 192.

418  John Lowry and Rod Edmunds of section 10, but the loss occurred before the breach was remedied, it may be possible for the insured to rely upon section 11. These are matters that may occupy the courts in future. In similar vein, section 11(3) looks to present particularly demanding challenges. On the one hand, it seems clear that if a term requires the installation of a burglar alarm and the insured fails to do this, the insurer will not be able to deny an indemnity on the basis of loss caused by flooding. However, the application of this provision may not always be so straightforward, and this is likely to generate a significant number of disputes and litigation. To give but one potentially problematic scenario. Take a household policy that contains a clause warranting that all outside doors should be locked by five bolt locks at night. Thieves break down the locked door, which is found to have only a three bolt lock, and they rob the house. In such circumstances, can the insured argue that the thieves were so well equipped that whatever precautions had been taken by way of locks, they would still have got in? Many other examples have already been contemplated by the Law Commissions,72 and in all probability there are others yet to be envisaged. Underlying all this is the fundamental area of uncertainty raised by section 11 – that is the extent to which, if at all, it injects ideas of causation into the law. The Law Commissions did not intend this to happen.73 In essence, they saw the provision, in its original manifestation, as preventing ‘reliance on breach of blatantly irrelevant warranties in order to escape liability for an unconnected loss’.74 Be that as it may, the drafting of the section leaves leeway for arguments that could nonetheless result in the introduction of a causation requirement via the back door. In their own assessment of the combined impact of the recommendations that ultimately became section 10 and the precursor to section11, the Law Commissions saw a reduction in the potential for dispute: We consider that overall there will be an increase in certainty for both parties. The new remedy for breach of warranty, together with the ‘type of loss’ recommendations should channel litigants on both sides towards less speculative arguments. At the point of risk assessment, insurers will know that certain terms are to be taken to have specific purposes and will only affect certain types of risk. Insurers therefore can, where necessary, take this into account in their pricing. It will also give policyholders some confidence when making an insurance claim and defending a breach of an irrelevant condition, as they will no longer have to rely on a particularly favourable interpretation of the contract from the courts.75

It remains to be seen how optimistic this is. In our view, there is much to suggest that the courts will play as prominent role in the ‘brave new world’ ushered in by the reforms as they did following Lord Mansfield’s foundational rulings.



72 See

the Law Commission Insurance Contract Law Issues Paper 2 Warranties (n 13) para 18.10. para 18.26. 74 ibid para 18.20. 75 ibid para 18.50. 73 ibid

The Reform of Insurance Warranties  419 Going back to Bluebon, Bryan J’s reasoning reflects the suspensive nature of warranties as they are now to be interpreted under the 2015 Act. Looking forward, the advent of section 10 may well downplay the need for future courts to engage in the contortions of the past as a means of avoiding automatic discharge of the insurers’ liability to pay for the loss. However, this may be replaced by new, and equally challenging, questions of construction surrounding which terms are designed to reduce loss of a particular kind. This may not have been problematic on the facts of the case, because presumably the insured could have successfully argued that non-compliance with the EIIW requirement could not have increased the risk of that fire occurring, and that therefore the insurers would have been liable to the claim. The point is not beyond dispute, and there are likely to be many instances where, as we have argued, the parties will inevitably be driven to test the parameters of section 11 in court.

420

20 The Right to Delivery of Goods under Contracts of Carriage MELIS ÖZDEL

Carriage of goods by sea is a ‘commercial’ activity, and non-shipping lawyers might be forgiven for assuming that the contractual terms of the sea carriage are based on freedom of contract. Astonishingly for non-shipping lawyers, this specialist area of commercial law is largely regulated by a patchwork of different sets of mandatory rules contained in various international conventions and national legislation.1 The historical reason behind the conventions was the onesided nature of the carriage contracts contained in or evidenced by bills of lading or other documents of title: these contracts habitually contained carriers’ exclusion clauses to free themselves of all liabilities whatsoever.2 The conventions provide a minimum and mandatory protection for ‘cargo interests’3 and strike a balance between the conflicting interests of carriers and cargo interests. The net result of the mandatory application of the Hague and the Hague-Visby Rules, which govern most bills of lading in sea carriage, is that any contract term that has the effect of removing or lessening the carrier’s contractual duties and obligations is struck down by Article III, rule 8 of the Rules. It must be noted, however, that the limitation of freedom of contract provided by these conventions also has its own limits: the mandatory application of the

1 Under English law, the Hague-Visby Rules are implemented with the enactment of the Carriage of Goods by Sea Act 1992 (COGSA 1992), which contains some minor alterations. Section 1(3) of the Act provides specifically that the amended rules have the force of law. Following the Act, their mandatory effect was confirmed by the House of Lords in Owners of Cargo on Board the Morviken v Owners of the Hollandia (The Morviken) [1983] 1 AC 565, [1982] 1 Lloyd’s Rep 1. 2 For a historical background to the Hague, the Hague-Visby and the Hamburg Rules, see M Özdel, ‘The EU and the Carriage of Goods by Sea under Private Law and EU Regulation’ in H Jessen and MJ Werner (eds), EU Maritime Transport Law (Hart Publishing 2016) paras 2 et seq. 3 In this chapter, the phrase ‘cargo interest(s)’ is used to refer to parties who have interest in the goods carried by sea, whether or not they have a contract of carriage with the sea carrier.

422  Melis Özdel Hague and the Hague-Visby Rules is confined to the well-known ‘tackle-to-tackle’4 period.5 Thus, once the goods are discharged from the vessel, freedom of contract allows parties to regulate their rights and obligations under the contract. This point is important, particularly for present purposes: in carriage of goods by sea, delivery and discharge are not interchangeable concepts. The contract of carriage by sea is a combined contract of transportation and bailment.6 For the contract to be discharged by performance, discharge of the goods, which marks the end of sea carriage, is not sufficient. As bailees, the carriers are required to surrender possession of the goods by divesting themselves of all power to compel any physical dealing with the goods.7 Usually, delivery occurs after discharge, that is outside the Hague8 or Hague-Visby sphere of responsibility.9 Consequently, the carriers’ rights and obligations in respect of the goods from the point at which they are discharged until they are ‘legally’10 delivered do not in principle form part of the Hague-Visby regime. This situation directly affects the carriers’ delivery obligation, leaving them with the freedom to agree more onerous terms in respect of their responsibility for the cargo and its delivery.11

4 On the application of the tackle-to-tackle rule, see Pyrene v Scindia [1954] 2 QB 402 (QBD). The ‘tackle-to-tackle’ rule does not always apply in a strict sense, particularly where the contract of carriage contains a FIOST (free in, out, stowed and trimmed) or a similar clause. The rules do not, therefore, require the carrier to take responsibility for loading and discharging. This is a matter for the parties’ contract: see Jindal Iron & Steel Co Ltd v Islamic Solidarity Shipping Co Jordan Inc (The Jordan II) [2004] UKHL 49, [2005] 1 Lloyd’s Rep 57. Regarding the tackle-to-tackle rule, another important qualification has recently been made in Volcafe Ltd v Compania Sud Americana de Vapores SA (trading as CSAV) [2016] EWCA Civ 1103, [2017] 1 Lloyd’s Rep 32. There, the Court of Appeal held that the loading took place for the purposes of the application of the Hague Rules, where the containers were filled in a different part of the port before actual loading. The decision was subsequently overruled by the Supreme Court but on different grounds: see [2018] UKSC 61, [2019] 1 Lloyd’s Rep 21. 5 The Conventions do not regulate contracts that, in addition to sea carriage, involve a different mode of carriage prior and/or subsequent to sea carriage. The Hamburg Rules only slightly extend the carriers’ period of responsibility: beyond tackle-to-tackle, their responsibility merely extends to the periods during which they are in charge of the goods at the ports of loading and discharge: see Art 4 of the Hamburg Rules. 6 See Great Eastern Shipping Co Ltd v Far East Chartering Ltd (The Jag Ravi) [2012] EWCA Civ 180, [2012] 1 Lloyd’s Rep 637 [43]. See also Volcafe Ltd and Others v Compania Sud Americana de Vapores SA [2018] UKSC 61, [2019] AC 358 [8]. 7 See The Jag Ravi and Volcafe (n 6). 8 For present purposes, this chapter will focus on the Hague-Visby Rules. 9 However, this may not always be the case; see, for instance, Deep Sea Maritime Ltd v Monjasa A/S (The Alhani) [2018] EWHC 1495 (Comm), [2018] 2 Lloyd’s Rep 563, where the cargo was discharged through a ship-to-ship transfer into a another vessel without production of a bill of lading. See also Cia Portorafti Commerciale SA v Ultramar Panama Inc (The Captain Gregos) [1990] 3 All ER 967, [1990] 1 Lloyd’s Rep 310. 10 See The Jag Ravi (n 6) [43]. 11 It should be noted that where a contract of carriage is contained in or evidenced by a bill of lading, the lawful holder of the bill of lading, as defined by s 5(2) of the COGSA 1992, has the right to ask the contractual carrier to deliver the goods once they arrive at the discharge port. Similarly, a consignee under a seawaybill or ship’s delivery order, who has rights of suit against the contractual carrier pursuant to COGSA 1992, also has a contractual right to delivery. The corollary of the cargo interest’s right is the carrier’s obligation to deliver the goods pursuant to the terms of the contract of carriage.

The Right to Delivery of Goods  423 The purpose of this chapter is to evaluate the limitation of freedom of contract provided by the Hague-Visby Rules, considering in particular the extent to which the carrier’s exclusion clauses apply to misdelivery claims. In this context, it will also shed light on the interaction between the contractual extension of the Hague-Visby Rules and the carrier’s exclusion clauses in the case of a misdelivery claim. For the purposes of addressing these issues, the chapter will first consider the important preliminary point of what constitutes a misdelivery, considering in particular the presentation rule and its exceptions.

I.  What Constitutes Misdelivery? Delivery of goods by the sea carrier is an indispensable part of every contract of carriage.12 Where the goods shipped aboard a vessel are covered by a negotiable bill of lading,13 the carrier is both entitled and obliged to deliver the goods only against production of the original bill of lading (this will be referred to as ‘the presentation rule’).14 As Lord Denning stated in Sze Hai Tong Bank Ltd v Rambler Cycle Company Ltd, ‘a shipowner who delivers without production of the bill of lading does so at his peril’.15 In practical terms, the presentation rule enables the carrier to ascertain who is entitled to get delivery of the goods covered by the bill of lading, which usually travels through the hands of numerous traders. The rule also protects the carrier from liability for misdelivery.16 In Glyn Mills Currie & Co v East and West India Dock Co,17 a set of three bills of lading was issued. Whilst the goods were in transit, the consignee, Cottam & Co, endorsed one part of the bills of lading to another party, Glyn Mills, as security for a loan. Although, on arrival of the goods, delivery was made against the unendorsed bill of lading, the carrier was not found liable in conversion. There was no liability mainly because (i) the contract of carriage under the bill of lading provided that ‘one accomplished the others to stand void’ and (ii) the goods were delivered without notice of a competing title.

12 Sze Hai Tong Bank Ltd v Rambler Cycle Company Ltd [1959] 2 Lloyd’s Rep 114 (PC), 121 (Lord Denning). 13 Under English law, a straight bill of lading needs to be surrendered where the bill of lading is silent on the issue of whether it has to be presented for delivery. See the obiter statements of Lord Bingham and Lord Steyn in JI MacWilliam Co Inc v Mediterranean Shipping Co SA (The Rafaela S) [2005] UKHL 11, [2005] 1 Lloyd’s Rep 347 [20] and [45], respectively. 14 Motis Export Ltd v Dampskibsselkabet AF 1912 Aktiesekkab [1999] 1 Lloyd’s Rep 837. 15 Sze Hai Tong Bank Ltd (n 12) 120. See also The Ines [1995] 2 Lloyd’s Rep 144, where Clarke J said obiter that even a delivery to the party entitled to possession would give rise to a breach of contract, if the party failed to obtain delivery by presenting the bill of lading. 16 SA Sucre Export v Northern River Shipping Ltd (The Sormovskiy 3068) [1994] 2 Lloyd’s Rep 266, 274 (Clarke J). 17 Glyn Mills Currie & Co v East and West India Dock Co (1882) 7 App Cas 591. Compare the facts in Feardon v Bowers (1882) 7 App Cas 591, where the carrier had notice of the competing titles.

424  Melis Özdel Viewed from the perspective of the lawful bill of lading holders,18 the ­presentation rule makes commercial sense, as it protects them from fraud to some extent19 while also giving some degree of control over the goods during sea transit.20 However, the presentation rule often acts as an unnecessary straitjacket in international trade, where there is usually a chain of sale contracts. The selling and reselling of goods in international trade seldom leave sufficient time for the documents to travel through the banking system and become available to the ultimate consignee before the arrival of the goods at the discharge port. In such cases, the shipowners typically deliver the goods to the ultimate consignee without production of the bill of lading once they are provided with an indemnity in respect of the consequences of such delivery.21 A carrier is in principle entitled to enforce such an indemnity once he is provided with a reasonable explanation22 for the absence of the original bills of lading.23 Where a letter of indemnity is issued in favour of a carrier for delivery of the goods,24 the undertakings in the letter will be engaged only when the goods are ‘delivered’ to the party identified in the letter.25 For delivery, the carrier need not physically hand the goods over to the receivers. However, they will not be deemed to be delivered unless they are handed over to the cargo receiver’s agent at the port of discharge and the carriers divest themselves of all power to control any physical dealing in the goods.26

A.  The Presentation Rule At first sight, the main source of the presentation rule appears to be contractual, although a closer look at the relevant cases, which are discussed in this section, 18 See the definition of lawful bill of lading holders in s 5(2) of COGSA 1992. 19 The presentation rule did not protect the cargo interest in The Future Express [1993] 2 Lloyd’s Rep 542. Following the enactment of COGSA 1992, The Future Express would have been decided differently. 20 A bill of lading can therefore be used as a document of title, with the effect that the transfer of the document transfers also the right to demand the cargo from the ship at discharge. Nonetheless, it must be emphasised that mere transfer of bills of lading is insufficient to transfer any property in or possessory rights to the goods. 21 As is clear from the decision in Kuwait Petroleum Corp v I&D Oil Carriers Ltd (The Houda) [1994] 2 Lloyd’s Rep 541, charterparty provisions on delivery of goods against production of a letter of indemnity are not usually taken to impose any express obligation on the charterers to discharge a cargo in the absence of the bill of lading. 22 Hence, the carrier has the right to refuse to deliver the goods even to the owner where the owner fails to satisfactorily explain why the bill of lading is not available for presentation: see, eg, Barclays Bank Ltd v Commissioners of Customs and Excise [1963] 1 Lloyd’s Rep 81, 88–89. 23 See The Jag Ravi (n 6) [51]. 24 Where a carrier is not a party to a contract contained in the letter of indemnity, it may still be possible for the carrier to enforce the letter against the cargo receiver pursuant to the Contracts (Rights of Third Parties) Act 1999. Hence, the shipowners in The Jag Ravi were entitled to enforce the letter of indemnity issued by the cargo receiver to the charterer where the indemnity provided in the letter was extended to the ‘servants and agents’ of the charterers. 25 Farenco Shipping Co Ltd v Daebo Shipping Co Ltd (Bremen Max) [2008] EWHC 2755 (Comm),[2009] 1 Lloyd’s Rep 81. 26 The Jag Ravi (n 6) [51].

The Right to Delivery of Goods  425 suggests that custom of the port27 can play an important role when deciding whether the presentation rule is applicable in a particular case.28 The hierarchical order between the contract terms regarding delivery and custom of merchants has not yet been subject to judicial scrutiny. Although one may be inclined to give more weight to the contractual terms, whether the custom of merchants can effectively displace an express contract term on the presentation of the bill of lading for delivery has yet to be judicially considered. In Glencore v MSC,29 the goods were misdelivered through the use of an electronic ‘early release system’ at a container terminal at Antwerp. The plaintiff cargo interests, who brought an action against the carrier for misdelivery, relied on the express term in the bill of lading, which provided that ‘one original Bill of Lading, duly endorsed must be surrendered by the Merchant to the Carrier (together with outstanding freight) in exchange for the Goods or a Delivery Order’. Before the action, the cargo interests had taken delivery of goods by using the same system for over a year. This did not, however, give rise to an estoppel, because the Court of Appeal found that there was no representation made by Glencore that delivery otherwise than against a bill of lading would in any case be acceptable.30 The Court also held that there was no custom of merchants that could displace the presentation rule expressly provided in the bill of lading, finding that the presentation of PIN codes for delivery was not a sufficiently well-established rule of conduct binding upon the parties.31 In The Sormovskiy,32 a set of three bills of lading that had been issued for the misdelivered cargo provided that any of the bills of lading being accomplished, the others were to stand void. It was implicit in this provision that one of the bills of lading could only be accomplished by being presented to the master or carrier for delivery.33 Nonetheless, the goods were delivered otherwise than against production of a bill of lading. Holding that this constituted misdelivery, Clarke J (as he then was) stated the rule that once a bill of lading is issued in order form, the presentation rule applies by implication if not expressly provided in the bill of lading. On the role of the custom of merchants, he said: [I]f there were a custom of the port … that cargo was always delivered … without the production of an original bill of lading, [such a] delivery … would probably amount to performance of the … [carrier’s] obligations under the contract of carriage.

27 On the meaning of ‘custom of the port’, see R Thomas, ‘“Custom of the Port” As a Category of Commercial Custom’ [2016] LMCLQ 436. 28 This does not mean, however, that the presentation rule can only arise exclusively either from contract or from custom of the port. In fact, the rule would appear to be derived from both of these sources. 29 Glencore International AG v MSC Mediterranean Shipping Co SA [2017] EWCA Civ 365, [2017] 2 Lloyd’s Rep 186. 30 ibid [67]–[69]. 31 ibid. 32 The Sormovskiy (n 16). 33 A similar view was also taken by Lord Denning in Sze Hai Tong Bank Ltd (n 12).

426  Melis Özdel However custom in this context means custom in its strict sense; that is it must be ­reasonable, certain, consistent with the contract, universally acquiesced in and not contrary to law.34

It seems fairly clear from this passage that Clarke J thought that the presentation rule can in certain circumstances give way to custom ‘in its strict sense’ requiring delivery without the production of the bill of lading. What is not clear is what should be understood by custom in ‘its strict sense’. As stated by Clarke J, a method of delivery that is in accordance with the practice in the port is not sufficient to override an express term in the bill of lading requiring delivery against the presentation of the bill of lading.35 Clarke J’s statements in The Sormovskiy could perhaps be taken to mean that if the custom at the port makes it impossible for the cargo to be delivered otherwise than without the production of the bill of lading, the delivery should be treated as contractual despite the presentation rule expressly stated in the bill of lading. Clarke J in The Sormovskiy also sought to shed light on the carrier’s obligation to deliver the goods where the bill of lading is lost or stolen. On the applicability of the presentation rule in such cases, he said: In trades where it is difficult or impossible for bills of lading to arrive at the discharge port in time the problem is met by including a contractual term requiring the master to deliver the cargo against a letter or indemnity or bank guarantee. That is common place and indeed there was a provision to that effect here. The simple rule [that in the absence of an express term of the contract the master must only deliver to the holder of the bill of lading] does require some exceptions because the bill of lading might have been lost or stolen. In order to cater for that problem it is no doubt necessary to imply a term that the master must deliver cargo without production of an original bill where it is proved to his reasonable satisfaction both that the person seeking delivery of the goods is entitled to possession and what has become of the bills of lading. The precise nature of the exceptions will no doubt require further consideration in the future.36

The suggested exceptions to the presentation rule met with judicial criticism. Both Rix J in Motis Export Ltd v Dampskibsselkabet AF 1912 Aktiesekkab37 and Mance LJ in East West Corporation v DKBS 191238 took the view that these ­exceptions, if recognised, would diminish the strength of the presentation rule, leaving the carriers with an obligation to deliver even where a suitable indemnity is not provided. In particular, these developments revealed that the master’s reasonable satisfaction that the person seeking delivery is entitled to possession could not be sufficient to discharge the carrier from liability for misdelivery. In particular, the decision in Motis Export Ltd v Dampskibsselkabet AF 1912 Aktiesekkab attested that, absent an express provision to the contrary, the

34 The

Sormovskiy (n 16) 275 (emphasis added).

35 ibid. 36 ibid

274. Export Ltd v Dampskibsselkabet AF 1912 Aktiesekkab (n 14). 38 East West Corporation v DKBS 1912 [2003] EWCA Civ 174, [2003] 1 Lloyd’s Rep 239. 37 Motis

The Right to Delivery of Goods  427 carrier’s reasonable belief is too low a threshold. There, the carrier was under an absolute obligation to deliver goods against the presentation of one original bill of lading covering the relevant cargo. Consequently, the carrier was held to be liable for conversion due to the delivery up of the goods against what appeared to be a genuine bill of lading, which later turned out to be a forgery. Although the misdelivery was not deliberate, the carrier was held liable mainly on the grounds that he had ‘intentionally’ parted with the goods, even though delivery was achieved by deception.

B.  Misdelivery and Block-chain-based Bills of Lading Looking forward, it is very likely that the carrier’s standard of duty to deliver will be the subject of judicial consideration once carriage of goods under blockchain-based bills of lading becomes a reality. Without getting deep into the technological structure of how block chain works, one might ask if the reasoning in Motis can equally apply in a case where a fraudster takes delivery of the goods by stealing the traders’ private keys in a block-chain bill of lading. It was striking in Motis that the Court sought to draw a line between theft and intentional delivery of goods by deception (with the use of a forged bill of lading, which was treated as a nullity). While the carrier in Motis would have been excused in the former situation under the general all-embracing exclusion clause in the contract of carriage, the carrier found himself in the latter situation and could not escape liability, since he had deliberately parted with the goods against the presentation of a forged bill of lading. Where a fraudster takes delivery of the goods by stealing the traders’ private keys in a block-chain bill of lading, will this be treated as theft or as intentional delivery of goods by deception? The difference between the two can appear illusory in such cases. The solution would appear to boil down to the question of whether the block-chain-based bill of lading could be treated as forged once the private keys are stolen. If so, the line of reasoning adopted in Motis could equally apply to the carrier in such a case and the carrier may not be excused for the misdelivery, particularly when the contract contains a wide but general exclusion clause. As is clear from the decision in Motis, the essential element of misdelivery is consensual passing of possession in goods without production of the appropriate documents (whether paper or electronic). The extent to which a carrier’s exclusion clause can provide sufficient protection to the carrier is discussed in the following section.

II.  The Limits to the Cargo Interest’s Rights to Delivery In a misdelivery action, one of the main hurdles faced by the cargo interest is the limitation of liability and exemption clauses. For proper analysis of the effect of such clauses, it is essential to analyse individually those cases where the

428  Melis Özdel misdelivery takes place when the Hague-Visby Rules are in force either mandatorily or by contract, and those cases where the delivery takes place outside the scope of application of these Rules. Before discussing the legal positions in these two separate cases, it is important to raise a preliminary question: can the Hague-Visby Rules, or at least some of the provisions therein, be applicable to a misdelivery that has taken place outside the sphere of application of the Rules? On the face of it, the answer appears to be a simple ‘Yes’, particularly where the parties contractually extend the operation of the Rules. However, where there is no such contractual extension, could the wording of the Rules be stretched to cover the periods prior to loading and after the goods have been discharged? This question is important, particularly in the case of a dispute arising from misdelivery. In such cargo disputes, there are two key articles that the carrier would be keen to rely on. One is the Article III rule 6 time-bar, which provides that the carrier shall be discharged from ‘all liability whatsoever in respect of the goods’, unless a cargo claim is brought within one year after delivery of the goods or the date when the goods should have been delivered.39 The other is Article IV rule 5 on package limitation, which provides that: Unless the nature and value of such goods have been declared by the shipper before shipment and inserted in the bill of lading, neither the carrier nor the ship shall in any event40 be or become liable for any loss or damage to or in connection with the goods in an amount exceeding the equivalent of 666.67 units of account per package or unit or units of account per kilo of gross weight of the goods lost or damaged, whichever is the higher.

The wording of these provisions would appear to be wide enough to cover post-discharge events, such as misdelivery, despite the fact that the Rules are in principle applicable only during the tackle-to-tackle period. It has been suggested that the Rules, and Article III rule 6 in particular, apply to such post-discharge events. Considering the main purpose of the Hague-Visby Rules, which is to reach international uniformity of approach on the carrier’s liability, there seems to be no reason to distinguish cases where delivery has taken place after the discharge of cargo from those where the delivery has taken place on discharge of the cargo. Viewed from this perspective, it makes sense to treat both cases as the same and apply at least the Article III rule 6 time-bar provision to a carrier

39 Emphasis added. In order to prevent cargo interests from circumventing the time bar by initiating an action against carriers in tort, the Hague-Visby Rules also introduce a special provision, Art IVbis 1, which provides that ‘the defences and limits of liability provided for in these rules shall apply in any action against the carrier in respect of loss or damage to goods covered by a contract of carriage whether the action be founded in contract or tort’. 40 Emphasis added. Note, however, the important exception to the carrier’s right to rely on limitation provided in Article IV rule 5(e), which reads: ‘Neither the carrier nor the ship shall be entitled to the benefit of the limitation of liability provided for in this paragraph if it is proved that the damage resulted from an act or omission of the carrier done with intent to cause damage, or recklessly and with knowledge that damage would probably result.’

The Right to Delivery of Goods  429 who has misdelivered the cargo after discharge. It can be argued that the Rules apply as an implied term of the bailment to cover the post-discharge period.41 Nonetheless, as will be seen from the cases discussed in sections II.A and B, the courts have shown a preference for not stretching the sphere of application of the Hague-Visby Rules by way of implying a term into the contract of carriage to that effect.

A.  Where a Misdelivery Takes Place within the Sphere of Application of the Hague-Visby Rules As will be recalled, misdelivery usually occurs after the goods are discharged, when the Hague-Visby Rules no longer apply mandatorily. Nonetheless, in some cases, misdelivery can occur before or simultaneously with discharge. The latter event was seen in The Alhani,42 where the owner of the vessel discharged a cargo of bunker fuel through a ship-to-ship transfer into another vessel, without production of the bill of lading covering the cargo. Similarly, in The Captain Gregos,43 the goods were misappropriated by the carriers within the tackle-to-tackle period, where they deliberately omitted to discharge the full cargo at the discharge port. Where a misdelivery takes place within the sphere of application of the Hague-Visby Rules, will the carrier’s breach be subject to the Hague-Visby scheme? In other words, does the carrier’s delivery obligation stand outside the Hague-Visby Rules entirely, regardless of whether the misdelivery took place within the sphere of application of the Rules? A closer look at Article II suggests that the answer is not as clear as it might at first seem. Article II of the Hague-Visby Rules provides that under every contract of carriage of goods by sea the carrier, in relation to the loading, handling, stowage, carriage, custody, care and discharge of such goods, shall be subject to the responsibilities and liabilities and entitled to the rights and immunities hereinafter set forth.

The provision can be understood to mean that description of the carrier’s obligation under Article II is limited to the stages of carriage only, starting with loading until the discharge of goods. Consequently, one might argue that the carrier’s obligation to deliver should in any case be outside the Hague-Visby scheme. However, where delivery takes place within the sphere of application of the Hague-Visby Rules, there seems to be no reason for giving such a limited effect to Article II, particularly since one of the main objects of every contract of carriage is usually considered to be delivery of the goods.44



41 See

R Aikens and M Bools, Bills of Lading, 2nd edn (Informa 2016) para 10.94. Alhani (n 9). 43 The Captain Gregos (n 9). 44 Motis Export Ltd v Dampskibsselkabet AF 1912 Aktiesekkab (n 14) 216–17. 42 The

430  Melis Özdel In The MSC Amsterdam,45 the misdelivery of a cargo of copper cathodes occurred after the discharge. The Hague-Visby Rules would have applied after the discharge if there had been a contract provision in the carriage contract to that effect. However, the application of the Rules was not extended by contract, and the carrier’s delivery obligation was therefore not subject to the Rules. At first instance, the judge’s obiter view was that had the application of the Rules been extended by contract, the carrier’s obligation would have been subject to the Hague-Visby Rules, with the effect that the carrier would have been able to rely on the limitations and exceptions under the Rules. In the Court of Appeal, Longmore LJ preferred not to express any views on this matter.46 In The Captain Gregos,47 the cargo interest framed the claim in tort and not for breach of the Rules. Nonetheless, Bingham LJ took the view that the conversion resulting in the loss of part of the cargo could amount to a breach of Article III rule 2 of the Hague-Visby Rules.48 This appears to be in line with the purpose behind Article IV rule 1bis, which provides: The defences and limits of liability provided for in these Rules shall apply in any action against the carrier in respect of loss or damage to goods covered by a contract of carriage whether the action be founded in contract or in tort.

The objective here appears to ensure that a cargo owner who can sue the carrier in contract49 is not better off suing in tort. Against this background, the carrier in The Captain Gregos, as bailee of the goods, was held to have failed to keep and care for the goods in breach of Article III rule 2. The important point was that the delivery obligation was held to be subject to the Hague-Visby Rules, as a result of which the carrier was able to rely on Article III rule 6. Article III provides that the carrier is to be discharged from ‘all liability whatsoever in respect of the goods’. On the meaning of the quoted phrase, Bingham LJ said that ‘all liability whatsoever in respect of the goods’ meant exactly what it said; the inference that the one year time bar was intended to apply to all claims arising out of the carriage of goods was strengthened by the consideration that art III, r 6 was intended to achieve finality.50

45 Trafigura Beheer BV v Mediterranean Shipping Co SA (The MSC Amsterdam) [2007] EWCA Civ 794, [2007] 2 Lloyd’s Rep 622. 46 ibid [27]. 47 The Captain Gregos (n 9). 48 This also appears to be in line with the view expressed by Devlin J in Adamastos Shipping Co Ltd v Anglo-Saxon Petroleum Ltd [1957] 1 Lloyd’s Rep 79, 87, where he said (i) the reference to ‘loss or damage’ in the Hague Rules should not relate only to physical loss of or damage to goods, and (ii) it should also cover the cases of misdelivery. 49 Hence, the Rules are intended to regulate the rights and duties of the parties to the carriage contract contained in or evidenced by bills of lading and similar document of title: see Art I(b) of the Rules. Bingham LJ stressed that the Rules can have no application in cases where there is no such contract between the parties. 50 The Captain Gregos (n 9) 315–16, 319.

The Right to Delivery of Goods  431 At this juncture, the crucial point is whether full effect could still be given to this phrase even in the case of an intentional misdelivery by the carrier. Regardless of the all-embracing phrase ‘all liability whatsoever’ in Article III rule 6, one can argue that to apply a time bar in a case where there is intentional misdelivery would defeat the main object of the carriage contract, proper delivery of the goods by the carrier. There may be an obvious attraction in this argument, particularly since a similar line of reasoning is usually followed51 when interpreting a carrier’s exclusion clause in the case of a misdelivery that has taken place outside the sphere of application of the Hague-Visby Rules. Nonetheless, acknowledging that the Rules, including the time-bar provision therein, have an international currency, Bingham LJ refused to limit the scope of application of the time bar with reference to the canons of construction at English common law. With a view to achieving the correct construction of this international convention, he relied on the legislative intention, as reflected in the travaux préparatoires, as well as in the Rules themselves,52 to apply the time bar to cases of wrong delivery. In so doing, he held that the carrier was entitled to rely on the time bar against the cargo interest, who had failed to bring a cargo claim during the one-year time period. On the whole, where the Rules apply to a sea carriage mandatorily, a carrier who has converted a cargo on board, whether negligently or intentionally, is able to avail himself of the time-bar provision, although the carrier will lose the benefit of the financial limitation, that is the package/unit limitation under Article IV rule 5(3) of the Rules.

i.  The Role of Article III rule 8 The faith of the carrier’s exclusion clause in a carriage contract mandatorily governed by the Hague-Visby Rules is a separate matter, which is largely concerned with the question of whether such a clause offends Article III rule 8 of the Rules, which provides that: Any clause, covenant, or agreement in a contract of carriage relieving the carrier or the ship from liability for loss or damage to, or in connection with, goods arising from negligence, fault, or failure in the duties and obligations provided in this article or lessening such liability otherwise than as provided in these Rules, shall be null and void and of no effect.

A brief reading of the provision clearly suggests that an exclusion clause seeking to relieve the carrier of liability for misdelivery is rendered null and void and of no effect. However, difficulties arise when considering the effect of a clause 51 See section II.B. 52 See also an interesting recent decision in Sea Tank Shipping AS v Vinnlustodin HF (The Aqasia) [2018] EWCA Civ 276, [2018] 1 Lloyd’s Rep 530, where the Court of Appeal relied on the legislative intention.

432  Melis Özdel seeking to provide substituted ways of delivery. It is clear that a clause stating that the carrier shall not be liable for misdelivery will be struck down by Article III rule 8, since any clause purporting to relieve the carrier of his obligation under Article III rule 2 is rendered null and void by Article III rule 8. What is not so clear is whether the same result would follow in the case of a clause defining what should be understood by proper delivery under the contract. Can such a clause be equally struck down, or should it be given effect? One can argue that where the application of a clause has the effect (whether direct or indirect) of removing or lessening the carrier’s liability, it should be struck down by Article III rule 8. Further support can be drawn from the fact that, even after the decision in Photo Production v Securicor,53 misdelivery continues to be judicially considered as a breach of fundamental character:54 this has been the main pillar of the court’s preference for giving limited effect to the carrier’s exemption clauses purporting to embrace also the cases of misdelivery.55 Nonetheless, these points would appear to be outweighed by one important consideration. On the nature of the carrier’s obligation to take care of the cargo under Article III rule 2, English courts have given effect to provisions defining the terms on which the carriage service is to be performed. In Pyrene v Scindia,56 Devlin J took the view that the extent to which loading and discharging could be brought within the carrier’s obligations had to be left to the parties themselves to decide. Following this decision, in Renton v Palmyra,57 the House of Lords saw no difficulty in giving effect to a strike clause allowing the carrier to discharge the cargo at a different safe port, emphasising that the Rules allowed the parties to determine the parts that each party had to play. More recently, the House of Lords in The Jordan II58 also followed this line of reasoning. Although these three cases are not directly related to the carrier’s obligation to deliver goods, they establish the main principle that Article III rule 8 only prohibits the carrier from contracting out of liability for doing what he has contractually undertaken. In this context, Article III rule 2 only operates for the purposes of setting out the manner in which the carrier must perform those duties that he has undertaken to perform. Consequently, a carriage term defining delivery may not necessarily be struck down by Article III rule 8, although its application may leave the cargo interest without any remedy when the cargo has been delivered to someone else pursuant to a delivery clause. The problem discussed is particularly relevant in cases where a charterparty provision giving the shipowner the right to deliver the goods against a letter of



53 Photo

Production Ltd v Securicor Transport Ltd [1980] AC 827 (HL). Export Ltd v Dampskibsselkabet AF 1912 Aktiesekkab (n 14) 216–17. 55 See also section II.B. 56 Pyrene v Scindia [1954] 1 Lloyd’s Rep 321. 57 Renton v Palmyra [1956] 2 Lloyd’s 379 (HL). 58 The Jordan II (n 4). 54 Motis

The Right to Delivery of Goods  433 indemnity is incorporated into the relevant bill of lading. For the reasons previously explained, such a clause would not necessarily be rendered void by Article III rule 8. However, unless the provision expressly and clearly provides to the contrary, the English court will not interpret such a provision to mean that the owner will not be liable vis-à-vis the lawful bill of lading holder for misdelivery when he has delivered the cargo to someone else against a letter of indemnity.59

ii.  Where the Rules have Contractual Force Before moving on to section II.B, it is worth discussing the cases where the Hague-Visby Rules have only contractual force. Such circumstances arise, not infrequently, on two main occasions: (i) where the transport document covering the cargo is a ship’s delivery order or a seawaybill; and (ii) where the Rules are incorporated to cover the post-discharge and/or pre-loading stages. It is common ground that where the Rules do not have a mandatory effect,60 Article III rule 8 will not by force of law strike out any offending term in the carriage contract. All contract provisions and the Hague-Visby Rules will stand on an equal footing, with the effect that any issue of inconsistency will have to be resolved pursuant to the general aids of contract construction at English common law. Against this background, there is an important practical issue that needs to be examined for present purposes: should the provisions in the Hague-Visby Rules still be interpreted as an international convention, even though they have only contractual force in the circumstances described above? As is clear from a long line of English cases, the Rules are not interpreted on the basis of any technical rules of English law.61 Instead, the courts are in principle guided by the following: The duty of a court is to ascertain the ordinary meaning of the words used, not just in their context but also in the light of the evident object and purpose of the Convention. The court may then, in order to confirm that ordinary meaning, have recourse to the travaux préparatoires and the circumstances of the conclusion of the Convention.62

Furthermore, on the use of the travaux préparatoires, the courts allow these to be determinative of the question of construction only ‘in an appropriate case involving truly feasible alternative interpretations of a convention’, provided that the

59 See n 19. 60 The Rules mandatorily apply to ‘a bill of lading or any similar document of title’ (see Art I(b) of the Rules), provided the conditions under Art X of the Rules are satisfied. Alternatively, the Rules apply by force of law to any non-negotiable receipt marked as such if it expressly provides that the Rules are to govern as if it were a bill of lading: see s 1(6)(b) of the Carriage of Goods by Sea Act 1971. See also the decision in Browner International Ltd v Monarch Shipping Co Ltd (The European Enterprise) [1989] 2 Lloyd’s Rep 185; and Özdel, ‘The EU and the Carriage of Goods by Sea’(n 2) [60]–[78]. 61 See, for instance, the reasoning in the House of Lords decision in The Jordan II (n 4). 62 CMA CGM SA v Classica Shipping Co Ltd (The CMA Djakarta) [2004] EWCA Civ 114, [2004] 1 Lloyd’s Rep 460 [75].

434  Melis Özdel court is satisfied that the travaux préparatoires clearly and indisputably point to a definite legal intention.63 On the whole, the Rules are construed on broad principles of general ­acceptation,64 although it may be difficult in most cases to know what broad principles of general acceptation should prevail over domestic principles. Nonetheless, the difference between these two approaches can be seen in this present context: in The Captain Gregos, the court treated the all-embracing but unqualified time-bar clause, Article III rule 6, as sufficient to cover the misappropriation by the carrier. In so doing, the court mainly relied on the guidance provided above as opposed to the aids to construction at common law, which are discussed in section II.B.

B.  Where a Misdelivery Takes Place Outside the Sphere of Application of the Hague-Visby Rules Following a wave of landmark cases on contract interpretation, a vast amount of academic scholarship has been produced in recent years on the topic of interpretation of commercial contracts. For present purposes, suffice to say that there are five general principles of construction, as summarised by Lord Hoffman’s judgment in ICS: (1) Interpretation is the ascertainment of the meaning which the document would convey to a reasonable person having all the background knowledge which would reasonably have been available to the parties … at the time of the contract. (2) The background … includes absolutely anything which would have affected the way in which the language of the document would have been understood by a reasonable man. (3) The law excludes from the admissible background the previous negotiations of the parties and their declarations of subjective intent … (4) … The background may not merely enable the reasonable man to choose between the possible meanings of words which are ambiguous but even (as occasionally happens in ordinary life) to conclude that the parties must, for whatever reason, have used the wrong words or syntax … (5) … [W]e do not easily accept that people have made linguistic mistakes, particularly in formal documents. On the other hand, … the law does not require judges to attribute to the parties an intention [that] they plainly could not have had …65

Following this, a first question that comes to mind is whether, and if so to what extent, these principles can apply in the context of maritime law. Leaving aside the cases where the question of construction is related to an international maritime 63 Fothergill v Monarch Airlines Ltd [1980] 2 Lloyd’s Rep 295. 64 Stag Line Ltd v Foscolo, Mango & Co Ltd (The Ixia) [1932] AC 328 (HL), 350 (Lord Macmillan). 65 Investors Compensation Scheme Ltd v West Bromwich Building Society (No 1) [1998] 1 WLR 896 (HL), 913. For discussion of the extent to which parties can contractually control these principles of contractual interpretation, see R Calnan, ‘Controlling Contractual Interpretation’, ch 4 in this volume.

The Right to Delivery of Goods  435 convention, such as the Hague-Visby Rules, these principles are frequently resorted to by the courts when interpreting a maritime contract.66 At first sight, there seems to be no reason for treating a maritime contract differently from a general commercial contract. On a closer look at the relevant case law, however, it will be seen that there have been some judicial attempts to treat maritime contracts sui generis, in some respects.67 In Tate & Lyle Ltd v Hain Steamship Co Ltd,68 the House of Lords applied the now discredited doctrine of ‘fundamental breach’ against a carrier that was liable for geographical deviation. Considering the seriousness of the breach, the carrier was stripped of the contractual limitations and exemptions in the contract of carriage. This issue has only recently been revisited by Carr J in The Sur,69 where the judge took the view that, following the House of Lords decision in Photo Production v Securicor, the seriousness of the breach is no longer sufficient for determining whether contractual limitations or exemptions can apply to particular breaches. Even after the decision in Photo Production v Securicor, misdelivery is considered to be a breach of fundamental character. This does not mean, however, that misdelivery cases are not assimilated into the ordinary law of contract. Whether and to what extent a carrier’s exclusion clause can apply to misdelivery is now a matter of contract construction. Because of the fundamental character of this breach, English courts have not allowed a carrier to escape liability for misdelivery by relying on a general exclusion clause. However, this does not mean that the carrier can never escape liability for such breaches: with an express wording in the exclusion clause, the carrier can indeed avoid liability. On the question of whether a general exclusion clause can protect the carrier in the case of theft, Clarke J’s judgment in The Ines70 provides guidance.71 There, the cargo was misdelivered without any bill of lading presented. The contract of carriage contained an exemption clause, which provided that ‘Goods in the custody of the carrier or his agent … after discharge … are in such custody at the sole risk of the owner of the goods and thus the carrier has no responsibility whatsoever for goods … subsequent to discharge from the ocean vessel.’ Clarke J said: In the context in which the word ‘whatsoever’ is used in cl 3 it does not seem to me to be apt to cover misdelivery. As I have already said, the clause seems to me to be concerned with the case where the goods are lost or damaged, and may include the case where they are stolen, but does not include delivery without production of an original bill of lading.72

66 See, for instance, the recent Court of Appeal decision in Caresse Navigation Ltd v Zurich Assurances Maroc (The Channel Ranger) [2014] EWCA Civ 1366, [2015] 1 Lloyd’s Rep 256. 67 Photo Production v Securicor (n 53) 845 (Lord Wilberforce). 68 Tate & Lyle Ltd v Hain Steamship Co Ltd (1936) 55 Ll L Rep 159. 69 Dera Commercial Estate v Derya Inc (The Sur) [2018] EWHC 1673 (Comm), [2019] 1 Lloyd’s Rep 57. 70 MB Pyramid Sound NV v Briese Schiffahrts GmbH & Co KG MS Sina (The Ines) (No 2) [1995] 2 Lloyd’s Rep 144. 71 See also Motis Export Ltd v Dampskibsselkabet AF 1912 Aktiesekkab (n 14). 72 The Ines (n 70) 152.

436  Melis Özdel The explanations above reveal that there is some level of certainty when it comes to the interpretation of the carrier’s exclusion clauses in the case of misdelivery. Nonetheless, difficulties arise on the issue of whether these exclusion clauses will be available to a carrier where there is no contract of carriage between the cargo interest and the carrier. This will be discussed in section III.

III.  Possible Avenues to Bring a Misdelivery Action It is important to start with the premise that there are two main grounds on which misdelivery claims can be based: contract and conversion. As is clear from the decision in Sze Hai Tong Bank Ltd v Rambler Cycle Co Ltd,73 a cargo interest can bring parallel claims, one in contract and the other in the tort of conversion. The carrier will obviously be in a position to rely on an exclusion clause against the cargo interest to escape liability, provided such a clause is suitably drafted to cover the cases of misdelivery. As long as there is a contract between the parties, the cargo interest will not be able to avoid the application of the exclusion clause by bringing an action exclusively in conversion.74 Difficulties arise where there is no contractual nexus between the carrier and the cargo interest. This was the case in East West Corporation v DKBS 1912,75 where the cargo claimant, the seller of the goods, had no title to sue in contract as it did not neatly fall within the definition of a lawful bill of lading holder under section 5(2)(a) of COGSA 1992. In East West, goods were shipped by the sellers under bills of lading naming certain Chilean banks as consignees. Upon discharge, the goods were held under the custody of the customs authorities in Chile pending payment of import duty. On payment of the customs duty by the buyers, the goods were delivered to the buyers without presentation of the original bills of lading. No payment was in fact made by the buyers to the sellers for the goods. On request, the Chilean banks delivered the bills of lading back to the sellers but did not endorse the bills of lading to them. Another interesting case where no contractual nexus was found between the cargo interest and the carrier is Elder Dempster & Co Ltd v Paterson Zochonis & Co Ltd.76 There, shippers of cargo on a time-chartered ship brought a cargo claim not against the time charterer, who was also the contractual carrier, but against the shipowners as actual carriers. Although the shipowners were not a party to the contract of carriage contained in or evidenced by the bills of lading, they sought to rely on the exception in the bill of lading. The decision of the House of Lords was that the shipowners were entitled to rely on the exclusion, despite the absence of a



73 Sze

Hai Tong Bank Ltd (n 12). v Merrett Syndicates Ltd (No 1) [1995] 2 AC 145. 75 East West (n 38). 76 Elder Dempster & Co Ltd v Paterson Zochonis & Co Ltd [1924] AC 522. 74 Henderson

The Right to Delivery of Goods  437 contractual relationship. The preferred reasoning is found in the judgment of Lord Summer, where his Lordship said: [I]n the circumstances of this case the obligations to be inferred from the reception of the cargo for carriage to the United Kingdom amount to a bailment upon terms, which include the exceptions and limitations of liability stipulated in the known and contemplated form of bill of lading.77

In Elder Dempster, the goods were shipped directly onto the vessel by the shipper. This created a direct bailment relationship between the shipper and the shipowner, thereby allowing the carrier to rely on the exclusion clause in the bill of lading.78 As has been illustrated in the House of Lords’ decision in The Starsin,79 the shipowner could have held possession of the goods as a sub-bailee had there been an initial bailment by the shipper to the time charterer (the contractual carrier under the bill of lading), followed by a sub-bailment by the time charterer to the shipowner.80 The terms of the sub-bailment would have been those contained in the time charterer’s bill of lading.81 Hence, any exclusion clause in the bill of lading would have been available to the shipowner.82 Bailment reasoning was also used in East West, where Mance LJ (now Lord Mance JSC) recognised bailment as a cause of action distinct from those in tort or in contract. Consequently, the doctrine of bailment on terms was applied in this case, giving the seller, as shipper and owner of the goods, the right to sue the carrier in bailment, while affording the carrier with the benefit of any relevant exemption clause contained in the bill of lading. In this case, there had been a contract of carriage between the shipper and the carrier, but the shipper lost its right of suit once the bill of lading was transferred to the Chilean banks, which were acting as its agents.83 The bailment reasoning gives the benefit of the contractual protections contained in the bill of lading to the carrier. This is particularly relevant for present purposes, where a carrier is faced with a cargo claim for misdelivery. Nonetheless, we should not lose sight of the fact that the application of the bailment reasoning has certain important limitations. The main limitation comes with the treatment

77 ibid 564. 78 Had the goods initially come into the charterer’s possession before shipment, the shipowner would have been the sub-bailee. 79 Homburg Houtimport BV v Agrosin Private Ltd (The Starsin) [2003] UKHL 12, [2003] 1 Lloyd’s Rep 571. 80 ibid 598. 81 ibid. Hence, when ascertaining on which terms the bailment relationship has been created, the courts ask on which terms the shipowner (as bailee or sub-bailee) took the goods into its possession. 82 The sub-bailment reasoning can also find room for application where there is transhipment and where the shipper sues the non-contractual carrier (ie the sub-bailee): see KH Enterprise v Pioneer Container (The Pioneer Container) [1994] 2 AC 324, [1994]1 Lloyd’s Rep 593 (PC). 83 On this matter Mance LJ in East West (n 38) said (at 247), ‘there was nothing in the statutory scheme of the 1992 Act to lend any support to the idea that after a statutory transfer of contractual rights by a principal to its agent, the principal could still sue in contract in its own name’.

438  Melis Özdel of bills of lading: although usually referred to as a symbol of the goods covered, the mere transfer of bills of lading is insufficient to transfer any property in or possessory rights to the goods.84 The law does not provide a remedy in tort or in bailment to a party who can establish no more than that he or she is a lawful holder of a bill of lading. The holder can have a contractual claim against the carrier under COGSA 1992. As has been stated by Mance LJ in East West, the transfer of contractual rights pursuant to COGSA 1992 cannot be treated as working an automatic transfer of any rights in bailment.85 Another difficulty with the bailment reasoning is that only a limited group of people can be considered as a bailee. While a shipowner can readily be treated as a bailee or sub-bailee, the same cannot be said of a time charterer of a vessel. Where the misdelivery claim is purely based in conversion, can the carriers avail themselves of the terms in the relevant bill of lading, including the exclusion clause contained therein? As is clear from the decision in The Captain Gregos (No 2),86 a claimant that has title to sue in conversion can effectively avoid the application of the terms of the bill of lading, including the carrier’s exclusion clause therein, to the misdelivery claim. In this case, the carrier’s argument was that it was entitled to rely on the time bar (under Article III rule 6 of the Hague-Visby Rules) on the bailment under the bill of lading. The Court of Appeal held that (i) the intermediate buyer, PEAG, was not the original bailor of the goods, despite the fact that it later obtained ownership once the goods were loaded onto the vessel; and (ii) the shipowners never attorned to PEAG as bailor.

IV. Conclusion Having reviewed the possible causes of action for misdelivery, one can argue that the recognition of various different legal bases for title to sue is welcome. Despite the comprehensive contractual regime recognised under COGSA 1992, a cargo interest may need to base its claim for misdelivery on a breach of bailment or conversion. However, the applicability of an exclusion clause can quite drastically change the outcome of an action for misdelivery. This gives rise to the question of why a cargo interest that can base its claim purely in conversion should be better off than those cargo interests that have a contractual and/or bailment relationship with the carrier. Apart from the legal technicalities, there seems to be no justification to treat cargo interests differently as long as they have title to sue, whether in contract, tort or bailment. As has further been demonstrated

84 See Leigh & Sillivan Ltd v Aliakmon Shipping Co Ltd (The Aliakmon) [1986] 1 AC 785, [1986] 2 Lloyd’s Rep 1 (HL). 85 East West (n 38) 254. 86 Cia Portorafti Commerciale SA v Ultramar Panama Inc (The Captain Gregos) (No 2) [1990] 2 Lloyd’s Rep 395 (CA).

The Right to Delivery of Goods  439 in this chapter, a carrier’s liability for misdelivery is also fragmented even in the context of  ­contractual claims for misdelivery. This is mainly because of the limited ­application of the Hague-Visby Rules to the tackle-to-tackle period. Despite the fact that all the different tools discussed in this chapter help the courts to respond to the ad hoc situations before them, these fragmented solutions are perhaps not commercially desirable, particularly in the interests of unification and standardisation of the laws governing the international carriage of goods by sea.

440

21 The Contents of Commercial Contracts: Terms Affecting Freedoms – A Response JACQUELINE COOK

I.  Themes Old and New in Commercial Contracts Listening to an expert group of judges, academics and well-known lawyers discussing contract law from a variety of thought-provoking angles at the conference that led to this book, is quite humbling for someone like me. I found myself trying to marry what I heard about terms affecting freedoms in commercial contracts with what I see, and have seen, in practice over many years.1 When I realised that lawyers in practice, who are dealing with both simple and complex issues and drafting every day, are at the cutting-edge of commercial contracts, it prompted me to write this chapter. Some of the themes identified here, for example good faith and interpretation, are not new, yet they are developing through both case law and more general use. More recent concepts, like relational contracts and behaviours and dealing with uncertainty, are finding their way into contracts because of the need for parties to identify and, where possible, quantify and allocate risk arising from the political environment and increased regulation. So, lawyers at all levels dealing with commercial and finance contracts have something valuable to contribute to the debate shaping English law. In fact, they are already shaping the drafting. Here are some key themes from the chapters in this volume from my perspective.

II.  Contractual Relationship and Behaviours How the courts interpret aspects of contracts in dispute forms the basis of case law  affecting contract, but those contracts that are not in dispute (or at least contracts that do not need or reach judicial or arbitral proceedings) affect how 1 Examples given are drawn both from the event and from the writer’s experience in finance transactions and commercial contracts for large assets, in a variety of practice areas. All references and citations are taken from the papers collected in this volume.

442  Jacqueline Cook parties behave and their actions during the life of a contract, even though these are often overlooked in the study and practice of contract law. It is refreshing to think of a contract in terms of a living state of affairs, rather than in a dispute situation.2 Many transactions set up a relationship between two or more parties for projects with milestones, targets and behaviours the parties have to meet during the life of the agreement. As well as the main thrust of an agreement, for example, in a facility agreement, borrowing money in return for paying interest and repaying principal, parties give ‘undertakings’ on what they will do and refrain from doing during the life of the transaction. While the parties do have freedom to contract at the outset, once the relationship is established (say as lender– borrower, supplier– purchaser where there is an ongoing supply) restrictions will inevitably feed into the contractual terms. In the contract, the parties will consensually restrict their own future actions and behaviours, moving from self-interest to the interest of the common goal of the contract, with different outcomes for each party. Here are the type of issues commercial contracts can build into the terms that affect behaviours: • provision of information, reports, financial information and accounts, information on assets; • requirements of confidentiality on all things or particular things: nondisclosure provisions in agreements, loan and facility agreements, employment contracts; • negative undertakings, that is an obligation not to do certain things: borrow, lend, grant security, dispose of any assets, act in competition, merge, work for another party, release trade secrets, release trade information, grant dividends and distributions, trade with others either not approved or within a certain criterion; • protection of assets or staff or persons: maintenance and repair; insurance; non-disposal of assets; keyman or key person insurance; health; environmental; modern slavery; standards in employment; Transfer of Undertakings (Protection of Employment) Regulations 2006 (SI 2006/246) (TUPE Regulations); employee health and safety; product liability and safety; fitness for purpose; trading standards and disclosure requirements. Loss of reputation as a threat, and almost as a remedy, in everyday contracts is a valid consideration in how and whether to comply with a contract, when set in the 2 The following chapters in this volume have examined these concepts: R Stevens, ‘Binding Our Future Selves’, ch 2; L Gullifer and G Penn, ‘The Boundaries of a Borrower’s Freedom to Act: Negative Covenants in Loan Agreements’, ch 8; L Miller, ‘“Ethical Clauses” in Global Value Chain Contracts: Exploring the Limits of Freedom of Contract’, ch 9; C MacMillan, ‘Private Law and Public Concerns: Non-Disclosure Agreements in English Contract Law’, ch 15; R Moorhead, ‘Professional Ethics and NDAs: Contracts as Lies and Abuse?, ch 16. For more on dispute resolution, refer to A Mills, ‘Choice of Court and Choice of Law Agreements: Freedom of Freedom of Contract’, ch 17 in this volume.

Terms Affecting Freedoms – A Response  443 real world of instant news and social media today. Reputation for a small business can be everything, so reputational damage, social media publicity – good or bad – and how the parties act to make up for a mistake or breach could enhance or bring down a business. Reputation is, of course, important for very large companies with a high profile too, such as airlines, food providers, and supermarkets. The global, social and environmental swell of opinion to put duties – whether moral, regulatory or legal – on to businesses, is affecting how businesses behave and is finding its way into contracts. This concept of ‘being a good business’ seemed to weigh more heavily after the financial crisis of 2008. Initially, this focused on transparency and money laundering in financial transactions. It is now even wider, encompassing (i) the minimum wage, (ii) disclosure of ‘people of significant control’ and the ‘ultimate beneficial owner’3 of a business with international reach, (iii) sanctions requirements for international ­transactions, (iv) green requirements, (v) compliance with modern slavery disclosures, (vi) health and safety requirements, and in some cases (vii) green and sustainability requirements. The rise of ‘ethical clauses’, clauses with a public interest aspect to them, is having a trickle effect from the regulatory or the moral code into contractual provisions. One or other party may feel the need to require certain behaviours of another contracting party, or may feel the pressure to demonstrate to the outside world and to its employees, and to the regulator, if it has one, that it is contracting on this basis. In some ways, this is the public face of the contract. There are different factors at play here: regulators’ requirements; reputation; using an ethical stance as a marketing tool to keep the market and contracting parties onside; a legal requirement to include the clause; a legal requirement to make a return or a disclosure; or an attempt to make sure the contracting party is morally and ethically on the same playing field.

III.  Breach and Termination – Inside or Outside the Contract Where something goes wrong between parties to a contract, resulting in a breach, no matter how large or small, the parties could look at remedies available to them in the agreement using contract law, and, in some cases, may also look at remedies available in tort. Breach of contract, frustration, estoppel and restitution may be possible options.4 By identifying risks due to uncertainty, quantifying

3 These concepts and disclosure requirements were introduced into English law on 6 April 2016 under pt 21A of and sch 1A to the Companies Act 2006. 4 Chapters in this volume that have covered aspects of this topic are: A Burrows, ‘Anti-Oral Variation Clauses: Rock-Solid or Rocky?, ch 3; J Morgan, ‘Opting for “Documentary Fundamentalism”: Respecting Party Choice for Entire Agreement and Non-Reliance Clauses’, ch 12; W Day,

444  Jacqueline Cook and allocating financial or practical responsibility for such risks means that within some commercial agreements, parties are trying to cover situations that could arguably be classed as frustration, or which may give grounds for an argument for restitution. Contracting parties could instead maintain their contractual relationship and keep the contract alive. In a way, the parties are legislating by themselves for the likes of unjust enrichment and restitution. The contracting parties, by setting out contractual terms dealing with the unexpected, could guide the courts as to how they would see their remedies applying to their relationship. In a facility agreement, for example, the lender will list events it considers to be ‘events of default’. The borrower may negotiate these with the lender before the agreement is executed. However, as many of these are also standard for the market they are in, it is more usual for the borrower to negotiate the commercial undertakings and representations to meet the practicalities of their particular business. Breach of an undertaking or making a misrepresentation could trigger an event of default. Rather than running straight to the courts, under the acceleration provisions the parties set out various options for what they can do, usually at the request of the security agent acting on the basis of the consent of the majority of or all lenders. So ‘inside’ the contract, the parties have set out whether: to put the loan on demand; to demand repayment of principal, interest and costs; to make a demand under a guarantee; to apply default interest; to replace a defaulting lender; to replace an impaired agent; or to instruct the security agent to enforce the security. ‘Outside’ the contract, it is still open to the parties to acquiesce and not take action on the basis of the breach. For example, they could negotiate an amendment to the financial covenants, waive a breach, or give more time to the borrower to provide missing information or to discuss the position. The parties could go to court to claim damages for breach of contract or seek specific performance, but in some ways these outcomes are not really the practical remedy either party is seeking in a financing transaction. That is why the market clauses on ‘Event of Default’ and ‘Acceleration’ have been developed and adopted.5 One interesting example is the contractual event of default that can arise in the event that a misrepresentation is made. This allows for a remedy not just in tort for misrepresentation, but also as a breach of contract, in that the misrepresentation triggers an event of default. This in turn triggers the remedies within the agreement itself, with optional routes the lender or security agent can take under the acceleration provisions. Take another example: a manufacture and supply agreement for a large, highvalue asset. Parties look at scenarios that could arise during the life of the contract, ‘­Disproportionate Penalties in Commercial Contracts’, ch 11; N Connolly, ‘Planning for Failure: Contract Design, Ineffective Bargains and Restitution’, ch 13; U Grušić and M Penades Fons, ‘Illegality in English Arbitration Law after Patel v Mirza’, ch 18, J Lowry and R Edmunds, ‘The Reform of Insurance Warranties: Looking Beyond the Past’, ch 19; M Ozdel, ‘The Right to Delivery of Goods under Contracts of Carriage’, ch 20. 5 See Loan Market Association (LMA) documentation for examples of provisions for different sectors in the syndicated loan market.

Terms Affecting Freedoms – A Response  445 which could give rise to an action for (i) breach of contract, (ii) frustration (iii) specific performance, or (iv) liquidated damages. In fact, the parties would often prefer to keep the contract in play and regulate the relationship between them on how to deal with different situations when things do not go to plan, especially for long-term contracts. Sometimes, neither would like to end the whole arrangement. So, for example, where there is a problem with the specification for the asset, the parties can set out contractually a method to propose a variation, discuss it, agree it, and then document and implement it as a variation procedure. Within the framework of the agreement, the parties can anticipate something going awry. Rather than waiting for and relying on the doctrine of frustration, they actively consider different routes that could be taken through the contract terms. The contract may include a methodology to vary the contract to the satisfaction of all parties to keep the contract alive, with the new circumstances incorporated into the scope of the contract. Identification of risks and situations that may arise is important. For example, in the event of Brexit, the parties could agree contractually who would be obliged to meet the costs if tariffs were to be imposed on importing an asset; and if those costs were not paid by the correct party, what the consequences would be. The process of risk identification, quantifying the costs and allocating them, means the dispute about who should pay does not reach the courts, as it is dealt with in the contract itself. This is clearly not possible in all situations: parties may not have foreseen an issue or set out how they would deal with it and what options could follow certain events. Remember the Icelandic volcano that grounded flights: was that an act of God or a foreseeable event – was it grounds for frustration? The English courts held that Brexit is not a ‘frustrating event’ in the context of a lease in favour of the European Medicines Agency6 (EMA) as tenant of premises for its headquarters in London. The lease would not be frustrated under English law by the UK’s withdrawal from the EU, even although the EMA claimed that under EU law7 it was required to move its headquarters out of the UK to another location within an EU Member State, namely to Amsterdam in The Netherlands. Many contracts we prepare, whether using market standard agreements or not (eg LMA language in syndicated loans, or a client’s preferred wording or form of contract), do anticipate things going wrong either as a minor breach or as a material breach of a contractual term. Agreements may set out a sophisticated approach to identifying the type of breach. Grace periods of, say, three to five days could be built in to grant extra time to deliver ongoing monitoring documentation, which could be considered as a minor breach in some cases. Specific provisions could be included to deal with expropriation, which is outside the control of the parties, or non-compliance with laws, stating remedies for the



6 Canary

Wharf (BP4) T1Ltd v European Medicines Agency [2019] EWHC 335 (Ch). (EU) 2018/1718.

7 Regulation

446  Jacqueline Cook parties within the remit of the agreement, say for the lender or security agent to take action to put the loan on demand, to make a demand for repayment or to enforce security. Remedies can include: giving time; waiving a breach; making an amendment to the contract to set new parameters; using an agreed variation process involving all parties with an agreed timetable; legislating in the contract to avoid the need to rely on the doctrine of frustration, for example through London Inter-Bank Offered Rate (LIBOR) replacement provisions; making a loan on demand; calling a guarantee; extending the time to perform; enforcing a guarantee; enforcing security and disposing of the asset; appointing an administrator to run and save the business; terminating the contract; allocating payments, or resale or disposal of an asset. All of these can be ‘inside’ the contract as methods of resolving difficulties or anticipated events. Collaboration between academia and practising lawyers in finance and commercial law – both in advising on transactions and litigation – and those proposing reforms to the law in this area would help to shape the law by reflecting how parties and their legal advisers use and apply contract law in practice. This way, large commercial contracts that are working well, not just those that are in dispute, which come before the courts and are reported in the law reports, could be used as examples of how parties contract and how they regulate remedies available to them inside the contract, before even considering reverting to court procedures.

IV.  Good Faith While English law does not have an express or implied duty of good faith in contract law, certain chapters in this volume gave examples of concepts akin to good faith duties in other jurisdictions, where a general doctrine of good faith is written into the civil code, and circumstances that could mean English law has its own variation.8 Some of the examples given and discussed were: • a duty to act honestly; • a duty to commit to collaboration for a common purpose; and/or • a duty to act in the spirit of negotiation. In English law, neither the duty nor the obligation to act in good faith is automatically implied in all contracts. If the parties wish to require that one or both of them act in good faith, they should state so expressly. During contract negotiations, were a party to refuse to agree to a duty to act in good faith as required by a contractual

8 The following chapters in this volume looked specifically at good faith: M Raczynska, ‘Good Faiths and Contract Terms’, ch 5; and PS Davies, ‘Excluding Good Faith and Restricting Discretion’, ch 6.

Terms Affecting Freedoms – A Response  447 term, would the other parties still wish to contract with that party? Would the counterparties question the other’s honesty and integrity in the real world? Semantics and language can be tricky to interpret for the lay person, and indeed for the legal community, as a party not acting in good faith is not necessarily the same as that party acting in bad faith. In some cases, bad faith could be seen to be similar to deceit or fraud. Nonetheless, English law has well-established doctrines of deceit and fraud, so is a doctrine of good faith really needed? Similarly, English common law respects the freedom of parties with unequal bargaining positions9 to contract, which of course in some contracts is very common. Having unequal bargaining power does not of itself mean that one party, the more powerful, is acting in bad faith. An obligation to act in good faith does appear in the legislation and the text of the Withdrawal Agreement10 for the UK to leave the European Union, where in Article 5 (Good faith), in public law terms, the state of the United Kingdom of Great Britain and Northern Ireland and the EU27 agree to act in good faith: The Union and the United Kingdom shall, in full mutual respect and good faith, assist each other in carrying out tasks which flow from this Agreement.

Could ‘good faith’ mean ultimate disclosure, fair play, and/or open and honest acts as part of the negotiating tactics? Or does ‘good faith’ mean something more along the lines of aiming for a mutual goal and aiming for even-handed dealings between the parties during negotiations, where negotiations are honest but may not disclose everything about the other parties’ plans?11 Some contracts in English law use other concepts in a similar way to good faith under other systems (eg contracts uberrimae fidei for insurance and employment, protections for each party under the law of surety and guarantee, relational contracts12 with the ultimate goal involving denial of or limitation of self-interest). Until recently, categorising a contract as relational would not necessarily lead to an implied duty of good faith. However, Fraser J, in Bates v Post Office Ltd,13 held that because a contract was relational, a term of good faith should be implied at law! Nine aspects are listed, but to me this one seems to top them all: 7. The contract will involve a high degree of communication, cooperation and predictable performance based on mutual trust and confidence, and expectations of loyalty.

9 Times Travel v Pakistan Airways [2019] EWCA Civ 828. 10 Agreement on the Withdrawal of the United Kingdom of Great Britain and Northern Ireland from the European Union and the European Atomic Energy Community, as agreed at negotiators’ level on 14 November 2018 (TF50 (2018) 55 – Commission to EU27), as amended on 17 October 2019 by the revised texts for the Protocol on Ireland/Northern Ireland (TF50 (2019) 64 – Commission to EU27); approved by the UK and EU Commissions and implemented through the European Union (Withdrawal Agreement) Act 2020. 11 Raczynska, ‘Good Faiths and Contract Terms’ (n 8). 12 See references and extracts in Davies, ‘Excluding Good Faith and Restricting Discretion’ (n 8). 13 Bates v Post Office Ltd [2019] EWHC 606 (QB) [725].

448  Jacqueline Cook When dealing with cross-border transactions where some jurisdictions do have an obligation to act in good faith, it is important to be aware of it, to work out where it should be expressly included in contracts governed by English law. The courts are ‘more ready to give effect to express terms to perform the contract in good faith’,14 even though a fairly restrictive approach is used. The courts do look at the question of good faith in relation to the exercise of powers and discretions. In this context, good faith can equate to the lack of ‘irrationality, capriciousness or arbitrariness’, with the courts balancing freedom and sanctity of contract with the ‘worst instances of opportunism’. The lesson in practice is to be as specific as possible. If, in the drafting, one party would expect another to consider particular pieces of information before exercising a discretion, this should be stated in the contract. Although not enforceable, provisions in some commercial and finance contracts are clearly drafted as an agreement to agree on an amendment or course of action in the future. We can see this in relation to the lead-up to the likely demise of sterling LIBOR from 2021, or possibly from late 2020, and the use of the LMA’s (Replacement of Screen Rate) wording for parties to agree to substitute a suitable alternative benchmark at a later date to calculate the interest payable on a loan.15 Having a common goal to keep the contract alive can affect the expected behaviours of the parties and the mutual trust and expected collaboration on the project.

V. Interpretation Could drafting become more certain just by having a well-thought-out interpretation clause? Could certainty come from more careful drafting? In Richard Calnan’s contribution to this volume, he raised many questions by reviewing a much overlooked boilerplate provision, in light of an increasing number of cases over the last few years on interpretation of contracts.16 Here are some of the thoughts aired by Richard Calnan and other contributors: • Should it be open to the parties to agree the rules of interpretation that they want to apply to the contract? • Should the courts apply the ordinary meaning unless there is clear mistake, such as a typographical error? • Should the parties set out a factual matrix or context in which they enter into the contract? 14 Raczynska, ‘Good Faiths and Contract Terms’ (n 8), also referencing M Arden, ‘Coming to terms with good faith’ (2013) 30 Journal of Contract Law 199, 200. 15 In LMA agreements the language used is ‘an amendment or waiver … made with the consent of the Agent (acting on the instructions of the Majority Lenders) and the Obligors’. 16 R Calnan, ‘Controlling Contractual Interpretation’, ch 4 in this volume.

Terms Affecting Freedoms – A Response  449 • In contracts nowadays, should parties also consider the content in relation to the news, world events, social media, etc? Further discussions looked at how common sense or context could be used to override the express terms of a contract. Richard Calnan recognises the changes in the courts’ methods of interpretation and how contextual interpretation is much more common. The court will look at the wording of the document in the context in which the document is set. This concept comes into focus sharply as lawyers start to revisit Interpretation and Construction provisions in light of Brexit. New terms such as ‘retained EU law’, being EU law applicable in the UK and frozen on ‘IP completion day’,17 will start to appear in contracts, as well as references to some of the myriad statutory instruments brought in to amend legislation to account for Brexit, on the basis of the Withdrawal Agreement18 with a transitional or extended transitional period. Richard Calnan has tabled a draft interpretation clause19 that looks at setting interpretation in context at the date when the parties enter the contract. No conclusions were drawn, but many issues were raised to provoke lawyers to think about this in relation to their industry standard documents and bespoke contracts. It was also noted that an entire agreement clause, used in some commercial contracts, is one way to confine interpretation to particular issues inside the contract, so that prior written or oral discussions are not part of the contract.

VI.  The Muddle of Reform Several chapters in this volume20 consider the reform of the law on assignment of receivables and what eventually became The Business Contract Terms (Assignment of Receivables) Regulations 2018 (SI 2018/1254)21 (‘the Regulations’). These Regulations are a restriction on freedom of contract due to the implementation of a policy decision from the UK Parliament into private law, and are a prime example of stop-start reform, in my view. In trying to implement a policy to allow small businesses access to finance via invoice discounting and forfaiting by selling receivables in return for cash, the first set of regulations to nullify any clause that prohibited the assignment of a ­receivable, having almost finished the journey through the legislative process, was

17 This is the end of the implementation period under the European Union (Withdrawal Agreement) Act 2020. 18 European Union (Withdrawal) Act 2018 and European Union (Withdrawal Agreement) Act 2020. 19 Calnan, ‘Controlling Contractual Interpretation’ (n 16). 20 See Stevens, ‘Binding Our Future Selves’ (n 2); N McBride, ‘“All Watched Over by Machines of Loving Grace”? The Inevitable Conflict between Contract Law and Free Speech in Cyberspace’, ch 14; Day, ‘Disproportionate Penalties in Commercial Contracts’ (n 4); S Green and A Sanitt, ‘Smart Contracts’, ch 10; and H Beale, ‘The New Override of Bans on Assignment of Receivables’, ch 7. 21 Under the Small Business Enterprise and Employment Act 2015.

450  Jacqueline Cook stopped and withdrawn just before the regulations came into force. The drafting approach adopted seems to favour (i) setting out a premise then (ii) excluding certain things from it (ie certain parties and certain types of contract). This method has been used in these Regulations to affect all contracts, but not those listed as excluded and those parties listed as excluded. Defining a concept in the negative, or as ‘not’ something else, is not a very clear way to proceed.22 Many claim that English law is clear and accessible, but in some cases this method of drafting itself can bring misunderstandings. Freedom of contract would allow one party or another to restrict the rights of one or more parties to assign the rights to a third party. This is not uncommon in finance transactions, where the party as borrower wants to know with which lenders it is contracting, especially since there are some lenders it would rather not contract with at all, for example following an assignment. The original intention to keep the policy reform concerning the assignment of receivables to small businesses, did not come through in the first set of regulations but has done so in the second set of the Regulations, which are now in force. What is challenging for the interpretation of contracts, focusing on small businesses for a moment, is that the story does not end with the terms of the contract. There will be other statutory provisions that will affect the terms of the contract. Yet parties and businesses have to be able to interpret and comply with their own contracts and adjust their behaviour required by the contract and English law, even where there is no dispute and no action before a court. In short, the legislation has introduced the need for all contracts to be analysed to assess whether the contract is subject to the Regulations and whether it contains the offending provision on assignment, which will then have no effect. Does this not put a burden on small, medium and large businesses and their advisers, to check if the Regulations apply? Perhaps this could be better done by drafting legislation that identifies the sector to be subject to the policy, and then introducing a regulatory requirement for that industry sector? Principles of contract law are being propounded only by reference to examples of clauses, or through cases that have fallen into a dispute. Having access to large commercial contracts to analyse, some academics and legal writers would see not just what the courts interpret in a disputed contract, but also how contracting parties look ahead and anticipate breach and real-life situations, providing different courses of action within the parameters of the contract itself.

22 For example, under reg 1(3) a ‘large group’ means a group that is not a small group or a medium group (within the meanings given by the Companies Act 2006). See reg 3 for exceptions for suppliers who are large enterprises or special purpose vehicles. Reg 4 lists contracts that are excluded, which is in some ways a response to objections raised from networks and trade bodies. It should be commended for narrowing the scope of the rules to just the purpose of the policy. For example, reg 2 does not apply to terms in a contract that is a contract for prescribed financial services, a contract for interests in land or national security, a petroleum licence, an option, future, swap or contract for differences, a transfer of ownership in a firm or in a contract outside a trade, business or profession.

Terms Affecting Freedoms – A Response  451 Many commercial parties do consider minor and material breaches, and options to extend time limits to keep the contract alive. It seems that we need to let our academics and lawyers involved in law reform, in and outside government, have the chance to analyse current commercial contracts from practice. But of course, without confidentiality agreements in place and the agreement of clients and contracting parties, it is only papers concerning contracts in disputes that will come before our judges and law makers. If one thing is clear, it is that this does not give the whole picture about the range of contractual terms being used every day in practice, which affect behaviours and restrictions on freedom of contract. So, looking ahead, how will contract law cope with the rise of coding, smart contracts or the automatic logic programme to buy and sell without a human element or the need for additional decisions or discretion? Can lawyers lead the discussion, or do we need technology specialists to show what is happening in reality, or indeed in the virtual sphere, so that we can impose or equate the processes to stages of the creation and enforcement of a contract under English law? Have lawyers already lost out to coders and programmers? And how will judges interpret these contracts and lawyers advise on them? Policy decisions are pushing more regulation on to companies and businesses, not just in the UK but in the EU, EEA and globally. The effects of additional regulation and compliance are appearing in private contractual terms, as we see with, for example, data and GDPR,23 bail-in24 provisions in finance documents, ‘people of significant control’25 provisions in corporate and in finance documents, obligations for disclosure on environmental compliance, health and safety, modern slavery, work practices and sustainability. Yes, there are good reasons to impose these policies on society and businesses, but should – and can – contract law be the instrument to implement them? Do we like the discipline of including provisions in contracts to enable due diligence and disclosure, with an eye to ‘knowing who our customer is’ or ‘knowing with whom we are contracting’? Will contracts then just continue to get longer and longer to cater for these new arrangements and disclosures? All in all, more dialogue between businesses, legal advisers, law makers, judges, academics and government would provide a more realistic picture of contract law as it is being used every day, and develop clearer and more realistic mechanisms for policy implementation and for reform in the future. Using sector-based links with trade bodies, academics and law makers, where people can feel they are making a difference, may be the most efficient way to bring contract law into focus with the current and future reality for commercial and finance transactions.

23 General Data Protection Regulation EU/2016/679. 24 EU Bank Recovery and Resolution Directive 2014/59/EU, implemented into the law of the UK on 1 January 2015. 25 See n 3.

452

INDEX Agreement see also Contract enforceable contract, need to distinguish, 13 Anti-oral variation clause consideration, effect of provision, 37 departure from, 45–6 doctrine of apparent authority, 46–7 generally, 35–6, 43–9 ineffectiveness, 42–3 informality, 45 representation, restriction on power to make, 47–9 validity, removal of uncertainty, 35 warranty, restriction on power to make, 47–9 Arbitration clause generally, 7, 381 illegality, see Illegality Assignment ban on see Non-assignment clause receivables, of, Regulations, 449 Autonomy see Freedom of contract Bill of lading see Delivery of goods Blockchain Bitcoin— Distributed Ledger Technology, use of see Distributed Ledger Technology meaning, 192 proof-of-work, use, 193 dispensing with, 195 legal supervision, dispensation, 199 meaning, 192 Breach of contract inside or outside contract, 443–6 insurance warranty see under Insurance warranty loan agreement see under Loan agreement termination see Termination Brexit frustrating event, whether, 445 impact on the content of contracts, 449 lease, whether affected, 445 non-disclosure agreement, 317–18, 333–4

Carriage of goods cargo interests, 421 contract— discharge, 422 nature, 422 conventions governing, historical background, 421 delivery see Delivery of goods, 421 tackle-to-tackle rule, 422 Choice of court see Freedom of contract Choice of law see Freedom of contract Consideration bona fide purchase, defence, 30 corresponding portions, identifying, 291 deeds, 26 dispensation, 25 enforceability, as requirement— generally, 27 past consideration, 28 third parties’ rights, 28–9 unilateral contract, 27–8 equity will not assist a volunteer, 30 evidential value, 26 exit from contract, 33 failure of— generally, 285 total see total failure below formality— as, 25 lack of, effect, 27 from whom moving, 29 generally, 25–7, 33–4 importance, 13–14 increasing and decreasing pacts, 31–3 invalid contract, 287–8 nearly universal application, 25 part payment, 31 past, 28 performance of promise, 31 pointlessness, 26 poor probative force, 26 promise— in exchange for, 29 provided by, 26

454  Index substantive nature, 29 third parties, 28–9 total failure— attenuation of rule, 289 avoiding restriction of rule, 289–90 generally, 285, 288–9 net enrichment, calculation, 289 recovery where, 288 severable part of consideration, 288 variations not supported by, 30–1 Contract agreement, need to distinguish, 13 autonomy see Freedom of contract behaviour— ethical clauses see Ethical clause self-interested see Self-interested behaviour breach see Breach of contract competitive cooperation, spirit of, 89 compromise, drawbacks, 240 consideration see Consideration contextual relationism, 239 effect on rights, 13 enforceability, 24–5 exiting, difficulties, 33–4 failure of, planning for— benefits, 260 costs involved 269 generally, 267–8, 290–1 restitution see Restitution unforeseen eventualities, 268–70 unjust enrichment see Unjust enrichment valid contracts as justifying grounds, 270–6 formalist approach, 239–40 formalities, effect, 24–5 frustration see Frustration good faith see Good faith honesty see Honesty interpretation see Interpretation length, 51 loan agreement see Loan agreement narrow textual approach, consequences, 241 negative obligation— generally, 139–40 loan agreement see under Loan agreement nullity, 24–5 offer and acceptance, need for, 18 party autonomy see Freedom of contract positive obligation, 139 privity see Privity

promise— promise, contract not constituting, 15 promissory theory, divergence from, 19 see also Promise below purpose, 74–5, 77, 83, 85, 370 relational contract scholarship, and, 239 relational fundamentalism, 240 rights under— conditions for existence, 31–2 increasing and decreasing pacts, 31–3 need for, 14–15 self-interested behaviour see Self-interested behaviour subjective approach to existence, 21 subsequent conduct, exclusion, 20 termination see Termination terms affecting freedoms, 2–3 textual formalism, 239 variation— anti-oral clause see Anti-oral variation clause whether need for further consideration, 30–1, 33 will, relevance, 10–11, 19 Contract law bilateralism, 176, 180 classical view, 65 individualism ethic, 65, 70, 84, 87, 167 need for, 15 public interest regulatory role, 165 relational ethic, 84–5 rules, importance, 13, 15 Cryptocurrency Bitcoin see Blockchain Damages bilaterality concept, 180 contract designed to facilitate recovery, 180 Delivery of goods bill of lading— document of title, 424 lost or stolen, 426 misdelivery see under misdelivery below cargo interests— limits on right to delivery see limits on right to below meaning, 421 deemed, 424 discharge from vessel— after, 422 different safe port, 432

Index  455 exclusion clause— direct bailment relationship, where, 437–8 null and void, where, 431 theft, where, 435 generally, 421–3, 438–9 Hague-Visby regime, relevance, 422 letter of indemnity, against, 424 liability— discharge, 430–1 misdelivery for see under misdelivery below limits on right to— exclusion clause see exclusion clause above generally, 427–9 Hague-Visby Rules, application— breach of Rules, 430 construction of Rules, 431, 434–6 contractual force, 433–4 different safe port, discharge at, 432 fundamental breach doctrine, use, 435 generally, 428–9, 434 letter of indemnity, delivery against, 432–3 limited nature, 439 mandatory application, 431 misdelivery outside sphere, 434–6 misdelivery within sphere, 429–34 substituted ways of delivery, 432 sui generis treatment of cases, 435 tackle-to-tackle period, in, 422, 429, 439 theft, where, 435 time-bar clause, 428, 431, 434 package limitation, 428 theft, where, 435 time limit for bringing claim, 428 tort claim, 430 misdelivery— avenues for bringing action, 436–8 bill of lading— block-chain-based, 427 forgery, 427 in absence of, 423, 424, 426, 429 lost or stolen, where, 426 presentation rule see presentation rule below claim for, grounds for bringing— contract, in, 436, 437, 438 conversion, in, 436, 438 direct bailment relationship, where, 437–8 generally, 436 parallel claims, 436

discharge, after, 430 electronic early release system, use, 425 generally, 423–4 Hague-Visby Rules, application of see under limits on right to delivery above letter of indemnity, 424 liability for— fragmented nature, 439 protection from, 423 limits on right to delivery see limits on right to above presentation rule see presentation rule below tackle-to-tackle period, within, 429, 439 time of, 429 presentation rule— absolute obligation to deliver, 427 accomplished bills, 425 applicability, 425–6 commercial sense, 424 custom, giving way to, 426 drawbacks, 424 effect, 423, 424 electronic early release system, 425 exceptions, 426–7 forged bill, 427 generally, 424–7 meaning, 423 void bills, 425 refusal, right of, 424 substituted ways of delivery, attempt to impose, 432 tackle-to-tackle rule, 422, 429, 439 Director de facto— meaning, 154 test for determining status, 154–5 loan agreement, duties under see under Loan agreement shadow— duties where also lender, 156 generally, 155 meaning, 155 shareholder as, 155 whether lender can be, 155–6 Dishonesty self-interested behaviour, as part of see under Self-interested behaviour (honest performance, exclusion of duty of)

456  Index Disproportionate penalty breach of contract— abolition of need for, Law Commission’s view, 216 limits on drafting around rule, 217–19 need for, retention and justification, 215–17 ‘classic’ penalty clause, effect, 222 Consumer Rights Act 2015, effect, 212 contractual clause as penalty, whether enforceable, 215 default interest, 213–14, 228, 231 deposits, 223 determining whether, 212, 217 deterrence clause— disproportionate secondary obligation, 230 generally, 227–8 legitimacy, 228 penalty clause, 236 proportionate nature of deterrence, 227 rejection, 227 rent increase for breach of side letter, 231 see also detriment as below detriment as— essence of, determining, 213 irrelevant to innocent party’s losses, 227–8 drafting to oust rule against, 218 essence of, 213 forfeiture, relief from, 222, 223 general test for, 227 generally, 211–12, 237 going beyond what parties have agreed, 237 interest provisions, whether enforceable, 231 justification— autonomy of parties, need to promote, 235 freedom after contracting, importance, 234–5 generally, 233–7 lack of, 233 redundant nature of doctrine, 234 legitimate interest— determining, 226–7 enforcement of primary obligations, in, 226, 227 generally, 225–7 substitution of primary obligations, in seeking, 226 loan extension, fees on, 217 non-compete covenants, 232 non-monetary detriments, 222

primary obligation— damages as default remedy, 229 enforceability, 219, 226 legitimate interest in enforcing etc see legitimate interest above payment obligation, enforcement, 230 positive and negative, distinguishing between, 226, 229–33 secondary obligation distinguished, 220–22 specific performance as exceptional remedy, 229, 230 third-party or public interest in enforcement of, 227 proportionality— application of test— damages see damages below default interest provisions, excessive, 231 negative primary obligations, 232–3 non-compete covenants, 232 positive primary obligations, 229–32 rent increase, 231 specific performance, whether appropriate, 229, 230 damages— compensatory basis, on, 233 injunction, in lieu of, 232 proportionality test, application of, 233 default interest provisions, excessive, 231 determining, 228 deterrence clause see deterrence clause above formulation of test, 223–5 freedom of contract, presumption in favour, 228–9 late payment of debt, damages for, 231 legitimate interest, analysis, 225–7 non-compete covenants, 232 rent increase, 231 understanding test, 225–9 public policy against, 236 reasonable pre-estimate of loss— detriment as, 213 where detriment not comprising, 214 reversal of transfer of title, clause as to, 222 secondary obligation— damages at law, 219 development of concept, 219 disproportionate, effect where, 230 generally, 219–20 meaning, 219

Index  457 primary obligation distinguished, 220–22 requirement additional to breach, 219 subject to penalty rule, 219 types subject to penalties rule, 222–3 sham doctrine, approach to, 218 substance over form approach, 218, 219 tests for, as quasi-statutory code, 213 tort law, 236–7 transfer of property, clause requiring, 222 unambiguous language, use, 218, 219 unenforceability, 223 Distributed Ledger Technology altruism, alignment with self-interest, 195 common characteristics, 192 directed acyclic graph, use, 196 double spending, prevention, 192 generally, 192 Genesis Block, 192 hash algorithms, types, 193–5 hash puzzle, 193 node— blockhash, use, 194 meaning, 192 nonce, use, 194 proof-of-work, 193, 195 use, 193 verification process, 195 proof-of-work, use, 193, 195 puzzle-solving approach, 193–5 total validation replacing control, 192 transparency, achieving, 192–3 trapdoor maths, use, 195 Entire agreement clause advantages, 242 agent— insurance broker as, 243 integration into firm, 243 boilerplate, as, 249–52 common law regulation, 255–8 compromise, drawbacks, 240 contextual relationism, 239 contra proferentem construction, 241, 255–8 courts’ involvement, 240 documentary fundamentalism, preference for, 266 due diligence, need for, 245 enforcement— arguments as to, 241, 252–5 courts’ powers, 266 renegotiation, courts’ powers on, 266

fraud and public policy— fraus omnia corrumpit, 248 generally, 246–9 generally, 239–42, 266 growing use, 245–6 intra-organisation control, and, 243 liability— exclusion, 244 strict, 248 logically unsustainable, whether, 252–5 merger clause, 246 misrepresentation— onerous statutory reforms, 245 oral, protection from, 244–5 rescission for, 245 stringent approach of English law, 245 narrow textual approach, consequences, 241 objections— boilerplate, clause as, 249–52 fraud and public policy, 246–9 logically unsustainable. 252–5 party autonomy, importance, 240, 241 pluralist solution, 240 protection of parties, approach to, 244 purpose and use, 242–6 reasons for use, 245, 246 relational contract scholarship, and, 239 relational fundamentalism, 240 statement made in negotiation, protection from, 243–4 statutory regulation— generally, 241–2 legislative reform, proposals for, 264–6 scope, confining, 266 textual drafting techniques, use, 246 textual formalism, 239 upholding, arguments as to, 241 widespread use, 242 Ethical clause contents, 164 core contractual obligations, distinguishable from, 164 distinctiveness— generally, 166 public dimension see public dimension below economic motivations driving use, 170 enforcement— contract law, limitations inherent in, 186–7, 190 contractual governance structures, and, 184

458  Index court and defendant in different jurisdictions, 179 damages— bilaterality concept, and, 180 contract designed to facilitate recovery, 180 possible development of new rule, 180 regulatory sanction, as, 180 reorientation of rule, 179–80 difficulties inherent in, 178, 179 exclusion clauses, prohibition etc, 185–6 generally, 165, 177–8 lead firm, by— certification scheme, compliance with, 187 contract law, use, 179, 181 contractual governance structures, and, 184 damages rules, ineffectiveness, 180 discretionary nature, 178 factual control of suppliers, and, 183–4 generally, 178–82 non-legal mechanisms, use, 187 obstacles, 178–182 privity, effect, 181 reasons for not enforcing, 178–9 reputation, claim for loss, 180 non-legal mechanisms— certification scheme, compliance with, 187 contract law’s weaknesses, 188, 189 drawbacks, 188–9 examples, 187 generally, 187–9 lack of political neutrality, 188 performance interest, claim based on, 180 private law, in, 186 regulatory dimension, and, 177–8 reputation, claim for loss, 180 third parties, and— arguments in favour of, 182 beneficiary of promise by lead firm, where, 184, 186 contractual governance structures, and, 184 crucial mechanism, as, 182 exclusion of right by lead firm, 185 factory, where duty to inspect, 183 factual control of suppliers, and, 183–4 generally, 182–7 intention of parties, importance, 185

intention to benefit third party, where, 185 obstacles, 182–3 successful claim, elements of, 184–5 transnational claim, problems associated with, 182–3 waiver of right to, 178, 179 ethical obligations— breach, effect of privity of claim, 181 cascading up or down the chain, 181 examples and effect, 164 factors explaining inclusion, 164 generally, 163–6, 189–90, 443 greenwashing tool, as, 179 limits on freedom of contract, and, 165 marketing tool, as, 443 meaning, 163 no obligation to contract ethically, 177 public dimension— anti-bribery terms and conditions, use, 172 aspects to, 166–7 collective interest— sensitivity to, 171 threat to, 170 contractualisation endangering public interest, 170 domestic legislation, effect, 171–2 freedom of contract, and, 170–1, 173 general interest, private parties defining, 169 generally, 166–74 ideological input, 169 international standards, use, 172–3 politically constructed freedom of contract, 173 protection of public interest, 167–8 protection of state, absence, 173 public nature of clause, 167 transnational sphere, public ordering in, 173 unfettered private power, concerns as to, 173 welfare-enhancing nature of contracts, 168 regulatory dimension— contracts as a form of regulation, 175–6 enforcement, and, 177–8 generally, 174–7 international instruments, disadvantages of using, 174 no obligation to contract ethically, 177

Index  459 private regulators, contracting parties as, 176 state power, diminishment, 174 transnational private regulation, and, 174 standards etc set by, 164 substantive content, origins, 172 third-party— enforcement by see under enforcement above interests, protection, 167 Fairness absence, 89 Force majeure clauses— approach to drafting, 282–3 contents, 281, 282 generally, 281–2, 291 partial performance of obligations, where, 282 remedies available, 282 scope, 281 modification of obligations, 282 Formality agreement, in, 24–5 consideration, and, 25, 27 Fraud on a power basis of doctrine— disagreements over, 107 implied term not comprising, 107 construction, as rule of, 107 context, importance, 86–87, 107 contract terms, need to examine, 111 fiduciary powers, not limited to, 86, 106 gateway issue, absence, 107 generally, 86, 112 good faith distinguished, 86–87, 106 implied term, not resting on, 86–87, 110 intention of donor, importance, 107 judicial analysis, 105–6 long-standing nature of doctrine, 86, 105 mandatory nature of rule— breach, consequences, 110 generally, 107, 110 prohibition on excluding, 111 meaning, 86, 105 narrow scope of doctrine, 86, 110 parties agreeing to further controls, 111 principle, nature of, 86, 106, 107 Free speech conservative voices see limits on the Internet below hate speech: meaning, 302

liberal gnosticism see Liberal gnosticism limits on the Internet— conservative voices, 294 contract law as vehicle for, 294 Facebook, 302 generally, 293–4 liberal gnosticism see Liberal Gnosticism protecting free speech see methods of protecting below Twitter, 302–3 US First Amendment, and, 310–11 Youtube, 303 methods of protecting— common callings, law on, 309–10 contractual arguments, 305–8 discretionary powers, reasonable exercise, 308 honesty in performance, whether implied, 307 implied obligations, 306–8 legislation, relevant, 305, 307 limits of contractual arguments, 308 non-contractual arguments, 308–10 private law of little use, 310–13 procedural terms, whether implied, 307 relational contract, implied term of honesty in, 307 tort, in, 308–9 understanding nature of contract, 305 private law— enjoyment of goods, protection of, 312–13 of little use in protecting, 310–13 potential for protecting, 312–13 purpose, 312 truth, importance of debating questions of, 304, 313 Freedom of contract agreement, contract’s contents as matter of, 10 arbitration clauses, enforceability, 7 balancing approach to, 12 contact law, role, 9 disproportionate penalties, rule against, 7 effect, 450 entire agreement clauses, 8 exclusion clauses, types, 8 freedom to contract distinguished from, 1 generally, 1–2, 12, 450 ideas embodied in, 1 illegality, multi-factorial approach to, 7 importance, 1 judicial interpretation, 1, 12

460  Index justifications for existence of, 10 liability, exclusion of— clauses operating as, 8 continuing existence of underlying obligations, 23–4 contract of carriage, 8, 9 damages, effect on right to, 24 insurance contracts, 8–9 penalties, approach to, 7 terms limiting or excluding, 7–8 limitations on, 442 limits of principle, 2, 4–5 meaning, 13 non-assignment clauses, 6 non-disclosure clauses, 6–7, 351–2 non-oral variation clauses, 5, 6 non-reliance clauses, 8 objective approach to contract’s contents, 10 obligations, control of extent, 90 online social media platforms, use, 7 other freedoms as basis of terms, 10 party autonomy— control, 6 moral philosophical approach to, 10–11 obligations, enforcement, 9 public policy considerations, 1, 6, 7, 12 reasons for entering into contract, 11 restitutionary liability, control, 8 sanctity of contract, and, 1 terms affecting freedoms, 2–4, 5 trends in English contract law, 4–5 unrestricted nature of, 5 see also Party autonomy in Private International Law Frustration advance payments, return of, 280–1 expenditure prior to, loss sharing, 281 legislation, 281 loss, court’s power to make allowance for, 281 restitution, power to limit, 281 valid contract, need for, 281 Gaming agreement insurance agreement, comparison with, 14 legal enforcement unnecessary, 14 nature, 14 Global value chain contract ethical clauses in see Ethical clause generally, 163 legal infrastructure necessary for, 163 network-like structure of chain, obstacles posed by, 180

Good faith abuse of right doctrine— adjudication model, and, 74 application, 70–2 purpose-based model, and, 70–2 adjudication model, see models below adjustment of amount owed to discharge obligation, 75 attitude, as, 70 Australia, in, 69, 70, 73 background duty, as, 80 bad faith, examples, 447 behaviour, as standard of, 80, 84–5 Brexit legislation, in, 447 Canada, in, 73–4 common goal to keep contract alive, where, 448 constituents, 78–9 contract terms, where no effect on, 70 control of contractual discretions via duty of, 66 cooperation, duty of, 91–2 costs, negotiation of extra, 91 counterparty, calculation of sums due from, 91 cross-border transactions, 448 disputes to be settled by ‘friendly discussion’, 91 doctrine, whether necessary, 87–88, 447 duties beyond honesty, incorporating, 100 duty— examples, 446 exclusion see exclusion below implied see implied duty below whether parties under, 111 exchange justice, 73 exclusion— contract silent as to good faith, where, 93 courts’ reluctance to exclude, 100–1 generally, 111–12 reasons for, 100 targeted exclusions, 101 see also Self-interested behaviour below express terms as to— absence, effect, 111 effectiveness, 66, 91 incorporation into contract, 111, 112 need for, 446 foreign jurisdictions, in, 68–9 France, in, 68, 71–2 future amendment or course of action, and, 448

Index  461 future legal developments, 88 generally, 65–8, 87–8, 111, 446–8 Germany, in, 74–5 honesty— as, 69–70 behaviour according with, 77 constituent of, as, 78 contracts lacking need for, 77, 78 exclusion of term as to see Self-interested behaviour (honest performance, exclusion of duty of) insurance contract, in, 77 judicial approach, 76 reasonable expectations of, 76–8 see also Honesty (exercise of contractual powers, and) implied duty— basis for implying, in fact, 78–1 basis for implying, in law, 81–5 cooperation, and duty of, 83 courts’ circumspect approach to, 81 courts’ wariness in implying, 99 courts’ willingness to imply, 66 generally, 67–8, 78, 111, 446 models, when taking account of, 81 no automatic implication, 446 overlapping but independent terms, 80 relational contract, in see relational contract, implied duty in below relevant contracts, 81 specific behaviour expected, as to, 84–5 improper purpose, and, 106 individualistic ethic, accommodated within, 70 interpretation model, see models below interpretation of contract in light of reasonable expectations of honesty, 76–8 interpretation clause as to, 60–1, 63, 64 judicial approach, 83 meaning, examples, 80, 447 medical practices, exchange of, 75 models— adjudication, 74–5 generally, 68–9, 88 implied duty, 72–4 interpretation, 69–70 purpose-based, 70–2 need for, where, 79 obligation— adjustment of amount owed to discharge, 75 no effect on, 70

powers and discretions, exercise, 448 principle— suggested adoption, 67 whether need for in English law, 68, 88 proper purpose, doctrine— abuse of right, and, 87 discretions, exercise, 85 exercise, 86 French courts’ approach to, 86–7 generally, 85–7, 88 improper purpose, exercise, 86 origins, 86 public decision making, analogies drawn with, 85 proper purpose rule, and, 86 purpose-based model of good faith, and, 86 purpose-based model see models below refusal to act in, likely effect, 446–7 rejection of duty, 65–6 relational contract, implied duty in— boundaries of contract, 96–7 characteristics of contract, 97–8 cooperation, duty, 83 criteria for, tightening, 85 established concept, as, 96 fact, implied in, 96 generally, 82–5, 95–9, 447 identifying relational contract, 97–9 law, implied at, 97, 99, 447 problems identifying, 82 purpose-based model, comparison with, 84 relevant types of contract, 95–6 specific behaviour, courts’ attempt to identify, 84 specific duties, courts’ attempt to identify, 84 unstable application of criteria, 60, 83 Scotland, in, 69 self-interested behaviour— imposing limits on, 70, 72–3 see also Self-interested behaviour silence, in case of, 93 situation-sense approach, 73 tools for interpreting concept of, 67–8 unequal bargaining power, relevance, 447 USA, in, 68 Goods carriage see Carriage of goods delivery see Delivery of goods

462  Index Hague-Visby Rules delivery of goods, and see under Delivery of goods (limits on right to) Honesty exclusion of term as to see Self-interested behaviour (honest performance, exclusion of duty of) exercise of contractual powers, and— capricious behaviour, 110 discretion— meaning, 108 target of, determining, 108–9 whether power counting as, 107–8 factual background, relevance, 108 fraud on a power see Fraud on a power generally, 76–78, 104–11 judicial review, unhelpful analogy with, 105 public law analogy, rejection of, 105 purpose, invoking concept of, 108 swap contract, 109–10 variation of interest rate, 108 Wednesbury unreasonableness, dismissal of approach, 104–5 good faith, as part of see under Good faith interpretation of contract in light of reasonable expectations of, 76–8 Illegality arbitration, relevance in, conflict of laws, and see conflict of laws below complex and controversial nature, 382 generally, 381–4 conflict of laws, and— common law, in— arbitration law, 386–8 generally, 384–6 foreign law, contract subject to, 385–6 generally, 383–384, 401 governing law— exceptions, 384–5 proper law of contract, 384 policy-approach to, 388–9 post-arbitral-award litigation— arbitrability rule, 395 balancing of competing policies, 395 breach of contract prior to illegality, 396 bribery, where, 396 enforcement and annulment proceedings distinguished, 397–8 EU law, violation, 394

flexible policy-based approach, 395–6 generally, 393–5 integrity of legal system despite arbitration, 399 legal certainty, relevance, 399–401 ossification risk, 399–401 public policy as ground to set aside, 394 voidable contracts, 396 pre-arbitral-award litigation— appropriate methodology, 392, 393 English law governing, where, 391, 393 EU law triggered, where, 391 foreign law governing arbitration agreement, 393 generally, 390–3 proportionality test, use of, 392, 393 public policy, agreement contrary to, 392 stay of proceedings, 390–1 types of invalid agreement etc, 391–2 turpitude, 389–90 Insurance agreement gaming agreement, comparison with, 14 Insurance warranty antecedence, 407–9 basis of contract clause, prohibition, 403 breach— date of remedying, 416 discharge of contract on, 416 modern approach, 405, 416 non-consumer contract, in, 416 causation requirement, whether part of law, 418 contractual construction— absence of word ‘warranty’, 410 avoidance of warranty, 413–14 contra proferentem rule, 412–13 cross-incorporation of terms, 410, 412 current legislation, under, 419 de minimis principle, 415 descriptive of risk, whether term, 411–12 equitable resolution, attempt at, 410 forensic deconstruction of language used, 414–15 generally, 406, 407 judicial contortions, 409–15 multi-sectioned policy, 412 nuanced approach, 414 plain language, 411, 413 suspensive condition, 406, 414, 415 temporary suspension of cover, 412 unreasonable or draconian terms, 411 varying approaches to, 409

Index  463 cross-incorporation of terms, 410, 412 duty to bring nature of term to insured’s attention, 417 electrical installation, testing, 405 fire damage, as to, 405, 408 historical background, 403–4, 407–9 legislation— causation requirement, whether part of law, 418 generally, 416–19 increase of risk clauses, 417 possible problems not covered by, 418 suspensive condition, warranty as, 416 life insurance, facts concealed, 409 meaning— difficulties over, 403 marine insurance purposes, for, 404 modern approach— breach, to, 405, 416 generally, 416–19 legislation see legislation above reduction in potential for dispute, hopes for, 418 multi-sectioned policy, 412 non-consumer contract, in, 416 proportionate response to breach, 405, 415 reform of— contractual construction prior to see contractual construction above generally, 403–5, 415 historical background, 403–7 legislation see legislation above problems faced by court, 405–7 twentieth century calls for, 404 strict compliance, by reference to, 404 suspensive condition, as, 406, 415, 416, 419 suspensory condition, as, 410 unwanted consequences, court’s attempt to avoid, 406 waiver, 404 Internet liberal gnosticism, and see Liberal gnosticism (control of speech on Internet) limits on free speech see under Free speech Interpretation background information, use, 51, 52–3, 57 Brexit, effect, 449 clauses— agreement over rules to be used, 448 ambiguities, dealing with, 59

background facts— context, interpretation of words in, 58 exclusion, 57 generally, 56–8 identifying, 57 limits on use, 57 relevance, 57 certainty of drafting, and, 448 construction, canons, 62–3 contents see contents of clause below contra proferentem rule, excluding, 63, 64 discretions, exercise, 62 draft example, 59–60, 63, 64, 449 eiusdem generis rule, 62–64 entire agreement see entire agreement clause below estoppel by convention, and, 63 example, 59–60, 63, 64, 449 good faith, duty, 60–1, 63, 64 implied terms, 61 lists, avoiding use, 63 ordinary meaning of words, as to, 58–9, 448 recitals, use of— generally, 57 see also background facts above rectification, as to, 63 use of— generally, 60–3 reasons for, 55–6 scope for, 55 clear drafting, need for, 54 common sense, use, 52, 449 context, importance— change in approach to, 53 generally, 52–5, 57, 58, 449 previous approach, 52–3 text making context relevant, 58 contra proferentem rule, excluding, 63, 64 eiusdem generis rule— excluding, 63, 64 use of, 62 entire agreement clause— binding nature, 54 effect, 54, 449 generally, 54 limit on effectiveness, 55 purpose, 54–5 expert translation evidence, and, 205 generally, 51–2, 63, 79, 88, 448–9 good faith, duty— generally, 60–1, 63, 64 see also Good faith

464  Index honesty, in light of expectations of, 76–8 implied terms, 61 industry-specific context, 205 lists, avoiding use, 63 matter of judgment, as, 54 objective meaning of language used, ascertaining, 203 purposive, 53 silence, of, 93 smart contract, of see under Smart contract strict approach to, 92 textual, 53 uncertainty in, 54 Legal duty exclusion of liability, continuing nature of, 23–4 liability for failure to comply with, 23 limits on imposition of, 16 philosophical analysis, 17 unenforceability, 23 Legal obligation see Legal duty Liberal gnosticism alternative viewpoints, need for, 304 beliefs inherent in, 297 communism as ultimate form of, 298 computer world as manifestation of, 299–301 control of speech on Internet— abusive behaviour, 302, 303 contract law, use, 302 cyber-bullying, 303 Facebook, 302, 304 generally, 301–2 harassment, 303 hate speech, 302, 304 hateful conduct, 302–3 methods, 301–2 suggested reforms to protect free speech, 304 suppression of contrary views, 303 terms of service, as method of, 302 Twitter, 302–3, 304 Youtube, 303, 304 criticisms, refusal to accept, 299, 301 dangers inherent in, 299 exclusion, end to cruelty of, 297 freedom of speech— control see control of speech on Internet above in conflict with, 304 methods of protecting see under Free speech

futility of, 298 generally, 294, 295–7 gnosticism— generally, 295 modern form, 295 Nietzsche’s approach, 295–6 philosophical roots, 295 Nietzsche’s opposition, 295–6 plasticity of profession, 300 potential objections, 297–9 prevalence and fervour of, 299–301 self-respect and self-worth, importance, 300 Solzhenitsyn’s analysis of good and evil, 298 summum mallum, 296–7 worst outcome from implementation, 299 Loan agreement breach of contract— acceleration of obligation to lend, 147–8 adverse effects for borrower, 147–8 borrower’s best interests, in, 150–1 cancellation of obligation to lend, resulting in, 147–8 consequences, 150 corporate borrower, by, 147 damages not a helpful remedy, 149 directors’ approach to actions leading to, 153 directors’ liability for inducing, 146–7 early repayment, reluctance to trigger, 151 efficient, 149–50 ex ante remedy, 145 ex post remedy, 145 further drawdowns, refusal to allow, 151 general remedies, 145–6 insolvency of borrower— consequences, 150 effect on lender, 147–8 lender’s reluctance to bring about, 150 lender’s need to know, 152 negative covenants to prevent see negative covenants in below standstill agreement, entry into, 150 directors’ duties— alignment of interests, 148–54 breach of contract, and see breach of contract above effect of negative covenants on behaviour, 153–4 lender, acting in interests of, 154 nature, 148–9 overly harsh contracts, as to, 157 restructuring loan terms, 151–2

Index  465 shaped by negative covenants, 151–3 shareholders’ interests, to act in, 149 waiver of covenant, seeking, 152–3 event of default, remedies etc, 144–5 financial covenants in, 154 lender’s liability as director, 154–6 negative covenants in— ‘agency problem’ view, 148 asset disposal, restrictions on, 158–9 borrowing, restrictions on, 159 breach of see breach of contract above classic view, 148 content— generally, 156–8 types of covenant, 158–61 directors’ duties see directors’ duties above examples, 143–4 generally, 143–4, 158, 161 ‘get-out’ clause for lender, as, 147 negative pledge clause, 159–61 prohibitions with exceptions, as, 156, 157 purpose, 145, 149 remedies for breach of see under breach of contract above return of capital to shareholder, prohibition of, 161 waiver, 152–3, 156–7 wish to avoid, 152 purpose of contractual terms in, 140–2 risks faced by lender— generally, 140–1 terms to alleviate, 141–2 terms in— event of default on breach, 144–5 financial covenants, 143 generally, 142 information, terms as to provision of, 142–3 lender’s risk, to alleviate, 141–2 negative covenants see negative covenants in above Negative obligation effect, 139 generally, 139–40 loan agreement in see under Loan agreement reason for including, 139 Non-assignment clause alternatives to, 117 conditions or restrictions, examples of imposition, 115 confidentiality clauses, use 122–3

consumer contracts, in, 121 contract, relevance, 115 contracts between the same parties, 129–30 damages, liability in, 123 desirability of override, 117 drawbacks, 117, 118, 119 empirical studies— generally, 118–19 Government’s response to, 119–120 foreign jurisdictions, in, 137 generally, 113 interests in land, contracts involving, 121 legal effect, 116 legislation in other jurisdictions, 113–14 limits on override, 115 negative pledge clauses, 130 prevalence, 113 problems, whether causing, 118 receivable— avoided, 117 meaning, 129, 130 reform of legislation— background, 119–20 complexity, 116 criticism of the process, 136–8 muddled nature, 449–51 small businesses, effect on, 126–7, 134–6 set-off, in case of, 119, 125–6 small businesses, effect of legislation on, 126–7, 134–6 special interest exceptions— background to, 127–8 contracts for differences, 128 electricity supply contracts, 128 national security, 128–9 petroleum licences, 128 supply chain finance, in case of, 118, 121–2 waiver, where, 118 Non-disclosure agreement background to development, 316 compensation payable under, 355, 359 concerns about, reasons for ignoring, 359 confidentiality clauses, use, 353 contract law, and— consideration, relevance, 321, 334 employment claims, settlement, 321 general contractual principles, 321–37 partial consideration, drafting to avoid preservation, 334 regulation of specific terms, 320 settlement agreements, restrictions in, 320

466  Index vitiating elements— contrary to public policy see contrary to public policy below duress, 323–5 illegality, 325–7 misrepresentation, 321 mistake, 321 penalties, 335–7 unconscionability, 322 undue influence, 322–5 weak protective measures, 337 whistle-blowing. 320 contrary to public policy— administration of justice, prejudicial to, 328 Brexit preparations, 333–4 bullying, 332 freedom of contract, and, 329 generally, 327–35 government contracts, 333–4 government targets, failure to meet, 331–2 human rights, contravention, 328, 329–30 illegality, 328 iniquitous behaviour, 328 judicial reluctance, 329 legislation, 328, 329, 330, 332, 333 loss of monetary advantage, fear of losing, 334 Ministerial Code, use, 333 NHS, in, 331–3 power and wealth, use— generally, 331, 339–40 see also professional ethics, and below racist behaviour, 330, 331 restitution not necessarily following, 335 sexist harassment, 330 variable nature of public policy, 328 whistle-blowing, 328, 330, 332, 333, 334 ethical approach to see professional ethics, and below exceptions to, examples, 355 financial pressures on claimant, 360 functions, 316, 317 generally, 315–16, 337–8, 339–40 high-profile cases, 315 legislation— generally, 320, 327, 328, 329, 330, 332, 333 lack of, 337 non-disparagement clause, and, 320 over-par offers, costs consequences, 360

penalties in— effect, 335–7, 356 enforceable, whether, 356 example, 356 perverting the course of justice, whether— aims of agreement, 345 dishonesty, 345 distancing tactics by lawyers, 345 generally, 344–6, 358 intention, 346 lawful threat, 346 lawyers straying from professional principles, 346 offence, 344 potentially legitimate strategy, 346 requirements, 345 scope, 345 serious aggravating features, 346 tendency to pervert: meaning, 344 test, 345 wrongful interference with witness, 346 Zelda Perkins case, and, 344–6 positive act, as, 344 professional ethics, and— balancing exercise, 350–1 Bar’s Code of Conduct, 347 benefit of uncertainty, reasons for giving, 350 case studies— Philip Green etc, 352–8, 360–1 Weinstein-Perkins Agreement, 341–4, 360–1 Zelda Perkins, testimony of, 340–341 cheating by lawyers, 340 civil claim, detrimental effect on, 343 clients’ instructions, whether bound to accept, 350 common law, and, 351–2 competing interests, proportionality test, 354 copy of agreement, refusing to hand over, 342, 343 costs arguments, 358–60 criminal complaint, prohibition on making, 342 criminal process, effect on participation in, 344 discreditable conduct, accusations of, 352 financial pressures on claimant, 360 freedom of contract, and, 351–2 generally, 339–40, 342 human rights, and, 353

Index  467 independence, acting with, 350 integrity, need to act with, 346, 347, 348 interpretation of clause, 350 judgment, need for, 351 lawful process, restrictions on cooperating with, 343 lawyers’ approach, 347, 350–1 legal proceedings, disclosure limited to, 342 limited form of disclosure insisted on, 342 medical practitioner forced to sign agreement, 342 onerous but unenforceable clause, 348 perspectives, 346–9 perverting the course of justice see perverting the course of justice, whether above police, effect of restriction on disclosure to, 343, 344, 345 potentially justifiable clause, 350 professional minimalism, 350 represented opponent, in case of, 349 responsibility, lawyer’s need to act with, 350–1 sexual abuse, 339 Solicitors’ Code of Conduct, and, 346–8 Solicitors Regulation Authority guidance, 348, 349, 361 sufficiently likely to prevent publication: meaning, 353 unenforceability, problems identifying, 347 unenforceable clause, use, 348, 349 unfair advantage, taking, 348–9 Weinstein case, 348–9 witness, problems in calling party as, 343 written consent to make disclosures, 342 public interest— disclosure in, 320 private rights, compromised by, 354 purpose, 315–16 use and abuse of— Brexit, over, 317–18 charities, effect on, 319 common use, 317 contract law to prevent abuse see contract law, and above generally, 316–19, 337–8 increased use, 317 legislation to prevent abuse, 319, 321 NHS, in, 317 private and public sectors, in, 318

public interests, concealment adverse to, 319 suppression of information as cause for concern, 317 warranties in settlement agreements, 320, 356, 357–8 widespread practice of accepting, 359 Non-oral modification clause see Anti-oral variation clause Non-reliance clause agent— insurance broker as, 243 integration into firm, 243 boilerplate, as, 249–52 common law regulation, 255–8 common theme, 242 compromise, drawbacks, 240 contextual relationism, 239 contra proferentem construction, 241, 255–8 courts’ involvement, 240 documentary fundamentalism, preference for, 266 due diligence, need for, 245 enforcement— arguments as to, 241, 252–5 courts’ powers, 266 renegotiation, courts’ powers on, 266 exclusion clause, as, 260 fraud and public policy— fraus omnia corrumpit, 248 generally, 246–9 intra-organisation control, and, 243 generally, 239–42, 266 growing use of, 245–6 liability— exclusion, 244 generally, 259–60 strict, 248 misrepresentation— onerous statutory reforms, 245 oral, protection from, 244–5 rescission for, 245 stringent approach of English law, 245 narrow textual approach, consequences, 241 objections to— boilerplate, clause as, 249–52 fraud and public policy, 246–9 logically unsustainable. 252–5 party autonomy, importance, 240, 241 pluralist solution, 240 protection of parties, approach to, 244 purpose and use, 242–6

468  Index reasons for use, 245, 246 relational contract scholarship, and, 239 relational fundamentalism, 240 statement made in negotiation, protection from, 243–4 statutory regulation— generally, 241–2, 258–9 legislative reform, proposals for, 264–6 reasonableness test, 262–4 scope— compromise over, 260–2 confining, 266 generally, 259 liability, problems defining etc, 259–260 maximal, 260 minimal, 259 textual formalism, 239 upholding, arguments as to, 241 widespread use, 242 Party autonomy in Private International Law benefits of giving effect to, 374 change in approach to, 363–5 choice of forum, effect, 364, 366 choice of law agreement, and, 366 choice of law and incorporation by reference, distinction between, 378–80 contractual party autonomy— consequences of distinguishing, 369–80 distinguished from, 366– relationship with, 367–9 development of, justification for, 364 deontological libertarianism, 372–3, 375 freedom of contract, as implication of, 364–5 generally, 363–5, 375–6, 380 justifications for, 369–76 limitations on, 376–7 litigation risks, costing and reducing, 374 meaning, 363, 365, 366 objection to, 363–4 party expectations, 371–2 private consequentialist libertarianism, 373–4, 37 public consequentialist libertarianism, 374–5 rejection of concept, 363, 367 regulatory competition, leading to, 375 rule of customary law, as, 367 unenforceable choices of court and law, 371 wide acceptance, 363 see also Freedom of contract

Penalty disproportionate see Disproportionate penalty Power to contract equal freedom principle, 17 generally, 13–14 lack of restrictions, 13 privity see Privity restrictions, 22 rights and recourse, 22–4 Privity common law, and, 22 doctrine of, whether anomalous, 21 generally, 21–2 rights, effect, 21 statutory protection, 22 Promise agreement, and see Agreement consideration, and, 26, 29, 31 contract as, 18 contract not constituting, 15 further promises— whether further benefits provided by, 30–1 whether need for consideration, 30–1 generally, 15–17 gratuitous, drawbacks, 26 morality of, 18 non est factum doctrine, and, 19–20 objectivity, 18–21 oral exchange of, difficulties in proving, 26 performance of, placing value on, 31 personal virtue, 19 promissory theory, contract’s divergence from, 19 responsibility for, need for, 19 service promisor already obliged to perform, 31 trust, giving rise to, 16 unilateral, insufficiency, 18 vow, as, 16 Promissory estoppel cause of action, prohibition on using as, 44 defence, as, 44 elements grounding, 44 fact-specific nature of doctrine, 44 judicial approach, 44–5 Proper purpose rule see Fraud on a power Receivables ban of assignment of see Non-assignment clause

Index  469 Relational contract characteristics, 97–8 good faith, implied duty of see Good faith (relational contract, implied duty in) identifying, 82–3, 97–9 test for, 99 Representation restriction on authority to make, 47–9 Reputation importance, 443 maintenance of, as inducement to perform agreement, 14, 443 Restitution availability, 271 bad bargain, avoidance, 273 boundary between contract and, 272–5 contractual gaps, effect, 273 counter-restitution, valuation, 289 English law, reasons for choosing, 270 excluded, where, 272 implied terms giving effect to, 275 inappropriate, where, 271 incompatibility criterion, implications, 273 money paid under a binding contract, 274–5 multi-party transactions, 275–6 overpayments, 274 part payments received, where, 289 partial failure, where, 288 performance of contract, effect, 272 primacy of contractual bargain, 271–2 risk, implicit allocation, 274 scope of contract, importance, 272 silence in contract, relevance, 274 termination allowing for, 272–3 valid contract as basis of, 270–1 See also Consideration (failure of, and total failure of) and Unjust enrichment Self-interested behaviour dishonesty see under honest performance, exclusion of duty of below exclusion of terms restricting— courts’ powers, restrictions on, 99 courts’ reluctance to exclude, 100–1 generally, 99–101 honest performance see honest performance, exclusion of duty of below implied terms, 100–1 reasons for, 100 targeted exclusion, 101

express restrictions on— barrier to enforcement, 90–1 construction of good faith term, 91 disputes to be settled by ‘friendly discussion’, 91 enforceable by court, 90 fact, implied in, 90 generally, 90–2 interpretation, approach to, 93 judicial analysis, 90 law, implied in, 90 right to include, 90 uncertainty as to, effect, 90–1 freedom of contract, effect, 65–6, 90 good faith, and, 68–70, 72–3, 75, 84, 87 honest performance, exclusion of duty— Canada, in, 103–4 casino, potential flaws in system, exploitation of, 102–3 dishonesty— interpreting meaning, 102 jury question, as, 102 objective element in, 102 spirit of mistrust, agreement to proceed in, 103 troubled concept, 101–2 generally, 101–4 justification, 101 objections, 101 spirit of mistrust, agreement to proceed in, 103 unconscionability as general doctrine, 104 honesty— dispensing with see honest performance, exclusion of duty of above exercise of contractual powers see Honesty (exercise of contractual powers, and) implied restrictions on— broad interpretative approach, need for, 92 burden of proof, 94 court’s reluctance to infer, 92, 94 examples where implied, 95 fact, implied in, 93–4 generally, 92–9 necessary, whether, 93, 94 ‘officious bystander’ test, 93 relational contracts see under Good faith silence, inference to be drawn from, 93 tailoring term, 94 obligations, control of extent, 90 parties’ right to act selfishly, 65, 89 restricted by contract, 66, 89, 90

470  Index Smart contract automated execution, effect, 191 code— actions, as, 203 agreement ‘subject to contract’, where, 200 arbitration agreement, comparison with, 201 common lexicon, lack of, 206 content, effect, 204 contract, as, 199–202, 203, 204 intention to create legal relations, whether, 199–200 not interpreted in isolation, 198 performative utterances, 203 platform, need for, 203–4 policy-based examples, comparison with, 201 Ricardian contract, 199–200, 201 common law, need for adaptation, 209 contract law— ineffectiveness, 191 whether having part to play, 202 contractual estoppel, and, 201 conventional contract, differing from, 196 decentralisation and distributed consensus— Bitcoin see under Blockchain blockchain see Blockchain decentralisation: meaning, 192 Distributed Ledger Technology, use of see Distributed Ledger Technology Ethereum, 192 nature, 192 disputes over, adjudication, 191 Distributed Ledger Technology, use of see Distributed Ledger Technology end-to-end residential property transaction, trial of, 196 formation, 197–8 fraud, effect, 202 generally, 191, 208–9 interpretation— code— actions, as, 203 performative utterances, as, 203 qualitatively different from human language, 205 coders, use, 206 combination of words used, relevance, 207 common lexicon, lack of, 206 complex forensic process, as, 204 conduct rather than words, contract as, 203

difficulties, 202 expert translation evidence, and, 205, 206, 207, 208 generally, 202–8 judicial, relevance, 204 lack of room for, 204 matters ignored as part of, 203 meaning, 202–3 objective approach, and, 206 performative utterances, code as, 203 precision in computer’s response, 207 technical problems, 205 legal supervision, avoidance, 199 machine logic, use, 196 meaning, 191–2, 197 negotiating powers, imbalance in, 202 novation, use, 209 performance— increased likelihood of, 191 time for, 202 purpose, 191 rectification, problems as to, 208–9 Ricardian contract, 198, 199–200, 201, 203 spectrum, 197–8 time between formation and performance, 202 transfer of title to assets, and, 196 trustless nature of, 192 types, 198 useful, where, 196 Termination advance payments— generally, 279–80 return, on failure of consideration, 280–1 breach, right to damages for, 277 clauses, 277–8, 291 consequences, contract setting out, 277 generally, 277 instalments due prior to, claim for, 277 performance obligations, structuring— advance payments, 279–80 deposits, 279–80 entire obligation, 278–9 timing, 279 prospective application, 277 restitution, right to, 277 right accruing prior to, 277 valid contract, need for, 277

Index  471 Terms affecting freedoms see Contract Trade receivables ban of assignment of see Non-assignment clause Unfairness acceptability of, 89 Unilateral contract enforcement difficulties, 27–8 mutual nature, 18 Unjust enrichment defence to, 271 distinct from enforcing contract, 283 English law, reasons for choosing, 270 generally, 283 invalid contract, 287 multi-party transactions, in case of, 275–6 proof of, 283 restitution see Restitution

Rome I Regulation, and, 270 Rome II Regulation, and, 270 valuing the enrichment— arm’s-length bargain, in case of, 283 contract price— cap on valuation, as, 284–5 evidence of value, as, 283 devaluation of goods or services, 284 qualifications to using bargain as evidence of value, 284 Variation anti-oral clause see Anti-oral variation clause consideration, whether further need for, 30–1, 33 Warranty restriction on authority to make, 47–9

472