Liability Insurance in International Arbitration: The Bermuda Form 9781509917259, 9781509917280, 9781509917266

The Bermuda Form This is the third revised edition of what was described by the English Court of Appeal in C v D as the

345 110 5MB

English Pages [483] Year 2021

Polecaj historie

Property and liability insurance entities 9781119699095, 1119699096

Nature, conduct, and regulation of the business -- Audit considerations -- Premiums -- The loss reserving and claims cyc

671 110 12MB Read more

Cross Examination in International Arbitration 0199681236, 9780199681235

A practical, self-teaching guide to effective cross-examination in international arbitration. Offers an introductory or

472 59 2MB Read more

Ensuring Corporate Misconduct: How Liability Insurance Undermines Shareholder Litigation 9780226035079

Shareholder litigation and class action suits play a key role in protecting investors and regulating big businesses. But

182 52 2MB Read more

International Cargo Insurance 9781843119470

844 59 12MB Read more

State Liability in Investment Treaty Arbitration: Global Constitutional and Administrative Law in the BIT Generation 9781472564948, 9781841138565

Today there are more than 2,500 bilateral investment treaties (BITs) around the world. Most of these investment protecti

198 9 3MB Read more

Public Health in International Investment Law and Arbitration 9781138025233, 9780415507493

Is a State free to adopt measures to protect the public health of its citizens? If so, what are the limits, if any, to s

765 76 2MB Read more

Ethics in International Arbitration [1 ed.] 0195337697, 9780195337693

International arbitration is a remarkably resilient institution, but many unresolved and largely unacknowledged ethical

894 86 6MB Read more

International Arbitration in Tax Matters [1 ed.] 9789087223427, 9789087223410

The last two decades have seen an unprecedented integration of national economies, the emergence of new players and the

171 30 967KB Read more

Fodor's Bermuda 1988 0679014721, 0340417870

228 39 79MB Read more

Attribution in International Law and Arbitration [1 ed.] 2019949145

Attribution in International Law and Arbitration clarifies and critically discusses the international rules of attributi

1,344 91 2MB Read more

Liability Insurance in International Arbitration: The Bermuda Form
9781509917259, 9781509917280, 9781509917266

Author / Uploaded
Richard Jacobs
Lorelie S Masters
Paul Stanley

Table of contents :
Foreword to the First Edition
Preface
Contents
About the Authors
Table of Cases
Table of Legislation
1. THE LEGAL AND ECONOMIC ORIGINS OF THE BERMUDA FORM
Policy Forms and Liability Problems
US Legal Decisions on Insurance Coverage Issues
The Creation of ACE and XL
Key Features of the Bermuda Form
2. THE BERMUDA FORM: ITS BASIC STRUCTURE
An Occurrence Reported Form
The Period of Cover: Coverage A and B
Limits
The Excess Point
The Inception and Retroactive Coverage Dates
The Exclusions
The Conditions
Schedules and Endorsements
3. CHOICE OF LAW ISSUES UNDER THE BERMUDA FORM
Identifying the Relevant Governing Law
The Division between Substance and Procedure
4. INTERPRETATION OF THE BERMUDA FORM AND THE MODIFICATION OF NEW YORK LAW
The Structure of the Modification of the New York Law Proviso
Interpretation of Insurance Contracts: The Basic Approach
Interpretation of the Modification of the New York Law Proviso
The ‘Forbidden Grounds’
Fruit of the Poisonous Tree?
5. THE COVERAGE CLAUSE
Insurance against Liability
The Insured
Ultimate Net Loss/All Sums Insurance … For Damages
Judgments
Settlements by the Policyholder
Mitigation of Damages
Allocation of Payments
Personal Injury, Property Damage and Advertising Liability
Personal Injury
Property Damage
Advertising Liability
Other Aspects of the Coverage Clause
6. THE DEFINITION OF ‘OCCURRENCE’
Introduction
‘Occurrence’: The Basic Definition
Aggregation
Expected or Intended Injury
7. FORTUITY, EXPECTED OR INTENDED, AND THE ‘MAINTENANCE DEDUCTIBLE’
Introduction
Fortuity and Related Doctrines
Expectation and Intention
The ‘Maintenance Deductible’
8. NOTICE OF OCCURRENCE
Introduction
Method of Giving Notice
Notice as a Condition of Coverage
Notice as a Trigger of Coverage: Legal and Practical Considerations
9. ARTICLE II: THE EXCESS POINT AND LIMITS OF LIABILITY
The Excess Point
The Other Insurance Condition
The Policy Limits
Joint Ventures, Partnerships and Minority Interests
10. THE EXCLUSIONS
Introduction
Interpretation of Exclusion Clauses: General Principles and the Effect of the Bermuda Form’s Modification of New York Law
Causation
The Specific Exclusions in the Bermuda Form: Introduction
The ‘Prior to Inception or Retroactive Coverage Date’ and ‘Other Insurance’ Exclusion
The ‘Workers’ Compensation, etc’ Exclusion
The ‘Professional Services’ Exclusion
The ‘Owned Property; Care, Custody or Control, etc’ Exclusion
The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)
The ‘Advertising’ Exclusion
The ‘War’ Exclusion
The ‘Toxic Substances’ Exclusion
The ‘Aircraft’ Exclusion
The ‘Watercraft’ Exclusion
The ‘Pollution’ Exclusion
The ‘Nuclear’ and the ‘Radioactive Contamination (Outside the United States)’ Exclusions
The ‘Erisa’ Exclusion
The ‘Repetitive Stress’ Exclusion
The ‘Securities, Antitrust, etc’ Exclusion
11. THE CONDITIONS
The Premium Condition
The Inspection Condition
The Cross-liability Condition
The Notice of Occurrence Condition
The Assistance and Co-operation Condition
The Appeals Condition
The Loss Payable Condition
The Representation Condition
The Other Insurance Condition
The Subrogation Condition
The Changes Condition
The Assignment Condition
The Cancellation Condition
The Currency Condition
The Arbitration Condition
The Conflicting Statutes Condition
The Law of Construction and Interpretation Condition
The Proration of Losses Condition
The Liability of the Company Condition
The Policy Extension Condition
The Reinstatement Condition
The Discovery Period Condition
The Expiration Date Condition
The Former Subsidiaries, Affiliates and Associated Companies Condition
The Notice Condition
The Headings Condition
12. MISREPRESENTATION AND NON-DISCLOSURE
Introduction
Non-Disclosure under New York Law
Misrepresentation under New York Law
Remedy for Misrepresentation
13. WAIVER AND ESTOPPEL AND RESERVATIONS OF RIGHTS
Introduction
Waiver
Estoppel
The ‘Changes’ Condition in the Bermuda Form
14. COMMENCING A BERMUDA FORM ARBITRATION AND APPOINTING ATTORNEYS AND ARBITRATORS
Introduction
Legal Representation
The Commencement of Arbitration
The Selection of an Arbitrator
Appendix: Notice to Commence Arbitration
15. THE COURSE AND CONDUCT OF A BERMUDA FORM ARBITRATION IN LONDON
Introduction
The Overall Shape of the Arbitration
The ‘Pleadings’ Stage
The First Order for Directions
Confidentiality
Discovery of Documents
Preliminary Issues or ‘Bifurcation’
Witness Statements
Expert Evidence
Preparation for the Substantive Hearing
The Substantive Hearing
The Award and Post-Award Events
Appendix 1: First Order for Directions
Appendix 2: Protective Order
16. DISCOVERY, PRIVILEGE AND WAIVER OF PRIVILEGE
General Principles
Discovery in Arbitration in Practice
Legal Professional Privilege
Waiver of Privilege
Confidential Documents
Other Forms of Discovery
17. INTEREST AND COSTS
Interest
Costs
Appendix
English and New York Law: Sources and Legal Approach
English Law for New York (And Other United States) Lawyers
New York Law for English Lawyers
Cultural Differences
Bibliography
Index

Citation preview

Liability Insurance in International Arbitration The Bermuda Form This is the third revised edition of what was described by the English Court of Appeal in C v D as the ‘standard work’ on Bermuda Form excess insurance policies. The Form, first used in the 1980s, covers liabilities for catastrophes such as serious explosions or mass tort litigation and is now widely used by insurance companies. It is unusual in that it includes a clause requiring disputes to be arbitrated under English procedural rules in London but subject to New York substantive law. This calls for a rare mix of knowledge and experience on the part of the lawyers involved, each of whom is required to confront the many differences between English and US law and legal culture. In addition, since the awards of arbitrators are confidential and are not subject to the scrutiny of the courts, the book helps professionals understand the Form’s lengthy and complex provisions. The book, first published in 2004, was the first comprehensive analysis of the Bermuda Form. It is frequently cited in Bermuda Form arbitrations and was the joint winner in 2012 of British Insurance Law Association Book Prize for the most notable contribution to literature in the field of law as it affects insurance. It offers a detailed commentary on how the Form is to be construed, its coverage, the substantive law to be applied, the limits of liability, exceptions, and, of course, the procedures to be followed during arbitration proceedings in London. The book will prove invaluable to lawyers, risk managers, and executives of companies which purchase insurance on the Bermuda Form, and to clients, lawyers or arbitrators involved in disputes arising therefrom.

ii

Liability Insurance in International Arbitration The Bermuda Form Third Edition

Sir Richard Jacobs

Judge of the Queen’s Bench Division of the High Court, England and Wales

Lorelie S Masters

Hunton Andrews Kurth LLP, Washington DC

and

Paul Stanley QC

Essex Court Chambers, London

HART PUBLISHING Bloomsbury Publishing Plc Kemp House, Chawley Park, Cumnor Hill, Oxford, OX2 9PH, UK 1385 Broadway, New York, NY 10018, USA 29 Earlsfort Terrace, Dublin 2, Ireland HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2021 1st edition 2004, 2nd edition 2011 Copyright © Sir Richard Jacobs, Lorelie S Masters and Paul Stanley QC, 2021 Sir Richard Jacobs, Lorelie S Masters and Paul Stanley QC 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/, 1998–2021. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication data Names: Jacobs, Richard (Judge) author. | Masters, Lorelie S, author. | Stanley, Paul, 1970- author. Title: Liability insurance in international arbitration : the Bermuda Form / Richard Jacobs QC, Lorelie S Masters, Paul Stanley QC. Description: Third edition. | Oxford ; New York : Hart, 2021. | Includes bibliographical references and index. Identifiers: LCCN 2021042190 (print) | LCCN 2021042191 (ebook) | ISBN 9781509917259 (hardback) | ISBN 9781509917266 (pdf) | ISBN 9781509917273 (Epub) Subjects: LCSH: Liability insurance—Law and legislation—United States. | Insurance claims—United States. | Liability insurance—Law and legislation—Great Britain. | Insurance claims—Great Britain. | Arbitration commercial arbitration. | Insurance companies—Law and legislation—Bermuda Islands. Classification: LCC K1271 .J33 2021 (print) | LCC K1271 (ebook) | DDC 346/.0865—dc23/eng/20211004 LC record available at https://lccn.loc.gov/2021042190 LC ebook record available at https://lccn.loc.gov/2021042191 ISBN: HB: 978-1-50991-725-9 ePDF: 978-1-50991-726-6 ePub: 978-1-50991-727-3 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. Any third party copyrighted material displayed in the pages of this book is done so on the basis of ‘fair dealing for the purposes of criticism and review’ or ‘fair use for the purposes of teaching, criticism, scholarship or research’ only in accordance with international copyright laws, and is not intended to infringe upon the ownership rights of the original owners.

In memory of Johnny Veeder QC 1948–2020

vi

Foreword to the First Edition VV VEEDER QC

Thirty years ago, one of England’s most eminent arbitration practitioners asserted that international commercial arbitrations in London only rarely raised issues on the conflicts of laws; and rightly so; such an arbitration, almost invariably, was then conducted before English arbitrators by English counsel instructed by English solicitors arguing over an English standard form of contract governed by English substantive law and English procedural law, with only the parties’ foreign names signifying the transnational character of the disputed transaction. This is no longer so. The changes came even before the English Arbitration Act 1996; and in the field of insurance, they included the early London arbitrations on the Bermuda Form. At first, not a few English lawyers and arbitrators were puzzled by the inter-locking logic of its different provisions and the split between English procedural law and New York substantive law, the latter subject to the famous proviso subject itself (it was thought) to English substantive law under English conflict rules. From the beginning, it proved impossible to conduct such London arbitrations without the active participation of US lawyers and arbitrators, who were equally puzzled at the cultural idiosyncrasies of English arbitral procedure: the practice (soon abandoned) of adducing expert witness evidence on New York law as opposed to arguing New York law in submission, the black-letter syntactical approach to contractual wording, the readingout of law reports and the common tea-breaks where, by ancient tradition (but also soon abandoned), the chocolate-covered biscuits were reserved for the arbitrators. Eventually, the two legal tribes acquired a better understanding of each other; but sadly, a better public understanding of the Bermuda Form remained elusive. That was not only due to the fact that for every new case, that complicated wording is to be read, again, for the first time; but also because English arbitrations are private and English arbitral awards under English law remain confidential. There is an article by a well-known insurance lawyer from Washington DC, and later there was a successful one-day London conference on the procedural aspects of the Bermuda Form; but until now there was never the fuller work undertaken by these three distinguished authors. They represent the symbiotic response of English and US legal specialists which the Bermuda Form requires for every London arbitration; and in their comprehensive coverage of both substance and procedure, they have performed a great public service to the many users of the Bermuda Form and their legal advisers everywhere. Not all will agree with their analysis or conclusions; but it is nevertheless high time that any dispute should begin with a better knowledge of the issues raised by the Bermuda Form’s successive wordings under English and New York law. The Bermuda Form remains a complex document. For the objective lawyer and impartial arbitrator, it is good to come to any new dispute with an open mind,

viii

Foreword to the First Edition

susceptible to new arguments and new approaches, but it is not good to come with an empty head, possibly clouded by samizdat rumours of different arguments and awards in other cases. It is time the Bermuda Form came out from the dark, even if, unlike Peter Pan, the Bermuda Form can never quite lose its shadow. With this most useful work, the authors have begun that most important and necessary process. London May 2004

Preface A certain measure of dull predictability is a virtue in a standard form in common use. When we wrote the first edition of this book nearly 20 years ago, the Bermuda Form still had a slightly rebellious, youthful quality about it: a specialist form, the preserve of specialist insurers which set out to challenge existing markets and practices. Tempus fugit. The prickly teenager is now a respectable citizen. Its originators are deeply embedded in the market they set out to challenge. The Form, or variations of it, turns up constantly. Aspects of it surface from time to time in court in both the United Kingdom and the United States. It has even generated a significant decision, relating to arbitrator impartiality, in the UK Supreme Court. It has proved itself, over the years, both adaptable and recalcitrant. Adaptable to successive waves of product-liability litigation. Recalcitrant in the sense that some of its central mysteries (for example, the expected-or-intended provisions, the ‘maintenance deductible’, the aggregation of post-termination injuries) continue to be controversial in concrete cases. Historians of risk management will no doubt consider that the decades from 1965 to 1995 stand in stark contrast to those that have followed. In this respect, the past really is another country. The way that the industrial enterprises of the 1950s and 1960s assessed and approached risks from things such as asbestos, pollution or new drugs seems astonishingly amateurish and thoughtless by today’s standards; the way that insurers, in their boxes at Lloyd’s or their offices in Hartford and elsewhere, handled underwriting equally so. Litigation was different too: more tractable in some respects (the typewriter and the pencilled note left for disclosure so much less than the email and the cloud), but more free-wheeling and innovative in others. In 1985 it was only a long professional lifetime since English law abandoned a Victorian capitalist conviction that a manufacturer owed a duty only to those who purchased directly from it, or that the American courts – slightly ahead of England – developed and advanced principles of strict liability. The succeeding decades look, by contrast, relatively stable. Writing this edition has been largely an exercise in incremental change, in pruning those discussions which can now (as times and wordings change) be regarded as only of historical interest, and adaptation to address those problems that now seem more prominent or where our views have changed. We are grateful to a number of practitioners who kindly reviewed chapters or provided ideas for issues to be considered: Professor Tom Baker, Peter Halprin, Rebecca Jacobs, Richard P Lewis, Jeremy Lawrence, Cary Lerman, Tom Newman, Scott Oostdyk and Jakob Reckhenrich. The previous edition of this book was awarded the British Insurance Law Association book prize jointly with the excellent work co-authored by David Scorey QC, with whom we have had many fruitful discussions over the years. We also thank our other colleagues, too numerous to mention, with whom we have taken the opportunity to discuss issues. Many people in London and Washington DC helped with

x

Preface

the laborious work of proofreading the manuscript and checking citations from sources on both sides of the Atlantic: Ava M Dixon, Alice Duddridge, David L Gunn, Stephanie E Meharg, Cynthia Smoot and Lori Wester. Our copy-editor, Jon Lloyd, read every word with immense care, identifying points which we needed to reconsider. Dr Roberta Bassi at Hart Publishing successfully, and with considerable patience, ensured that we completed this edition. We remain very grateful to Richard Hart who, in 2003, had no doubt that our proposed book deserved publication. Our partners (Pamela, Jack and Daniel) have continued to provide support and encouragement. So have our children (Rebecca, Benjamin, Hannah, Ian, David, Benedict and Zachary), who now (as adults) have a greater understanding of the Bermuda Form and why we were interested in writing about it. Tempus fugit and mors omnibus instat. Johnny Veeder QC sadly died while this edition was being prepared. He was a giant in the world of international arbitration, making immense contributions both on a practical and an intellectual level. He was one of the first people in the world to handle a dispute under the Bermuda Form in the early 1990s, and dealt with many disputes under the Form thereafter. He gave enormous help and encouragement when we wrote the first edition, including trying (not always successfully) to stop us expressing strong views on issues which were more difficult than we appreciated. He was the most generous colleague anyone could have wished to have, and he is missed by many around the world whose work was improved, and life enriched, by knowing and being able to talk to him. We are delighted to reprint his foreword to the first edition. He probably still would not, as he delicately put it, agree with all our analysis and conclusions, but he would be very pleased that they are available to whoever wishes to read them. This book is dedicated to him. Richard Jacobs Lorelie S Masters Paul Stanley London and Washington DC 1 November 2021

Contents Foreword to the First Edition�� vii Preface�� ix About the Authors�� xv Table of Cases�� xvii Table of Legislation��li 1. THE LEGAL AND ECONOMIC ORIGINS OF THE BERMUDA FORM�� 1 Policy Forms and Liability Problems�� 1 US Legal Decisions on Insurance Coverage Issues�� 4 The Creation of ACE and XL�� 11 Key Features of the Bermuda Form�� 13 2. THE BERMUDA FORM: ITS BASIC STRUCTURE�� 20 An Occurrence Reported Form�� 20 The Period of Cover: Coverage A and B�� 22 Limits�� 24 The Excess Point�� 24 The Inception and Retroactive Coverage Dates�� 25 The Exclusions�� 25 The Conditions�� 26 Schedules and Endorsements�� 26 3. CHOICE OF LAW ISSUES UNDER THE BERMUDA FORM�� 27 Identifying the Relevant Governing Law�� 28 The Division between Substance and Procedure�� 40 4. INTERPRETATION OF THE BERMUDA FORM AND THE MODIFICATION OF NEW YORK LAW�� 46 The Structure of the Modification of the New York Law Proviso�� 47 Interpretation of Insurance Contracts: The Basic Approach�� 49 Interpretation of the Modification of the New York Law Proviso�� 52 The ‘Forbidden Grounds’�� 54 Fruit of the Poisonous Tree?�� 58 5. THE COVERAGE CLAUSE�� 61 Insurance against Liability�� 61 The Insured�� 63 Ultimate Net Loss/All Sums Insurance … For Damages�� 64 Judgments�� 67 Settlements by the Policyholder�� 70

xii

Contents

Mitigation of Damages�� 78 Allocation of Payments�� 78 Personal Injury, Property Damage and Advertising Liability�� 84 Personal Injury�� 84 Property Damage�� 86 Advertising Liability�� 89 Other Aspects of the Coverage Clause�� 90 6. THE DEFINITION OF ‘OCCURRENCE’�� 91 Introduction�� 91 ‘Occurrence’: The Basic Definition�� 93 Aggregation�� 98 Expected or Intended Injury��106 7. FORTUITY, EXPECTED OR INTENDED, AND THE ‘MAINTENANCE DEDUCTIBLE’��107 Introduction��107 Fortuity and Related Doctrines��109 Expectation and Intention��111 The ‘Maintenance Deductible’��124 8. NOTICE OF OCCURRENCE��132 Introduction��132 Method of Giving Notice��134 Notice as a Condition of Coverage��135 Notice as a Trigger of Coverage: Legal and Practical Considerations��145 9. ARTICLE II: THE EXCESS POINT AND LIMITS OF LIABILITY��151 The Excess Point��151 The Other Insurance Condition��159 The Policy Limits��162 Joint Ventures, Partnerships and Minority Interests��163 10. THE EXCLUSIONS��166 Introduction��166 Interpretation of Exclusion Clauses: General Principles and the Effect of the Bermuda Form’s Modification of New York Law��166 Causation��169 The Specific Exclusions in the Bermuda Form: Introduction��177 The ‘Prior to Inception or Retroactive Coverage Date’ and ‘Other Insurance’ Exclusion��178 The ‘Workers’ Compensation, etc’ Exclusion��179 The ‘Professional Services’ Exclusion��179 The ‘Owned Property; Care, Custody or Control, etc’ Exclusion��180 The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)��188 The ‘Advertising’ Exclusion��200 The ‘War’ Exclusion��200 The ‘Toxic Substances’ Exclusion��205

Contents

xiii

The ‘Aircraft’ Exclusion��206 The ‘Watercraft’ Exclusion��207 The ‘Pollution’ Exclusion��207 The ‘Nuclear’ and the ‘Radioactive Contamination (Outside the United States)’ Exclusions��215 The ‘Erisa’ Exclusion��216 The ‘Repetitive Stress’ Exclusion��216 The ‘Securities, Antitrust, etc’ Exclusion��217 11. THE CONDITIONS��219 The Premium Condition��219 The Inspection Condition��220 The Cross-liability Condition��222 The Notice of Occurrence Condition��223 The Assistance and Co-operation Condition��223 The Appeals Condition��226 The Loss Payable Condition��227 The Representation Condition��238 The Other Insurance Condition��239 The Subrogation Condition��239 The Changes Condition��240 The Assignment Condition��241 The Cancellation Condition��244 The Currency Condition��245 The Arbitration Condition��246 The Conflicting Statutes Condition��249 The Law of Construction and Interpretation Condition��250 The Proration of Losses Condition��250 The Liability of the Company Condition��251 The Policy Extension Condition��251 The Reinstatement Condition��252 The Discovery Period Condition��254 The Expiration Date Condition��256 The Former Subsidiaries, Affiliates and Associated Companies Condition��256 The Notice Condition��257 The Headings Condition��258 12. MISREPRESENTATION AND NON-DISCLOSURE��259 Introduction��259 Non-Disclosure under New York Law��262 Misrepresentation under New York Law��263 Remedy for Misrepresentation��272 13. WAIVER AND ESTOPPEL AND RESERVATIONS OF RIGHTS��278 Introduction��278 Waiver��280 Estoppel��285 The ‘Changes’ Condition in the Bermuda Form��287

xiv

Contents

14. COMMENCING A BERMUDA FORM ARBITRATION AND APPOINTING ATTORNEYS AND ARBITRATORS��290 Introduction��290 Legal Representation��290 The Commencement of Arbitration��295 The Selection of an Arbitrator��302 Appendix: Notice to Commence Arbitration��309 15. THE COURSE AND CONDUCT OF A BERMUDA FORM ARBITRATION IN LONDON��311 Introduction��311 The Overall Shape of the Arbitration��311 The ‘Pleadings’ Stage��314 The First Order for Directions��317 Confidentiality��319 Discovery of Documents��320 Preliminary Issues or ‘Bifurcation’��321 Witness Statements��322 Expert Evidence��325 Preparation for the Substantive Hearing��328 The Substantive Hearing��330 The Award and Post-Award Events��333 Appendix 1: First Order for Directions��335 Appendix 2: Protective Order��340 16. DISCOVERY, PRIVILEGE AND WAIVER OF PRIVILEGE��342 General Principles��342 Discovery in Arbitration in Practice��347 Legal Professional Privilege��350 Waiver of Privilege��358 Confidential Documents��375 Other Forms of Discovery��375 17. INTEREST AND COSTS��382 Interest��382 Costs��391 Appendix��395 English and New York Law: Sources and Legal Approach��395 English Law for New York (And Other United States) Lawyers��395 New York Law for English Lawyers��399 Cultural Differences��402 Bibliography��405 Index��409

About the Authors Sir Richard Jacobs was educated at Cambridge University and was formerly a QC and member of Essex Court Chambers in London. He practised as a barrister between 1980 and 2018, specialising in commercial law, including insurance and arbitration law. He appeared as leading counsel in court and arbitration proceedings, and acted as arbitrator in various disputes including Bermuda Form arbitrations. In 2018, he was appointed a judge of the Queen’s Bench Division of the High Court, England & Wales, where he sits principally in the Commercial Court in London. Lorelie S Masters was educated at Georgetown University and Notre Dame Law School. She is a partner in Hunton Andrews Kurth LLP in Washington, DC, where she advises and represents policyholders in insurance coverage and litigation. She is co-author of a treatise entitled Insurance Coverage Litigation (Wolters Kluwer, 2nd edition, 1999 and Supps). Paul Stanley QC was educated at Cambridge University and Harvard Law School and is a barrister at Essex Court Chambers in London, specialising in commercial law, including insurance and arbitration law. He appears as leading counsel in court and arbitration proceedings and acts as arbitrator in various disputes including Bermuda Form arbitrations. He is the author of The Law of Confidentiality: A Restatement (Hart Publishing, 2008).

xvi

Table of Cases Australia Haynes v Hurst (1927) 27 NSWSR 480��13.15 R v Bonython (1984) 38 SASR 45��15.42 Canada Mutual Life Ins Co of New York v Ontario Metal Products Co Ltd [1925] LRAC 344��12.17 European Court of Human Rights Campbell v United Kingdom (1992) 15 EHRR 137��16.28 Foxley v United Kingdom (2000) 31 EHRR 637��16.28 Hong Kong First Laser Ltd v Fujian Enterprises (Holdings) Company Ltd [2012] HKCU 1397 (HKCFA)��3.34 New Zealand B v Auckland District Law Society [2003] UKPC 38, [2003] 2 AC 736��16.27, 16.90 United Kingdom A v B [2010] EWHC 1626 (Comm), [2010] 2 CLC 1, 132 ConLR 73��15.70 A v B (No 2) [2006] EWHC 2006 (Comm), [2007] 1 Lloyd’s Rep 237��3.17 A and B v C, D and E (Taking Evidence for a Foreign Seated Arbitration) [2020] EWCA Civ 409, [2020] 1 WLR 3504, [2020] 4 All ER 874; [2020] EWHC 258 (Comm), [2020] Bus LR 426�� 16.97, 16.100 Accident Exchange v McLean [2018] EWHC 23 (Comm), [2018] 4 WLR 26��16.67

xviii

Table of Cases

Ace Capital Ltd v CMS Energy Corporation [2008] EWHC 1843 (Comm), [2009] Lloyd’s Rep IR 414, [2008] 2 CLC 318��14.33 Al Wahab, The. See Amin Rasheed Shipping Corp v Kuwait Insurance Co Ali Shipping Corporation v Shipyard Trogir [1999] 1 WLR 314, [1998] 2 All ER 136 (CA)��3.34, 15.24 Amec Capital Projects Ltd v Whitefriars City Estates Ltd [2004] EWCA Civ 1418, [2005] 1 All ER 723, [2005] BLR 1��14.38 Amin Rasheed Shipping Corp v Kuwait Insurance Co; The Al Wahab [1984] AC 50, [1983] 3 WLR 241�� 3.20, 3.33 Anderson v Bank of British Columbia (1876) 2 Ch D 644��16.33, 16.39 Angelic Grace, The. See Angeliki Charis Compania Maritima SA v Pagnan SpA Angeliki Charis Compania Maritima SA v Pagnan SpA (The Angelic Grace) [1995] 1 Lloyd’s Rep 87��3.08 Aragon, The. See Segovia Compagnia Naviere SA v R Pagnan & Fratelli Asbestos Insurance Coverage Cases, Re; sub nom Sedgwick Group plc v Johns-Manville Fibreboard Corp [1985] 1 WLR 331, [1985] 1 All ER 716 (HL)�� 16.103 ASM Shipping Ltd of India v TTMI Ltd of England [2005] EWHC 2238 (Comm), [2006] 2 All ER (Comm) 122, [2006] 1 Lloyd’s Rep 375, [2005] ArbLR 5��14.38 Aspen Insurance UK Ltd v Pectel Ltd [2008] EWHC 2804 (Comm), [2009] Lloyd’s Rep IR 440, [2009] 2 All ER (Comm) 873��8.21 Assetco plc v Grant Thornton UK LLP [2019] EWHC 592 (Comm), [2019] 1 Costs LR 197��17.03 Assimina Maritime Ltd v Pakistan Shipping Corp (The Tasman Spirit) [2004] EWHC 3005 (Comm), [2005] 1 All ER (Comm) 460, [2005] 1 Lloyd’s Rep 525�� 16.103 Associated Electric and Gas Insurance Services Ltd v European Reinsurance Co of Zurich [2003] UKPC 11, [2003] 1 WLR 1041��3.34, 3.35, 15.24, 15.26 AstraZeneca Insurance Co Ltd v XL Insurance (Bermuda) Ltd & ACE Bermuda Insurance Ltd [2013] EWCA Civ 1660, [2014] 2 All ER (Comm) 55, [2014] Lloyd’s Rep IR 509�� 1.19, 3.03, 3.30, 5.29 Atlas Power Ltd v National Transmission and Despatch Co Ltd [2018] EWHC 1052 (Comm), [2019] 1 All ER (Comm) 931, [2018] 2 Lloyd’s Rep 113��11.85 Axa Reinsurance (UK) Ltd v Field [1996] 1 WLR 1026,[1996] 3 All ER 517 (HL)��6.33 B v Auckland District Law Society [2003] UKPC 38, [2003] 2 AC 736��16.27, 16.90 Balabel v Air India [1988] 1 Ch 317, [1988] 2 WLR 1036 (CA)�� 16.34, 16.35, 16.36 Baltic Universal, The. See Structural Polymer Systems v Brown Barings PLC v Coopers & Lybrand [2000] EWCA Civ 148, [2000] 1 WLR 2353, [2000] 3 All ER 910��16.90 Barings plc & ors (No 5), Re; Secretary of State for Trade and Industry v Baker [1999] 1 BCLC 433 (ChD)��15.42 Baschet v London Illustrated Standard Co [1900] 1 Ch 73 (ChD)��3.40 Benfield Holdings Ltd v Richardson [2007] EWHC 171 (QB), [2007] 2 WLUK 521�� 16.101

Table of Cases

xix

Benhurst Finance Ltd v Colliac [2018] EWHC 2188 (QB), [2018] 6 WLUK 641��16.98 Berezovsky v Abramovich [2011] EWHC 1143 (Comm), [2011] 5 WLUK 145��16.86 BNP Paribas v Deloitte and Touche LLP [2003] EWHC 2874 (Comm), [2004] 1 Lloyd’s Rep 233�� 16.103 Bofors (AB) v AB Skandia Transport [1982] 1 Lloyd’s Rep 410, [1982] Com LR 63 (QBD Comm Ct)��3.33 Bonacina, Re [1912] 2 Ch 394 (CA)��3.21 Bourns Inc v Raychem Corp [1999] 3 All ER 154, [1999] CLC 1029 (CA)��16.25, 16.91 Boys v Chaplin [1971] AC 356, [1969] 3 WLR 322 (HL)��3.38 Bremer Vulkan Schiffbau und Maschinenfabrik v South India Shipping Corporation Ltd [1981] AC 909, [1981] 2 WLR 141 (HL)��3.06 Brennan v Sunderland City Council [2009] ICR 479 (EAT)��16.82 British American Tobacco Investments Ltd v United States of America [2004] EWCA Civ 1064, [2004] 7 WLUK 986��16.24, 16.25 Brotherton v Aseguradora Colseguros SA [2003] EWCA Civ 705, [2003] 2 All ER (Comm) 298, [2003] Lloyd’s Rep IR 746��12.30 Brown v Guardian Royal Exchange Assurance plc [1994] 2 Lloyd’s Rep 325, [1994] NPC 8 (CA)��16.71, 16.73 Browne v Dunn (1893) 6 R 67 (HL (Scot))��15.64 Buttes Gas and Oil Co v Hammer (No 3) [1981] QB 223, [1980] 3 WLR 668 (CA)��16.63 C v D [2007] EWCA Civ 1282, [2008] 1 WLR 239, [2008] 1 Lloyd’s Rep 239�� 1.27, 3.10, 3.17, 11.81, 11.85 Carl Zeiss Stiftung v Rayner & Keeleer Ltd (No 2) [1967] 1 AC 853, [1966] 3 WLR 125 (HL)��3.34 Carter v Boehm (1766) 3 Burr 1905, 97 ER 1162 (Ct of KB)��A.27 Central Estates (Belgravia) Ltd v Woolgar [1972] 1 WLR 1048, [1972] 3 All ER 610 (CA)��13.15 CIA Barca de Panama SA v George Wimpey & Co Ltd [1980] 1 Lloyd’s Rep 598 (CA)��16.67 Cia Tirrena di Assiccurazioni SpAv Grand Union Insurance Co [1991] 2 Lloyd’s Rep 143 (QBD Comm Ct)��13.15 City of Gotha v Sothebys [1998] 1 WLR 114, (1997) 141 SJLB 152 (CA)��16.90 Civil Aviation Authority v R (on the Application of Jet2.Com Ltd) [2020] EWCA Civ 35, [2020] QB 1027, [2020] 2 WLR 1215��16.34, 16.84 Clay, In re (Clay v Booth) [1919] 1 Ch 66 (CA)��14.33 Cofely Ltd v Bingham [2016] EWHC 240 (Comm), [2016] 2 All ER (Comm) 129, [2016] BLR 187��14.44 Commerce and Industry Insurance Co of Canada v Lloyd’s Underwriters [2002] 1 WLR 1323, [2002] 1 Lloyd’s Rep 219 (QBD Comm Ct)��15.07, 16.97, 16.98, 16.100 Commercial Union Assurance Co plc v Mander [1996] 2 Lloyd’s Rep 640, [1997] CLC 32 (QBD Comm Ct)�� 16.59, 16.63, 16.66, 16.67, 16.69, 16.71, 16.72, 16.73, 16.75

xx

Table of Cases

Compagnie Financière et Commerciale du Pacifique v Peruvian Guano Co (1882) 11 QBD 55 (CA)��16.11 D, In the matter of [2008] UKHL 33, [2008] 1 WLR 1499��7.18 Department of Economics Policy & Development of the City of Moscow v Bankers Trust Co [2003] EWHC 1377 (Comm), [2003] 1 WLR 2885; [2004] EWCA Civ 314, [2005] QB 207 (CA)��15.24 Derby & Co Ltd v Weldon (No 7) [1990] 1 WLR 1156, [1990] 3 All ER 161��16.21 Digicel (St Lucia) Ltd v Cable and Wireless Plc [2009] EWHC 1437 (Ch), [2009] 6 WLUK 466�� 16.43, 16.77, 16.81, 16.82, 16.83 Director of the Serious Fraud Office v Eurasian Natural Resources Corp Ltd [2018] EWCA Civ 2006, [2019] 1 WLR 791, [2019] 1 All ER 1026��16.33, 16.37, 16.40 Dolling-Baker v Merrett [1990] 1 WLR 1205, [1991] 2 All ER 890 (CA)��15.27, 16.93 Dreymoor Fertilisers Overseas Pte Ltd v Eurochem Trading GmbH [2018] EWHC 2267 (Comm), [2018] 2 Lloyd’s Rep 536, [2018] 2 CLC 576�� 16.101 Drinkwater v The Corporation of the London Assurance Co (1767) 2 Wils 363, 95 ER 863��10.96 Dtek Trading SA v Morozov [2017] EWHC 94 (Comm), [2017] 1 Lloyd’s Rep 126, [2017] 1 CLC 53��16.99 Duncan, Re; Garfield v Fay [1968] P 306, [1968] 2 WLR 1479��16.25 Dunlop Slazenger International Ltd v Joe Bloggs Sports Ltd [2003] EWCA Civ 901, [2003] 6 WLUK 197��16.81, 16.82 Durham v BAI (Run Off) Ltd [2012] UKSC 14, [2012] 1 WLR 867��6.16 Edo Corporation v Ultra Electronics Ltd [2009] EWHC 682, [2009] 2 Lloyd’s Rep 349�� 16.103 Egmatra AG v Marco Trading Corp [1999] 1 Lloyd’s Rep 862, [1998] CLC 1552 (QBD Comm Ct)�� 1.28, 3.16 El Du Pont de Nemours & Co v IC Agnew and others [1987] 2 Lloyd’s Rep 585, [1987] 2 FTLR 487 (CA)��14.33 El-Farargy v El-Farargy [2007] EWCA Civ 1149, [2007] 3 FCR 711��14.38 Emmott v Michael Wilson and Partners Ltd [2008] EWCA Civ 184, [2008] 2 All ER (Comm) 193, [2008] Bus LR 1361��14.43 Enka Insaat Ve Sanayi AS v OOO Insurance Company Chubb [2020] UKSC 38, [2020] 1 WLR 4117 (HL); [2020] EWCA Civ 574, [2020] 2 Lloyd’s Rep 389��3.08, 3.09, 3.10, 3.11, 3.17, 3.44, 17.10, 17.12, 17.13, 17.14 Evans Marshall & Co Ltd v Bertola SA [1973] 1 WLR 349, [1973] 1 All ER 992 (CA)��3.21 Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22, [2003] 1 AC 32 (HL)��6.04 Farm Assist Ltd (In Liquidation) v Secretary of State for Environment Food & Rural Affairs [2008] EWHC 3079 (TCC), [2009] BLR 80, [2009] PNLR 16��16.43, 16.77 Fiona Trust & Holding Corp v Privalov [2007] UKHL 40, [2007] 4 All ER 941��3.06, 14.33 Firma C-Trade SA v Newcastle Protection and Indemnity Association [1991] 2 AC 1, [1990] 3 WLR 78 (HL)�� 10.106

Table of Cases

xxi

First Energy (UK) Ltd v Hungarian International Bank Ltd [1993] 2 Lloyd’s Rep 194, [1993] BCLC 1409 (CA)��10.12 Formica v Export Credit Guarantee Department [1995] 1 Lloyd’s Rep 692 (QBD)�� 16.63, 16.66, 16.67, 16.69 Front Comor, The. See West Tankers Inc v RAS Riunione Adriatica di Sicurta SpA Fuld, In the Estate of, Hartley v Fuld (No 3) [1968] P 675, [1966] 2 WLR 717��3.32 Fulham Leisure Holdings v Nicholson Graham & Jones [2006] EWHC 158 (Ch), [2006] 2 All ER 599��16.84 Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd [2001] EWCA Civ 1047, [2001] 2 All ER (Comm) 299��5.40 General Accident Fire and Life Assurance Co v Tanter; The Zephyr [1984] 1 WLR 100, [1984] 1 All ER 35��16.84 General Mediterranean Holdings SA v Patel [1999] 3 All ER 673, [2000] 1 WLR 272�� 15.50, 16.27, 16.28 Glenn v Watson [2016] EWHC 3259 (Ch), [2017] 4 WLR 48��16.81 Good Luck, The; Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd (Disclosure) [1992] 2 Lloyd’s Rep 540 (Note)��16.63 Gray v Barr, Prudential Assurance Co (Third Party) [1971] 2 QB 554, [1971] 2 WLR 1334��3.23 Guidant LLC v Swiss Re International SE [2016] EWHC 1201 (Comm), [2016] 1 CLC 767��14.31, 14.32 Guinness Peat Properties Ltd v Fitzroy Robinson Partnership [1987] 1 WLR 1027, [1987] 2 All ER 716 (CA)��16.64 H (Minors) (Sexual Abuse: Standard of Proof), Re [1996] AC 563, [1996] 2 WLR 8 (HL)��7.18 Halliburton Co v Chubb Bermuda Insurance Ltd (Formerly Ace Bermuda Insurance Ltd) [2020] UKSC 48, [2020] 3 WLR 1474, [2021] 1 Lloyd’s Rep 1�� 14.31, 14.32, 14.38, 14.40, 14.41, 14.42, 14.44, 14.48 Harbour Assurance Co (UK) Ltd v Kansa General International Insurance Co Ltd [1993] QB 701, [1993] 3 WLR 42 (CA)��3.06 Harris v The Society of Lloyd’s [2008] EWHC 1433 (Comm), [2009] Lloyd’s Rep IR 119��16.90 Hayes v Dowding [1996] PNLR 578 (ChD)��16.76, 16.78 Hellenic Industrial Development Bank SA v Atkin (The Julia) [2002] EWHC 1405 (Comm), [2003] Lloyd’s Rep IR 365��17.03 Hellespont Ardent, The. See Red Sea Tankers Ltd v Papachristidis and others Helow (AP) v Secretary of State for the Home Department [2008] UKHL 62, [2008] 1 WLR 2416��14.38 Henderson v Merrett Syndicates Ltd (No 2) [1997] LRLR 247 (QBD Comm Ct)��1.15 Highgrade Traders Ltd, Re [1984] BCLC 151 (CA)��16.40 HLB Kidsons (A Firm) v Lloyd’s Underwriters (Costs) [2007] EWHC 2699 (Comm), [2008] 3 Costs LR 427��17.33, 17.35 Hosking v Apax Partners LLP [2018] EWHC 2732 (Ch), [2019] 1 WLR 3347, [2018] 5 Costs LR 1125��17.38 Hydrosan Ltd, Re [1991] BCLC 418, [1991] BCC 19��16.70

xxii

Table of Cases

Ikarian Reefer, The; National Justice Compania Naviera SA v Prudential Assurance Co [1993] 2 Lloyd’s Rep 68, [1993] FSR 563��15.47 Inco Europe Ltd v First Choice Distribution [2000] 1 WLR 586, [2000] 2 All ER 109 (HL)��A.04 Investors Compensation Scheme v West Bromwich Building Society [1998] 1 WLR 896, [1998] 1 All ER 98 (HL)��4.11 ISTIL Group Inc v Zahoor [2003] EWHC 165 (Ch), [2003] 2 All ER 252, [2003] CP Rep 39��16.32, 16.39 Italia Express, The. See Ventouris v Mountain Jacobs v Coster [2000] Lloyd’s Rep IR 506, [1998] CLY 3374 (CA)��8.29 Jaffray and Others v Society of Lloyd’s [2002] EWCA Civ 1101, (2002) 146 SJLB 214��16.20 James-Bowen v Commissioner of Police of Metropolis [2018] UKSC 40, [2018] 1 WLR 4021, [2018] 4 All ER 1007��16.66, 10.67 Julia, The. See Hellenic Industrial Development Bank SA v Atkin Kanchenjunga, The. See Motor Oil Hellas (Corinth) Refineries SA v Shipping Corp of India Kastor Navigation Co Ltd v Axa Global Risks (UK) Ltd [2004] EWCA Civ 277, [2004] 2 Lloyds Rep 119��17.35 Kazakhstan Kagazy plc v Zhunus (formerly Zhunussov) [2018] EWHC 369 (Comm), [2018] 2 WLUK 675��17.03 Kennedy v Cordia (Services) LLP [2016] UKSC 6, [2016] 1 WLR 597, 2016 SLT 209��15.42 Kershaw v Whelan [1996] 1 WLR 358, [1996] 2 All ER 404 (QBD)��16.76 Korner v Witkowitzer [1950] 2 KB 128, [1950] 1 All ER 558 (CA)��3.33 Kuwait Airways Corp v Kuwait Insurance Co SAK [1996] 1 Lloyd’s Rep 664 (QBD Comm Ct)��6.33 Kuwait Airways Corp v Kuwait Insurance Co SAK [2000] 1 Lloyd’s Rep IR 678��17.03 Laker Airways Inc v FLS Aerospace Ltd [1999] 2 Lloyd’s Rep 45, [1999] CLC 1124 (QBD Comm Ct)�� 14.38, 14.41, 14.47, A.05 Lancashire County Council v Municipal Mutual Insurance Ltd [1997] QB 897, [1996] 3 WLR 493 (CA)��3.23 Lawrence v Campbell (1859) 4 Drewry 485��16.25 Layher Ltd v Lowe [2000] Lloyd’s Rep IR 510, 58 Con LR 42 (CA)��8.29 Lesotho Highlands Development Authority v Impregilo SpA & ors [2002] EWHC 2435 (Comm), [2003] 1 All ER (Comm) 22 (HC); [2003] EWCA Civ 1159, [2003] 2 Lloyd’s Rep 497, (CA); [2005] UKHL 43, [2006] 1 AC 221 (HL)�� 3.06, 3.43, 3.44, 11.55, 11.80, 17.04, 17.11, 17.12, 17.15 Lillicrap v Nalder & Sons [1993] 1 WLR 94, [1993] 1 All ER 724 (CA)��16.76 Lincoln National Life Insurance Co v Sun Life Assurance Co of Canada [2004] EWHC 343 (Comm), [2004] 1 Lloyd’s Rep 737, [2004] 2 CLC 36��16.20 Locabail (UK) Ltd v Bayfield Properties Ltd [2000] QB 451, [2000] 2 WLR 870 (CA)��14.38, 14.42 Lumbermens Mutual Casualty Co v Bovis Lend Lease Ltd [2004] EWHC 2197 (Comm), [2005] 1 Lloyd’s Rep 494��3.30

Table of Cases

xxiii

MAC Hotels Ltd v Rider Levett Bucknall UK Ltd [2010] EWHC 767 (TCC)��16.82, 16.83 Mackender v Feldia [1967] 2 QB 590, [1966] 3 All ER 847��3.21 Makers UK Ltd v The Mayor and Burgesses of the London Borough of Camden [2008] EWHC 1836 (TCC), [2008] BLR 470, [2008] 3 EGLR 1��14.35 Mann v Lexington Insurance Co [2001] Lloyd’s Rep IR 179, [2001] 1 Lloyd’s Rep 1 (CA)��6.33 Marc Rich & Co v Portman [1996] 1 Lloyd’s Rep 430 (QBD Comm Ct); [1997] 1 Lloyd’s Rep 225 (CA)��12.27 Martrade Shipping & Transport GmbH v United Enterprises Corp [2014] EWHC 1884 (Comm), [2015] 1 WLR 1, [2014] 2 Lloyd’s Rep 198��17.04 Matthews v Smallwood [1910] 1 Ch 777��13.15 MDIS v Swinbank [1999] Lloyd’s Rep IR 516 (CA)��3.30 Medicaments and Related Classes of Goods (No 2), Re [2000] EWCA Civ 350, [2001] 1 WLR 700, [2001] ICR 564��14.38 Messier Dowty Ltd v Sabena SA [2000] EWCA Civ 48, [2000] 1 Lloyd’s Rep 428, [2000] CP Rep 72��3.42 Michael Wilson & Partners Ltd v Emmott [2008] EWCA Civ 184, [2008] 1 Lloyd’s Rep 616�� 15.24, 15.25, 15.27, 16.93 Miliangos v George Frank (Textiles) Ltd [1976] AC 443, [1975] 3 WLR 758 (HL)��3.43 Morris v Banque Arab International D’Investissement SA [2001] 1 BCLC 263, [2002] BCC 407 (ChD)��16.30 Motor Oil Hellas (Corinth) Refineries SA v Shipping Corp of India (The Kanchenjunga) [1990] 1 Lloyd’s Rep 391 (HL)��13.02 Motorola Solutions Inc v Hytera Communications Corp Ltd [2021] EWCA Civ 11, [2021] 2 WLR 679��16.23 Multiplex Constructions (UK) Ltd v Cleveland Bridge UK Ltd [2008] EWHC 2280 (TCC), [2009] 1 Costs LR 55, 122 Con LR 88��17.33 National Westminster Bank Plc v Rabobank Nederland [2007] EWHC 1742 (Comm), [2008] 1 All ER (Comm) 243, [2008] 3 Costs LR 396��17.03, 17.38 Nea Karteria Maritime Co Ltd v Atlantic and Great Lakes Steamship Corp (No 2) [1981] Com LR 138��16.81 Nederlandse Reassurantie Groep Holding NV v Bacon & Woodrow (a firm) and others [1995] 1 All ER 976, [1995] 2 Lloyd’s Rep 77 (QBD)��16.24 Noble Assurance Co v Gerling-Konzern General Insurance Co [2007] EWHC 253 (Comm), [2007] 1 CLC 85, [2007] ArbLR 44�� 1.19, 9.44, 11.81 Norbrook Laboratories Ltd v Tank [2006] EWHC 1055 (Comm), [2006] ArbLR 50��14.35, 14.38 Norwich Union Life Insurance Society v British Railways Board (1987) 283 EG 846��3.20 NRG v Bacon & Woodrow [1995] 1 All ER 976, [1995] 2 Lloyd’s Rep 77 (QBD)��16.90 Nukila, The; Promet Engineering (Singapore) Pte Ltd v Sturge [1997] 2 Lloyd’s Rep 146 (CA)��10.78 Oak Property Co v Chapman [1947] 1 KB 886, [1947] LJR 1327 (CA)��13.15

xxiv

Table of Cases

Omega Group Holdings Ltd & others v Kozeny & others [2002] CLC 132 (QBD Comm Ct)�� 16.101 Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co Ltd [1995] 1 AC 501, [1994] 3 WLR 677 (HL)��A.08 Panayiotou v Sony Music Entertainment (UK) Ltd [1994] Ch 142, [1994] 2 WLR 241 (DC)�� 16.99, 16.103 Paragon Finance v Freshfields [1999] 1 WLR 1183, [1999] Lloyd’s Rep PN 446 (CA)�� 16.24, 16.27, 16.47, 16.76, 16.77, 16.78, 16.84 Phrantzes v Argenti [1960] 2 QB 19, [1960] 2 WLR 521��3.40 Pickering, Re (1880) 25 Ch D 247��16.70 Pilkington UK Ltd v CGU Insurance PLC [2004] EWCA Civ 23, [2004] Lloyd’s Rep IR 891��5.59, A.23 PJSC Tatneft v Bogolyubov [2020] EWHC 2437 (Comm), [2021] 1 WLR 403, [2021] 2 All ER 224��16.34 Police Authority for Huddersfield v Watson [1947] KB 842, [1948] LJR 182 (DC)��A.23 Porter v Magill [2001] UKHL 67, [2002] 2 AC 357��14.38 Practice Statement (Judicial Precedent) [1966] 1 WLR 1234, [1966] 3 All ER 77 (HL)��A.10 R v Derby v Magistrates Court, ex parte B [1996] 1 AC 487, [1995] 3 WLR 681 (HL)��16.27, 16.47 R v Gough (Robert) [1993] AC 646, [1993] 2 WLR 883 (HL)��14.38 R (Morgan Grenfell Ltd) v Special Commissioner [2002] UKHL 21, [2002] 2 WLR 1299, [2003] 1 AC 563��16.27, 16.28 R (on the application of Kelly) v Warley Magistrates’ Court [2007] EWHC 1836 (Admin) [2008] 1 WLR 2001��16.28 RBS Rights Issue Litigation [2016] EWHC 3161 (Ch), [2017] 1 WLR 1991, [2017] Lloyd’ds Rep FC 83�� 16.24, 16.25, 16.31, 16.37 Red Sea Tankers Ltd v Papachristidis and others; The Hellespont Ardent [1997] 2 Lloyd’s Rep 547 (QBD (Comm Ct)��15.42 Reliance Industries Ltd v Enron Oil and Gas India Ltd [2002] 1 Lloyd’s Rep 645 (QBD Comm Ct)�� 1.28, 3.15, 3.16 Rio Tinto Zinc Corp v Westinghouse Electric Corp [1978] AC 547, [1978] 2 WLR 81 (HL)�� 16.100 Roberta, The (1937) 58 Ll LR 159��3.32 Rolls Royce plc v Unite the Union [2009] EWCA Civ 387, [2010] 1 WLR 318��14.33 St Pierre v South American Stores Ltd [1937] 3 All ER 349��3.33 Sanghi Polyesters Ltd (India) Ltd v The International Investor (KCFC) Kuwait [2000] 1 Lloyd’s Rep 480, [2001] CLC 748��3.16 Savings & Investment Bank Ltd v Finken [2003] EWCA Civ 1630, [2004] 1 All ER 1125��16.23 Science Research Council v Nasse [1980] AC 1028, [1979] ICR 921 (HL)��16.93 Secretary of State for the Home Department v Raytheon Systems Ltd [2014] EWHC 4375 (TCC), [2014] 12 WLUK 774��15.69 Secretary of State for the Home Department v Rehman [2001] UKHL 47, [2003] 1 AC 153��7.18

Table of Cases

xxv

Secretary of State for Trade & Industry v Baker & ors (No 5) [1999] 1 BCLC 433 (ChD (Companies Ct))��15.42 Segovia Compagnia Naviere SA v R Pagnan & Fratelli; The Aragon [1977] 1 Lloyd’s Rep 343 (CA)��4.11 Silver Dry Bulk Co v Homer Hulbert Maritime Co Ltd [2017] EWHC 44 (Comm), [2017] 1 All ER (Comm) 791, [2017] 1 Lloyd’s Rep 154��16.99 Skandia International Corp v NRG Victory Reinsurance Ltd [1998] Lloyd’s Rep IR 439, [1998] 2 All ER 434 (CA)��3.30 Smith v Kvaerner Cementation Foundations Ltd [2006] EWCA Civ 242, [2007] 1 WLR 370��14.47 Society of Lloyd’s v Jaffray [2002] EWCA Civ 1101 (2002) 146 SJLB 214 (CA); (2000) November 3 (QBD)�� 1.06, 1.15, 16.20 Structural Polymer Systems v Brown; The Baltic Universal [2000] Lloyd’s Rep IR 64, [1999] CLC 268 (QBD (Comm Ct))��3.30 Svenska Handesbanken v Sun Alliance and London Insurance PLC [1995] 2 Lloyds Rep 84 (QBD (Comm Ct))��16.64 Tajik Aluminium Plant v Hydro Aluminium [2005] EWCA Civ 1218, [2006] 1 WLR 767�� 16.103 Tasman Spirit, The. See Assimina Maritime Ltd v Pakistan Shipping Corp Tate and Lyle Food & Distribution Ltd v Greater London Council [1982] 1 WLR 149, [1981] 3 All ER 716��17.03 Tchenguiz Imerman v Imerman [2012] EWHC 4047 (Fam), [2014] 1 FLR 232, [2013] Fam Law 966��16.21 Thornton Springer v NEM Insurance Co Ltd [2000] Lloyd’s Rep IR 590, [2000] 2 All ER 489��3.30 Three Rivers District Council and others v Bank of England (No 5) [2003] EWCA Civ 474, [2003] QB 1556��16.37, 16.39 Three Rivers District Council and others v Bank of England (No 6) [2004] UKHL 48, [2005] 1 AC 610�� 16.33, 16.34, 16.36, 16.37, 16.47 Triple Point Technology Inc v PTT Public Co Ltd [2018] EWHC 45 (TCC), [2018] 1 Costs LO 111��17.03 TSB Bank plc v Robert Irving & Burns [2000] 2 All ER 826, [1999] Lloyd’s Rep IR 528 (CA)��16.71 UMS Holding Ltd v Great Station Properties SA [2017] EWHC 2398 (Comm), [2018] 1 All ER (Comm) 856, [2017] 2 Lloyd’s Rep 421��15.70 USP Strategies plc v London General Holdings Ltd [2004] EWHC 373 (Ch), [2004] 3 WLUK 205��16.90 Unilever plc v Procter & Gamble Co [2000] 1 WLR 2436, [2000] 1 All ER 783 (CA)��16.23 Van der Giessen-de-Noord Shipbuilding Division BV v Imtech Marine & Offshore BV [2008] EWHC 2904 (Comm), [2009] 1 Lloyd’s Rep 273��15.69 Ventouris v Mountain; The Italia Express [1991] 1 WLR 607, [1991] 3 All ER 472 (CA)��16.34 Vitkovice v Korner [1951] AC 869, [1951] 2 All ER 334 (HL)��3.33 W Ltd v M SDN BHD [2016] EWHC 422 (Comm), [2017] 1 All ER (Comm) 981, [2016] 1 Lloyd’s Rep 552��14.38

xxvi

Table of Cases

Walker v Rome [1999] 2 All ER (Comm) 961, [2000] 1 Lloyd’s Rep 116 (QBD Comm Ct)��17.26 Watts v Watts [2015] EWCA Civ 1297, [2015] 12 WLUK 728��14.47 Waugh v British Railways Board [1980] AC 521, [1979] 3 WLR 150 (HL)��16.40 Wentworth v Lloyd (1864) 10 HLC 589��16.41 West London Pipeline & Storage Ltd v Total UK Ltd [2008] EWHC 1729 (Comm), [2008] 2 CLC 258�� 16.20, 16.21, 16.42 West Tankers Inc v RAS Riunione Adriatica di Sicurta SpA (The Front Comor) [2007] UKHL 4, [2007] 1 Lloyd’s Rep 391��3.08 Westwood Shipping Lines Inc v Universal Schiffahrtsgesellschaft mbH [2012] EWHC 3837 (Comm), [2013] 1 Lloyd’s Rep 670��15.25 WH Holding Ltd and West Ham United Football Club Ltd v E20 Stadium LLP [2018] EWCA Civ 2652, [2018] 11 WLUK 511��16.21, 16.39 Wheeler v Le Marchant (1881) 17 Ch D 675��16.39 Winterthur Swiss Insurance Co v AG (Manchester) Ltd (In Liquidation) [2006] EWHC 839 (Comm), [2006] 4 WLUK 329�� 16.33, 16.34, 16.39, 16.64, 16.66, 16.67, 16.69, 16.73 X Corp Ltd v Y (a firm), 16 May 1997 (unreported)��5.30 XL Insurance Co Ltd v Owens Corning [2001] 1 All ER (Comm) 530, [2000] 2 Lloyd’s Rep 500 (QBD Comm Ct)��3.10 XL Insurance Ltd v Toyota Motor Sales USA Inc (14 July 1999) (unreported)��14.30 Yorkshire Water Services Ltd v Sun Alliance & London Insurance Plc [1997] 2 Lloyd’s Rep 21, [1997] CLC 213 (CA)��A.27 Youell v Bland Welch & Co Ltd [1992] 2 Lloyd’s Rep 127 (CA)��10.06 Zephyr, The. See General Accident Fire and Life Assurance Co v Tanter Zuma’s Choice Pet Products Ltd v Azumi Ltd [2017] EWCA Civ 2133, [2017] 12 WLUK 374��14.47 United States of America 35 Park Ave v Campagna 48 NY2d 813 (NY 1979)��12.41 Abex Corp v Maryland Cas Co 790 F2d 119 (DC Cir 1986)��1.15 ABI Asset Corp v Twin City Fire Insurance Co No 96 Civ 2067, 1997 WL 724568 (SDNY 18 November 1997)��10.18 AC&S, Inc v Aetna Cas & Sur Co 576 F Supp 936 (ED Pa 1983), aff ’d 764 F2d 968 (3d Cir 1985)��1.10 Aerojet-General Corp v Superior Court 257 Cal Rptr 621, supplemented on denial of reh’g, 258 Cal Rptr 684 (Ct App 1989)��10.49 Aetna Cas & Sur Co v Abbot Labs, Inc 636 F Supp 546 (D Conn 1986)��1.15 Aetna Cas & Sur Co v Bell 390 F2d 612 (1st Cir 1968)��11.51 Aetna Cas & Sur Co v Crown 181 AD2d 883 (App Div 1992)��5.36 Aetna Cas & Sur Co v General Time Corp 704 F2d 80 (2d Cir 1983)�� 5.16, 5.17 Aetna Cas & Sur Co v Lumbermens Mut Cas Co 527 NYS2d 143 (NY App Div 1998)��5.05

Table of Cases

xxvii

Agent Orange Liability Litigation, In re 597 F Supp 740 (EDNY 1984), aff ’d, 818 F2d 145 (2d Cir 1987)��5.30 AH Prop LLC v N H Ins Co 945 NYS2d 391 (App Div 2012)��8.20 Aiena v Olsen 194 FRD 134 (SDNY 2000)��16.57 Air Et Chaleur SA v Janeway 757 F2d 489 (2d Cir 1985)��5.41 AIU Ins Co v Superior Court (FMC Corp) 799 P2d 1253 (Cal 1990)��10.49 Alaz Sportswear v Public Mut Ins Co 600 NYS2d 63 (NY 1993)��12.05 Albert J Schiff Associates, Inc v Flack 435 NY2d 972 (NY1980)��13.01, 13.08, 13.19 Album Realty Corp v American Home Assurance Co 592 NYS2d 657 (NY 1992)�� 10.12, 10.19, 10.20, 10.24 Ali v Federal Ins Co 719 F3d 83 (2d Cir 2013)��9.27 AllCity Ins Co v Sioukas 378 NYS2d 711 (App Div 1976), aff ’d, 393 NYS2d 993 (NY 1977)��13.05 Allianz Ins Co v Otero 353 F Supp 2d 415 (SDNY 2004)��2.24 Allstate Ins Co v Gross 27 NY2d 263 (NY 1970)��13.03 Allstate Insurance Co v Mugavero 79 NY2d 153, 581 NYS2d 142 (1992)�� 7.28, 7.34 Allstate Insurance Co v Sullivan 646 NYS2d 359 (App Div 1996)��5.36 Allstate Insurance Co v White Metal Rolling & Stamping Corp 466 F Supp 419 (EDNY 1979)��3.27 Allstate Insurance Co v Zuk 571 NYS2d 429 (NY 1991)�� 7.24, 7.28, 7.48 Alyeska Pipeline Co v Wilderness Soc’y 412 US 240 (1975)��11.87 Ambassador Group Inc Litig, In re 738 F Supp 57 (EDNY 1990)��5.22 American Home Assurance Co v International Ins Co 661 NYS2d 584 (NY 1997)��8.14 American Home Assurance Co v Libbey-Owens-Ford Co 786 F2d 22 (1st Cir 1986)�� 5.16, 5.61 American Home Prods Corp v Liberty Mutual Ins Co 565 F Supp 1485 (SDNY 1983), aff ’d as modified 748 F2d 760 (2d Cir 1984)�� 1.04, 1.10, 1.14, 1.15, 4.13, 4.17, 5.57, 5.60 American Ins Co v Fairchild Indus Inc 56 F3d 435 (2d Cir 1995)��8.13 American Int’l Specialty Lines Ins Co v Towers Fin Corp No 94 Civ 2727, 1997 WL 90627 (SDNY 12 Sept 1997)��12.32 American Nat’l Fire Ins Co v Mirasco Inc 249 F Supp 2d 303 (SDNY 2003), vacated in part on other grounds sub nom Mirasco Inc v American Nat’l Fire Ins Co 144 F App’x 171 (2d Cir 2005)��5.49, 10.07, 10.17 American Re-Ins Co v United States Fidelity & Guar Co 837 NYS2d 616 (App Div 2007)��16.55 American States Ins Co v Koloms 687 NE2d 72 (Ill 1997)��10.109, 10.111, 10.113 American States Ins Co v Nethery 79 F3d 473 (5th Cir 1996)�� 10.111, 10.115 American W Home Ins Co v Gjonaj Realty & Mgmt Co 192 AD3d 28 (App Div 2020)��5.22 Ames v New York Union Ins Co 14 NY 253 (NY 1856)��11.42 Amrep Corp v American Home Assurance Co 440 NYS2d 244 (App Div 1981)��12.19

xxviii

Table of Cases

Amtrol Inc v Tudor Ins Co No 01-10461-DPW, 2002WL 31194863 (D Mass 2002)��5.61 Ansonia Assoc Ltd P ’ship v Public Serv Mut Ins Co 257 AD2d 84 (NY App Div 1999)��5.39, 11.39 Apache Foam Prods v Continental Ins Co 528 NYS2d 449 (App Div 1988)��10.75 Appalachian Ins Co v Gen Elec Co 8 NY3d 162 (NY 2007)��6.33 Arbitration in London England between Norfolk S Ry and General Sec Ins Co and ACE Bermuda Ltd, In re 626 F Supp 2d 882 (ND Ill 2009)��16.98 Argo Corp v Greater NY Mut Ins Co 794 NYS3d 704 (NY 2005)�� 8.12, 8.14 Armstrong World Indus, Inc v Aetna Cas & Sur Co 52 Cal Rptr 2d 690 (Cal Ct App 1996)��10.81 Arno Schefl er v Livestock Cas Ins Co 44 AD 2d 811 (App Div 1974)��11.39 Arrowood Indem Co v Atlantic Mut Ins Co 948 NYS2d 581 (App Div 2012)��11.66 Arthur A Johnson Corp v Indem Ins Co of N Am 7 NY2d 222 (NY 1959)�� 6.33, 6.37 Asbestos Ins Coverage Litig, In re (Armstrong World Indus, GAF, Fibreboard) Judicial Council Coordinated Proceeding No 1072, Phase III (Cal Super Ct 24 Jan 1990); aff’d sub nom Armstrong World Indus, Inc v Aetna Cas & Sur Co 52 Cal Rptr 2d 690 (Ct App 1996)��1.15 Ashline v Genesee Patrons Cooperative Insurance Society 638 NYS2d 217 (App Div 1996)��12.23 Atlantic Cement Co v Fidelity & Cas Co of NY 459 NYS2d 425 (App Div 1983), aff ’d 481 NY2d 329 (NY 1984)��4.24 Auto-Owners Ins Co v Jensen 667 F2d 714 (8th Cir 1981)�� 7.25, 7.28 Automobile Ins Co of Hartford v Cook 808 NYS2d 176 (2006)��7.28 Avondale Indus, Inc v Travelers Indem Co 774 F Supp 1416 (SDNY 1991)�� 8.10, 8.12 AXA Equitable Life Ins Co v Malen 986 NYS2d 132 (App Div 2014)��12.41 Banker’s Trust Co v Hartford Accident & Indemnity Co 518 F Supp 371 (SDNY), vacated due to settlement 621 F Supp 685 (SDNY 1981)��10.46, 10.47 Barnum v Merchants’ Fire Ins Co 97 NY 188 (NY 1884)��11.49, 11.50 Barrett v State Mut Life Assurance Co 396 NYS2d 848 (App Div 1977), aff ’d 407 NYS2d 478 (NY 1978)��12.19 Barrier Systems Inc v AFC Enterprises Inc 694 NYS2d 440 (App Div 1999)��12.34 Barton v Home Ins Co 42 Mo 156 (1868)��10.96 Bausch & Lomb, Inc v Lexington Ins Co 414 F App’x 366 (2d Cir 2011)�� 6.34, 6.35 Bausch & Lomb Inc v Utica Mut Ins Co 625 A2d 1021 (Md 1993)��10.49 Bazar v Great Am Indem Co 306 NY 481 (1954)��13.27 Beazley Ins Co v Ace Am Ins Co 880 F3d 64 (2d Cir 2018)�� 10.23, 10.24, 10.38 Bebber v CNA Ins Cos 729 NYS2d 844 (Sup Ct 2001)��10.09, 10.26 Belesi v Connecticut Mut Life Ins Co 707 NYS2d 663 (App Div 2000)��12.35 Belt Painting Corp v TIG Ins Co 100 NY2d 377, 763 NYS2d 790, 795 NE2d 15 (NY July 1, 2003)��10.111, 10.112, 10.113, 10.116 Belt Painting Corp v TIG Ins Co 742 NYS2d 332, 293 AD2d 206 (NY App Div May 13 2002)��4.11, 10.04, 10.05, 10.08, 10.10, 10.108, 10.109, 10.110 Berger v Manhattan Life Insurance Co 805 F Supp 1097 (SDNY 1992)�� 12.21, 12.23, 12.29 Bethlehem Steel Co v Turner Construction Co 161 NYS2d 90 (NY 1957)��4.19

Table of Cases

xxix

Bi-Economy Market Inc v Harleysville Ins Co of NY 10 NY3d 187 (NY 2008)��5.39 Bifulco v Great N Ins Co No 99-CV-0119E(M), 2001 WL 877335 (WDNY 3 July 2001)�� 12.14, 12.21, 12.32 Biloz v Tioga Cty Patrons’ Fire Relief Ass’n 21 NYS2d 643 (Sup Ct 1940)��13.27 Bird v St Paul Fire & Marine Insurance Co 224 NY 47 (NY 1918)��10.12, 10.13, 10.16, 10.20, 10.22 Bituminous Cas Corp v Advance Adhesive Tech Inc 73 F3d 335 (11th Cir 1996)�� 10.111 Bituminous Cas Corp v Cowen Constr Inc 55 P3d 1030 (Okla 2002)�� 10.109 Black v Goodwin, Loomis & Britton Inc 681 A 2d 293 (Conn 1996)��5.31 Blackfield v Underwriters at Lloyd’s London 245 Cal App 2d 271 (1966)��10.60 Board of Educ Yonkers City Sch Dist v CNA Ins Co 647 F Supp 1495 (SDNY 1986), aff ’d 839 F2d 14 (2d Cir 1988)��4.24 Board of Regents of University of Minnesota v Royal Ins Co of Am 517 NW2d 888 (Minn 1994)�� 10.109, 10.115 Boeing Co v Aetna Cas & Sur Co 784 P 2d 507 (Wash 1990)��4.07 Borden Inc v Howard Trucking Co 454 So 2d 1081 (La 1984)��5.16 Borel v Fibreboard Products Corp 493 F2d 1076 (5th Cir 1973)�� 1.05, 5.31 Bowes & Co v American Druggists Ins Co 467 NYS2d 202 (App Div 1983)��12.41 Boyd v Otsego Mut Fire Ins Co 510 NYS2d 371 (App Div 1986)��12.05 Brady v Nally 151 NY 258 (1896)��3.27 Brandon (National Mut Ins Co), In re 743 NYS2d 53 (NY 2002)��8.14 Breed v Insurance Co of N Am 46 NY2d 351, 413 NYS2d 352, 385 NE2d 1260 (App Div 1978)�� 4.09, 4.13, 4.22, 10.05 Bronx Sav Bank v Weigandt 1 NY2d 545, 154 NYS2d 878, 136 NE2d 848 (NY 1956)��12.08, 12.13 Broome Cty v Travelers Indem Co 125 AD3d 1241 (NY App Div 2015)�� 10.112 Brown v American Motorists Ins Co 930 F Supp 207 (ED Pa 1996), aff ’d, 111 F3d 125 (3d Cir 1997)�� 10.111, 10.115 Brown v Metro Life Ins Co 343 NYS2d 443 (App Div 1973)��12.25 Brunswick Corp v St Paul Fire & Marine Ins Co 509 F Supp 750 (ED Pa 1981)��11.68 Buckeye Check Cashing, Inc v Cardegna, 546 US 440 (2005)��3.06 Bullseye Rest Inc v James River Ins Co 387 F Supp 3d 273 (EDNY 2019)��8.17 Bunge Corp v London & Overseas Insurance Co 394 F2d 496 (2d Cir 1968)��5.37, 11.39 Burlington Ins Co v NYC Transit Auth 29 NY3d 313 (NY 2017)��10.11, 10.24 Burroughs Wellcome Co v Commercial Union Ins Co 632 F Supp 1213 (SDNY 1986)�� 5.16, 5.33, 5.57 CACI Int’l Inc v St Paul Fire & Marine Ins Co 566 F3d 150 (4th Cir 2009)��10.92 Calabro v Stone 225 FRD 96 (SDNY 2004)��16.57 Calvert Ins Co v S & L Realty Corp 926 F Supp 44 (SDNY 1996)�� 10.110 Campagnola v Mulholland 556 NYS2d 239 (NY 1990)��11.87 Campese v National Grange Mut Ins Co 689 NYS2d 313 (App Div 1999)�� 12.19, 12.22, 12.23, 12.25 Campbell v Met Prop & Cas Ins Co No 98 Civ 5328 NRB, 2000 WL 297174 (SDNY 21 Mar 2000), aff’d in part, rev’d in part on grounds 239 F3d 179 (2d Cir 2001)��5.57

xxx

Table of Cases

Carey Canada, Inc v Aetna Cas & Sur Co No Civ A 84-3113, 1988 US Dist Lexis 8997 (DDC 1988)��10.81 Carey Canada Inc v California Union Ins Co 720 F Supp 1018 (DDC 1989), aff ’d in part, rev’d in part, remanded, 940 F2d 1548 (DC Cir 1991)�� 10.104 Carfax Inc v Ill Nat’l Ins Co 7 NYS3d 478 (App Div 2018)�� 10.134 Carlson v American Int’l Group Inc 30 NY3d 288 (NY 2017)��8.16, 13.04 Carpinone v Mutual of Omaha Ins Co 697 NYS2d 381 (App Div 1999)��12.19, 12.22, 12.23, 12.32 Carrier Corp v Allstate Ins Co 63 Misc 3d 1212(A), 2018 NY Slip Op 51965(U) (Sup Ct 21 November 2018), aff’d as modified 133 NYS3d 697 (App Div 2020)��5.57 Carroll v Charter Oak Ins Co 38 Barb 402 (NY Sup Ct 1862)��11.66 Carter Wallace Inc v Admiral Ins Co 712 A 2d 1116 (1994)��9.31 Casey v General Accident Insurance Co 178 AD2d 1001, 578 NYS2d 337 (App Div 1991)�� 4.07, 4.22, 10.05 Cassandra Group, In re 338 BR 583 (Bankr SDNY 2006)��17.19 Castillo v Prince Plaza 43 Misc 3d 335 (NY App Div 2014)��8.17 Castillo v Prince Plaza 142 AD3d 1127 (NY App Div 2016)��8.17 Castle Village Owners Corp v Greater NY Mut Ins Co 878 NYS2d 311 (App 2009)��10.46 Celotex Corp v AIU Ins Co (In re Celotex Corp) 175 BR 98 (Bankr MD Fla 1994)�� 10.104 Century Indem Co v Keyspan Corp 15 Misc 3d 1132(A), (Sup Ct 2007) (unreported disposition), aff’d 58 AD3d 573 (NY App Div 2009)�� 8.11, 8.28, 8.29 Cepeda v Varveris 651 NYS2d 185 (App Div 1996)�� 10.110 Certain Underwriters at Lloyd’s v Milberg LLP 2009 WL 3241489 (SDNY 30 September 2009)��12.41 Champion Int’l Corp v Continental Cas Co 400 F Supp 978 (SDNY 1975), aff ’d 546 F2d 502 (2d Cir 1976)��4.24 Cherkes v Postal Life Ins Co 138 NYS2d 88 (App Div 1955), aff ’d 309 NY 964 (1956)��12.15 Chester Med Diagnosis PC v Kemper Cas Ins Co 873 NYS2d 232 (Civ Ct 2008)��11.46 Chevron Corp, In re 709 F Supp 2d 283 (SDNY 6 May 2010)��16.98 Chicago Insurance Co v Halcond 49 F Supp 2d 312 (SDNY 1999)��12.09, 12.13 Chicago Insurance Co v Kreitzer & Vogelman 265 F Supp 2d 335 (SDNY 2003)��13.07, 13.14 Christiania Gen Ins Corp v Great Am Ins Co 979 F2d 268 (2d Cir1992)��8.29, 12.10, 12.19, 12.32 Chubb Ins Co of NJ v Hartford Fire Ins Co No 97 CIV 6935 LAP, 1999 WL 760206 (SDNY 1999)��7.27 Cincinnati Ins Co v Roy’s Plumbing, Inc No 16-2511, 2017 US App LEXIS 9729 (2d Cir 31 May 2017)�� 10.112 Cincotta v National Floor Insurer Assocs 452 F Supp 928 (EDNY 1977)��10.16 City of Johnstown v Bankers Standard Insurance Co 877 F2d 1146 (2d Cir 1989)�� 7.07, 7.08, 7.11, 7.23, 7.24, 7.25, 7.32, 7.34

Table of Cases

xxxi

City of Kirtland v World Ins Co 540 NE2d 282 (Ohio Ct App 1988)��5.19 City of Utica v Genessee Management Inc 934 F Supp 510 (NDNY 1996)��13.10 City of Ypsilanti v Appalachian Ins Co 547 F Supp 823 (ED Mich 1982), aff ’d, 725 F2d 682 (6th Cir 1983)��5.19 Clendenin Bros Inc v United States Fire Ins Co 889 A2d 387 (Md 2006)�� 10.111 Clifford Chance LLP v Indian Harbor Insurance Co 836 NYS2d 484 (Sup Ct 2006), aff ’d, 838 NYS2d 62 (App Div 2007)�� 5.47, 5.51 CNA Reinsurance of London Ltd v Home Insurance Co No 85 Civ 5681, 1990 WL 3231 (SDNY 10 Jan 1990)�� 12.31, 12.37, 13.12, 13.14 Coastal Caisson Corp v EE Cruz/NAB/Frontier-Kemper No 05 civ 7462 (DLC), 2007 WL 2285936 (SDNY 2007), vacated on other grounds 346 F App’x 717 (2d Cir 2009)��17.10, 17.22 Cohen v Monumental Life Ins Co 194 SE 2d 867 (NC App 1973)��10.93 Coleman v New Amsterdam Cas Co 160 NE 367 (NY 1928)��11.21 Collins v Isaksen 633 NYS2d 539 (App Div 1995)��13.27 Coastal Power International Ltd v Transcontinental Capital Corp 10 F Supp 2d 345 (SDNY 1998)��17.19, 17.21 Colonial Tanning Corp v Home Indem Co 780 F Supp 906 (NDNY 1991)��10.49 Comer v Murphy Oil USA 585 F3d 855 (5th Cir 2009), reh’g granted, 598 F3d 208 (5th Cir 2010)�� 10.117 Commerce & Indus Ins Co v Gun Hill Mgmt Inc No 10147910, 2010 WL 4530221 (Sup Ct 2010)��12.43 Commercial Union Assurance Co v Glass-Lined Pipe Co 372 So 2d 1305 (Ala 1979)��10.85 Commercial Union Ins Co v International Flavors & Fragrances Inc 822 F2d 267 (2d Cir 1987)�� 8.11, 8.27 Commercial Union Ins Co v Pittsburgh Corning Corp 553 F Supp 425 (ED Pa 1981); rev ’d on other grounds, 789 F2d 214 (3d Cir 1986)��1.10 Commercial Union Ins Co v Sepco Corp 765 F2d 1543 (11th Cir 1985)��1.10 Commonwealth Asssociates v Letsos 40 F Supp 2d 170 (SDNY 1999)��17.19, 17.21 Conergic Corp v Dearborn Mid-W Conveyor Co 144 AD3d 516 (NY App 2016)��8.17 Conn v Am Elec Power Co; Open Space Inst v Am Elec Power Co 582 F3d 309 (2d Cir 2009)�� 10.117 Consolidated Edison Co of NY v Allstate Insurance Co 98 NY2d 208, 746 NYS2d 622 (2002)�� 1.13, 4.13, 5.48, 5.52, 5.54, 7.17, 7.24 Continental Casualty Co v Employers Ins Co of Wausau 871 NYS2d 48 (App Div 2008), leave to appeal denied, 13 NY2d 710 (2009), rev’d on other grounds 85 AD3d 403 (NY APP Div 2011)��5.57 Continental Casualty Co v Rapid-American Corp 593 NYS2d 966, 609 NE2d 506 (NY 1993)��5.42, 5.57, 7.24, 7.25, 7.26, 7.27, 7.28, 10.08, 10.110, 10.111 Continental Casualty Co v Stradford 847 NYS2d 631 (App Div 2007), aff ’d as modified, 871 NYS2d 607 (NY 2008)��11.21 Continental Casualty Co v Stronghold Ins Co 77 F3d 16 (2d Cir 1996)��11.42, 11.46, 11.50, 11.51, 11.52

xxxii

Table of Cases

Continental Insurance Co v Arkwright Mutual Insurance Co 102 F3d 30 (1st Cir 1996)��4.17, 10.12, 10.22 Continental Insurance Co v Colangione 484 NYS2d 929 (App Div 1985)��7.24 Continental Insurance Co v Helmsley Enterprises, Inc 622 NYS2d 20 (App Div 1995)��12.34, 13.14 Contra National Elec Mfrs Ass’n v Gulf Underwriters Ins Co 162 F3d 821 (4th Cir 1998)�� 10.111 Cook v Evanson 920 P2d 1223 (Wash Ct App 1996)�� 10.115 Cornell v TV Div Corp 17 NYS2d 69 (NY 1966)��5.41 Country-Wide Ins Co v Excelsior Ins Co 46 NYS3d 96 (App Div 2017)��10.23 County of Broome v Aetna Casualty & Surety Co 540 NYS2d 620 (App Div 1989)�� 7.25, 7.28 County of Broome v Travelers Indem Co 88 AD2d 720 (NY App Div 1982), aff’d, 58 NY2d 753 (NY 1982)��10.56 County of Columbia v Continental Ins Co 83 NY2d 618, 628 (NY 1994)��4.13 County of Fulton v United States Fidelity & Guar Co 600 NYS2d 972 (App Div 1993)��10.09 Courtney v New York City Ins Co 28 Barb 116 (NY Sup Ct 1858)��11.66 County of Niagara v Netherlands Ins Co 839 F App’x 581, (2d Cir 8 Dec 2020) (unpublished decision)��13.03 Cronk v State of New York 420 NYS2d 113 (Ct Cl 1979)��3.27 Crown, Cork & Seal Co, Inc v Aetna Cas & Sur Co No 1292, slip op (Pa Ct CP 2 Aug 1983)��1.10 Curanovic v NY Central Mut Fire Ins Co 762 NYS2d 148 (App Div 2003)��10.25, 12.23 Cutrone v American General Life Ins Co of New York 606 NYS2d 491 (App Div 1993)�� 12.06, 12.19, 12.22 Damanti v A/S Inger 314 F2d 395 (2d Cir 1963)��5.26 Danaher Corp v Travelers Indem Co 414 F Supp 3d 436 (SDNY 2019)��5.57 Daubert v Merrell Dow Pharmaceuticals Inc 509 US 579 (1993)��15.42 Dayton Indep Sch Dist v National Gypsum Co 682 F Supp 1403 (ED Tex 1988), rev’d on other grounds, sub nom WR Grace & Co v Continental Cas Co 896 F2d 865 (5th Cir 1990)�� 1.05, 1.15, 5.26, 10.81, 10.83 De Feo v Merchant 454 NYS2d 576 (NY City Ct 1982)�� 11.112 DeForte v Allstate Ins Co 442 NYS2d 307 (App Div 1981)��4.17 Denburg v Parker Chapin Flattau & Klimpl 624 NE2d 995 (NY 1993)��5.31 Deni Assocs of Fla Inc v State Farm Fire & Cas Ins Co 711 So 2d 1135 (Fla 1998)�� 10.109 Designcraft Jewel Indus Inc v St Paul Fire & Marine Ins Co 59 AD2d 857, 399 NYS2d 225 (NY App Div 1977), aff’d, 46 NY2d 796 (NY 1978)��12.08 Deutsche Bank Nat’l Trust Co v Quicken Loans Inc 810 Fd 861 (2d Cir 2015)��11.52 Dimambro-Northend Assocs v United Constr Inc 397 NW 2d 547 (Mich Ct App 1986)��5.16 Domtar Inc v Niagara Fire Ins Co 563 NW2d 724 (Minn 1997)��7.25 Don Clark, Inc v US Fidelity & Guar Co 545 NYS2d 968 (Sup Ct 1989)��10.46 Donaldson v Urban Lead Interests, Inc 564 NW2d 728 732 (Wis 1997)�� 10.118

Table of Cases

xxxiii

Do-Re-Knit Inc v National Union Fire Ins Co of Pittsburgh Pa 491 F Supp 1334 (EDNY 1980)��11.39 Drennan v Sun Indemnity Co of New York 271 NY 182 (NY 1936)��13.25, 13.27, 13.29 Dryden Mut Ins Co v Brockman 687 NYS2d 504 (App Div 1999)��7.27 Dynatec Consulting, Inc v Burlington Ins Co No 655241, 2019 WL 199768 (Sup Ct 7 January 2019), aff’d as modified 184 AD3d 475 (NY App Div 2020)��13.03 Eagle-Picher Indus, Inc v Liberty Mut Ins Co 523 F Supp 110 (D Mass 1981), 682 F2d 12 (1st Cir 1982), cert denied 460 US 1028 (1983)�� 1.10, 1.11 Eagle-Picher Indus, Inc v Liberty Mut Ins Co 829 F2d 227, 232 (1st Cir 1987)�� 1.10, 1.11 Eastern Baby Stores Inc v Central Mut Ins Co No 08-3368-cv, 337 F App’x 10 (2d Cir 2009)��8.10 ECDC Envt’l LC v New York Marine & Gen Ins Co No 96 Civ 6033, 1999 WL 777883 (SDNY 1999)��5.16 Egiazaryan v Zalmayev 290 FRD 421 (SDNY 2013)��16.55 Elat Inc v Aetna Casualty & Surety Co 654 A2d 503 (NJ Super Ct App Div 1995)��11.66 Eli Lilly & Co v Home Ins Co 482 NE2d 467 (Ind 1985)��1.15 Eljer Manufacturing Inc v Liberty Mutual Insurance Co 972 F2d 805 (7th Cir 1992)�� 5.58, 5.59, 5.61, 10.88, 10.89 El Paso Corp v La Comision Ejecutiva Hidro-Elecctrica Del Rio Lempa 341 F App’x 41 (5th Cir 6 August 2009)��16.98 Endicott Johnson Corp v Liberty Mut Ins Co 928 F Supp 176 (NDNY 1996)�� 6.34, 6.35 Ennar Latex Inc v Atlantic Mut Ins Co No 94 Civ 150, 1995 WL 325640 (SDNY 30 May 1995)��10.94, 10.99 Entron, Inc v Affiliated FM Ins Co 749 F2d 127 (2d Cir 1984)��17.18 Equitable Life Assurance Soc of US v O’Neil 413 NYS2d 714 (App Div 1979)��12.25 Erie Railroad Co v Tompkins 304 US 64 (1938)��1.10, 17.21, A.15 Esicorp Inc v Liberty Mutual Ins Co 266 F3d 859 (8th Cir 2001)��5.61 ESPN Inc v Office of Commissioner of Baseball 76 F Supp 2d 383 (SDNY 1999)��12.34 Essex House v St Paul Fire & Marine Ins Co 404 F Supp 978 (SD Ohio 1975), vacated in part on other grounds, Mirasco Inc v Am Nat’l Fire Ins Co 144 F App’x 171 (2d Cir 2005) (unpublished)��10.17 Evanston Ins Co v GAB Bus Serv 521 NYS2d 692 (App Div 1987)��3.27 Excel Graphics Techs Inc v CFG/AGSCB 75 Ninth Avenue LLC 767 NYS2d 99 (App Div 2003)��13.27 Exeter Bldg Corp v Scottsdale Ins Co 913 NYS2d 733 (NY App Div 2010)��10.78 Fabozzi v Lexington Ins Co 601 F3d 88 (2d Cir 2010)��11.42, 11.46 Federal Ins Co v Kozlowski 795 NYS2d 397 (App Div 2005)��5.22, 12.30 Federal Ins Co v Tyco Int’l Ltd 784 NYS2d 920 (Sup Ct 2004), 792 NYS2d 497 (App Div 2005)��5.22, 12.42 Feldman v Friedman 661 NYS2d 9 (App Div 1997)�� 12.19, 12.23, 12.32

xxxiv

Table of Cases

Fin Guar Ins Co v Putnam Advisory Co LLC No 2020 WL 264146 (SDNY 17 January 2020)��12.41 Firemen’s Ins Co v Nat’l Union Fire Ins Co of Pittsburgh, Pa 904 A2d 754 (NJ App Div 2006)��10.78 First Fin Ins Co v Allstate Interior Demolition Corp 193 F3d 109 (2d Cir 1999)��3.32, 12.05, 12.19, 12.23, 12.32 First Inv’rs Corp v Liberty Mut Ins Co 955 F Supp 274 (SDNY 1997), aff ’d 153 F3d 162 (2d Cir 1998)��5.18 First Penn Banking & Trust Co v United States Life Ins Co 421 F2d 959 (3rd Cir 1969)��13.10 Fisher v American Family Mut Ins Co 579 NW2d 599 (ND 1998)��5.06 Flat Ridge 2 Wind Energy LLC v Those Underwriter at Lloyd’s No 650508/2014, 2014 WL 3373435 (NY Sup Ct 30 June 2014)�� 11.46, 11.50, 11.52 Fluor Corp v Superior Ct of Orange Cty 354 P3d 302 (Cal 2015)��11.71, 11.72 Forest City Dillon Inc v Aetna Cas & Sur Co 852 F2d 168 (6th Cir 1988)��10.81 Frazier v Royal Ins Co 110 F Supp 2d 110 (NDNY 2000)��13.05, 13.24 Friedman v Prudential Life Ins Co 589 F Supp 1017 (SDNY 1984)��12.32 Frontier Insulation Contractors Inc v Merchants Mutual Insurance Co 667 NYS2d 982 (NY 1997)�� 10.124 Fulton Boiler Works Inc v Am Motorists Ins Co 828 F Supp 2d 481, 489 (NDNY 2011)��5.57 Garcia v Abrams 471 NYS2d 161 (App Div 1984)��11.21 Gardner v Ryder Truck Rental Inc 690 NYS2d 614 (App Div 1999)��13.05 Garfield Slope Housing Corp v Public Serv Mut Ins Co 973 F Supp 326 (EDNY 1997)�� 10.110 Gaston County Dyeing Mach Corp v Northfield Ins Co 524 SE2d 558 (NC 2000)�� 10.111 Geer v Union Mutual Life Insurance Co 273 NY 261, 7 NE2d 125 (NY 1937)�� 12.17, 12.19, 12.20, 12.21, 12.22 General Accident Fire & Life Assurance Co v Bongiorno 161 NYS2d 551 (Sup Ct 1957), aff ’d, 177 NYS2d 1019 (App Div 1958), aff ’d, 201 NYS2d 778 (NY 1960)��13.27 General Accident Ins Co of Am v Manchester 497 NYS2d 180 (App Div 1986)�� 7.24, 7.48 General Accident Ins Group v Cirucci 414 NYS2d 512 (1979)��13.16 George A Fuller Co v US Fidelity & Guar 200 AD2d 255, 613 NYS2d 152 (NY App Div 1994)��10.76, 10.78 Gerrish Corp v Universal Underwriters Ins Co 947 F2d 1023 (2d Cir 1991)��10.46, 10.49 Gilbert Frank Corp v Federal Ins Co 525 NYS2d 793 (NY 1988)��13.07 Gittelson v Mutual Life Ins Co of NY 41 NYS2d 478 (App Div 1943)��4.11, 10.04 Giuliani v Metropolitan Life Insurance Co 56 NYS2d 475 (App Div 1945)�� 12.17, 12.20, 12.22 Glatt v Union Cent Life Ins Co No 92 Civ 1227 (SWK), 1994 WL 329985 (SDNY 11 July 1994)��12.15 Glickman v New York Life Ins Co 291 NY 45, 50 NE2d 538, 148 ALR 454��12.17

Table of Cases

xxxv

Globecon Group LLC v Hartford Fire Insurance Co No 03-civ-0023, 2004 US Dist Lexis 13130 (SDNY 13 July 2004), aff ’d in part, vacated in part, remanded 434 F2d 165 (2d Cir 2006)��11.66, 11.73 Go Medical Indus Pty Ltd v CR Bard Inc No 3:95 MC 522, 1998 WL 1632525 (D Conn 1998)��16.57 Goldberg v Manufacturesrs Life Ins Co 672 NYS2d 39 (App Div 1998)��12.41 Grace (WR) & Co v Continental Cas Co 896 F2d 865 (5th Cir 1990)��1.05 Granite Ridge Energy LLC v Allianz Global Risk US Ins Co 979 F Supp 2d 385 (SDNY 2013)��17.19, 17.20 Granite State Ins Co v R&Q Reins Co No 654494/2013, 2015 NY Misc Lexis 2616 (Sup Ct 21 July 2015)��16.55 Great Am Ins Co v Houlihan Lawrence Inc 449 F Supp 3d 354 (SDNY 2020)��10.23, 10.24 Great Am Ins Co v Lerman Motors Inc 491 A2d 729, (NJ Super Ct App Div 1984)��5.16 Great Am Restoration Servs Inc v Scottsdale Ins Co 78 AD 3d 773, 911 NYS2d 142 (App Div 2010), remanding, No 110122008, 2011 WL 11718178 (NY Sup Ct 11 Jan 2011)�� 10.112 Great Canal Realty Corp v Seneca Ins Co 800 NYS2d 521 (NY 2005)�� 8.12, 8.14 Great Northern Ins Co v Dayco Corp 637 F Supp 765 (SDNY 1986)�� 4.27, 5.49, 10.07, 10.14, 10.16, 10.17, 10.18 Greaves v Public Serv Mut Ins Co 155 NE2d 390, 5 NY2d 120 (NY 1959)��4.22, 10.05 Green v Santa Fe Industries Inc 576 F Supp 269 (SDNY 1983)��A.15 Green Door Realty Corp v TIG Ins Co 329 F3d 282 (2d Cir 2003)��8.12 Greene v United Mut Life Ins Co 238 NYS2d 809 (Sup Ct 1963), aff ’d, 258 NYS2d 323 (App Div 1965)��12.25 Greenhomes Am LLC v Farm Family Cas Ins Co 936 NYS2d 829 (App Div 2012)��11.66 Guaranty Trust Co of NY v York 326 US 99 (1995)��17.21 Gulf Ins Co v Transatlantic Reins Co 788 NYS2d 44 (App Div 2004)��16.53, 16.55 Gulf Mississippi Marine Corp v George Engine Co 697 F2d 668 (5th Cir 1983)��10.81, 10.83 Hahn Automotive Warehouse Inc v American Zurich Insurance Co 18 NY3d 765 (NY 2012)��11.41, 11.52 Hallmark Capital Corp, In re 534 F Supp 2d 951 (D Minn June 1, 2007)��16.98 Hallock v State of New York 474 NE2d 1178 (NY 1984)��5.31 Halpin v Insurance Co of N Am 120 NY 73 (NY 1890)��12.14 Hancock Labs, Inc v Admiral Ins Co 777 F2d 520 (9th Cir 1985)��1.10 Hanover Ins Co v Vermont Mut Ins Co 69 F Supp 3d 302 (NDNY 2014)��5.57 Hanwei Guo, Application and Petition of 965 F3d 96 (2d Cir 2020)��16.98 Harbor Insurance Co v Tishman Construction Co 578 NE2d 1197 (Ill App Ct 1991)��10.59 Hardware Mut Ins Co v Mason-Moore-Tracy, Inc 194 F2d 173 (2d Cir 1952)��10.56 Harford Cty v Harford Mut Ins Co 610 A2d 286 (Md 1992)��1.15 Harleysville Worcester Ins Co v Wesco Ins Co 314 F Supp 3d 534 (SDNY 2018), aff’d 722 F App’x 90, (2d Cir 2019)��8.17

xxxvi

Table of Cases

Harris v Allstate Ins Co 309 NY 72 (1955)��4.17 Hartford Accident & Indem Co v Village of Hempstead 422 NYS2d 47 (NY 1979)��3.23 Hartford Accident & Indem Co v Wesolowski 33 NY2d 169 (NY 1973)��4.13 Hartford Ins Co v County of Nassau 46 NY2d 1028, 416 NYS2d 539 (NY 1979)�� 13.03 Hartford Fire Ins Co v California 509 US 764 (1993)�� 1.02, 1.24, 10.29 Haslauer v North Country Adirondack Coop Ins Co 654 NYS2d 447 (App Div 1997)��8.20, 13.08 Hatco v WR Grace & Co 801 F Supp 1334 (DNJ 1992)��8.10 Hawthorne v South Bronx Community Corp 576 NYS2d 203 (NY App Div 1991)��5.05 Hay v Star Fire Ins Co 77 NY 235 (NY 1879)��11.42 Health-Chem Corp v National Union Fire Insurance Co of Pittsburgh 559 NYS2d 435 (Sup Ct 1990)��5.51 Hearn v Rhay 68 FRD 574 (1975)��16.24 Heiser v Union Cent Life Ins Co No 94CV179, 1995 WL 355612 (NDNY 1995)��13.18 Henkel Corp v Hartford Accident & Indemnity Co 62 P3d 69 (Cal 2003)�� 11.69, 11.70, 11.71, 11.72, 11.73 Herald Square Loft Corp v Merrimack Mut Fire Ins Co 344 F Supp 2d 915 (SDNY 2004)��10.11 High Point Design LLC v LM Ins Corp No 14-cv-787, 2016 WL 426594 (SDNY 3 Feb 2016)��5.51 Hoechst Celanese Corp v Certain Underwriters at Lloyd’s 673 A 2d 164 (Del 1996)��5.63 Hoechst Celanese Corp v National Fire Ins Co of Pittsburgh, Pa No 89C-SE- 35, 1994 WL 721642 (Del Super Ct 13 April 1994)��8.10 Hoechst Celanese Corp v National Union Fire Ins Co of Pittsburgh, Pa v Centaur Ins Co No 89C-SE-35, 1991 WL 190313 (Del Super Ct Sept 10, 1991)��11.88 Hoechst Celanese Corp v National Union Fire Insurance Co of Pittsburgh, Pa 623 A 2d 1118 (Del Super Ct 1992), No 89C-SE-35, 1994 Del Super LEXIS 571, 1994 WL 721633 (22 Apr 1994)�� 3.23, 7.23, 7.26, 7.31, 16.24 Holy Angels Academy v Hartford Ins Group 487 NYS2d 1005 (Sup Ct 1985)�� 10.133 Home Decor Furniture & Lighting Inc v United Nat’l Group No 05CV02005, 2006 WL 3694554 (EDNY 2006)��13.18 Home Insurance Co v American Home Products Corp 551 NYS2d 481 (NY 1990)��3.23 Home Insurance Co v American Insurance Co 537 NYS2d 516 (App Div 1989)�� 10.14, 10.21, 10.22, A.20 Home Insurance Co of Illinois (New Hampshire) v Spectrum Information Technologies Inc 930 F Supp 825 (EDNY 1996)�� 12.08, 12.10, 12.11, 12.14, 12.16, 12.23, 12.26, 12.27 Hopeman Bro Inc v Continental Cas Co 307 F Supp 3d 433 (ED Va. 2018)��5.57 Horne v Radiological Health Servs PC 371 NYS2d 948 (Sup Ct Spec Term 1975), aff ’d 379 NYS2d (App Div 1976)��13.14 HRH Construction Corp v Bethlehem Steel Corp 412 NYS2d 366 (NY 1978)��3.27

Table of Cases

xxxvii

HS Equities Inc v Hartford Accident & Indemnity Co 609 F2d 669 (2d Cir 1979)��5.37 Hyatt Corp v Occidental Fire & Cas Co 801 SW2d 382 (Mo Ct App 1990)��5.19 I Q Originals, Inc v Boston Old Colony Ins Co 447 NYS2d 174 (App Div), aff ’d 458 NYS2d 540 (NY 1982)��4.24 IBM Poughkeepsie Employees Federal Credit Union v Cumis Ins Soc’y Inc 590 F Supp 769 (SDNY 1984)��4.09 Idle Aire Techs Corp, Re No 08-10960 (KG), 2009 Bankr Lexis 353 (Bankr D Del 2009)�� 10.115 IJ White Corp v Columbia Cas Co 105 AD3d 531 (NY App Div 2013)��10.76 Imperial Cas & Indem Co v High Concrete Structures Inc 858 F2d 128 (3d Cir 1988)��10.81 In the Matter of an Arbitration between Sec Life Ins Co of Am 228 F3d 856 (8th Cir 2000)�� 16.101 Incorporated Village of Cedarhurst v Hanover Ins Co 653 NYS2d 68 (NY 1996)�� 10.110 Independent Petrochemical Corp v Aetna Cas & Sur Co 654 F Supp 1334 (DDC 1986), aff ’d in part, rev’d in part on other grounds 944 F2d 940 (DC Cir 1991)��1.15 Indian Harbor Ins Co v City of San Diego 972 F Supp 2d 634 (SDNY 2013), aff’d, 586 F App’x 726 (2d Cir 2014)��8.20 Insurance Antitrust Litigation 938 F2d 919 (9th Cir 1991), aff ’d in part, rev’d in part and modified sub nom Hartford Fire Insurance Co v California 509 US 764 (1993)�� 1.02, 1.24, 10.29 Insurance Co v Wilkinson 80 US 222 (Supreme Ct)��13.21 Insurance Co of N Am v Forty-Eight Insulations, Inc 451 F Supp 1230 (ED Mich 1978), aff’d 633 F2d 1212 (6th Cir 1980), clarified 657 F2d 814 (6th Cir 1981), cert denied 454 US 1109 (1981)�� 1.10, 1.11, 1.13 Insurance Co of N Am v Lindsey 372 NYS2d 164 (Sup Ct 1975)��13.08 Intel Corp v Advanced Micro Devices Inc 542 US 241 (2004)��16.98 Intel Corp v Hartford Accident & Indemnity Co 692 F Supp 1171 (ND Cal 1988), aff ’d in part, rev’d on other grounds, 952 F2d 1551 (9th Cir 1991)��10.47 Intelligent Digital Sys LLC v Beazley Ins Co 906 F Supp 2d 80 (EDNY 2012)��13.23 International Ins Co v Newmont Mining Co 800 F Supp 1195 (SDNY 1992)��16.55 International Flavors & Fragrances Inc v Royal Insurance Co of Am 844 NYS2d 257 (App Div 2007)��5.53 International Hormones Inc v Safeco Insurance Co of America 394 NYS2d 260 (App Div 1977)��10.69, 10.82 IQ Originals, Inc v Boston Old Colony Ins Co 447 NYS2d 174 (App Div), aff ’d 458 NYS2d 540 (NY 1982)��4.24 Isadore Rosen & Sons Inc v Security Mutual Insurance Company of New York 339 NYS2d 97 (NY 1972)�� 5.37, 5.38, 11.39 James v Allstate Ins Co 578 NYS2d 18 (App Div 1991)��8.10 Jenkins v Etlinger 447 NYS2d 696 (App Div 1982)��5.41 Jenkins v John Hancock Mut Life Ins Co 257 NY 289 (1931)��12.19 Johnson v Studyvin 839 F Supp 1490 (D Kan 1993)��5.16 Joint E & S Dist Asbestos Litig, In re 78 F3d 764 (2d Cir 1996)��5.37

xxxviii

Table of Cases

Jones Truck Lines v Transport Ins Co No Civ A 88-5723, 1989 US Dist Lexis 5092 (ED Pa 1989)��10.47 Joseph R Loring & Assocs Inc v Continental Cas Co 56 NY2d 848, 453 NYS2d 169 (NY 1982)��3.27 Journal Publ’g Co v Am Home Assurance Co 740 F Supp 1015 (SDNY 1990)��11.41 JP Morgan Chase & Co v Indian Harbor Ins Co 947 NYS2d 17 (App Div 2012)��16.53 JP Morgan Sec Inc v Vigilant Ins Co 58 NYS3d 38 (App Div 2017)��5.37 Juliano v Health Maintenance Org of NJ 221 F3d 279 (2d Cir 2000)��13.08 Kansas v Colorado 556 US 98, 129 S Ct 1294 (2009)��11.87 Kantor v Nationwide Life Ins Co 227 NYS2d 703 (Sup Ct 1962)��12.30 Katz Communications Inc v Evening News Ass’n 705 F2d 20 (2d Cir 1983)��5.41 Kaufman v Cohen 760 NYS2d 157 (App Div 2003)��12.41 Keck v Metropolitan Life Ins Co 264 NYS 892 (App Div 1933), aff ’d 264 NY 422 (1934)��12.19 Keene Corp v Insurance Co of North America 513 F Supp 47 (DDC 1981), rev’d 667 F2d 1034 (DC Cir 1981), cert denied 455 US 1007 (1982)�� 1.07, 1.10, 1.11, 1.12, 1.14, 1.15 Kenai Corp, In re 136 BR 59 (SDNY 1992)��5.22 Kent Farms Inc v Zurich Ins Co 969 P2d 109 (Wash Ct App 1998)�� 10.115 Keyspan Gas E Corp v Munich Reinsurance Am Inc 23 NY3d 583 (NY 2014)��13.04 Keyspan Gas E Corp v Munich Reinsurance Am Inc 31 NY3d 51, 73 NYS3d 113 (NY 2018)�� 1.13, 5.54 Kingsway Financial Services Inc v PriceWaterhouseCoopers LLP No 03-Civ-5560 (RMG) (HBP) 2008 WL 4452134 (SDNY 2008)��16.57 Klapper v Hanover Ins Co 240 NYS2d 284 (Sup Ct 1970)��10.56 Kroski v Long Island Sav Bank 689 NYS2d 92 (App Div 1999)��12.09, 12.13 Lac d’Amiante du Quebec, Ltee v American Home Assurance Co 613 F Supp 1549 (DNJ 1985), vacated as to one defendant on other grounds 864 F2d 1033 (3d Cir 1988)��1.10 Landshire Fast Foods of Milwaukee Inc v Employers Mut Cas Co 676 NW2d 528 (Wis Ct App 2004)�� 10.115 Langer v Metro Life Ins Co 290 NY 601 (NY 1943)��12.08 Laquila Construction Inc v Travelers Indemnity Co of Illinois 66 F Supp 2d 543 (SDNY 1999), aff ’d, 216 F3d 1072 (2d Cir 2000)��10.63 Lauder v First Unum Life Ins Co 284 F3d 375 (2d Cir 2002)��13.08 Lavandier v Landmark Ins Co 810 NYS2d 45 (App Div 2006)��6.34 Leamy v Berkshire Life Ins Co 383 NYS2d 564 (NY 1976)��12.03, 12.19 Lefrak Organisation Inc v Chubb Custom Ins Co 942 F Supp 949 (SDNY 1996)�� 10.108 Lenhard v Genesee Patrons Co-Op Ins Co 818 NYS2d 644 (App Div 2006)��12.23 Lewis v Ocean Accident & Guar Corp 224 NY 18 (1918)��4.11, 10.04 Liberty Mut Ins Co v Those Certain Underwriters at Lloyd’s, London 650 F Supp 1553 (WD Pa 1987)��11.41, 11.43 Life Receivables Trust v Syndicate 102 at Lloyd’s of London 549 F3d 210 (2d Cir 2008)�� 16.101 Linde Thomson Langworthy Kohn & Van Dyke PC v Resolution Trust Corp 5 F3d 1508 (DC Cir 1993)��16.56

Table of Cases

xxxix

Lindenbaum v Equitable Life Assurance Soc’y of the US 174 NYS2d 421 (App Div 1958)��12.25 Liquidation of Midland Ins Co, In re 171 AD3d 564 (NY App Div 2019)��5.54 Liquidation of Midland Ins Co, In re 269 AD2d 50 (NY App Div 2000)��9.31 Lissauer v Guidone Specialty Mut Ins 161 AD3d 974 (NY App Div 2018)��10.24 Liverpool and London & Globe Ins Co v Kearney 180 US 132 (1901)��4.22 Lone Star Indus Inc v Liberty Mut Ins Co 689 F Supp 329 (SDNY 1988)��13.19 Long Island Lighting Co v Allianz Underwriters Ins Co 805 NYS2d 74 (App Div 2005)��8.12 Lowenstein Dyes & Chems Inc v Aetna Life & Cas Co 524 F Supp 574 (EDNY 1981), aff ’d, 742 F2d 1437 (2d Cir 1983)��5.14, 10.70 Lowville Producer’s Dairy Co-Op Inc v American Motorists Ins Co 604 NYS2d 421 (App Div 1993)��10.75, 10.89 Lowy v Travelers Prop & Cas Co No 99 Civ 2727, 2000 US Dist Lexis 526702 (SDNY 2 May 2000)��5.22 Lugo v AIG Life Insurance Co 852 F Supp 187 (SDNY 1994)��13.18 Luria Brothers & Co v Alliance Assurance Co 780 F2d 1082 (2d Cir 1986)�� 3.30, 5.26, 5.27, 5.30, 5.31, 5.36, 5.37, 11.21, 11.39, 13.02, 13.10, 13.17 McCarthy v AMEX Assurance Co 636 NYS2d 475 (App Div 1996)��4.13 McCarthy v Olin Corp 119 F3d 148 (2d Cir 1997)��5.04 McDaniels v American Bankers Ins Co 643 NYS2d 846 (App Div 1996)��12.23 McGinniss v Employers Reinsurance Corp 648 F Supp 1263 (SDNY 1986)��5.22 McGrail v Equitable Life Assur Soc’y of the United States 55 NE2d 483 (NY 1944)��3.27 McGroarty v Great Am Ins Co 36 NY2d 358 (1975)��7.24 MacKinnon v Truck Insurance Exchange 73 P3d 1205 (Cal 2003)�� 10.110, 10.111, 10.113, 10.114 McNaught v Equitable Life Assurance Society of United States 136 AD 774, 121 NYS 447 (App Div 1910)�� 12.34, 12.39, 13.15 McNeilab v North River Ins Co 645 F Supp 525 (DNJ 1986), aff ’d, 831 F2d 287 (3d Cir 1987)��10.84 McPherson v Buick Motor Co 145 NYS 462 (App Div 1914), aff ’d 217 NY 382 (1916)��1.05 Mack-Cali Realty v NGM Ins Co 119 AD3d 905 (NY App Div 2014)��10.24 Madawick Contracting Co v Travelers Ins Co 307 NY111 (NY 1954)��4.13 Malloy’s Estate, In re 278 NY 429 (1938)��3.27 Mapes Indus Inc v United States Fidelity & Guar Co 560 NW2d 814 (Neb 1997)��10.69 Marino v New York Tel Co 944 F2d 109 (2d Cir 1991)��8.16 Marino v NewYork Tel Co No 88 Civ 5817, 1992 WL 212184 (SDNY 24 Aug 1992)��13.19 Maroney v NY Cent Mut Fire Ins Co 5NY3d 467 (NY App Div 2005)��10.24 Martin v City of Cohoes 371 NYS2d 687 (NY 1975)��3.27 Maryland Casualty Co v Wausau Chem Corp 809 F Supp 680 (WD Wis 1992)��10.49

xl

Table of Cases

Maryland Casualty Co v Williams 377 F3d 389��17.21 Maryland Casualty Co v WR Grace & Co 794 F Supp 1206 (SDNY 1991), 23 F3d 617 (2d Cir 1994), 128 F3d 794 (2d Cir 1997)�� 5.60, 5.61, 5.62, 5.64, 8.28, 10.63, 10.75, 10.82, 10.83 Massachusetts v Environmental Protection Agency 549 US 497 (2007)�� 10.118 Maxum Indem Co v A-One Testing Labs Inc 150 F Supp 3d 278 (SDNY 2015)��5.57 MBIA Inc v Fed Ins Co 652 F3d 152 (2d Cir 2011)��5.36, 13.08 Meadows Indemnity Co v Nutmeg Insurance Co 157 FRD 42 (MD Tenn 1994)�� 16.101 Meagher v Executive Life Ins Co 607 NYS2d 361 (App Div 1994)��12.22, 12.25 MCI Llc v Rutgers Cas Ins Co No 06 Civ 4412 (THK), 2007 US Dist Lexis 59241 (SDNY 2007)��13.16, 13.17 Medical Facilities Inc v Pryke 62 NY2d 116 (NY 1984)��11.43, 11.46 Medway Power Ltd, Re Application of 985 F Supp 402 (SDNY 1997)��16.98 Merchants Mutual Casualty Co v Wildman 209 NYS2d 242 (App Div 1960)��13.30 Merchants Mut Ins Co v Hoffman 452 NYS2d 398 (NY 1982)��8.12 Merriam v United States 107 US 437 (1883)��4.17 Messersmith v American Fidelity Co 133 NE 432, 232 NY 161 (1921)�� 4.13, 7.24, 7.54 Metro Prop & Cas Ins Co v Sarris No 1:15CV0780, 2017 WL 3252812 (NDNY 28 July 2017)��7.27 Meyer (A) & Sons Corp v Zurich Ins Grp 74 NY2d 298 (NY 1989)��4.13 MHR Capital Partners LP v Presstek Inc 12 NYS3d 640 (NY 2009)��11.41 Micha v Merchants Mut Ins Co 463 NYS2d 110 (App Div 1983)��11.46 Mighty Midgets Inc v Centennial Insurance Co 403 NYS2d 768 (App Div 1978), aff’d 416 NYS2d 559 (NY 1979)�� 8.12, 8.24, 8.25, 13.27 Miller v Continental Ins Co 40 NY2d 675 (NY 1976)��7.24 Miller v Shugart 316 NW 2d 729 (Minn 1982)��5.31 Mills v Everest Reinsurance Co 410 F Supp 2d 243 (SDNY 2006)��12.41 Minerva Enters Inc v Bituminous Cas Corp 851 SW2d 403 (Ark 1993)�� 10.111 Mirasco Inc v American Nat’l Fire Ins Co 144 Fed App’x 171 (2d Cir 2005)��5.49, 10.07, 10.17 Mitchell v New York Hospital 214 473 NYS2d 148 (NY 1984)��3.27 Molycorp Inc v Aetna Cas & Sur Co 431 NYS2d 824 (App Div 1980)��5.49, 10.18 Mooney v Nationwide Mut Ins Co 577 NYS2d 506 (App Div 1991)��12.30 Morales v Castlepoint Ins Co 4 NYS3d 297 (App Div 2015)��12.19 Morgan Stanley Group, Inc v New England Ins Co 36 F Supp 2d 605 (SDNY 1999), aff ’d in part, rev’d in part, 225 F3d 270 (2d Cir 2000)��5.32 Morris Park Contracting Corp v National Union Fire Ins Co of Pittsburgh, Pa 822 NYS2d 616 (App Div 2006)�� 8.11, 8.12, 8.14, 8.28 Morton Int’l Inc v General Accident Ins Co of Am 629 A2d 831 (NJ 1993)��10.29 Moshiko Inc v Seiger & Smith 529 NYS2d 284 (App Div), aff ’d 533 NYS2d 52 (NY 1988)��4.07 Mount Vernon Fire Ins Co v Creative Housing Ltd 645 NYS 433 (NY 1996)��10.23 Mount Vernon Fire Ins Co v Jones No CV 95 3295, 1997 WL 37033 (EDNY 1997)��10.23, 10.26

Table of Cases

xli

Mount Vernon Fire Ins Co v Rennie No 97 CV 2778, 1999 WL 33113 (EDNY 1999)��10.23, 10.26 Mount Vernon Fire Ins Co v William Monier No 95 Civ 0645, 1996 WL 447747 (SDNY 1996)��13.18 Mraz v Canadian Universal Ins Co 804 F2d 1325 (4th Cir 1986)��1.15 MRI Broadway Rental Inc v US Mineral Prods Co 92 NY2d 421 (NY 1998)��5.64 Mt Hawley Ins Co v Abraham Little Neck Dar Group 825 F Supp 3d 384 (EDNY 2011)��8.14 Mt McKinley Ins Co v Corning Inc 946 NYS2d 136 (App Div 2012)�� 6.33, 6.35 Munzer v St Paul Fire & Marine Ins Co 538 NYS2d 633 (App Div 1999)��7.48 Murnane Bldg Contractors Inc v Zurich Am Ins Co 966 NYS2d 486 (App Div 2013)��5.51 MutRede Houses Inc v Greater NY Mut Ins Co 611 NYS2d 550 (App Div 1994)��8.20 Mutual Redevelopment Houses, Inc v Greater NY Mut Ins Co 611 NYS2d 550 (App Div 1994)��13.08, 13.18 Myers v Cigna Prop & Cas Ins Co 935 F2d 551 (SDNY 1997)��11.42 Myers v Equitable Life Assurance Soc’y 401 NYS2d 325 (App Div 1978)��12.32 Nassau Trust Co v Montrose Concrete Prods Corp 56 NY2d 175 (NY 1982)��13.22 National Broadcasting Co v Bear Stearns & Co 165 F3d 184 (2d Cir 1999)��16.98 National Union Fire Ins Co of Pittsburgh Pa v CBI Indus Inc 907 SW2d 517 (Tex 1995)�� 10.109 National Union Fire Insurance Co of Pittsburgh, Pa v Turner Construction Co 117 AD3d 103 (NY App Div 2014)��10.78 Nationwide Mut Fire Ins Co v Pascarella 993 F Supp 134 (SDNY 1998)��12.23 Nationwide Mut Ins Co v Davis No 99 Civ 11928, 2000 WL 964939 (SDNY 11 July 2000)�� 8.10–8.11 Nature’s Plus Nordic A/S v Natural Organics, Inc 108 F Supp 3d 52 (EDNY 2015), aff ’d, 646 F App’x 25, (2d Cir 15 Apr 2016)��17.19, 17.21 Nav-Its Inc v Selective Ins Co 869 A2d 929 (NJ 2005)�� 10.111 Nesmith v Allstate Ins Co 24 NY3d 520 (NY 2014)��6.34 New Castle County v Continental Cas Co 725 F Supp 800 (D Del 1989), aff ’d in part, rev’d in part on other grounds, 933 F2d 1162 (3d Cir 1991)��10.48 New Hampshire Insurance Co v Jefferson Insurance Co of NY 624 NYS2d 392 (App Div 1995)��10.23 New York Cen Mut Life Ins Co v Wood 827 NYS2d 760 (App Div 2007)��7.28 New York Presbyterian Hosp v Allstate Ins Co 814 NYS2d 687 (App Div 2006)�� 11.112 New York Marine & General Ins Co v Tradeline (LLC) and Deepak Fertilisers and Petrochemicals Corp Ltd No 98 CIV 7840; 2000 WL 739567 (SDNY 7 June 2000), aff ’d in part, rev’d in part 266 F3d 112 (2d Cir 2001)��12.39 Niagara Frontier Transp Auth v Encon Underwriting Agency, Inc 586 NYS2d 53 (App Div 1992)��11.46, 11.50 NL Indus Inc v Painewebber Inc 720 F Supp 293 (SDNY 1989)��13.27 No Hero Enters BV v Loretta Howard Gallery Inc 20 F Supp 3d 421 (SDNY 2014)��11.50 North Am Philips Corp v Aetna Cas & Sur Co No 88C-JA-155, 1995 WL 628447 (22 Apr 1995)��8.10

xlii

Table of Cases

North River Ins Co v ACE Am Reins Co 361 F3d 134 (2d Cir 2004)��17.21 North River Ins Co v Columbia Cas Co No 90 Civ 2518, 1995 US Dist Lexis 53 (SDNY 1995)��16.55 North River Ins Co v Philadelphia Reins Corp 797 F Supp 363 (DNJ 1992), rev’d on other grounds 52 F3d 1194 (3d Cir 1995)��16.55 Northern Ins Co of NY v Allied Mut Ins Co 955 F2d 1353 (9th Cir 1992)��11.66, 11.69, 11.70 Northrop Corp v Triad Int’l Marketing SA No 03 Civ 1730 LAKGWG 842 F2d 1154 (9th Cir 2004)��17.21 Northville Indus Corp v Nat’l Union Fire Ins Co of Pittsburg, Pa 657 NYS2d 564 (NY 1997)��10.09 Nothhelfer v American Surety Co of NY 302 NY 910 (1951)��13.27 Nu-Way Envt’l Inc v Planet Ins Co No 95 Civ 573 (HB), 1997 WL 462010 (SNDY 1997)��5.22 NY Cty Data Entry Worker Prod Liab Litig, In re 616 NYS2d 424 (Sup Ct 1994)��5.31 O’Dowd v American Sur Co of NY 165 NYS2d 458 (NY 1957)��5.05 Ogden Corp v Travelers Indemnity Corp 681 F Supp 169 (SDNY 1988)��10.63, 10.74, 10.77 Olds-Olympic Inc v Commercial Union Insurance Co 918 P2d 923 (Wash 1996)��10.50 Olezeski v Fingers Lakes-Seneca Coop Ins Co 629 NYS2d 873 (App Div 1995)��12.25, 12.26 Olin Corp v Certain Underwriters at Lloyd’s London 468 F3d 120 (2d Cir 2006)��5.64 Olin Corp v Certain Underwriters at Lloyd’s London 347 F App’x 622 (2d Cir 2009)��5.64 Olin Corp v American Home Assur Co 704 F3d 89 (2d Cir 2012)��5.64 Olin Corp v Insurance Co of N Am 743 F Supp 1044 (SDNY 1990)�� 8.10, 8.11, 8.12, 8.13, 8.28 Olin Corp v Ins Co of N Am 762 F Supp 548 (SDNY 1991), aff’d 966 F2d 718 (2d Cir 1992)��7.23 Olin Corp v Lamorak Ins Co 332 F Supp 3d 818 (SDNY 2018), reconsideration denied, No 84cv1968, 2018 WL 4360775 (SDNY 17 Aug 2018)�� 7.23, 7.27 Open Space Inst v American Elec Power Co 582 F3d 309 (2d Cir 2009)�� 10.117 Operadora DB Mexico SA DE CV, Application of, In re No 6:09-cv-383-Orl-22GJK, 2009 WL 2423138 (MD Fla 4 August 2009)��16.98 O’Reilly v The Guardian Mutual Life Ins Co of NY 60 NY 169 (NY 1875)��11.43, 11.45 Outlet Embroidery Co v Derwent Mills 172 NE 462 (NY 1930)��4.18 Overson v US Fidelity & Guar Co 587 P2d 149 (Utah 1978)��10.57 Owens Illinois, Inc v Aetna Cas & Sur Co 597 F Supp 1515 (DDC 1984)��1.10 Owens-Corning Fiberglass Corp v American Centennial Ins Co 74 Ohio Misc 2d 183 (Ct Com Pl 1995)��7.48 Oxus Gold PLC, In re No Misc 06-82, 2006 WL 2927615 (DNJ 11 October 2006)��16.98 Oy Saimma Lines Logistics Ltd v Mozaica-New York Inc 193 FRD 87 (EDNY 2000)��17.21 Pac Ins Co v Burnet Title Inc 380 F3d 1061 (8th Cir 2004)��5.19

Table of Cases

xliii

Pan American World Airways, Inc v Aetna Casualty & Surety Co 505 F2d 989 (2d Cir 1974)��4.28, 10.07, 10.14, 10.15, 10.16, 10.94, 10.95, 10.96, 10.97, 10.98, 10.99, 10.100, A.20 Panasia Estates Inc v Hudson Ins Co 856 NYS2d 513 (2008)��5.39 Paramount Communications Inc v Gibraltar Cas Co 612 NYS2d 156 (App Div 1994)��8.28 Paramount Ins Co v Rosedale Gardens Inc 743 NYS2d 59 (App Div 2002)��8.11, 8.24, 8.29 Park-Ohio Indus Inc v Home Indem Co 975 F2d 1215 (6th Cir 1992)�� 10.115 Parker Prods Inc v Gulf Ins Co 486 SW2d 610 (Tex Ct App 1972), aff ’d, 498 SW2d 676 (Tex 1973)��10.83 Parkset Plumbing & Heating Corp v Reliance Insurance Co 448 NYS2d 739 (App Div 1982), supplemented by No 83 Civ 7451, 1991 WL 283547 (SDNY 1991), rev’d on other grounds, 23 F3d 617 (2d Cir 1993)��10.63 Parmar v Hermitage Ins Co 21 AD3d 538, 800 NYS2d 726 (App Div 2005)��12.23 Patch v Stanley Works 488 F2d 483 (2d Cir 1971)��17.18 Pavia v State Farm Mut Auto Ins Co 82 NY2d 445 (NY 1993)��5.39 Penn Mut Life Ins Co v Remling 702 NYS2d 375 (App Div 2000)��12.19 Pennsylvania Gen Ins Co v Austin Powder Co 510 NYS2d 67 (NY 1986)��11.60 Perry v Metropolitan Life Ins Co 153 NYS 459 (App Div 1915)��12.29 People ex rel New York C & H R RR v Walsh 105 NE 136 (NY 1914)��4.09 PepsiCo Inc v Continental Casualty Co 640 F Supp 656 (SDNY 1986)�� 5.22, 5.50 Pfeffer v Harleysville Grp Inc 502 F App’x 28 (2d Cir 2012)��8.25 Pfizer Inc v Stryker Corp 385 F Supp 2d 380 (SDNY 2005)�� 5.48, 5.51 Pfizer Inc Sec Litig, In re No 90 Civ 1260, 1993 WL 561125 (SDNY 23 December 1993)��16.56, 16.57 Piliero v Allstate Ins Co 209 NYS2d 90 (App Div 1960)��4.13 Pittsburgh Corning Corp v Travelers Indem Co No 84-3985 (ED Pa 21 Jan 1985), 1988 US Dist Lexis 2052, 1988 WL 5291 and 5301 (ED Pa 1988)�� 1.10, 1.15 Pittston Co Ultramar America Ltd v Allianz Insurance Co 124 F3d 508 (3d Cir 1999)��7.11 Pittway Corp v American Motorists Ins Co 370 NE2d 1271 (Ill App Ct 1977)��10.70, 10.72 Playtex Inc v Columbia Casualty Co No Civ A 88C-MR-233, 1993 WL 390469 (Del Super Ct 1993)��7.34 Popkin v Sec Mut Ins Co of NY 367 NYS2d 492 (App Div 1975)�� 10.133 Porter v American Optical Corp No. 75-2202, 1977 US Dist LEXIS 12735 (ED La 1977), 641 F2d 1128 (5th Cir 1981), cert denied 454 US 1109 (1981)��1.10 Prima Paint Corp v Flood & Conklin Mfg Co 388 US 396 (1967)��3.06 Process Plants Corp v Beneficial Nat’l Life Ins Co 385 NYS2d 308 (App Div 1976), aff ’d 397 NYS2d 1007 (NY 1977)��12.19, 12.20, 12.30, 12.32 Progressive Cas Inc Co v Featherstone Foods Inc No 18 Civ 7923 (AKH), 2019 WL 8333518 (SDNY 1 August 2019)��13.05, 13.16 Proserver Ins Co v Ryba 10 NY3d 635 (NY 2008)��8.16

xliv

Table of Cases

Prudential Prop & Cas Ins Co, Matter of (King) 604 NYS2d 136 (App Div 1993)��5.38, 11.39 Public Serv Elec & Gas Co v Certain Underwriters at Lloyd’s of London No 88-4811 (DNJ 30 September 1994)��7.48 Public Util Dist No 1 v International Ins Co 881 P 2d 1020 (Wash 1994)��5.31 Puritan Ins Co v Eagle SS Co 779 F2d 866 (2d Cir 1985)��12.05 QBE Ins Co v Adjo Contracting Corp 121 AD3d 1064 (NY App Div 2014)��10.76, 13.03, 13.16 Queen Insurance Co v Globe & Rutger Fire Insurance Co 263 US 487 (1924)��10.14 Ramirez v Allstate Ins Co 811 NYS2d 19 (App Div 2006)�� 6.34, 6.35 Ramlochan v Scottsdale Ins Co 55 NYS3d 369 (App Div 2017)��8.20 Randell Elec Inc v State of New York 150 AD2d 875 (NY App Div 1989)��5.31 Rapid-American Corp v Allstate Insurance Co No QDS-22236162, slip op (NY Sup Ct 30 Jul 1999)��7.27 RD Maidman Family LP v Scottsdale Ins Co 783 NYS2d 205 (Sup Ct 2004)��10.46 Red Panther Chem Co v Insurance Co of Pa 43 F3d 514 (10th Cir 1994)�� 10.111 Reese v Schneider 201 NYS2d 857 (Sup Ct 1960)��13.27, 13.28 Regent Ins Co v Holmes 835 F Supp 579 (D Kan 1993)�� 10.111 Rekemeyer v State Farm Mut Auto Ins Co 796 NYS2d 13 (NY 2005)��8.14 Reliance Ins Co v Nat’l Union Fire Ins Co 691 NYS2d 458 (App Div 1999)��10.38 Renfrew Ctr v Blue Cross & Blue Shield of Central NY Inc No 94 Civ 1527 (RSP/GJD), 1997 US Dist Lexis 5088 (NDNY 1997)��11.66 Republic Ins Co v Masters, Mates & Pilots Pension Plan 77 F3d 48 (2d Cir 1996)��12.32 Republic of Kazakhstan, Re Application of 168 F3d 880 (5th Cir 1999)��16.98 Reynolds Metal Co v Aetna Casualty & Surety Co 696 NYS2d 563 (App Div 1999)�� 8.12, 8.24, 8.26, 8.28 Rhinebeck Bicycle Shop, Inc v Sterling Ins Co 151 AD2d 122, 125 (NY App Div 1989)��2.24 Richardson v Nationwide Mutual Insurance Co 826 A2d 310 (DC 2003), vacated, rehearing en banc granted 832 A2d 752 (DC 2003), vacated due to settlement of the parties, 844 A2d 344 (DC 2004)�� 10.10, 10.111 Ris v National Union Fire Ins Co of Pittsburgh Pa No 86 Civ 9718, 1989 WL 76199 (SDNY 1989)��12.25 RJC Realty Holding Corp v Republic Franklin Ins Co 756 NYS2d 631 (App Div 2003), rev’d, 777 NYS2d 4 (NY 2004)��10.23 Rockland Exposition, Inc v Great American Assurance Co 746 F Supp 2d 528 (SDNY 2010), aff’d, 445 F App’x 387 (2d Cir 2011)��8.26 Roman Catholic Diocese of Brooklyn v National Union Fire Ins Co of Pittsburgh, Pa 21 NY3d 139, 969 NYS2d 808 (NY 2013)�� 1.13, 4.13, 6.33, 6.34, 13.03, 13.16 Roofers’ Joint Training, Apprentice & Educ Comm’n of W NY v General Accident Ins Co of Am 713 NYS2d 615 (App Div 2000)�� 10.08, 10.108 Rooney v Tyson 674 NYS2d 616 (NY 1998)��A.16 Rosenau v Idaho Mut Ben Ass’n 145 P2d 227 (Idaho 1944)��10.93 Royal Indem Co v Smith 173 SE2d 738 (Ga App 1970)��10.55

Table of Cases

xlv

Roz Trading Ltd, Re 469 F Supp 2d 1221 (ND Ga 2006)��16.98 Ruggerio v Aetna Life & Casualty Co 484 NYS2d 106 (App Div 1985)��10.23, 10.25 St Paul Fire & Marine Ins Co v Vigilant Ins Co 919 F2d 235 (4th Cir 1990)��9.31 St Paul Surplus Lines Ins Co v Life Fitness No C3-99-9980, 2001 WL 588949 (Minn Dist Ct 2001)��5.19 Safeco Ins Co v Munroe 527 P2d 64 (Mont 1974)��5.16 Safeway Stores Inc v National Union Fire Ins Co of Pittsburgh, Pa 64 F3d 1282 (9th Cir 1995)�� 5.47, 5.50 Saks v Nicosia Contracting Corp 625 NYS2d 758 (App Div 1995)��7.27 Same Day Delivery Ser Inc v Penn Star Ins Co 151 F Supp 3d 380 (SDNY 2015)��8.12, 8.13, 8.26 Savage v Sun Life Assur Co of Canada 57 F Supp 620 (WD La 1944)��10.93 Savoy Medical Supply Co v F&H Manufacturing Corp 776 F Supp 703 (EDNY 1991)��10.45 Scalia v Equitable Life Assurance Society of the United States 673 NYS2d 730 (App Div 1998)��12.35, 13.14 Schefler v Livestock & Cas Ins Co 44 AD2d 811 (NY App Div 1974)��5.38 Schirmer v Penkert 840 NYS2d 796 (App Div 2007)��12.23 Schumann v State of New York 610 NYS2d 987 (NY Ct Cl 1994)�� 10.108 Scottsdale Indem Co v Beckerman 992 NYS2d 117 (App Div 2014)��10.23, 10.24 Seaboard Surety Co v Gillette Co 486 NYS2d 873, 64 NY2d 304, 476 NE2d 272 (NY 1984)��10.05, 10.08 Securities & Exch Comm’n v Credit Bancorp Ltd 147 F Supp 2d 238 (SDNY 2001)�� 12.06, 12.30, 12.32, 12.36, 13.09, 13.10, 13.12, 13.13, 13.14, A.25 Security Mut Ins Co v Acker-Fitzsimons Corp 340 NYS2d 902 (NY 1972)�� 8.12, 8.14 Selective Ins Co of Am v County of Rensselaer 26 NY3d 649 (NY 2016)��4.13 Servidone Construction Corp v Security Insurance Co of Hartford 64 NY2d 419, 447 NE2d 441, 448 NYS2d 139 (NY 1985)�� 5.28, 5.32 Seward Park Housing Corp v Greater NY Mut Ins Co 43 AD3d 23 (App Div 2007)��11.39 Shalimar Contractors Inc v American States Ins Co 975 F Supp 1450 (MD Ala 1997), aff ’d, 158 F3d 588 (11th Cir 1998)�� 10.115 Shamah v Schweiger 21 F Supp 2d 208 (EDNY 1988)��17.22 Shell Oil Co v Winterthur Swiss Ins Co 15 Cal Rptr 2d 815 (Ct App 1993)�� 10.48, 10.49 Sherri v Nat’l Sur Co 243 NY 266 (NY 1926)��11.39 Show Car Speed Shop, Inc v United States Fidelity & Guar Co 596 NYS2d 608 (App Div 1993)��4.17 Sindell v Abbott Labs 607 P 2d 924 (Cal 1980)��6.04 Sirignano v Chicago Ins Co 192 F Supp 2d 199 (SDNY 2002)��13.11 Slabic v Hendrickson 556 NYS2d 236 (Sup Ct 1990)��9.31 Slayko v Security Mutual Insurance Co 98 NY2d 289 (NY 2002)��3.27 Slotkin v Citizens Cas Co of NY 614 F2d 301 (2d Cir 1980)��5.41 Smirlock Realty Corp v Title Guar Co 421 NYS2d 232 (App Div 1979), aff ’d as modified, 437 NYS2d 57 (NY 1981)��12.25 Smith v New York Cent Mut Fire Ins Co 785 NYS2d 776 (App Div 2004)��7.28

xlvi

Table of Cases

Smith v Zurich Gen Accidental & Liab Ins Co 105 NYS2d 713 (App Div 1951), aff ’d, 303 NY 948 (1952)��13.27 Smith Pontiac-GMC Truck Ctr Inc v Hartford Accident & Indem Co 599 NYS2d 308 (App Div 1993)��5.18 Sokolowski v Aetna Life & Cas Co 670 F Supp 1199 (SDNY 1987)��5.19 Sommer v Guardian Life Ins Co of Am 28 NY 508 (NY 1938)��12.08 Sonkin Assocs v Columbian Mut Life Ins Co 541 NYS2d 611 (App Div 1989)��12.22 Sony Fin Servs LLC v Multi Video Group Ltd No 03 Civ.1730 LAK GWG, 2005 WL 91310 (SDNY 2005)��17.21 Soto v State Farm Ins Co 613 NYS2d 352 (NY 1994)��3.23 SP Duggal Corp v Aetna Cas & Sur Co 181 AD2d 472 (App Div 1992)��4.13 Sphere Drake Ins Co v YL Realty Co 990 F Supp 240 (SDNY 1997)�� 10.110 Squibb (ER) & Sons, Inc v Accident & Cas Ins Co 241 F3d 154 (2d Cir 2001)��1.04, 1.05, 5.17 SR International Business Ins Co v World Trade Center Props LLC No 01 CIV 9291 (JSM) 2002 WL 1334821 (SDNY 2002)��16.57 SR International Business Ins Co v World Trade Ctr Props LLC 375 F Supp 2d 238 (SDNY 2005)��11.66 Stanbery v Aetna Life Ins Co 98 A2d 134 (NJ Super Law Div 1953)��10.93 Standard Asbestos Mfg & Insulating Co v Royal Indem Ins Co No CV-80-14909, slip op (Mo Cir Ct 3 Apr 1986)��1.15 State v Central Mut Fire Ins Co 542 NYS2d 402 (App Div 1989)��10.48, 10.49 State Farm Auto Ins Co v Blanco 617 NYS2d 898 (App Div 1994)��5.36 State Farm Mut Auto Ins Co v Del Pizzo 586 NYS2d 310 (App Div 1992)��5.36 State Farm Mut Auto Ins Co v Roberts 697 A 2d 667 (Vt 1997)��4.24 State Farm Mut Auto Ins Co (Callisto), Matter of 255 AD2d 876 (NY App Div 1998)��5.38, 11.39 State of New York v Amro Realty Corp 697 F Supp 99 (NDNY 1988), aff ’d in part and rev’d in part on other grounds 936 F2d 1420 (2d Cir 1991)��8.10, 12.06, 13.07, 13.08, 13.10, 13.17, 13.18 State of NY v NY Cent Mut Fire Ins Co 147 AD2d 77 (NY App Div 1989)��10.46 Steadfast Ins Co v Purdue Frederick Co No X08CV020191697S, 2006 WL 1149185 (Conn Super Ct 11 April 2006)��7.34 Steen v Niagara Fire Ins Co 89 NY 315 (NY 1882)��11.43 Stonewall Insurance Co v Asbestos Claims Management Corp No 86 Civ 9671, 1992 US Dist Lexis 7607, 73 F3d 1178, (2d Cir 1995), modified on denial of reh’g, 85 F3d 49 (2d Cir 1996)��1.13, 1.15, 5.22, 5.31, 5.33, 5.48, 5.53, 5.57, 5.61, 5.64, 7.07, 7.09, 7.11, 7.16, 7.17, 7.23, 7.32, 7.34, 7.39, 7.48, 7.64, 10.07, 10.62, 10.69, 10.75, 10.76, 10.81, 10.82, 10.83 Stonewall Insurance Co v National Gypsum Co No 86 Civ 9671, 1991 US Dist Lexis 19146 (SDNY 1991), 1992 US Dist Lexis 7607, 1992 WL 123144 (SDNY1992)�� 5.33, 7.39, 7.48

Table of Cases

xlvii

Stoney Run Co v Prudential LMI Commercial Ins Co 47 F3d 34 (2 Cir 1995)��10.108, 10.110, 10.111 Sturges Mfg Co v Utica Mut Ins Co 371 NYS2d 444 (NY 1975)�� 4.22, 5.59, 5.62, 5.63, 10.05, 10.70, 10.85 Sullins v Allstate Ins Co 667 A2d 617 (MD 1995)�� 10.111 Superintendent of Fin Servs of NY v Guar Ins Co No 450023/13, 2013 NY Misc LEXIS 2442, (Sup Ct 10 June 2013)��11.52 Sylvania Twp Bd of Trustees v Twin City Fire Ins Co No L-03-1075, 2004 WL 226115 (Ohio Ct App 2004)��5.19 Technicon Elecs Corp v American Home Assurance Co 554 NYS2d 531 (NY 1989)�� 10.110 Tennenbaum v Ins Corp of Ireland 579 NYS2d 351 (App Div 1992)��12.19, 12.22 TerraMatrix Inc v US Fire Ins Co 939 P2d 483 (Colo Ct App 1997)�� 10.115 Testa v Utica Fire Ins Co 610 NYS2d 85 (App Div 1994)��12.19 Texaco A/S of Denmark v Commercial Ins Co of Newark 160 F3d 124 (2d Cir 1998)�� 5.26, 5.37 Texaco A/S SA v Commercial Ins Co No 90 CIV 2722 (JFK), 1995 US Dist Lexis 15818 (SDNY 26 October 1995)��11.66 Texas E Transmission Corp, In re 870 F Supp 1293 (ED Pa July 9, 1992)��10.47 Theatre Guild Prods Inc v Ins Corp of Ireland 267 NYS2d 297, (App Div 1966), aff ’d, 19 NY2d 656 (NY 1967)��4.13 Thomas J Lipton Inc v Liberty Mutual Ins Co 34 NY2d 356, 357 NYS2d 705, 314 NE2d 37 (NY 1974)�� 4.13, 4.22, 5.05, 5.16, 5.59, 5.62, 5.63, 7.54, 10.05, 10.82, 10.83, 10.85 Thomas J Sette Construction Corp v National Fire Insurance Co of Hartford 196 NYS2d 144 (Sup Ct 1959)��13.27 Thrasher v US Liab Ins Co 225 NE2d 503 (NY 1967)��11.21 Throgs Neck Bagels Inc v GA Insurance Co of New York 671 NYS2d 66 (App Div 1998)�� 4.09, 4.11, 4.27, 10.04, 10.05, 10.07, 10.18 TIG Insurance Co v Town of Cheektowaga 142 F Supp 2d 343 (WDNY 2001)�� 13.04, 13.08, 13.17 Todd Shipyards Corp v Turbine Servs Inc 674 F2d 401 (5th Cir 1982)��10.68, 10.81, 10.83 Tortuga Cas Co v Nat’l Union Fire Ins Co of Pittsburgh, Pa 604 A2d 419 (Del 1991) (unpublished decision) (Table)��9.33, 11.88 Tower Ins Co of NY v Commissary Direct Inc 63 Misc 3d 1229(A) (Sup Ct 13 May 2019)�� 8.17, 8.18 Tower Ins Co of NY v Lin Hsin Long Co 50 AD3d 305, 855 NYS2d 75 (App Div 3 April 2008)��8.29 Town of Massena v Healthcare Underwriters Mut Ins Co 749 NYS2d 456 (App 2001)��3.23 Travelers Indem Co v Israel 354 F2d 488 (2d Cir 1965)��11.66 Travelers Indem Co v Northrop Grumman Corp 413 F Supp 3d 263 (SDNY 2019)��8.20

xlviii

Table of Cases

Travelers Indem Co v Northrop Grumman Corp 416 Supp 3d 290 (SDNY 2019), denying in part, granting in part, motions for reconsideration No 16 Civ 8778 (LBS), 2020 WL 1469550 (SDNY 26 Mar 2020)��8.20 Travelers Ins Co v Eljer Manufacturing, Inc 757 NE2d 481 (Ill 2000)�� 5.59, 5.61, 10.85 Travelers Ins Co v Stanton 645 NYS2d 948 (App Div 1996)��7.34 Trend Export Funding Corp v Foreign Credit Insurance Ass’n 670 F Supp 480 (SDNY 1987)��13.27 TRT/FTC Communications Inc v Insurance Co of Pa 847 F Supp 28 (D Del 1993), aff ’d 9 F3d 1541 (3d Cir 1993)��10.99 Truax & Hovey Ltd v Aetna Cas & Sur Co 504 NYS2d 934 (App Div 1986)��10.82 Tuminelli v First Unum Life Ins Co 648 NYS2d 967 (App Div 1996)��12.25 Turner Constr Co v American Mfrs Mut Ins Co No 01 Civ 2899, 2005 US Dist Lexis 7644, 2005 WL 1022429 (SDNY 29 April 2005)�� 8.11, 8.26 Unigard Sec Ins Co v North River Ins Co 584 NYS2d 290 (NY 1992)��8.14 Union Carbide Corp v Affiliated FM Ins Co 958 NYS2d 311 (Sup Ct 2010), aff ’d 955 NYS2d 572 (App Div 2012)�� 7.27, 7.64 Union Indem Ins Co of NY, In re 611 NYS2d 506 (App Div 1994), aff ’d 651 NYS2d 383 (1996)��12.32 Uniroyal Inc v Home Ins Co 707 F Supp 1368, (EDNY 1988)�� 1.05, 1.15, 3.30, 5.25, 5.26, 5.27, 5.28, 5.30, 5.31, 5.32, 5.36, 5.53, 5.57, 10.93 United Prop Inc v Home Ins Co 311 NW 2d 689 (Iowa Ct App 1981)��5.16 United Spec Ins Co v Premier Contracting of NY, Inc 375 F Supp 3d 389 (SDNY 2019)��13.05 United States v Brennan 938 F Supp 1111 (EDNY 1996); rev’d, 183 F3d 139 (2d Cir 1999)��4.07 United States v Gullo 672 F Supp 99 (WDNY 1987)��15.27 United States v Conservation Chem Co 653 F Supp 152 (WD Mo 1986)��10.47, 10.49 United States Fidelity & Guaranty Co v Mayor’s Jewelers of Pompano Inc 384 So 2d 256 (Fla Dist Ct App 1980)��10.71 United States Fidelity & Guaranty Co v Nevada Cement Co 561 P2d 1335 (Nev 1977)��10.72 US Fidelity & Guar Co v Wilkin Insulation Co 550 NE2d 1032 (Ill Ct App 1989)��5.33 United States Fidelity & Guaranty Co v Wilkinson Insulation Co 578 NE2d 926 (Ill 1991)��10.83 US Fire Insurance Co v NY Marine & General Insurance Co 706 NYS2d 377 (App Div 2000)��10.23 US Life Ins Co in City of NY v Blumenfeld 938 NYS2d 84 (App Div 2012)��12.36 US Gypsum Co v Admiral Ins Co 643 NE2d 1226 (Ill Ct App 1994)��5.31 US Liab Ins Co v Dehkterman No 00 CV 2273, 2002 WL 31780174 (EDNY 2002)��10.23 US Liab Ins Co v Farley 626 NYS2d 238 (App Div 1995)��5.57 US Underwriters Ins Co v 203–11 West 145th Street Realty Corp No 99 Civ 8880, 2001 WL 604060 (SDNY 2001)�� 10.23, 10.26, 11.21 US Underwriters Ins Co v City Club Hotel LLC 789 NYS2d 470 (NY 2004)��11.87 US Underwriters Ins Co v Val-Blue Corp 623 NYS2d 834 (NY 1995)��10.23

Table of Cases

xlix

US Underwriters Ins Co v Zeugma Corp No 97 Civ 8031, 1998 WL 633679 (SDNY 1998)��10.23, 10.26 Universal Cable Prods, LLC v Atlantic Spec Ins Co No 17-56672, 2019 WL 3049034 (9th Cir 12 July 2019)��10.93 Universal Cable Productions Inc v Atlantic Specialty Ins Co 929 F3d 1143 (9th Cir 12 July 2019)��10.93, 10.97 Utica City Nat’l Bank v Gunn 118 NE 607 (NY 1918)��4.18 Utica Mut Ins Co v Prudential Prop & Cas Ins Co 477 NYS2d 657 (App Div 1984)��7.17 Vale Chem Co v Hartford Accident & Indem Co 490 A 2d 896 (Pa Super Ct 1985), rev’d on other grounds 516 A 2d 684 (Pa 1986)��1.10 Vander Veer v Continental Cas Co 356 NYS2d 13 (NY 1974)��12.19, 12.20 Vanguard Ins Co v Polchlopek 275 NYS2d 515 (NY 1966)��11.21 Vari Builders Inc v United States Fidelity & Guar Co 523 A2d 549 (Del Super Ct 1986)��10.62, 10.64 Vella v Equitable Life Assurance Soc’y 887 F2d 388 (2d Cir 1989)�� 12.10, 12.11, 12.14 Verlus v Liberty Mut Ins Co No 14-cv-2493, 2015 WL 7170484 (SDNY 12 November 2015)��6.35 Vigilant Insurance Co v Bear Stearns Cos 855 NYS2d 45 (2008)��5.24, 11.36 Vigilant Ins Co v I.Techs Inc 696 NYS2d 596 (App Div 1998)��10.11 Viking Pump, In re 146 A3d 1046 (Del 2015)��A.16 Viking Pump, In re 27 NY3d 244 (NY 2016)�� 4.13, 5.54, 6.33, 7.55, A.16 Viking Pump, In re 33 NYS3d 118 (NY 2016)��1.13 Viking Pump Inc, In re 148 A3d 633 (Del 2016)��5.57 Viking Pump, Inc v Century Indem Co 2 P3d 76 (Del Ch 2009)��11.73 Viking Pump Inc v Century Indem Co No 10C-06-141, 2013 WL 7098824 (Del 31 October 2013)��5.57 Villa Charlotte Bronte Inc v Commercial Union Ins Co 487 NYS2d 314 (NY 1985)��11.18 Vista Eng’g Corp v Everest Indem Ins Co 170 AD 3d 915 (App Div 2019)��8.16, 13.04 Vitkus v Beatrice Co 127 F3d 936 (10th Cir 1997)��5.31 Wachtel v Equitable Life Assurance Society 284 NY 345 (NY 1935)��11.43 Wakefield Park Inc v Ram Mutual Ins Co 731 NW2d 154 (Minn Ct App 207)�� 10.109 Wallace v 600 Partners Co 86 NY2d 543 (NY 1995)��12.43 Waltuch v Conticommodity Servs Inc 88 F3d 87 (2d Circ 1996)��5.22 Watkins Glen Central School District v National Union Fire Insurance Co of Pittsburgh, Pa 732 NYS2d 70 (App Div 2001)��10.23, 10.26 Wausau Underwriters Ins Co v Old Republic Gen Ins Co 122 F Supp 3d 44 (SDNY 2015)��8.17 WE O’Neil Construction Co v National Union Fire Insurance Co of Pittsburgh Pa 721 F Supp 984 (ND Ill 1989)��10.60 Wechsler v Hunt Health Sys Ltd 330 F Supp 2d 383 (SDNY 2004)��17.21 Wedtech Corp v Federal Ins Co 740 F Supp 214 (SNDY 1990)��5.22 Weedo v Stone-E-Brick Inc 405 A 2d 788 (NJ 1979)�� 10.62, 10.63, 10.64 West v Palmetto State Life Ins Co 25 SE2d 475 (SC 1943)��10.93 West Am Ins Co v Tufco Flooring East Inc 409 SE2d 692 (NC App 1991)�� 10.111

l

Table of Cases

West, Weir & Bartel, Inc v Mary Carter Paint Co 307 NYS2d 449 (NY 1969)��4.19 Western Alliance Ins Co v Gill 686 NE2d 997 (Mass 1997)�� 10.111 Westview Assocs v Guaranty Nat’l Ins Co 95 NY2d 334, 717 NYS2d 75 (NY 2000)�� 4.13, 10.08, 10.110, 10.111 Weyerhauser Co v Aetna Cas & Sur Co 874 P2d 142 (Wash 1984)��7.11 White v City of NY 598 NYS2d 759 (NY 1993)��8.12 Williams v State 770 SW 2d 948 (Tex App Ct 1989)��15.27 Winnick v Equitable Life Assurance Soc of US 494 NYS2d 509 (App Div 1985)��12.25, 12.26 Wisconsin Dep’t of Corrections v Schacht 524 US 381 (1988)��A.14 Wisconsin Elec Power Co v California Union Ins Co 419 NW2d 255 (Wis Ct App 1987)��1.15 Wisconsin Label Corp v Northbrook Prop & Cas Ins Co 607 NW2d 276 (Wis 2000)��5.61 Wittner v IDS Ins Co of NY 466 NYS2d 480 (App Div 1983)��12.21, 12.23 Worcester Ins Co v Bettenhauser 95 NY2d 185, 188 (NY 2000)��13.03 Worldcom Inc Securities Litig, In re 354 F Supp 2d 455 (SDNY 2005)��5.22, 12.30, 12.42 World Trade Center Properties v Hartford Fire Insurance Co 345 F3d 154 (2d Cir 2003)��10.12 Wright v Brocket 571 NYS2d 660 (Sup Ct 1991)��15.27 Wright v Evanston Ins Co 788 NYS2d 416 (App Div 2005)��7.54 Wyoming Sawmills Inc v Transportation Insurance Co 578 P2d 1253 (Or 1978)��5.61 Yohannon v Keene Corp 924 F2d 1255 (3d Cir 1991)��17.21 York-Buffalo Motor Express v National Fire & Marine Insurance Co 294 NY 467 (1945)��5.52 Younis Bros & Co v Cigna Worldwide Ins Co 899 F Supp 1385 (ED Pa 1995), aff ’d 91 F3d 12 (3d Cir 1996), judgement enforced on remand 167 F Supp 2d 743 (ED Pa 2001)��10.99 Zandri Construction Co Inc v Firemen’s Ins Co of Newark 440 NYS2d 353 (App Div 1981), aff ’d, 446 NYS2d 45 (NY 1981)��10.63 Zeig v Massachusetts Bonding & Insurance Co 23 F2d 665 (2d Cir 1928)��9.27, 11.31 Zeldman v Mutual Life Insurance Co of New York 53 NYS2d 792 (App Div 1945)��12.35 Zielinski v Associated Mut Ins Co 629 NYS2d 894 (App Div 1995)��12.26 Zilkha v Mutual Life Ins Co of NY 732 NYS2d 51 (App Div 2001)��12.19, 12.22 Zurich Ins Co v Northbrook Excess & Surplus Ins Co 494 NE2d 634 (Ill App Ct 1986), aff ’d sub nom Zurich Ins Co v Raymark Indus, Inc 514 NE2d 150 (Ill 1987)��1.15 Zurich Ins Co v Shearson Lehman Hutton, Inc 618 NYS2d 609 (NY 1994)�� 3.20, 3.23 Zurich Ins Co v White 633 NYS2d 415 (App Div 1995)��7.27 ICSID Case No ARB/05/25 (available at http://icsid.worldbank.org/ICSID)��14.47

Table of Legislation European European Convention on Human Rights��14.38, A.09 Art 6��A.09 Art 8��16.28 Rome Convention on the Law Applicable to Contractual Obligations��3.07, 3.36, 11.90 Art 1(2)(d)��3.07 Art 8��11.90 Rome I Regulation, Regulation (EC) 593/2008, [2008] OJ L177/6�� 3.07, 3.18, 3.40, 11.90 Art 1(2)(e)��3.07 Art 7��3.18 Art 10��3.21 Art 11��11.90 Art 12(1)(c)�� 3.38, 3.40 Art 12(1)(d)��3.36 Art 18(1)��3.32 Art 20��3.20 International Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters 847 UNTS 231�� 16.98, 16.100 Art 1��16.98 UNCITRAL Model Law on International Commercial Arbitration��A.03 Art 16(1)��3.06 United Kingdom: Primary Legislation Arbitration Act 1950�� 3.01, 3.12, 14.27, 15.70(App 1) Arbitration Act 1975�� 3.01, 3.12, 15.70(App 1) s 1��1.28 Arbitration Act 1979�� 3.01, 3.12, 15.70(App 1)

lii

Table of Legislation

Arbitration Act 1996�� 1.28, 3.07, 3.13, 3.14, 3.16, 3.18, 3.31, 3.33, 3.36, 3.39, 14.08, 14.13, 14.15, 14.24, 14.25, 14.27, 14.28, 15.04, 15.05, 15.16, 15.24, 15.70(App 1), 15.70(App 2), 16.24, 16.99, 16.103, 17.04, 17.33, A.03, A.08, A.09 s 2(1)�� 3.12, 3.16 s 3��3.12 s 4��11.85, 17.08 s 4(1)��3.13 s 4(5)��17.12, 17.13 s 7��3.06 s 9��3.08 s 9(1)��14.28 s 13(1)��3.36 s 13(4)��3.36 s 16(1)��14.15 s 16(3)��14.28 s 16(5)�� 14.01, 14.15, 14.49 s 17��14.18 s 18��3.16, 14.18, 14.24, 14.26, 14.28 s 18(2)��14.28 s 19��3.16 s 21(5)��14.28 s 24��3.16, 14.38 s 24(1)��14.28, 14.38 s 25(3)��14.28 s 28(2)��14.28 s 30��3.07 s 31��3.07 s 32��3.16 ss 33–41��3.13 s 33��3.13, 14.34, 14.38 s 33(1)(b)��15.19 s 34��3.32, 11.82, 12.27, 15.02, 15.05, 15.09, 16.02, 16.26, 16.27, 16.28, 16.101 s 34(1)��3.14, 15.04, 15.43, 16.02 s 34(2)��3.14, 16.02 s 34(2)(a)�� 15.02, 16.101 s 34(2)(c)��15.02, 15.15 s 34(2)(d)�� 15.02, 16.02, 16.24, 16.26 s 34(2)(e)�� 15.02, 16.97, 16.106 s 34(2)(f)��15.02 s 34(2)(g)��15.02 s 34(2)(h)��15.02 s 37��17.39 s 37(2)��3.13 s 40��3.13

Table of Legislation

liii

s 43��3.13, 16.97 s 43(1)�� 16.103 s 44�� 16.97, 16.99, 16.100 s 44(2)(a)�� 16.99, 16.100 s 46��3.07, 3.18, 3.21, 3.28, 3.32, 4.04 s 46(1)��3.18 s 46(1)(b)��3.18 s 46(2)��3.19 s 46(3)��16.26 s 47��15.31 s 48(4)��3.43, 11.80 s 49��3.44, 17.01, 17.02, 17.08, 17.11 s 49(1)��17.02 s 49(2)��17.02, 17.08 s 49(3)�� 17.02, 17.03, 17.04, 17.05, 17.08, 17.10, 17.11, 17.13, 17.14, 17.15, 17.16, 17.17, 17.23, 17.24, 17.25 s 49(3)(a)�� 17.02, 17.26, 17.40 s 49(3)(b)��17.02 s 49(4)�� 17.02, 17.26, 17.40 s 49(5)��17.02 s 49(6)��17.02 s 50��11.83 s 52(4)��15.69 s 56��3.13 s 56(2)��14.28 s 57��15.70 s 59(1)(a)��17.27 s 59(1)(c)��17.27 s 60��3.13, 3.45, 11.87, 15.22, 17.31 s 61��17.28 s 61(2)��17.32, 17.34 s 63(3)��17.39 s 63(4)��14.28 s 63(5)(a)��17.38 s 63(5)(b)��17.38 s 64��17.41 s 67��3.07, 3.16, 11.85, 15.70 s 67(1)��14.28 s 68��3.16, 11.85, 14.38, 15.69, 15.70, 17.11 s 68(1)��14.28 s 69�� 1.28, 3.16, 15.70 s 69(1)��3.16, 11.84 s 82(1)��3.16 s 93��14.25 s 108��A.03 Sch 1��3.13, 11.85

liv

Table of Legislation

Contracts (Applicable Law) Act 1990��3.07 Sch 1 Article 8��3.21 Article 10��3.21 Article 10 (1)(c)��3.38 Article 10 (1)(d)��3.36 Evidence (Proceedings in Other Jurisdictions) Act 1975��16.98 Foreign Limitation Periods Act 1984��3.37 Human Rights Act 1998��A.09 Insurance Act 2015��12.01, A.08, A.27 s 13A��17.04 Late Payment of Commercial Debts (Interest) Act 1988��17.04 Law Reform (Miscellaneous Provisions) (Scotland) Act 1990��A.03 Marine Insurance Act 1906��A.08 Senior Courts Act 1981��17.03 s 35A��17.13 United Kingdom: Statutory Instruments Civil Procedure Rules 1998 (SI 1998/3132)��16.10 Pt 18 – Further Information�� 16.105 Pt 31 – Disclosure and Inspection of Documents��16.10 Part 31 PD��16.10 r 31.6��16.10 Pt 34 – Witnesses, Depositions and Evidence for Foreign Courts r 34.8��16.96 r 34.13��16.99 Pt 35 – Experts and Assessors Pt 35 PD��15.47 r 35.10��15.50 Pt 44 – General Rules about Costs r 44.2(2)(a)��17.33 Pt 51 – Transitional Arrangements and Pilot Schemes Pt 51 PD 51U – Disclosure Pilot for the Business And Property Courts��16.12 Pt 62 – Arbitration Claims��14.28 r 62.2(1)��14.28 r 62.5��3.12 r 62.5(1)(b)�� 16.100 Law Applicable to Contractual Obligations and Non Contractual Obligations (Amendment etc) (EU Exit) Regulations 2019 (SI 2019/834)��3.07, 11.90 Rules of the Supreme Court 1965 (SI 1965/1776) Ord 62, r 3��17.33 UKSC Practice Direction 6 para 6.3.4��A.10

Table of Legislation

lv

United States of America California Insurance Code, §520��11.71, 11.72 Cal Water Code, §102��10.49 Clayton Act, Pub L No 36-212, 38 Stat 730 (15 Oct 1914), codified at 15 USC §§ 12–27a et seq�� 10.133 Clean Air Act 42 U.S.C. §7401 et seq (1970)�� 10.118 Comprehensive Environmental Response, Compensation & Liability Act (CERCLA or “Superfund”), Pub L No 96-510, 94 Stat 2767 (11 Dec 1980), codified at 42 USC §§ 9601 et seq��1.05, 10.44 Employee Retirement Income Security Act 1974 (ERISA), Pub L No 93-406, 88 Stat 829 (2 Sept 1974), codified at 29 USC §§ 1001 et seq��3.20, 10.126 Federal Arbitration Act, Pub L No 80-282, 61 Stat 669 (30 July 1947), codified at 9 USC §§ 1–14�� 3.06–3.07, 16.101, A.17 § 2��3.07 § 3��3.07 § 7�� 16.101 §§ 201–216��3.06 § 201��3.07 § 206��3.08 § 216��3.07 Federal Employers’ Liability Act (FELA), Pub L No. 56-219, 34 Stat 232 (11 June 1906), codified at 45 USC §§ 51–60��10.35 Federal Rules of Civil Procedure, Pub L No 97-462, 96 Stat 2527 (12 Jan 1983), codified at United States Code tit 28��15.11, 16.14 r 26��15.49 r 26(b)��16.11 r 26(b)(1)��16.11 r 26(b)(4)��15.49 r 26(b)(4)(B)��15.49 r 56��12.09 Federal Trade Commerce Act, Pub L No 63-203, 38 Stat 717 (26 Sept 1914), codified at 15 USC §§ 41–58�� 10.133 Hart-Scott-Rodino Antitrust Improvement Act, Pub L No 94-435, 90 Stat 1383 (30 Sept 1976), codified at 15 USC §§ 15c–15h, 18a, 66�� 10.133 Investment Advisers Act 1940�� 10.132 Investment Company Act 1940�� 10.132 Jones Act, Pub L No 66-261, 41 Stat 988 (5 June 1920), codified at 46 USC §§ 861 et seq��10.35 Lanham Patent Act, Pub L No 76-287, 53 Stat 1212 (5 Aug 1939), codified at 35 USC §§ 134, 135, 141, 145, 146�� 10.133 Longshoremen’s and Harbor Workers’ Compensation Act, Pub L No 69-803, 44 Stat 1424 (4 Mar 1927), codified at 33 USC §§ 901 et seq��10.35

lvi

Table of Legislation

New York Civil Practice Law and Rules (CPLR)�� 17.08, 17.09, 17.17, 17.18, 17.19, 17.22 § 101��17.06, 17.18 § 213��12.40 § 213(1)��12.40 § 213(2)��11.46 § 213(6)��12.43 § 213(8)��12.40 § 3101(a)(1)��15.49 § 3101(d)(2)��15.49 § 3212��12.09 § 5001�� 17.01, 17.06, 17.08, 17.19 § 5001(a)��17.06, 17.21 § 5001(b)��17.06, 17.25 § 5001(c)��17.06 § 5002�� 17.01, 17.06, 17.08 § 5003�� 17.01, 17.06, 17.08 § 5004��3.44, 17.01, 17.05, 17.06, 17.08, 17.23 New York Constitution Art 6, §§ 1–17��A.12 Art 6, § 7��A.12 New York Insurance Law 1909��12.22 § 59��12.17 New York Insurance Law 1940��12.17, 12.19 § 149�� 12.17, 12.20, 12.22 § 149(2)��12.17 § 149(3)��12.17 § 167(8)��13.05 New York Insurance Law�� 3.01, 3.22, 3.23, 4.01 § 1101��7.11 § 3105�� 12.04, 12.07, 12.08, 12.12, 12.16, 12.17, 12.19, 12.20, 12.21, 12.22, 12.31 § 3105(a)�� 12.03, 12.08, 12.12 § 3105(b)�� 12.04, 12.08, 12.12, 12.16, 12.17, 12.19, 12.20, 12.28, 12.30 § 3105(c)�� 12.04, 12.08, 12.12, 12.17, 12.23, 12.27 § 3105(d)��12.08, 12.12 § 3420�� 8.15, 8.16, 8.17, 8.19, 13.04 § 3420(a)�� 8.15, 8.17, 8.20 § 3420(a)(5)�� 8.15, 8.17, 8.19 § 3420(c)(2)(A), (B)��8.15 § 3420(d)��8.16, 13.03, 13.04, 13.05, 13.16 § 3420(d)(2)��13.03 New York Navigation Law §§170, 172(12), (18)��10.49 New York Sess Laws 1088 (McKinney) 2008 s 8��8.16

Table of Legislation

lvii

Public Utility Holding Company Act 1935�� 10.132 Robinson-Patman Act 10.84, Pub L No 74-692, 49 Stat 1526 (19 June 1936), codified at 15 USC §§ 13a, 13b, 21a�� 10.133 Securities Act 1933�� 10.132 Securities Exchange Act 1934�� 10.132 Sherman Act, 26 Stat 209, ch 647 (2 July 1890), codified at 15 USC §§ 107 et seq�� 10.133 Trust Indenture Act 1939�� 10.132 United States Code (28 USC) § 41��A.13 § 112��A.13 § 1254��A.13 § 1257��A.12 § 1332��A.14 § 1332(c)(1)��A.14 § 1407��14.46 § 1782�� 16.98, 16.101, 16.104 § 2122��14.46 § 2201��14.33 United States Constitution Seventh Amendment��A.25

lviii

1 The Legal and Economic Origins of the Bermuda Form POLICY FORMS AND LIABILITY PROBLEMS In the mid-1980s the excess casualty insurance market in the United States collapsed. This was due to a number of reasons, including the impact of liability claims, the cyclical ‘boom’ and ‘bust’ nature of the insurance industry, and poor investment results in the early 1980s.1 A number of insurance companies rose from the ashes, fuelled in part by capital from large US manufacturing companies which, as policyholders, needed access to high-level excess casualty cover and thus sought to help stabilise capacity in the excess casualty insurance market. The effort to create this ‘alternative’ excess liability insurance market was spearheaded by insurance brokerage Marsh & McLennan and bankers JP Morgan. These two companies created ACE Insurance Company Ltd (ACE) in 1985 to provide excess liability catastrophe coverage at layers in excess of US$100 million. Due to ACE’s success and the need for excess liability insurance at layers below ACE’s US$100 million attachment point, XL Insurance Ltd (XL) was created following the ACE model and began underwriting in May 1986. Both were based in Bermuda and began to write insurance on a freshly drafted and novel policy form which rapidly became known as the Bermuda Form. The drafting of that form was commissioned by Marsh & McLennan, with input from other insurance professionals.2 1 See, eg, Barbara D Stewart, Richard E Stewart and Richard SL Roddis, ‘A Brief History of Underwriting Cycles’ in J David Cummins, Scott Harrington and Robert Klein (eds), Cycles and Crisis in Property/Casualty Insurance: Causes and Implications for Public Policy (Kansas City, Missouri, National Association of Insurance Commissioners) 400, available at www.naic.org/documents/prod_serv_special_cyc_pb.pdf (addressing ways to control ‘[t]he disruptive effects of competition’ in the insurance industry by restricting entry, ‘stabilizing market shares and standardizing product’). This study was commissioned by the National Association of Insurance Commissioners as part of a longer study of related issues. 2 A lawyer, Thorn Rosenthal of Cahill Gordon in New York City, is often credited as the primary drafter of the Bermuda Form policy. The broad concept of the Form was conceived by Bob Clements of Marsh. Thorn Rosenthal worked with a committee of brokers from Marsh (Robert Redmond, Al Holzgruber, Tom Keating, Vince Stahl and Myra Tobin) and other industry experts. Thorn Rosenthal worked on the later versions of the XL form and the ACE 5 form which was developed at the same time as XL 004. For further background on the formation of ACE and XL, based in part on interviews with those involved, see Catherine R Duffy, Held Captive: A History of International Insurance in Bermuda (Toronto, Oakwell Boulton 2004) chs 33–36. See also the discussion of the development of Bermuda insurance market in David Scorey, Richard Geddes and Chris Harris, The Bermuda Form: Interpretation and Dispute Resolution of Excess Liability Insurance, 2nd edn (Oxford, OUP, 2018) chs 1–2 (hereafter Scorey, Geddes and Harris, Bermuda Form: Interpretation and Dispute Resolution).

1.01

2

The Legal and Economic Origins of the Bermuda Form

1.02

The collapse of the existing US casualty insurance market,3 and many of the distinctive features of the Bermuda Form that followed, had their origins in actual or perceived problems for insurers that arose as a result of the interpretations given by courts in the United States to the existing Comprehensive General Liability (CGL) insurance policies,4 particularly in the context of asbestos but also other latent-injury claims. CGL insurance policies, like the umbrella and excess liability insurance policies that often followed form to CGL or other underlying liability insurance policies, used standardised terms drafted by insurance industry organisations such as the Insurance Services Offices Inc in the United States and the Non-Marine Association for Lloyd’s and the London market.5

1.03

Until shortly after the Second World War, general liability and product liability insurance policies (often called public liability insurance in the United States) granted coverage for liability for bodily injury and property damage that were ‘caused by accident’.6 Coverage was normally afforded on a first-dollar basis or with a small deductible, and most of the US market bought primary insurance only from US insurance companies. Policy limits were specified for each accident, as well as on an aggregate basis for liability for ‘bodily injury’ or ‘property damage’. After the War, two insurance brokers in Canada and the United States and one underwriting organisation at Lloyd’s introduced the ‘umbrella’ policy, written at low excess levels on a broad form with attractive pricing. This policy form granted coverage on the basis of an ‘occurrence’. Umbrella liability insurance was the first type of excess liability insurance widely marketed and purchased by US policyholders as additional coverage above the primary or ‘working’ layer.7

1.04

Within a few years of the umbrella policy’s introduction, major US insurance companies copied it, and occurrence coverage began to drive ‘accident’ coverage out of the market. In 1966, the American rating bureaus – the organisations that set premium rates and drafted policy forms for nearly all US insurance companies – changed the standard primary general liability policy to an occurrence basis.8 The occurrence policy responds 3 A detailed economic analysis of the cause of the ‘liability insurance crisis’ in the mid-1980s is beyond the scope of this book. Apart from the decisions of the US courts, other factors contributed to severe economic problems for insurers; for example, the inflation which followed the 1974 decision of the Organization of the Petroleum Exporting Countries (OPEC) to raise oil prices substantially. See also Richard E Stewart, ‘The Eighth Cycle’, Marsh & McLennan Newsletter, December 1985. 4 In the revision of the primary CGL policy form released in the United States on 1 January 1986 under the auspices of the insurance industry trade group, the Insurance Services Office, Inc (ISO), the insurance industry changed the name of the CGL policy from the ‘Comprehensive General Liability’ policy to the ‘Commercial General Liability’ policy. 5 For a history of CGL standardised form language in the United States and the London market, see In re Insurance Antitrust Litigation 938 F2d 919 (9th Cir 1991), aff ’d in part, rev’d in part and modified sub nom Hartford Fire Insurance Co v California 509 US 764 (1993). See also Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) chs 1, 4, 14 and 15 (hereafter Masters and Stanzler, Insurance Coverage Litigation). 6 For additional history on standard-form CGL and umbrella insurance, see Masters and Stanzler, Insurance Coverage Litigation, ch 13. 7 Legend has it that underwriters in the London market used the term ‘umbrella’ as shorthand in transatlantic cabling. The term now generally refers to a broad form of excess catastrophe liability insurance that will, as its hallmark feature, ‘drop down’ in place of underlying primary insurance when the primary insurance does not apply to cover a loss. For additional background on CGL primary, umbrella and excess insurance, see ibid, ch 13. 8 The rating bureaus were the predecessors of ISO. The Insurance Rating Board (IRB), the Mutual Insurance Rating Board (MIRB) and the National Bureau of Casualty Underwriters (NBCU) were the rating bureaus in the United States involved in drafting CGL policy language and obtaining requisite state regulatory approval. See ibid, chs 1 and 4.

Policy Forms and Liability Problems

3

to – or, in insurance parlance, is ‘triggered by’ – liability from bodily injury or property damage that takes place during the policy period. The bodily injury or property damage must be caused by an ‘occurrence’, which is defined as ‘accident, including continuous or repeated exposure to conditions’ that is neither expected nor intended from the standpoint of the insured.9 The insurance industry’s movement from ‘accident’ to ‘occurrence’-based coverage was, as was the development of liability insurance generally, a response to expansions in tort liability throughout the twentieth century. At the turn of the twentieth century, tort liability had been based on privity of contract, but, with the increasing dominance of the industrial economy, expanded towards strict or virtually strict liability.10 By the early 1980s, policyholders had been held liable, and they and their insurers had paid, for strict liability in tort, often in a mass-tort context. A number of problems existed at around that time. In liability for medical malpractice, there had been a migration of the law, rather like that for products, from negligence to a concept more like strict liability, coupled with inflation in the cost of the service (medical care) that often measured damages. In asbestos bodily injury, the ‘exclusive remedy’ provisions of workers’ compensation laws broke down.11 In the environmental context, the United States Congress in 1981 had enacted the Comprehensive Environmental Recovery, Compensation, and Liability Act (CERCLA),

9 See, eg, 1973 CGL Form, reproduced in ibid at A18. The occurrence, or cause of the bodily injury or property damage, need not take place during the policy period; instead, occurrence coverage is activated when bodily injury or property damage takes place during the policy period. Courts have found that the bodily injury or property damage needed to trigger a CGL policy may be microscopic and need be ‘proved in retrospect ….’. American Home Prods Corp v Liberty Mut Ins Co 748 F2d 760, 765–66 (2d Cir 1984) (applying New York law). See also ER Squibb & Sons, Inc v Lloyd’s & Cos, 241 F3d 154, 166–69 (2d Cir 2001) (per curiam) (rejecting insurer argument that bodily injury must have outward or ‘physical effect’, but can include predisposition to illness or disease (there, from diethylstilbestrol (DES) in second- and third-generation children)) (applying New York law). 10 The expansion of tort liability throughout the twentieth century traces its roots back to the landmark decision by the New York Court of Appeals, New York’s highest court, in MacPherson v Buick Motor Co 145 NYS 462 (App Div 1914), aff ’d, 217 NY 382 (1916). The court there held a manufacturer liable for injuries resulting when it put an ‘inherently dangerous’ product into the stream of commerce: If the nature of a thing is such that it is reasonably certain to place [life] and limb in peril when negligently made, it is then a thing of danger. Its nature gives warning of the consequences to be expected. If to the element of danger there is added knowledge that the thing will be used by persons other than the purchaser, and used without new tests then, irrespective of contract, the manufacturer of this thing of danger is under a duty to make it carefully … We are dealing now with the liability of the manufacturer of the finished product, who puts it on the market to be used without inspection by his customers. If he is negligent, where danger is to be foreseen, a liability will follow. We are not required at this time to say that it is legitimate to go back of the manufacturer of the finished product and hold the manufacturers of the component parts. To make their negligence a cause of imminent danger, an independent cause must often intervene; the manufacturer of the finished product must also fail in his duty of inspection. MacPherson 217 NY at 389–90. These expansions of tort liability were motivated in part by a conviction that manufacturers should be required to bear some of the societal and other costs arising from products manufactured on a mass basis. See also discussion of the history CGL insurance in Masters and Stanzler, Insurance Coverage Litigation, §§ 1.01, 14.02. 11 An important turning point in relation to asbestos bodily injury claims was the decision in Borel v Fibreboard Products Corp 493 F2d 1076 (5th Cir 1973). The US Court of Appeals for the Fifth Circuit applied the doctrine of strict liability in asbestos disease cases and subjected producers to joint and several liability, adopting a theory of enterprise liability. In order to state a viable claim, a claimant needed only to show exposure to the defendants’ asbestos or asbestos-containing product, and an asbestos-related disease.

1.05

4

The Legal and Economic Origins of the Bermuda Form

imposing retroactive liability without fault for environmental damage at a polluted site. In asbestos ‘property damage’ cases, claims against insurers arose from the cost of removal of asbestos-containing building materials (ACBM) from buildings, despite the absence of scientific evidence that asbestos was actually doing any harm if it was encapsulated and not disturbed or friable.12 Other mass torts that had imposed liability at that time included Agent Orange, a defoliant that the US armed forces had used in Vietnam,13 and DES (diethylstilbestrol), a drug taken by pregnant women to prevent miscarriage.14 1.06

These claims presented serious problems for insurers who had written ‘long-tail’ occurrence-based liability insurance policies. Because liability insurance policies insure the policyholder’s liability, they generally pay under the law and economic circumstances as they exist at the time liability is decided. That may be years after the insurance policy was underwritten and priced. During the time interval between pricing and underwriting on the one hand, and paying claims on the other, general liability insurers are exposed to the risk of adverse legal and economic change. In the 10 or 15 years before the ‘liability crisis’ of the 1980s, the cumulative effect of the expansion of liability law during the post-war period, the decisions of the US courts on insurance coverage, the economic inflation after the oil embargo of 1973, poor investment results and overcapacity worked its way through insurance company financial accounts. These trends reduced insurers’ capital and demonstrated that their premium rates were far too low. Insurers became unwilling to take on new risks, or even keep the risks they had, on anything like the old pricing. In addition to the serious troubles of the day, there was no assurance that the problem was not going to continue or even get substantially worse. The refusal of insurers to renew or to take on new accounts, the sudden and dramatic escalation of insurance premiums, the tightening of underwriting criteria and the sudden contraction in capacity took place most forcefully in the parts of the liability insurance market with the greatest exposure to adverse change and the least reliable data, such as excess general liability insurance. It was these developments that led to the need for and creation of new sources of capacity, including ACE and XL.

US LEGAL DECISIONS ON INSURANCE COVERAGE ISSUES 1.07

In the late 1970s and early 1980s, courts in the United States decided a number of significant cases involving insurance policy interpretation in the context of asbestos bodily injury claims, resulting in several different trigger theories.15 By that time, asbestos claimants were suing asbestos producers in increasing numbers,16 albeit nothing compared 12 See, eg, Dayton Indep Sch Dist v National Gypsum Co 682 F Supp 1403 (ED Tex 1988), rev’d on other grounds sub nom WR Grace & Co v Cont’l Cas Co 896 F2d 865 (5th Cir 1990). The Environmental Protection Agency in the United States eventually decided that building materials containing asbestos already in place, if undisturbed, should usually remain in place in schools and other public buildings. 13 Uniroyal, Inc v Home Ins Co 707 F Supp 1368 (EDNY 1988). 14 See, eg, ER Squibb & Sons, Inc v Lloyd’s & Cos 241 F3d 154 (2d Cir 2001) (per curiam) (applying New York law). 15 For a further discussion of trigger theories, see Masters and Stanzler, Insurance Coverage Litigation, ch 4. 16 In the late 1970s, claims against the principal asbestos producers were running at approximately 1,000 a year. At the time, these figures seemed very substantial, but by the late 1980s they were running at approximately 24,000 per year and in the early 1990s at 60,000 per year. For further details, see Society of Lloyd’s v Jaffray 3 November 2000 (Cresswell J, unreported) ch 16; [2002] EWCA Civ 1101 (CA).

US Legal Decisions on Insurance Coverage Issues

5

with those seen later. Disputes arose between the asbestos producers, who were on the receiving end of claims from injured individuals, and their general liability insurers. The primary question, in the early claims involving delayed manifestation of bodily injury, concerned trigger of coverage, that is, the event that activates coverage. As shown in those early cases, insurance companies on the risk early in a period of continuing injury argued that ‘occurrence’ policies are triggered only at the time of the ‘manifestation’ of injury. Insurance companies on the risk later in that period argued that ‘occurrence’ policies were triggered only at the time of first exposure. Policyholders rejected the choice between ‘manifestation’ and ‘exposure’ offered by insurers, arguing that CGL policies were triggered at all times during the course of injury, from first exposure to the manifestation of injury or disease – the so-called continuous trigger.17 For example, an asbestosis sufferer might have been exposed to asbestos for a period of 20 years prior to manifestation of the disease in year 21. The asbestos producer might have bought insurance coverage for that entire period, or some part of it. If the ‘exposure’ theory of insurance liability were adopted, then all the insurance policies in force during the period of exposure (or alternatively at the time of a claimant’s first exposure) to asbestos would respond.18 If the ‘manifestation’ theory were adopted, then the only insurance policies that would respond would be those on the risk in year 21. It was in the economic interests of the insurers who had underwritten in the exposure years to support the ‘manifestation’ theory, and vice versa. Apart from these economic interests, many in the insurance market held strong views as to how occurrence policies should respond. If the exposure or continuous trigger theory were adopted, so that a number of insurance policies potentially were triggered, further questions would arise as to the extent to which each policy would respond. For example, would the insured producer be able to claim the full amount of its liability (subject of course to policy limits) from any insurer that had written during the exposure years, leaving that insurer to seek contribution from other insurers? Or would the policyholder be required to ‘pro-rate’ the amount of its liability among all the insurers during those years, leaving each insurer with (using a time-on-therisk model) a one-twentieth share (assuming coverage had been bought during each of the exposure years)? What would happen if the policyholder had not bought coverage during each of the 20 years, but had only bought it only for (say) five of those years? Could the policyholder in these circumstances recover its full liability (defence costs and judgments/ settlements) from any of the insurers who had written coverage during the five years? If pro rata allocation were applied, would the full loss be divided among the five insurers and, if so, how? Or would the policyholder be required to bear three-quarters of the loss, leaving only one-quarter to be prorated among the insurers? 17 See, eg, Keene Corp v Ins Co of N Am 667 F2d 1034, 1039–41 (DC Cir 1981) (Hartford argued for a ‘first-exposure-only’ trigger while INA, Liberty and Aetna advanced a manifestation-only trigger of coverage. This theory arose in part as a result of insurance industry ‘drafting’ documents that policyholders received in discovery in early insurance coverage cases). See, eg, Masters and Stanzler, Insurance Coverage Litigation, §§ 1.03, 1.04, 4.01[A][7], 4.07[F]. 18 Two ‘exposure only’ theories have developed. Under an exposure trigger, all policies in effect during exposure to injurious conditions are triggered. If the claimant’s exposure to asbestos or another injury-causing agent continues over a period of years, multiple policy periods are triggered. In contrast, a ‘first exposure’ trigger typically activates only one policy period: that during which the claimant’s first exposure to the injurious agent takes place.

1.08

6

The Legal and Economic Origins of the Bermuda Form

1.09

These and related questions came before the US courts in the years leading up to the 1985 liability insurance crisis, and indeed they continue to arise in coverage litigation to this very day. Despite successes, many of these decisions on trigger and allocation issues were adverse to the insurance industry. Insurers thus came to view courts in the United States as giving further (and in some cases extreme) effect to social engineering to ensure that injured parties, and the companies that they were suing, were not left without a financial remedy. The coverage decisions during that period not only contributed to the liability insurance crisis, but also shaped many aspects of the policy wording of the Bermuda Form – as discussed further below, examples include the dispute resolution mechanism (London arbitration); the choice of law (New York, perceived as a more neutral or pro-insurer jurisdiction, particularly as modified); the redefinition of an ‘occurrence’; limited environmental and punitive damage coverage; and the specific exclusion of asbestos and other known problems.

1.10

In the United States, insurance is a matter of state law and, therefore, can vary from state to state.19 The courts in the United States reached different results in construing the standard-form CGL insurance policy provisions that came before them.20 There were, however, a number of themes, adverse to insurers, which ran through many of the decisions.21 Generally speaking, courts rejected the ‘manifestation-only’ theory. Instead,

19 Although insurance is an issue of state law in the United States, many insurance disputes are brought in federal court. The federal courts have jurisdiction under federal ’diversity jurisdiction’, which allows federal courts to entertain a case when all plaintiffs are completely ‘diverse’ to all defendants. Parties are ‘diverse’ for these purposes when the citizenship of the plaintiffs (meaning for corporations, both principal place of business and state of incorporation) differs from the citizenship of all defendants. In such cases, the “Erie Doctrine” applies, requiring a federal court sitting in diversity jurisdiction to apply the law of the state whose law is determined to apply. See Erie RR Co v Tompkins, 304 US 64 (1938). 20 See the Appendix (description of the hierarchy of the US court system). 21 Principal decisions on trigger and allocation in the late 1970s and early 1980s, prior to the first Bermuda Form in 1985, include the following: (1) Porter v American Optical Corp, No 75-2202, 1977 US Dist Lexis 12735 (ED La 1977), aff ’d in part, rev’d in part, 641 F2d 1128 (5th Cir 1981): the federal trial court applied a manifestation-only trigger of coverage for asbestos bodily injury claims. The US Court of Appeals for the Fifth Circuit reversed the district court’s application of the manifestation-only trigger of coverage and applied the exposure trigger for asbestos bodily injury claims. The court expressly concurred with the Sixth Circuit’s decision in Forty-Eight and pro-rated liability among all insurers on the risk during the exposure period. (2) Insurance Co of N Am v Forty-Eight Insulations Inc, 451 F Supp. 1230 (ED Mich 1978), aff ’d, 633 F2d 1212 (6th Cir 1980), clarified on reh’g, 657 F2d 814 (6th Cir 1981): the federal trial court applied exposure theory of coverage for asbestos bodily injury claims and apportioned liability on a pro rata basis on the basis of the relative lengths of their respective coverage. The US Court of Appeals for the Sixth Circuit upheld application of exposure theory and concluded that insurers had a duty to defend asbestos bodily injury claims brought against the manufacturers of asbestos. The court pro-rated liability among all insurers on the risk during the exposure period, with the burden of any uninsured years falling on the policyholder. (3) Eagle-Picher Indus, Inc v Liberty Mut Ins Co 523 F Supp 110 (D Mass 1981), aff ’d, 682 F2d 12 (1st Cir 1982): the federal trial court held that coverage for asbestos bodily injury is triggered when signs or symptoms become manifest, as determined by the date of actual diagnosis or, with respect to those cases in which no diagnosis was made prior to death, the date of death. The US Court of Appeals for the First Circuit affirmed the federal trial court’s application of the manifestation theory of coverage for asbestos bodily injury claims, rejecting Keene, but modified the appropriate definition of ’manifestation date’ to be not when the disease was actually diagnosed, but when it becomes reasonably capable of medical diagnosis. Remanded to the federal trial court for factual investigation as to when manifestation took place. In July 1984, on remand, the trial court determined that asbestos-related disease becomes reasonably capable of medical diagnosis ‘six years prior to the date of actual diagnosis’ (Eagle-Picher Indus, Inc v Liberty Mut Ins Co 829 F2d 227, 232 fn 5 (1st Cir 1987)).

US Legal Decisions on Insurance Coverage Issues

7

courts adopted trigger-of-coverage theories that activated coverage in multiple policy years, calling the theory by a variety of names. A number of courts adopted the exposure theory, or some variant of it. Unless adopted as a first-exposure-only trigger, this typically had the effect of activating coverage in multiple years and exposing more insurance (4) Keene Corp v Insurance Co of N Am, 513 F Supp 47 (DDC 1981), rev’d, 667 F2d 1034 (DC Cir 1981) (Keene): on appeal, US Court of Appeals for the District of Columbia Circuit applied a three-step or continuous trigger of coverage for asbestos bodily injury claims. In doing so, the appellate court reversed the decision of the federal trial court limiting coverage to the exposure period. (5) Commercial Union Ins Co v Pittsburgh Corning Corp, 553 F Supp 425 (ED Pa 1981), rev’d on other grounds, 789 F2d 214 (3d Cir 1986): the federal trial court applied the exposure theory of coverage to asbestos bodily injury cases, holding in addition that the primary insurer owed a continuing duty to defend despite exhaustion of liability limits. In 1988, in the case of Pittsburgh Corning Corp v Travelers Indemnity Co, No 84-3985, 1988 WL 5291 (ED Pa 21 January 1988), the same court broadened its decision to incorporate the continuous trigger, following the reasoning adopted in Keene. (6) American Home Prods Corp v Liberty Mut Ins Co 565 F Supp 1485 (SDNY 1983), aff ’d in part, rev’d in part, 748 F2d 760 (2d Cir 1984): the federal trial court refused to apply Keene to a DES case, criticising Keene as ‘result-oriented’ and instead applied an ‘injury-in-fact’ trigger triggering coverage when disease was diagnosable and compensable. The US Court of Appeals for the Second Circuit affirmed the federal trial court’s reading of the policy language as calling for a ‘injury-in-fact’ trigger, but rejected the federal trial court’s requirement that ‘injury-in-fact’ be ‘diagnosable’ or ‘compensable’ and remanded the case to the federal trial court for further proceedings to determine when ‘injury-in-fact’ took place. (7) Crown, Cork & Seal Co v Aetna Cas & Sur Co, No 1292, slip op (Pa Ct CP Aug. 2, 1983) (Mealey’s, Litigation Report: Asbestos 7022 (12 August 1983)): the Pennsylvania state trial court, in an unpublished decision, triggered all insurance policies in effect during exposure, exposure-in-residence and manifestation to require the insurer to indemnify and defend the policyholder, following Keene. (8) AC&S, Inc v Aetna Cas & Sur Co 576 F Supp 936 (ED Pa 1983), aff ’d in part, rev’d in part, 764 F2d 968 (3d Cir 1985): the federal trial court followed Keene and held that coverage was triggered by exposure, exposure-in-residence and manifestation, and that the duty of liability insurer to defend is separate from, and broader than, the duty to indemnify, and insurers must continue to defend their policyholders even after the exhaustion of policy limits. The US Court of Appeals for the Third Circuit affirmed the federal trial court’s decision that exposure, exposure-in-residence and manifestation all trigger coverage, following Keene, but overturned the district court’s decision that there was an unlimited duty to defend, holding that an insurer has no duty to defend if it is established at the outset of the action that the insurer cannot possibly be liable for indemnification because policy limits have been exhausted, and explicitly refusing to determine whether or under what circumstances an insurer may terminate its defence of a claim in ‘mid-course’. (9) Owens-Illinois, Inc v Aetna Cas & Sur Co 597 F Supp 1515 (DDC 1984): the federal trial court concluded that it was bound by the DC Circuit’s decision in Keene and applied the continuous or triple trigger theory of coverage for asbestos bodily injury claims. (10) Vale Chem Co v Hartford Accident & Indem Co 490 A2d 896 (Pa Super Ct 1985), rev’d on other grounds, 516 A2d 684 (Pa 1986): the Pennsylvania trial court found DES claims alleged continuous injury and triggered all insurance policies in effect from ingestion to discovery of cancer in offspring, following Keene. The Pennsylvania Supreme Court reversed on procedural grounds. (11) Commercial Union Ins Co v Sepco Corp, 765 F2d 1543 (11th Cir 1985): United States Court of Appeals for the Eleventh Circuit upheld exposure trigger of coverage, following Porter. (12) Lac D’Amiante du Quebec, Ltee v American Home Assurance Co 613 F Supp. 1549 (DNJ 1985), vacated as to one defendant on other grounds, 864 F2d 1033 (3d Cir 1988): the federal trial court in New Jersey adopted the continuous/triple trigger of coverage for asbestos property damage and bodily injury claims. (13) Hancock Labs, Inc v Admiral Ins Co 777 F2d 520 (9th Cir 1985): a coverage dispute arising from allegedly defective heart valves. The US Court of Appeals for the Ninth Circuit rejected the Keene theory on the grounds that it was adopted because it was difficult to discern from medical evidence when and how an injury takes place from asbestos inhalation. Instead, the court applied the exposure theory, holding that the bodily injury took place when the defective heart valve was implanted. Because insurance law is an issue of state, not federal law, the US Supreme Court routinely rejected appeals in coverage cases by denying petitions for writs of certiorari in these cases, including: Insurance Co of N Am v Forty-Eight Insulations Inc 633 F2d 1212 (6th Cir 1980), clarified on reh’g, 657 F2d 814 (6th Cir 1981), cert denied, 454 US 1109 (1981); Porter v Am Optical Corp 641 F2d 1128 (5th Cir 1981), cert denied, 454 US 1109 (1981); Keene Corp v Ins Co of N Am 667 F2d 1034 (DC Cir 1981), cert denied, 455 US 1007 (1982); Eagle-Picher Indus, Inc v Liberty Mut Ins Co 682 F2d 12 (1st Cir 1982), cert denied, 460 US 1028 (1983).

8

The Legal and Economic Origins of the Bermuda Form

policies to claims. Some courts adopted a ‘continuous trigger’ (sometimes referred to as ‘triple trigger’), whereby liability under each insurance policy in force from first exposure through manifestation was triggered. If a policy was triggered, then a separate question arose as to how much it should pay. On this issue, typically called ‘allocation’, some courts were willing to adopt a ‘joint and several’ approach to payment by insurers, rather than pro rata allocation across policy years.22 1.11

Both the continuous trigger and ‘joint and several’ or ‘all sums’ allocation were adopted in the landmark decision in Keene Corp v Insurance Co of North America.23 Two aspects of this decision were unfavourable to insurers. On trigger of coverage, Keene was the first decision to hold that asbestos bodily injury takes place, and triggers coverage, not only when a person was exposed to asbestos fibres and when illness manifested itself, but also at any stage in between.24 Prior to Keene, the battle appeared to be between the exposure and manifestation theories.25 But Keene encompassed both of these triggers and (by including any stage in between) went beyond both of them.

1.12

Second, Keene adopted what the court called a joint-and-several basis of allocating liability among insurance policies that were triggered to cover the policyholder’s liability for continuing damage or injury over a period of years. An insurer that had issued a CGL insurance policy during any part of the period of continuing injury or damage was liable under Keene-type allocation for the full amount of the injury suffered by the individual who had claimed against the policyholder. Thus, the court recognised a distinction under the policy language between the ‘trigger’ and ‘scope’ of coverage. The time of the bodily injury or property damage was the ‘trigger’ (the event that activated coverage), but the policy language (which included no allocation provision) allowed an interpretation whereby once ‘triggered’, the scope of the policy’s coverage was determined separately. The point was starkest where some injury or damage took place during the policy’s operative period, but substantial damage from the same cause also took place both before and after it. By pointing to policy language and insurance-industry drafting documents that (as the court concluded) separated the questions (i) whether there had been injury or damage during the policy period (trigger) and (ii) what amount of loss had taken place during the policy period (scope), the court in Keene found that an insurer whose insurance policy was triggered was liable for ‘all sums’ of the policyholder’s liability regardless of whether some of the damage took place outside of the policy period in question. This approach, commonly known as the ‘joint and several’26 22 This ‘all sums’ or, as Keene called it, ‘joint-and-several’ allocation leads to ‘stacking’ of limits: multiple policy years can be called upon to pay for the policyholder’s liability. 23 Keene Corp v Insurance Co of North America 667 F2d 1034 (DC Cir 1981). 24 There is some irony in the fact that this landmark decision involved Keene Corporation (Keene). This company was by no means a mainstream producer of asbestos products and had apparently sold only US$750,000 worth of asbestos products during its entire existence. Like many asbestos producers (and, in recent years, other companies), Keene was ultimately driven into bankruptcy by asbestos liability, although the decisions on coverage by the DC Circuit and other courts on Keene’s coverage allowed it to continue operating many more years than otherwise would have been the case. 25 Early cases on coverage endorsing an exposure (Forty-Eight) or manifestation (Eagle-Picher) trigger adopted the position sought by the policyholder, as that theory maximised its coverage. 26 This terminology made all insurers whose insurance policies had been ‘triggered’ potentially liable for the full amount of the policyholder’s loss, just as ‘joint and several’ liability may make two tortfeasors each liable for the full amount of their victim’s loss arising from the combination of their torts.

US Legal Decisions on Insurance Coverage Issues

9

or ‘all sums’27 allocation of liability, was controversial. It was, and has been, bitterly resented by the insurance industry, and Keene became shorthand not only for ‘continuous trigger’, but also for ‘all sums’ or ‘joint and several’ allocation. The alternative approach, which was current as an alternative, treated the ‘trigger’ and ‘scope’ of coverage together, and held the insurer liable only for such portion of the loss as may be attributed to that part of the injury or damage as took place during the policy period.28 This is typically called pro rata allocation and for years it was generally considered to be the law in New York on allocation. This conclusion arose from the decision by the New York Court of Appeals, the ultimate arbiter of New York law, in Consolidated Edison Co of New York v Allstate Ins Co.29 The court concluded that the policy language at issue called for pro rata, time-on-the-risk allocation.30 Recognising that the policy language there was not standard-form, the Con Ed court specifically stated that its decision there was ‘not the last word on proration’ in New York.31 The court confirmed that point in 2016 in its unanimous decision in In re Viking Pump.32 In that case, New York’s high court adopted ‘all sums’ allocation, under ‘different policy language’, affirming that ‘[i]n determining a dispute over insurance coverage, [courts] look first to the policy language’;33 as a result, there is no ‘blanket rule’ on the issue of allocation in New York.34

1.13

In the 1980s, New York federal courts had rejected Keene’s continuous trigger approach in favour of an ‘injury-in-fact’ trigger, which is often considered more favourable to insurers and which remains relevant to issues that can arise under the Bermuda Form as to whether and when personal injury and property damage have taken place.35

1.14

The result of the coverage litigation on trigger and allocation issues36 was that insurers found themselves liable for massive amounts on insurance policies written over the

1.15

27 Because it made all insurers liable for the entirety of the insured’s loss on the basis of the reference to the insurance company’s insuring agreement to pay the policyholder’s liability for ‘all sums’ that the insured became liable to pay on account of bodily injury or property damage caused by an occurrence. 28 Insurance Co of N Am v Forty-Eight Insulations Inc 633 F2d 1212 (6th Cir 1980), clarified on reh’g, 657 F2d 814 (6th Cir 1981), cert denied, 454 US 1109 (1981). 29 Consolidated Edison Co of NY v Allstate Ins Co 746 NYS2d 622 628–31 (NY 2002) (Con Ed). Before New York’s highest court, the New York Court of Appeals, addressed the issue of allocation in Con Ed, various federal courts, applying New York law, had adopted pro rata allocation. See, eg, Stonewall Ins Co v Asbestos Claims Mgmt Corp 73 F3d 1178 (2d Cir 1995), modified on other grounds, 85 F3d 49 (2d Cir 1996) (Stonewall). 30 Con Ed 746 NYS2d at 630 (policy language did not specifically ‘mandate’ pro rata allocation). See also Keyspan Gas E Corp v Munich Reins Corp 73 NYS3d 113 (NY 2018) (applying pro rata allocation after analysis of policy language at issue). 31 Con Ed 746 NYS2d at 630. 32 In re Viking Pump 33 NYS3d 118 (NY 2016) (In re Viking Pump). 33 Ibid at 125–26 (alteration in original) (quoting Roman Catholic Diocese of Brooklyn v National Union Fire Ins Co of Pittsburgh, Pa 969 NYS2d 808, 813 (NY 2013)). 34 In re Viking Pump 33 NYS3d at 124–25. The court also upheld ‘vertical exhaustion’, rejecting insurer arguments that all primary policies in all triggered years had to be exhausted before any excess policies would respond, finding no support for that position in the policy language (ibid at 128–32). 35 American Home Products 565 F Supp 1485 (SDNY 1983), aff ’d as modified, 748 F2d 760 (2d Cir 1984). See further ch 5. 36 For later decisions in the 1980s, see, for example: (1) Aetna Cas & Sur Co v Abbott Labs, Inc 636 F Supp. 546 (D Conn 1986): the federal trial court in Connecticut applied an ‘injury-in-fact’ trigger of coverage to a coverage for DES liability.

10

The Legal and Economic Origins of the Bermuda Form

previous decades, and in respect of which insurers thought that they had closed their books. Insurers found themselves facing huge liabilities from these ‘long-tail’ claims for which they had not adequately reserved. The background is reflected not only in well-known decisions in the United States,37 but also in a number of decisions of the English courts in the Lloyd’s context.38

(2) Standard Asbestos Mfg & Insulating Co v Royal Indem Ins Co, No CV-80-14909, slip op at 9 (Mo Cir Ct Apr. 3, 1986) (Mealey’s Litigation Report: Insurance at 2,424 (13 May 1986)): the Missouri state trial court, in an unpublished decision, applied an injury-in-fact trigger to activate policies in effect at the time of injurious exposure only, rejecting Keene and following Forty-Eight. (3) Abex Corp v Maryland Cas Co 790 F2d 119 (DC Cir 1986): the US Court of Appeals for the District of Columbia Circuit applied an ‘injury-in-fact’ trigger, refusing to follow Keene after concluding that New York law applied to Abex’s insurance policies, and instead following American Home Products. (4) Zurich Ins Co v Northbrook Excess & Surplus Ins Co 494 NE2d 634 (Ill App Ct 1986), aff ’d sub nom Zurich Ins Co v Raymark Indus, Inc, 514 NE2d 150 (Ill 1987) (Raymark): the Appellate Court of Illinois upheld the trial court’s decision that insurance coverage for asbestos-related claims was triggered both by exposure and by manifestation, but not exposure in residence which some have called ‘dual trigger’. The court overturned the trial court’s decision that insurers had a duty to defend claims even after the limits of their policies were exhausted by the payments of judgments or settlements, even in respect of pending claims. The Supreme Court of Illinois affirmed the lower courts’ findings that coverage attaches at the time of bodily injury, which the court determined is concurrent with exposure, or disease. The trigger applied by Raymark has been called a ‘double trigger’ because the court did not explicitly find coverage for what is called ‘injury-in-residence’. The Supreme Court held that trigger may need to be decided on a case-by-case basis (ibid at 161). The Supreme Court also rejected the policyholder argument for an ‘unlimited duty to defend’, upholding the appellate court’s finding that the duty to defend ends when the policy’s limits are exhausted. (5) Mraz v Canadian Universal Ins Co 804 F2d 1325 (4th Cir 1986) (applying Maryland law): the federal appeals court, predicting Maryland law, applied manifestation trigger to liability for environmental property damage. The Maryland Court of Appeals, Maryland’s highest court, later rejected the decision by the US Court of Appeals for the Fourth Circuit as contrary to Maryland law: Harford Cty v Harford Mut Ins Co, 610 A2d 286 (Md 1992). (6) In re Asbestos Ins Coverage Cases, Judicial Council Coordination Proceeding No 1072, Phase III (Cal Super Ct filed 24 January 1990), aff ’d sub nom Armstrong World Indus, Inc v Aetna Cas & Sur Co, 52 Cal Rptr 2d 690 (Ct App 1996): the California trial court applied a Keene-type result and broadened the period of continuous trigger found in Keene to include the period from first exposure to asbestos or asbestos-containing products until date of death or date of claim, whichever takes place first. The court followed Keene, requiring any insurer on risk during continuous injury to pay the policyholder’s liability for the entire claim up to the policy limits. The court allowed the policyholders to choose which insurer would be obligated to defend and held that ‘all sums’ did not require the policyholder to bear responsibility for any uninsured or self-insured periods. In addition, the court held that insurers were not required to defend actions, or even to continue to defend pending cases, once policy limits have been exhausted, although it held that the cost of providing a defence does not reduce primary policy limits. (7) Pittsburgh Corning Co v Travelers Indem Co, No 84-3985, 1988 WL 5291 (ED Pa 21 January 1988) (Pittsburgh Corning I); and Pittsburgh Corning Co v Travelers Indem Co, No 84-3985, 1988 WL 5301 (ED Pa 21 January 1988) (Pittsburgh Corning II). In Pittsburgh Corning I, the federal trial court determined that coverage for asbestos property damage is triggered at the time of discovery of damage. In Pittsburgh Corning II, the federal trial court determined coverage for asbestos bodily injury, sickness or diseases is triggered ‘if any part of the injurious process – from time of exposure to time of manifestation – occurred within’ the policy period (1988 WL 5301, at *3). (8) Eli Lilly & Co v Home Ins Co 482 NE2d 467 (Ind 1985): the Indiana Supreme Court upheld the application of the continuous trigger to liability for pharmaceutical bodily injury. (9) Independent Petrochemical Corp v Aetna Cas & Sur Co 654 F Supp 1334 (DDC 1986), aff ’d in part, rev’d in part on other grounds, 944 F2d 940 (DC Cir 1991): the federal trial court, applying Missouri and New York law, adopted an injury-in-fact trigger, in reliance on federal courts decisions in American Home Products. The federal appeal court affirmed the trial court’s decision on trigger.

The Creation of ACE and XL

11

THE CREATION OF ACE AND XL The interpretation of CGL occurrence insurance policies by the US courts, and the other developments already discussed, had major repercussions for the insurance industry. Beginning around 1984, commercial buyers of excess liability and directors’ and officers’ liability insurance began to see a significant reduction in worldwide insurance capacity. Insurance rates skyrocketed, and this trend continued into 1985, at which time the commercial insurance market to a large extent ceased writing high excess general liability coverage for most companies in the United States, particularly the perceived high-risk chemical and pharmaceutical industries. In the light of the absence of available insurance capacity in the commercial market, the major broking firm Marsh & McLennan, together with JP Morgan Guaranty Bank, worked to create new insurance companies to help fill this void. Both Marsh and Morgan agreed to provide start-up services to the new insurance companies to help accelerate their entrance into the market. Because the capital markets declined to provide support, the concept relied on the policyholders of the new insurance entities to provide the start-up capital to form the companies. This new capital would come from an initial group of ‘sponsor’ companies, together with a required reserve premium39 that non-sponsor companies would have to pay in addition to their annual premium. Whilst both Marsh and Morgan were rewarded (for example, by way of options and discounted equity) for their promotional efforts, Marsh’s overriding goal was to provide its brokerage clients with insurance coverage.

(10) Uniroyal, Inc v Home Ins Co 707 F Supp 1368 (EDNY 1988): the federal trial court applied an injuryin-fact trigger to liability for injuries from Agent Orange. The decision is also one of the key decisions for the principle in New York law holding that a policyholder need not prove the case made by plaintiffs in the underlying cases in order to obtain its insurance coverage. (11) Dayton Indep Sch Dist v National Gypsum Co 682 F Supp 1403 (ED Tex 1988), rev’d on other grounds sub nom WR Grace & Co v Continental Cas Co 896 F2d 865 (5th Cir 1990) (applying Texas and New York law): the federal trial court applied a continuous trigger to liability for asbestos property damage. (12) Wisconsin Elec Power Co v California Union Ins Co 419 NW2d 255 (Wis Ct App 1987): the Wisconsin appellate court applied the continuous trigger to liability for injury caused by emanations from a faulty power supply system. 37 For example, the decision in Stonewall 73 F3d 1178 (2d Cir 1995), modified on other grounds, 85 F3d 49 (2d Cir 1996) gives a good overview of the decisions in New York on coverage for both asbestos personal injury and property damage. 38 See, eg, Henderson v Merrett (No 2) [1997] LRLR 247, where Lloyd’s Names (who were private individuals who joined the Lloyd’s market and participated in syndicates which underwrote risks in the hope of making a profit) sued their managing agent, Merrett, for negligently writing run-off contracts and negligently closing years of account by reinsurance to close; and Society of Lloyd’s v Jaffray [2002] EWCA Civ 1101, where Lloyd’s Names unsuccessfully accused Lloyd’s of fraudulent misrepresentation in the context of asbestos liabilities. The judgment of Cresswell J (3 November 2000) at first instance is unreported, but available on legal databases. It contains a wealth of detail as to the development of asbestos claims, both bodily injury and property damage: see in particular ch 16 of the judgment. 39 Condition V(y) of the XL 001 Form required the Named Insured to purchase stock in XL for at least 100 per cent of the aggregate amount of the premiums for the first annual period. This requirement was sometimes deleted and replaced by an endorsement, the Reserve Premium Endorsement, which required the payment of a reserve premium equal to the aggregate premiums applicable for the first annual period. The 001 Form also contained a condition (Article V(r)) which provided for the proration of certain losses: see ch 11, para 11.92 below.

1.16

12

The Legal and Economic Origins of the Bermuda Form

1.17

Marsh and Morgan were successful in selling this concept to their client base and other interested companies. In late 1985 ACE was formed, followed by XL in May 1986. Many sponsor companies invested US$20 million or, in some cases, US$10 million in either or both companies.

1.18

Marsh led the development of the ACE and XL policy forms with the assistance of the New York law firm Cahill Gordon. The ACE policy form was introduced in late 1985, and XL was created thereafter to provide limits below those provided by ACE. When XL commenced writing business in May 1986, it adopted the ACE form in its entirety. The policy form sought to meet the needs of its sponsor companies and other clients,40 whilst at the same time trying to avoid exposing the new insurers to financial ruin from the very liability catastrophe problem they were established to help solve. Marsh understood that, for these new companies to succeed, they needed to provide significant amounts of excess catastrophe liability insurance for products and other liabilities. At the same time, because XL and ACE provided only high-excess layers (initially, ACE excess of at least US$100 million and XL excess of at least US$50 million, later reduced to US$25 million), a mechanism had to be provided that would allow a policyholder to access the limits for product claims arising from a single product defect, as most individual product claims would not reach the ACE or even the lower XL layers, even when reduced to an excess of US$25 million. Therefore, the new policy had to provide a means for the aggregating claims resulting from a common defect. Policyholders also generally sought at least a limited amount of sudden and accidental pollution coverage, and directors’ and officers’ liability insurance. In addition, policyholder companies were looking for these new insurance companies to provide coverage and premium stability over time, something which the traditional insurance market was perceived as having failed to deliver. This led to an unusual policy form that was a hybrid of occurrence, accident, discovery and claims-made concepts.

1.19

The policy forms used by ACE and XL have developed over the years. The first Bermuda Form, used by both ACE and XL, was 001. The current XL form is 004, and ACE has used five forms over the years. The later forms for both companies retain most of the distinguishing features that characterise the original policy form. The XL and ACE forms diverged in certain respects over the years,41 but the current versions are substantially the same. The Bermuda Forms pioneered by ACE and XL have been adopted by others in the insurance industry, not only in Bermuda but also by companies and Lloyd’s syndicates operating in the London market and European insurers.42 A typical Bermuda Form policy will now usually also include an increasingly large number of ‘endorsements’, containing terms which supplement or vary the basic policy terms.43 Such endorsements often seek to achieve 40 See, eg, XL’s 1993 Annual Report: ‘Part of XL’s mission is providing clients with confidence in the Company’s ability to meet their catastrophic liability claims needs.’ 41 For example, for a time, under the ACE 004 Form, ACE required arbitration of disputes in Bermuda, not London. 42 Some insurance companies utilise the XL Forms with no or virtually no adaptation or alteration. Other companies have made more substantial alterations. Amongst the companies which have written business on a version of the Bermuda Form are Swiss Re, Allianz, Zurich, Gerling, Starr Excess and Oil Casualty Insurance Ltd (OCIL). The Forms have also been adopted for areas of business other than general liability: for example, XL’s professional liability policy is substantially based on the Bermuda Form. 43 See, eg, the ‘Limited Liability Entity Endorsement’ in Noble Assurance Co v Gerling Konzern General Insurance Co [2007] EWHC 253 (Comm), [2007] 1 CLC 85.

Key Features of the Bermuda Form

13

a fine-tuning of the basic wording, but they can unintentionally create radical changes to a balanced policy form or introduce inconsistencies. It is therefore necessary to consider proposed endorsements with caution, so that alterations to the existing balance are understood. It is also not unusual for Bermuda Form policies to be made subject to English law. Again, unless caution is exercised, this can give rise to problems. The Bermuda Form is drafted against a background of US insurance law, and some of the concepts (for example the ‘expected or intended’ provision) are unfamiliar in an English law context.44 Originally, as stated above, ACE sold general liability insurance coverage only in excess of US$100 million, and XL sold general liability insurance policies in excess of a US$50 million retention with a $75 million limit. XL initially applied higher retentions for certain classes of business, but by the late 1980s chemical and pharmaceutical companies could buy coverage attaching at US$25 million. EXEL Ltd, the parent company of XL, went public in 1991, and ACE Limited (the parent of ACE) went public in 1993. The shares in both companies rose significantly in value after their public offerings. Both companies thereafter grew significantly, in part as the result of acquisitions. In 2016, Ace Limited acquired the Chubb Corporation, and the business now trades as Chubb. In 2018, Axa acquired the XL Group Ltd and the division is now known as Axa XL.45

1.20

KEY FEATURES OF THE BERMUDA FORM The ‘Bermuda Form’ policy, as originally drafted and issued by ACE, is properly to be 1.21 regarded as a balanced policy form, aiming to hold the ring fairly between the interests of policyholders and the interests of investors, as the same industrial corporations were in both roles. The key policy provisions can frequently be traced to something that had gone wrong or proved very expensive in the liability insurance crisis of the mid-1980s. The Bermuda Form is a monument to learning from unpleasant experience. The architecture of the new policy was driven by a desire to preserve key features of occurrence coverage, whilst avoiding the problem of ‘stacking’. Occurrence Reported The effect of the decisions on the CGL occurrence policies had been to present insurers with a massive but unknown ‘tail’ of liability. This is because occurrence policies respond 44 In addition, as shown by the decision of the English Court of Appeal in AstraZeneca Insurance Co Ltd v XL Insurance (Bermuda) Ltd & ACE Bermuda Insurance Ltd [2013] EWCA Civ 1660, a change in the choice of law from New York to English law can have unintended consequences. There, the Court of Appeal applied the rule, established by English case law, that an insurer is required to pay for only actual legal liability, placing the burden on the policyholder to prove, if it settled, that it would have been subject to actual legal liability. The law in New York does not, in contrast, require the policyholder to ‘prove the case against itself in order to obtain cover for settlements: Luria Bros & Co v Alliance Assurance Co 780 F2d 1082, 1091 (2d Cir 1986). The English Court of Appeal recognised that the applicable principle under the New York law principle differed from English law: see [2013] EWCA Civ 1660, [2014] 2 All ER (Comm) 55, [25]. 45 Although both companies initially eschewed operations in the United States, following a merger in 1999, XL established the subsidiary XL America based in Stamford, Connecticut. Chubb has executive offices in New York, Zurich, London and other locations.

1.22

14

The Legal and Economic Origins of the Bermuda Form

to bodily injury or property damage that takes place during the policy period, even if the injury or damage is undetected and develops gradually over a period of years. It was for this reason that insurers who had written policies in, say, 1950 found that they were having to meet claims arising from the exposure of an individual to asbestos during 1950. Because asbestos-related diseases46 developed over years and were not detected for years or even decades, these liabilities generally did not emerge until the late 1970s or later. It is, therefore, not surprising that in insurance company accounts drawn up in the 1950s, 1960s and early 1970s there were no reserves for them. Insurers found that they were having to meet claims on policies on which their books had been closed.47 Indeed, especially in the early days of the asbestos problems, it often proved difficult actually to trace or find the policies themselves. 1.23

By the mid-1980s, the US insurance drafting organisation ISO promulgated a ‘claims made’ form for general liability insurance sold to large companies in the United States.48 ‘Claims made’ insurance is also now the standard basis upon which professional liability risks are written in the United Kingdom. In their simplest form, policies written on this basis respond to a claim which is made against the policyholder during the currency of the policy. If no claim is made during the annual policy period, the insurer will keep the premium as a profit. The insurer may renew the business for the following year, and again the policy will respond if a claim is made against the insured in that year. If the insurer discontinues cover, or the policyholder goes to another insurer, without any claim having been made during an annual policy period, then there will never be a claim for which the insurer has to respond. In this way, the problem of a massive but unknown tail of claims does not arise: the insurer will know, at the end of each annual policy period or soon thereafter,49 whether a claim has been made against a policyholder.

1.24

At the time of the liability insurance crisis in the mid-1980s, however, the insurance-buying public and state insurance regulators in the United States resisted the insurance industry’s adoption of a claims-made trigger for all general liability insurance. The new claims-made form met with vociferous objections from corporate policyholders, brokers and regulators in the United States, and it even became the subject of major antitrust litigation.50

46 That is, asbestosis, pleural plaques, mesothelioma and other asbestos-related cancers. 47 The problem was particularly acute for Lloyd’s Names (see above n 38). Part of the Lloyd’s system was that the syndicate would if possible close its books by effecting a reinsurance to close; viz a contract of reinsurance with the Names who were members of the syndicate in a later year. These Names, many of whom did not join Lloyd’s until the 1980s, found themselves having to meet very substantial liabilities on old Lloyd’s policies, most of which were written before they joined, and some of which were written before they were even born. 48 See, eg, Masters and Stanzler, Insurance Coverage Litigation, chs 1 and 4. 49 Claims-made policies invariably contain provisions requiring the prompt notification of losses or at least during the policy period. Policies written in the United States generally allow the insured an extended reporting period for at least a short period of time. This extended reporting period applies after the claims-made coverage expires or is cancelled and provides that a claim made during the extended reporting period is deemed to have taken place on the last day of the policy period. State insurance departments in the United States required that insurers add an extended reporting period to the proposed claims-made form before they would give the requisite regulatory approvals to the new claims-made form in the 1980s. See Masters and Stanzler, Insurance Coverage Litigation, ch 1. 50 In re Ins Antitrust Litig 938 F2d 919 (9th Cir 1991), aff ’d in part, rev’d in part and modified sub nom Hartford Fire Ins Co v California 509 US 764 (1993).

Key Features of the Bermuda Form

15

As a result, the industry presented both claims-made and occurrence forms for approval to regulators in 1985.51 The Bermuda Form policy is neither pure occurrence nor pure claims-made, but is a hybrid of the two. The policy is at heart an occurrence policy, and indeed expressly uses the expression ‘occurrence’. In broad terms, it covers occurrences that take place during the policy period, with a start-point and an end-point. The start-point is the inception date of the policy52 and the end-point is the moment when the policyholder stops buying the basic cover granted by the policy,53 or the insurer stops selling it. The policy is envisaged to be a continuous policy, in the sense that it continues from year to year. In contrast, a claims-made policy stops at one year end, and starts again afresh, with a new policy period, if the policy is renewed. The policy period of a Bermuda Form policy automatically renews unless one party cancels it.54 Thus, a Bermuda Form policy has a policy period that may span years with a number of Annual Periods. Each Annual Period requires a new premium and provides new limits of liability. Accordingly, the start-point and the end-point of a Bermuda Form policy may be many years apart.

1.25

Thus far, the Bermuda Form has all the features of an occurrence policy. But it also has the beneficial feature, from the insurer’s standpoint, that the occurrence must be reported during the policy period. Hence, if the policyholder stops buying the basic cover granted by the policy and has not reported an occurrence during the policy period, the insurer can, subject to one proviso, close its books on the policy since no claim can in future trigger the cover. The proviso is that the insured has an option to purchase an extended ‘reporting’ or discovery period, known as Coverage B. This buys the policyholder extra time to report occurrences which took place during the currency of the basic period of cover, referred to as Coverage A.

1.26

Dispute Resolution The liability insurance crisis of the mid-1980s was viewed by many insurance people 1.27 at the time as largely attributable to decisions by American judges and juries that both

51 The insurance industry continues to sell a substantial amount of occurrence-based CGL insurance in the United States. 52 Sometimes a policyholder purchases retroactive coverage. A retroactive date defines the starting point of the period during which the bodily injury or property damage covered by the policy must take place. In other words, bodily injury in claims covered by the policy must commence after the retroactive date. The retroactive date may be the same as the inception date of the policy or may be a date that is earlier than the inception date. See chs 2 and 6 below. 53 Cover is afforded in respect of the period of ‘Coverage A’. When the policy would otherwise terminate, the policyholder has the option to purchase Coverage B. In substance, this provides an extended reporting period in respect of occurrences which have taken place during the Coverage A period. It does not extend to fresh occurrences which take place during Coverage B. Complications arise in respect of ‘batch’ or ‘integrated’ occurrences, and the start and end points for these. See ch 6 below. 54 The parties will need to agree upon terms for continuation, such as the premium (see ch 2 below) and the discussion of the cancellation and policy extension conditions (see ch 11, paras 11.74–11.59 and 11.94–11.96 below).

16

The Legal and Economic Origins of the Bermuda Form

expanded tort liabilities and broadened insurance coverage beyond that which insurers believed was contemplated when they wrote and sold the policies. To address this problem, the decision-making process on disputes with policyholders was moved from the US court system to arbitration in London.55 The substantive law selected was that of New York, which was considered to be, as between insurers and policyholders, more developed and neutral than that of many other states of the United States. By modifying New York law in some respects,56 the drafters of the Bermuda Form sought to correct certain perceived imbalances in favour of the policyholder under New York law, and to limit resort to materials extrinsic to the language of the policy itself.57 1.28

The removal of disputes to a London arbitral forum has benefited English legal practitioners, who have become involved (whether as counsel or arbitrators) in disputes which they might otherwise not see. However, a more important side effect is that no body of case law has built up in relation to the interpretation of the Bermuda Form. Although English law58 does permit appeals against arbitration awards in limited circumstances, these circumstances are confined to awards where there is an error of English law. Since the substantive law applied by arbitrators in relation to the Bermuda Form is New York law, there is, therefore, no possibility of an appeal59 of a decision of an English court on the interpretation of the Bermuda Form. As far as the US courts are concerned, the Bermuda Form aims to keep disputes between insurer and policyholder away from those courts. Although some US litigation has addressed the question of the interpretation of the Bermuda Form (for example, as a result of a contribution claim by one insurer against XL or ACE), there has hitherto been no significant American case law on the Form. This may change to some extent as Bermuda Form policy provisions are used increasingly in policy forms used by other insurance companies. Detailed Policy Language

1.29

The drafting of the Bermuda Form reflects a concern that the language of existing policies was often too brief and vague, leaving room for the courts to construe policy language liberally. The Bermuda Form, by contrast, seeks to achieve clarity by elaboration. It may first strike a reader as over-elaborate and over-drafted. There are clearly

55 In C v D [2007] EWCA Civ 1282, [2008] 1 Lloyd’s Rep 239, Longmore LJ referred to this passage and suggested that ‘it might equally be true that the selection of New York law as the proper law of the contract may show a certain disenchantment with the substantive law of insurance in England, a matter which the Law Commission is currently addressing’. Given that those involved in creating the Bermuda Form were US brokers and lawyers, and given that the principal purchasers of Bermuda Form policies would be US companies, it was natural to choose the law of one of the states of the United States. The choice did not, in fact, reflect disenchantment with English law. 56 See ch 4 below. 57 The extent to which this was achieved is also considered in ch 4 below. 58 See English Arbitration Act 1996, s 69. This largely codifies the principles as to appeals established by the case law under the English Arbitration Act 1975, s 1. 59 See, eg, Egmatra AG v Marco Trading Corp [1999] 1 Lloyd’s Rep 862; Reliance Industries Ltd v Enron Oil and Gas India Ltd and Another [2002] 1 Lloyd’s Rep 645. The 004 Form contains a waiver of any right of appeal which, under the Arbitration Act 1996, is valid even though concluded before any dispute had arisen. The earlier policy forms do not contain an equivalent waiver. See further ch 3, para 3.16 below.

Key Features of the Bermuda Form

17

areas where it is not easy to work out how particular clauses fit together, and also where a literal interpretation of policy language might produce a result that is at odds with what appears to be the intention of the policy. There is also duplication or surplusage in the drafting, particularly of the original policy form. These matters are more easily understood once one appreciates the climate in which the language first came to be drawn up and the challenges which the Bermuda Form was created to address. The liability insurance crisis had been, to a significant extent, caused or aggravated by the immense and dramatic mass-tort liabilities for certain products, such as asbestos, DES and intrauterine devices. The Bermuda Form singles these out for exclusion by name. Pollution liability was beginning to be a huge problem at the time when the Form was first drafted, and the policy contains a pollution exclusion which preserves some limited pollution cover, again reflecting the balance that needed to be struck between the interests of insurer and insured.

1.30

Aggregation of Claims One of the many problems faced by insurers in the years prior to the liability crisis 1.31 was the aggregation or ‘stacking’ of policy limits. An insurer – for example, an insurer who had written a policy to an asbestos producer for a number of years – could find that each year’s policy would be exposed and required (or ‘stacked’) to pay for the policyholder’s liability and mammoth defence costs. If the policy contained an aggregate limit, it would prevent payment of claims after the aggregate limit of the policy was exhausted. However, the limit alone did not sufficiently limit the insurer’s liability for a number of reasons. First, before the 1980s, most CGL policies, primary or excess, paid defence costs in addition to limits (and many do still today). These costs were high given the thousands of asbestos and other long-tail claims in states all across the country. Because the limits were exhausted only by the payment of settlements or judgments, the insurer could be liable for hundreds of thousands or even millions of dollars in defence costs, before a dollar was paid that eroded the limits. Second, some liabilities such as environmental liabilities were typically not subject to any aggregate limits at all. In those situations, the policy could be required to respond to many occurrences, subject to a per-occurrence limit but no aggregate limit. Third, in cases of continuing injury, all policy periods on the risk during the period of injury could be triggered. Once one policy year was exhausted, the policyholder would then simply turn to other policies or years where aggregate limits had not been exhausted. If the relevant jurisdiction took the ‘all sums’ approach,60 then the policyholder could in many states also pick and choose the policy years that would respond. The policies were not written on the basis that there was an overall aggregate, across policy years, or insurers in respect of a particular problem. Accordingly, all the limits of all insurers whose policies were triggered were in practice cumulative. Insurers found themselves liable on each year’s policy up to

60 See

paras 1.10–1.13 above.

18

The Legal and Economic Origins of the Bermuda Form

the policy limits, and thus found themselves paying on many insurance policies over many policy years in respect of similar claims. 1.32

‘Anti-stacking’ was a cornerstone of the new policy form. The Bermuda Form sought to eliminate ‘stacking’ in a number of ways. The key was the occurrence-first-reported trigger. A simple way to avoid cumulation of limits in a liability policy is to specify a single moment as the trigger and to sweep into the single triggered policy all the financial consequences of the policyholder’s liability in respect of the underlying problem. Accordingly, the Form requires the policyholder to group related events together or ‘integrate’ them into a single year, that being the year in which the policyholder determined that the claims were likely to implicate the policy and gave notice of that occurrence to the insurer. The policy gives a measure of discretion and judgment to the insured. The policyholder does not have to report every liability claim that is made, but only those that are ‘likely to involve this policy’.61

1.33

As a result, the Bermuda Form policy is at risk, in connection with a related claim or series of claims, only for one set of limits in the year in which the claim is reported. This feature of sweeping all related injuries or losses into a single policy year is commonly called ‘occurrence integration’ or ‘batching’ or ‘batch occurrence’. The ‘batching’ provision of the Bermuda Form, however, benefits both parties, thus reflecting the balanced nature of the Form. The batching clause enables the policyholder to add together a large number of small occurrences, with the result that the policyholder can exceed the very high retention that underlies the excess insurance provided by Bermuda Form policies and would otherwise defeat coverage for many individual claims. The insurer is protected, in that it is obligated to pay only one limit for any particular problem. Expected or Intended Injury and the ‘Maintenance Deductible’

1.34

The Bermuda Form contains a clause which has come to be known in Bermuda insurance industry custom and practice as the ‘maintenance deductible’.62 No Bermuda Form policy actually uses that term. If shown the policy for the first time, even an experienced lawyer would be hard put to locate the clause to which the expression relates. In fact, the relevant clause is the part of the provision in the definition of ‘occurrence’ that concerns injury or damage that is expected and intended. The concept of excluding injury or damage that the policyholder expected or intended was a well-known feature of CGL insurance policies, and was carried over into the first version of the Bermuda Form. It has remained there, with some significant development, ever since.

1.35

The idea that an insurer should not be liable for losses that the policyholder expected or intended is something which commands general acceptance. For example, few would quarrel with the notion that a policyholder who deliberately causes harm should not be able to recover from insurance – for example, the policyholder who maliciously pollutes

61 See 62 See

ch 8 below for a discussion of the reporting clause. ch 7 below.

Key Features of the Bermuda Form

19

a water supply. However, the application of this basic principle becomes much harder in less obvious situations. A classic example involves a drugs company that manufactures a product, say a vaccine, which is beneficial to huge numbers of people. Many drugs, such as vaccines, may cause some harm to a very small number of people who, for one reason or another, react adversely to the product. The vaccine may be successfully used by millions of people each year, but cause harm to an average of eight to ten people a year, all of whom can be expected to bring claims. An impartial observer might suggest that it would be fair for the drugs manufacturer to bear the risk of paying for the ‘noise-level’ eight to ten claims per year, but that if, for some unexpected reason, the level of claims rose significantly (perhaps to 20, 50, 100 or more), then the company’s insurance should respond.

1.36

In very broad terms, the ‘maintenance deductible’ concept in the Bermuda Form was an innovative solution to this recognised problem. These provisions are considered in Chapter 7 below, but for present purposes it is important only to note how the Bermuda Form sought to strike a balance between the legitimate interests of policyholder and insurer. Absent the revised expected or intended language and the ‘maintenance deductible’ concept, which originally operated as a proviso to the classic ‘expected/intended’ language of the policy, the insurer might have said to the policyholder that the marketing of a product with a proven history of losses meant that the policyholder expected or intended all the damage that resulted, whether or not there was a later unanticipated ‘spike’ in claims. Accordingly, this concept was aimed at preserving the existence of cover for a product with a known historical incidence of losses. At the same time, however, it was intended to put the ‘noise-level’ claims onto the shoulders of the policyholder, whilst providing the company with insurance protection for the later unexpected ‘spike’ in claims.

1.37

2 The Bermuda Form: Its Basic Structure 2.01 The Bermuda Form, as originally drafted and as further developed, is a complex document, whose various parts are intricately related. The revisions to the original Bermuda Form have been in the nature of refinements and clarifications, rather than a radical restructuring of the policy. This chapter describes the basic structure of the Form, and serves as an introduction both to the Form and to the chapters which follow.

AN OCCURRENCE REPORTED FORM 2.02 The Bermuda Form is neither purely an occurrence policy nor purely a claims made policy. Instead it is a distinctive mixture of certain features of each, using as the ‘trigger’ for coverage the concept of an ‘occurrence reported’. 2.03 A pure ‘occurrence’ form looks for a temporal connection between the time when the injury giving rise to the policyholder’s liability takes place and the period of the policy. Accordingly, in traditional occurrence-based coverage, coverage is activated or triggered when covered injury or property damage takes place during the policy period. The date when injury first comes to the policyholder’s attention, or is reported to the insurer, is irrelevant; what matters is when the injury actually happened. By contrast, a pure claims made form looks for a temporal connection between the time when the claim is made and the period of the policy. Traditional claims made coverage is triggered when a claim is made against the policyholder, or notified to the insurer, or both, during the policy period.1 2.04 The Bermuda Form uses the concept of an occurrence, and for the same purpose of placing some temporal limitation on the insurers’ liability. Thus, under the 004 Form,2 an ‘Occurrence’ is defined in Article III(V) and exists if and only if: (a) except with respect to actual or alleged Personal Injury or Property Damage arising from the Insured’s Products, there is an event or continuous, intermittent or repeated exposure to conditions which event or conditions commence on or subsequent to the Inception Date, or the Retroactive Coverage Date, if applicable, and before the Termination Date of Coverage A, and which cause actual or alleged Personal Injury, Property Damage or Advertising Liability; 1 See, eg, Lorelie S Masters, ‘Inconsistencies in Trigger, Arbitration, and Exhaustion Provisions in Claims Made Policies: A Cautionary “Tail”’, 5 Coverage No 3 (June/July 1995) (available in LexisNexis). 2 The ‘Occurrence’ definition in earlier Bermuda Forms (Forms 001, 002A, and 003) is found in Article III(e) and, as far as temporal limitations are concerned, is broadly similar in effect.

An Occurrence Reported Form

21

(b) actual or alleged Personal Injury to any individual person, or actual or alleged Property Damage to any specific property, arising from the Insured’s Products takes place on or subsequent to the Inception Date, or the Retroactive Coverage Date, if applicable, and before the Termination Date of Coverage A.

Accordingly, to qualify as an ‘Occurrence’ under paragraph (a) of this definition – which provides, for example, premises operations coverage – the ‘event or continuous, intermittent or repeated exposure to conditions’ must commence on or subsequent to the inception date or retroactive date of the policy and before the termination date of Coverage A (of which more below). To qualify under paragraph (b), which provides products liability coverage, the injury must take place on or subsequent to the inception date or the retroactive date and prior to the termination date of Coverage A.3 There is one important qualification or extension to the ‘occurrence’ definition, arising from the ‘batching’ provision which is found in the last paragraph of Article III(e) of the 001 Form and which became the ‘Occurrence Integration’ or ‘Integrated Occurrence’ provisions of the later Forms. Those clauses provide, in essence, that where damage results or allegedly results from a common defect, all the injuries resulting from that defect are treated as forming one occurrence ‘irrespective of the period … over which [they] occur’. These batching provisions are in effect deeming provisions that treat as a single occurrence what would otherwise be a series of separate occurrences. The batching provisions do not identify the existence of an occurrence, but are directed at the number of such occurrences and are a distinguishing characteristic of the Bermuda Form.

2.05

The batching provisions have various effects. First, because Bermuda Form policies are always excess of a large ‘per occurrence’ retention or deductible,4 the policyholder is able to exhaust that retention or deductible by combining, or ‘batching’, many small claims if they relate to the same defect. The provision thus benefits the policyholder by allowing it to access the high excess limits provided by the policy’s coverage, by a group of claims that, individually, would not exceed the retention. Second, because there is a ‘per occurrence’ limit, the insurer is not obliged to pay more than one limit for claims resulting from the same defect or event.5 In this way, the provision benefits the insurer by making clear that only one policy period applies to an occurrence involving continuing injury.

2.06

However, the existence of an ‘occurrence’ within the relevant time frame is not itself sufficient to activate coverage. In order for the insurers to be under an obligation to indemnify, the policyholder must give notice of that occurrence. Thus, Article I of the policy, the insuring agreement, states that an indemnity is provided only against

2.07

3 The ‘Occurrence’ definition is discussed in ch 6 below. 4 As discussed in ch 1, ACE initially provided catastrophe cover excess of a US$100 million or higher retention. XL typically attached initially at excess of a US$50 million or higher retention, but later sold insurance policies excess of US$25 million retention. 5 The 002 and 003 Bermuda Forms included (Article V(t)) a Reinstatement of Limits provision that allowed policyholders to buy and thus ‘reinstate’ a second limit of liability when a policyholder gave notice during the policy period of an occurrence that impaired the original limit. The second limit applied during the same Annual Period in which notice was given, upon payment of a reinstatement premium which was a maximum of 125 per cent of the total annual premium. As discussed in ch 11, Article VI(R) in the 004 Form also provides for the reinstatement of the aggregate limit.

22

The Bermuda Form: Its Basic Structure

an occurrence of which notice is given prior to either (i) the expiration of Coverage A or (ii) the expiration of Coverage B, if Coverage B is purchased. A separate provision of the policy requires that the policyholder give notice of occurrence ‘as soon as practicable’ once certain employees become aware of an occurrence ‘likely to involve this Policy’.6 2.08

Although there is a natural tendency to focus on the possibility that notice might be given late, questions may arise, depending upon which Bermuda Form is being considered, as to whether a policyholder can give notice early. The problem of early notice is best appreciated by imagining a policyholder who has decided not to extend Coverage A. By giving notice prior to its expiry of every conceivable claim, whether or not each claim appeared at that stage likely to involve the policy, the policyholder could (in effect) obtain coverage (subject to the policy’s temporal limitations) against that and any similar claims in perpetuity and for no premium. For instance, a policyholder might have been sued for US$1,000 by one person alleging a particular defect in a product. By giving notice of a ‘batch’ or ‘integrated’ occurrence prior to the expiration of Coverage A, the policyholder could try to argue that, if subsequent claims emerged,7 all of them would be treated together as a single occurrence and all would be covered – even without purchasing Coverage B.

THE PERIOD OF COVER: COVERAGE A AND B 2.09

It is the practice of insurance companies writing on the Bermuda Form to conduct annual reviews with their policyholders and to have meetings with them which are often called ‘renewal’ meetings. There is therefore a tendency to think of the policy as being renewed annually. Strictly speaking, however, what happens under the Bermuda Form is that Coverage A of the Policy is being extended. The Bermuda Form is not framed in terms of a period during which the contract subsists, but in terms of the period during which the coverage exists. It can be thought of either as a continuing contract, which subsists for so long as any coverage under it subsists, or as a contract that has continuing effects, dependent upon the period of the coverages it defines.8 The coverage continues until cancelled; separate premiums and limits apply to each ‘Annual Period’.

2.10

The total coverage period is divided into two distinct parts. Coverage A is the primary period. It commences, in effect, on the inception date or, if the parties have so agreed, on the retroactive coverage date. Thus, the policy will, if triggered, provide coverage for bodily injury or property damage taking place on or after the inception or retroactive coverage date. The coverage period continues from year to year, in ‘Annual Periods’,9

6 Article V of the 004 Form and Article V(d) of earlier Forms. See ch 8 below. 7 If the relevant occurrences took place during the period of Coverage A, there would be coverage for them. If the relevant occurrences, forming part of the batch, took place after Coverage A had expired, questions arise as to whether these can nonetheless form part of the overall batch: see below ch 6, paras 6.37–6.46. 8 See, eg, (i) the cancellation condition, Article VI(L) of the 004 Form and Article V(m) of earlier Forms; (ii) the policy extension condition, Article VI(Q) of the 004 Form and Article V(s) of earlier Forms; (iii) the discovery period condition, Article VI(S) of the 004 Form and Article V(u) of earlier Forms. These and related conditions are discussed in ch 11 below. 9 Article III(C) of the 004 Form and Article III(l) and (n) of earlier Forms.

The Period of Cover: Coverage A and B

23

until the coverage expires upon cancellation10 or the agreement of the parties not to extend the policy. Coverage B begins when Coverage A ends and provides what is often thought of as an extended reporting period for claims that arise after the expiration of Coverage A. It too continues for Annual Periods for as long as the policyholder wishes to purchase it, at premiums fixed by the policy based on the premium in the last period for which Coverage A existed. The primary differences between Coverage A and Coverage B are as follows. Coverage A can be extended only by mutual agreement. Coverage B can be obtained or extended at the option of the policyholder. It follows, too, that while the insurance company can use negotiations to extend Coverage A as an opportunity to impose revised premiums or terms, it cannot do so for extensions of Coverage B; the terms of Coverage B are fixed on the expiry of Coverage A.11

2.11

Coverage under the contract is for occurrences which are subject to the temporal limitations described above. For example, in relation to personal injury and property damage arising from the policyholder’s products, the damage must take place during Coverage A, and damage that takes place prior to the commencement of Coverage A is not covered. Coverage B does not extend the temporal limitations during which an occurrence must take place. Instead, it is an extended period in which to report occurrences that took place during the relevant time period prescribed for occurrences; hence its alternative name, the ‘discovery’ period.12

2.12

During Coverage A, a new aggregate limit of the policy is ‘reinstated’ every year, as provided in Article VI(R). The 004 Form provides in Article VI(R) for optional reinstatement of the aggregate limit under certain circumstances.13 During Coverage B, there is no yearly reinstatement; there is a single aggregate limit for all the occurrences reported during the last year of Coverage A and the whole of Coverage B.14

2.13

Because there is no obligation on the part of the insurers to extend Coverage A, the expiry of each Annual Period provides an opportunity for the insurers to reconsider whether, and if so on what terms, they are prepared to extend cover for a further Annual Period. Article VI(Q) expressly refers to such extensions being made after agreement of premium and other terms and conditions between the policyholder and the insurer.15

2.14

10 Article VI(L) of the 004 Form and Article V(m) of earlier Forms. The policy might be cancelled for a number of reasons, including non-payment of the premium and the commencement of proceedings in breach of the arbitration clause. Unless cancellation is for non-payment of the premium or breach of the arbitration clause, the effect of cancellation is to terminate Coverage A, but leave the policyholder with the option of purchasing Coverage B with effect from cancellation. 11 Article VI(S) of the 004 Form and Article V(u) of earlier Forms. The premium payable is set out in Schedule D. 12 Article III(J) of the 004 Form and Article III(m) and (o) of earlier Forms; Article VI(S) of the 004 Form and Article V(u) of earlier Forms. 13 The 004 Form sets forth specific terms and conditions that apply to such reinstatements in Article VI(R)(2)(a)–(c). For example, the policyholder must elect this reinstatement in writing, identifying the amount reinstated (but not in excess of the original aggregate limit) and paying a reinstatement premium calculated pursuant to Article VI(R)(2)(a). See further ch 11, paras 11.97–11.101 below. 14 Article VI(R) of the 004 Form and Article V(t) of earlier Forms. 15 Article V(s) of earlier Forms. The Cancellation Condition (Article VI(L) of the 004 Form and Article V(m) of earlier Forms) seems to suggest that, unless the policyholder positively gives notice to cancel at the anniversary date, coverage will continue, presumably on the same terms as before. It is not altogether easy to dovetail this provision with the policy extension condition (Article VI(Q) of the 004 Form and Article V(s) of earlier Forms), which appears to make extension conditional upon positive agreement as to the terms for the next Annual Period. In practice Article VI(Q)/Article V(s) always gives the insurer the right to reconsider the terms on which cover will be extended.

24

The Bermuda Form: Its Basic Structure

It is in this loose sense that the policy is subject to annual ‘renewal’ discussions. If those discussions result in any change to the policy terms (for example, an increase in the retention or in limits), whether the new or the old terms apply to a particular occurrence will depend on when notice of that occurrence is given. Changes will not apply to occurrences of which the policyholder has already given notice to the insurer.16 Changes will apply (absent some other agreement) to occurrences which have already taken place, but of which the policyholder has not already notified the insurer.

LIMITS 2.15

The Bermuda Form has limits which apply (i) to each occurrence, (ii) to all occurrences reported in any annual period during Coverage A, and (iii) to all occurrences reported during the last year of Coverage A and the whole of Coverage B.

2.16

Thus, Article II17 imposes a limit of liability in respect of ‘each occurrence’ as set out in Declaration 2(a). Because of the ‘batching’ or ‘integrated occurrence’ provisions, multiple injuries and claims arising from the same cause may constitute a single occurrence. Article II also imposes an overall limit of liability for all occurrences reported during each Annual Period, and during the whole discovery period and its immediately preceding Annual Period (that is, the last Annual Period of Coverage A and the whole of the period of Coverage B). During the period of Coverage A, a separate aggregate limit applies annually, as discussed above. The limits operate for all parties insured under a single policy collectively, not for each insured.

THE EXCESS POINT 2.17

The Bermuda Form is, as various provisions emphasise, intended to operate as ‘excess’ insurance.18 It does not, however, simply operate with a fixed excess point; it is best regarded as ‘floating’ above a fixed minimum excess point. Thus, Article II(A)(2) of the 004 Form19 makes the insurer liable only for such portion of ultimate net loss as exceeds the per occurrence retention set forth in the Declarations. This is a fixed monetary sum and represents the minimum excess point under the policy. This is because Article II(A)(1) of the 004 Form20 makes the insurer liable only for such portion of the ultimate net loss as exceeds the greater of the ‘limits of the … underlying insurances and any self-insured retentions listed, or which should have been listed, on the present and/or any prior Schedule B’ or ‘the per occurrence retention amount listed in Item 2 of the Declarations’. Schedule B lists a series of (known) insurance policies that are regarded as ‘underlying’. 16 Article II(A) of the 004 Form and the concluding words of Article I of earlier Forms. See further ch 8 below. 17 See ch 9 below. 18 See, eg, the ‘Other Insurance’ and ‘Subrogation’ Conditions: Article VI(H) of the 004 Form and Article V(i) of earlier Forms; Article VI(I) of the 004 Form and Article V(j) of earlier Forms. The excess point and the ‘other insurance’ provisions are considered in more detail in ch 9 below. 19 Article II(a)(2) of earlier Forms. 20 Article II(a)(1) of earlier Forms.

The Exclusions

25

Until those policies and the retention have been exhausted, the Bermuda Form policy does not respond. Article VI(H) of the 004 Form21 provides that if there are other ‘valid and collectable’ insurances – whenever issued – available to respond to a particular loss, then the Bermuda Form policy ‘shall be in excess of and shall not contribute with such other insurance’. The effect of this provision is that if there are insurances not listed on Schedule B, the excess point for the policy in effect ‘moves up’ to make the policy excess of those other insurances. This is intended to have two effects: first, that the policyholder cannot make a claim under the policy for any loss covered by other ‘valid and collectable’ insurance until that insurance has been exhausted; and, secondly, that other insurers cannot make a claim against those writing on the Bermuda Form for contribution. The policy also attempts to spell out what will happen if this second objective fails to be achieved.

2.18

THE INCEPTION AND RETROACTIVE COVERAGE DATES The inception date is explicitly set out in the policy’s Declarations, as Declaration 3. It is used for various purposes. It is the start of the first Annual Period, and its anniversary is the start of each subsequent Annual Period.22 Unless a retroactive coverage date is agreed, the inception date is the date on or after which damage must take place if it is to form part of an occurrence under the second paragraph of the occurrence definition, and the date on or after which an ‘event’ or ‘exposure to conditions’ must take place in order for the consequences of that event or conditions to form an occurrence under the first paragraph.23

2.19

The ‘retroactive coverage date’,24 if it exists, is a date (ex hypothesi earlier than the inception date) that takes the place of the inception date for the purpose of determining whether damage forms part of an occurrence. The availability of retroactive coverage may be attractive to a policyholder who, for example, wishes to change insurers and wants to ensure that there is no gap in coverage between the old insurance and the new.

2.20

THE EXCLUSIONS The Bermuda Form includes a lengthy list of exclusions, which are set out in Article IV of the Form. The applicability of the exclusions in any particular case gives rise to a mix of issues which are discussed in Chapter 10.

21 Article V(i) of earlier Forms. 22 See the definition of ‘Annual Period’ in Article III(C) of the 004 Form and Article III(l) and (n) of earlier Forms. 23 Article III(V) of the 004 Form and Article III(e) of earlier Forms. 24 Article III(AB) of the 004 Form. The expression is used, but not specifically defined, in earlier Forms.

2.21

26

The Bermuda Form: Its Basic Structure

THE CONDITIONS 2.22

The Bermuda Form also contains a lengthy list of conditions, which interrelate with other parts of the policy.25 These conditions are largely ‘purpose built’ in that, unlike some of the exclusions, their origins are not to be found in other policy forms. The governing law and arbitration provisions are particularly important, and they give rise to the issues considered in particular in Chapters 3, 14, 15 and 16.

SCHEDULES AND ENDORSEMENTS 2.23

The Bermuda Form has several form Schedules. Schedule A identifies additional insureds. Schedule B identifies underlying insurance. Schedule C identifies insured watercraft. Schedule D identifies factors used to calculate premiums applicable to Coverage B.

2.24

Bermuda Form policies, like other insurance policies, often include endorsements that supplement or amend provisions in the body of the policy or identify additional insured entities.26

25 Article VI of the 004 Form and Article V of earlier Forms. 26 New York cases state the common sense point that policy language modified by endorsement ‘generally speaking, takes precedence over the form policy language’. See, eg, Allianz Ins Co v Otero 353 F Supp 2d 415, 424–25 (SDNY 2004) (citing Rhinebeck Bicycle Shop, Inc v Sterling Ins Co 151 AD2d 122, 125 (NY App Div 1989)).

3 Choice of Law Issues under the Bermuda Form Insurance policies written on the Bermuda Form have a distinctly international character. The policyholder and the insurers are often based in different countries. The policy includes an express choice of New York law (with certain modifications) and invariably provides for disputes to be settled by arbitration, usually in England and sometimes in Bermuda. The thrust of the contractual provisions can be gleaned from the ‘Construction and Interpretation’ and ‘Arbitration’ clauses in the current version of the Bermuda Form.1 The former is a choice of law provision which provides: This Policy, and any dispute, controversy or claim arising out of or relating to this Policy, shall be governed by and construed in accordance with the internal laws of the State of New York, except insofar as such laws: (1) may prohibit payment in respect of punitive damages hereunder; (2) pertain to regulation under the New York Insurance Law or regulations issued by the Insurance Department of the State of New York pursuant thereto, applying to insurers doing insurance business, or issuance, delivery or procurement of policies of insurance, within the State of New York or as respects risks or insureds situated in the State of New York; or (3) are inconsistent with any provision of this Policy; provided, however, that the provisions, stipulations, exclusions and conditions of the Policy are to be construed in an evenhanded fashion as between the Insured and the Company; without limitation, where the language of this Policy is deemed to be ambiguous or otherwise unclear, the issue shall be resolved in the manner most consistent with the relevant provisions, stipulations, exclusions and conditions (without regard to authorship of the language, without any presumption or arbitrary interpretation or construction in favor of either the Insured or the Company or reference to the ‘reasonable expectation’ of either thereof or to contra proferentem and without reference to parol or other extrinsic evidence). To the extent that New York law is inapplicable by virtue of any exception or proviso enumerated above or otherwise, and as respects arbitration procedure pursuant to Condition N, the internal laws of England and Wales should apply.

The arbitration provision (Condition N) provides that: Any dispute, controversy or claim arising out of or relating to this Policy or the breach, termination or invalidity thereof shall be finally and fully determined in London, England under the provisions of the Arbitration Acts of 1950, 1975 and 1979 and/or any statutory modification or

1 Bermuda

Form 004, Articles VI(N) and VI(O).

3.01

28

Choice of Law Issues under the Bermuda Form

amendments thereto, for the time being in force, by a Board composed of three arbitrators to be selected as follows …2

3.02

It might at first sight seem somewhat eccentric for a standard form insurance policy to provide for disputes under a contract governed by New York to be arbitrated in London. The historical background to the Bermuda Form explains this. The insurers were anxious to avoid US courts, which were perceived, rightly or wrongly, as too pro-policyholder. Arbitration was seen as potentially more efficient and less expensive than US court litigation, and arbitration awards do not create legal precedent. The insurance companies typically appointed English barristers or retired judges as their party-appointed arbitrators. This was perhaps because London arbitration offered the industry the opportunity to have their new policy form interpreted by arbitrators who would come to the policy form afresh and without any preconceptions that might be derived from extensive involvement in insurance disputes in the United States. By providing for New York law in confidential London arbitration, the chances of the Bermuda Form being subjected to interpretation by the courts (either in the United States or England) would be minimised if not negated. In particular (as discussed later in this chapter), appeals to the English courts on issues of law would not be possible, since no question of English substantive law would ever be involved. On the other hand, buyers of insurance (envisaged at the outset as being the large US corporations) were not required to submit to English law as the substantive law of the contract. Instead, the policy provided for the law of New York – a state whose insurance law was perceived as less pro-policyholder than others, but was still more benevolent to policyholders than English law in certain respects3 and was familiar to companies based in the United States.

3.03

This chapter concerns some of the consequences of this international character of Bermuda Form arbitrations, and the respective roles that the various systems of law may have over questions that arise in the course of resolving a dispute. Since the important demarcation is normally between English law, which is the centre of gravity of the dispute resolution provisions, and New York law, which is the centre of gravity of the provisions of substantive law, reference will mostly be to these systems of law. However, some versions of the Form have provided for Bermuda arbitration, so that reference to ‘England’ and ‘English law’ should sometimes be read as references to ‘Bermuda’ and ‘Bermudian law’.4

IDENTIFYING THE RELEVANT GOVERNING LAW 3.04

There are six main areas in which it may be necessary to identify a governing law for the purposes of a dispute under the Bermuda Form. First, if the question arises as to whether 2 The clause then provides in detail for the manner of appointment of arbitrators and for various other matters relating to the arbitration. 3 Notably in the law relating to non-disclosure and misrepresentation. 4 For example, ACE’s Form at one time provided for arbitration in Bermuda under Bermudian law. Parties may agree to change the governing law. Doing so can affect how (or whether) the coverage applies and deserves careful consideration. See, eg, AstraZeneca Insurance Co Ltd v XL Insurance (Bermuda) Ltd and ACE Bermuda Insurance Ltd [2013] EWCA Civ 1660, [2014] 2 All ER (Comm) 55, (choice of law revised to English law and arbitration clause waived by the parties).

Identifying the Relevant Governing Law

29

there is a valid arbitration agreement, or whether a particular dispute is covered by it, it is necessary to identify the law governing the arbitration agreement. Second, in the course of the arbitration, it may be necessary for the tribunal to identify the legal rules and principles which govern its conduct of the arbitration, referred to here as the ‘internal procedural law’ of the arbitration. Third, it may be necessary to identify the system of law governing judicial supervision of the arbitration process, and the courts having jurisdiction to supervise the process. That would happen, for instance, if problems arose concerning the appointment of an arbitrator or there was some challenge to the arbitration tribunal’s conduct of the proceedings. We refer to this system of law as the ‘curial law’. Fourth, it may be necessary to identify the legal rules that govern any substantive issues of conflict of laws that arise in the course of the arbitration proceedings – that is to say, the system that dictates the choice of law rules to be applied by the tribunal. We refer to those rules as the ‘law governing choice of law’. Fifth, it is necessary to identify the legal rules governing the dispute. We refer to this as the ‘applicable substantive law’. Sixth, it may be necessary to consider the legal rules that governed the legal proceedings out of which the insured’s liability arose. We refer to this last set of rules as the ‘law governing the underlying claim’. This taxonomy is complicated enough as it is, but even so it might be argued that it is technically incomplete. At least in England, writers have identified a number of other systems of law that may apply in an arbitration.5 For present purposes, at least, the division set out above is sufficient.

3.05

The Law Governing the Arbitration Agreement Most legal systems – including all those likely to have a bearing on a dispute under the Bermuda Form6 – now recognise the doctrine of the ‘separability’ of an arbitration clause. Under this doctrine, an arbitration clause is treated for various contractual purposes as if it were a contract that is separate from the rest of the contract in which it is physically embedded. The result is that the arbitration agreement may be valid even if the contract in which it is found is void or voidable (for instance, because of misrepresentation or because it infringes public policy). The arbitration agreement may survive the termination of the substantive contract (for instance, its frustration). It also follows that the validity

5 See Michael J Mustill and Stewart C Boyd, Commercial Arbitration, 2nd edn (London, Butterworths, 1989) 60–62; Michael J Mustill and Stewart C Boyd, Commercial Arbitration: 2001 Companion (London, Butterworths, 2001) 122; cf Lord Collins of Mapesbury et al (eds), Dicey, Morris & Collins: The Conflict of Laws, 15th edn (London, Sweet & Maxwell, 2012) vol 1, 16-006–16-010 (hereafter Dicey, Morris & Collins: The Conflict of Laws). 6 England: see English Arbitration Act 1996, s 7; Fiona Trust & Holding Corp v Privalov [2007] UKHL 40, [2007] 4 All ER 941; Lesotho Highlands Development Authority v Impregilo SpA [2005] UKHL 43, [2006] 1 AC 221, [21] (Lord Steyn) (describing separability as ‘part of the very alphabet of arbitration’); and (at common law) Bremer Vulkan Schiffbau und Maschinenfabrik v South India Shipping Corporation Ltd [1981] AC 909; Harbour Assurance Co (UK) Ltd v Kansa General International Insurance Co Ltd [1993] QB 701. Bermuda: UNCITRAL Model Law, Article 16(1). For the position under the US Federal Arbitration Act (9 USC §§ 201–16), see Prima Paint Corp v Flood & Conklin Mfg Co 388 US 395 (1967); Buckeye Check Cashing, Inc v Cardegna 546 US 440 (2005).

3.06

30

Choice of Law Issues under the Bermuda Form

and interpretation of the arbitration agreement is not necessarily governed by the same legal rules as govern the substantive provisions of the contract. Where, therefore, some question arises as to the validity or meaning of the arbitration agreement, it becomes necessary to identify the law governing that arbitration agreement. 3.07

Under the Bermuda Form, there are two obvious candidates: New York law, as the main applicable law for the substantive parts of the policy,7 or English law, as the law governing the procedural aspects of the arbitration.8 How the choice between these two candidates is approached will depend on where the issue arises, since it depends on the choice of law rules of the forum deciding the question. An English court, therefore, will apply English law choice of law rules to issues concerning the validity or interpretation of the arbitration agreement.9 An arbitration tribunal appointed under the Bermuda Form would also probably be bound to apply English choice of law rules in deciding any question as to its jurisdiction.10 To do otherwise would make little sense because either party would have the right to challenge the tribunal’s decision on such an issue before the English court,11 which will naturally apply its own choice of law rules. But if an issue as to the validity of the arbitration clause arises in, say, a US court – perhaps in the context of an application to stay proceedings brought against a Bermuda Form insurer under the Federal Arbitration Act12 – the US court typically would apply its own choice of law rules to decide whether the arbitration agreement is valid and enforceable and whether a particular dispute falls within it.13 In that sense, there is no universally definitive answer to the question ‘What law governs the arbitration agreement?’ since the way that this question is approached will depend on where it arises. Below we consider the English law rules, as they are the rules that would be applied by any arbitration tribunal called upon to decide its own jurisdiction.

7 See below paras 3.20–3.21. 8 See below paras 3.12–3.15. 9 English common law. The Rome Convention (which applies by virtue of the Contracts (Applicable Law) Act 1990 to questions concerning choice of law in contracts made before 17 December 2009) and the Rome I Regulation, Regulation (EC) 593/2008, [2008] OJ L177/6, (which, prior to the United Kingdom’s departure from the European Union, applied directly as a matter of EU law) are both expressly inapplicable to arbitration and jurisdiction clauses: Rome Convention, Article 1(2)(d); Rome I Regulation, Article 1(2)(e). One is therefore left with the common law rules. It should be noted, however, that the rules in the Rome Convention and the Rome I Regulation have (subject to some amendments) been retained as part of English law notwithstanding the United Kingdom’s departure from the EU: see the Law Applicable to Contractual Obligations and Non Contractual Obligations (Amendment etc.) (EU Exit) Regulations 2019 (SI 2019/834). For a short summary of the position after Brexit, see Civil Procedure (generally known as the ‘White Book’) (London, Sweet & Maxwell, 2021) vol 1, para 6HJ.18. 10 Not as to the substance of the case. As to that, s 46 of the Arbitration Act 1996 provides its own choice of law regime: see para 3.18 below. But a jurisdictional question is different, because there is an unfettered right under s 67 of the Act to have such cases decided de novo by the court, so it would make little sense for a tribunal to apply choice of law rules other than those that the court would be bound to apply. 11 English Arbitration Act 1996, s 67. Under the English Arbitration Act 1996, a tribunal may, and normally will, decide any issue as to its own jurisdiction that arises in the course of the arbitration: ss 30, 31. However, its decision is never conclusive, but is always subject to judicial challenge. 12 9 USC §§ 201, 216. 13 Ibid § 2 (validity) and § 3 (scope).

Identifying the Relevant Governing Law

31

In practice, this question has most often been confronted when one party does something, such as bringing proceedings in a court, which the other argues is a breach of the arbitration agreement. Most systems of law provide ways of requiring parties to abide by an agreement to arbitrate.14 English courts, however, have not been inclined to leave such matters to the law of the place where the inconsistent proceedings are started, but have been willing to exercise a jurisdiction to enjoin persons who agreed to arbitration in England from acting contrary to their agreement.15 In that context, it may be necessary to identify the law that governs the arbitration agreement in order to decide whether the proceedings that the court is being asked to enjoin are or are not being brought in breach of a valid and binding agreement to arbitrate.16

3.08

One might think that was a simple question which would admit a simple answer. But it has produced a tangle of rather unsatisfactory cases, culminating in the Supreme Court’s decision in Enka Insaat Ve Sanayi AS v OOO Insurance Company Chubb.17 The court stressed that there are three stages to the necessary analysis: (a) have the parties expressly chosen a law to govern the agreement; (b) if not, have they impliedly done so; and (c) if not, with what system of law is it most closely connected? In general, if there is an express choice of law to govern the contract, the presumption is that it governs all the terms of the contract including the arbitration provision. Although depeçage (having some terms of a contract governed by one system of law and some by another) is possible, it is not lightly presumed, and the separability of the arbitration agreement is not itself a reason to presume it.18 Partly because of separability, however, the presumption against depeçage is weaker for arbitration clauses than for other provisions.19 But the mere fact that the parties select a seat different from the country whose law governs the agreement is not itself sufficient grounds for concluding that they intended a different choice of law.20

3.09

Where does that leave the Bermuda Form, which had been the progenitor of certain cases which had adopted an analysis which linked the choice of law to the choice of seat?21 In our view, the reasoning of those cases cannot stand. The prima facie position,

3.10

14 For instance, by a stay of proceedings under s 9 of the English Arbitration Act 1996 or by the mechanisms provided by the Federal Arbitration Act, eg, 9 USC § 206. 15 See, eg, Angeliki Charis Compania Maritima SA v Pagnan SpA (The Angelic Grace) [1995] 1 Lloyd’s Rep 87; West Tankers Inc v RAS Riunione Adriatica di Sicurta SpA (The Front Comor) [2007] UKHL 4, [2007] 1 Lloyd’s Rep 391; Enka Insaat Ve Sanayi AS v OOO Insurance Company Chubb [2020] UKSC 38, [2020] 1 WLR 4117, [173]–[185]. 16 Enka Insaat ve Sanayi AS v OOO Insurance Company Chubb [2020] EWCA Civ 574 [2020] 2 Lloyd’s Rep 389 [53] (‘The anti-suit injunction jurisdiction is concerned to protect and enforce the integrity of the arbitration agreement. In order to do so it must necessarily interrogate the substantive jurisdiction of the arbitration tribunal’) and [64]. In this respect, the approach of the Supreme Court was similar to that of the Court of Appeal: [2020] UKSC 38, [2020] 1 WLR 4117, [173]–[185]. 17 [2020] UKSC 38, [2020] 1 WLR 4117. 18 Ibid at [40]. 19 Ibid at [40], [41]. 20 Ibid at [43]–[45], [53]. 21 XL Insurance Co Ltd v Owens Corning [2001] 1 All ER Comm 530; C v D [2007] EWCA Civ 1282, [2008] 1 WLR 239. The Bermuda Form has been a fertile source of disputes about this because insurance companies with arbitration clauses in their policies, who are sued in the United States, often seek anti-suit injunctions from the English courts on the grounds that such arbitration agreements have been breached.

32

Choice of Law Issues under the Bermuda Form

therefore, is that the arbitration agreement in the Bermuda Form is governed by New York law. There are two possible complications. First, it might be argued that the express reference to the (English) arbitration acts makes a difference. But that seems improbable, since the Supreme Court poured cold water on the idea that the choice of curial law was of significance.22 Second, the Supreme Court did accept that if an arbitration agreement would be invalid under the law chosen to govern the agreement as a whole, that might be a useful indication of separability.23 However, that is a fairly unpromising approach, since New York law (as opposed to the laws of certain other states in the United States) does not invalidate arbitration agreements in insurance contracts. At first sight, XL v Owens Corning24 might look like such a case. But the argument there was highly case-specific (namely that New York’s interpretation of the requirement that an agreement be ‘in writing’ was more testing than the standard applied in England to the same question). 3.11

If that is right, then the correct conclusion is probably that the arbitration agreement in the Bermuda Form is governed by New York law, like the rest of the agreement. This, however, will not put paid to the cottage industry of anti-suit injunctions. In Enka, the Supreme Court made it clear that the courts of the seat could be expected to enforce arbitration agreements (if valid under their proper law).25 English law gives the High Court ‘long arm’ jurisdiction for this purpose,26 It is routinely and enthusiastically exercised, and may be exercised on an ex parte basis. The justification for not giving notice of the application is usually that the other party (usually the policyholder) is likely to seek an injunction of its own (an ‘anti anti-suit injunction’) if it is provided with notice. Moreover, defendants to such proceedings will be alert to the fact that their superficially most promising defence – that the agreement is not valid under its proper law – is dangerous: if successful, it may point towards the conclusion that, after all, English law governs. The Internal Procedural Law of the Arbitration

3.12

There is no doubt that the internal procedural law applicable to any arbitration under the Bermuda Form is that found in the English Arbitration Act 1996. This is expressly referred to in the Bermuda Form (either directly or, in early versions of the Form, by reference to any statutory ‘modifications or amendments’ of the Arbitration Acts 1950, 1975 and 1979). In any case, the Act is of mandatory application to any arbitration whose ‘seat’ is in England,27 and the English courts have supervisory powers to ensure that it is applied.

22 Enka (above n 15) at [94]. 23 Ibid at [96]. 24 [2001] 1 All ER Comm 530. 25 See above n 15. 26 Civil Procedure Rules 1998, Rule 62.5. 27 English Arbitration Act 1996, ss 2(1), 3. But it is not necessary that all the hearings and deliberations of the tribunal should take place in England. The tribunal may decide to hold hearings at any convenient location without thereby varying the ‘seat’.

Identifying the Relevant Governing Law

33

The statutory rules governing the conduct of arbitration proceedings under the English Arbitration Act 1996 are found in sections 33–41 of the Act. They give very considerable leeway to the parties and the tribunal. There is a minimal framework of mandatory provisions.28 The tribunal must ‘act fairly and impartially’ and ‘adopt procedures suitable to the circumstances of the case, avoiding unnecessary delay and expense, so as to provide a fair means for the resolution of the matters falling to be determined’.29 The parties must co-operate by doing ‘all things necessary for the proper and expeditious conduct of the arbitral proceedings’, including complying ‘without delay’ with any procedural directions given by the tribunal.30 These leave a wide field within which the parties and the tribunal are largely31 free to shape the arbitral procedure as they think fit. The Act provides various default powers and rules, but these should be modified to meet the particular circumstances of the dispute.

3.13

A central provision of the Act is section 34(1), which gives the tribunal power to decide ‘all procedural and evidential matters’. That power expressly includes matters such as whether there should be pleadings, whether documents should be disclosed (and if so on what basis), whether the strict rules of evidence should be applied, and how submissions should be made.32 The Act therefore does not require that the tribunal should follow the procedures or practices that would be applied in an English court, and as a practical matter the procedures adopted are normally different in various respects. For instance, an English court would require any question of foreign law (including New York law) to be ‘proved’ as a matter of fact by calling expert witnesses who would be examined and crossexamined.33 Although an arbitration tribunal could in theory follow this procedure, and arbitrations have followed this procedure, the usual modern practice in arbitrations under the Bermuda Form is to deal with New York law by way of submission rather than to hear ‘evidence’ on it.34

3.14

This chapter does not consider the practical aspects of the conduct of a Bermuda Form arbitration.35 The distinction between issues of ‘procedure’ (governed by English law) and issues of ‘substance’ (governed by New York law) is, however, discussed in more detail below.36

3.15

28 English Arbitration Act 1996, s 4(1) and Sched 1. 29 Ibid, s 33. 30 Ibid, s 40. 31 There are a few other mandatory provisions of the Act relevant to procedure during the course of a reference, involving what constitutes an expense of the arbitration (s 37(2)), the availability of compulsory process to secure the attendance of witnesses (s 43), the power to withhold an award if the tribunal’s fees have not been paid (s 56) and the award of costs (s 60). 32 English Arbitration Act 1996, s 34(2). The list is not exhaustive. 33 See Dicey, Morris & Collins: The Conflict of Laws, vol 1, ch 9. 34 For judicial recognition of this practice, see Reliance Industries Ltd v Enron Oil and Gas India Ltd [2002] 1 Lloyd’s Rep 645, 649 fn 8. 35 This topic is covered in chs 14–16 below. For a general description, including cultural differences that might surprise US attorneys arbitrating in London, see J Dasteel and R Jacobs, ‘American Werewolves in London’ (2002) 18 Arbitration International 165. 36 See below paras 3.30–3.45.

34

Choice of Law Issues under the Bermuda Form

Curial Law: Judicial Supervision of the Arbitral Process 3.16

Because the juridical seat of the arbitration is in England,37 the arbitration is subject to the supervisory jurisdiction of the English High Court under the English Arbitration Act 1996.38 The English court would exercise, if necessary, its powers to appoint or remove arbitrators.39 It would deal with any issue as to the tribunal’s jurisdiction40 and with any complaint of serious procedural irregularity in the course of the hearing.41 The most famous peculiarity of the English supervisory jurisdiction – the possibility of an appeal to the court on a point of law42 – will not arise: in the first place because such an appeal is expressly excluded by agreement;43 and in the second place because ‘New York law’ would be regarded, for these purposes, as fact rather than law.44

3.17

The agreement to arbitration in England is treated by the English courts as amounting to an agreement only to seek remedies from the English courts in relation to such matters.45 The Court of Appeal has approved first instance authority describing the choice of an arbitral seat as ‘analogous to an exclusive jurisdiction clause’ so that any claim for a ‘remedy going to the existence or scope of the arbitrator’s jurisdiction or as to the validity of an existing interim or final award is agreed to be made only in the courts of the place designated as the seat of the arbitration’.46 An application to any other court may be enjoined. This would not, presumably, preclude either an application to enforce an award outside England or any defence to such an application. But it will preclude active steps being taken outside England by way of positive challenge to an arbitration award. Law Governing Choice of Law

3.18

The choice of law rules applicable to a given dispute are normally regarded as a matter for the law of the forum.47 In the context of an international arbitration taking place in England and governed by the English Arbitration Act 1996, the starting point is the Act itself, which contains an express provision for international arbitrations, different 37 See para 3.12 above. 38 English Arbitration Act 1996, s 2(1). 39 Ibid, ss 18–19 (appointment), 24 (removal). 40 Ibid, ss 32, 67. Section 32 enables an application to be made, with the consent of the parties or the tribunal’s permission, in the course of the reference. Section 67 relates to challenges to an award on the ground that it was made without jurisdiction. 41 Ibid, s 68. 42 Ibid, s 69. 43 Ibid, s 69(1): ‘Unless otherwise agreed by the parties’ an appeal lies. 44 English Arbitration Act 1996, s 82(1); Egmatra AG v Marco Trading Corp [1999] 1 Lloyd’s Rep 862, 865; Sanghi Polyesters Ltd (India) v The International Investor (KCFC) Kuwait [2000] 1 Lloyd’s Rep 480, 483. Where the parties prefer to proceed on the basis that the chosen foreign law is the same as English law (but without agreeing to vary the applicable law), the resulting decision is not a decision of ‘law’ for the purposes of an appeal: Reliance Industries Ltd v Enron Oil and Gas India Ltd [2002] 1 Lloyd’s Rep 645. 45 C v D (above n 21) at [16]–[17]. This aspect of the case was not doubt in Enka (above n 15). 46 A v B (No 2) [2006] EWHC 2006 (Comm), [2007] 1 Lloyd’s Rep 237, para [111], approved in C v D (above n 21) at [17]. 47 Subject to renvoi, which does not apply in contract cases. See below n 53.

Identifying the Relevant Governing Law

35

(in some respects) from the rules that would be applied by an English court.48 For present purposes, two provisions matter. First, under section 46(1), the tribunal is directed to apply the law chosen by the parties. One need not, under the Bermuda Form, look any further, therefore, than the express choice of New York law. Second, under section 46(1)(b), the Act permits the parties to opt for a decision to be made according to non-legal principles, should they wish to do so. That is significant, in this context, because it removes any lingering doubt about the permissibility of adopting a ‘pick and mix’ approach to choice of law, whereby the parties do not simply choose a system of law, but adopt that system with modifications by providing for disputes to be determined ‘in accordance with such other considerations as are agreed’. Since, strictly, it is English legal rules, and specifically the statutory choice of law rules applicable in arbitrations, that govern the choice of law question, the interpretation and validity of the choice of law provisions are matters for English law, not New York law. New York choice of law rules, in particular, are irrelevant. As a matter of basic principle, a choice of contractual governing law does not include the choice of law rules of the selected legal system;49 the doctrine of renvoi is not applied to contracts.50 That position is placed beyond any doubt by the express reference to the ‘internal’ law of New York, which excludes recourse to New York’s private international law.51

3.19

Applicable Substantive Law The basic choice of law. By express choice, the contract is governed by the ‘internal laws of the State of New York’ subject to certain modifications. Why the reference to New York’s ‘internal’ law? That is not entirely clear. However, it is unlikely to mean that New York state case law should apply to the exclusion of decisions by the federal trial and appellate courts applying New York law. In the United States, insurance law is largely a subset of general contract law and is a matter of state rather than federal law.52 It therefore seems more likely that the reference to the ‘internal’ law of the state of New York is intended to make it clear that New York’s private international law rules are not to be applied.53 48 An English court will apply the Rome I Regulation (above n 9) in respect of contracts made after 17 December 2009, since this (subject to relatively minor modifications) has been retained as English law notwithstanding the United Kingdom’s departure from the EU. In matters related to insurance, special rules (Article 7) may apply, but they are unlikely to be significant in the case of the sort of risk to which the Bermuda Form relates; in any event, for the reasons given in the text, the provisions of s 46 of the Arbitration Act 1996 take precedence. 49 English Arbitration Act 1996, s 46(2). 50 See below n 53. 51 See further below para 3.20. 52 See, eg, Zurich Ins Co v Shearson Lehman Hutton, Inc 618 NYS2d 609, 613 (NY 1994). There is federal law in the United States on other issues. Federal courts interpret federal statutes and a federal common law on those issues exists. For example, employee benefit plans generally are subject to the Employee Retirement Income Security Act (ERISA), 29 USC §§ 1001 et seq, a federal statute. But this does not affect the issues discussed in this chapter. 53 This is consistent with the general approach taken to choice of law in contract: see, eg, Amin Rasheed Shipping Corp v Kuwait Ins Co [1984] AC 50, 61–62; Rome I Regulation (above n 9), Article 20; American Law Institute, Restatement (Second) of Conflict of Laws (St Paul, MN, American Law Institute Publishers, 1971 and Supps) s 187(4). It might therefore be objected that expressly spelling it out in the policy would be surplusage, but arguments about surplusage are always weak in commercial contracts, since drafters ‘frequently use many words … out of a sense of caution’ (Norwich Union Life Ins Soc v British Railways Board (1987) 283 EG 846).

3.20

36

Choice of Law Issues under the Bermuda Form

This reference may have sought to make clear that only the ‘local’ law of New York is to be applied, meaning its substantive law, and not New York’s ‘whole law’, which would also include choice of law rules.54 3.21

English law regards issues of initial validity of contract, and allied questions such as the effect of misrepresentation or duress upon a contract, as matters governed by the ‘putative’ applicable law; that is, the law that would be applicable assuming a contract had been made. This now has legislative warrant for cases in court,55 reflecting what appears to have been the position at common law.56 When the English Arbitration Act refers to the law chosen by the parties being used to determine ‘the dispute’, it therefore includes within the ambit of the ‘dispute’57 a dispute about the validity of the contract. Therefore, this, too, will be governed by New York law.58

3.22

Exclusion of insurance regulatory law. Recent versions of the Bermuda Form exclude New York law insofar as it ‘pertains to regulation under the New York Insurance Law’ or to regulations issued by the Insurance Department of the State of New York applying to insurers doing business in the state. The focus is exclusively on regulatory statutes. It would not, for instance, be open to the parties to rely on some breach of New York regulatory law to invalidate the policy. It does not exclude those parts of the New York Insurance Law which state or amend the law governing the rights and obligations of insurers and policyholders among themselves. For instance, it does not stand in the way of the application of those parts of the New York Insurance Law that deal with the effect of misrepresentation and non-disclosure or the effect of a breach of warranty. Drawing the line between ‘regulatory’ law and other law is not, however, always easy.59 In our view, the approach must look to the substance and not merely to the form of the law.

3.23

Modification of New York law: coverage for punitive damages. Under New York law, punitive damages60 are not insurable, as a matter of public policy.61 As a matter of construction, the parties to the Bermuda Form clearly intended punitive damages to be insured. A New York court might well decline to apply a choice of law provision which attempted to circumvent this policy, just as it might decline to enforce the policy’s extension of

54 Rejection of the ‘whole law’ helps avoid the renvoi. It is by no means unprecedented, in our experience, to find an express exclusion of renvoi in a contractual choice of law provision. 55 Contracts (Applicable Law) Act 1990, Sched 1, Article 8 (for contracts concluded prior to 17 December 2009) and Rome I Regulation (above n 9), Article 10 (for contracts concluded thereafter), both of which have been retained as English law after the United Kingdom’s departure from the EU. 56 Re Bonacina [1912] 2 Ch 394 (consideration); Mackender v Feldia [1967] 2 QB 590; Evans Marshall & Co Ltd v Bertola SA [1973] 1 WLR 349. 57 English Arbitration Act 1996, s 46. 58 In American private international law, an expressly chosen law is not necessarily applied to such issues. But that does not matter, since it is English, not New York law that governs choice of law issues. 59 See the discussion about late notice and the relevance of prejudice in para 8.17 below. 60 Usually known in English law as ‘exemplary’ damages. 61 Soto v State Farm Ins Co 613 NYS2d 352 (NY 1994); Hartford Accident & Indem Co v Village of Hempstead 422 NYS2d 47 (NY 1979); Town of Massena v Healthcare Underwriters Mut Ins Co 749 NYS2d 456 (App Div 2001); cf Zurich Ins Co v Shearson Lehman Hutton, Inc 618 NYS2d 609 (NY 1994); Home Ins Co v American Home Products Corp 551 NYS2d 481 (NY 1990). See also Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1996 and 2021 Supp) ch 8 (hereafter Masters and Stanzler, Insurance Coverage Litigation).

Identifying the Relevant Governing Law

37

coverage for punitive damages. In addition, the punitive damages for which insurance is sought may have been awarded under the law of a state other than New York; in such instances, some courts have found under New York law that this issue is governed by the law of the state making the award of punitive damages.62 But since English law governs the validity of the choice of law provision, that does not matter. In Lancashire County Council v Municipal Mutual Insurance Ltd,63 the Court of Appeal held that English law does not prohibit insurance of liability for punitive damages, even where the punitive damages are awarded by an English court. There might be difficulty if the insurance policy sought to provide indemnity where the policyholder was found to have committed a crime,64 but that is unlikely to be an issue in the sort of case with which the Bermuda Form is concerned. Lancashire County Council v Municipal Mutual Insurance Ltd was directed at a case where the liability to pay punitive damages was vicarious, and formally leaves open the question of whether a tortfeasor whose personal conduct was opprobrious could recover. However, liability normally will be vicarious where a company makes a claim under the Bermuda Form, and there seems to be little reason why a policyholder who otherwise meets the requirements for coverage (including the absence of any intention to cause injury) should not be indemnified. An arbitration tribunal is not, therefore, precluded from giving effect to the clear intention of the parties that the Bermuda Form should respond to losses caused by awards of punitive damages. Applying English law, and English public policy, to the choice of law provision, an arbitration tribunal can disregard the New York public policy that would deny recovery in such cases. If it were ever necessary to seek judicial enforcement of an award giving indemnity for punitive damages in a jurisdiction – whether New York or elsewhere – which regards such liability as uninsurable, problems might arise. But, fortunately, judicial enforcement of arbitration awards under the Bermuda Form is rarely necessary, and even if necessary, the enforcement would be unlikely to take place in New York.

3.24

Modification of the canons of construction: validity. As noted above, the choice of law provision, quoted below from the 004 Form, seeks to spell out an approach to interpretation that differs, in various ways, from the approach ordinarily taken by the New York courts when interpreting insurance contracts:

3.25

[T]he provisions, stipulations, exclusions and conditions of this Policy are to be construed in an evenhanded fashion as between the Insured and the Company; without limitation, where the language of this Policy is deemed to be ambiguous or otherwise unclear, the issue shall be resolved in the manner most consistent with the relevant provisions, exclusions and conditions (without regard to authorship of language, without any presumption or arbitrary interpretation or construction in favor of either the Insured or the Company [or reference to the ‘reasonable expectations’ of either thereof or to contra proferentem]65 and without reference to parol [or other extrinsic]66 evidence).67 62 See, eg, Hoechst Celanese Corp v National Union Fire Ins Co of Pittsburgh, Pa No 89C-SE-35, 1994 Del Super Lexis (22 April 1994). See also Masters and Stanzler, Insurance Coverage Litigation, ch 8. 63 Lancashire County Council v Municipal Mutual Insurance Ltd [1997] QB 897. 64 Ibid at 907; Gray v Barr [1971] 2 QB 554 (‘deliberate, intentional and unlawful violence’). 65 These words do not appear in the original versions of the Form. 66 These words do not appear in the original versions of the Form. 67 Article VI(N). Early Bermuda Form policies included this provision by endorsement.

38

Choice of Law Issues under the Bermuda Form

3.26

Later versions of the Bermuda Form add ‘without reference to the reasonable expectations of either of the parties’ and extend to ‘extrinsic’ as well as ‘parol’ evidence.

3.27

If this clause came before a New York court, the question might arise as to the extent to which the modification of New York law principles of contract interpretation was valid in the light of New York public policy. For example, would a policy term purporting to preclude the use of parol or extrinsic evidence be void as against public policy?68 How would a New York court react to the clause in the light of the principle that a literal construction of the policy language is not appropriate if such a construction thwarts the clear purpose of the contract or leads to an absurd result?69 Is it permissible for an insurance company drafter to oust the ordinary contra proferentem rule, which is a rule rooted in fairness?

3.28

However, all this is irrelevant where the arbitration is taking place in England, for English law makes it quite clear that the parties are free to agree to whatever rules they wish governing the contract.70 And English law would certainly not regard any of the modifications to the canons of construction as offensive to English public policy. In practice, arbitration tribunals in Bermuda Form arbitrations apply modified New York law as provided for in the clause. The effect of this modification is considered in Chapter 4 below.

68 In the light of our conclusions in paras 3.24 and 3.27, we do not address this issue in the detail which it might otherwise deserve. The general principle is that contract terms freely negotiated are valid unless they violate New York public policy. In Slayko v Security Mutual Insurance Co 98 NY2d 289 (NY 2002), the New York Court of Appeals upheld a clause in a homeowners’ policy excluding liability for criminal activity and said that ‘when statutes and Insurance Department regulations are silent, we are reluctant to inhibit freedom of contract by finding insurance policy clauses violative of public policy’. See also Joseph R Loring & Assocs Inc v Continental Cas Co 453 NYS2d 169 (NY 1982), where the New York Court of Appeals held that a claims made policy did not violate public policy. See also the authorities which indicate that parties can waive rules of evidence or in other ways make the law that will bind them: Brady v Nally 151 NY 258, 264–65 (1896); Mitchell v New York Hospital 473 NYS2d 148, 151 (NY 1984); Martin v City of Cohoes 371 NYS2d 687, 690 (NY 1975); In re Malloy’s Estate 278 NY 429, 433 (1938). In HRH Construction Corp v Bethlehem Steel Corp 412 NYS2d 366, 369 (NY 1978), the court held that it was open to parties to agree the ‘substantive rule on the basis of which the award was to be made in their arbitration proceedings’ provided that it did not ‘run afoul of public policy’. If the Bermuda Form modification of New York law were to be challenged on the grounds of violation of New York public policy, questions might arise as to whether New York public policy applied in the context of a highlevel excess cover between a policyholder and insurer neither of whom might be domiciled in New York, and who had agreed to submit their disputes to London arbitration. For authorities which could perhaps be relied upon in support of a challenge, see Cronk v State of NY 420 NYS2d 113 (Ct Cl 1979) (a clause pre-empting the court from considering legally competent evidence was held to be void as against public policy); Allstate Ins Co v White Metal Rolling & Stamping Corp 466 F Supp 419 (EDNY 1979) (an ‘entire agreement’ clause did not preclude extrinsic evidence of declarations of intention to resolve an ambiguity or equivocation). See also Lorelie S Masters, ‘Arbitration Clauses in Liability Policies: A Ticket to Ride’ (1996) 9 John Liner Rev, No 4, 33; Mitchell Dolin and Ethan Posner, ‘Understanding the Bermuda Excess Liability Form’ (1998) 1 Journal of Insurance Coverage 76–77. 69 McGrail v Equitable Life Assur Soc’y of the US 55 NE2d 483 (NY 1944); Evanston Ins Co v GAB Bus Serv 521 NYS2d 692 (App Div 1987). 70 English Arbitration Act 1996, s 46. See above paras 3.18 and 3.19. Reinsurance contracts often contain ‘honourable engagement’ or similar clauses which require the application of fairness principles rather than a system of law. Against this background, the Bermuda Form’s application of a modified system of law is neither unique nor strange.

Identifying the Relevant Governing Law

39

Law Governing the Underlying Claim Because the Bermuda Form is a liability policy, a claim under it rests upon some legal liability that has been asserted against the policyholder by third parties. As a result, it sometimes becomes necessary to consider legal systems other than England and New York as part of the factual background of the case. For example, a policyholder who is seeking to establish that a particular settlement was ‘reasonable’ may need to explain why there was a perceived risk of liability under the state law applicable to the settled claim. Or if the question arises as to whether the insured expected (or ought to have anticipated) liability of a certain magnitude, it may be necessary to consider the law that was or would have been applied to deciding those actual or potential claims.

3.29

Where this issue arises, however, it is only ever as a matter of factual background. Under English law, it may be necessary, depending upon the wording of the liability insurance policy, for the policyholder to prove that there was a liability. Where a case is settled, it is not enough to show that the policyholder faced a risk that a particular court or a particular jury would or might have reached an adverse verdict; the anticipated verdict must be shown to be, in some objective sense, ‘right’ as a matter of law.71 This approach, which rests on a questionable jurisprudential premise,72 is the product of a rarefied legal consciousness divorced from the practical realities of life before juries in US state courts and is totally foreign to New York’s approach to this question as it arises under liability insurance policies.73 The law applicable to the underlying cases is relevant only for the light it throws (which may sometimes be oblique) on the practical risks that the policyholder faced in litigation. As such, it might sometimes be helpful for a tribunal, which typically includes members who have no practical experience in civil jury litigation, to be instructed on such issues by means of expert evidence. This expert evidence can help place the legal principles in a practical context and take them out of the purely theoretical context that might exist were the tribunal simply to hear submissions on such questions. But each case varies, and the tribunal should adopt whatever procedure seems best fitted to the circumstances.

3.30

71 Skandia International Corp v NRG Victory Reinsurance Ltd [1998] Lloyd’s Rep IR 439; MDIS v Swinbank [1999] Lloyd’s Rep IR 516; Thornton Springer v NEM Ins Co Ltd [2000] Lloyd’s Rep IR 590; Structural Polymer Systems v Brown [2000] Lloyds Rep IR 64, 68; Lumbermens Mutual Casualty Co v Bovis Lend Lease Ltd [2004] EWHC 2197 (Comm), [2005] 1 Lloyd’s Rep 494 [44]; AstraZeneca Insurance Co Ltd v XL Insurance (Bermuda) Ltd [2013] EWCA Civ 1660, [2014] 2 All ER (Comm) 55 (where a Bermuda Form policy was governed by English law). 72 It assumes that there is a ‘right legal answer’ to any case divorced from the result the institutions of the system would actually generate. Cf Justice Holmes, ‘The Path of the Law’ (1897) 10 Harvard Law Review 457: ‘Take the fundamental question, what constitutes the law? The prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law.’ 73 See, eg, the decision in Luria Brothers & Co v Alliance Assurance Co 780 F2d 1082, 1085 (2d Cir 1986) (applying New York law): ‘In order to recover the amount of the settlement from the insurer, the insured need not establish actual liability to the party with whom it has settled so long as a potential liability on the facts known to the [insured is] shown to exist, culminating in a settlement in an amount reasonable in view of the size of possible recovery and degree of probability of claimant’s success against the [insured].’ See also Uniroyal Inc v Home Ins Co 707 F Supp 1368, 1378 (EDNY 1988); and ch 5 below.

40

Choice of Law Issues under the Bermuda Form

THE DIVISION BETWEEN SUBSTANCE AND PROCEDURE 3.31

Since procedural questions are governed by English law, in the form of the English Arbitration Act 1996, while substantive questions are governed by New York law, it can sometimes matter whether a particular issue is treated as substantive or procedural.74 Since that question arises as part of the choice of law process, it is a matter of English law.

3.32

Burden of proof. At common law, there was some doubt as to whether issues of burden of proof were procedural or substantive.75 In England, in court proceedings concerning contracts, the question is now resolved by legislation. The law applicable to the contract applies ‘to the extent that it contains, in the law of contract, rules which raise presumptions of law or determine the burden of proof’.76 It is for English law to decide whether a rule about burden of proof should be regarded as part of the law of contract (in which case it is governed by the applicable law) or part of the foreign legal system’s general legal rules (in which case it is not).77 This rule does not formally apply in arbitration, but the tribunal is free to decide to apply it.78 Since this legislative rule may reflect the common law79 and is consistent with the views of most commentators, it is suggested that the same approach would be taken. For most practical purposes, New York rules relating to the burden of proof will be applied, since these rules are part of the relevant law of contract. So a tribunal will apply New York law in determining whether a particular provision is to be treated as an exclusion from coverage (where the burden lies on the insurer) or part of the primary definition of coverage (where it lies on the policyholder). And it will apply New York law whereby the burden of proving material misrepresentation lies on the insurer.80

3.33

Interpretation. It is unfortunate that rules of contractual interpretation sometimes masquerade as evidential rules; it is said that a particular line of argument is barred by the ‘parol evidence rule’ or that particular material offered to assist in interpretation is ‘inadmissible’ for that purpose. It is well established that, despite this terminology, the relevant rules are rules of substance not procedure.81 What facts are relevant to interpretation is 74 The distinction between substance and procedure remains part of the orthodox English approach to the conflict of laws: see Dicey, Morris & Collins: The Conflict of Laws (above n 5) vol 1, paras 7-002–7-005 (Rule 19). In the United States, the Restatement (Second) of Conflict of Laws abandoned reliance upon the classification of rules as ‘substance’ or ‘procedure’ because it was thought to produce crude decision-making: s 122, comment (b). As will become apparent, modern English practice makes finer distinctions than a straightforward division by reference to ‘substance’ and ‘procedure’ might suggest. 75 There is authority that the burden of proof is a procedural issue: The Roberta (1937) 58 Ll L Rep 159, 177; Re the Estate of Fuld (No 3) [1968] P 675, 696–97. But Dicey, Morris & Collins: The Conflict of Laws, citing certain comments of Lorenzen with approval, consider that there is ‘much to be said for treating them as substantive’, at least where their effect is to shape substantive rights. Dicey, Morris & Collins: The Conflict of Laws (above n 5), vol 1, para 7-084. 76 Rome I Regulation (above n 9) Article 18(1). 77 See CGJ Morse, ‘The EEC Convention on the Law Applicable to Contractual Obligations’ in Yearbook of European Law, vol 2 (Oxford, Clarendon Press, 1982) 107, 156. 78 Either directly under s 46 (as a choice of law rule) or indirectly under s 34 as a procedural rule. 79 See n 75 above. 80 See, eg, First Fin Ins Co v Allstate Interior Demolition Corp 193 F3d 109 (2d Cir 1999) (applying New York law). 81 St Pierre v South American Stores Ltd [1937] 3 All ER 349, 351; AB Bofors v AB Skandia [1982] 1 Lloyd’s Rep 410, 412; Amin Rasheed Corp v Kuwait Ins [1984] AC 50. But the contrary decision in Korner v Witkowitzer [1950] 2 KB 128, aff ’d as Vitkovice v Korner [1951] AC 869 is problematic.

The Division between Substance and Procedure

41

determined by the applicable law (or such other rules as the parties have, as the English Arbitration Act 1996 permits, selected). Insofar as any question of ‘admissibility’ arises, it arises because evidence that is irrelevant should not be considered. To adopt any different classification would, as the editors of Dicey point out, ‘be tantamount to distorting the foreign law’.82 Thus, it is New York law, as modified by the parties, that governs such questions. Estoppel. In English law, estoppels are sometimes said to be rules ‘of evidence’. But it seems clear that they are not ‘ordinary’ rules of evidence. They often arise out of, affect or define substantive rights: ‘It is true that estoppels can be described as rules of evidence or as rules of public policy to stop the abuse of process by relitigation. But that is to look at how estoppels are given effect to, not at what is the nature of the private law right which the estoppel recognises and protects.’83 Although there is authority for the proposition that estoppel is an issue for the lex fori (that is, the law of the court hearing the matter), it is not unequivocal even in the context of estoppel per rem judicatam.84 It may be that some estoppels are properly regarded as substantive and some as procedural.85 Nor is it clear whether, if ‘procedural’, the discretion given to tribunals as to the procedures to be followed in arbitration extend to the question of whether an estoppel should be recognised in particular circumstances. In Ali Shipping Corp v Shipyard Trogir,86 the English Court of Appeal assumed that English law, including the rule requiring identity of the parties to found an issue estoppel, would apply to determine whether an estoppel was created by an earlier arbitration award. But it was not explained whether this was because English law was the procedural law of the arbitration or because English law applied to the substance of the claim, and it does not appear to have been argued that any other principle could be applied.

3.34

Whatever the position with regard to estoppel, principles of waiver, election and affirmation are certainly regarded as substantive principles of law, since they are so intimately linked with substantive rights. And, whether procedural or substantive, decisions about estoppel are for the arbitrators.87

3.35

Limitation. The classification of rules governing the limitation of actions has been notoriously problematic. The traditional English common law approach has been to treat (most) English limitation periods as procedural, but to recognise the possibility that a

3.36

82 Dicey, Morris & Collins: The Conflict of Laws (above n 5), vol 2, para 32-145. 83 Associated Electric and Gas Insurance Services Ltd v European Reinsurance Co of Zurich [2003] UKPC 11, [2003] 1 WLR 1041, at [15] (Lord Hobhouse). 84 Carl Zeiss Stiftung v Rayner & Keeler Ltd (No 2) [1967] 1 AC 853, 919, per Lord Reid: ‘It is quite true that estoppel is a matter for the lex fori but the lex fori ought to be developed in a manner consistent with good sense.’ The principle of estoppel per rem judicatam applies where there has been a previous court decision (a res judicata) on a matter. 85 See Dicey, Morris & Collins: The Conflict of Laws (above n 5), vol 1, para 7-038, describing the categorisation of estoppel as an ‘entirely open question’. In First Laser Ltd v Fujian Enterprises (Holdings) Company Ltd [2012] HKCU 1397, the Hong Kong Court of Final Appeal held that estoppel by convention is a question of substance, not procedure. 86 Ali Shipping Corp v Shipyard Trogir [1999] 1 WLR 314. 87 Associated Electric and Gas Insurance Services Ltd v European Reinsurance Co of Zurich [2003] UKPC 11, [2003] 1 WLR 1041. For a discussion of US law on these issues, see Masters and Stanzler, Insurance Coverage Litigation, ch 12.

42

Choice of Law Issues under the Bermuda Form

foreign limitation period might be substantive, depending on whether it was interpreted as barring ‘the right’ or ‘the remedy’. This led to considerable complexity and to problems of so-called ‘cumulation’ and ‘gap’. Since the Foreign Limitation Periods Act 1984, limitation has (subject to certain qualifications) generally been treated as a substantive question, subject to the lex causae, an approach which is also adopted by EU private international law.88 This approach is, under the English Arbitration Act 1996, applied in arbitration.89 Thus, in a case under the Bermuda Form, the relevant limitation period is the six-year period laid down by New York law.90 3.37

Remedies. Remedies are tricky; they cannot be neatly categorised as matters of substance or of procedure. A distinction is customarily drawn between claims for damages and other remedies.

3.38

The question of what sorts of damage are compensable was at common law regarded as a matter for the lex causae.91 Questions about the precise proof and quantification of damages were then again a matter for the lex fori; one was back in the realm of procedure and evidence. In contractual cases before English courts, there is now a legislative rule that damages, insofar as they are subject to legal rules, are a matter for the applicable law,92 but always within the ‘limits’ of the court’s procedural powers. Moreover, the applicable law governs only the ‘consequences of breach, including assessment of damages in so far as it is governed by rules of law’.93 This seems to leave open the application of the lex fori to issues about how detailed facts should be proved. It is hard to see any material difference between this and the common law position. In an arbitration, the tribunal will be entitled to use its procedural discretion to decide how questions of ‘detailed proof’ or ‘assessment’ should be approached.

3.39

Other remedies are more controversial. Obviously, the English Arbitration Act 1996 governs to the extent that it places statutory limits on the type of remedy the tribunal may order. But that is not a practical difficulty, since the range of available options is wide. Where there may be room for debate is on the issue whether, in a particular case, a given remedy should be deployed – and if so on what terms. Suppose that New York law permits an insurer to rescind a contract for misrepresentation without returning the premium, as is sometimes (though in our view wrongly)94 suggested. Should an arbitration tribunal pursue that course or should it permit rescission only on terms which require return of the premium, as English law does? Or suppose that the insurer seeks a ‘negative declaration’ that it will not be liable if certain events take place – for instance, if the attachment point is exceeded in the future. Should the tribunal apply the English

88 Rome I Regulation (above n 9), Article 12(1)(d). The Rome Convention was to similar effect: Contracts (Applicable Law) Act 1990, Sched 1, Article 10(1)(d). 89 English Arbitration Act 1996, s 13(1), (4). 90 For the application of the limitation period to a claim for payment by the policyholder, see below paras 11.40–11.52. In relation to the limitation period for rescission for misrepresentation, see below paras 12.40–12.43. 91 Boys v Chaplin [1971] AC 356, 379, 393, 394–95. 92 Rome I Regulation (above n 9) art 12(1)(c); prior to 17 December 2009: Contracts (Applicable Law) Act 1990, Sched 1, Article 10(1)(c). 93 Ibid (emphasis added). 94 See below ch 12, paras 12.29–12.37.

The Division between Substance and Procedure

43

approach to deciding when such negative declarations are appropriate or should it be guided by the circumstances in which such relief would be granted in New York? The traditional English view is that choice between remedies is dictated by the principles customarily applied by the forum. On this theory, an English court or tribunal could order specific performance of a contract in circumstances where the lex causae would not, or could restrict the claimant to damages in circumstances where the lex causae would grant specific performance, and so on.95 But, even in court, there are limits: the remedy granted must not be such as to alter fundamentally the nature and scope of the right as it is conceived under the lex causae.96 Moreover, in cases before the English courts, the Rome I Regulation seems to go rather further, since it mandates that (within the limits of the powers conferred upon the court deciding the case) ‘the consequences of breach’ are governed by the applicable law.97 The modern position, then, seems to be that even courts, in contract cases at any rate, should so far as possible replicate the remedies that would be granted by a court in the state whose law is to be applied.

3.40

Although that legislation does not apply directly in an arbitration, there seems to be good reason why an international arbitration tribunal should generally prefer to produce an award which is as close as possible to that which the applicable law chosen by the parties would arrive at. Arbitral seats are selected largely for their neutrality; their awards may well not even be enforced or performed at the seat. Even if the normal remedy which would be a departure from that granted under the chosen law does not totally pervert the right, it is not unlikely to change it quite significantly. For example, if New York law permitted an insurer to rescind a policy yet retain the premium, an order requiring the return of the premium as the price of rescission would unjustifiably alter the parties’ rights.

3.41

Negative declarations are trickier. The restrictions on the grant of such declarations do not derive from any particular conception of rights under the contract or even the juridical ‘consequence of breach’, but from variable conceptions of the proper role of a judicial body, often with constitutional overtones. The qualms a court in New York may feel about granting such declarations do not obviously carry over to a private tribunal sitting in England. Nor, indeed, do the qualms that an English court may feel.98 As public tribunals, courts have to juggle various policy considerations which may militate in favour of a fairly restrictive approach to ‘hypothetical’ or ‘advisory’ remedies. They must consider their constitutional role compared to that of the legislature, the proper use of publicly funded court time, and the suitability of a rather inflexible and formal procedure for deciding such questions. Arbitration tribunals do not work within the same constraints: they are privately funded, constitutionally uninfluential and capable of moulding their procedure more flexibly than a court can. There are aspects of granting relief in relation to disputes which may be hypothetical that they should worry about, such as the cost and expense for an unwilling party of a procedure which may serve no practical purpose.

3.42

95 Baschet v London Illustrated Standard Co [1900] 1 Ch 73. 96 Phrantzes v Argenti [1960] 2 QB 19, 35. 97 Rome I Regulation (above n 9) Article 12(1)(c). A court could not use the lex causae to acquire a remedial power it lacks, but only to choose between the exercise of its various remedial powers. 98 Or, perhaps, used to feel – English courts now regard negative declarations with more fortitude than previously: see, eg, Messier Dowty Ltd v Sabena SA [2000] 1 Lloyd’s Rep 428.

44

Choice of Law Issues under the Bermuda Form

But they are not in the same position as any court, and it might well be said that they should not necessarily follow the approach taken even by the courts of the forum in deciding whether it is appropriate that a declaration should be granted. 3.43

Currency. The Bermuda Form provides expressly for the currency in which premiums and losses are payable: US dollars, unless the parties have otherwise agreed.99 The contractual entitlement of a creditor to be paid in a particular currency is a substantive right,100 and for many years this right has been recognised by way of awards made in the currency of the contract. Section 48(4) of the Arbitration Act 1996 gives a tribunal the power to order ‘the payment of a sum of money, in any currency’. In Lesotho Highlands Development Authority v Impregilo SPA, the majority of the House of Lords held that section 48(4) of the Arbitration Act did not give a tribunal a broad discretion to award payment in any currency it thought fit, regardless of the terms of the contract.101 In the unlikely scenario of an arbitral tribunal failing to give effect to these provisions, it would not be possible to appeal against the tribunal’s decision. This is because it would amount to an error of law rather than a case where the tribunal had exceeded its powers.102

3.44

Interest. The award of interest is usually regarded in England as a procedural matter governed by English law.103 English law permits the award of simple or compound interest under section 49 of the Arbitration Act 1996. Where a claim is successful, interest will usually be awarded. An issue which has come into focus in recent years is whether Bermuda Form tribunals should award interest (i) at the 9 per cent (simple interest) rate which is mandated by s 5004 of the New York Civil Practice Law and Rules, or (ii) at a rate (whether simple or compound) that companies of a similar status as the successful claimant would have had to pay to borrow the sum awarded over the period it has been wrongly withheld. Substantial sums of money can turn on this issue: at a time of low interest rates, the 9 per cent rate has for some time exceeded the normal borrowing rate. This issue is addressed in detail in Chapter 17 below.

3.45

Costs and attorneys’ fees. Whether costs (both the fees and costs of the tribunal and those paid to a party’s own lawyers) should be awarded is a procedural matter governed by English law. Early versions of the Bermuda Form contained a provision whereby each party bore its own costs of representation, and the parties shared the costs of

99 Condition M in the XL 004 version of the Form. In the original Bermuda Form (Form 001), the equivalent condition simply stated that the premiums and losses were payable in US currency. 100 Miliangos v George Frank (Textiles) Ltd [1976] AC 443. 101 Lesotho Highlands Development Authority v Impregilo SPA [2005] UKHL 43, [2006] 1 AC 221. 102 Ibid, reversing the decision of the English Court of Appeal in Lesotho Highlands Development Authority v Impregilo SpA and Others [2003] EWCA Civ 1159, [2003] 2 Lloyd’s Rep 497. An appeal for error of law is not possible under the 004 Form: see further paras 3.16 above and 11.84 below. 103 Interest is discussed in detail in ch 17 below, paras 17.1–17.23. There is some conflict in the English authorities, but the preponderance of authority favours the view that awards of interest are governed by the lex fori: see the summary of the authorities in Lesotho Highlands [2003] EWCA Civ 1159 [2003] 2 Lloyd’s Rep 497, reversed but not on this point [2005] UKHL 43 [2006] 1 AC 221. It was unnecessary for the House of Lords to reach any view on whether the power to award interest is procedural or substantive, since in either case (as the majority held) the tribunal had not exceeded its jurisdiction. In Enka (above n 15), the Supreme Court held that the rate of interest is a matter of procedure and is therefore governed by English law: see below ch 17, paras 17.13–17.15.

The Division between Substance and Procedure

45

the arbitration. Unless explicitly reconfirmed by mutual agreement after the dispute has arisen,104 this provision is invalidated by the English Arbitration Act 1996.105 It is, of course, usual in England for the unsuccessful party to pay the successful party’s reasonable costs and those of the tribunal. Recent versions of the Bermuda Form generally make no attempt to alter this usual practice. The question of costs is addressed in greater detail in Chapter 17 below. Privilege. Questions of privilege often arise in arbitrations under the Bermuda Form. The issue is usually whether the policyholder ought to be required to disclose documents generated by or for lawyers in the underlying proceedings. In principle, the question of whether such documents should be disclosed is a procedural question governed by English law. If a document is privileged under English law principles, the tribunal may not require its disclosure.106 To that extent, privilege is a matter for English law. However, just because a document is not privileged to English eyes does not mean that the tribunal must require it to be disclosed. The tribunal has a discretion. It may be appropriate to consider foreign rules of privilege in deciding how that discretion should be exercised. A tribunal might, therefore, decide not to require disclosure of documents that are not privileged in England, but are privileged under some other rules. For example, where litigation is ongoing and disclosure would cause prejudice to the policyholder in the underlying litigation, the prejudice faced by the policyholder in disclosing the document might outweigh any prejudice to the insurer resulting from the document being withheld.107

104 For instance, by including a provision in an agreed procedural order. 105 Section 60. 106 The tribunal’s broad discretion over procedure does not extend to ordering disclosure of privileged documents. 107 A detailed discussion of privilege, including further discussion of the applicable law, is contained in ch 16 below.

3.46

4 Interpretation of the Bermuda Form and the Modification of New York Law 4.01 One of the most distinctive features of the Bermuda Form is that, besides selecting a system of law to govern interpretation, it attempts to modify the principles that would normally be applied by that system of law (Article VI(O) of Form 004, subdivided into sections marked [A], [B], and [C] for the purposes of discussion in this chapter): This Policy, and any dispute, controversy or claim arising out of or relating to this Policy, shall be governed by and construed in accordance with the internal laws of the State of New York, [A] except insofar as such laws: (1) may prohibit payment in respect of punitive damages hereunder; (2) pertain to regulation under the New York Insurance Law, or regulations issued by the Insurance Department of the State of New York pursuant thereto, applying to insurers doing insurance business, or issuance, delivery or procurement of policies of insurance, within the State of New York or as respects risks or insureds situated in the State of New York; or (3) are inconsistent with any provision of this Policy; [B] provided, however, that the provisions, stipulations, exclusions and conditions of this Policy are to be construed in an evenhanded fashion as between the Insured and the Insurer; without limitation, where the language of this Policy is deemed ambiguous or otherwise unclear, the issue shall be resolved in the manner most consistent with the relevant provisions, stipulations, exclusions and conditions (without regard to authorship of the language, and without any presumption or arbitrary interpretation or construction in favor of either the Insured or the Insurer or reference to the ‘reasonable expectations’ of either thereof or to contra proferentem and without reference to parol or other extrinsic evidence). [C] To the extent that New York law is inapplicable by virtue of any exception or proviso enumerated above or otherwise … the internal laws of England and Wales shall apply.1

4.02

The validity of this approach to choice of law is discussed elsewhere.2 It remains to consider its effect in practical terms. The departures from New York law as it would ordinarily be applied by a New York court take various forms. In some areas – for instance, the recovery of punitive damages3 – the Bermuda Form provides that New York law is not

1 Article

VI(O) of the 004 Form. See also similar provisions in Article V(q) of earlier Forms. ch 3, paras 3.18–3.27, especially paras 3.25–3.27. 3 New York law precludes coverage for punitive damages: see ch 3, para 3.23. 2 See

The Structure of the Modification of the New York Law Proviso

47

to apply at all. This is the effect of the exceptions in section [A] of the provision. Instead, in these respects, English law is to apply, as section [C] of the provision makes clear. The exceptions in [A] and [C] pose some interpretative problems, but in practice do not frequently give rise to significant issues.4 There is, however, another technique used to modify New York law that is more troublesome in practice: the proviso in the section we have marked [B]. These are provisions that do not so much exclude New York law in favour of some other system of law as provide canons of construction by which, when applying New York law, difficult questions of interpretation are to be addressed. The canons of construction in the policy differ in some respects from those that New York law, left to its own devices, would apply; to that extent, they modify New York law as it applies to the Form. But the proviso does not operate by disapplying New York law and replacing it with principles drawn from some other legal system (whatever they might be), but rather by operating directly and systematically on New York law itself. The modification includes both a positive statement of principles that should be applied and a negative list of principles that should not be applied.

4.03

Although this is unusual, it is not legally troublesome so far as a contract that is subject to arbitration under English law is concerned. Section 46 of the English Arbitration Act 1996 requires an arbitral tribunal to decide the dispute ‘(a) in accordance with the law chosen by the parties as applicable to the substance of the dispute, or (b) if the parties so agree, in accordance with such other considerations as are agreed by them or determined by the tribunal’. Clearly, then, the parties are not obliged to choose a single system of law; they can construct a framework for decision in any way they please. In the early years of the Bermuda Form, it was in vogue to argue that the modifications in the clause should be ignored. Such arguments met with little success and have now, rightly, fallen out of use.

4.04

THE STRUCTURE OF THE MODIFICATION OF THE NEW YORK LAW PROVISO Many of the problems of interpreting the proviso are reduced if one pays careful attention to its structure. Its starting point is not negative, but positive: the Form is to be ‘construed in an evenhanded fashion as between the Insured and the Insurer’. What follows by way of a series of negative injunctions not to make use of particular types of reasoning (extrinsic evidence and so forth) is not free-standing, but put forward as a (non-exhaustive)5 gloss on, or enlargement of, this basic and fundamental principle. In construing the ambit of what might be called the ‘forbidden canons’ of New York law, it is important to keep this point in mind, for the prohibitions are not random, but follow from this positive injunction, which may greatly assist in interpreting their proper ambit. Moreover, even the itemisation of the ‘forbidden canons’ of construction is not primarily negative. For apart from the instructions not to use various techniques for resolving vagueness and ambiguity, the interpreter is also given a positive instruction: to arrive at

4 For

their effect, see ch 3, paras 3.22–3.24, and ch 8, para 8.17. ‘without limitation’.

5 Hence

4.05

48

Interpretation of the Bermuda Form and the Modification of New York Law

the resolution ‘most consistent with the relevant provisions, stipulations, exclusions and conditions’, and once again this positive principle – to apply the policy in an evenhanded fashion – helps one to understand the negative points that follow. 4.06

The starting point, therefore, is not negative, but doubly positive. The Form is to be interpreted in an ‘evenhanded fashion’, and difficult points are to be resolved by giving the best interpretation possible to the provisions of the policy themselves. It might be thought that this latter injunction is merely circular: what is interpretation other than the process of deciding which among any number of possible meanings is ‘most consistent with’ the words that have been used? The ‘forbidden canons’ give the clue. In various respects the interpretation of insurance policies under New York law is affected by what the drafters of the Bermuda Form evidently believed were extraneous and arbitrary considerations – a public policy to maximise coverage, or to protect the interests of the policyholder specially, or to resolve doubts against the insurer, and so forth. These principles are to be abandoned as, in the drafters’ view, they stand in the way of ‘evenhanded’ interpretation of the contract language itself.

4.07

A coherent case could no doubt be made that there is no inherent antipathy between canons such as these and the principle that interpretation is the search for the ‘most consistent’ and ‘evenhanded’ meaning of the contract language.6 It could be argued, for example, that the principle that words are construed contra proferentem exists precisely to redress the imbalance that otherwise exists between those who draft contracts that are essentially contracts of adhesion and those who accept them. But that may depend on how the principles are applied. It is one thing, for instance, to apply contra proferentem where there would otherwise be no reason to prefer one possible interpretation over another, but quite another to apply contra proferentem in such a way as to determine a choice between two possible meanings without any regard for which would otherwise be the more probable. The drafters of the Bermuda Form took a jaundiced view that these canons, even if they were originally grounded in a sound approach to particular problems of interpretation consistent with the notion that it is no more and no less than a search for meaning, have come to be applied in a way that is ‘arbitrary’. The drafters probably felt that these rules systematically favoured the interests of the policyholder in ways that make sense in a consumer context where the policyholder’s ignorance and weaker bargaining power call for protection, but that are inappropriate to the sort of commercial relationship with which the Bermuda Form is likely to be concerned.7 6 New York courts would presumably take this view, being equally committed to an approach that does not extend coverage beyond the fair intent and meaning of the agreement: Casey v General Accident Ins Co, 578 NYS2d 337, 338 (App Div 1991) (citing Moshiko Inc v Seiger & Smith, 529 NYS2d 284, 288–89 (App Div), aff ’d, 533 NYS2d 52 (NY 1988)). Much depends on how ‘fair intent’ is understood, on the relative emphasis placed on ‘fair’ and ‘intent’, and on the particular conception of fairness that is in play. 7 Such considerations motivate proponents of the ‘sophisticated policyholder’ defence used in coverage litigation in US jurisdictions – an argument that has met with mixed success. This argument is obviously difficult to apply in cases involving a standard form insurance policy, as it suggests that the form means one thing when sold to a small business and another thing when sold to a large, multinational corporation. See, eg, Boeing Co v Aetna Cas & Sur Co, 784 P2d 507, 514 (Wash 1990). The Restatement of the Law, Liability Insurance rejected the ‘sophisticated policyholder exception’: American Law Institute, Restatement of the Law, Liability Insurance (St Paul, MN, American Law Institute Publishers, 2019) s 4, cmt h (hereafter Restatement Liability Insurance). Because the Bermuda Form is designed specifically for relatively ‘sophisticated’ policyholders, these objections arguably do not apply. But even so, one should be careful: the fact that a policyholder is a large and

Interpretation of Insurance Contracts: The Basic Approach

49

Thus, for instance, although rules that redress an imbalance between insurer and insured may be valid tools when dealing with the relationship between ordinary consumers of insurance and large insurance companies, it is by no means obvious that the sort of commercial enterprises that purchase high-level excess coverage require or should expect such protection. It is in this context – the context for which the Bermuda Form was written – that the ‘forbidden canons’ are perceived to operate arbitrarily. In this context, they may stand in the way of interpretation properly so called, and substitute for it what amounts to a rewriting, rather than a reading, of the policy language. At root, then, the instructions as to interpretation should be approached positively: apply New York law, but in such a way as to produce an interpretation that respects the interests of both parties and is the best reading of the policy, not a rewriting of it for either party.

4.08

INTERPRETATION OF INSURANCE CONTRACTS: THE BASIC APPROACH Commentators sometimes present the process of interpretation as if it involved two separate stages: a first stage, in which an attempt is made to ascertain the ‘plain meaning’ of the contract; and a second stage – reached only if the search for a ‘plain’ meaning has failed – in which the interpreter ‘resorts’ to wider considerations or ‘rules of construction’ in order to ascertain meaning. Thus, for example, Ostrager and Newman suggest that ‘as a general rule the language of an insurance policy will be given its plain meaning and there will be no resort to rules of construction unless an ambiguity exists’.8 In our view this approach is apt to be misleading. The object of construction is always to ascertain the meaning of the contract, that is, the intention of the parties at the time of contracting as objectively expressed in the words they have used. In this regard, an insurance contract does not differ from any other contract.9 The objective, therefore, is not first to interpret the words of the contract and then – if that fails – to make use of ‘rules of construction’. Rather, the rules are to be regarded (as the Restatement (Second) of Contracts puts it) as ‘aids’ in interpretation.10

sophisticated company in its own sphere of business does not mean it is necessarily sophisticated in insurance matters. See, eg, United States v Brennan, 938 F Supp 1111, 1121 (EDNY 1996), rev’d, 183 F3d 139 (2d Cir 1999). So far as relative sophistication goes, it is only to be expected that the insurance company – whose whole business is insurance – will be considerably more sophisticated than the policyholder is. 8 Barry R Ostrager and Thomas R Newman, Handbook on Insurance Coverage Disputes, 19th edn (New York, Wolters Kluwer, 2019) s 1.01[a] (emphasis added). For a contrary view, see Jeffrey W Stempel, Interpretation of Insurance Contracts (Boston, Little, Brown & Co, 1994) ch 11. See too Restatement Liability Insurance, ss 3–4 and comments thereto (stating rules of insurance policy interpretation). 9 People ex rel New York C & H R RR v Walsh, 105 NE 136 (NY 1914); Throgs Neck Bagels Inc v GA Ins Co, 671 NYS2d 66, 69 (App Div 1998) (citing Breed v Insurance Co of N Am, 413 NYS2d 351, 355 (App Div 1978)). See also Uniroyal, Inc v Home Ins Co, 707 F Supp 1368 (EDNY 1988); IBM Poughkeepsie Employees Federal Credit Union v Cumis Ins Soc Inc, 590 F Supp 769, 772 (SDNY 1984). 10 American Law Institute, Restatement (Second) of Contracts (St Paul, MN, American Law Institute Publishers, 1981) s 202 (hereafter Restatement (Second) of Contracts).

4.09

50

Interpretation of the Bermuda Form and the Modification of New York Law

4.10

The two-stage approach tends to come to the fore precisely where the interpreter is seeking to apply not a single set of rules to ascertain contractual meaning, but two sets of rules: one which aims to ascertain a range of permissible meanings, as a prelude for the possible application of a second set of rules whose purpose is to give priority to those available meanings that will best effectuate some extraneous purpose. That is precisely the position in which New York courts often find themselves in insurance cases, because of the potency of the principle that boilerplate contracts are to be construed against the drafter and allied principles. But the two-stage approach is not the position embodied in the Bermuda Form, and as a result the whole notion of interpretation in ‘stages’ – one that filters out all those meanings which are ‘reasonably possible’ and one that then selects among those reasonably possible meanings – is generally unhelpful. In construing the Form, one does not first of all seek its ‘plain and natural’ meaning and then, if that search fails, do something else; one seeks always to provide an ‘evenhanded’ interpretation that is the best or most consistent with all the relevant provisions.

4.11

By the same token, it is wrong to equate the ‘plain’ or ‘natural’ meaning ‘of words with the true or best meaning of a contractual provision.11 Of course, the meaning and use of words in ordinary language are important factors.12 Careful consideration, however, may sometimes convince one that the parties have used words in an obscure, unnatural or even inaccurate way. That may be so even though the words do have a ‘plain’ meaning. To take a famous example from another context, the words ‘east of the Panama Canal’ do have a plain meaning, and if one consults an atlas with that plain meaning in mind, it is ‘plain’ that the Gulf of Mexico is not east of the Panama Canal. It may nevertheless be apparent, if one understands the words used in context, that the parties intended the Gulf of Mexico to be treated as ‘east of the Panama Canal’.13 That conclusion, however, cannot be reached without considering the words in context, by reference not only to the lexicographical meaning of the words, but also to their purpose in the document in which they were used. If resort to the ‘plain’ meaning of words is intended to preclude such an approach to construction, then ‘plain meaning’ is inconsistent with the goal of interpreting a contract so as to give effect to the parties’ mutual intentions. If, on the other hand, resort to ‘plain meaning’ is no more than a synonym for that approach, it is hardly useful.

4.12

Interpretation can never be properly understood as if it were mere decoding. Even after a sentence has been parsed grammatically and syntactically, and the meaning or possible meaning of every word is understood, there is often a choice to be made. Even a simple sentence, Ben likes fish, may mean something different depending on whether it is uttered in the context of a discussion of the relative merits of pets (Brenda likes dogs; Ben likes

11 For example: ‘The meaning which a document … would convey to a reasonable man is not the same thing as the meaning of its words. The meaning of words is a matter of dictionaries and grammars; the meaning of the document is what the parties using those words against the relevant background would reasonably have been understood to mean.’ Investors Compensation Scheme v West Bromwich Building Society [1998] 1 WLR 896, 913 (Lord Hoffmann). 12 Throgs Neck Bagels, 671 NYS2d at 69 (citing Gittelson v Mut Life Ins Co of NY, 41 NYS2d 478, 481–82 (App Div 1943); and Lewis v Ocean Accident & Gvar Corp, 224 NY 18, 21 (1918) (Cardozo, J)). See also Belt Painting Corp v TIG Ins Co, 742 NYS2d 332, 334 (NY 2003). 13 Segovia Compagnia Naviera SA v R Pagnan & Fratelli [1977] 1 Lloyd’s Rep 343.

Interpretation of Insurance Contracts: The Basic Approach

51

fish) or of food (Brenda likes beef; Ben likes fish).14 A great deal depends on context. But what counts as context is itself variable. Sometimes a particular phrase may, as it were, create its own context. We naturally contextualise the rule dogs must be carried on the escalator differently from the rule passes must be carried in the building: the words themselves suggest different purposes and contexts. Sometimes context depends on the setting of the particular utterance among others (for example, whether Ben’s liking for fish occurs in an alimentary or a veterinary context). Sometimes we may need to bring in context from outside the utterance altogether: Ben likes fish, and we know Ben is a cat or a vegetarian. Sometimes considerations of knowledge or context may lead us to conclude that a particular phrase has been used idiosyncratically, artificially or even sloppily. But always the meaning of the utterance as a whole will be more than merely the meaning of the words. Under New York law, an insurance policy ‘is a contract which, like any other contract, must be construed to effectuate the parties’ intent as expressed by their words and purposes’.15 The Bermuda Form’s starting point, which is its language, is also the starting point under New York law. The first thing a New York court looks to is the language of the policy.16 It is the intention of the parties ‘as derived from the language employed’ that is to be sought.17 In doing this, one looks not just to individual words, but to the policy as a whole,18 and to the contested words in that context,19 and avoids reading the policy in a way that makes some of its provisions meaningless20 or redundant.21 The objective is to understand what coverage was agreed, not to vary it or extend it beyond the ‘obvious intent and meaning of the agreement’.22 Coverage should not be rendered illusory or ‘nearly illusory’,23 and indemnity should not be ‘reduced to a shadow’.24 So it should not be thought that the notion of an ‘evenhanded interpretation’ which pays fidelity to the text of the policy as a whole is in any fundamental sense foreign to New York law. Although those who drafted the Bermuda Form took issue with some of the peripheral

14 This sort of surprising ambiguity is neatly summarised by the aphorism of computer scientists, derived from some research at Harvard in the 1960s: time flies like an arrow; fruit flies like a banana. See Steven Pinker, The Language Instinct (London, Penguin, 1995) 209. 15 American Home Prods Corp v Liberty Mut Ins Co 565 F Supp 1485, 1493 (SDNY 1983), modified 748 F2d 760 (2d Cir 1984). A brief but useful statement of the relevant principles is to be found in In re Viking Pump 27 NY3d 244, 257–58 (NY 2016). 16 Selective Ins Co of Am v County of Rensselaer 26 NY3d 649, 655 (NY 2016) (citing Consolidated Edison Co of NY v Allstate Ins Co 98 NY2d 208 (NY 2002) and Breed v Insurance Co of N Am 46 NY2d 351 (NY 1978)). 17 Hartford Accident & Indem Co v Wesolowski, 33 NY2d 169, 171 (NY 1973). 18 SP Duggal Corp v Aetna Cas & Sur Co 181 AD2d 472, 474 (App Div 1992); A Meyer & Sons Corp v Zurich Ins Grp 74 NY2d 298 (NY1989). 19 Piliero v Allstate Ins Co 209 NYS2d 90, 91 (App Div 1960). 20 County of Columbia v Continental Ins Co 83 NY2d 618, 628 (NY 1994). 21 In re Viking Pump, 27 NY3d 244, 257–58 (NY 2016); Roman Catholic Diocese of Brooklyn v National Union Fire Ins Co of Pittsburgh Pa 21 NY3d 139, 148 (NY 2013); Westview Assocs v Guaranty National Ins Co 95 NY2d 334, 339 (NY 2000); Theatre Guild Prods Inc v Ins Corp of Ireland, 267 NYS2d 297, 300-01 (App Div 1966), aff ’d, 19 NY2d 656 (NY 1967). English courts have, especially in recent years, generally been less impressed by arguments from surplusage which is endemic in insurance drafting. 22 McCarthy v AMEX Assurance Co 636 NYS2d 475, 477 (App Div 1996). 23 Madawick Contracting Co v Travelers Ins Co 307 NY111 (NY 1954); Thomas J Lipton v Liberty Mut Ins Co 34 NY2d 356, 358 (NY 1974). 24 Messersmith v American Fidelity Co 232 NY 161 (NY1921) (Justice Cardozo).

4.13

52

Interpretation of the Bermuda Form and the Modification of New York Law

doctrines by which New York courts may attempt to resolve ambiguities, the courts have been consistently clear that these are all supplementary principles that operate in the shadow only where the language fails to provide an answer; the mere fact of a dispute, of genuine differences of subjective interpretation, or of the theoretical possibility of greater clarity does not make a contract ambiguous. Nothing in the Bermuda Form invalidates any of those basic principles.

INTERPRETATION OF THE MODIFICATION OF THE NEW YORK LAW PROVISO 4.14

In our view, then, the basic approach required by New York law when interpreting insurance policies – which is to construe them, as any contract, so as to give effect to the parties’ mutual intentions as objectively expressed – is entirely compatible with the approach set out in Article VI(O) and its predecessors. Where the two systems part company is in the use of a number of devices (construction contra proferentem, or against the insurer, and so forth) that arguably depart from this basic principle and substitute for it an approach to construction that gives priority to other interests, such as the protection of the ‘consumer’ of insurance. New York law often involves a two-stage process, in which the first stage (‘finding’ meaning in the document) is relevant largely because of its ability to control the extent of judicial freedom at the second stage (‘giving’ meaning to the document in order to further policy objectives decided upon by the interpreter). The Bermuda Form aims to dispense with that second stage. But it does not at all follow that what is left – the search for meaning – can be carried out mechanically; that it is a matter merely of grammar and dictionary work; or that the interpretation ‘most consistent with’ the relevant provisions of the policy will be that which occurs to one first, or which involves the most ‘common’ use of words, or which rests on the most ‘plain’ or ‘natural’ view of them.

4.15

In summary, the effect of Article VI(O) is as follows: (i) it is consistent with the proposition that interpretation aims at ascertaining the mutual objectively expressed intention of the parties at the time of the contract, as expressed in the words they have used. This is the basic doctrine of New York law and remains so under the Bermuda Form; (ii) it permits the use of all the basic tools of interpretation, properly so called – including those that require sensitivity to be paid to context and purpose – which are designed to achieve that aim; (iii) it does not envisage a mechanical approach to interpretation, or commit the parties to the ‘plain’ or ‘natural’ meaning of the words, divorced from their context; (iv) it is incompatible with a ‘two-stage’ approach, whereby the interpreter searches first for the range of possible meanings and then selects among them on the basis of criteria other than the parties’ mutual intentions; (v) it precludes approaches that are designed to impose rather than uncover meaning;25 (vi) it does not, however, seek to produce an entirely ‘value-free’ interpretation, as it specifies within the text and as part of the intention of the parties the goal that interpretation should be ‘evenhanded’.

25 It

demands that interpretation be exegesis rather than eisegesis.

Interpretation of the Modification of the New York Law Proviso

53

A number of points flow from this. First, nothing in Article VI(O) commits one to interpret each provision of the Form on its own, without regard to the other provisions. On the contrary, under New York law, an insurance policy must be interpreted as a whole.26 Nothing in Article V(O) contradicts this. The ‘relevant’ provisions of the contract are not limited to those whose meaning is immediately in dispute.

4.16

Second, as the Restatement (Second) of Contracts puts it:

4.17

Words and other conduct are to be interpreted in the light of all the circumstances, and if the principal purpose of the parties is ascertainable it is given great weight.27

An insurance policy is to be construed on the basis of the meaning of the words employed ‘read in the context of the policy as a whole, the purposes sought to be accomplished, and the relevant surrounding circumstances’.28 In the context of an insurance policy, that includes paying attention to the ‘risk, subject matter and purpose of the policy’ when interpreting the words used.29 It is sometimes suggested that the Bermuda Form’s prohibition on the use of ‘parol or extrinsic evidence’ precludes reference to the ‘circumstances’ surrounding the contract. This, we suggest, is not correct. New York law has long regarded the underlying circumstances as relevant to the interpretation of a contract, prior to and quite distinct from any question of whether extrinsic or parol evidence may be considered. When interpreting contracts, New York courts seek to avoid any interpretation that would take the language and ‘rob it of its efficacy as an implement to be used in the furtherance of a business purpose’.30 In applying that principle, regard is paid to ‘surrounding circumstances’ which may ‘stamp upon a contract a popular or looser meaning’31 of words than that which – viewed in isolation – they might bear. It is indeed ‘questionable whether a word has a meaning when divorced from the circumstances in which it is used’.32

4.18

Reference to such background circumstances, the context in which both parties would have understood they were acting, and the objectives both would have had in view at the time the contract was made, is not regarded as involving any reference to extrinsic

4.19

26 Restatement (Second) of Contracts, s 202(2). For application of the principle in an insurance context under New York law, see the cases cited at n 18 above. 27 Restatement (Second) of Contracts, s 202(1). 28 American Home Products 565 F Supp at 1493. See also Continental Ins Co v Arkwright Mut Ins Co 102 F3d 30, 34 (1st Cir 1996) (citing Harris v Allstate Ins Co 309 NY 72, 75–76 (1955): ‘The words of the policy are to be read in context, the language construed fairly and reasonably with an eye to the object and purpose to be achieved by the writing’). In long-standing precedent, the US Supreme Court stated the ‘fundamental rule that in the construction of contracts the courts may look not only to the language employed, but to the subject matter and the surrounding circumstances and may avail themselves of the same light which the parties possessed when the contract was made’: Merriam v United States 107 US 437, 441 (1883). 29 Show Car Speed Shop, Inc v United States Fidelity & Guar Co 596 NYS2d 608, 609 (App Div 1993) (citing DeForte v Allstate Ins Co 442 NYS2d 307, 309 (App Div 1981)). According to Restatement Liability Insurance (s 2(2)): ‘Except as this Restatement or applicable law otherwise provides, the ordinary rules of contract interpretation apply to the interpretation of liability insurance policies.’ 30 Outlet Embroidery Co v Derwent Mills, 172 NE 462, 463 (NY 1930) (Cardozo CJ). 31 Utica City Nat’l Bank v Gunn 118 NE 607, 607–08 (NY 1918) (Cardozo J). 32 E Allan Farnsworth, Contracts, 2nd edn (Boston, Little, Brown & Co, 1990) 256; Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) at s 2.06.

54

Interpretation of the Bermuda Form and the Modification of New York Law

evidence. Indeed, the New York Court of Appeals has itself referred to such circumstances in construing contracts precisely in order to determine that extrinsic evidence will not be admissible. Thus, in West, Weir & Bartel, Inc v Mary Carter Paint Co,33 the court considered what both parties ‘anticipated’ would be the commercial reaction of third parties to a plan and to the history of the parties’ relationship. In Bethlehem Steel Co v Turner Construction Co,34 the court considered the purpose and effect of a particular provision concerning price increases against the background of findings about Bethlehem Steel’s historic practices with regard to price increases. In neither case did the court show any sign that it considered the use of material of that sort as a resort to ‘extrinsic’ or ‘parol’ evidence. This view is surely right. Objective information about the factual circumstances, known to and in the contemplation of both parties at the time the contract was made, is not extrinsic evidence of what particular words mean. It is not, indeed, evidence of meaning at all, or subjective intention, but material that may assist in clarifying the meaning of the words themselves. Some conception of the background against which the words were framed, and the purpose for which they were framed, will almost always be necessary to make sense of them. Put differently, the objection to extrinsic evidence is that it substitutes for the question ‘What did the parties mean by what they said?’ the following question: ‘What did the parties mean to say?’ Evidence of the objective circumstances against which the parties acted, however, does not have this vice; it simply flows from the need to understand the policy as a purposeful communication. 4.20

The objection just disposed of is sometimes summed up in the catchphrase that interpretation must remain ‘within the four corners of the contract’, with the insinuation that, as soon as one considers the background, aim and genesis of the transaction, one is moving outside that sacred quadrangle. But the maxim – though found in some of the cases – is as unhelpful as such maxims normally are. No contract is self-interpreting. Language is a social medium, and one inevitably passes outside the ‘four corners’ of the particular document if one is to make sense of it. Even the most enthusiastic proponent of quadrangular interpretation is likely to use dictionaries, previously decided cases, and common-sense notions of purpose and reason. One could, indeed, hardly manage without such material; meaning is never hermetically sealed in a single document. Given that, the question is not whether one moves outside at all, but how far one can legitimately move outside. The long tradition of the common law, including New York law, is that one may take account of matters such as the background circumstances and commercial purposes of the parties without engaging in anything that could be stigmatised as involving recourse to the ‘extrinsic evidence’.

THE ‘FORBIDDEN GROUNDS’ 4.21

The approach set out above suggests what is common to the various ‘forbidden canons’ which the Bermuda Form outlaws. Each of them is an example of a principle of so-called

33 West,

Weir & Bartel, Inc v Mary Carter Paint Co 307 NYS2d 449, 452–53 (NY 1969). Steel Co v Turner Construction Co 161 NYS2d 90, 93–94 (NY 1957).

34 Bethlehem

The ‘Forbidden Grounds’

55

‘construction’, which may be seen as moving beyond interpretation in the sense of ‘reading’ the contract and into the realm of interpretation as a device to mould the meaning of the contract in order to give effect to objectives that are not fundamentally rooted in any conviction about what the parties’ words mean, but rather what they should or could have said. It is only in this sense that the various ‘forbidden grounds’ can be understood as involving any element of the ‘arbitrary’. Contra Proferentem The rule that ambiguities are to be interpreted against the party who drafted the document (contra proferentem) is an ancient one that is applicable to contracts of all sorts.35 In practical terms, in an insurance context, it becomes a rule that ambiguities are to be construed against the insurer, who is normally the proferens. (Though that rule might have other sources as well, such as a perception that, as the purpose of an insurance contract is to provide coverage, to maximise coverage is to effectuate the parties’ mutual intent.)36 Concern about the use of non-negotiable standard terms framed in a one-sided fashion by insurers goes back well into the last century; resolution of doubts against the insurance company is a time-honoured way to redress the balance. Thus, in 1901, the US Supreme Court referred to the general rule that where a policy of insurance is so framed as to leave room for two constructions the words used should be interpreted most strongly against the insurer.37

This is now a ‘universal rule’ in jurisdictions in the United States,38 including New York,39 and one on which the resolution of disputes about policy interpretation often turns. It is, however, expressly excluded under the Bermuda Form, whether it appears in the guise of a rule that the policy is to be interpreted ‘against the insurer’ or as a rule that the policy is to be interpreted based on the authorship of the language, or under its original Latin garb. Although strict contract theory distinguishes between a presumption against the author, a presumption against the insurer and a presumption against the proferens, in practice these principles merge together. All are excluded under every version of the Form.

35 Restatement (Second) of Contracts, s 206. 36 It may be argued that this is, as it stands, a questionable argument. The purpose of an insurance contract is undoubtedly to provide some coverage; any interpretation that renders coverage nugatory or discretionary would normally be suspect on that ground. But the purpose of the contract is not to provide limitless coverage, and defining what is not covered is as important as defining what is covered. One could just as well say – with equal lack of cogency – that because the purpose of exclusions in an insurance policy is to narrow the coverage granted, they ought to be interpreted as broadly as possible. While consideration of the purpose and background will often be helpful, it must be balanced. 37 Liverpool and London & Globe Ins Co v Kearney 180 US 132, 136 (1901). 38 Insurance Coverage Litigation (above n 32) at s 2.04. See also Restatement Liability Insurance, s 4 and cmt a. 39 The cases are legion: see, eg, Sturges Mfg Co v Utica Mut Ins Co 371 NYS2d 444, 448–49 (NY 1975); Thomas J Lipton, Inc v Liberty Mut Ins Co 357 NYS2d 356, 361 (NY 1974); Greaves v Public Serv Mut Ins Co 155 NE2d 390, 391–92 (NY 1959); Casey v General Accident Ins Co 578 NYS2d 337, 338 (App Div 1991) (citing Breed v Insurance Co of N Am 413 NYS2d at 355).

4.22

56

Interpretation of the Bermuda Form and the Modification of New York Law

Reasonable Expectations 4.23

The Form’s reference to the parties’ ‘reasonable expectations’ as one of the ‘forbidden canons’ is, at first sight, odd, particularly for lawyers not steeped in the litigation of large insurance coverage claims in the United States. It is hard to see why reflection on the parties’ reasonable understanding of the policy could be objectionable. Indeed, in many ways reference to the parties’ expectations appears to be the essence of interpretation. The explanation is that ‘reasonable expectations’ here is not to be taken simply at face value, but is probably a reference to a technical doctrine in insurance law in the United States, first outlined in a seminal article in the Harvard Law Review40 by Professor (later Judge) Keeton in 1970: The objectively reasonable expectations of applicants and intended beneficiaries regarding the terms of insurance contracts will be honored even though painstaking study of the policy provisions would have negated those expectations.41

4.24

The key words, which explain the inclusion of the doctrine in the list of prohibited grounds here, are italicised. At least as it has come to be applied, the ‘reasonable expectations’ doctrine may treat as ‘reasonable’ expectations that would be unjustified if the party in question (in practice, almost always the policyholder) had actually engaged in a sophisticated interpretation of the contract as a whole. So understood, the doctrine embodies the view that it is, as it were, ‘reasonable’ for a layperson to misunderstand a complex document, and that in certain circumstances this misunderstanding, as long as it is ‘reasonable’, should be treated as a correct reading of the language. This may be inconsistent with the principle that a policy should be interpreted as a whole. It is also inconsistent with the principle that interpretation should be evenhanded, because (despite some exiguous authority to the effect that the reasonable expectations of the insurance company might be relevant)42 the overwhelming tendency of courts applying the rule has been to prefer the interpretation likely to favour the policyholder over that likely to favour the insurance company.43 The Bermuda Form seeks to prevent the use of any doctrine using the ‘reasonable expectations’ that a particular provision might engender in the mind of an unsophisticated policyholder.44 The insurer does not generally object to the use of what the policyholder might objectively expect45 – that emphasis on objective expectation is quite consistent with the Bermuda Form’s approach.

40 Robert E Keeton, ‘Insurance Law Rights at Variance with Policy Provisions’, (1970) 83 Harvard Law Review 961. 41 Ibid at 967 (emphasis added). 42 State Farm Mut Auto Ins Co v Roberts 697 A2d 667, 672 (Vt 1997). 43 The question remains about what could the policyholder objectively expect. Some courts applying the doctrine require that the policyholder’s expectations be ‘objectively reasonable’; see, eg, Insurance Coverage Litigation (above n 32) at s 2.05. 44 The provision excluding use of the reasonable expectations doctrine is not present in earlier Forms. 45 See, eg, Atlantic Cement Co v Fidelity & Cas Co of NY, 459 NYS2d 425 (App Div 1983), aff ’d, 481 NYS2d 329 (NY 1984); IQ Originals, Inc v Boston Old Colony Ins Co 447 NYS2d 174 (App Div), aff ’d, 458 NYS2d 540 (NY 1982). See also Board of Educ, Yonkers City Sch Dist v CNA Ins Co 647 F Supp 1495 (SDNY 1986), aff ’d, 839 F2d 14 (2d Cir 1988); Champion Int’l Corp v Continental Cas Co 400 F Supp 978 (SDNY 1975), aff ’d, 546 F2d 502 (2d Cir 1976).

The ‘Forbidden Grounds’

57

However, it is clear that the insurer objects to the proposition that the decision-maker can conjure such objective expectations from some portion of the policy, even though a painstaking analysis of the whole policy would negate them. Insurers also tend to object to the further proposition that the intentions of either party, and especially of just one party rather than their mutual intentions, are relevant. It is not inappropriate to ponder the ‘reasonable expectations’ that would be mutually shared by the parties if they did make such a painstaking study, or forming the background to the transaction. Indeed, it is hardly possible to conceive of any process of interpretation that would not, consciously or subconsciously, reflect on that question. It is the ‘reasonable expectations’ doctrine, as a technical doctrine, that is targeted. Parol or Extrinsic Evidence Some comments on the exclusion of ‘parol’ and ‘extrinsic’ evidence have already been made above.46 Article VI(O) and its predecessors do not require that the Bermuda Form be interpreted in a contextual vacuum. The sort of material that, in our view, remains relevant and is not to be treated as ‘extrinsic’ evidence includes: the legal background against which the Bermuda Form was drafted (including the long history of the ‘comprehensive’ or later ‘commercial’ general liability (CGL) policy Form and its interpretation), and whose doctrines it sometimes adopts, sometimes spurns and sometimes modifies; the legal system within which tort claims – against which policyholders seek coverage – are made against manufacturers in the United States; the practicalities of defending and settling such claims; the structure of the insurance programme of which the Bermuda Form is part; the lines of business in which the insured is engaged (which are invariably known to both parties before inception); and the sort of liability to which such businesses characteristically give rise. What the provision does seek to prevent is evidence offered in an attempt to show, directly, the subjective meaning the parties to the contract (a fortiori one of them) had in using particular words. Thus, evidence of, for instance, the subjective intentions of the drafter of the contract, the meaning attached to the contract by one or other of the parties, or even oral expressions of opinion between the parties as to the meaning of the contract – though they might be permitted in case of ambiguity under New York law – are not admissible in relation to interpretation as such.

4.25

It might be thought that the exclusion of extrinsic evidence is not explicable as being designed simply to ensure that the Bermuda Form is construed so as to give effect to the mutual intention of the parties, without regard to extraneous matters. After all, extrinsic evidence is often used to elucidate precisely those intentions. It is at least arguable, however, that, as it has come to be applied, the use of extrinsic evidence goes beyond that function. There is a narrow line between seeking the parties’ intentions as expressed in a document and seeking those intentions as they might have been expressed in the

4.26

46 The terms are not identical, though they overlap. ‘Parol’ evidence traditionally means evidence of oral negotiations offered to vary or qualify the meaning of a written contract. ‘Extrinsic’ evidence is somewhat wider and includes but is not limited to what is called ‘parol evidence’.

58

Interpretation of the Bermuda Form and the Modification of New York Law

document. The first exercise is one of interpretation; the second is one of psychological reconstruction. Ordinary speech may not draw a very clear distinction between ‘what I meant by what I said’ and ‘what I meant to say’, but the distinction is real, especially when the ‘mutual intentions’ of the parties are to be looked at objectively. No doubt extrinsic evidence can occasionally throw real light on meaning. But the line between construction and reconstruction is easily crossed. The Bermuda Form seeks to ensure that the line is not crossed, and therefore it excludes evidence that is considered highly likely to do so. Doubtful Principles 4.27

The Bermuda Form identifies three specific doctrines – the contra proferentem rule, the ‘reasonable expectations’ doctrine in its technical sense, and the use of parol and extrinsic evidence – as prohibited. Beyond this, the clause prohibits ‘arbitrary’ constructions and presumptions. The reach of this prohibition may be controversial. Some ‘presumptions’ (such as the rule that the policyholder must bring itself within the basic terms of the insuring provisions, but the insurer must prove any relevant exclusion or ground negating coverage)47 are obviously not arbitrary. Others cause difficulty. For instance, is the rule that exclusions are interpreted strictly and narrowly48 an arbitrary presumption? It undoubtedly has strong historical connections with the contra proferentem rule. Nevertheless, it has become firmly established in its own right, and it does not necessarily seem arbitrary to approach a document on the basis that provisions carving out exceptions from cover prima facie granted by the policy ought generally to be rather closely confined. At the very least, care must be taken to ensure that the effect of applying those exceptions is to maintain an ‘evenhanded’ interpretation that reflects the grant of cover and that pays due regard to the drafter’s decision, as a matter of technique, to present the exclusion as such rather than as a primary defining term of that cover. There may be room for an intermediate position. That position recognises that there is nothing ‘arbitrary’ in resolving genuine ambiguities in exclusions on the basis that the parties would not intend an exclusion to apply unless it did so clearly and squarely, but eschews hyper-narrow meanings.

FRUIT OF THE POISONOUS TREE? 4.28

The modification of New York’s ordinary principles of construction raises one further difficulty: how far does it go? In developing general principles of insurance law, or authoritatively deciding the meaning of a particular phrase, New York courts frequently use one of the principles – such as interpretation contra proferentem – that the Bermuda Form

47 Throgs Neck Bagels 671 NYS2d 66 at 70 (App Div 1998). See also Great Northern Ins Co v Dayco Corp 637 F Supp 765, 777 (SDNY 1986). 48 See below ch 10, paras 10.03–10.06.

Fruit of the Poisonous Tree?

59

eschews.49 Does that mean that the relevant New York law authority should be ignored? Should one try to speculate about what the New York court would have done but for the forbidden principle? Or is the particular doctrine or interpretation a definite rule of New York law, so as to continue to have validity when applied to the Bermuda Form? There is no easy answer to these questions and what follows is only one view. We consider that the correct approach is as follows. First, where New York law has developed a doctrine – which is not itself a rule of construction – that is stated and applied generally, the doctrine should be applied to cases under the Bermuda Form, even though one of the doctrine’s intellectual roots is connected to a principle that would not itself be applied under Article V(O) or its predecessors. Were it to be otherwise, given the pervasive effect of these principles on New York courts’ approach to all insurance cases, there would probably be very little New York law left to apply. Yet the parties, though clearly aware of the existence of the ‘forbidden canons’ of construction in New York law and presumptively aware of their pervasive importance, nevertheless elected to apply New York law as the primary law. The proviso to that election should not be allowed to emasculate the basic choice, as it well might if any doctrine ‘tainted’ by the use of principles such as contra proferentem were, for that reason, simply abandoned. The contract would then be left to be interpreted in a legal vacuum, contrary to the parties’ express intentions, and the proviso would have swallowed the basic principle.

4.29

Second, where New York law has established an interpretation of a particular term commonly used in insurance policies generally and specifically in the Bermuda Form, one may presume that the interpretation was being used in the same sense in the Bermuda Form as well. That remains true even though the particular interpretation was reached originally by recourse to the ‘forbidden canons’. Where, for example, the drafter of this Form has chosen to use terms with a long history of use and interpretation in the standard CGL policy language, it is likely that the terms are used in the same sense in the Bermuda Form. The decision to use a particular form of words is likely to be best viewed as an attempt to preserve the same meaning for those words in the Bermuda Form, quite apart from any question about how New York law originally arrived at that meaning. Where a term is, in effect, a term of art in New York insurance law, it ought to be given a meaning in the Bermuda Form consistent with its meaning elsewhere.50

4.30

Third, however, where attempts are made to reason by analogy from decisions that construe different contractual language – especially where the decisions are sparse or decided by New York trial courts or other low-level New York courts – careful scrutiny may be appropriate. It might be easy to predict what a New York court that has decided

4.31

49 For instance, some of the decisions as to who bears the burden of proof seem to be influenced by this, though the issue of the burden of proof (who must prove?) is a free-standing issue quite apart from any question of construction (what must be proved?). See, eg, Pan Am World Airways, Inc v Aetna Cas & Sur Co, 505 F2d 989, 999–1001 (2d Cir 1974) (applying New York law). Yet it would be highly artificial to regard the rules of burden of proof as depending on the application of the contra proferentem rule or to suppose that the ordinary rules like those on the burden of proof did not apply to the Bermuda Form. 50 Eg, the long history of the ‘expected and intended’ provisions (discussed in ch 7) and the pollution exclusion (discussed in ch 10).

60

Interpretation of the Bermuda Form and the Modification of New York Law

such a case would do with the different language in the Bermuda Form; nevertheless, that prediction is of very limited value if the court’s decision were to rely upon one of the principles of interpretation from which the Bermuda Form departs. On the other hand, such decisions might well be relevant as background material – for instance, because they cogently explain a particular choice of words in the Bermuda Form, perhaps as a deliberate departure from earlier authority. Moreover, one must be careful not to adopt the fallacious argument that, because a particular decision was based on contra proferentem reasoning, the decision would, in the absence of such reasoning, have reached the opposite result. Because principles such as interpretation contra proferentem often provide a convenient short cut to choosing between competing viable interpretations, a New York court will not necessarily have expressed any view, or even considered, what choice should be made in the absence of such a principle. It is wrong, in general, to assume that because a court found both X and Y to be reasonable interpretations, and chose X because it favoured the policyholder, it would necessarily have chosen Y as the better interpretation, all things considered. X might be, on its own merits, the better interpretation, but the court may never have needed to address the issue. 4.32

In summary, we do not think it sensible to ignore any New York authority just because it makes use of the ‘forbidden canons’, or to abandon any established substantive doctrine of New York law merely because it rests on those grounds, or to consider afresh the meaning of words where New York authority that is directly in point exists. Where clear New York authority exists, it transcends its origins. Where clear New York authority does not exist, but a party invites the tribunal to reason by analogy from decisions about different language, the particular approach to interpretation under the Bermuda Form may well weaken the force of the analogy, and care will be needed to handle such authority in an appropriate way.

5 The Coverage Clause Article I of the Bermuda Form sets out the principal obligation of the insurer. It does so in language which at first sight appears relatively simple:

5.01

INSURING AGREEMENTS 1. COVERAGE [The insurance company] shall, subject to the limitations, terms, conditions and exclusions below, indemnify the Insured for Ultimate Net Loss the Insured pays by reason of liability: (a) imposed by law, or (b) of a person or party who is not an Insured assumed by the Insured under contract or agreement, for Damages on account of: (i) Personal Injury (ii) Property Damage (iii) Advertising Liability encompassed by an Occurrence, provided: COVERAGE A: notice of the Occurrence shall have been first given by the Insured in an Annual Period during the Policy Period in accordance with Article V of this Policy, or COVERAGE B: notice of the Occurrence shall have been first given during the Discovery Period in accordance with Article V of this Policy, but only if the Discovery Period option has been elected in accordance with the provisions of this Policy.

Each of the critical words in Article I is, however, subject to extensive definition elsewhere in the Form. Central to Article I is the definition of an ‘Occurrence’, which is discussed in greater detail in Chapter 6. The ‘Occurrence’ definition also includes the requirement that the relevant damage should not be expected or intended, and a proviso to that requirement, which preserves coverage for certain injury or damage and also creates what has generally been called the ‘maintenance deductible’. That aspect is considered in Chapter 7. This chapter discusses aspects of Article I, the coverage clause, other than the ‘Occurrence’ definition.

5.02

INSURANCE AGAINST LIABILITY The opening words of the Bermuda Form make it clear that the policy is a liability policy. When the Form was first introduced, the Declarations page of the 001 Form described it as ‘Umbrella Liability Insurance’. At the top of that Declarations page was the statement

5.03

62

The Coverage Clause

that the policy ‘is a claims made policy with an open-end discovery period on the terms and conditions set forth herein’. At the front of the 004 Form, a notice that appears on a separate page and typed all in capital letters describes it as ‘a standalone indemnity policy’. 5.04

The 004 Form covers the liability of the insured which is either ‘imposed by law’ or ‘of a person or party who is not an Insured assumed by the Insured under contract or agreement’. In either case, the liability must be ‘for Damages’, which is broadly defined in Article III(G) to include ‘all forms of compensatory damages, monetary damages and statutory damages, punitive or exemplary damages and costs of compliance with equitable relief’. The concept of a liability ‘imposed by law’ covers a very wide range of circumstances in which a policyholder will be responsible for ‘personal injury’, ‘property damage’ or ‘advertising liability’. A claimant who has suffered injury or damage is likely to advance a number of legal theories as to why the policyholder is liable, and these may include theories of negligence or other tortious conduct, strict liability and breaches of statute. These various sources of liability will be likely to come within the concept of ‘liability imposed by law’, although separate questions may arise as to whether breaches of particular statutes or other conduct result in the application of the policy exclusions.1 For example, in a typical products liability case, the allegation will be that the insured has manufactured and put into the stream of commerce a product that is inherently defective and ultimately causes injury or damage to end users. Such conduct, if proved, is likely to engage the strict products liability laws in various states.2 There can be no doubt that such liability would be a liability ‘imposed by law’ for damages.

5.05

In many cases of liability, there will no direct contractual relationship between the policyholder and the claimant, and the policyholder’s liability therefore flows from common law or statutory law rather than from the terms of a particular contract between the parties. There will, however, be other cases where there is a direct contract between the policyholder and the claimant: for example, the policyholder may have supplied a defective product to another manufacturer, who has incorporated the defective product into another product.3 Or there may be a contract between the policyholder and another company which leads to injury or damage to a third party: for example, a product is supplied to a wholesaler or distributor who sells the product on to a claimant who is injured, or the policyholder negligently provides architectural or engineering services to a client, and a third party is injured.4 In all these cases, liabilities arising out of the performance of the contract are likely to be liabilities ‘imposed by law’ and therefore within the coverage clause. Common law liabilities which have their origin in a contractual relationship between the parties are recoverable under an insurance policy covering liabilities ‘imposed by law’, even if a separate and independent liability flows from a contractual

1 For example, Article IV(P), the Securities Antitrust exclusion. 2 See, eg, in New York, McCarthy v Olin Corp 119 F3d 148, 154, 155 (2d Cir 1997) (New York strict products liability law requires harm from an alleged defect in the manufacture, warning or design of a product). 3 See, eg, Thomas J Lipton Inc v Liberty Mut Ins Co 357 NYS2d 705 (NY 1974) (defective noodles incorporated into the buyer’s soups resulted in liability for damage to the soups). 4 As contemplated by Article IV(C), which excludes errors and omissions in the rendering of professional services other than architectural or engineering services.

The Insured

63

provision such as an express indemnity.5 Thus, in the case where an original manufacturer of a product is liable to another manufacturer or distributor, who is in turn liable to consumers who have been injured, the original manufacturer’s liability at common law to pay damages to its customer would be a liability for damages ‘imposed by law’. If, as frequently happens, the manufacturer provides an express indemnity to a customer, or a subcontractor provides an indemnity to a contractor, then this would qualify for coverage under Article I as a liability ‘of a person or party who is not an Insured assumed by the Insured under contract or agreement’. Such language is typically included in insurance policies in order to cover the situation where one party has provided an indemnity to another.6 The inclusion of coverage for assumed liabilities avoids the need for an inquiry into whether, apart from the specific assumption of liability, a liability was ‘imposed by law’.

5.06

THE INSURED The Bermuda Form provides an indemnity to ‘the Insured’. The Declarations page of the policy will contain the name of the ‘Named Insured’. The company seeking insurance will typically be a substantial corporation with a number (possibly a very large number) of subsidiaries. Article III, the definitions section of the policy, contains an extended definition of ‘Insured’ so as to include subsidiaries and affiliates whose accounts are consolidated into the Named Insured’s annual financial statements in accordance with generally accepted accounting principles in the United States.7 The Insured under the policy will also include, for example, any subsidiary or affiliate of a foreign Named Insured where the accounts of the subsidiary or affiliate would have been so consolidated, had the foreign Insured drawn up its accounts in accordance with such accounting principles.

5.07

Article III(P)(2)(b) of the 004 Form8 also extends the definition of Insured to cover any subsidiaries, affiliates or associated companies which are listed in Schedule A to the policy. At the time that the Named Insured – that will usually be the ultimate parent company in the group – completes the application for insurance, it will have the opportunity to identify and list the subsidiaries that it wishes to cover. The insurer may, of course, specifically exclude a particular subsidiary from the coverage that it is willing to provide.

5.08

5 See, eg, O’Dowd v American Sur Co of NY 165 NYS2d 458 (NY 1957); Aetna Cas & Sur Co v L umbermens Mut Cas Co 527 NYS2d 143 (NY App Div 1988); Hawthorne v South Bronx Cmty Corp 576 NYS2d 203 (NY App Div 1991); Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) ss 17.01–02 (hereafter Masters and Stanzler, Insurance Coverage Litigation). 6 See, eg, Fisher v American Family Mut Ins Co 579 NW2d 599, 603–04 (ND 1998); and see generally Masters and Stanzler, Insurance Coverage Litigation, ch 17; Barry R Ostrager and Thomas R Newman, Handbook on Insurance Coverage Disputes 19th edn (New York, Wolters Kluwer, 2019) ss 7.01 and 7.07 (hereafter Ostrager and Newman, Handbook on Insurance Coverage Disputes). 7 Article III(P)(2)(a) of the 004 Form; Article III(a)(1)(B)(i) of earlier Forms. 8 Article III(a)(1)(B)(ii) of earlier Forms.

64

The Coverage Clause

5.09

The definition of Insured also covers officers, directors and employees, but only with respect to liability for acting within the scope of their duties as such.9 Where the Named Insured, or its subsidiaries, have agreed by contract to provide insurance to an outside entity, the policy provides coverage to that entity in respect of certain operations and facilities.10 The person or entity legally responsible for the use of automobiles owned or hired by certain insureds is also, subject to exceptions, within the definition of ‘Insured’.11 Article III(P)(6) of the 004 Form12 deals with the position of joint ventures. This provision should be read together with the provisions of Article II relating to joint ventures.13

5.10

Article III(P)(7)14 deals with the situation where, during an Annual Period, there is an acquisition, or the formation of a new company, or a merger. Certain acquisitions will automatically come within the cover, without any additional premium. However, the Bermuda Form imposes limits on the size of the new or acquired or merged entity which will qualify for free cover. A policyholder may think it sensible to notify its insurers of the acquisition and obtain confirmation that coverage applies. The converse situation, where the Named Insured sells or otherwise disposes of a company during the course of an Annual Period, is dealt with in Article VI(T).15

ULTIMATE NET LOSS/ALL SUMS INSURANCE … FOR DAMAGES 5.11

Article I of the original Bermuda Form (as well as the 002 and 003 Forms) sets out an agreement by the insurer to indemnify the Insured for all sums which the Insured shall be obligated to pay by reason of liability imposed upon the Insured by law or assumed under contract or agreement by the Insured for damages … resulting from [an occurrence].16

This language is similar to the ‘all sums’ language used in the Comprehensive General Liability (CGL) occurrence wordings, and which was in part responsible for the court decisions, on issues such as ‘trigger of coverage’ and ‘allocation’, that imposed substantial liabilities on insurers for long tail claims under those policies.17 The impact of the 9 Article III(P)(3) of the 004 Form; Article III(a)(1)(C) of earlier Forms. The policy is not, however, a Directors and Officers (D & O) policy. D & O insurance will typically provide cover against ‘wrongful acts’, whereas the Bermuda Form policy is a general liability policy that provides coverage for liability for personal injury, property damage and advertising liability. 10 Article III(P)(4) of the 004 Form; Article III(a)(1)(D) of earlier Forms. 11 Article III(P)(5) of the 004 Form; Article III(a)(1)(E) of earlier Forms. 12 Article III(a)(2) of earlier Forms. 13 See below ch 9, paras 9.40–9.44. 14 Article III(a)(3) of earlier Forms. 15 Article V(w) in earlier Forms: see below ch 11, para 11.111. 16 Article I of Forms 001, 002 and 003 (emphasis added). 17 As discussed in ch 1, other provisions of CGL policies led to these results, including the definition of occurrence and the timing language under those policies that activated, or triggered, coverage at the time of injury or damage, and were often read to trigger many policy periods in the case of injury or damage continuing over a period of years. The abbreviation ‘CGL’ also refers, generally and in this chapter, to later ‘Commercial General Liability’ policies: see above ch 1, para 1.02. See also Masters and Stanzler, Insurance Coverage Litigation, ch 4, discussing case law in the United States on these issues.

Ultimate Net Loss/All Sums Insurance … For Damages

65

‘all sums’ wording in these earlier Bermuda Forms is, however, substantially tempered by the wording of the occurrence definition, as well as the requirement that notice of occurrence must be reported during the currency of the policy. In the 004 Form, however, the language of ‘all sums resulting from an occurrence’ has been replaced with an indemnification ‘for Ultimate Net Loss the Insured pays by reason of liability … for Damages on account of (i) Personal Injury (ii) Property Damage (iii) Advertising Liability encompassed by an Occurrence’. In many and perhaps most cases, this change of language will not make any real difference. However, it is possible to envisage arguments that, on certain facts, the 004 wording is narrower in scope.18 The definition of Ultimate Net Loss in the 004 Form is, however, widely drawn:19

5.12

the total sum which the Insured shall become obligated to pay for Damages on account of Personal Injury, Property Damage and/or Advertising Liability which is, and/or but for the amount thereof would be, covered under the Policy less any salvage or recoveries.20

The word ‘Damages’ is also a defined term21 and it is again given in wide terms:

5.13

all forms of compensatory damages, monetary damages and statutory damages, punitive or exemplary damages and costs of compliance with equitable relief, other than governmental (civil or criminal) fines or penalties, which the Insured shall be obligated to pay by reason of judgment or settlement for liability on account of Personal Injury, Property Damage and/or Advertising Liability covered by this Policy, and shall include Defense Costs.22

It is not necessary for the policyholder to show that each element of an underlying claimant’s claim falls within the definition of, for example, personal injury or property damage. Article I and the definition of ‘Ultimate Net Loss’ both refer to a liability for damages ‘on account of’ personal injury and property damage. The ‘Damages’ definition uses the same language. The words ‘on account of’ require no more than a causal connection between the policyholder’s liability for damages and personal injury or property damage. In Lowenstein Dyes & Cosmetics Inc v Aetna Life & Casualty Co, the court stated that the words ‘on account of’ expressed only ‘a causal relationship, without requiring that the plaintiff suing an insured b[e] the party that has sustained the actual physical injury or property damage’.23 In that case, Lowenstein, a manufacturer, had supplied defective hair dyes to a business (Angel Hall). Angel Hall claimed to have suffered various losses in consequence, including money spent hiring beauty operators to restore the miscoloured hair of customers who had used Lowenstein’s products. Since the miscolouring of the customers’ hair constituted personal injury, Aetna was obliged to defend Angel Hall’s action against Lowenstein. Lowenstein’s potential liability to Angel Hall was ‘on account of’ personal injury, even though the customers had not sued Lowenstein directly and even 18 See further the discussion in ch 6 of what is encompassed by an ‘occurrence’. 19 Article III(AD). 20 ‘Ultimate Net Loss’ is a term widely used in other excess liability insurance policies and is often defined in a similar fashion. Some of the problematic cases on trigger and allocation have involved excess insurance policies using similar policy language. 21 Article III(G) of the 004 Form; Article III(g) and (i) of earlier Forms. 22 The term ‘Defense Costs’ is defined in Article III(H) of the 004 Form. There is no separate definition in the earlier Forms, but the ‘Damages’ definition in those Forms is longer. 23 Lowenstein Dyes & Cosmetics Inc v Aetna Life & Cas Co 524 F Supp 574, 578 (EDNY 1981), aff ’d, 742 F2d 1437 (2d Cir 1983) (unpublished table decision).

5.14

66

The Coverage Clause

though the remedial expenses were economic in nature (ie, the sums paid to the beauty operators) and had been incurred by Angel Hall rather than Lowenstein. We suggest that the words ‘on account of’ are therefore synonymous with wording in other policies such as ‘because of’ or ‘by reason of’.24 5.15

A claimant who has suffered personal injury may also suffer consequential losses, such as lost earnings, which are recoverable from the policyholder in the underlying litigation. We consider that such losses would be recoverable under the Bermuda Form, because the liability of the policyholder would be for damages ‘on account of’ personal injury. Another example is a case where the policyholder’s products have caused physical damage to a claimant’s house. The cost of repairing the house itself may be relatively small, but the claimant may have suffered other losses for which the policyholder is liable: for example, survey costs, architects’ fees, rental of alternative accommodation during the period of repairs, mental anguish, legal expenses and so on. All of these losses consequent on the property damage are in principle recoverable. Although they might not themselves be described as ‘property damage’, they are clearly ‘on account of’ property damage. If, however, no underlying personal injury, property damage or advertising liability has occurred, there can be no claim under the Bermuda Form.

5.16

The seminal New York case on this subject is Thomas J Lipton Inc v Liberty Mutual Insurance Co.25 In that case, the insured supplied contaminated noodles that were incorporated into Lipton’s soup products. Lipton then sued the insured for damage to the soup products. The New York Court of Appeals held that the insured could recover under its general liability policy for each of the categories of consequential damages claimed against it by Lipton, including the value of Lipton’s time in withdrawal, recall and destruction of the contaminated products; costs incurred in notification of the trade and general public; and loss of profits. Subsequent New York case law has upheld claims for indemnity against consequential loss in a variety of contexts.26

5.17

Factual questions can arise as to whether there is a sufficient causal connection between the ‘Personal Injury’ or ‘Property Damage’ and the Insured’s ‘liability … for Damages’, so that the latter is ‘on account of’ the former. In addressing such issues of causation, New York courts have taken a fairly generous approach. In Aetna Casualty and Surety Co v General Time Corp, the US Court of Appeals for the Second Circuit described a 24 As illustrated by one of the examples given in the Oxford English Dictionary: ‘We should celebrate two legged carrots and corkscrew runner beans, which are wonderful to eat but usually rejected on account of their looks.’ 25 Thomas J Lipton Inc 357 NYS2d 705. 26 See, eg, Aetna Cas & Sur Co v Gen Time Corp 704 F2d 80, 82–84 (2d Cir 1983) (lost profits were covered consequential damages where the insured’s defective electric motors damaged valves into which they were incorporated); ECDC Env’t LC v NY Marine & Gen Ins Co No 96 Civ 6033 (BSJ), 1999 WL 777883, at *2–*3 (SDNY 29 Sept 1999) (the policy covered any consequential damages arising from clean-up of discharged dredge into port, including lost profits); Burroughs Wellcome Co v Commercial Union Ins Co 632 F Supp 1213, 1221–22 (SDNY 1986) (policy coverage for DES injury also included liability for consequential damages). Courts in other jurisdictions have taken a similar approach: eg, American Home Assur Co v Libbey-Owens-Ford Co 786 F2d 22, 23, 25–26 (1st Cir 1986); Great Am Ins Co v Lerman Motors Inc 491 A2d 729, 730, 731, 733 (NJ Super Ct App Div 1984); Safeco Ins Co v Munroe 527 P2d 64, 68 (Mont 1974); United Props Inc v Home Ins Co 311 NW2d 689, 692 (Iowa Ct App 1981); Johnson v Studyvin 839 F Supp 1490, 1496 (D Kan 1993); Dimambro-Northend Assocs v United Constr Inc 397 NW2d 547, 550–51 (Mich Ct App 1986); Borden Inc v Howard Trucking Co 454 So 2d 1081, 1090–91 (La 1983).

Judgments

67

policy provision – which defined ‘property damage’ to mean ‘injury to or destruction of tangible property (other than property owned by the named Insured) and all direct and consequential loss resulting therefrom’ – as being ‘simply a succinct statement of the general rule of law which New York has adopted’.27 However, New York courts have permitted policyholders to recover losses consequent on personal injury or property damage, even though such losses may at first sight appear somewhat distant or remote from the acts which gave rise to the liability. In ER Squibb & Sons Inc v Lloyd’s and Cos,28 the Second Circuit held that the insurance covered Squibb’s liability for injuries suffered by individuals (third-generation offspring) who had not even been born at the time of Squibb’s tortious conduct. In that case, first-generation women had ingested a drug, and one issue was whether Squibb was liable for consequences which were distant in time; namely, injuries suffered by the grandchildren as a result of the fact that their grandmothers had ingested the drug. The court rejected the argument that there was something unfair, and perhaps unmanageable, about requiring coverage for claims that were in a sense far removed from the causative acts. The underlying tort law governing consequential claims itself provided limits on the liability of Squibb and hence insurers. If such law did impose liability on Squibb, then the damages payable were ‘by reason of’ liability and there was no reason to conclude that Squibb had intended to self-insure in respect of that distant element of its liability. The case therefore suggests that underlying litigation is the place for assessing the strength of the connection between the alleged conduct and the personal injury and consequential damages. However, it is not sufficient simply to show that there has been personal injury somewhere along the line which leads to the policyholder’s liability. If the underlying loss caused by the policyholder’s conduct is uncovered economic loss – for example, a decline in the value of investments – emotional distress consequential thereon is equally uncovered.29

5.18

JUDGMENTS If a case proceeds to judgment, and judgment is given against the policyholder, the policyholder will wish to claim for the amounts payable under the judgment, including any liability for the legal expenses of the successful plaintiff, as well as the costs incurred in mounting the unsuccessful defence. The amounts payable under the judgment will come within the broad definition of ‘Damages’ under Article III(G) of the 004 Form, which includes ‘all forms of compensatory damages, monetary damages and statutory damages, punitive or exemplary damages and costs of compliance with equitable relief’. Similar wording appears in earlier versions of the Form.30 If the court orders the policyholder to reimburse the successful claimant for some or all of its legal costs of the litigation, 27 Aetna v General Time 704 F2d 80 at 83–84. 28 ER Squibb & Sons Inc v Lloyd’s & Cos 241 F3d 154 (2d Cir 2001). 29 First Inv’rs Corp v Liberty Mut Ins Co 955 F Supp 274, 278–79 (SDNY 1997), aff ’d, 153 F3d 162 (2d Cir 1998). See also Smith Pontiac-GMC Truck Ctr Inc v Hartford Accident & Indem Co 599 NYS2d 308 (App Div 1993) (finding that the policyholder’s breach gave rise only to economic loss rather than injury from an automobile accident that occurred subsequently). 30 Article III(g) of the 001 and 003 Forms; Article III(i) of the 002 Form.

5.19

68

The Coverage Clause

then the sums payable would also come within the broad definition of ‘Damages’.31 In the United States, however, this would be somewhat unusual in the absence of a statute or contract providing for such recovery. The ‘English rule’ that an unsuccessful party generally pays the costs of the successful party in litigation is not the typical rule. In the absence of statutory or other authority specifically authorising such recovery, US courts generally follow the ‘American rule’ requiring each party to bear its attorneys’ fees and costs of litigation.32 Under the ‘Loss Payable’ Condition, Article VI(G), the policyholder must have paid the amount of the judgment in order to make a claim under the policy.33 ‘Defense Costs’ 5.20

The ‘Damages’ definition in the 004 Form states specifically that it ‘shall include Defense Costs’. Such costs are again broadly defined in Article III(H) of the 004 Form, and include ‘reasonable legal costs and other expenses incurred by or on behalf of the Insured in connection with the defense of any actual or anticipated Claim’, but exclude the ‘salaries, wages and benefits of the Insured’s employees and the Insured’s administrative expenses’. In previous versions of the Form, the policyholder’s own legal and other costs formed part of a much longer definition of ‘Damages’. It appears clear from this wording in the various versions of the Bermuda Form that coverage extends to the policyholder’s own legal and other expenses in defending the underlying action, and it would indeed be surprising if a liability insurance policy did not cover such costs.

5.21

A question arises, however, as to how the coverage for the policyholder’s ‘Defense Costs’ relates to other parts of the Bermuda Form wording – namely the earlier part of the ‘Damages’ definition and the Loss Payable Condition – which require a judgment or settlement. Thus, the ‘Damages’ definition refers to compensatory and other damages ‘which the Insured shall be obligated to pay by reason of judgment or settlement for liability on account of Personal Injury, Property Damage and/or Advertising Liability covered by this Policy and shall include Defense Costs’. Strictly speaking, the policyholder’s liability for its own defence costs in the underlying action will arise independently of a judgment or settlement, Thus, if judgment is given against the policyholder, the policyholder’s defence costs will form no part of that judgment, and the policyholder will not be ‘obligated’ by that judgment to pay those costs. Similarly, if the policyholder settles a claim, it arguably will not become ‘obligated’ by that settlement to pay its own defence costs in the underlying action.34 Indeed, the cases in which the policyholder needs to seek an indemnity from

31 City of Ypsilanti v Appalachian Ins Co 547 F Supp 823, 827–28 (ED Mich 1982), aff ’d, 725 F2d 682 (6th Cir 1983) (unpublished table decision); Hyatt Corp v Occidental Fire & Cas Co 801 SW2d 382 (Mo Ct App 1990); Sokolowski v Aetna Life & Cas Co 670 F Supp 1199, 1209–10 (SDNY 1987) (relying on Pac Ins Co v Burnet Title Inc 380 F 3d 1061, 1065–66 (8th Cir 2004)); St Paul Surplus Lines Ins Co v Life Fitness No C3-99-9980, 2001 WL 588949 (Minn Dist Ct 29 Jan 2001); City of Kirtland v W World Ins Co 540 NE2d 282, 285 (Ohio Ct App 1988); Sylvania Twp Bd of Trustees v Twin City Fire Ins Co No L-03-1075, 2004 WL 226115, at *6 (Ohio Ct App 6 Feb 2004). 32 See below ch 11, para 11.87 n 129. 33 See below ch 11, para 11.32. 34 If the settlement provides for each party to bear its own costs, it could be argued that the policyholder is ‘obligated’ to pay those defence costs ‘by reason of the settlement’. The contrary view is that the obligation to

Judgments

69

its insurer are those where the judgment or settlement is adverse to the policyholder, and therefore where the costs of defence are not being recovered from the claimant. Thus, if it were to be suggested that defence costs were recoverable only in circumstances where the judgment or settlement obligated the policyholder to pay those costs, the coverage for defence costs would in practice be illusory. These considerations, as well as the structure of the ‘Damages’ definition,35 suggest that the coverage for ‘Defense Costs’ may not be linked to judgments adverse to the policyholder or settlements concluded by the policyholder. If so, in a case where the defence costs are sufficient in themselves to exceed the policyholder’s retention, the policyholder could ask for reimbursement of its defence costs prior to judgment or settlement.36 The contrary argument is that both the ‘Damages’ definition and the Loss Payable condition require a judgment or settlement in order to crystallise the policyholder’s entitlement to recover not simply the sums payable under that judgment or settlement, but also the defence costs incurred in relation to the claims resolved by the judgment or settlement.

5.22

A related question is whether defence costs are recoverable in circumstances where the policyholder has mounted a successful defence, perhaps at very significant cost, in relation to the underlying claims. In such cases, there will be no adverse judgment against the policyholder. It would be surprising if the policyholder were to be better off by losing a case (in which case the insurer would have to pay both the amount of the judgment and the policyholder’s defence costs) than by winning the case (so that no judgment sum would be payable and need to be claimed under the insurance).37 We suggest that the

5.23

pay the defence costs arises by reason of the (antecedent) agreement between the policyholder and its lawyers, rather than by reason of the settlement. 35 The words ‘and shall include Defense Costs’ appear right at the end of the sentence, and are separate from the earlier part of the sentence which relates to those matters which ‘the Insured shall be obligated to pay by reason of judgment or settlement’. 36 A number of cases from New York and other jurisdictions suggest that, even in the absence of a duty to defend, the insurer may (under a policy which covers defence costs) be obliged to pay the policyholder’s defence costs for potentially covered claims as they are incurred, leaving policy defences to be litigated at a later stage. If the insurer later succeeds on the policy defences, then the insurer may be entitled to recoup the payments. The cases in New York do, however, contain different approaches to the question, and it may not be possible to reconcile all the cases. See PepsiCo Inc v Continental Cas Co 640 F Supp 656, 659–60 (SDNY 1986), questioned on other grounds by Waltuch v Conticommodity Servs Inc 88 F3d 87 (2d Cir 1996); McGinniss v Employers Reinsurance Corp 648 F Supp 1263, 1271 (SDNY 1986); In re Ambassador Grp Inc Litig 738 F Supp 57, 63 (EDNY 1990); Wedtech Corp v Federal Ins Co 740 F Supp 214, 221 (SDNY 1990); In re Kenai Corp 136 BR 59, 63 (SDNY 1992); Stonewall Ins Co v Asbestos Claims Mgmt Corp 73 F3d 1178, 1218–19 (2d Cir 1995), modified on denial of reh’g, 85 F3d 49 (2d Cir 1996); Nu-Way Envt’l Inc v Planet Ins Co No 95 Civ 573 (HB), 1997 WL 462010, at *3 (SDNY 12 August 1997); Lowy v Travelers Prop & Cas Co No 99 Civ 2727 (MBM), 2000 WL 526702, at *2–*4 (SDNY 2 May 2000); Federal Ins Co v Tyco Int’l Ltd 784 NYS2d 920 (Sup Ct 2004); In re World Com Inc Sec Litig 354 F Supp 2d 455, 464 (SDNY 2005); Federal Ins Co v Kozlowski 792 NYS2d 397, 403 (App Div 2005). Where defence costs are paid pursuant to a duty to defend, the New York Appellate Division has held that these cannot be recouped in the absence of an express provision to that effect: American W Home Ins Co v Gjonaj Realty & Mgmt Co 192 AD3d 28,41 (App Div 2020). The Restatement of the Law, Liability Insurance states that the ‘default rule’ in the United States is no recoupment of defence costs unless ‘otherwise stated in the insurance policy or otherwise agreed to by the insured’. American Law Institute, Restatement of the Law, Liability Insurance (St Paul, MN, American Law Institute Publishers, 2019) s 21 and s 21 cmt a (hereafter Restatement Liability Insurance). 37 Such a result would also potentially discourage settlement and contravene the public policy in New York favouring settlement: see below n 52.

70

The Coverage Clause

words of the definition of Damages ‘and shall include Defense Costs’ are sufficiently broad to cover costs incurred in mounting both successful and unsuccessful defences. The definition of ‘Damages’ does not draw a distinction between the costs incurred in these two situations. The words ‘obligated to pay by reason of judgment’ does not produce this distinction; as discussed above, the obligation to pay defence costs – whether of a successful or unsuccessful defence – does not arise ‘by reason of judgment’. Nor, we suggest, does the Loss Payable Condition produce this distinction; that clause is concerned with the timing of the insurer’s obligation to pay, not with the categories of loss that the policyholder is entitled to recover.

SETTLEMENTS BY THE POLICYHOLDER 5.24

A policyholder will often seek to settle the underlying claim, rather than risk the consequences of an adverse jury verdict and attendant legal fees and costs. The definition of ‘Damages’ in Article III of the 004 Form (and equivalent definitions in earlier Forms) expressly refers to settlements. It is not unusual, however, for issues to arise under liability policies with regard to coverage for a settlement concluded by the policyholder. For example, the insurer may contend that the policyholder actually had no legal liability to the underlying claimant and therefore should not have settled. Or the insurer may contend that the policyholder did not obtain its approval for a settlement and should have; see the Loss Payable condition in the Bermuda Form.38 In some of the decided cases, the policyholder has faced arguments based upon a failure to obtain consent to a settlement in circumstances where the insurer has reserved rights under the policy and has not expressed a view when told of a proposed settlement by the policyholder who wants to settle an underlying claim. Questions can also arise as to whether the attachment point under the policy has been reached and hence whether a settlement is within the scope of the approval provision. For example, does the approval provision apply to a settlement that is below the attachment point of the policy? Does it apply only to settlements that the insurer is called upon to pay? Legal Liability to Pay

5.25

Under New York law, a policyholder who seeks to recover in respect of a settlement does not have to show facts proving that it was actually liable to the underlying claimant 38 Article VI(F) of the 004 Form; Article V(g) of earlier Forms. Article VI(F) provides that the ‘Insured’s liability covered hereunder shall have been fixed and rendered certain either by final judgment against the Insured after actual trial or by settlement approved in writing by Underwriters’ (emphasis added). As discussed in more detail in ch 11 below, paras 11.34–11.39, the Loss Payable Condition uses the word ‘approved’ rather than ‘consent’, and this distinction may be relevant to the question of whether approval can be sought subsequent to the conclusion of a settlement. We there express the view that the policyholder does not breach a condition of the policy by failing to obtain approval prior to settlement: cf Vigilant Ins Co v Bear Stearns Cos 855 NYS2d 45 (NY 2008). Although it is sensible to seek approval prior to concluding a settlement, or at least to obtain a statement from the insurer that it will not subsequently rely upon lack of approval as a defence to the policyholder’s claim, we suggest that approval can also be sought subsequently. The insurer needs to respond consistently with its duty of good faith and fair dealing: see below paras 5.37–5.39 and ch 11, para 11.39.

Settlements by the Policyholder

71

(that is, to prove liability against itself). New York law recognises that, because settlement agreements typically do not concede liability or specify the precise basis for the settlement, it would be unreasonable and unfair to require a policyholder to prove the case against itself or to prove that the case made by the claimant in the underlying claim is true. Rather, a policyholder need demonstrate only that settled claims are of a ‘type’ included within the policy’s coverage and that the settlement was reasonable. The policyholder can recover in respect of a settlement so long as it had a potential liability for a claim that would itself have been a covered loss, and the settlement was reasonable in the light of that risk.39 Under the Bermuda Form, this general approach is reinforced by the references in the ‘Occurrence’ definition to alleged injury or damage. Accordingly, a policyholder is not required to litigate all the facts of a settled claim to obtain insurance coverage under New York law. In Luria Brothers & Co v Alliance Assurance Co, the US District Court for the Eastern District of New York, applying New York law, found that the policyholder was ‘potentially liable’ under the allegations of the complaint and that ‘the policies covered this type of liability’.40 The US Court of Appeals for the Second Circuit, sitting in New York, affirmed this decision, holding that:

5.26

In order to recover the amount of the settlement from the insurer, the insured need not establish actual liability to the party with whom it has settled ‘so long as … a potential liability on the facts known to the [insured is] shown to exist, culminating in a settlement in an amount reasonable in view of the size of possible recovery and degree of probability of claimant’s success against the [insured]’.41

The decision in Luria was applied in Uniroyal Inc v Home Insurance Co.42 The policyholder, a manufacturer of ‘Agent Orange’, sought insurance coverage for its defence costs and the settlement of a personal injury class action by Vietnam veterans and their families. The insurer had refused to participate in the settlement and failed to defend the policyholder. In moving for summary judgment, the insurer (Home) argued that the policyholder was required to prove ‘actual injury’ in order to show a covered loss, because the underlying class action tort lawsuit was settled without factual findings establishing the policyholder’s liability in the underlying action. The court rejected this argument, recognising that if the policyholder were forced to prove its liability in order to get insurance coverage, it would be placed in the ‘hopelessly untenable position of having to refute liability in the underlying action until the moment of settlement, and then of turning about face to prove liability in the insurance action’.43

39 Uniroyal Inc v Home Ins Co 707 F Supp 1368, 1378 (EDNY 1988). For a review of issues that can arise in relation to settlements, see Thomas R Newman and Aidan McCormack, ‘Challenging an Insured’s Settlement’ (Winter 1997) 47(2) Federation of Insurance & Corporate Counsel Quarterly No 2. 40 Luria Bros & Co v Alliance Assur Co 780 F2d 1082, 1087 (2d Cir 1986). 41 Ibid at 1091 (quoting Damanti v A/S Inger 314 F2d 395, 397 (2d Cir 1963)). See also Texaco A/S Denmark v Commercial Ins Co of Newark 160 F 3d 124, 128 (2d Cir 1998) (following Luria); Dayton Indep Sch Dist v National Gypsum Co 682 F Supp 1403, 1406–07 (ED Tex 1988) (applying New York law), rev’d on other grounds sub nom WR Grace & Co v Cont’l Cas Co 896 F2d 865 (5th Cir 1990); Uniroyal 707 F Supp at 1379. 42 707 F Supp 1368. 43 Ibid at 1378–79.

5.27

72

5.28

The Coverage Clause

In so holding, the court distinguished the decision of the New York Court of Appeals in Servidone Construction Corp v Security Insurance Co of Hartford,44 holding that Servidone: required only that the claim settled by the insured be a ‘covered loss’ under the policy, and not that the insured independently prove the truth of the underlying claim. In Servidone the Court of Appeals held that a claim of a type excluded by the policy could not be indemnifiable if settled; it never held that an otherwise covered claim once settled, must be proven anew by the insured.45

The court in Uniroyal also noted that often the evidence needed to prove ‘actual injury’ would be unavailable to the policyholder. The court was considering insurance policy language in which the insurer agreed to pay ‘all sums which the insured shall be obligated to pay by reason of the liability … assumed under contract or agreement’. Similar language appears in Bermuda Forms 001–003. The court in Uniroyal also held that a settlement of an underlying personal injury tort action constitutes a ‘contract or agreement’ within the meaning of the policy, so long as the occurrence is covered by the policy. 5.29

The legal position under New York law is therefore fundamentally different from the approach taken under English law, where there is considerable authority to the effect that a policyholder who has settled a claim must, in an action against its liability insurer, prove liability against itself.46 In AstraZeneca Insurance Co Ltd v XL Insurance (Bermuda) Ltd, the English Court of Appeal applied this principle to a Bermuda Form policy, notwithstanding arguments to the contrary based upon the policy wording.47 Since that case concerned a policy governed by English law, neither the decision nor its reasoning applies where, as is more usually the case, the policy is governed by New York law.

5.30

Questions can of course arise as to whether, on the facts, a particular settlement was reasonable. In In re Agent Orange Product Liability Litigation,48 the court took into account the procedural posture of the litigation, the difficulty any plaintiff would have in establishing its case against any one or more of the defendants, the uncertainties associated with a trial, and the unacceptable burden that continued litigation would place on counsel for the plaintiffs and defendants and on the courts. In Uniroyal, the court noted that the reasonableness of a settlement can include an analysis of the probability of loss and the probable size of the loss. The reasonableness of a settlement is judged on the facts known to the insured at the time of the settlement.49 Accordingly, the court in Uniroyal held that a settlement was nevertheless reasonable despite the fact that, following the settlement, a court had rejected the claims of certain claimants. Evidence as to the

44 Servidone Constr Corp v Sec Ins Co of Hartford 488 NYS2d 139 (NY 1985); see Ostrager and Newman, Handbook on Insurance Coverage Disputes, s 2.05[c]. 45 Uniroyal 707 F Supp at 1379. 46 The English law authorities are reviewed in AstraZeneca Insurance Co Ltd v XL Insurance (Bermuda) Ltd [2013] EWCA Civ 1660, [2014] 2 All ER (Comm) 55. See too ch 3, para 3.29 above. 47 [2013] EWCA Civ 1660, [2014] 2 All ER (Comm) 55. 48 In re Agent Orange Prod Liab Litig 597 F Supp 740 (EDNY 1984), aff ’d, 818 F2d 145 (2d Cir 1987). 49 Uniroyal 707 F Supp at 1379 (citing Luria).

Settlements by the Policyholder

73

reasonableness of a settlement is likely to consist principally of factual evidence, for example, from a senior executive of the policyholder or a senior individual within the policyholder’s legal department. If a policyholder adduces evidence of this kind, q uestions can arise as to whether privilege has been or should be waived in relation to legal advice concerning the underlying litigation or settlement.50 It is possible that a tribunal would be willing to hear expert evidence directed towards the reasonableness of a settlement, for example, statistically based evidence. However, an arbitral tribunal with a majority of English members might well consider that the reasonableness of a settlement was very much a matter of common sense, to be decided by reference to the particular facts of the dispute, and that expert evidence would not be of much assistance.51 If factual evidence as to why the litigation was settled is tendered from the lawyers handling the underlying litigation for the policyholder, the evidence would to some extent cross the boundaries between factual and expert evidence in any event. The decisions in Luria and Uniroyal reflect the long-standing New York public policy favouring settlement of claims.52 The policy of encouraging settlements (in New York and in the United States generally) is so strong that, in Stonewall Insurance Co v Asbestos Claims Management Corp, the court found that the insurance companies were bound by the underlying settlement agreement, even though that agreement resulted in the policyholder paying claims for damages resulting from the use of someone else’s products.53 The court found that, because the formula used to determine liability resulted in an overall cost reduction, it was a reasonable way of resolving the claims.54

50 For a discussion of issues of discovery, privilege and waiver of privilege, see below ch 16. 51 See below ch 15, paras 15.42–15.50, where the question of expert evidence in the context of an English arbitration is discussed. When issues of the reasonableness of settlements arise under English law, the courts sometimes take an objective approach: see X Corp Ltd v Y (A Firm) (Moore Bick J) 16 May 1997 (unreported). However, this is a matter of substantive English law rather than procedure and should not be transplanted into the context of an insurance policy governed by the substantive law of New York. 52 See, eg, Denburg v Parker Chapin Flattau & Klimpl 624 NE2d 995 (NY 1993) (‘Strong policy considerations favor the enforcement of settlement agreements’); In re NY Cty Data Entry Worker Prod Liab Litig 616 NYS2d 424, 427 (Sup Ct 1994) (‘It is well settled that strong policy considerations favor settlements’), citing Hallock v State of NY 474 NE2d 1178 (NY 1984); Randell Elec Inc v State of NY 150 AD2d 875, 876 (NY App Div 1989). Other states in the United States follow this same public policy: see, eg, Vitkus v Beatrice Co 127 F3d 936, 944–45 (10th Cir 1997) (applying Illinois law, citing Luria) (the policyholder need only ‘demonstrate that it had a “reasonable anticipation of liability when it settled the underlying cases”’), citing US Gypsum Co v Admiral Ins Co 643 NE2d 1226, 1244 (Ill Ct App 1994); Black v Goodwin Loomis & Britton Inc 681 A2d 293, 299 (Conn 1996) (holding that stipulated judgment was not contrary to public policy, rejecting the insurer’s challenge to settlement on the grounds of fraud or collusion); Miller v Shugart 316 NW2d 729 (Minn 1982) (same); Pub Util Dist No 1 v Int’l Ins Co 881 P 2d 1020, 1032 (Wash 1994) (en banc) (the court enforced the policyholder’s coverage for settlement which was covered by the policies and with regard to which insurers failed to any prejudice from the policyholder’s settlement in violation of a ‘no action clause’ in the policies). 53 Stonewall 73 F3d 1178 at 1206–07. 54 In the context of asbestos claims, the development of the case law was such that asbestos producers were sometimes held liable even in the absence of proof that the injured claimant had used their particular products; eg, with the court applying liability on the basis of the producer’s generic share of the market. This development, which started before the ‘liability crisis’, perhaps explains some of the provisions of the Bermuda Form which refer to both the insured’s products or ‘similar products causing or allegedly causing’ personal injury; see, eg, the occurrence definition in the 001 Form. This approach, called ‘enterprise liability’, arose out of a case that led to the initial flood of asbestos cases in the 1970s: Borel v Fibreboard Paper Prods Corp 493 F2d 1076 (5th Cir 1973).

5.31

74

The Coverage Clause

5.32

It is still, however, necessary for the policyholder to prove that the underlying claim fell within the scope of the insurance policy.55 Therefore, the policyholder in these situations generally must show: (i) that the policy applies to the claim; and (ii) that the settlement was reasonable.56

5.33

A related aspect of these principles is that, in a mass tort case involving many actions with similar facts, the courts in the United States typically make their determination on the basis of representative complaints.57 Courts in New York (and indeed elsewhere in the United States) recognise that, particularly in mass tort insurance coverage cases, which may involve coverage for hundreds of thousands of underlying claims, it is impractical, unnecessary and virtually impossible to require the trier of fact to examine each and every individual underlying claim. It might be argued that this is a rule of procedure rather than substance and that an English tribunal should not automatically apply it.58 In practice, however, it is a sensible approach that is likely to commend itself to an arbitral tribunal. Approval of Settlements

5.34

The Loss Payable Condition in the Bermuda Form envisages that the insurers’ approval of settlements will be sought and obtained. Bermuda Forms 001–003 did not make it clear whether approval would be needed in respect of all underlying settlements, however remote from the layer being written by the Bermuda Form insurers. The approach which appears more appropriate is that approval would be needed only for those settlements that directly affect those insurers because they are settlements which the Bermuda Form insurers are called upon to pay. In practice, it is unlikely that a high-level catastrophe insurer (for example, an insurer writing excess of US$100 million) would want, or even expect, to have its approval sought for low-level settlements that are unlikely to involve the excess layer and in any event are likely to be handled by the policyholder’s primary or lower-level excess insurers. A policyholder might consider it sensible, in order to avoid argument, to notify its Bermuda Form (and other excess) insurers even of low-level settlements. Under the 004 Form, the position appears clearer. The Loss Payable Condition refers to ‘the Insured’s liability covered hereunder’, thereby indicating that consent is required only in respect of the higher-level liabilities that the Bermuda Form policy actually insures.59

5.35

What if the Bermuda Form insurer, upon notification of a proposed settlement, takes no notice or declines to comment, or simply reserves its rights? An insurer will often wish 55 See, eg, Servidone Constr Corp v Security Ins Co 488 NYS2d 139 (NY 1985); Morgan Stanley Group Inc v New England Ins Co 36 F Supp 2d 605, 608 (SDNY 1999), aff ’d in part, rev’d in part, 225 F3d 270 (2d Cir 2000) (affirming summary judgment for insurer on one claim; vacating the summary judgment for the insurer on other claims, remanding for further factual findings); Uniroyal 707 F Supp at 1379. 56 See Morgan Stanley 36 F Supp 2d at 608. In Servidone, the insurer alleged that the claim fell within a policy exclusion, and accordingly bore the burden of establishing that this was the case. 57 See, eg, Stonewall Ins Co v National Gypsum Co No 86 Civ 9671, 1992 WL 123144, at *7 fn 10 (SDNY 27 May 1992), aff ’d in part, rev’d in part by Stonewall 73 F3d 1178, quoting US Fidelity & Guar Co v Wilkin Insulation Co 550 NE2d 1032,1036 (Ill Ct App 1989); Burroughs Wellcome 632 F Supp at 1218 (finding scrutiny of underlying claims in a mass tort case unnecessary where the facts underlying the complaints were similar). 58 See above ch 3, paras 3.30–3.45 for a discussion of the difference between issues of substance and procedure. 59 See below ch 11, para 11.34 and, generally in relation to the Loss Payable Condition, paras 11.27–11.55.

Settlements by the Policyholder

75

not to be drawn into a debate about whether or not the policyholder should settle and might simply advise it to act prudently. For example, the claim may be a long way from the insurer’s layer, or the insurer may not want to spend time or money considering or investigating it, or the company may even have reserved its rights under the policy quite generally in the hope of preserving all possible defences in the event that a claim is eventually made. Under New York law, an insurer who is asked to consent, but takes a hands-off approach, cannot later turn around and invoke its lack of consent as a reason for not paying the claim. In effect, by not responding, the insurer has told the policyholder that it is on its own. Provided that the policyholder acts reasonably thereafter in settling the claim,60 it will be able to recover. Thus, an insurer is estopped from denying coverage on the basis of failure to procure consent to settlements if the policyholder makes repeated attempts to obtain the insurer’s consent or participation in the underlying negotiations and the insurer never responds. In Allstate Insurance Co v Sullivan,61 the policyholder wrote several letters regarding settlement that the insurer ignored, and the court held that the insurer was estopped from raising the defence of failure to obtain consent to settlement. In MBIA Inc v Federal Insurance Co, the US Court of Appeals for the Second Circuit said that by an insurer’s ‘unreasonable delay, silence or conduct, it can either waive a consent requirement or acquiesce in a settlement’, thereby precluding reliance on a clause requiring consent to settlements.62

5.36

A number of New York cases have addressed the question of settlement in the context of an insurer having denied coverage or breached its duty to defend. For example, in Bunge v London & Overseas Insurance Co,63 the court, applying New York law, held that it was well settled that ‘at least after a denial of liability by an insurer, the insured may enter into a settlement with a third party without prejudicing its rights against the insurer’.64 In order for this principle to apply, it is not necessary that the insurer has expressly denied coverage, nor that its denial relate to the particular claim. In Isadore Rosen & Sons Inc v Security Mutual Insurance Co of New York,65 the insurer had delayed responding to the claim. The New York Court of Appeals held that the insurer’s obligation to act in good faith for the policyholder’s interests could be ‘breached by neglect and failure to act protectively when the insured is compelled to make a settlement at his peril; and unreasonable delay by the insurer, in dealing with a claim, may be one form of refusal

5.37

60 See the discussion of Luria and Uniroyal above. 61 Allstate Ins Co v Sullivan 646 NYS2d 359, 360 (App Div 1996). See also State Farm Mut Ins Co v Del Pizzo 586 NYS2d 310 (App Div 1992). In this context, the insurer’s silence may be treated as acquiescence to settlement (at 311–12: the insurer’s failure to respond to letters from the policyholder notifying of settlement may be deemed an acquiescence to the settlement). 62 MBIA Inc v Fed Ins Co 652 F3d 152, 169 (2d Cir 2011) (applying New York law). See also Aetna Cas & Sur Co v Crown 181 AD2d 883 (App Div 1992); State Farm Mut v Del Pizzo 586 NYS2d 310 (App Div 1992); State Farm Auto Ins Co v Blanco 617 NYS 2d 898 (App Div 1994). 63 Bunge v London & Overseas Ins Co 394 F2d 496 (2d Cir 1968). 64 Ibid at 497–98; see also Luria 780 F2d at 1091; In re Joint E & S Dist Asbestos Litig 78 F3d 764, 779 (2d Cir 1996) (applying New York law). 65 Isadore Rosen & Sons Inc v Security Mut Ins Co of NY 339 NYS2d 97 (NY 1972). See also JP Morgan Sec Inc v Vigilant Ins Co 58 NYS 3d 38 (App Div 2017).

76

The Coverage Clause

to perform which could justify settlement by the insured’. In HS Equities Inc v Hartford Accident & Indemnity Co,66 the policyholder had become aware, primarily through conversations with its insurance broker, of the insurer’s general position as to why there was no coverage under the relevant policy, and the insurer’s arguments for taking that position. Although this may not have been a denial of liability for the specific claim, the court held that the expression of the insurer’s general position was ‘the equivalent of a denial of liability’ in relation to the specific underlying action. In Texaco A/S Denmark v Commercial Insurance Co,67 the insurer’s decision to allocate claims to a particular set of policies meant that the insurer ‘effectively denied coverage’ under another set of policies. The rationale of these cases applies equally, we suggest, to a case where the insurer simply reserves its rights as to coverage and declines to participate in settlements. In both situations, the policyholder must decide how best to litigate the underlying actions without the input or approval of the insurer. 5.38

What if, however, the insurer has not declined coverage, or has reserved its rights, but expressly refuses approval for the underlying settlement? This scenario is perhaps likely to arise only in circumstances where the policyholder and insurer are cooperating in the defence of the claim, or where no issue has arisen between the insurer and policyholder as to liability under the policy. A policyholder in this situation may find it more difficult to argue that the refusal of approval does not operate to bar the claim, at least unless the refusal is arbitrary or unreasonable.68 It is not a situation where the insurer has denied or reserved the right to deny coverage and therefore, directly or implicitly, told the policyholder that it is on its own. In principle, and subject to the obligation of good faith and fair dealing,69 there is no reason why a liability insurer should not tell its policyholder to fight an underlying case in circumstances where the insurer has agreed to cover an adverse judgment and has bargained for a provision requiring its approval of settlements.70 Problems can arise particularly in circumstances in which the proposed settlement would require the insurer to pay a total loss under the policy. In these circumstances, the insurer has no real incentive to agree to a settlement: it will be no worse off if the policyholder fights the case and loses. Equally, the policyholder has every incentive to agree to the settlement because the settlement may be funded in whole or in substantial part by its insurers, and the policyholder will avoid the risk of a more severe judgment which may exceed the total limits in its insurance programme.

66 HS Equities Inc v Hartford Accident & Indem Co 609 F2d 669 (2d Cir 1979). 67 Texaco A/S Denmark 160 F3d 124. 68 Schefler v Livestock & Cas Ins Co 44 AD2d 811 (NY App Div 1974); Matter of Prudential Prop & Cas Ins Co 604 NYS2d 136 (App Div 1993); Matter of State Farm Mut Auto Ins Co (Callisto) 255 AD2d 876, 876–77 (NY App Div 1998). 69 The duty to act in good faith for the insured’s interests was referred to in Isadore Rosen & Sons Inc v Security Mutual Insurance Co 339 NYS2d 97. It therefore appears that a breach of this duty may prevent an insurer from relying upon its lack of consent to the policyholder’s settlement. Arguably, an unreasonable or arbitrary refusal to approve a settlement would constitute a breach of that duty: see above the cases cited in n 68. 70 The Appeals Condition (Article VI(E) of the 004 Form and Article V(f) of earlier Forms) deals with the situation where there is a possible appeal, but not with the claim in the trial court. See also ch 11, paras 11.23–11.26 below.

Settlements by the Policyholder

77

English professional liability insurance policies often include a clause71 that provides for the resolution of a dispute between policyholder and insurer as to whether the case should be fought. The Bermuda Form, however, includes no equivalent clause. Nevertheless, an insurer’s freedom of action is likely to be constrained in practice by the possibility that a refusal of consent to a settlement, followed by a substantial judgment against the policyholder, may lead to an action by the policyholder for bad faith based on the insurer’s wrongful refusal to settle,72 or for consequential damages resulting from a breach of the covenant of good faith and fair dealing.73 Under the Bermuda Form, the law of New York likely applies to these issues.

5.39

In England, in Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2),74 the Court of Appeal explored, under English law, a related issue in some detail. The court there addressed the question of whether a reinsurer can properly refuse consent to a reasonable settlement. The reinsured had settled an underlying claim without the reinsurer’s consent. The reinsurer defended the claim under a clause that provided: ‘No settlement and/or compromise shall be made and liability admitted without the prior approval of Reinsurers.’ The reinsurer contended that nothing constrained its decision to grant or withhold consent. The majority of the Court of Appeal rejected this argument and decided to imply a term that imposed some restraints. The court held that any withholding of approval by reinsurers should:

5.40

take place in good faith after consideration of and on the basis of the facts giving rise to the particular claim and not with reference to considerations wholly extraneous to the subject matter of the particular reinsurance.75

The court gave examples of situations where the refusal of consent would be based on reasons extraneous to the underlying claim; for example, attempting to influence an insurer’s attitude in relation to a matter arising under another quite separate reinsurance, or (perhaps) seeking to prolong claims for as long as possible, however obvious it might be that they would have to be met in full.76 Under New York law, these examples may

71 In particular, a ‘QC clause’ which binds the parties to an opinion of an English Queen’s Counsel (a rank conferred upon senior lawyers, generally barristers) as to whether a case should be fought: see Birds et al (eds), MacGillivray on Insurance Law, 14th edn (London, Sweet & Maxwell, 2018) para 30-103. 72 See, eg, Masters and Stanzler, Insurance Coverage Litigation, ss 11.10[A], 11.13[G][6]; Ostrager and Newman, Handbook on Insurance Coverage Disputes, s 12.04. Under New York law, an action for breach by the insurer of its duty of good faith in relation to the settlement of claims requires the policyholder to prove a ‘gross disregard’ of the insured’s interests – that is, a deliberate or reckless failure to place on an equal footing the interests of its insured with its own interests when considering a settlement offer: Pavia v State Farm Mut Auto Ins Co 82 NY2d 445 (NY 1993); Ansonia Assoc Ltd P’ship v Public Serv Mut Ins Co 257 AD2d 84, 86 (NY App Div 1999). As to the position of excess insurers, see the discussion in Masters and Stanzler, Insurance Coverage Litigation s 13.11; and Ostrager and Newman, Handbook on Insurance Coverage Disputes, ss 13.05 ff. An insurer in a layer above the policy may also complain that the underlying insurer did not protect the excess insurer’s interests. 73 Bi-Economy Mkt Inc v Harleysville Ins Co of NY 10 NY3d 187 (NY 2008); Panasia Estates Inc v Hudson Ins Co 856 NYS2d 513 (NY 2008). 74 Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2) [2001] EWCA Civ 1047, [2001] 2 All ER (Comm) 299. 75 Ibid, [67]. 76 Ibid, [68].

78

The Coverage Clause

qualify as unreasonable or arbitrary conduct which may preclude reliance on a consent provision.77

MITIGATION OF DAMAGES 5.41

A related question is whether the policyholder has mitigated its loss. The Bermuda Form does not contain a policy provision specifically dealing with mitigation of damages, but insurers sometimes criticise the policyholder for failing to mitigate its loss. Under New York law, the insurer bears the burden of proving that the policyholder could have lessened its damages.78 If this point is taken, the insurer must establish not only that the policyholder unreasonably failed to mitigate damages, but also that reasonable efforts would have reduced its damages.79 If the effort, risk, sacrifice or expense that must be incurred to avoid or minimise a loss is such that ‘under all the circumstances a reasonable man might well decline to incur it, a failure to do so imposes no disability against recovering full damages’.80

ALLOCATION OF PAYMENTS 5.42

Difficult questions arise where a policyholder makes payments in respect of claims that are both covered and uncovered.81 For example, the policyholder’s product may have caused personal injury to a particular individual both prior to and subsequent to the inception of the policy period. The former injury would be outside the coverage provided by the ‘Occurrence’ definition and specifically excluded by Exclusion IV(A) of the 004 Form. The latter injury would be a covered occurrence. Or a claim may be made against the policyholder on the basis of a number of theories of liability, some of which would be covered (for example, claims for negligence) and some of which would, if proved, be uncovered (for example, liability arising from dishonesty or other matters specified in Exclusion IV(P)). Or the policyholder may receive a number of claims, some of which may not be capable of being aggregated as an integrated occurrence.

77 See above n 68. See also Restatement Liability Insurance, s 24 and cmts thereto. 78 Air et Chaleur SA v Janeway 757 F2d 489, 494–95 (2d Cir 1985) (citing Jenkins v Etlinger 447 NYS2d 696, 698 (App Div 1982)); Cornell v TV Dev Corp 17 NYS2d 69, 74 (NY 1966). See discussion in Masters and Stanzler, Insurance Coverage Litigation, ch 10. 79 Katz Commic’ns Inc v Evening News Ass’n 705 F2d 20, 26 (2d Cir 1983) (applying New York law). 80 Slotkin v Citizens Cas Co of NY 614 F2d 301, 313 (2d Cir 1979). 81 In the absence of specific contractual provisions, this is to be distinguished from issues under New York law relating to allocation of defence costs. The New York Court of Appeals has refused to allocate defence costs at least with regard to liability policies that (unlike the Bermuda Form) include a duty to defend. The court in Continental Casualty Co v Rapid-American Co 593 NYS2d 966 (NY 1993) found that precluding the assignment of a share of defense costs to the policyholder under a CGL policy helped ensure that the policyholder received a ‘complete defense’ under the primary CGL policies at issue there (at 973). The rule stated in the Restatement Liability Insurance similarly does not support proration of defence costs in CGL policies: Restatement Liability Insurance, s 20 and specifically s 20(3) (the insurer selected to defend must provide a ‘full defense until the duty to defend is terminated’, subject to an insurer’s rights of contribution). For further discussion of allocation of defence costs, see below para 5.51.

Allocation of Payments

79

If the claims are the subject of judgment against the policyholder, then the jury verdict may enable a differentiation to be made between the covered and uncovered claims. For example, the jury may reject the case of dishonesty and return a verdict based upon negligence, or vice versa. An initial question arises, however, as to whether the decision of the jury is binding upon either the policyholder or the insurer in the subsequent arbitration of the claim under the Bermuda Form policy. Thus, if the jury has been persuaded that the policyholder’s conduct was deliberate or dishonest, and awards punitive damages in consequence, the insurer may say that the insurance claim is barred because of the ‘expected or intended’ provision or the ‘dishonesty’ exclusion. Can the policyholder then seek to persuade the tribunal that the jury’s verdict was perverse and that it was in fact guilty of nothing worse than negligence? Equally, if the jury verdict is favourable to the policyholder on these issues, can the insurer say that nevertheless the claim should fail because of ‘expected or intended’ or the dishonesty exclusion? In each case, there may be evidence available which supplements that which was put before the jury and which puts a different complexion on the facts.

5.43

In a case where a policyholder is seeking recovery for sums paid under a judgment, the claim is being made in respect of an actual liability that has been established. In some cases, an issue which is critical to the outcome of the policyholder’s claim may have been squarely before the jury. For example, a claim may have been made against the policyholder on the basis of both negligent and dishonest or fraudulent conduct. If the jury has rejected dishonesty and fraud, but upheld the claim in negligence, then the basis of the policyholder’s liability under the particular judgment would be negligence alone. In such circumstances, we doubt whether the insurer could seek to prove that the liability arose from fraud or dishonest conduct under Exclusion IV(P)(3) or (9), since the jury’s decision means that this was not the basis of the liability imposed on the policyholder. Equally, if the jury’s verdict was that there was dishonesty or fraud, and damages were awarded on that basis, then the policyholder’s liability for such damages would seem to be excluded. In many cases, however, the jury may not have needed to address a factual issue which is critical to the claim under the policy, or the questions before the jury may not coincide with the questions relevant to the determination of whether there is coverage under the policy. In such cases, there would need to be an investigation of the facts in order to determine whether coverage applies. Indeed, the Loss Payable Condition provides that: ‘Underwriters may examine the underlying facts giving rise to a judgment or settlement by the Insured to determine if, and to what extent, the basis for the Insured’s liability under such judgment or settlement is covered by this Policy.’82

5.44

Different problems arise if there is a settlement of the underlying claims. The settlement may take place before there is any jury verdict at all in any underlying case. Or the settlement may be prompted by a decision against the policyholder in a jury trial, but before judgment is entered. In such cases, there will be no jury verdict or judicial determination setting forth the basis of the policyholder’s liability. In some cases, the settlement

5.45

82 Article VI(F) of the 004 Form. This provision does not appear in the Loss Payable Condition in earlier Forms.

80

The Coverage Clause

will simply be an overall settlement of the claimant’s claims, without any differentiation between covered and uncovered claims. In other cases, the settlement may seek to anticipate and forestall potential arguments by the insurer – for example, by providing that the settlement sums are in respect of the negligence claims and that claims based upon dishonesty are withdrawn without payment. It is, perhaps, doubtful whether such a settlement would preclude an insurer from contending that the settlement was in fact a settlement of all the claims. Nevertheless, the question of how to allocate the settlement between the covered and uncovered claims would remain, as it would in a case where the settlement was by its express terms a settlement of all claims. 5.46

The Bermuda Form contains no express term that provides for the approach to be taken in these situations. In the 004 Form, but not the earlier versions, Article VI(D)(3) identifies the problem: If liabilities, losses, costs and/or expenses are in part covered by this Policy and in part not covered by this Policy, the Insured and Company shall use their best efforts to agree upon a fair and proper allocation thereof between covered and uncovered amounts, and the Insured shall cooperate with such efforts by providing all pertinent information with respect thereto.

5.47

Although this clause does not explain how to allocate,83 it does make clear that there should indeed be an allocation between covered and uncovered claims or amounts. Neither party could therefore contend that a situation of a covered and uncovered claim automatically produced the consequence that one or other party should bear the loss; that consequence should flow only if the facts meant that this was the ‘fair and proper’ allocation.

5.48

New York courts have, in a number of different contexts, addressed the question of covered and uncovered claims and allocation. Allocation issues arise both in relation to the allocation of sums paid in settlement and for defence costs. The starting point for all allocation issues is always the policy language,84 but the trend of the case law has been to treat these matters as raising questions of fact.85

5.49

Where there is a settlement of covered and uncovered claims, it could be argued that an ‘all or nothing’ approach should be taken. Under this approach, the issue would be: what was the efficient or proximate cause of the loss? The policyholder’s recovery might then depend upon the relative strength of the covered and uncovered claims. If the covered claims were of greater strength and merit than the uncovered claims, the policyholder would recover the full amount. If the covered claims were less strong and meritorious, the

83 Safeway Stores Inc v Nat’l Union Fire Ins Co of Pittsburgh Pa 64 F3d 1282, 1289 (9th Cir 1995). In Clifford Chance Limited Liability Partnership v Indian Harbor Insurance Co 836 NYS2d 484 (Sup Ct 2006), aff ’d, 838 NYS2d 62 (App Div 2007), the court considered a clause that not only provided for allocation, but that identified the basis upon which allocation was to be carried out. 84 Consolidated Edison Co of NY Inc v Allstate Ins Co 746 NYS2d 622, 628–29 (NY 2002). It is always important to consider whether the terms of the policies considered in the case law are sufficiently similar to the Bermuda Form wording. For example, the Bermuda Form coverage for ‘alleged’ personal injury or property damage may, particularly in the context of coverage for defence costs, result in coverage that is wider or different from that provided by the policies at issue. See, eg, Stonewall 73 F3d 1178. 85 See, eg, the summary of the cases in Pfizer Inc v Stryker Corp 385 F Supp 2d 380 (SDNY 2005).

Allocation of Payments

81

policyholder would not recover. If both sets of claims were equally strong and meritorious, then the loss would result from concurrent causes and there would be recovery: If a loss is proximately caused by an event covered by the policy, the insurer is liable. If the loss is caused by an event excluded from coverage, the insurer is not liable. When, however, a loss occurs through the concurrence of two independent events, one covered and one excluded, the loss will be covered.86

This approach was not, however, suggested by either party in PepsiCo v Continental Casualty Co.87 In that case, the question of allocation arose in the context of claims against covered and non-covered parties. PepsiCo and its directors and officers had settled a class action securities case. The policy covered the directors and officers, but not PepsiCo itself. PepsiCo nevertheless argued that the insurer should pay the full amount of the settlement, because the directors and officers were jointly and severally liable for the entire amount. The court held, however, that it was extremely unlikely that the d irectors and officers would have agreed to the settlement if they had been ‘expected to pay in toto’. The court therefore held that the settlement should be allocated according to the ‘relative exposures’ of the respective parties to the class action.88 Since the settlement had been concluded in good faith, the insurer bore the burden of proving that all or a portion of the total paid in settlement was excluded from the policy coverage. The court also held that the insurers had no liability for that ‘portion’ of the defence costs that were attributable to the defence of PepsiCo itself.

5.50

In Health-Chem Corp v National Union Fire Insurance Co of Pittsburgh,89 the court considered whether there should be an allocation of defence costs. It held that allocation was permitted ‘if factually possible’. The insurer was held to bear the burden of proving that all or a specific portion of the expenses had been incurred in the defence of a noncovered party. In Pfizer Inc v Stryker Corp,90 a lump sum settlement had been agreed by Pfizer with a group of claimants. Under its agreement with Stryker, Pfizer was entitled to an indemnity for claims arising from the use of a particular product sold after the date when a business had been sold to Stryker. Only some of the claimants had used the product after this date. The court held that the claimants’ own allocation of the lump sum settlement as between themselves was factual evidence relevant to the determination

5.51

86 American Nat’l Fire Ins Co v Mirasco Inc 249 F Supp 2d 303, 325 (SDNY 2003), vacated in part on other grounds sub nom Mirasco Inc v American Nat’l Fire Ins Co 144 F App’x 171, 173 (2d Cir 2005). See too Great N Ins Co v Dayco Corp 637 F Supp 765, 780 (SDNY 1986) (‘Where a policy expressly insures against direct loss and damage by one element but excludes loss or damage caused by another element, the coverage extends to the loss even though the excluded element is a contributory cause’). Cf Molycorp Inc v Aetna Cas & Sur Co 431 NYS2d 824 (App Div 1980) (‘where two causes lead to loss, one within and without coverage, the relevant inquiry is to determine which of the two was the dominant and efficient cause of the loss, generally a factual issue to be determined by the trier of the facts’). 87 PepsiCo v Continental Casualty 640 F Supp 656. 88 For a discussion of the factors that may be relevant in considering the relative exposures of parties and nonparties, see Safeway Stores Inc v National Union 64 F3d 1282; Ostrager and Newman, Handbook on Insurance Coverage Disputes, s 20.07 [a][2]. Allocation, of course, has long been an issue that US courts have specifically addressed under D & O liability policies, and some courts have declined to apply such cases outside the D & O policy context. 89 Health-Chem Corp v Nat’l Union Fire Ins Co of Pittsburgh, Pa 559 NYS2d 435 (Sup Ct 1990). 90 Pfizer Inc 385 F Supp 2d 380.

82

The Coverage Clause

of the amount that Pfizer could recover. In relation to defence costs, the court noted that there was little authority on the allocation of covered and non-covered expenses, and held that the party seeking indemnity had the initial burden to establish that an expense was incurred in the defence of a covered claim. These decisions were applied in High Point Design v LM Insurance Corp.91 The policyholder together with three other unrelated companies had incurred defence costs. The parties were agreed that there was no coverage for costs solely incurred in relation to the non-covered parties. They disagreed as to whether costs, which were for the benefit of all parties, were covered in full (as the policyholder contended) or were to be allocated pro rata between the four parties (as the insurer contended). The court reviewed the prior case law and upheld the policyholder’s argument. The insurer was therefore liable to the policyholder for the fees and expenses incurred in defending the underlying action, less any additional expenses that would not have been incurred but for the presence of the non-covered defendants. Once the policyholder established ‘prima facie proof’ that a particular expense was incurred in the underlying defense, the insurer had the burden to prove that it was an additional expense which should be allocated and thus exempted. In Clifford Chance Limited Liability Partnership v Indian Harbor Insurance Co,92 the court stated that insurers were entitled to ‘allocate loss, or settlement costs, between covered and non-covered claims, or parties, where there is a factual basis for the allocation’. It also noted that little case law existed in New York outlining the factors and methods to be employed in determining a proper allocation. In that case, however, the policy contained an express provision that allocation was to be approached on a relative exposure basis. 5.52

The question of allocation has arisen in New York in contexts other than allocation of a global settlement of covered and uncovered claims. In Consolidated Edison Co of New York Inc v Allstate Insurance Co,93 the court considered CGL insurance coverage for the policyholder’s liability for environmental contamination of a site. Continuous harm had taken place over a lengthy period, and it was impossible to determine the extent of the property damage that resulted from an occurrence during any particular policy period. The court rejected the policyholder’s argument that each available policy should respond in full to the claim. Instead, interpreting the specific language at issue in that case, it adopted a pro rata approach, which essentially allocated a proportion of the loss to each of the available policies under the non-standard policy language at issue in that case. This acknowledged ‘the fact that there is uncertainty as to what actually transpired during any particular policy period’. The court prorated liability based on the amount of time the policy was in effect in comparison to the overall duration of the damage: a ‘time on the risk’ approach. The New York Court of Appeals in Consolidated Edison contrasted the situation there with its earlier decision in York-Buffalo Motor Express v National Fire & Marine Insurance Co,94 finding that, unlike York-Buffalo, which involved ‘concurrent insurance’

91 High Point Design LLC v LM Ins Corp No 14-cv-787, 2016 WL 426594 (SDNY 3 Feb 2016). See also Murnane Bldg Contractors Inc v Zurich Am Ins Co 966 NYS2d 486 (App Div 2013). 92 Clifford Chance Limited Liability Partnership v Indian Harbor Insurance Co 836 NYS2d 484. 93 Consolidated Edison Co of New York Inc v Allstate Insurance Co 746 NYS2d 622 (NY 2002) (Consolidated Edison). 94 York-Buffalo Motor Express v National Fire & Marine Ins Co 294 NY 467 (NY 1945).

Allocation of Payments

83

for an automobile accident,95 the situation in Consolidated Edison ‘involve[d] successive insurers and a continuous harm’.96 The court considered that, whilst allocation was not explicitly mandated by the policies, it was ‘consistent with the language of the policies’. It also recognised that there were different ways to prorate liability among successive policies, and that courts had differed on how to treat self-insured retentions, periods of no insurance, periods where no insurance was available and settled policies. Therefore, the court concluded that its decision in Consolidated Edison ‘is not the last word on proration’ in New York.97 The US Court of Appeals for the Second Circuit applied pro rata allocation under the standard CGL occurrence-based policy language with regard to the policyholder’s liability for progressive personal injury caused by asbestos in Stonewall Insurance Co v Asbestos Claims Management Corp.98 The court there required the policyholder to bear a proportionate share of the loss in relation to certain periods where it decided to ‘self insure’.99 In Uniroyal Inc v The Home Insurance Co,100 the court also took a pro rata approach to a continuous occurrence (comprising deliveries of Agent Orange defoliant) that spanned two policy periods, again under standard CGL policy language.

5.53

In In re Viking Pump,101 the New York Court of Appeals emphasised the importance of the policy language in all policy disputes and specifically in determining issues of allocation. The policies in that case contained ‘Prior Insurance and Non-Cumulation of Liability’ provisions,102 which the court considered to be inconsistent with pro rata application. The court therefore applied its interpretation of the policy language in adopting a different approach to that taken in the Consolidated Edison case. It applied an ‘all sums’ or ‘joint and several’ approach which entitled the policyholder to recover in full under the policies, in respect of its total asbestos liabilities, since the policies were in effect during

5.54

95 As explained in Consolidated Edison at 776 NYS2d at 629. 96 Ibid. 97 Ibid at 630. 98 Stonewall 73 F3d 1178. 99 Ibid at 1203–04. Adopting the ‘unavailability exception’, the court there refused to assign liability to the policyholder for periods in which the policyholder had been unable to purchase CGL insurance for its asbestos related exposures (ibid). 100 Uniroyal Inc v The Home Insurance Co 707 F Supp at 1391–94. The decision that there was a single occurrence, giving rise to an allocation issue, was doubted in International Flavors & Fragrances Inc v Royal Insurance Co of America 844 NYS2d 257 (App Div 2007). 101 In re Viking Pump 27 NY3d 244 (NY 2016). See too Keyspan Gas E Corp v Munich Reinsurance Am Inc 31 NY3d 51 (NY 2018); In re Liquidation of Midland Ins Co 171 AD3d 564, 564 (NY App Div 2019) (following Viking Pump). 102 For example, ‘Prior Insurance and Non-cumulation of Liability’ clauses in certain excess policies at issue, referred to as ‘Condition C’, states as follows: It is agreed that if any loss covered hereunder is also covered in whole or in part under any other excess Policy issued to the Insured prior to the inception date hereof[,] the limit of liability hereon … shall be reduced by any amounts due to the Insured on account of such loss under such prior insurance. Subject to the foregoing paragraph and to all the other terms and conditions of this Policy in the event that personal injury or property damage arising out of an occurrence covered hereunder is continuing at the time of termination of this Policy the Company will continue to protect the Insured for liability in respect of such personal injury or property damage without payment of additional premium. Quoted from Viking Pump 27 NY3d at 252–53.

84

The Coverage Clause

the periods when the damage occurred. The clauses in question had two aspects. First, they provided for the reduction of amounts payable in the event that a covered loss was also covered by a prior policy. The 004 Form contains no equivalent wording. We consider that a pro rata allocation would likely be applied in the case of continuing injury or damage which takes place both prior to and during the policy period. The former (injury or damage prior to the policy period) is excluded under Article IV(a), but the latter (injury or damage during the policy period) is covered. Second, the clauses provided for continued coverage where an occurrence was continuing at the time of termination of the coverage. In this respect, the policies have a similarity with the Bermuda Form provisions which entitle and require the policyholder to aggregate injuries and damages ‘irrespective of the period (without limiting the effect of Exclusion IV.A) or area over which the actual or alleged Personal Injuries, Property Damages … occur”.103 Accordingly, pro rata allocation is not in our view appropriate in respect of injury or damage taking place both during and after the termination of a Bermuda Form policy.

PERSONAL INJURY, PROPERTY DAMAGE AND ADVERTISING LIABILITY 5.55

Under the coverage clause, the liability of the policyholder must arise from personal injury, property damage or advertising liability. Each of these is a defined term.

PERSONAL INJURY 5.56

Personal injury is broadly defined so as to mean ‘Bodily Injury, mental injury, mental anguish, shock, sickness, disease, disability, false arrest, false imprisonment, wrongful eviction, detention, malicious prosecution, discrimination, humiliation, and libel, slander or defamation of character of invasion of rights of privacy’.104 It therefore includes not only bodily injury105 but also other matters that would not normally be thought of as personal injury.

5.57

Very often, the question of whether personal injury exists will not be difficult to answer. Difficulties can arise, however, in the event that a product causes a progressive injury or illness taking place over a period of years. New York law applies an ‘injury-in-fact’ standard to the question of whether coverage is triggered under CGL occurrence policies, which of course are triggered by ‘bodily injury’ or ‘property damage’ that takes

103 Article III(V)(2), discussed in ch 6 below, paras 6.37–6.46. 104 Article III(W) of the 004 Form; Article III(b) of earlier Forms. 105 ‘Bodily Injury’ is a defined term in the 004 Form (Article III(E)), but not in earlier versions. It is defined as ‘physical injury to the body of a person including death at any time resulting therefrom’. It is relevant to distinguish between ‘bodily injury’ and ‘personal injury’ when considering whether there has been property damage under the Form, since property damage includes losses consequent upon evacuation arising from actual or threatened ‘bodily injury’.

Personal Injury

85

place during the policy period.106 Thus, in American Home Products Corp v Liberty Mutual Insurance Co,107 the court held that a real but undiscovered injury, proved in retrospect to have existed at the relevant time, would trigger coverage, irrespective of the time the injury became diagnosable. Since the ‘injury-in-fact’ trigger of coverage is now firmly established under New York law, this is the appropriate test to apply under the Bermuda Form when a tribunal is considering whether and when an actual injury occurred. In cases of progressive illness, this injurious process can be a continuous one so that there can be injuries in fact occurring over a lengthy period, thereby potentially triggering policies covering different periods.108 The question of whether there is an injury in fact will likely depend upon medical or other expert evidence as to the development of the relevant disease.109 Indeed, Stonewall Insurance Co v Asbestos Claims Management Corp110 involved appeals in separate cases where the judge in one case, and the jury in the other, had considered the same disease and come to different conclusions on ‘injury in fact’.111 However, since the Bermuda Form coverage is also concerned with ‘alleged’ injuries,112 a tribunal may need to consider not only any medical evidence as to whether or when an actual injury occurred, but also the underlying plaintiff’s allegations on these issues.

106 See, eg, Masters and Stanzler Insurance Coverage Litigation, ch 4. 107 American Home Prods Corp v Liberty Mut Ins Co 565 F Supp 1485 (SDNY 1983), aff ’d as modified, 748 F2d 760, 764 (2d Cir 1984). See also Cont’l Cas v Rapid-Am 593 NYS2d 966 (NY 1993). See also for discussion of injury-in-fact trigger and proving injury: Stonewall 73 F3d 1178; Campbell v Met Prop & Cas Ins Co No 98 Civ 5328 NRB, 2000 WL 297174 (SDNY 21 Mar 2000), aff ’d in part, rev’d in part on grounds, 239 F3d 179 (2d Cir 2001); Uniroyal 707 F Supp 1368; Burroughs Wellcome 632 F Supp 1213; Continenta’l Cas Co v Employers Ins Co of Wausau 871 NYS2d 48, 62–63 (App Div 2008), leave to appeal denied, 13 NY2d 710 (2009); rev’d on other grounds, 85 AD3d 403 (NY APP Div 2011); Maxum Indem Co v A-One Testing Labs Inc 150 F Supp 3d 278 (SDNY 2015); Hanover Ins Co v Vermont Mut Ins Co 69 F Supp 3d 302 (NDNY 2014); Viking Pump Inc v Century Indem Co No 10C-06-141, 2013 WL 7098824 (Del 31 October 2013 (applying New York law); In re Viking Pump Inc 148 A3d 633 (Del 2016) (applying New York law); Hopeman Bro Inc v Continental Cas Co 307 F Supp 3d 433, 466–69 (ED Va. 2018) (applying New York law); Carrier Corp v Allstate Ins Co 63 Misc 3d 1212(A), 2018 NY Slip Op 51965(U) (Sup Ct 21 November 2018), aff ’d as modified, 133 NYS3d 697 (App Div 2020); Danaher Corp v Travelers Indem Co 414 F Supp 3d 436, 456–57 (SDNY 2019). See also Ostrager and Newman, Handbook on Insurance Coverage Disputes, s 9.03 [b][4]. 108 Stonewall 73 F3d at 1197; US Liab Ins Co v Farley 626 NYS2d 238 (App Div 1995); Fulton Boiler Works Inc v Am Motorists Ins Co 828 F Supp 2d 481, 489 (NDNY 2011). The New York Court of Appeals has yet to rule on trigger of coverage. 109 Hopeman Brothers 307 F Supp 3d at 466–69 (applying New York law). Compare Carrier Corp 63 Misc 3d 1212(A), where the court did not require a trial of the issue of when asbestos caused injuries in fact, holding that subclinical damage was sufficient and that injuries occurred from the time of first exposure until either assertion of a claim for damages or death. Both of these cases are relevant in the context of arguments, sometimes advanced by insurers, based upon the decision of the Appellate Division in Continental Casualty Co v Employers Insurance Co of Wausau 871 NYS2d 48, often known as ‘Keasbey’. The essence of these arguments is that Keasbey is controlling New York law as to when an injury in fact occurs in the context of asbestos injuries and that subclinical changes cannot amount to such injury. In Hopeman, the court held that issues as to when injury in fact occurs raised factual questions for determination at trial. In Carrier Corporation, the court considered that this issue did not require further evidence, since there was sufficient precedent and medical evidence in the case to establish that injuries in fact occurred from the time of first exposure. On this issue and the decision in Keasbey, see also Viking Pump v Century Indem2013 WL 7098824 (applying New York law); In re Viking Pump 148 A3d 633 (applying New York law). 110 73 F3d 1178. 111 Ibid at 1197–200. 112 See below ch 6.

86

The Coverage Clause

PROPERTY DAMAGE 5.58

The 004 Form defines the term ‘Property Damage’113 as follows: (1) physical damage to or destruction of tangible property including the loss of use thereof at any time resulting therefrom; (2) loss of use of tangible property which has not been physically damaged or destroyed arising from physical damage to or destruction of other tangible property; or (3) losses consequent upon evacuation arising from actual or threatened ‘Bodily Injury’ to or destruction of tangible property.

This definition is very similar to the definition in earlier Forms114 and has its roots in the definition of ‘property damage’ first used in the 1973 CGL Form. Typically, a policyholder’s claim will concern subparagraph (1) of the Property Damage definition – that is, physical damage to or destruction of tangible property.115 5.59

In many cases, it will be obvious that physical damage to tangible property has taken place, and when. The US case law shows, however, that difficult questions can arise as to whether ‘property damage’ has taken place; whether that property damage can properly be described as ‘physical injury (or damage) to tangible property’ or ‘loss of use’; and, if so, when that damage took place. The timing of ‘property damage’ may be particularly important, and policyholders in many cases have sought to show that their policies have been triggered, and insurers have sought to show either that no policies have been triggered or that policies issued by insurers in other time periods are the policies that have to respond. For example, the issue was critical in the Eljer litigation116 (discussed further below), where the policyholder, Eljer (formerly called US Brass), was anxious to trigger its old occurrence policies, since it had been unable to buy coverage for the claims in question in later years. Eljer faced massive liability for property damage to claimants’ houses from plumbing leaks arising from the manufacture and design of a faulty plumbing system. If Eljer could show that ‘property damage’ began at the time of installation of the defective system, its old occurrence policies would be triggered. But if the relevant damage took

113 Article III(AA) of the 004 Form; see Article III(c) of earlier Forms. 114 Article III(c) of earlier Forms. One change is that the 004 Form uses the expression ‘physical damage’ instead of ‘physical injury’. 115 For an illustration of a situation coming within sub-para (2), see the example given by Judge Posner in Eljer Manufacturing Inc v Liberty Mutual Insurance Co 972 F2d 805, 810 (7th Cir 1992) (a defective crane collapses and blocks access to a claimant’s restaurant business). 116 See the judgment of the US Court of Appeals for the Seventh Circuit in Eljer v Liberty Mutual 972 F2d 805 (applying the laws of both New York and Illinois, which were agreed by the parties to be the same) (Posner J). The case involved the federal appellate court ‘predicting’ how, in relation to certain of the policies in issue, the state Supreme Court of Illinois, the arbiter of Illinois law, would decide the issue in question under Illinois law. The Supreme Court of Illinois later decided the same issue in another case involving Eljer’s insurance and came to a different conclusion from that of Judge Posner with regard to the post-1982 policies; this was based on the Illinois court’s conclusion on Illinois law, which governed the policyholder’s policies on the risk after 1982. See Travelers Ins Co v Eljer Mfg Inc 757 NE2d 481 (Ill 2000). Under the New York Court of Appeals’ decisions in Thomas J Lipton 357 NYS2d 705 and Sturges v Utica Mutual 371 NYS2d 444 (see below para 5.62), incorporation of a defective product into a larger entity or product constitutes ‘physical injury’ and thus property damage in the pre-1982 CGL policies. Illinois law, in contrast, as determined in Travelers v Eljer, does not consider incorporation of a defective product, without more, to constitute ‘physical injury’.

Property Damage

87

place only at the time of a leak, the liability of the insurers with policies during the early years of the claim would be minimised. The debate therefore in some respects paralleled the well-known time of exposure/manifestation debate that arose in relation to asbestos bodily injury and property damage claims. With regard to Eljer, the US Court of Appeals for the Seventh Circuit took one view, while the Illinois Supreme Court took another in a later case involving the same claims. The result was that some occurrence insurers (those whose policies were governed by New York law) were held liable for claims where ‘physical damage’ was found to have taken place prior to manifestation as shown by a leak, but many others were not (those whose policies were governed by Illinois law).117 For the purposes of the Bermuda Form, in a case that is not straightforward, it may be necessary to answer the same two questions as are posed in many of the US cases on the CGL Form – that is, (i) has ‘property damage’ within the meaning of the policy taken place, and (ii) if so, when? The starting point in answering these questions is again the decision in American Home Products,118 which applies an ‘injury-in-fact’ test to the question of whether there has been property damage or injury. The injury-in-fact approach was also applied in the context of asbestos property damage in Maryland Casualty Co v WR Grace & Co,119 where the court rejected arguments that there should be a distinction, in this respect, between asbestos property-damage and asbestos bodily-injury claims.

5.60

A related question is whether the injury in fact is in fact ‘physical injury to or destruction of tangible property’ (the definition in Forms 001–003) or ‘physical damage to or destruction of tangible property’ (the definition in Form 004). The pre-1973 CGL occurrence policy wordings had defined property damage as ‘injury to or destruction of tangible property’. The definition was altered in 1973, adding ‘physical’, to define ‘property damage’ in part as ‘physical injury to tangible property’.120 Again, q uestions may arise, on particular facts, as to whether damage is ‘physical’. In Maryland Casualty v WR Grace, the US Court of Appeals for the Second Circuit, applying New York law, held that ‘physical’ damage to buildings took place when asbestos products were installed therein.121 In the Seventh Circuit decision in Eljer, the US Court of Appeals for the Seventh Circuit, applying New York law, concluded that ‘physical damage to tangible property’ took place when defective plumbing systems were installed in buildings.122 The Illinois Supreme Court, on identical facts, reached a different conclusion under Illinois law, holding that physical damage only took place at the time of the leak.123

5.61

117 The English Court of Appeal has considered a similar issue concerning ‘physical damage’ in relation to a policy governed by English law. Relevant US cases, including the Eljer decisions, were cited. The court preferred the views of the Supreme Court of Illinois, applying Illinois law, to those of the Seventh Circuit, applying New York law: Pilkington UK Ltd v CGU Insurance PLC [2004] EWCA Civ 23, [2004] Lloyd’s Rep IR 891. 118 American Home Products Corp v Liberty Mutual Insurance Co 748 F2d 760. See also the authorities cited above n 107 and the discussion in para 5.57 above. 119 Maryland Cas Co v WR Grace & Co 23 F3d 617 (2d Cir 1994). 120 Eljer v Liberty Mutual 972 F2d 805; Travelers v Eljer 757 NE2d 481. 121 23 F3d 617. See also Stonewall 73 F3d 1178. 122 972 F2d at 814. 123 757 NE2d at 502. See further: Wyoming Sawmills Inc v Transportation Ins Co 578 P2d 1253 (Or 1978); Esicorp Inc v Liberty Mut Ins Co 266 F3d 859 (8th Cir 2001); American Home v Libbey-Owens-Ford 786 F2d at 25; Wisconsin Label Corp v Northbrook Prop & Cas Ins Co 607 NW2d 276 (Wis 2000); Amtrol Inc v Tudor Ins Co No 01–10461–DPW, 2002WL 31194863 (D Mass 2002); Scott C Turner, ‘Insurance Coverage for Incorporation of Defective Construction Work or Products’ 18 APR Construction Law 29.

88

The Coverage Clause

5.62

In answering these various questions, it may be relevant to consider another principle of New York insurance law, known as the incorporation doctrine. This principle has been used to determine that ‘property damage’ or ‘physical’ injury begins when an allegedly defective product is incorporated into other property. In Sturges Manufacturing Co v Utica Mutual Insurance Co,124 a manufacturer of defective ski straps sought to recover for liability to the manufacturers of the ski-bindings into which the straps had been incorporated. The policy used the pre-1973 CGL policy language, which defined ‘property damage’ as ‘injury to or destruction of tangible property’.125 The policyholder argued that its strap, as a defective component of the binding, diminished the value of the unit as a whole, and that such damage constituted ‘property damage’, which was covered by the policy. The New York Court of Appeals held that this contention might be sustainable on the facts, holding that ‘if, as a result of defective straps, the value of the bindings were reduced beyond the cost or strap replacements, then that differential was harm not to the straps but to the bindings themselves’. Accordingly, if a defective product is incorporated into a greater whole, there is property damage to the whole if it has been diminished in value by an amount greater than the value of the defective product itself. The US Court of Appeals for the Second Circuit applied this decision in Maryland Casualty Co v WR Grace & Co,126 holding that ‘damage in fact occurs upon installation in buildings of products containing asbestos’. Both of these decisions address the pre-1973 policy wording, and the question may arise under the Bermuda Form as to whether, on a particular set of facts, the damage which takes place on incorporation is ‘physical’ injury or damage, and whether this concept is actually different from ‘injury’ in the pre-1973 form.

5.63

Many of the relevant cases in this area are considered in the decision of the Delaware Supreme Court in Hoechst Celanese Corp v Certain Underwriters at Lloyd’s London.127 The case also involved claims for liability arising out of the same plumbing system that gave rise to the Eljer litigation. Hoechst Celanese supplied the raw material plastic that Eljer used to make its defective plumbing fittings. The question arose as to whether the policyholder could recover against its occurrence insurers, who were on the risk until early 1986, on the basis that property damage began at installation.128 The occurrence insurers argued for a leak trigger, which would greatly minimise their liability because most leaks took place after 1985. The trial judge, on a summary judgment application, upheld the insurers’ argument. On appeal, the Delaware Supreme Court reversed the lower court’s ‘leak only’ trigger, holding that the timing of ‘property damage’, defined as ‘physical injury to tangible property’, presents a question of fact under New York law’s injury-in-fact trigger. The court concluded that ‘in this case, property damage sufficient to trigger insurance coverage may occur as early as installation of plumbing systems into houses’, and remanded the case for a jury determination as to when an ‘injury in fact’ took place.129 The jury later held that injury in fact began at installation and continued

124 Sturges

v Utica Mutual 371 NYS2d 444. See also Thomas J Lipton 357 NYS2d 705. 1973 definition also continued to define ‘loss of use’ as ‘property damage’. See above para 5.58. 126 Maryland Casualty v WR Grace 23 F3d 617. 127 Hoechst Celanese Corp v Certain Underwriters at Lloyd’s London 673 A2d 164 (Del 1996). 128 The policyholder’s coverage changed from occurrence based coverage to claims made coverage in 1986. 129 673 A2d at 170. 125 The

Advertising Liability

89

thereafter until the damage manifested itself at the time of leak, presumably on the basis that there was a progressive breakdown of the plastic fitting which ultimately led to the leak. Other cases decided under New York law have concluded as a matter of law that ‘physical injury’, and thus ‘property damage’, begins at the time a defective product is incorporated into a larger whole.130 The case law shows that, as with personal injury, there may be a continuous injurious process so that property damage occurs over a lengthy period, thereby potentially triggering policies covering different periods. Maryland Casualty v WR Grace131 and Stonewall132 both concerned the installation of asbestos in buildings. The Second Circuit held that the property damage to buildings occurred on installation. Even though asbestos fibres may have escaped continuously after installation, this did not mean that there was any incremental or continuous property damage. The position is different in environmental pollution cases where contamination may continue to increase or spread, whether or not the contamination is based on active pollution or the passive migration of contamination into the soil and groundwater.133 The existence and timing of injury in fact in any particular case therefore raises questions of fact dependent upon the particular circumstances. Again, a tribunal may need to consider not only the actual facts as to whether or when property damage occurred, but also the underlying plaintiff’s allegations on these issues.

5.64

ADVERTISING LIABILITY The Bermuda Form defines ‘Advertising Liability’ as damages on account of various matters, such as libel, slander or defamation, ‘committed or alleged to have been committed in any advertisement, publicity article, broadcast or telecast and arising out of the Insured’s advertising activities’.134 The full text of the definition is as follows:

5.65

‘Advertising Liability’ means liability for Damages on account of: (1) (2) (3) (4)

libel, slander or defamation, any infringement of copyright or of a title or of slogan, piracy or misappropriation of ideas under an implied contract, or any invasion of right of privacy,

committed or alleged to have been committed in any advertisement, publicity article, broadcast or telecast and arising out of the Insured’s advertising activities.

This coverage is becoming increasingly important with the expanding reliance on internet communications and the focus on intellectual property and privacy issues. There has been extensive litigation in US courts on issues under similar provisions in the CGL Form.135

130 Sturges v Utica Mutual 371 NYS2d 444; Thomas J Lipton 357 NYS2d 705. See above para 5.62. 131 23 F3d 617. 132 73 F3d 1178. See also MRI Broadway Rental Inc v US Mineral Prods Co 92 NY2d 421 (NY 1998). 133 Olin Corp v Certain Underwriters at Lloyd’s London 468 F3d 120 (2d Cir 2006); Olin Corp v Certain Underwriters at Lloyd’s London 347 F App’x 622 (2d Cir 2009); Olin Corp v American Home Assur Co 704 F3d 89 (2d Cir 2012) (all applying New York law). 134 Article III(A) of the 004 Form; Article III(d) of earlier Forms. 135 See eg, Masters and Stanzler, Insurance Coverage Litigation, ch 16.

5.66

90

The Coverage Clause

OTHER ASPECTS OF THE COVERAGE CLAUSE 5.67

The coverage clause brings into play other concepts that are discussed in more detail elsewhere. Under the 004 Form, personal injury, property damage or advertising liability must be ‘encompassed by an Occurrence’, and earlier Forms use the expression ‘resulting from’ an occurrence. The occurrence definition is discussed in detail in Chapter 6. The clause also imposes the requirement for reporting of the occurrence during the period of Coverage A or during the discovery period (Coverage B). The reporting requirement is considered in Chapter 8. The distinction between Coverage A and Coverage B is discussed in Chapter 2 and also, in the context of specific clauses, in Chapter 11. Article 1 of Forms 001–003 contained concluding words that identified the terms which would apply to any particular occurrence, namely those applicable at the time notice in respect of the relevant occurrence was first given. This is an important feature of the Form. The equivalent provision, now somewhat expanded, appears in Article II of the 004 Form. It is discussed in that context in Chapter 9 below.

6 The Definition of ‘Occurrence’ INTRODUCTION One of the distinctive features of the Bermuda Form is that it is an ‘occurrence reported’ form. This feature refers to the mechanism by which coverage under the Form is activated or, in the terminology used in coverage disputes in the United States, ‘triggered’. The Bermuda Form’s occurrence reported trigger is a hybrid of two earlier triggers, or types, of general liability insurance: the ‘occurrence trigger’, for which coverage is activated when some event, usually ‘bodily injury’ or ‘property damage’, ‘occurred’ during the policy period;1 and the ‘claims made’ trigger, under which coverage arises when a claim or potential claim is first made during the policy period. As the successor to the occurrence policies in use during the second half of the twentieth century, the Bermuda Form inevitably inherits some of the historical baggage in its approach to ‘occurrence’. It both adopts and departs from some of the language used by those policies (in particular the CGL form). And in various respects it is best understood as a reaction – partly positive and partly negative – to the doctrines and principles that were established or fought over in the context of those occurrence policies.

6.01

The occurrence definition, now found in Article III(R) and (V) of Form 004, has undergone considerable refinement since the Bermuda Form was first drafted. This chapter focuses on the current version of the Bermuda Form.2 One marked tendency has been to distinguish more carefully, with each revision, between different situations likely to arise in practice and different functions to be performed by the policy terms, and to address each more specifically. Sometimes, however, one will encounter variations ‘in the wild’.

6.02

1 Standard ‘Comprehensive’ or later ‘Commercial’ General Liability (CGL) policies typically include several coverages. Occurrence coverage for bodily injury and property damage liability is triggered when bodily injury or property damage takes places during the policy period. The ‘occurrence’, which is the cause of the injury or damage, need not take place during the policy period and, in fact, as shown by legion cases on trigger of coverage litigated in the United States, may take place years before the bodily injury or property damage manifests or is discovered. As discussed in this chapter, unlike the standard CGL policy, the Bermuda Form uses notice in part as a trigger of coverage and thus does not maintain the same distinction between the triggering event (bodily injury or property damage) and the cause of the injury or damage (the ‘occurrence’). Standard CGL policies typically include other coverages, including Advertising Injury and Personal Injury Coverage, and Contractual Liability Coverage. Each of these coverages is triggered by some event other than bodily injury or property damage. For a discussion of those coverages, their triggers and issues arising under them in standard CGL policies historically used in the United States, see Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) chs 16 and 18. 2 A more detailed consideration of the earlier versions can be found in the 1st edition of this book, at paras 6.34–6.41.

92

The Definition of ‘Occurrence’

Many insurers, beyond ACE (now Chubb) and XL (now Axa XL), have adopted the Bermuda Form. They have sometimes used old versions and have sometimes made ad hoc adaptations of their own. So it is always important to consider the particular wording that has been used. 6.03

One of the difficulties with the ‘Occurrence’ definition, however, is that even in its revised form, it fulfils so many functions. It fixes the temporal limit of the coverage, especially the extent to which the policy will respond to problems from the past. It contains provisions that aggregate related losses for the purpose of applying limits and retentions. And it defines the type of losses to which this policy – as a form of liability policy – responds, that is to say, the ‘insured perils’. It touches upon (and in earlier policy forms, attempted comprehensively to address) issues of fortuity and intentional harm. This multiplicity of functions inevitably makes for complexity. It is often important to ask of a particular provision in the ‘Occurrence’ definition why that provision is there.3

6.04

Still greater complexity is added by the multiplicity of different factual circumstances that may give rise to claims against a policyholder and that the ‘Occurrence’ definition seeks to address. It must deal with cases where the policyholder has caused real injury, with cases where there is injury (but not, in truth, caused by the policyholder) and with cases where there is no real injury at all (but injury is alleged). It must deal with cases where the injury takes place at one moment, with cases where it develops over a prolonged period and with cases where it is hard to tell when the injury started. It must deal with cases where courts stretch or expand the traditional notions of causation to impose liability under novel theories such as ‘market share’ liability,4 or by imposing joint and several liability where causation is in doubt.5 It must often deal with cases where the facts are in doubt or are never established. It must deal with cases where liability results from a single distinct event (such as an explosion) and with cases where liability arises as a result of numerous events taking place over a prolonged period of time (such as typical product liability cases). It must deal with cases where the plaintiff in the liability case claims compensation not for an injury that has actually happened, but for the increased risk that injury might occur.

6.05

Earlier attempts at defining ‘occurrences’ or ‘accidents’ tended to assume a simpler world. It seems relatively easy to define an occurrence for the conventional tort, with a clearly injured plaintiff and a simple physical causal link between that plaintiff’s injury and some act by the defendant. In the real world of mass torts in the United States, however, the link is often indirect or obscure, causation is very likely to be controversial, and the very fact of injury may be disputable.6 It is in this real, and occasionally surreal, world that the purchasers of Bermuda Form coverage do their business and incur their liabilities, and it is with this real world that the occurrence definition has to deal. 3 Of course, one should not make the mistake of assuming that everything that is an ‘occurrence’ is necessarily going to entitle the policyholder to indemnity. Other provisions of the policy – such as the exclusions – may preclude or restrict coverage even where there is an occurrence. 4 See, eg, Sindell v Abbott Labs 607 P2d 924 (Cal 1980) (diethylstilbestrol (DES)). 5 An approach now taken in England; see, eg, Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22, [2003] 1 AC 32, HL (asbestos). 6 For instance, although many plaintiffs in the breast implant litigation had recognised autoimmune diseases, some were allegedly suffering from nebulous ‘syndromes’ whose very existence is questionable.

‘Occurrence’: The Basic Definition

93

‘OCCURRENCE’: THE BASIC DEFINITION The basic definition of ‘Occurrence’ is found in Article III(V)(1). This is the definition on which all the later parts of the Form build:

6.06

(1) An ‘Occurrence’ exists if, and only if: (a) except with respect to actual or alleged Personal Injury or Property Damage arising from the Insured’s Products, there is an event or continuous, intermittent or repeated exposure to conditions which event or conditions commence on or subsequent to the Inception Date, or the Retroactive Coverage Date, if applicable, and before the Termination Date of Coverage A, and which cause actual or alleged Personal Injury, Property Damage or Advertising Liability; (b) actual or alleged Personal Injury to any individual person, or actual or alleged Property Damage to any specific property, arising from the Insured’s Products takes place on or subsequent to the Inception Date, or the Retroactive Coverage Date, if applicable, and before the Termination Date of Coverage A.

This part of the definition is concerned primarily with two questions. First, it is defined in terms of the peril against which coverage is provided, and assists in defining the coverage for that peril: the risk that the insured will be held liable for ‘actual or alleged’ personal injury, property damage or advertising liability. Second, the provision defines the chronological limits of an occurrence. It does so using two distinct and mutually exclusive definitions: (i) one that applies to anything other than product liability claims (subparagraph (a): the ‘first limb’ of the ‘Occurrence’ definition); and (ii) one that applies only to product liability claims (sub-paragraph (b): the ‘second limb’ of the ‘Occurrence’ definition). The two limbs have a somewhat similar structure; at the heart of each is a schematic structure where X causes Y. In limb one, X is an ‘event’ or ‘exposure to conditions’. In limb two, X is one of the insured’s products. In limb one, Y is actual or alleged personal injury, property damage or advertising liability. In limb two, Y is actual or alleged personal injury or property damage. The two limbs differ, therefore, principally in what causative event they treat as relevant and in how they define the chronological ambit of the policy: limb one defines it by reference to the time when X, the cause, takes place, whereas limb two defines it by reference to when Y, the consequence, takes place.

6.07

The Common Factors: Personal Injury, Property Damage and Advertising Liability One thing the two limbs share is the need to establish the existence of actual or alleged personal injury, property damage or (in the case of limb one) advertising liability. Each is a defined term. The relevant definitions are discussed in Chapter 5.

6.08

The Causative Factor: Events and Products The second limb of the basic ‘Occurrence’ definition, dealing with product liability, talks simply of injury or damage ‘arising from the Insured’s Products’. It includes no qualification as to how the injury should arise. The injury may result from a defect in the design

6.09

94

The Definition of ‘Occurrence’

or manufacture of the product, or a deliberate or known feature of the product. There is no doubt that injury ‘arises from’ the use of a gun and ammunition, and if a gun is used to injure somebody, then the second limb of the ‘Occurrence’ definition is undoubtedly met. It does not, of course, follow that there would be coverage for the injury, because the injury might be expected, but that is a different question. For the purposes of the second limb of the ‘Occurrence’ definition, the method by which injury happens is irrelevant, provided it can be said to ‘arise from’ the product. 6.10

The first limb of Article III(V)(1) is more specific: it requires the injury, damage or advertising liability to be caused by an ‘event’7 or by ‘continuous, intermittent or repeated exposure to conditions’.

6.11

Both the first and the second limbs of Article III(V)(1) refer to ‘actual or alleged’ injury, damage or liability ‘caused by’ or ‘arising from’ the relevant thing or event. Read pedantically, this might be taken to suggest that, while a mere allegation of injury or damage would be sufficient (whether or not there is any real injury), the causal link must be genuine. But further reflection shows that this cannot be right. Between alleged (but in fact non-existent) injury and a real event or product, there can never be anything more than an alleged or hypothetical causal connection. The reference to ‘actual or alleged’ injury and damage caused by or arising from an event or product permits either injury or causation, or both, to exist only by allegation and not in reality. It therefore includes all of the following cases: actual injury actually caused by a product or event, actual injury allegedly caused by a product or event, and alleged injury allegedly caused by a product or event. If (as sometimes happens) a manufacturer is held liable for a non-existent injury suffered by a person who has never actually used its product, there is still an occurrence, provided the plaintiff alleged that there was injury caused by the product.8 The Temporal Factor

6.12

Both limbs of the ‘Occurrence’ definition incorporate temporal requirements, but they are differently defined. The conditions reflect the nature of the policy as an occurrences reported policy. One may compare a pure occurrence policy and a pure claims made policy. In the former, what matters is when the triggering event (the bodily injury or property damage for purposes of the CGL Coverage Part) took place. In the latter, what matters is when the claim was made. In the Bermuda Form, both questions matter: the former is

7 The terms of paragraph (2), which refers to ‘alleged event’, suggest that an ‘alleged event’ would suffice too. 8 Earlier versions of the Bermuda Form had the opposite confusion: they referred to ‘injury’ (without the addition of any adjective) which the use of a product ‘causes, allegedly causes or is deemed to cause’. Logically it would be possible to regard this to require real injury, albeit permitting merely alleged causation; this is a slightly more troublesome ambiguity than that in Form 004, as an alleged causal relationship can exist in relation to real injury in a way that a real causal relationship to alleged injury cannot. But, as with the current definition, the fact that the qualifying adjective or adverb attaches only to one of injury or causation should not preclude its application to both requirements. The core meaning is the same in both cases. It includes (i) actual injury actually caused, (ii) alleged injury allegedly caused, and (iii) actual injury allegedly caused, and would include (iv) alleged injury actually caused, if that concept made any sense.

‘Occurrence’: The Basic Definition

95

dealt with here in the ‘Occurrence’ definition, while the latter is dealt with in the notice provision9 and the insuring clause, read together. Under the first limb, the question is whether the event or conditions ‘commence’ on or after the inception date and before the termination date. If they do, then there is an occurrence – even if the conditions or the event continue after the termination date, and regardless of when the injury or damage takes place. Conversely, if the event or conditions ‘commence’ before the inception date, there is no occurrence – even if the injury takes place later. In effect, by choosing ‘commencement’ as the criterion controlling whether the event or conditions produce an occurrence, the Bermuda Form chooses a fixed point of time at which to judge whether there is an occurrence, regardless of the fact that both the occurrence and its consequences may be prolonged, possibly over many years.

6.13

For example, suppose that a factory explosion took place in 1975 and that it released toxic materials which caused cancer in those exposed many years later. Whether treated as an ‘event’ (the explosion) or as an ‘exposure to conditions’ (the toxic chemical released into the environment), the temporal requirement of the first limb of the occurrence definition is not satisfied under a policy with a typical retroactive coverage date of, say, 1986. Conversely, if the factory explosion had taken place in, say, 1995, at a time when the policyholder had Bermuda Form coverage, it would not matter (so far as the occurrence definition is concerned) that the injury did not happen until many years later. It is not necessary, in cases under the first limb of the ‘Occurrence’ definition, to ask ‘when did the injury occur?’. Difficult questions may, however, arise in marginal cases, where one attempts to identify when an event or a particular set of conditions ‘commenced’. But that question is often more easily answered than the question of when ‘injury’ took place. However, it may be necessary to ask when the injury takes place in order to apply the provisions of the policy with regard to ‘integrated occurrences’.

6.14

The second limb of the ‘Occurrence’ definition always requires one to ask precisely that question. Thus, what injury is included in an occurrence under the products liability limb depends on when the injury ‘takes place’. This approach raises a number of important practical points and invites a number of contrasts with the first limb of the ‘Occurrence’ definition.

6.15

The essential problem is that in many cases, ‘injury’ is a vague term in the context of diseases. The process by which the bones in X’s leg are broken by Y’s car takes place in moments. The process by which X develops cancer, for instance, is not. Although the details are controversial, between the point at which X’s body is functioning ‘properly’ – fully regulating the healthy reproduction of cells – and the moment at which X begins to experience the symptoms of illness, or at which a tumour could be seen or felt, there are many intermediate stages involving changes at the cellular or sub-cellular level. Exactly what those changes are, when they occur, how far any particular change could properly be described as involving ‘injury’ and when they might be detectable is often uncertain and controversial.10

6.16

9 See below ch 8. 10 See above ch 5, para 5.57; and (under English law), eg, Durham v BAI (Run Off) Ltd [2012] UKSC 14, [2012] 1 WLR 867 at [5]–[6].

96

The Definition of ‘Occurrence’

6.17

The Bermuda Form makes no attempt to reduce what may be a continuous injury process to a single point in time for the purpose of deciding whether there is an occurrence. The temporal condition does not ask when injury began or when it ended. It simply indicates that there is an occurrence if injury takes place between the inception date and the termination date of the Bermuda Form’s Coverage A. It is that injury, and only that injury, that will be prima facie ‘encompassed by’ the occurrence for the purposes of the coverage clause (though again this may be modified by the aggregation provisions in the policy). Thus, under the first limb of the ‘Occurrence’ definition, one asks ‘When did the event commence?’, with a view either to including the injury consequent to that event in toto within an occurrence, or taking it outside the scope of coverage in toto. In contrast, the second limb demands a more nuanced answer.

6.18

If injury commences before the inception date, continues throughout the period of coverage and ends only after its termination, there is an occurrence. But the occurrence is not coextensive with the totality of the injury. It ‘encompasses’, in the language of Article I, only that part of the injury that takes place between inception and termination of coverage. If ‘injury’ here means both ‘actual injury’ and ‘alleged injury’ – as it is suggested it does – then it must also follow that the question may not necessarily be ‘when did injury take place?’, but ‘when did injury allegedly take place?’. Determining this can be problematic, where the injury is merely alleged. It may indeed be common ground between the insurer and the policyholder that there was no injury at all. It is tempting to say that in such cases one looks merely to the allegations. But it not uncommonly happens, especially in cases where the injury and its causation are obscure, that the plaintiffs in the liability claims have not been too scrupulous in specifying precisely when the injury allegedly took place. In those cases, tribunals will need to do the best they can to fill in the detail of the allegations by reference to what allegations were made (both in the pleadings and in evidence), and if necessary to what allegations might plausibly have been made. In other words, it may be necessary to consider the ‘inner logic’ of the underlying plaintiffs’ case. If one takes it as a premise that product X caused injury Y, how and when might the injury plausibly be hypothesised to have happened? That exercise might seem one in fantasy, in circumstances where it is actually common ground that X did not cause Y, and possibly even that Y never took place at all. So long as it is understood, however, as a process of putting flesh on the bones of allegations that may never have been extensively explored or reasoned out, it appears legitimate and may be necessary.

6.19

In some cases, the date of ‘actual’ injury will coincide with that of ‘alleged’ injury: the plaintiff was injured when he or she claims to have been. Other cases may involve no actual injury, but only alleged injury. In either case, despite difficulties in establishing the relevant date, there is only one date to establish. Problematic situations nevertheless arise. What if the plaintiff alleges injury at one time, but the policyholder maintains that the injury actually took place at some other time; that is, what happens if the dates of ‘actual’ and ‘alleged’ injury do not coincide? This may seem improbable, but it is in fact rather common. For instance, suppose a plaintiff complains of disease developed during the period of Bermuda Form coverage. The plaintiff alleges that the insured’s product caused the disease and posits a process of continuous degenerative disease from first exposure to the product in the 1960s over many years, with symptoms only emerging years later. If correct, the ‘injury’ arguably took place both before and during the policy period.

‘Occurrence’: The Basic Definition

97

There is an occurrence, but it encompasses only part of the alleged injury. On the other hand, the policyholder wishes to produce evidence showing that the disease actually took place only shortly before the onset of symptoms, so that all the injury took place during the period of coverage. In principle, it seems to us, a policyholder must be entitled to rely on any of the alternatives provided. It is therefore open to the policyholder to prove when the injury actually took place, even if that differs from the date when the injury was alleged to have taken place. However, the policyholder must always bear in mind that, to establish an occurrence, it must prove both actual or alleged injury and actual or alleged causation of that injury by the insured’s products. These are connected requirements. If the policyholder can prove actual injury actually caused by its product, then it is entitled to recover on the basis of that injury, despite the fact that the mechanism of injury is different from that posited by the plaintiff in the underlying liability case. If, however, the policyholder is forced to rely on a merely alleged mechanism of causation, then the injury identified must be consistent with this alleged mechanism. In other words, in cases where what actually happened diverges from what allegedly happened, one must pick one version as a whole and stick to it consistently. One cannot combine one theory of injury with a different and incompatible theory of causation.

6.20

The Extent of the ‘Occurrence’ As has already been noted, the two limbs differ in a number of respects: (i) as to what they regard as the necessary causative precursor (in limb one, an event or exposure to conditions, and in limb two, simply the insured’s products); (ii) in the potential liabilities they cover (advertising liability under limb one only); (iii) in the reference point to which the temporal factor is keyed (in limb one, the event, and in limb two, the injury); and (iv) in the way they frame the temporal factor (in limb one, by reference to its commencement, and in limb two, by reference to any part of its progress). Close examination reveals another fundamental difference, namely in what is – at root – the occurrence. Under the first limb, the occurrence is the event; under the second limb, the occurrence is the injury or alleged injury.

6.21

This might seem to be a wholly academic point, but it interacts with the temporal condition and the insuring clause in an interesting way. Under the first limb, where the occurrence is the event, then provided that event commences within the necessary timeframe, it is an occurrence. The event may continue after the termination of the policy, and the injury it produces may take place at any time. Whatever happens, the entire event is the occurrence, and all the injury resulting from that event (whenever it takes place) is capable of falling within the occurrence. In contrast, under the second limb, the occurrence is, on the face of it, limited to the injury that takes place within the timeframe defined as beginning at the inception of the coverage and ending with the termination date. That may or may not be all the injury that the underlying claimant has suffered. Another interesting dichotomy emerges. Under the first limb of the definition, the occurrence is the event (so that, if there is one event there is only one occurrence, no matter how many people are injured by it); conversely, under the second limb, it appears that injury

6.22

98

The Definition of ‘Occurrence’

to each individual person constitutes a separate occurrence. Thus, as on the basis of Article III(V)(1) alone, a plant explosion that injures hundreds of people is still one occurrence, the occurrence is the explosion, and the explosion is an occurrence irrespective of the period over which injuries take place. In contrast, if the same defect in a product injures hundreds of people, there will be hundreds of occurrences – a separate occurrence for each injury. This, however, is not the end of the story – for the aggregation provisions of Bermuda Form may modify that conclusion.11 Those provisions provide an important qualification to the analysis of ‘occurrence’ under limb two, and it is to those provisions that we now turn.

AGGREGATION 6.23

Article III(V)(2) of the 004 Form provides: Except as provided in paragraph (3) below, where an Occurrence exists and a series of and/or several actual or alleged Personal Injuries, Property Damages and/or Advertising Liabilities occur which are attributable directly, indirectly or allegedly to the same actual or alleged event, condition, cause, defect, hazard and/or failure to warn of such, all such actual or alleged Personal Injuries, Property Damages and/or Advertising Liabilities shall be added together and treated as encompassed by one Occurrence irrespective of the period (but without limiting the effect of exclusion IV.A) or area over which the actual or alleged Personal Injuries, Property Damages and/or Advertising Liabilities occur or the number of such actual or alleged Personal Injuries, Property Damages and/ or Advertising Liabilities.

6.24

This Article is the key to the aggregation provisions of the Bermuda Form. Their application may depend (under Form 004)12 on Article III(V)(3), that is, on the notice of the occurrence referring to an integrated occurrence. Under Article III(V)(3), unless the policyholder’s notice specifies that the occurrence is an integrated occurrence, injuries that commence more than 30 days apart are not aggregated.13 To be aggregated, injury (including the injury that forms the subject of advertising liability) or damage to two or more persons or items of property must meet two sets of criteria: (i) the injuries must either commence within 30 days of each other (in which case it is unnecessary to give notice of an integrated occurrence) – such injury is not an integrated occurrence under the definition in Article III(R) – or be the subject of notice of an integrated occurrence; and (ii) the injuries must be related to each other in the way specified in Article III(V)(2).

6.25

Thus, ‘aggregation’ and ‘integrated occurrence’ are not synonymous, though they are related. Injuries commencing within 30 days of each other are to be aggregated, whether

11 Aggregation is important, as discussed elsewhere, because of its impact on the attachment point, or what we have called in this book the excess point, and the limits of the policy; see ch 9 below, and chs 1 and 2 above. 12 The original Bermuda Form (Form 001) did not distinguish between ‘occurrences’ and ‘integrated occurrences’, and in fact did not include a separate definition of ‘integrated occurrence’, so that provisions that were essentially the same as these – though with slight differences in drafting – applied automatically. See Article III(e) of the 001 Form. 13 See further ch 8 below.

Aggregation 99

or not notice of them as an integrated occurrence is given; indeed, notice of them as an integrated occurrence would not be technically correct, since such injuries do not fall within the definition of an integrated occurrence. If the policyholder’s notice does not refer to an integrated occurrence, then the aggregation provisions are not rendered entirely redundant. What then happens, however, is that instead of aggregation applying to all injuries, it applies only to groups of injury commencing within 30 days of each other; in other words, such notice separates what might be a single integrated occurrences into successions of smaller (but still, at least potentially, aggregated) occurrences. On this basis the policyholder might still, in theory at least, be entitled to recover something if it could be shown that, within any single 30-day period,14 the total injuries exceeded the attachment point.

6.26

If aggregation is permissible, because of the time at which the injuries commenced or because notice of an integrated occurrence was given, one then looks to Article III(V)(2) to consider the scope of the aggregation. It is this paragraph that defines the scope of aggregation. The key question here is whether the injuries, damage or liability are attributable ‘directly or indirectly or allegedly’ to the ‘same actual or alleged event, condition, cause, defect, hazard and/or failure to warn’. These are broad words – especially when one takes into consideration the gloss on what is meant by them which is introduced at the end of the provision.15 It should be noted that they primarily stress the factor causing the injury, and not the injury itself; injuries of very different sorts may have a common cause – for instance, where a plant explosion causes both property damage and personal injuries of various types. Still, difficult questions can sometimes arise, as illustrated by pharmaceutical products. There, claimants will often allege a variety of different injuries or symptoms, some of which may be evidently similar and some of which may be quite distinct. At the same time, there may be no precise specification of the defect – for example, each claimant says, in effect, the product was defective ‘because it caused my injury’. There is, then, an intermingling of injury and defect, and a rather delicate question may arise as to how the various different injuries should be grouped together. It is sometimes tempting to see, in such cases, a bewildering range of defects, each corresponding to a particular side effect: the defect of causing headaches, the defect of causing nausea, the defect of causing blindness, the defect of causing heart failure and so forth. But, quite apart from the fact that there may often be either evidence or at least suspicion that these various symptoms are actually part and parcel of the same process, it is inconsistent with the basic approach of Article III(V), which tends towards broad-based aggregation

6.27

14 At first sight, the period looks to be 60 days, because Article III(V)(3) only disaggregates injuries that meet the criteria of Article III(V)(2) if they take place more than 30 days earlier or later than another injury. So if one has injuries on days 1, 2, 32, 60 and 64, those on days 2 and 60 could both be aggregated with those on day 32. However, that would be wrong, because although days 2 and 60 can fit happily with day 32, they cannot sit happily together. In effect, the injuries must be divided up into groups of injuries such that no injury is a member of more than one group, and no injury in any group takes place more than 30 days earlier or later than any other injury in the same group. Yet, there might well be more than one division that would meet those criteria, and in principle the policyholder would be entitled, in our view, to make use of any arrangement that satisfied the criteria. 15 See below para 6.30. Earlier versions of the Form used different terms to discuss the aggregation concepts: eg, Form 003, Article II(f) (‘Occurrence Integration’ definition; ‘batch occurrence’ (undefined) in Article V(d) (‘Notice of Occurrence and Claim’); Form 001, Article II(f) (‘Same Event, Condition or Cause’).

100

The Definition of ‘Occurrence’

and which looks to defect not injury, to see a multiplicity of ‘defects’ on the basis of a multiplicity of ‘symptoms’. The right approach in these cases, it is suggested, is generally to favour aggregation of all the injuries, unless they are clearly unrelated.16 In particular, complainants who allege that they have suffered injury as a result of the particular formulation of a drug should often be regarded as complaining about the ‘same’ defect (namely, inappropriate formulation) and to have suffered as a result of the same cause (namely, exposure to an injurious substance), even though there may be a rather wide range of symptoms. Any other approach tends to become excessively complicated in practice, since plaintiffs in the liability cases often exhibit a range of symptoms and describe them in different ways, and the various different complaints often overlap. 6.28

This cannot, however, be an iron rule. In many product liability cases, the relevant defect is a failure to warn. Here, too, a measure of common sense is needed. If the warnings that should have been given relate to distinct and medically unrelated matters (for instance, cardiovascular risk and skin problems), aggregation is probably inappropriate. If, however, the warnings relate to similar or related matters, that seems unrealistic. Cardiovascular risk, for instance, will often manifest in a number of ways, such as strokes and heart problems. Or a product may cause or be alleged to cause blood clots in both veins and arteries. In practical terms, although no doubt a heart attack is not a stroke and although a venous blood clot is not an arterial blood clot, these should be treated as the result of the ‘same’ failure to warn. What must be warned of is risk, and where the risks are closely related, aggregation is appropriate.

6.29

It should be noted that the question posed in relation to aggregation (is the cause or defect the same?) differs from a question arising under the ‘Occurrence’ definition in the Bermuda Form about whether injury is expected or intended,17 namely, whether the injury is ‘fundamentally different in nature’ from an expected injury. This, too, will often matter in pharmaceutical cases. Suppose, for instance, that a drug is placed on the market with the expectation that it will cause headaches in some patients. As it happens, the drug causes a very wide range of side effects, including not only headaches and nausea, but in some cases long-term psychological damage as a result of changes in brain chemistry. For the purpose of aggregation, we would be inclined to treat all these various complaints as arising from the same alleged defect (namely, the formulation of the drug). For the purpose of the application of the ‘expected or intended’ provisions, however, the headaches and nausea (which were ‘expected’) might well be regarded as ‘fundamentally different in nature’ from the brain damage. In other words, there can be no a priori assumption that, because injuries result from the same ‘cause’ or ‘defect’, they are all to be treated as having the same ‘nature’. There is nothing odd in maintaining a distinction between the tests to be applied in operating the expected or intended provision, and those to be applied for the purpose of aggregation, since the underlying rationales are different in each case. 16 For example, one would not aggregate a claim about choking on a tablet with a claim about illness by ingesting it. The basic mechanism of injury in each case is obviously different, and the defect of physical form that underlines the first complaint differs from the complaint about chemical formulation that underlies the second. There might be other cases where it was clear that the causal mechanism, and thus the defect, must be quite different. 17 See further ch 7 below.

Aggregation 101

These points are underlined by the gloss that is placed on the expression ‘same actual or alleged event, condition, cause, defect, hazard and/or failure to warn of such’ found later in Article III(V)(2). That provision explains that, so far as losses resulting from ‘the design, formulation, manufacture, distribution, use, operation, maintenance and/or repair of an Insured’s Product’ or failure to warn in that context, that expression

6.30

means any such design, formulation, manufacture, distribution, use, operation, maintenance, repair and/or failure to warn, as the case may be, as to which such losses, injuries or damages are directly, indirectly, or allegedly attributable. As respects Advertising Liability, multiple or repeated broadcasts or publications of the same or similar materials shall constitute “the same as actual or alleged event, condition, cause or defect.

This provision reinforces the point that aggregation does not depend on the identity of the product or on the nature of the injury. Two different products may share the same defect. For instance, a manufacturer may produce two different models of a product, which, despite the extensive differences, share the same defect, for instance, in the material used. Injuries caused or allegedly caused by both products would then be aggregated. The same defect may produce different injuries, possibly very different injuries, as pointed out above. Nor does aggregation focus on why a particular aspect of design or formulation is a defect. Different plaintiffs may have different reasons for their complaint and yet still be complaining about the same defect. That a particular drug causes cancer and that it causes heart disease may be different aspects of the self-same ‘defect’ (namely, the inappropriate formulation of the drug), even though the nature of the injuries and the particular mechanism by which the defect causes harm is different in each case.

6.31

A similar reminder of the breadth of the aggregation provisions is given by that part of the paragraph that glosses the application of the basic criteria to advertising liability. The gloss makes it clear that the same ‘actual or alleged condition, cause or defect’ is to be regarded as present where the same material is repeatedly published or broadcast, which seems obvious, and also where ‘similar’ material is republished or rebroadcast. All this goes to demonstrate that aggregation is broad.

6.32

In general, in determining the number of occurrences, New York law proceeds on the basis that under a policy providing cover for ‘accident’18 or ‘occurrence’,19 the court must ask whether there has been one ‘unfortunate event’ or more than one. Like English law,20 it has treated an ‘unfortunate event’ as marked by a ‘close temporal and spatial relationship between the incidents giving rise to injury or loss, and whether the incidents can be viewed as part of the same causal continuum, without intervening agents or factors’.21 There is no difficulty in applying that jurisprudence to the aggregation provision of the Bermuda Form policy insofar as it refers to the same ‘event’ as the aggregating factor. Indeed, it probably does no more than reflect common-sense views about the meaning of ‘event’ in this context. But that is not the end of the matter. As the New York Court

6.33

18 Arthur A Johnson Corp v Indem Ins Co of N Am 7 NY2d 222 (NY 1959). 19 Roman Catholic Diocese of Brooklyn v National Union Fire Ins Co of Pittsburgh Pa, 21 NY3d 139 (NY 2013). 20 See, eg, Kuwait Airways Corp v Kuwait Insurance Co SAL [1996] 1 Lloyd’s Rep 664; Axa Reinsurance v Field [1996] 1 WLR 1026 (HL); Mann v Lexington Insurance Co [2001] Lloyd’s Rep IR 179 (CA). 21 Appalachian Ins Co v Gen Elec Co 8 NY3d 162, 173 (NY 2007).

102

The Definition of ‘Occurrence’

of Appeals has noted, it is the language of the policy that is critical.22 Parties are free to define occurrence in a manner that groups incidents based on different approaches.23 The Bermuda Form does not make singularity of ‘event’ the sole touchstone for aggregation. It goes on to posit a number of other potential singularities: of cause, of condition, of defect or of failure to warn. 6.34

The ‘Occurrence’ definition itself speaks in equivalent terms of an ‘event’ and ‘continuous, intermittent or repeated exposure to conditions’. How far a single case arises from the same conditions may itself call for interpretation. The New York Court of Appeals has made it clear that where a policy uses such language, ‘the definition encompasses and anticipates multiple claims, losses and incidents within the meaning of a single occurrence’.24 How far those words go may be a matter for argument. The Court of Appeals has recognised that they are intended to cover exposure to some injurious agent or phenomenon (such as asbestos, lead-based paint, heat, moisture and radiation), and declined to treat them as applicable to multiple instances of child abuse in the Roman Catholic Church.25 Where there are ‘conditions’, courts may still require some temporal and spatial connection in order to determine that they are the same conditions; thus, exposure to a harmful chemical at one site may not be combined with exposure to the same chemical somewhere else entirely, but exposure of a number of people to a hazard at the same place may be.26

6.35

However, quite apart from the obvious point that these are all fact-specific instances, they are also wording-specific. The Bermuda Form goes considerably further than merely referring to ‘event’ and ‘same conditions’, or even ‘substantially the same conditions’. It includes other and quite distinct aggregating factors, such as ‘cause’ and ‘defect’. So, although the US Court of Appeals for the Second Circuit held in one case that a manufacturer’s liability to numerous consumers for a defective product did not arise out of ‘the same general harmful condition’,27 such claims would certainly be aggregated under the Bermuda Form by virtue of the language that focuses on ‘defect’. Although we are aware of no cases deciding the point, it is arguable that the aggregation based on the same

22 In re Viking Pump 27 NY 3d 244, 250, 256 (NY 2016); Roman Catholic Diocese 21 NY3d at 148. 23 Appalachian Ins v General Electric 8 NY3d at 173; Mt McKinley Ins v Corning 946 NYS2d 136 (App Div 2012). 24 Roman Catholic Diocese 21 NY3d at 150. 25 Ibid. 26 See, eg, Bausch & Lomb, Inc v Lexington Ins Co 414 F App’x 366 (2d Cir 2011); Endicott Johnson Corp v Liberty Mut Ins Co, 928 F Supp 176 (NDNY 1996); Lavandier v Landmark Ins Co 810 NYS2d 45 (App Div 2006); Ramirez v Allstate Ins Co 811 NYS2d 19 (App Div 2006); Nesmith v Allstate Ins Co 24 NY3d 520 (NY 2014). 27 Bausch & Lomb v Lexington 414 F App’x 366. Contrast other cases where clauses providing for the grouping of claims have permitted a wider approach: eg, Endicott Johnson v Liberty Mut 928 F Supp 176, where repeated dumping at each of two waste sites was held to constitute one occurrence at each side; Ramirez v Allstate Ins Co 811 NYS2d 19 (App Div 2006), permitting the aggregation of claims made by different people who ingested lead at different times at the same location; Mt McKinley Ins v Corning 946 NYS2d 136, permitting grouping of asbestos claims arising from exposure to an asbestos condition at a common location at approximately the same time; Verlus v Liberty Mut Ins Co No 14-cv-2493, 2015 WL 7170484 (SDNY 12 November 2015), permitting aggregation for an attack on two people by different dogs but at the same time. The language of the clauses in these cases is much narrower than the aggregation provisions of the Bermuda Form.

Aggregation 103

‘cause’ could be applied in a case where multiple breaches of duty had a single originating cause, such as a systematic defect in the insured’s policies or procedures, even if that led to several incidents. Disaggregation Occurrences that fall under limb two of the ‘Occurrence’ definition always start life as a series of separate occurrences, with one occurrence for each individual person. In those cases, unless injuries are aggregated under Article III(V)(2) and (3), they will remain as separate occurrences. In contrast, occurrences that fall under limb one of the definition (being attributable to an event) generally start life as integrated occurrences28 – in the sense that all the injury attributable to a single event forms part of one occurrence or event by virtue of the definition in Article III(V)(1) alone. The occurrence naturally encompasses within it all the injury flowing from that event. In those cases, the significance of paragraphs (2) and (3) of Article III(V)29 is that the injury may be disaggregated, so that what is on the face of it one occurrence under Article III(V)(1) falls to be treated as if it were more than one occurrence.

6.36

Effects of Aggregation or Disaggregation Where injuries are aggregated, they are to be ‘added together’ and ‘treated as encompassed within one ‘Occurrence’ irrespective of the period (but without limiting the effect of Exclusion IV(A)) or area over which’ they take place. It is sometimes said that this approach ‘redefines’ the occurrence. The drafting technique used here – namely that of treating multiple injuries as encompassed by one occurrence – is not surprising in itself. After all, it has long been clear that even in the most prosaically old-fashioned cover, a single ‘accident’ or ‘occurrence’ may give rise to multiple injuries.30 But it gives rise to certain difficulties of interpretation.

6.37

The essential question is this: can injuries which would not themselves qualify as occurrences have that status conferred upon them by the aggregation provision? Or does the aggregation provision operate only so as to enable injuries which would otherwise constitute many occurrences to be treated as one occurrence? Suppose, for instance, that a policy is purchased for a single year, 2019, by the manufacturer of a drug. During the policy year, the manufacturer becomes aware that its drug causes injury and gives notice of an integrated occurrence. It also removes the drug from the market. The policy is not renewed. However, the injuries involved characteristically take place some time after the product in question has been prescribed and used by the patients. In due course, the policyholder faces claims which may be divided into five categories: (a) injuries which

6.38

28 The exception is for advertising liability, where there may easily be a succession of ‘events’ (individual publications), which require to be aggregated by virtue of paragraph (2). 29 See above paras 6.23–6.24. 30 See Arthur A Johnson 7 NY2d 222 (NY 1959).

104

The Definition of ‘Occurrence’

took place before inception; (b) injuries which took place during 2019 to people who used the drug before 2019; (c) injuries which took place during 2019 to people who used the drug during 2019; (d) injuries which took place after 2019 to people who used the drug before inception; and (e) injuries which took place after 2019 to people who used the drug in 2019. 6.39

It seems clear that injuries in categories (b) and (c) are covered; in each case, the injury occurs ‘on or subsequent to the Inception Date and before the Termination Date of Coverage A’. It also seems clear that, apart from the aggregation provision, injuries in categories (a), (d) and (e) would not be covered: category (a) because the injuries take place ‘before the Inception Date’ (and are for good measure excluded under exclusion IV(A)), and categories (d) and (e) because they take place after the termination of Coverage A. Nevertheless, might it be argued that, since there is some injury which qualifies as an occurrence, these other injuries (although not occurrences ‘in their own right’) can be treated as encompassed by the occurrence that does exist – taking at face value the promise that this is to be done ‘irrespective of the period’ over which they take place?

6.40

With respect to category (a) (pre-inception) injuries, the answer seems clear: they cannot be. After all, the injunction to aggregation ‘irrespective of the period’ is made expressly subject to Exclusion IV(A), which excludes pre-inception injury.

6.41

However, the answer with respect to post-inception injury is not so clear. Indeed, the very fact that the operation of the clause is made subject to Exclusion IV(A) might be thought to tell in favour of permitting aggregation to expand coverage to post-termination injuries. After all, that qualification would not be necessary if the drafter had been confident that, even without it, only post-inception injury was to be covered. And it is notable by its absence in relation to post-termination injury. Moreover, there are good commercial reasons why post-termination injury should be treated separately. It is common practice, once an occurrence has been notified, to exclude coverage for that occurrence going forward, and not unknown for further coverage to be declined or offered only at a prohibitive premium. The ‘claims made’ aspect of the policy might be thought to support the idea that, once notice of occurrence has been given, the insured has ‘locked in’ coverage for that particular problem, regardless of when injuries occur. It could be different, of course, if the injuries occurred because the insured chose to keep a harmful or allegedly harmful product on the market even after giving notice of occurrence, but that particular problem relates to the date of use of the product, not the date of injury – and is more directly addressed by Article III(L)(3).31

6.42

Against this, it can be said that the overall intent of the aggregation provisions is aggregation of occurrences, not redefinition. It supplements the basic ‘Occurrence’ definition – it does not transform it. To allow the existence of some (possibly minor) injury during the policy period to operate as a hook by which the insured can snare coverage for injuries occurring possibly many years later might be thought unreasonable.

6.43

In the first edition of this book, we expressed the view that this latter argument was correct, so that the injunction to add injuries together ‘irrespective’ of the area over which

31 See

below ch 7, para 7.66.

Aggregation 105

they took place must be regarded as subordinate to the first part of the occurrence definition, in effect an injunction to add together only injuries which would otherwise qualify as occurrences. But it has to be admitted that both the linguistic arguments (especially the explicit reference to Exclusion IV(A)) and the arguments from function have some force. It would be surprising if the giving of notice of occurrence might, de facto, enable the insurer to curtail coverage for future injuries resulting from a known problem, even though the policyholder might well be powerless to prevent them. There is, however, an alternative (and possibly better) way in which that oddity might be avoided. Under Article VI(Q) (the Policy Extension Condition), cancellation or non-extension of Coverage A

6.44

shall not affect the rights of the Insured as respects any Occurrence or Integrated Occurrence of which notice was given … prior to such cancellation or non-extension.

That provision clearly contemplates post-expiration injury being encompassed in an occurrence. It would seem to follow that if the policyholder gives notice of an ‘Occurrence’ before expiration of Coverage A or Coverage B, then all the policyholder’s rights in relation to that ‘Occurrence’ are to be preserved, notwithstanding termination. In other words, in respect of that ‘Occurrence’, the policy is to be treated as if termination had not taken place. If that reading were to be adopted, then it could be said that provided the policyholder gives notice of an integrated occurrence before termination, even ‘posttermination’ injuries (if otherwise covered) are eligible to be included in it – as far as that occurrence is concerned, it is as if termination had not taken place. The inclusion of post-expiration injury is also contemplated by the concluding words of the Article IV(Q): and shall not limit whatever rights the Insured otherwise would have under this Policy as respects actual or alleged Personal Injury, Property Damage or Advertising Liability included in such Occurrence or Integrated Occurrence taking place subsequent to such cancellation or non-extension.

That approach, if correct, solves the commercial and practical unfairness which would result if a policyholder, aware that its product had caused some injury, gives notice of a batch occurrence, which provokes either the termination of cover or the introduction of an exclusion for that batch in future coverage. But it would do so without requiring one to read the language of the ‘Occurrence’ definition itself as permitting post-termination injury to be counted, but by the device (artificial, no doubt, but effective) of treating injuries which would otherwise fall within the batch occurrence as not being post-termination injuries at all. It should be noted, however, that it would not permit the insured to include in the occurrence injuries which resulted from sales that occurred after notice was given, for such injuries would be excluded specifically and separately under Article III(L)(3). Such a construction seems to strike a fair balance between the legitimate interests of both parties.

6.45

The current version of the Form also avoids a different problem. However reasonable it may seem to suppose that the insured must have cover from injuries that occur, perhaps long into the future, as a result of those products which have been placed in circulation before Notice of Occurrence is given, it seems far less reasonable to suppose that it could continue to enjoy coverage into the indefinite future for injuries caused by products which

6.46

106

The Definition of ‘Occurrence’

are marketed after notice has been given, at a time when the insured already knows that there is or may be a significant problem. It is now quite clear, thanks to the ‘Commercial Risk Exclusion’ (Article III(L)(3), discussed in Chapter 7), that it does not.

EXPECTED OR INTENDED INJURY 6.47

Purportedly as part of its definition of what injury is to be regarded as encompassed within an occurrence, Article III(V)(2) provides that ‘any actual or alleged Personal Injury, Property Damage or Advertising Liability which is Expected or Intended by any Insured shall not be included in any Occurrence’. In early versions of the Bermuda Form, the entire edifice relating to ‘expected or intended’ injury – with its proviso as to when there might be recovery notwithstanding some element of expectation or intention – was part of the ‘Occurrence’ definition, greatly complicating it. As a matter of technique, these provisions are now dealt with in separate definitions, although it is this reference that makes them technically operative. These provisions are discussed in Chapter 7.

7 Fortuity, Expected or Intended, and the ‘Maintenance Deductible’ INTRODUCTION The Bermuda Form has from the outset contained, as part of the definition of an occurrence, a qualification relating to injury and damage which is ‘expected or intended’. This qualification has itself, however, been subject to a further qualification enabling the policyholder to recover under the policy, notwithstanding such expectation and intention. Some people refer to this further qualification as the ‘maintenance deductible’. The concept that a policyholder should not recover for damage that is expected or intended is a familiar one to US lawyers; ‘Comprehensive’ or later ‘Commercial’ General Liability (CGL) policies have contained a similar provision since the occurrence concept was incorporated into the CGL policy by endorsement in the 1950s and early 1960s and in the standard form in 1966.1 Therefore, a significant body of case law, both in New York and elsewhere, discusses how ‘expected or intended’ provisions in standard CGL forms are to be interpreted. By contrast, the ‘maintenance deductible’ originated with the Bermuda Form, and there is no US case law relating to the interpretation of similar clauses.

7.01

Since the ‘maintenance deductible’ was a novel feature of the 001 Bermuda Form, it is perhaps not surprising that its wording has developed significantly as the various Forms have been produced. The development also reflects, no doubt, the scope for argument which the original wording created. The wording of the 004 Form, whilst broadly embodying the original concept of the ‘maintenance deductible’, thus seeks to spell out in some detail how the clause works, in a way that was not attempted in the earlier Forms.

7.02

The relevant provision of the 001 Form appeared in the occurrence definition in Article III(e) as follows:

7.03

and which injury, damage or liability is neither expected nor intended by the Insured. Where certain injury, damage or liability is expected or intended by the Insured or the Insured has

1 See, eg, Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) chs 1, 4 and 7 (hereafter Masters and Stanzler, Insurance Coverage Litigation). The standard CGL policy in the United States defined ‘occurrence’ for many years as ‘an accident, including continuous or repeated exposure to conditions which results in bodily injury or property damage neither expected nor intended from the standpoint of the insured’ (ibid, App A-18). Later, in 1986, the insurance industry revised the standard CGL policy to make the ‘expected or intended’ provision an exclusion, helping to clarify the burden of proof (ibid App A-31).

108

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

historically experienced a level or rate of injury, damage or liability associated with given products or operations and injury, damage or liability fundamentally different in nature or vastly greater in order of magnitude occurs, such injury, damage or liability shall not by virtue of such expectation, intent or historical experience be deemed expected or intended to the extent and only to the extent it is different or incrementally greater.

These provisions gave rise to arguments particularly in cases involving integrated or batch occurrences, and the insurers sought to clarify the application of the coverage in such situations through a process culminating in the 004 Form. 7.04

The equivalent provision in the 002 Form, again Article III(e), was redrafted so that the final sentence referred to ‘actual or alleged injury, damage or liability’. When this change was introduced, XL wrote to its policyholders describing it as ‘clarification of underwriting intent’ rather than ‘coverage enhancement’.2

7.05

The concept was unchanged in the 003 Form, but considerably expanded in the 004 Form. The occurrence definition, Article III(V)(2), states: provided, however, that any actual or alleged Personal Injury, Property Damage, or Advertising Liability which is Expected or Intended by any Insured shall not be included in any Occurrence.

Article III(L) further defines the nature of expectation and intent as follows: (1) Nature of Expectation or Intent Personal Injury, Property Damage or Advertising Liability shall be ‘Expected or Intended’ where: (a) actual or alleged Personal Injury, Property Damage or Advertising Liability is expected or intended by an Insured; (b) as respects an Integrated Occurrence, an Insured has historically experienced a level or rate of actual or alleged Personal Injury or Property Damage; or (c) as respects an Integrated Occurrence, an Insured expects or intends a level or rate of actual or alleged Personal Injury or Property Damage (irrespective of whether or not the Insured expects or intends Personal Injury to any specific individual or Property Damage to any specific property); provided, however, that in the case of subparagraph (b) and/or (c) above, if actual or alleged Personal Injury or Property Damage fundamentally different in nature or at a level or rate vastly greater in order of magnitude occurs, all such actual or alleged fundamentally different or vastly greater Personal Injury or Property Damage shall not be deemed ‘Expected or Intended’ (subject to paragraph (3) below). (2) Timing of Determination ‘Expected or Intended’ is determined with reference to what is Expected or Intended (as set forth in paragraph (1) above): (a) at the time of any action (or inaction) by any person so acting (or failing to act) on behalf of an Insured (including, without limitation, the sale by an Insured of any Insured’s Products) concerning the consequences thereof; the expectation or intent of any individual person shall be attributed to an entity Insured only if and to the extent that such person is acting (or failing to act) within the scope of their duties on behalf of such entity,

2 Letter

from XL to its policyholders, dated 29 September 1988.

Fortuity and Related Doctrines

109

(b) at the Inception Date by any Executive Officer, and/or (c) as respects any liability of a person or party who is not an Insured assumed by an Insured under a contract or agreement, by an Insured at the time of such assumption. (3) Commercial Risk As respects any Integrated Occurrence arising out of the Insured’s Products, actual or alleged Personal Injury or Property Damage similar to, and not vastly greater in order of magnitude than, that included in such Integrated Occurrence arising out of sales, if any, of such products by the Insured after the date of the Notice of Integrated Occurrence shall be deemed Expected or Intended. No inference shall be drawn from the giving of a Notice of Integrated Occurrence or from this paragraph (3) that actual or alleged Personal Injury or Property Damage arising out of sales of such products by the Insured prior to the date of such Notice of Integrated Occurrence either was or was not Expected or Intended.

The concept of expected or intended injury or damage is also referred to in the Toxic Substances and Pollution Exclusions of the 004 Form.3 Accordingly, a provision precluding recovery for damage which is expected or intended is linked to a qualification (the ‘maintenance deductible’) that enables a recovery to be made, at least to some extent, in certain circumstances. This coupling involves strange bedfellows. The concept of damage which is expected or intended can be seen as a specific contractual application of a more general principle that, generally speaking, insurance applies only if losses are fortuitous. This is, at first sight at least, unrelated to the principle that the parties can agree upon a deductible or excess point at which the insurance will attach. It is the linking of two unrelated concepts that is in part responsible for some of the puzzling aspects of the ‘maintenance deductible’ provision of the contract. In order to unravel the clause, it is sensible to start by identifying some more general principles which, under New York law, define fortuitous losses.

7.06

FORTUITY AND RELATED DOCTRINES A number of related doctrines have arisen from concerns over the ‘moral hazard’ inherent in insurance – the possibility that the policyholder will seek to insure or recover for a loss falsely.4 Thus, in the first-party property context, courts historically have held that insurance applies only to a fortuitous loss. Under the doctrine of fortuity, insurance does not apply to cover a loss to property that was already destroyed when the insurance incepts. In recent years, insurers have raised fortuity as a defence to coverage in the context of third-party liability insurance too and have argued that CGL insurance does not apply to a ‘known loss’ or ‘known risk’. Courts applying New York law have recognised that other doctrines are related to the concept of fortuity.

7.07

For example, recovery is barred for misrepresentation or fraudulent concealment of a material fact by the policyholder. This would include the situation where the policyholder

7.08

3 Articles IV(H) and IV(K). 4 See, eg, Stonewall Ins Co v Asbestos Claims Mgmt Corp 73 F3d 1178, 1215–16 (2d Cir 1995), modified on other grounds, 85 F3d 49 (2d Cir 1996) (applying New York law); City of Johnstown v Bankers Std Ins Co 877 F2d 1146, 1152–54 (2d Cir 1989).

110

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

fraudulently conceals that the damages, for which recovery is sought, had occurred prior to the inception of the policy.5 The question of non-disclosure and misrepresentation is discussed further in Chapter 12 below. 7.09

In addition, there is the distinct and fact-intensive question as to whether the policyholder expected or intended the damage at issue. This inquiry may overlap with the question of whether the policyholder knew of loss at the time the policy took effect, but it is a distinct – and some courts have said overriding – point.6 This defence arises from the policy language. CGL policies have, over the years, used both the concept of ‘accident’ to define the liability-inducing event against which the insurance will pay and have also excluded damage which is expected or intended.7 The question of expectation or intention has frequently arisen in coverage disputes over multi-million dollar liabilities and can be particularly hard-fought and difficult to litigate in pollution claims where the policyholder seeks to recover the costs of cleaning up sites allegedly contaminated by harmful substances. This issue is important because it will act to prevent coverage if this defence is accepted. Fortuity, ‘Known Loss’ and ‘Known Risk’

7.10

Fortuity, ‘known loss’ and ‘known risk’ are related to expected or intended defences, and are often offered as defences in addition to expected or intended. They do not find solid support in New York law and, though subsidiary arguments to expected or intended, they can be addressed first relatively briefly.

7.11

Although the fortuity doctrine and its cousins arose in the context of first-party property insurance, some courts in the United States have also applied the concept to third-party liability insurance.8 There is no exclusion or other provision in the standard CGL policy that gives rise to or mentions fortuity, ‘known loss’ or ‘known risk’.9 Some courts have rejected the idea that ‘known loss’ applies at all to liability insurance, concluding that it is contrary to the concept of insurance which policyholders purchase to protect against potential liability. Other courts have applied the doctrine, but limit its reach and recognise that it is a variant of expected or intended. In Pittston Co Ultramar America Ltd v Allianz Insurance Co, the court held that the known loss doctrine will bar coverage only when the

5 City of Johnstown 877 F2d at 1153. 6 Stonewall 73 F3d at 1215. 7 Standard form CGL policies used ‘accident’ (often undefined) and were ‘written on an accident basis’ until 1966, when the insurance industry drafting organisations in the United States incorporated a definition of ‘occurrence’ which was defined as ‘an accident including injurious exposure to conditions’. Masters and Stanzler, Insurance Coverage Litigation, App A-10; see also the historical discussion of the ‘occurrence’ concept, ibid at chs 1, 4 and 15. 8 See Stonewall 73 F3d at 1215. 9 See, eg, Weyerhauser Co v Aetna Cas & Sur Co 874 P2d 142 (Wash 1984) (declining ‘to add language to the words of an insurance contract’); see also Kenneth Abraham, Environmental Liability Insurance Law (New Jersey, Prentice Hall Law & Business, 1991) 141 (‘The mere fact that the insured knew of a risk at the time of purchase is not a basis standing alone for denying coverage of liability for harm resulting from risk’). Cf NY Insurance Law, s 1101, which, in the context of licensing of insurers, defines an ‘insurance contract’ as an agreement to confer benefit or pecuniary value on another party ‘upon the happening of a fortuitous event’.

Expectation and Intention

111

legal liability of the policyholder is a certainty at the inception of the policy.10 Policyholders buy insurance because of the risk of loss; therefore, a policyholder’s knowledge of the risk of losses does not bar coverage.11 Thus, even where the policyholder may know, prior to the policy’s inception, that it faces a potential liability, the ‘known loss’ defence will not preclude coverage if there remain uncertainties concerning such matters as the number of allegedly damaged properties, the number of claims, the likelihood of successful claims and the amount of the ultimate loss.12 The US Court of Appeals for the Second Circuit found the ‘known risk’ defence to be incompatible with New York law: We do not agree that the cases offered by appellees stand for the proposition that knowledge of a risk makes that risk uninsurable … We find no basis for concluding that New York courts would embrace the ‘known risk’ theory urged upon us by appellees. In our view, to do so might well swallow up the more narrow doctrines regarding (1) concealment and misrepresentation; and (2) damages that are ‘expected’ or ‘intended’ by the insured.13

EXPECTATION AND INTENTION With this background, one can turn to the detail of the expected or intended language in the Bermuda Form. The expected or intended language in the various Forms is a coda to the ‘Occurrence’ definition, and the ‘maintenance deductible’ language provides a further qualification. Accordingly, the clause envisages a two-stage enquiry: first, as to whether actual or alleged personal injury or property damage was either expected or intended; and, second (that is, if it was expected or intended), an inquiry into whether the actual or alleged personal injury or property damage was ‘fundamentally different in nature or vastly greater in order of magnitude’ than the level expected or intended by the insured or its historical experience.14

7.12

The first stage of the enquiry – expectation or intention – will raise factual questions as to whether the policyholder had the relevant expectation or intent. When the insurer raises this argument, the dispute will often involve replaying, in the coverage dispute, issues that the claimants raised in underlying claims against the policyholder. The claimants (and their attorneys) who advanced the underlying claims may well have attempted to portray the policyholder’s actions as deliberate wrongdoing in the hope of recovering common

7.13

10 Pittston

Co Ultramar Am Ltd v Allianz Ins Co 124 F3d 508, 518 (3d Cir 1997). 73 F3d at 1215. The Restatement Liability Insurance states:

11 Stonewall

the fortuity that the liability insurance market requires is not uncertainty about whether a harm has occurred that might lead to an insured liability, but rather uncertainty about whether there is an insured liability that will reach the level of coverage provided by the insurance policy in question. Even if a harm has already occurred and even if it is foreseeable that the insured will be the subject of a liability action, there remains substantial uncertainty about whether there in fact will be a liability action, whether the plaintiff will succeed in the action, and, if so, for how much and when the judgment will be entered. American Law Institute, Restatement of Law, Liability Insurance (St Paul, MN, American Law Institute Publishers, 2019) (hereafter Restatement Liability Insurance), s 46, cmt a (emphasis in original). 12 Pittston Co Ultramar Am Ltd v Allianz 124 F3d at 518–19 (citing Stonewall 73 F3d at 1215). 13 City of Johnstown 877 F2d at 1152–53. 14 Under the 004 Form, the second stage of the enquiry has no application in the context of Advertising Liability.

112

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

law punitive damages or treble damages under a state statute. Even where the underlying claimants succeeded only in establishing negligence, an insurer may nevertheless try to ratchet up the allegation of negligence into a case of expected or intended damage. 7.14

Several legal issues recur in the case law concerned with allegations that damage was expected or intended. First, who bears the burden of proving (or disproving) expectation and intention? Second, is the expectation or intention to be judged subjectively or objectively? Third, whose expectation or intent is relevant? In addition, what is it that must be expected or intended in order to avoid coverage? All of these issues are controversial. Burden of Proof

7.15

It is a controversial question as to where the burden of proof lies. On the one hand, insurers argue that, since the clause is part of the coverage definition, the policyholder must demonstrate that there is coverage, and this requires the policyholder to prove a negative by showing the lack of any expectation or intention. On the other hand, policyholders typically argue that the relevant part of the ‘Occurrence’ definition acts as an exclusion from cover and that (in accordance with normal principles of New York law) the insurer bears the burden of establishing that an exclusion applies. Furthermore, the provision concerning expectation and intention is contained in a proviso to the ‘Occurrence’ definition, and it would be appropriate for the party relying on the proviso to prove its application. In practice, of course, cases rarely turn on questions of the incidence of the burden of proof. This is particularly so in the context of an issue relating to the factual application of the expected or intended provision. In any arbitration where this is raised as an issue, both parties are likely to direct a significant volume of factual and evidence towards the question of expectation and intention. It is highly likely that the tribunal will be able to come to a conclusion on the effect of that evidence without resort to the question of where the burden of proof lies.

7.16

Again, the authorities on the burden of proof are not unanimous, but we suggest that the burden is on the insurer, for two reasons: first, there is a general reluctance to impose a burden upon anybody to prove a negative; and, second, the relevant provision is, in substance, an exclusion from coverage, and it is the substance that matters.15

7.17

There is, however, a 2002 New York Court of Appeals decision which needs to be considered in this context. In Consolidated Edison Co v Allstate Insurance Co,16 the court held that the answer to the burden of proof question ‘centers on the language of the policies’ and decided that, under the policy language at issue there, the policyholder bore the burden of proof on expected or intended.17 The Bermuda Form is, however, unlike Consolidated Edison’s accident-based and occurrence-based excess liability policies – none of which specifically excluded intended or expected harm, or actually defined

15 See

Stonewall 73 F3d at 1205 (applying New York law), and the cases there cited. Edison Co v Allstate Ins Co 746 NYS2d 622 (NY 2002). 17 Ibid at 625–27. 16 Consolidated

Expectation and Intention

113

the terms ‘accident’ or ‘occurrence’ by reference to the policyholder’s expectation or intention.18 The language of the Bermuda Form – ‘neither expected nor intended by the Insured’ – in its plain meaning states that the expectation or intention is that of the policyholder and can by contrast be read as having an ‘exclusionary effect’.19 In their case, Consolidated Edison was left to argue that the policies’ requirement of an ‘accident’ or an ‘occurrence’ on its own operated as an exclusion. The Court in Consolidated Edison refused to import the ‘neither expected nor intended from the standpoint of the insured’ language into the non-standard language at issue: ‘In our view, the contention that the requirement of an ‘accident’ or ‘occurrence’ ‘itself operates as an exclusion is unpersuasive.’20 The Bermuda Form policy is on this point similar to the insurance policies addressed by the US Court of Appeals for the Second Circuit in Stonewall, which specified that property damage must be ‘neither expected nor intended from the standpoint of the insured’. Consequently, Stonewall and other decisions placing the burden of proof on the insurer should, we suggest, be held applicable by a tribunal under the Bermuda Form language. The issue of where the burden of proof lies is distinct from the question of the standard of proof to be applied in considering an issue of expected or intended. In English arbitration proceedings, a tribunal is likely to regard the standard of proof as being a procedural matter which is governed by English law as the lex fori. The issue will therefore be resolved on the balance of probabilities. An allegation of ‘expected or intended’ may, particularly if a subjective test is applied, carry a connotation that the policyholder has behaved in a reprehensible manner. Even in such cases, the standard of proof is the balance of probabilities, albeit that ‘cogent evidence is generally required to satisfy a civil tribunal that a person has been fraudulent or behaved in some other reprehensible manner. But the question is always whether the tribunal thinks it more probable than not’.21

7.18

Subjective or Objective Standard The question whether ‘expectation or intention’ is a subjective standard (what the insured, however unreasonably, actually expected or intended subjectively speaking) or an objective one (what the insured ought to have expected, if reasonable) is also often controversial. But in our view, the Form is clear.

7.19

In two respects, the standard is unquestionably objective. First, under Article III(L)(1)(b), if the insured has ‘historically experienced’ a level or rate of actual or alleged injury, that level or rate is deemed to be expected or intended. What level or rate, if any, has been experienced is an objective matter. It does not matter whether the insured is subjectively

7.20

18 Thus, the insurance policies at issue in Consolidated Edison did not use standard form CGL policy language which since 1966 has referred to damage ‘neither expected nor intended from the standpoint of the insured’. 19 Stonewall 73 F3d at 1205 (citing Utica Mut Ins Co v Prudential Prop & Cas Ins Co 477 NYS2d 657, 660 (App Div 1984)). 20 746 NYS2d at 626–27. 21 Secretary of State for the Home Department v Rehman [2001] UKHL 47 [2003] 1 AC 153, [55] (Lord Hoffmann); Re H [1996] AC 563, 586–87 (Lord Nicholls); In the Matter of D [2008] UKHL 33 [2008] 1 WLR 1499.

114

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

aware of it or subjectively expects it to continue into the future. The historical experience of a level or rate is equated with expectation or intention, whether or not it actually led to any expectation or intention. Second, under Article III(L)(3), injuries arising from the Insured’s products are ‘deemed’ to be expected if they occur after Notice of Integrated Occurrence has been given, if they are ‘similar to, and not vastly greater in order of magnitude than’ those included in that occurrence. Again, that is an objective question. Even if the insured thought that the notice was purely precautionary, and that its claims experience would not continue, similar injuries are, simply by virtue of the notice, deemed to be expected. 7.21

Those elements of objectivity, however, do not demonstrate that in other respects the standard is an objective one. Indeed, they support the proposition that in general the Form has subjective intention and subjective expectation firmly in view. We suggest that, as a matter of construction and logic, ‘intention’ and ‘expectation’ are normally subjective, not objective, concepts. Furthermore, the 001 to 003 Forms expressly say ‘expected or intended by the Insured’. The 004 Form says, in Article III(V)(2): ‘Expected or Intended by any Insured.’ The focus upon the ‘Insured’ can also be seen in Article III(L)(1)(a) and (c) of the 004 Form, where the nature of expectation or intent is defined.22 This language suggests that the question is whether the relevant individuals within the policyholder’s organisation did in fact expect or intend to cause injury – not whether they ought to have done so.

7.22

The point is reinforced, in the case of the 004 Form, by the provisions which relate to the ‘Timing of Determination’ (Article III(L)(2)). Sub-paragraph 2(a) states that ‘the expectation or intent of any individual person shall be attributed to an entity Insured only if and to the extent that such person is acting (or failing to act) within the scope of their duties on behalf of such entity’. Sub-paragraph 2(b) refers to the expectation or intention of ‘any Executive Officer’ at the inception date of the policy. Both sub-paragraphs therefore focus on the expectation or intention of particular individuals within the insured’s business. If the state of mind of particular individuals is to be examined, as Article III(L) (2) indicates, it seems difficult to see how the enquiry can be as to what they ought to have expected or intended, rather than what they actually expected or intended. An objective standard is also not easy to reconcile with the fact that the policy (see Article III(G)) is intended to cover punitive damages, which are most commonly awarded under US law for injury deemed to result from reckless or intentional acts.

7.23

The authorities in the United States as a whole are not unanimous on the issue of whether the standard for ‘expected or intended’ is subjective or objective. The subjective standard is, however, the predominant view,23 and we suggest that it is the better view – both in general and specifically under New York law and the Bermuda Form. Two US Court of Appeals authorities applying New York law squarely support the view that to be 22 Article III(L)(1)(a) refers to ‘actual or alleged Personal Injury, Property Damage … expected or intended by an Insured’; and (c) refers to ‘an Insured expects or intends a level or rate of actual or alleged Personal Injury or Property Damage’. 23 The Restatement Liability Insurance (s 32(4)) states: ‘Unless otherwise stated in the insurance policy, words in an exclusion regarding the expectation or intent of the insured refer to the subjective state of mind of the insured.’ See also ibid s 46, cont’d (‘the subjective standard … is the majority rule’).

Expectation and Intention

115

‘expected’, it is not enough that the policyholder ought to have foreseen injury – it is necessary that the policyholder did in fact expect or intend the injury: City of Johnstown v Bankers Standard Insurance Co24 and Stonewall Insurance Co v Asbestos Claims Management Corp.25 In City of Johnstown, the court concluded that recovery is barred only if the insured intended the damages … or if it can be said that the damages were, in a broader sense, ‘intended’ by the insured because the insured knew that the damages would flow directly and immediately from its intentional act.26

These federal appellate decisions represent that court’s ‘prediction’ as to how the highest state court in New York – the New York Court of Appeals – would rule on the issue if it were presented.27 Particularly given the New York Court of Appeals’ frequent reliance on federal appellate precedents, we consider that they can be considered as persuasive discussions of New York law. In Continental Casualty Co v Rapid-American Corp,28 the New York Court of Appeals cited City of Johnstown with approval and held that the narrower subjective standard applies to the ‘expected or intended’ issue: For an occurrence to be covered under the CNA policies, the injury must be unexpected and unintentional. We have read such policy terms narrowly, barring recovery only when the insured intended the damages … A person may engage in behavior that involves a calculated risk without expecting that an accident will occur – in fact, people often seek insurance for just such circumstances.29

There is a long line of New York authority, in the context of both ‘accident’ and ‘expected or intended’, which holds that negligence, the taking of a calculated risk or some other ‘foreseeability’ standard does not prevent a successful claim by the policyholder.30 An objective standard cannot easily be reconciled with these cases. 24 877 F2d 1146 (2d Cir 1989). 25 73 F3d 1178 at 1205. 26 877 F2d at 1150. See also Olin Corp v Ins Co of N Am 762 F Supp 548, 556 (SDNY 1991), aff ’d, 966 F2d 718 (2d Cir 1992) (‘Under the majority view, damages resulting from the intentional acts of the insured are covered, provided that the insured did not intend to cause the resultant damages. Thus “a pollution related loss arising from an ongoing business practice of discharging pollutants, even if the discharge amounted to wilful and intentional malfeasance, was [typically held to be] covered as long as the ultimate loss was neither expected nor intended”’) (emphasis in original) (quoting Note, ‘The Pollution Exclusion Clause through the Looking Glass’ (1986) 71 Geo LJ 1237, 1250); see also Olin Corp v Lamorak Ins Co 332 F Supp 3d 818, 844 (SDNY 2018), reconsideration denied, No 84cv1968, 2018 WL 4360775 (SDNY 17 Aug 2018) (‘the focus is on whether Olin subjectively expected and intended the specific damage that resulted from its waste disposal practices’). 27 Stonewall 73 F3d at 1205; City of Johnstown 877 F2d at 1150 (both predicting how the New York Court of Appeals would resolve the issue as required under the ‘Erie Doctrine’; see below, Appendix, paras A.15–A.16). See also the decision of the Delaware Superior Court, applying New York law, in Hoechst Celanese Corp v National Union Fire Insurance Co of Pittsburgh, Pa No 89C SE 35, 1994 WL 721633 (Del Super Ct 22 Apr 1994). 28 Continental Cas Co v Rapid-American Corp 593 NYS2d 966 (NY 1993). Consolidated Edison addressed only the burden of proof for expected or intended and noted that the insurance policies at issue there did ‘not explicitly define “accident” or “occurrence” as “unintended” or “unexpected”’ (746 NYS2d at 626). 29 593 NYS2d 966, 970 (NY 1993). The insurance policy there defined ‘occurrence’ as an event ‘which unexpectedly and unintentionally results’ in damage. Ibid at 968. 30 Messersmith 133 NE 424; McGroarty v Great Am Ins Co 36 NY2d 358 (1975); Miller v Continental Ins Co 40 NY2d 675 (1976); Continental Ins Co v Colangione 484 NYS2d 929 (App Div 1985); Gen Accident Ins Co of Am v Manchester 497 NYS2d 180 (App Div 1986); Allstate Ins Co v Zuk 571 NYS2d 429 (NY 1991). See further para 7.28 below.

7.24

116

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

7.25

One earlier Appellate Division decision, County of Broome v Aetna Casualty & Surety Co,31 contains dicta that might be taken to support an ‘objective’ approach which treats injury as expected where it ought to have been anticipated as a ‘substantial probability’. But the question was not expressly before the court or directly addressed, there was no attempt to analyse the competing authorities, and the actual decision was based on the policyholder’s subjective knowledge. As the US Court of Appeals for the Second Circuit, applying New York law, rightly pointed out in City of Johnstown, the County of Broome court failed to consider existing New York Appellate Division and Court of Appeals authority bearing on the issue. County of Broome has, in any event, been overtaken by the Court of Appeals’ decision in Rapid-American, in which the City of Johnstown analysis was approved.

7.26

Thus, the subjective standard is that adopted by the only decision of the New York Court of Appeals directly bearing on the issue (Rapid-American) and by two well-reasoned federal court decisions by the US Court of Appeals for the Second Circuit decisions applying New York law, one of which has been approved by the New York Court of Appeals. Moreover, the subjective standard has been applied by courts outside New York applying New York law32 and is the majority view outside New York.33 Although isolated instances of lower courts preferring (or appearing to prefer) an objective standard can be identified, the balance of authority in favour of the subjective standard is in our view compelling.

7.27

There is, however, a later and somewhat curious decision of the New York Supreme Court (which is the ‘trial court’ within the New York state system), namely Rapid-American Corp v Allstate Insurance Co.34 The decision is not reported in official law reports, but it is sometimes relied upon by insurers in support of an objective approach. The trial judge held in that case that the Court of Appeals’ decision in the earlier Continental Casualty v Rapid-American case was at best ‘obiter dictum’, suggesting that the Court of Appeals had not ‘reached’ the relevant issue. In our view, however, the lower court in Rapid-American v Allstate misconstrued the Court of Appeals’ decision in Continental Casualty v Rapid-American, interpreting it as having addressed only the insurer’s argument that prior jury verdicts awarding punitive damages to underlying claimants collaterally estopped the policyholder ‘from contesting intent to cause injury’.35 What the Court of Appeals actually decided in Continental Casualty v Rapid-American was that the insurer had failed to adduce any evidence of the policyholder’s expectation or intention, other than the prior jury verdicts – which the court found were insufficient on their own. On that basis, the Court of Appeals affirmed the Appellate Division’s ruling that the trial court’s summary judgment in the insurer’s favour was inappropriate.36

31 County of Broome v Aetna Cas & Sur Co 540 NYS2d 620 (App Div 1989). The court in County of Broome relied upon the decision of a federal Court of Appeals applying Minnesota law: Auto-Owners Ins Co v Jensen. In a subsequent case, the Minnesota Supreme Court rejected this analysis of Minnesota law: Domtar Inc v Niagara Fire Ins Co 563 NW2d 724, 735 fn 6 (Minn 1997). 32 For example, Hoechst Celanese 1994 WL 721633, at *1. 33 See discussion of cases in Masters and Stanzler, Insurance Coverage Litigation, ch 7. 34 Rapid-American Corp v Allstate Ins Co No QDS-22236162, slip op at 6 (NY Sup Ct 30 July 1999), reprinted in 13 Mealey’s Litigation Reports: Insurance, No 37 (3 August 1999). 35 Rapid-American v Allstate, slip op at 6 (citing Continental Casualty v Rapid-American 593 NYS2d at 970). 36 Continental Casualty v Rapid-American 593 NYS2d at 970.

Expectation and Intention

117

The Court of Appeals did, accordingly, reach the relevant issue, and ruled in favour of a subjective approach. Following the Court of Appeals’ decision in Continental Casualty v Rapid-American and recognising it as authoritative, intermediate appellate courts in New York have applied a subjective standard.37 It is also important to understand what is meant by ‘objective’ and ‘subjective’ in the present context. Whether the standard is objective or subjective, expectation of injury or damage is different from awareness of some possibility of injury or damage, or even a ‘reasonable foreseeability’ of damage. Courts and commentators favouring an objective approach have adopted a variety of different formulations, such as whether there is a ‘substantial probability’ that damage will take place.38 It should be noted, however, that even courts applying an objective standard do not suggest that the standard is merely that of ‘reasonable foreseeability’. That is certainly true in New York. Ordinary negligence does not constitute an intention to cause damage; neither does a calculated risk amount to an expectation of damage.39 Indeed, the New York Court of Appeals has made it clear that even reckless risk taking does not amount to ‘expectation’ of injury, even where the insurance policy in question expressly makes the test one of ‘reasonable expectation’. In Allstate Insurance Co v Zuk,40 the court was construing a non-standard policy that precluded coverage where injury was ‘reasonably expected to result’ from the policyholder’s actions. The policyholder had killed a friend and had been convicted of second-degree manslaughter. Nevertheless, the Court of Appeals concluded, there might be coverage for the civil action for wrongful death: A person may engage in behavior that involves a calculated risk without expecting – no less reasonably – that an accident will occur. Such behavior, which may be reckless for criminal responsibility purposes, does not necessarily mean that the actor reasonably expected the accident to result.41 37 See, eg, Union Carbide Corp v Affiliated FM Ins Co 955 NYS2d 572,575 (App Div 2012) (‘Plaintiff’s “calculated risk” in manufacturing and selling its products despite its awareness of possible injuries and claims does not amount to an expectation of damage’, citing Continental Casualty v Rapid American); Dryden Mut Ins Co v Brockman 687 NYS2d 504, 505 (App Div 1999) (‘neither the insureds nor their agent engaged in the assault, and there is no evidence that the resulting bodily injury was expected or intended by the insureds’); Zurich Ins Co v White 633 NYS2d 415, 417 (App Div 1995) (‘we find that [the insured] clearly intended to paint the bridge but by virtue of the acts undertaken both before and after the claims of overspraying were revealed, did not intend to cause the resultant damages’); Saks v Nicosia Contracting Corp 625 NYS2d 758, 759 (App Div 1995) (‘Defendant [insured] may have intended to build the house where it is, but there is nothing in the record to suggest that defendant intended to build the house outside the boundary lines of plaintiffs’ property. The damages to the neighboring lot and to plaintiffs’ property clearly were not intended by d efendant’). See too Chubb Ins Co of NJ v Hartford Fire Ins Co No 97 CIV 6935 LAP, 1999 WL 760206, *7–*8 (SDNY 1999); Metro Prop & Cas Ins Co v Sarris, No 1:15CV0780, 2017 WL 3252812 (NDNY 28 July 2017); Olin v Lamorak Ins 332 F Supp 3d at 844. 38 See, eg, the dicta in County of Broome 540 NYS2d at 622 (citing Auto-Owners Ins Co v Jensen 667 F2d 714, 719–20 (8th Cir 1981)). 39 Continental Casualty v Rapid-American 593 NYS2d at 970. 40 Allstate Ins Co v Zuk 571 NYS2d 429 (NY 1991). 41 Ibid at 432. See, too, Automobile Ins Co of Hartford v Cook 818 NYS2d 176 (2006); NY Cent Mut Life Ins Co v Wood 827 NYS2d 760 (App Div 2007). In other cases – for example, deliberate sexual abuse or assault – courts have held that injuries were expected or intended as a matter of law. In such cases, harm to the victim is inherent in the nature of the wrongdoer’s acts, and the harm flows directly and immediately from the wrongdoer’s intentional acts. In such cases, to do the act is necessarily to do the harm. Allstate Ins Co v Mugavero 581 NYS2d 142 (1992); Smith v New York Cent Mut Fire Ins Co 785 NYS2d 776 (App Div 2004). It is doubtful whether this line of authority will be of relevance to the facts of most Bermuda Form disputes.

7.28

118

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

Thus, even if an objective or ‘reasonable person’ standard is adopted, it is not sufficient to show that damage was ‘foreseeable’ to the degree that would suffice to establish negligence – even gross negligence. Even if an objective test is adopted, the damage must reasonably be anticipated not merely as a ‘risk’, but a substantial probability. However, as noted above, the Bermuda Form language by its terms appears to preclude application of an objective standard. 7.29

The conclusion – that a subjective approach is appropriate – also accords with commercial good sense. Although in theory (for instance, in relation to manufacturing defects) there might be liability under US law for entirely unforeseen events, in most cases that will not be so. Those who purchase Bermuda Form coverage are sophisticated companies, often producing complex products in fields such as pharmaceutical or chemical production where risks are extensively investigated. They are at their most vulnerable to liability in those cases where (objectively) a reasonable person would have understood that there was a risk – and either taken steps to avoid it altogether or to warn of it – but the insured did not. If coverage is barred in such cases because alleged injury ought to have been expected, then the cover is reduced to vanishing point, applicable only if the insured has been held liable without any actual negligence on its part. Such a policy would be as useless as automobile liability insurance which protected only constantly careful drivers or professional indemnity cover that protected only constantly skilful doctors. If the intention of the Form were to provide only such cover, it would be pointless. It is one thing to exclude coverage for those who deliberately cause harm, yet quite another to exclude coverage for those who do so inadvertently but carelessly.

7.30

The objective position may be relevant evidentially. An insurer may certainly invite a tribunal to conclude that because a risk was obvious, the insured did in fact expect or intend it to eventuate, and the more obvious the risk, the more persuasive that argument. But one should not confuse method of proof with the target of proof. The target is actual, that is to say, subjective, expectation or intention. Who Must Expect or Intend?

7.31

CGL insurance is sold to businesses. As a result, the question also arises of who, within the policyholder’s organisation, must have actually expected or intended the injury or damage. Must the insurers’ proof show that senior management expected or intended the damage in question? Or, as insurers sometimes argue, can the knowledge or intent of any company employee be imputed to the corporation? The 004 Form states in Article III(L)(2) that ‘the expectation or intent of any individual person shall be attributed to any entity Insured only if and to the extent that such person is acting (or failing to act) within the scope of their duties on behalf of such entity’. It further states that expectation or intent at the inception of the policy must be that of ‘any Executive Officer’. Earlier Forms do not address the issue. The New York Court of Appeals has not settled the issue in the context of insurance coverage disputes. However, courts applying New York law have held that the relevant expectation or intent is that of a corporation’s senior management: The Court finds that it would not be appropriate for defendant insurers to show ‘intent or expectation’ of a ‘low level’ employee to the ‘intent or expectation’ of the corporation.

Expectation and Intention

119

The Court thus holds that it must be the ‘intent or expectation’ of an official of the insured in a management position. This is in accord with New York law that the knowledge of the lower employees may be imputed to the corporation. However, their intent or expectations are not imputed to the corporation. The question of who constitutes an appropriate ‘official in a management position’ is an issue of fact for trial.42

What Must be Intended or Expected? Under the 001 Form, it is the injury or damage or liability which must be expected or intended. It is not clear whether, under the 001 Form, the intention or expectation is directed towards allegations of injury or damage.43 The 002 Form introduced expressly – albeit in the ‘maintenance deductible’ part of the clause – the concept of expecting or intending allegations of damage, whereas the language of the 001 Form was concerned with expected or intended ‘injury, damage or liability’. The 001 Form may connote actual personal injury, actual property damage and actual advertising liability. The position is, however, not clear, since the word ‘liability’ in the expression ‘injury, damage or liability’ may encompass allegations of damage.

7.32

The 004 Form addresses in more detail the question of what must be expected or intended in Article III(L), headed ‘Nature of Expectation or Intent’, in three sub-paragraphs prior to the ‘maintenance deductible’ proviso. Article III(L)(1)(a) refers to actual or alleged personal injury, property damage or advertising liability as the subject matter of the expectation or intention. Sub-paragraphs (b) and (c), which apply only to integrated occurrences and which are linked to the ‘maintenance deductible’ proviso, refer to the concept of a ‘level or rate’ of actual or alleged personal injury or property damage.44 It is clear that none of these formulations look to the act giving rise to the injury, but instead concern the consequences of the act. The causative act will very often be intentional; for example, a manufacturer does not market products by accident, but would not rationally expect or intend those products to injure consumers.

7.33

Article III(L)(1)(a): Read in the context of the Article as a whole, Article III(L)(1)(a) requires an expectation or intention of actual or alleged personal injury to a specific individual person. A classic example in the New York case law, where insurers may be entitled to defend on the basis of ‘expected or intended’, is where there has been a physical or sexual assault on another person. In such a case, there is likely to be an intention or expectation of injury to that individual, not least because harm to the victim is inherent

7.34

42 Hoechst Celanese 1994 WL 721633, at *1. 43 The relevant words are not framed in terms of expectation of ‘allegations’, ‘claims’ or even ‘alleged injuries’; they are framed in terms of expectation of ‘injuries’. As a matter of background, the origin of these very words lies in the ‘expected or intended’ language of the familiar standard CGL form. They were aimed at excluding coverage where the policyholder was found to have expected or intended the damage in question, not simply allegations of damage (eg, Stonewall 73 F3d at 1205; City of Johnstown 877 F2d at 1150–51). It is possible, however, to make a contrary argument when looking at the clause as a whole, in particular the ‘maintenance deductible’ wording. 44 Article III(M)(1)(b) and (c) are therefore not concerned with expectation or intention in relation to advertising liability.

120

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

in the act performed.45 In the product liability context, too, there is authority supporting the requirement of an intention to cause (or an expectation of) injury to a specific person or property.46 7.35

In contrast, it is clear that neither of the other sub-paragraphs of Article III(L), which are concerned with integrated occurrences, requires an intention to cause or an expectation of actual or alleged injury to a specific person or property. Article III(L)(1)(b) refers generally to historical experience, and this is not concerned with particular individuals. The concluding bracketed words of Article III(L)(1)(c) make it clear that there can be expectation or intention of a level or rate ‘irrespective of whether or not the Insured expects or intends Personal Injury to any specific individual or Property Damage to any specific property’.

7.36

Article III(L)(1)(b) is concerned with historical experience of a level or rate of actual or alleged personal injury or property damage. Here, as noted above, the Bermuda Form departs from New York law’s subjective approach to ‘expected or intended’, since the question of whether such level or rate has been historically experienced is an objective factual question. We suggest that this question is usually likely to be answered by reference to a policyholder’s claims history, the material parts of which will often be requested at the time when the policy is written. The claims history for particular products will likely provide information as to both the volume of claims and the amounts paid by way of judgment or settlement. The word ‘level’ is, in ordinary speech, a general word, and it is capable of being applied to a number of metrics, including volume and financial consequences such as amounts paid. The level is not, or not necessarily, the same as the ‘rate’, which will usually require reference to other data. For example, data as to sales made will likely enable a calculation to be made as to the rate at which claims have been made as a percentage of sales of the product in question.

7.37

Issues can arise as to whether a single event, or a limited number of events, can properly be regarded as historical experience of a level or rate. For example, a new product may be developed and be the subject of trials prior to marketing – an example being the clinical trials of a new drug. If one or a handful of adverse events are seen during the trial process, it is doubtful whether this would constitute historical experience of a level or rate, even though it might be possible to calculate a rate from the data generated by the trial. The concept of ‘historical’ experience of a level or rate cannot mean simply that ‘something that has happened’, for many things may happen once, or occasionally but unpredictably. It must mean experience of the past which is sufficiently consistent to warrant making predictions of the future. Only fairly prolonged, consistent and stable historical experience

45 See, eg, Allstate Ins Co v Mugavero 79 NY 2d 153 (NY 1992); Travelers Ins Co v Stanton 645 NYS 2d 948 (App Div 1996). 46 Stonewall Ins Co v National Gypsum Co No 86 Civ 9671, 1991 WL 320046 19146, at *4 (SDNY 31 December 1991) (where one reason for rejecting the insurer’s argument based upon ‘known loss’ (rather than expected or intended) was that the policyholder did not know ‘which buildings would bring suit, what claims would be made, [and] what amounts of liability would be imposed’). See too City of Johnstown 877 F2d at 1151; Playtex Inc v Columbia Cas Co No Civ A 88C-MR-233, 1993 WL 390469 (Del Super Ct 1993); Steadfast Ins Co v Purdue Frederick Co No X08CV020191697S, 2006 WL 1149185, at *4 (Conn Super Ct 11 April 2006) (applying New York law).

Expectation and Intention

121

could be enough, in and of itself, to be treated as equivalent to expectation or intention, which is the subject of the clause as a whole. Issues can also arise as to the relevance of historical experience of products different from those which have given rise to the underlying claims, and thence the integrated occurrence which is the subject of the claim against the insurer. Article III(L)(1)(b) begins with the words ‘as respects an Integrated Occurrence’, and this indicates that the historical experience should concern the product which is the subject of that integrated occurrence. But there may be questions of degree. If, as will often be the case, the new product is an updated formulation or design of a previous product, then the historical experience of that earlier product may be relevant. If, however, a completely new product is brought to the market, then it may not be appropriate to look at the historical experience of other similar products. Yet, even in that latter case, the policyholder may have a relevant expectation of a level or rate of actual or alleged injury in respect of the new product, based upon its experience with other similar products, so that Article III(L)(1)(c) applies even if (b) does not.

7.38

Article III(L)(1)(c) reverts to a subjective question – ie, whether the policyholder expected or intended a level or rate of actual or alleged personal injury or property damage. Under New York law, an awareness of a statistical risk that an injury might take place is not sufficient to amount to expectation or intention.47 However, if a policyholder knows that its statistics indicate that a particular product has resulted in a particular level or rate of claims, it may not be difficult for a tribunal to conclude that its knowledge of this statistical risk also evidenced an expectation of the requisite level or rate of actual or alleged injury.

7.39

In the foregoing discussion, we have treated the expression ‘actual or alleged Personal Injury or Property Damage’ as essentially synonymous with claims. Whilst this is a controversial issue, we consider that (bearing in mind the approach to interpretation under Article VI(O)) it is the appropriate interpretation in the light of the policy as a whole, the ‘Occurrence’ definition as a whole and the ‘maintenance deductible’ proviso discussed in more detail below.

7.40

In order to understand the controversy, it may be helpful to provide an illustration as to how the question can arise in practice. Issues as to the application of Article III(L)(1)(b) and (c), and the maintenance deductible proviso, will most usually arise in the context of product liability claims against a manufacturer of products. In the United States, legal liability attaches to a ‘defective product’ which causes injury. Products may be defective in

7.41

47 Stonewall Ins Co v National Gypsum Co No 86 Civ 9671 (SDNY 1992), 1992 WL 123144, at *10 (SDNY 27 May 1992), aff ’d in part, rev’d in part on other grounds by Stonewall 73 F3d 1178 (2d Cir 1995), modified, 85 F3d 49 (2d Cir 1996). The jury were directed as follows: ‘An injury is not expected or intended simply because the insured is aware of a statistical risk that an injury might occur. To use another example, an automobile manufacturer may be aware of the statistical risk that a buyer of a car will have an accident causing serious injury because of an undiscovered defect in the car. That injury is not, however, expected or intended. On the other hand, an automobile manufacturer who sells a car knowing that the brakes are very seriously defective may be said to have expected or intended the accident and injury that subsequently occur.’ This direction did not require the policyholder to have an expectation of injury to an identified person.

122

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

three ways. Taking, by way of example, a pharmaceutical company which manufactures and sells medicines, a company may produce a defective product as a result of: (i) manufacture (eg, the pill making machines malfunction and omit an important ingredient); (ii) design (eg, an unsafe ingredient is used); or (iii) failure to warn (eg, users are not told to avoid using the product with another product where the combination is harmful, or not told to avoid the product if they suffer from particular conditions in mind). The company will, however, provide warnings to potential users of possible dangers associated with the use of the product, and generally speaking these will enable the company to avoid liability for failure to warn. In the case of the pharmaceutical company, there are detailed and highly regulated labelling warnings on prescription products which are directed to prescribing physicians. If the warning labelling adequately alerts the physician to the risks and side effects of the medicine, based upon the then-current medical knowledge and state of the art testing, no legal liability should attach, because the product is not then ‘defective’, since by definition there was no failure to warn. 7.42

The absence of any liability for failure to warn does not, however, mean that the product has been harmless to all users. For example, it is well known that all medicines may have undesirable side effects in some portion of the population. The case of vaccine against DPT (diphtheria, pertussis and tetanus) provides an excellent illustration of a product that is immensely valuable to humans generally, but that will be harmful to a very small number of individuals who are vaccinated. Similarly, the contraceptive pill is very effective in preventing unwanted pregnancies, but a very small proportion of the population of women users may suffer a blood clot as a result of taking the pill. A lawn-mower manufacturer may also know that a small number of users will suffer injuries, possibly through misuse of the product. In all these cases, individuals within the general population of users may have been injured, but the warnings may have been sufficient to protect the manufacturer from claims, at least in any significant number, for a period of many years, or from any significant legal liability in respect of those claims.

7.43

Subsequently, however, the position may change as a result of some modification made to the product, or because developing scientific knowledge results in new and different warnings or information being given to users of the product. The efforts of the US plaintiff bar may mean that a manufacturer who had previously been virtually free of claims or liability for a particular product now finds that it is facing mass tort litigation with enormous consequential expense. If a claim is then made under the manufacturer’s Bermuda Form policy, the insurer may contend that all of the actual or alleged injuries within the mass tort were expected or intended. The argument is, in essence, that the manufacturer had always known that its product was causing injuries and that there has been no change in the level or rate of such injuries, merely because there is now a very much larger volume of claims or liability. The basis of the argument is, therefore, that an expectation as to a rate or level of ‘actual or alleged personal injury’ or ‘actual or alleged property damage’ is not the same as, and indeed is distinct from, an expectation as to the rate or level of claims.

7.44

We suggest that this approach is too narrow. The overall context of the expected or intended provisions is a liability policy which responds to judgments, settlements and defence costs. None of these would exist unless there were claims, and there can be little

Expectation and Intention

123

doubt that the purpose of the policy is to provide protection to the policyholder against claims. The more immediate context is the ‘Occurrence’ definition, which provides for the aggregation of all actual or alleged personal injuries which are ‘attributable directly, indirectly or allegedly to the same actual or alleged event, condition, cause, defect, hazard and/or failure to warn of such’. This definition goes on to explain that concept as meaning ‘any such design, formulation [etc] and/or failure to warn, as the case may be, as to which losses, injuries or damages are directly, indirectly or allegedly attributable’. These concluding words show that the drafter clearly had in mind the financial liabilities resulting from claims based on a variety of theories. Furthermore, both the ‘Occurrence’ definition and Article III(L) contain numerous references to ‘alleged’ personal injury and property damage. It is clear, therefore, that the policy covers something other than ‘actual’ personal injury. Where a person makes a claim that he or she has been injured, that is an ‘alleged’ injury. The most obvious example of an alleged injury, relevant in the Bermuda Form context, is where a claim is advanced in legal proceedings.48 The claim remains an alleged injury until it is proved or admitted. At that point, it could be reasonably described as an ‘actual injury’. The distinction between actual and alleged injuries, in the context of the Bermuda Form liability policy, is therefore that some injuries reach the stage where it is proved or admitted that the relevant product caused the injury, and others (usually most) never reach that stage and are simply ‘alleged injuries’. To similar effect is the requirement, within the Occurrence definition, for the aggregation of losses which are attributable ‘directly, indirectly or allegedly’ to the same actual or alleged event, condition, cause etc. These various provisions, referencing alleged personal injury or property damage, and losses attributable ‘allegedly’ to the same alleged event etc, also reflect the principle of New York law that a policyholder has coverage for reasonable settlements of underlying claims without having to prove the existence of an actual liability to the third party; ie, a reasonable settlement of an alleged liability.49 This may include the case where claims are made purely or largely as a result of the fact that the policyholder has a deep pocket, rather than because of anything actually wrong with the policyholder’s product.

7.45

We therefore suggest that, in the context described above, the concept of an expectation of a level or rate of ‘actual or alleged personal injury’ or ‘actual or alleged property damage’ cannot easily or satisfactorily be separated from an expectation of a level or rate of claims. If the policyholder is expecting no or minimal claims, there is in our view no difficulty in saying that this is the policyholder’s expectation of the level of alleged personal injury or property damage.

7.46

48 The Oxford English Dictionary defines the word ‘alleged’ to mean: ‘That is claimed or asserted without proof, or pending proof, to be that which is denoted by the noun or phrase being modified; that is subject of an allegation. Freq. with reference to illicit or illegal behaviour.’ The definition is illustrated by a number of examples, including a legal one: ‘You defended me once before sir … for stealin’ a watch.’ ‘For the alleged stealing of a watch, you mean’, corrected the barrister as he pocketed his fee. ‘Alleged be blowed!’ replied the prisoner promptly. ‘I’ve got the watch at ’ome now!’. 49 See the discussion of New York law on coverage for a policyholder’s liability for settlements in ch 5 above, paras 5.22–5.29.

124

7.47

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

In addition, as discussed in the following section concerning the ‘maintenance deductible’, an approach which focuses on the volume or financial cost of claims enables the maintenance deductible proviso to operate in a straightforward and simple way. It is necessary and sufficient simply to look at the routine financial costs of the ‘actual or alleged’ personal injury or property damage which the policyholder has historically experienced or (if different) expects, and this financial cost is then an additional ‘maintenance deductible’ which is over and above the excess point in the policy. The rationale is that the insurer is not insuring the policyholder’s routine cost of doing business, but is insuring its catastrophic losses. At What Time?

7.48

When is the relevant time for ascertaining expectation or intention? Article III(L)(2) now deals with the timing for the relevant expectation and intention – a point that was not expressly dealt with in the original wording of the clause. This is drafted widely: ‘at the time of action (or inaction) by any person so acting (or failing to act) on behalf of the Insured (including, without limitation, the sale by an Insured of any Insured’s Products) concerning the consequences thereof’. Despite the wide wording, it is likely that the time of sale will be the principal point in time which, in practice, will be relevant to a determination of expectation or intention in relation to product liability. It is at that time that the policyholder carries out the act that gives rise to its liability.50

7.49

Under Article III(L)(2), the timing also includes ‘at the Inception Date by any Executive Officer’. This is a different point of time from the time of sale, but the attribution of knowledge and expectation is limited to that of ‘any Executive Officer’.

THE ‘MAINTENANCE DEDUCTIBLE’ Bermuda Form 00451 7.50

As a matter of drafting, the ‘maintenance deductible’ is a qualification of the expected or intended language in Article III(L)(1)(b) and (c), which are both concerned with 50 Stonewall 73 F3d at 1215 (expectation and intent measured ‘at the time of the acts causing the injury’); Munzer v St Paul Fire & Marine Ins Co 538 NYS2d 633 (App Div 1999) (‘it cannot be ascertained as a matter of law that plaintiffs “intended” or expected to cause the damage that resulted … from their manner of operating the plant’); General Accident Ins v Manchester 497 NYS2d at 182 (‘[The test is whether the insured] intentionally caused [the] damage … or if he had expected that his actions would result in such damage’); Stonewall v National Gypsum 1992 WL 123144, at *10 (‘automobile manufacturer who sells a car knowing that the brakes are very seriously defective may be said to have expected or intended the accident and injury that subsequently occur’); Allstate Ins Co v Zuk 571 NYS2d 429, 432 (NY 1991). For non-New York cases indicating that the insured’s expectation of injury is judged at the time of the performance of the acts giving rise to the injury, see, eg, Owens-Corning Fiberglass Corp v American Centennial Ins Co 74 Ohio Misc 2d 183, 233–34 (Ct Com Pl 1995); Public Serv Elec & Gas Co v Certain Underwriters at Lloyd’s of London No 88-4811, slip op, at 12–13 (DNJ 30 September 1994), reprinted in 8 Mealey’s Litigation Reports: Insurance, No 47 (1994). 51 In the previous editions of this book we addressed in more detail the language of the earlier versions of the Form. Since disputes under the earlier versions are now increasingly rare, we focus in this edition on the 004 Form.

The ‘Maintenance Deductible’

125

integrated occurrences. Thus, if there were no historical experience, or expectation or intention, of a level or rate of actual or alleged personal injury or property damage, then the ‘maintenance deductible’ does not swing into play at all. If, however, the relevant historical experience or expectation or intention existed, then the qualification allows for a recovery in circumstances where it could otherwise be contended that at least some of the damage was expected or intended. Where does the term ‘maintenance deductible’ come from? It is a curious phrase: the policy contains no such expression, and the provisions as to deductibles are contained elsewhere.52 The answer is that the expression was used in the Bermuda market at around the time that the Bermuda Form was first drafted and marketed. The expression is intended to convey the concept that the policyholder has to bear certain ‘noise level’ routine claims. To describe the clause as a ‘maintenance deductible’, however, ironically involves the generous use of parol evidence of just the sort that the drafters of the Bermuda Form wanted to prohibit under the modified governing law provision.53 Anyone reading the clause for the first time would not dream of describing Article III(L) as containing a ‘maintenance deductible’, or perhaps would not even understand the concept. Instead, it would be viewed for what it is; namely, a proviso to the expected or intended language of the policy.

7.51

What is the reason for the existence of this ‘maintenance deductible’? The provision recognises that manufacturers will often know that their product will generate a relatively predictable number of claims. To revert to the examples previously given, a lawn-mower manufacturer will know that some users of its product will be injured, and may know that it will receive a certain number of claims. The qualification was originally conceived, as we understand it, with drug manufacturers particularly in mind.54 When drugs are marketed, a drug company may often know that its drugs will sometimes have certain side effects: indeed, the labelling of drugs will invariably contain a warning as to possible side effects. The qualification is aimed at permitting recovery in appropriate cases, even where this is so. Thus, if the drug company finds, for example, that the side effects of a drug affect many more people than it has consistently experienced in prior years, or that there are much more serious side effects than predicted, the policy provides coverage for the additional actual or alleged injuries (‘vastly greater in magnitude’ or ‘fundamentally different in nature’) that the policyholder does not expect or intend. Without this sort of protection, insurers might argue that such injuries are entirely incapable of being covered. The concept of expectation or intention may seem easy enough to apply where the issue concerns the pollution of a site. However, it becomes much harder to apply when one is dealing with what might be described as the law of large numbers. Take the example of the drug company, which expects one person a year to have an adverse reaction.

7.52

52 In the declarations part of the policy and in Article II (Limits of Liability). 53 Article V(q) of the 001 Form prohibits reference to parol evidence. See also Article VI(O) of the 004 Form. 54 The maintenance deductible language in the 001 Form uses the term ‘associated with’, which is a term well known to apply in the context of drugs. There might be scientific uncertainty as to whether a particular side effect was caused by a drug; until the scientific uncertainty was resolved, the side effects would be regarded as being associated with the product in question. Also, individuals may have particular susceptibilities so that they react adversely to a drug. It might be wrong to describe the drug as causing their injuries, but their injuries would be described as being ‘associated with’ that drug.

126

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

The difficulty is that it is impossible, in advance, to identify that single individual. If 10 or 100 individuals unexpectedly have an adverse reaction in a particular year, the insurer might say that the damage to each of them was expected or intended. It is in these circumstances that the proviso is intended to operate so as to preserve coverage for the additional unanticipated extent of the injury or damage.55 7.53

How does the protection actually work? There are a number of controversial issues which arise in this context.

7.54

If a policyholder receives many more claims than expected, at what point are these ‘vastly greater in order of magnitude’ than the claims that it was expecting? Insurers sometimes suggest that an order of ‘magnitude’ means 10 times more and that ‘vastly greater’ therefore connotes something at least or perhaps substantially greater than 10 times the policyholder’s historical experience or expected level of claims, and perhaps even something which is many ‘orders of magnitude’ greater.56

7.55

It is doubtful, in our view, whether the policy should be so construed, applying the principles of interpretation in Article VI(O)). The Bermuda Form does not formulate the requirement in such terms and does not use the expression ‘an order of magnitude’. There is, in our view, nothing in the language which leads to the conclusion that the parties agreed to equate ‘vastly greater’ with a requirement that the additional liability be 10 times more than past experience or intention. New York law requires a policy to be interpreted according to common speech.57 It is therefore not to be interpreted according to the speech of mathematicians using the expression ‘an order of magnitude’ in its technical sense. If the intention had been to impose an approach of mathematics to the relevant question, then it would have been easy to say ‘vastly greater than 10 times’. The word ‘magnitude’ means (as defined by the Oxford English Dictionary): ‘The quality or fact of being great’. It is also there defined, and is used in ordinary speech, to mean ‘size’ or ‘quantity’ or an ‘amount’. The expression ‘in order of’ simply means ‘in terms of’. It is used in the maintenance deductible provision so as to draw a contrast between greatness or size and ‘nature’ – the latter being addressed by the expression ‘fundamentally different in nature’. Interpreting the words ‘in order of magnitude’ as meaning ‘in terms of greatness’ or ‘in terms of size’ does no more than to give those words their ordinary meaning in common speech, where the expression ‘order of magnitude’ is commonly used by people without any knowledge of or intention to refer to its technical mathematical meaning. The concluding words of the proviso (‘all such actual or alleged … vastly greater Personal Injury …’) omit reference to ‘in order of magnitude’, and therefore suggest that what really matters is the question of whether what happened can be described as vastly greater. We suggest that the wording should be applied unconstrained by any technical mathematical meaning. 55 See the discussion on the intent of the Bermuda Form in ch 1 above generally and in para 1.53. 56 In mathematics, one order of magnitude denotes a case where y is about 10 times greater than x, and two orders of magnitude denotes a case where y is about 100 times greater than x. The argument, particularly the more extreme versions of it, may have the consequence that the coverage becomes illusory or nearly so. New York law does not favour an approach of reducing indemnity to a mere shadow or making it nearly illusory: see Messersmith 133 NE at 433; Thomas J. Lipton Inc v Liberty Mut Ins Co 357 NYS 2d 705, 706 (NY 1974); Wright v Evanston Ins Co 788 NYS2d 416 (App Div 2005). 57 In re Viking Pump Inc 27 NY 3d 244 (NY 2016).

The ‘Maintenance Deductible’

127

We have previously suggested that the word ‘level’ is a general word capable of being applied to a number of metrics including volume and financial consequences.58 It is, however, sometimes suggested that this word should be narrowly interpreted, for example, as applying to only to the physical severity of the actual or alleged personal injury or property damage. On this narrow approach, however, it is difficult to see how the technical mathematical meaning of an order of magnitude could be applied, since (for example) the increased severity of a personal injury, compared to what was previously expected, cannot readily be calibrated by reference to a mathematical scale. Furthermore, increased physical severity of the actual or alleged personal injury will, in many cases, enable the policy to respond under the provision concerning claims that are ‘fundamentally different in nature’, and it is therefore inappropriate to give the word ‘level’ a restricted meaning.

7.56

What if the policyholder receives the same number of claims, the injuries are the same, but the extent of the liability is vastly greater; for example, because juries award punitive damages whereas they had not previously done so? Again, substantial arguments can be made in support of the proposition that the policy responds. In our view, there is no real difficulty in saying that the policy should respond, not least because the ‘level’ of the claims is vastly greater.

7.57

In that context, it can reasonably be argued that the focus of the clause as a whole is not on the validity or even the number of claims, but on the policyholder’s expectations and historical experience of the ordinary financial consequences, in terms of payments to settle cases and the costs of litigation, flowing from what was ‘expected’. The essence of the clause is financial. Yet insurers sometimes seek to take a narrow view, suggesting that attention should be restricted only to the number of claims, without taking into account things like their severity, cost and so forth. This narrow approach is, in our view, not consistent with the purpose of the ‘expected or intended’ provisions of the occurrence definition, which aims to produce a ‘maintenance deductible’ to protect the insurers from the ‘ordinary’ or ‘routine’ costs of litigation associated with a product that ought to be absorbed as ‘business expenses’. An increase ‘in order of magnitude’, when applied to the ‘allegations of injury’,59 and in particular the expected level thereof, need not apply simply to the number of claims. It may apply to other measurable qualities, such as size, seriousness, cost and so forth: an allegation may be ‘greater’ in more senses than one.

7.58

The discussion in the foregoing paragraphs has focused principally on the number and nature of the claims made against the policyholder. However, as discussed above, controversy also arises as to whether claims or litigation costs are the focus of the maintenance deductible. The point can be illustrated by the example previously given of the drug company which, over a number of years, has received reports that patients have or may have suffered side effects associated with a particular product, but has rarely if ever been sued in respect thereof. What is the position if, at a certain point in time, plaintiff attorneys in the United States come to consider that these side effects are worthy of substantial litigation, and the drug company then receives a significant volume of claims unlike that

7.59

58 See above para 7.36. 59 The concept of alleged injury appears, expressly, in the 002 Form. However, as discussed in para 7.32 above, the word ‘liability’ in the 001 Form may possibly be interpreted as having the same effect.

128

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

it experienced in the past? Those who argue for a distinction between claims on the one hand and actual or alleged injuries on the other suggest that the maintenance deductible does not simply provide protection against a particular product becoming the target of the plaintiff bar. Thus, in this illustration, the drug company would need to show that what was ‘vastly greater in order of magnitude’ was not simply its expectation of the volume of claims, but its expectation of the volume (presumably the combined volume) of actual or alleged injuries (that is, including those which previously had not developed into claims). On this approach, where the policyholder had a previous expectation of actual injuries which had not developed into claims, the policy would not respond a vastly greater increase in the number of claims referable to the same number of injuries. In order to ground recovery, the policyholder would have to show that a vastly greater number of injuries had occurred, irrespective of the volume of claims. 7.60

For the reasons already outlined, we consider that this interpretation does not give a sufficient or satisfactory meaning to the policy. By referring to ‘alleged injury’, the focus becomes more generally upon the expectation of allegations of injury or claims. To say that the ‘alleged injuries’ (= allegations of injury = claims) are ‘greater in order of magnitude’ than those expected means that the alleged injuries are more numerous than expected.60

7.61

In the above example, where there has been no or minimal claims history, the policyholder can in our view properly say that both the rate and level of alleged injuries is vastly greater in order of magnitude than its expectation of the level or rate of alleged injuries.61 The policy can then work in a straightforward and simple way, by comparing the expected routine financial cost of the actual or alleged injuries which the policyholder had historically experienced or expected, and the financial costs in the new circumstances. This has the merit of being a like-for-like comparison.

7.62

The contrary approach requires an investigation into the policyholder’s expectation as to the level or rate of something other than claims, even though the policyholder’s knowledge of the actual or alleged injuries caused (or allegedly caused) by its product is likely to be most readily be found in the number of claims that it receives. The approach can then lead into consideration of the data that may, in relation to some industries, provide some information as to how many injuries are associated with a particular product. For example, in the case of the pharmaceutical industry, epidemiological studies may be available which provide data in respect of particular side effects or adverse medical events, and these may in some cases provide an incidence rate expressed by reference to the number of years of use of the product. It may then be suggested that the relevant comparison

60 We also suggest that the maintenance deductible applies where the claims are more severe, or more costly, or more difficult to deal with, or in any other commercially relevant way ‘greater’ (= ‘worse’) than expected, not simply that they are more numerous; see above, paras 7.52–7.60 and ch 1, paras 1.32–1.35. 61 Indeed, in many mass tort cases the policyholder will be facing allegations from the claimants that the failure to warn has meant that the product has injured vastly greater numbers of people compared to those who would have been injured if appropriate warnings had been given. The policyholder may therefore be able to rely upon allegations made by the underlying plaintiffs in support of its argument that the alleged injuries are vastly greater in order of magnitude.

The ‘Maintenance Deductible’

129

involves comparing an incidence rate based on the claims made against the policyholder and the incidence rate contained in epidemiological studies, so that the policy responds only if the former is vastly greater in order of magnitude than the latter. We doubt the utility or validity of this approach. A comparison between the claims received, the cost of which is ultimately reflected in the policyholder’s Ultimate Net Loss, and industry data such as those contained in epidemiological studies, is not a like-for-like comparison. In contrast, a comparison between the expected routine cost of claims and the amounts actually paid in a situation where the claims are vastly greater in order of magnitude is relatively simple and involves a comparison of figures which are truly comparable.62 Furthermore, this approach avoids any difficulties of trying to work out which particular individuals have suffered injuries which are within the maintenance deductible and which have not. This would be a complex, if not fruitless, quest in the circumstances of a mass tort, where a large number of plaintiffs will make allegations that they have been injured or damaged, but where the severity of alleged injury is likely to vary significantly between them and where the amounts paid in settlement to each of them will likely be very different.

7.63

We suggest that this approach to the Bermuda Form policy provides an appropriate balance. It means that the insurer does not have to bear the policyholder’s regular cost of doing business: its routine ‘maintenance’ costs. But equally, the policyholder is protected from the fortuity of a significant increase in claims activity which was vastly greater than its historical experience or expectation. There is no difficulty in principle in regarding a vastly increased number of claims (for example, in a situation which becomes a mass tort) as a fortuity; indeed, this is the very fortuity for which high-level catastrophe coverage is purchased. Where a policyholder has an uncertain exposure (eg, where there is uncertainty as to the prospective number of injuries, the number of claims, the likelihood of successful claims and the amount of the policyholder’s ultimate losses), the policyholder is ‘fully entitled to replace the uncertainty of its exposure with the precision of insurance premiums and leave it to the insurers’ underwriters to determine the appropriate premiums’.63

7.64

The relevant time for judging the expectation is (as suggested above) the time when the products are marketed. This matter cannot be judged, for example, at each year’s renewal, since then there would otherwise hardly ever be any cover at all under the policy. As claims grew, even from events that took place entirely in the past and about which the policyholder could do nothing, that volume of claims would be added to the ‘maintenance deductible’, so that the prospects of cover would perpetually recede. That is not required by the terms of Article III(L) or even by the logic (unexpressed in these terms anywhere in the policy) of a ‘maintenance deductible’. The logic of the ‘maintenance deductible’ is

7.65

62 It would be possible to make a comparison between epidemiological data at different points in time, eg, at the time when a drug was marketed and the subsequent time when the claims are made. However, it is difficult to see how such a comparison has any relevance to a claim under a liability insurance policy. It would also lead to the necessity of the parties identifying or obtaining reliable epidemiological data for both points in time. 63 Stonewall 73 F3d at 1215; Union Carbide v Affiliated FM 958 NYS2d 311 (Sup Ct 2010), aff ’d, 955 NYS2d 572 (App Div 2012).

130

Fortuity, Expected or Intended, and the ‘Maintenance Deductible’

that the policyholder should not be able to recover costs that were so routine, predictable and even certain that they ought to have been regarded as ‘business expenses’. A policyholder who knows that each sale carries a certain likelihood of litigation should regard those costs as part of its overhead; it should take those costs into account when deciding whether to do a particular line of business. That requires one to judge expectation at the time of sale. If the policyholder sells something believing (let us suppose, for the sake of argument, reasonably believing) that the average litigation cost will be US$1, but turns out to be US$100, the clause justifies the deduction of US$1 from any claim against the insurer, but not more. By the time the claim is actually made, many years later, the policyholder may know that its earlier calculation was wrong and that the costs of litigation were actually much higher. But it is precisely against that kind of miscalculation that the policyholder wishes to insure.64 Article III(L)(3) 7.66

Article III(L)(3) contains a deeming provision which is applicable to sales made after the date when notice of an integrated occurrence has been given. The thinking behind this further proviso appears to be the notion that if the policyholder has actually given notice of occurrence in relation to a particular product, any sales of the product thereafter should, to a substantial extent at least, be at the risk of the policyholder. This may, however, seem somewhat harsh where the policyholder has taken steps to modify its product in the light of the claims being made. The deeming provision only applies where the actual or alleged personal injury or property damages is similar to and not vastly greater in order of magnitude than that which is included in the integrated occurrence. In the light of that requirement, there is clearly room for argument as to the precise impact of the deeming provision. A Moving Target

7.67

For the reasons explained above, the relevant expectation falls to be assessed at the time that the insured markets the particular products that cause injury. In many cases that will be over a prolonged period. During that period both expectation and historical experience may be changing. The insured’s expectation and experience in relation to products sold in 2015 may be different from that in relation to products sold in 2022. Nevertheless, when it comes to assessing a ‘maintenance deductible’, it will be necessary to do so on an aggregated basis. There is a pool of total liability loss. It includes some part that is expected and some part that is ‘vastly greater’ than what was expected. How can this be divided up when there is no single point of time at which a definitive total expectation existed for the pool as a whole?

64 See above, paras 7.40–7.47 and 7.54–7.64 for discussion of the question of whether the maintenance deductible is concerned with claims or litigation costs.

The ‘Maintenance Deductible’

131

The first question for a tribunal in such circumstances is to consider whether it can safely decide that, making generous allowance for the insurer, the loss that has occurred will exhaust the policy limits. Where the loss is very large and the expectation or historical experience is minor, that is often possible. In our experience, tribunals have rarely had to attempt any more precise assessment. Either what has happened is so close to what was expected (or deemed expected) that there is no coverage at all, or what has happened is so catastrophic that the ‘deductible’ is financially irrelevant to the outcome of the case.

7.68

Nevertheless, the exercise can theoretically arise. We doubt that it is possible to assess the relevant sum in any mathematically precise way; there are too many uncertainties and too much flexibility in the standard to be applied. Faced with such a case, the tribunal can do no more than reach a reasonable assessment given the material available. In that task many simplifying assumptions may need to be made. In this task it has one advantage: the allowable recovery is for such alleged injury as is vastly greater in order of magnitude. The tribunal is therefore likely to take the information from which it has derived its conclusions that alleged injury was expected, or the historical experience, and make a fairly generous allowance in favour of the insurer to determine what part is ‘vastly greater’. Even so, however, mathematical precision should not be expected.

7.69

Forms 001–003 The wording of the earlier Forms is discussed in the previous editions of this book. We consider that the expected or intended provisions, and the maintenance deductible, work in broadly similar ways in the various versions of the Bermuda Form.

7.70

8 Notice of Occurrence INTRODUCTION 8.01

The Bermuda Form contains an important provision relating to the giving of notice of an occurrence. In its current version,1 paragraphs A–C of Article V provides: A. NOTICE AS SOON AS PRACTICABLE If any Executive Officer shall become aware of an Occurrence likely to involve this Policy, the Named Insured shall, as a condition precedent to the rights of any Insured under this Policy, give written notice thereof to the Company in the manner provided in Section D of this Article V. Such notice shall be given as soon as practicable and, in any event, during the Policy Period or the Discovery Period, if applicable, and in accordance with Paragraph 2(b) of Exclusion K, if applicable. Failure to provide written notice as prescribed above shall result in a forfeiture of any rights to coverage hereunder in respect of such Occurrence. B. PERMISSIVE NOTICE Any Insured may at any time during the Policy Period or Discovery Period give notice of an Occurrence to the Company in the manner provided in Section D of this Article V. C. PERMISSIVE NOTICE OF INTEGRATED OCCURRENCE The Insured may at its option give written notice to the Company of any Occurrence as an ‘Integrated Occurrence’ by designating it as such and giving such notice in the manner provided in Section D of this Article V. Once the Insured gives Notice of Integrated Occurrence, all Personal Injury or Property Damage that falls within the Integrated Occurrence (as provided in the terms, conditions and exclusions of this Policy) shall be treated as such for all purposes under this Policy irrespective of whether this Policy has been terminated after the Insured has given Notice of Integrated Occurrence. The limit of liability applicable to such Integrated Occurrences shall be the limit described in Article II of this Policy.

8.02

A policyholder’s notice of occurrence fulfils two related, but distinct, functions. It operates to activate, or ‘trigger’, coverage,2 and in doing so fixes the limits and retentions that will be applied to a claim or set of claims. The fact that notice was given, the terms in which it was given and the time at which it was given are thus important factors in fixing the terms that will apply to any subsequent claim for indemnity. Notice also operates as

1 Article V of the 004 Form. In earlier versions of the Bermuda Form, the notice provision was not a separate article, but was one of the policy’s Conditions (Condition V(d)). 2 Article I of the 004 and earlier Forms.

Introduction 133

a condition of coverage; that is to say, the failure to give notice in a timely fashion may completely defeat a claim for coverage. From these two broad functions, a number of points emerge. First, the notice of occurrence provision is intimately linked to, and indeed referred to in, the coverage clause that is the very first provision of the policy. Thus, Article I contains the essential obligation to indemnify, including the proviso that ‘notice of the Occurrence shall have been first given by the Insured in an Annual Period during the Policy Period in accordance with Article V of this Policy’.3 This is one of the defining features of the Bermuda Form, showing, as already explained,4 that the Bermuda Form is an occurrence reported policy which requires that notice of the occurrence be given during the currency of the policy (Coverage A), or during the extended reporting period (Coverage B) if that is purchased. Notice is therefore essential in order to trigger the policy.

8.03

Second, the notice of occurrence provision is intimately linked to the Bermuda Form’s provisions relating to retentions and limits in two ways.5 On the one hand, the monetary amount of the retention and limits of the policy will be determined by the date of notice of occurrence. This is important in the context of a policy which is intended to be extended for a number of years, and in which the retention and limits may change over time. Indeed, the time of notice will determine not just the retention and limits, but more generally the relevant terms, conditions and exclusions of the policy, which may also change. In addition, the date of notice will determine whether claims arising out of distinct occurrences exhaust the annual aggregate limit: if notice of three distinct occurrences is given in a single policy year, only one aggregate limit will be available for all those three occurrences. Finally, whether a notice is given as notice of an ‘integrated’ or a simple occurrence may determine whether claims from that occurrence can be ‘batched’ for the purpose of exhausting a retention.

8.04

Third, the notice provision seeks to impose a requirement, typical of liability insurance policies in general, that notice of an occurrence should be given promptly. Insurers regard this point as an important protection and, as discussed later in this chapter, New York law, although revised now by statute, has historically been notably strict in enforcing such protection. Accordingly, Article V requires notice from the Named Insured ‘as soon as practicable’. A failure to comply with this requirement is likely to give rise to a ‘late notice’ defence to a claim by a policyholder, a defence that may succeed without proof by the insurer that it was prejudiced by the timing of the notice. The strictness is mitigated by the limitation on the scope of those whose knowledge of the occurrence counts for these purposes, by the qualification that the insured need only give notice of occurrences ‘likely to involve this policy’ and by the further possibility that the insured may have an acceptable excuse for late notification. Nevertheless, this (as soon as practicable) qualification has historically acted as an important protection for the insurer and a significant potential pitfall for the policyholder. Earlier versions of the Form also required, in addition to

8.05

3 Article I of the 004 Form. The proviso in Article I also permits notice to be given during the ‘Discovery Period’ if the policyholder has elected for such coverage. Article I of the earlier Forms contained similar provisions. 4 See generally chs 2 and 5 above. 5 See Article II A of the 004 Form; Articles I and II of the earlier Forms; and ch 9 below.

134

Notice of Occurrence

notice of occurrence, notice of claims which were likely to involve the policy. This meant that a policyholder would in practice have to keep the insurer updated after notice of occurrence was served, and therefore presented a further pitfall to a policyholder. This additional requirement does not appear in the 004 Form. 8.06

Fourth, to obtain coverage for certain pollution risks, the policyholder must give written notice, in accordance with Article V(d), within 40 days of the commencement of a discharge of pollutants.6 This requirement flows from and emphasises the limited extent of the pollution coverage offered by the Bermuda Form.

METHOD OF GIVING NOTICE 8.07

Whatever the function of the notice, its formal requirements are set out in Article V(D) of the 004 Form: D. MANNER OF NOTICE (1) Notice of Occurrence must explicitly be designated as such in writing and must be directed to the Company’s Claims Department at the address set forth in Item 8(a) of the Declarations. (2) Information (including, without limitation, information about pending and/or prior claims, reserves or payments, loss runs, etc.) submitted (whether face to face, by mail, telex, courier, facsimile or otherwise) to the Company’s underwriter(s) (whether in an initial or annual renewal application/submission or otherwise) shall not constitute Notice of Occurrence. All material directed to the Company at the address indicated in Item 8(b) of the Declarations shall be deemed to have been submitted to the Company’s underwriters (unless otherwise acknowledged by the Company in writing).

8.08

The notice must ‘explicitly be designated as such in writing’. Accordingly, general information provided to insurers will not suffice to constitute a notice of occurrence. Furthermore, the notice ‘must be directed to the Company’s Claims Department’ at an address specified in the Declarations. The clause thus draws a distinction between the insurance company’s claims department, to whom notice of occurrence must be given, and its underwriters.7 Whilst information, such as claims and loss runs, will typically be provided to underwriters in the course of a policy’s existence, in particular at the time of renewing for another ‘Annual Period’, this information will not, under the current version of the Form, constitute notice of occurrence.

8.09

Quite apart from being explicitly designated as a notice of occurrence, the notice must – if it is to be a notice of integrated occurrence – be explicitly designated as such. Such a notice should not only identify the fact that some sort of integrated occurrence has taken place, but should also identify what the occurrence is said to consist of. Considerable care

6 See ch 10, which considers the policy exclusions including the pollution exclusion. Earlier versions of the Bermuda Form did not expressly link the written notice requirement in the pollution exclusion to the notice of occurrence provision. The 004 Form eliminates what was arguably an ambiguity. 7 Article V(D) of the 004 Form. Earlier versions of the Form did not differentiate between the insurance company’s claims department and underwriters.

Notice as a Condition of Coverage

135

should be given to drafting such notices to make sure that they describe the integrated occurrence in question accurately and comprehensively, but not too broadly. For instance, it would be unwise to define an integrated occurrence as applying to ‘claims relating to product X’, as this would be too broad. On the other hand, it would be equally unwise to define the claims too narrowly, as experience shows that what is recognisably a single mass tort may see considerable variation in the exact theories of injury, defect, causation and liability as time progresses. If the notice has been too narrowly framed, the insurer may try to argue that claims founded on these later theories are not within the integrated occurrence, so that separate retentions apply or some or all of the claims fall outside the coverage period altogether. However, tribunals applying Article V should bear in mind that notices of occurrence are often given before the full scale and extent of an incipient liability problem is recognised. A notice that is on its face extremely broad should not be construed as relating to issues that were quite outside the scope of those anticipated at the time it was given. Equally, however, one should be slow to conclude that claims fall outside the scope of a notice if they are the result of the evolutionary development that characteristically takes place in mass tort claims over time.

NOTICE AS A CONDITION OF COVERAGE Liability insurance policies almost invariably include a provision requiring notice of an occurrence or claim.8 New York courts frequently address factual issues about whether: (i) the timing of the policyholder’s notice was reasonable under all of the circumstances; (ii) the notice was in fact ‘late’; and (iii) circumstances exist which mitigate or excuse the delay.9 To some degree these issues overlap, and all must (of course) be judged against the terms of the particular policy in question. Courts in New York have sometimes said that there are two questions to be addressed in evaluating late notice defences.10

8.10

First, the court must decide when the obligation to give notice accrued. This must necessarily depend upon the nature and terms of the policy. A primary insurer typically has a responsibility to investigate and defend the claim. Primary policies therefore regularly require immediate notice of any accident or claim involving the insurance. An excess insurer, by contrast, is not interested in every accident, but only in those that may be serious enough to involve it. Accordingly, excess policies usually require an insured to give notice of claims that appear ‘likely to involve’ the excess. New York courts have asked

8.11

8 Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation 2nd edn (New York, Wolters Kluwer, 1996 and 2021 Supp) ch 5 (hereafter Masters and Stanzler, Insurance Coverage Litigation); Barry Ostrager and Thomas Newman, Handbook on Insurance Coverage Disputes, 19th edn (New York, Wolters Kluwer, 2019) ss 4.01–4.05. 9 For example, Hatco v WR Grace & Co 801 F Supp 1334, 1372 (DNJ 1992); Avondale Indus Inc v Travelers Indem Co 774 F Supp 1416, 1430 (SDNY 1991); New York v Amro Realty Corp 697 F Supp 99, 104–05 (NDNY 1988), aff ’d in part, rev’d in part on other grounds, 936 F2d 1420 (2d Cir 1991); North Am Philips Corp v Aetna Cas & Sur Co No 88C-JA-155, 1995 WL 628447, at *5 (Del Super Ct 22 April 1995); Hoechst Celanese Corp v National Fire Ins Co of Pittsburgh, Pa No 89C-SE- 35, 1994 WL 721642, at *3 (Del Super Ct 13 April 1994); James v Allstate Ins Co 578 NYS2d 18, 19 (App Div 1991). 10 Olin Corp v Insurance Co of N Am 743 F Supp 1044, 1053 (SDNY 1990); Nationwide Mut Ins Co v Davis No 99 Civ 11928, 2000 WL 964939, at *2–*3 (SDNY 11 July 2000). In other cases, the courts have elided the two questions; see, eg E Baby Stores Inc v Central Mut Ins Co 337 F App’x 10 (2d Cir 2009).

136

Notice of Occurrence

whether the circumstances known to the insured would have suggested to a reasonable person the possibility of a claim on a primary policy, or the reasonable possibility of a claim that would trigger the excess insurer’s coverage. This approach has therefore been partly subjective, in that it depends upon the facts known to the insured, and partly objective in that it considers the reaction of a reasonable person to those facts.11 The burden of proving the reasonable possibility is on the insurer who asserts that the notice was late.12 The amount of damages claimed against the insured in the ad damnum clause of the complaint is not determinative, given the commonplace practice of exaggerating damages requests in personal injury actions.13 The analysis of whether there is a reasonable possibility of a claim on an excess policy may be far from straightforward, for example, in environmental contamination cases where issues of pro rata allocation over a number of policy years will need to be factored into the analysis.14 8.12

Second, the courts have asked whether the insured did in fact provide timely notice. Where the insurance policy provided for notice ‘as soon as practicable’, courts have required notice within a reasonable time under all the circumstances.15 Where the policyholder seeks to explain a delay that would otherwise make the notice untimely, the courts have placed upon the policyholder the burden of establishing the reasonableness of the proffered excuse.16 New York courts have recognised, as possible reasons for a notification that would otherwise be late, the insured’s good faith belief that it faced no possibility of liability arising out of the occurrence or its good faith belief that there was no insurance coverage available for the claim.17 This encompasses a case where the insured has a reasonable good faith belief that the excess coverage would not be triggered by the underlying claim or claims.18 Where the insured relies upon a good faith belief, for example, that it would not be held liable, it may be relevant on the issue of reasonableness to inquire whether and to what extent the insured inquired into the circumstances of the accident or occurrence.19 Where an insured seeks to excuse delay on the basis that the excess policies would not be reached by the underlying claim, a policyholder should 11 Commercial Union Ins Co v International Flavors & Fragrances Inc 822 F2d 267, 272 (2d Cir 1987) (applying New York law); Olin v INA 743 F Supp at 1055; Nationwide v Davis 2000WL 964939, at *2–*3; Paramount Ins Co v Rosedale Gardens Inc 743 NYS2d 59 (App Div 2002); Turner Constr Co v American Mfrs Mut Ins Co No 01 civ 2899, 2005 WL 1022429, at *12 (SDNY 29 April 2005); Morris Park Contracting Corp v National Union Fire Ins Co of Pittsburgh, Pa 822 NYS2d 616 (App Div 2006). 12 See, in the context of a summary judgment application, Century Indem Co v Keyspan Corp 15 Misc 3d 1132(A), at *7 (Sup Ct 2007) (unreported disposition), aff ’d, 58 AD3d 573, 574 (NY App Div 2009) (denying the insurer’s motion for summary judgment under policies seeking notice ‘“reasonably likely” to implicate excess coverage’). 13 Morris Park Contracting v National Union 822 NYS2d at 619. 14 Century Indemnity v Keyspan 15 Misc 3d at *7, and 58 AD3d at 574. 15 For example, Sec Mut Ins Co v Acker-Fitzsimons Corp 340 NYS2d 902 (NY 1972); Mighty Midgets Inc v Centennial Ins Co 416 NYS2d 559 (NY 1979); Great Canal Realty Corp v Seneca Ins Co 800 NYS2d 521, 522 (NY 2005); see also Same Day Delivery Ser Inc v Penn Star Ins Co 151 F Supp 3d 380 (SDNY 2015); Olin v INA 743 F Supp at 1044–53. 16 Argo Corp v Greater NY Mut Ins Co 794 NYS2d 704 (NY 2005); Great Canal Realty 800 NYS2d at 522. 17 Merchants Mut Ins Co v Hoffman 452 NYS2d 398 (NY 1982); White v City of NY 598 NYS2d 759, 760 (NY 1993); Reynolds Metal Co v Aetna Cas & Sur Co 696 NYS2d 563, 567–68 (App Div 1999). See also Avondale Indus Inc v Travelers Indem Co 774 F Supp 1416, 1430–31 (SDNY 1991). 18 Morris Park Contracting v National Union 822 NYS2d at 620. 19 Great Canal Realty 800 NYS2d at 522; Green Door Realty Corp v TIG Ins Co 329 F3d 282, 286–88 (2d Cir 2003) (applying New York law).

Notice as a Condition of Coverage

137

adduce some evidence that the timing of its notice was the result of a deliberate determination to that effect.20 Courts in New York have found that even short delays in giving notice may, depending on the facts, provide the insurer with a defence to a claim. These can be as short as one, two or four months.21 These cases, particularly before the amendment of New York’s Insurance Law discussed below, held that coverage was forfeit if notice was found to be late.

8.13

The Relevance of Prejudice Unlike the majority of US jurisdictions which apply the so-called ‘modern rule’ on notice,22 the common law rule, which typically held sway under New York law prior to the statutory amendment described below, does not require the insurer to show that the lateness of the notice caused it any prejudice.23 New York courts have traditionally applied the principle that an insurer did not need to show prejudice without regard to whether primary coverage or a high-level excess policy is at issue.24 However, as indicated above, a policyholder that could show that the timing of its notice was reasonable under all of the circumstances, or that it had a reasonable excuse for giving notice when it did, may be able to retain coverage. With regard to excess insurance, New York courts have found notice timely if, before notice, the policyholder’s liability had not been ‘reasonably likely’ to reach the excess layer in question.25

8.14

In 2008, an amendment was made to section 3420 of the New York Insurance Law. This amendment provides that no policy shall be ‘issued or delivered in this state’26 unless it contains ‘in substance the following provisions or provisions that are equally or more favorable to the insured’. One such provision is that timely notice shall not invalidate a claim, ‘unless a failure to provide timely notice has prejudiced the insurer’.27 If prejudice is alleged, then the burden of proving prejudice or the absence of prejudice depends upon whether notice was provided within two years of the time required under the policy.

8.15

20 Long Island Lighting Co v Allianz Underwriters Ins Co 805 NYS2d 74 (App Div 2005); Morris Park Contracting v National Union 822 NYS2d at 620. 21 Olin v INA 743 F Supp at 1053 American Ins Co v Fairchild Indus Inc 56 F3d 435, 440 (2d Cir 1995); Same Day Delivery v Penn Star 151 F Supp 3d at 380. 22 See, eg, Masters and Stanzler, Insurance Coverage Litigation (above n 8) ch 5. 23 Eg, Security Mut Ins Co v Acker-Fitzsimons Corp 340 NYS2d 902 (NY 1972); Argo Corp 794 NYS2d 704; Great Canal Realty 800 NYS2d at 522. See also Mt Hawley Ins Co v Abraham Little Neck Dar Group 825 F Supp 3d 384 (EDNY 2011). See below, paras 8.15–8.16. 24 American Home Assur Co v International Ins Co 661 NYS2d 584 (NY 1997). Interestingly, New York’s highest court, the New York Court of Appeals, applied the modern rule on notice, requiring a showing of prejudice, to reinsurance companies: Unigard Sec Ins Co v North River Ins Co 584 NYS2d 290 (NY 1992). For the rationale for the rule, see Argo v Greater NY Mut Ins 794 NYS2d 704. In In re Brandon (Nationwide Mut Ins Co) 743 NYS2d 53 (NY 2002) and Rekemeyer v State Farm Mut Auto Ins Co 796 NYS2d 13 (NY 2005), the New York Court of Appeals accepted that the no prejudice rule should to some extent be relaxed in a case involving the supplementary uninsured/underinsured motorists provisions of an auto policy. With regard to excess insurance, see, eg, Morris Park Contracting v National Union 822 NYS2d at 622, and the cases cited therein. 25 See para 8.28 below and the cases cited in Masters and Stanzler, Insurance Coverage Litigation (above n 8) s 5.06. 26 NY Insurance Law, s 3420(a). 27 Ibid, s 3420(a)(5).

138

Notice of Occurrence

An irrebuttable presumption of prejudice arises if the insured’s liability has been determined by a competent court or in arbitration, or the insured has settled the claim.28 8.16

This legislation will only potentially impact on Bermuda Form policies where the policy has been issued or delivered in New York State.29 The requirement for the policy or contract to be ‘issued or delivered in this state’ is reiterated by section 8 of the amending statute.30 This provides that the amending act applies ‘to policies issued or delivered in this state on or after’ 180 days after it became law. The amendment therefore applies to policies issued or delivered in New York on or after 17 January 2009. A policy that does not meet this requirement therefore falls outside the scope of the amendment.31 The New York Court of Appeals has given a wide interpretation to the meaning of ‘issued or delivered’ in section 3420. The section therefore applies where both insureds and risks are located in New York.32 A Bermuda Form policy will often cover a large US company with a substantial business presence in New York and which creates risks in that state.33 Accordingly, that requirement of the statute may well be fulfilled.

8.17

The question of whether these statutory provisions will assist a policyholder requires consideration of Article VI(O). This provides a Bermuda Form policy is governed by New York law, except insofar as such laws ‘pertain to regulation under New York Insurance Law … applying to insurers doing business, or issuance, delivery or procurement of policies of insurance, within the State of New York’. It is open to argument whether the relevant provisions of section 3420(a) ‘pertain to regulation’. They do not concern what might ordinarily be thought of as a regulatory matter, such as licensing or similar requirements. However, it might be said that a broad interpretation could be given to the expression ‘pertain to regulation’, so that these provisions fall within the ambit of the exception and are therefore inapplicable.34 It could be argued that the opening words of section 3420(a) have something of a regulatory flavour: they prohibit the issue or delivery of policies in New York unless they contain certain provisions. However, it is clear that these provisions, in particular section 3420(a)(5), affect the parties’ substantive rights, ie, the ability of an insurer to rely upon a late notice defence. The statute is not concerned with any regulatory consequences of issuing or delivering policies in New York. Rather, the straightforward position is that an insurer may not deny coverage under a liability policy based on the failure of the insured to give timely notice of claim unless the insurer 28 Ibid, s 3420(c)(2)(A) and (B). 29 See Marino v NY Tel Co 944 F2d 109 (2d Cir 1991) (applying New York law), holding that New York Insurance Law, s 3420(d), relating to disclaimer, was inapplicable to a policy neither delivered nor issued for delivery in New York. 30 2008 NY Sess Laws 1088 (McKinney). 31 If the amendment were in principle applicable, a question would have arisen as to the position in relation to a Bermuda Form policy originally issued prior to 17 January 2009, but continued by virtue of a ‘renewal’ subsequent to that date. In such cases, it would seem that a notice subsequent to the renewal should be subject to the ‘no prejudice’ rule. 32 Carlson v American Int’l Group Inc 30 NY 3d 288 (NY 2017). 33 Ibid at 305, 306 (interpreting ‘issued or delivered’ in the amended statute to apply ‘to policies that cover insureds and rights located in’ New York, including where the policy ‘was specifically written to cover risks in New York created by the insured’s extensive operations in th[e] state’, citing Proserver Ins Co v Ryba 10 NY3d 635, 642 (NY 2008)); Vista Eng’g Corp v Everest Indem Ins Co 170 AD 3d 915 (App Div 2019). 34 See the discussion in David Scorey, Richard Geddes and Chris Harris, The Bermuda Form: Interpretation and Dispute Resolution of Excess Liability Insurance, 2nd edn (Oxford, OUP, 2018) para 9.45 (hereafter Scorey et al, The Bermuda Form).

Notice as a Condition of Coverage

139

suffers prejudice as a result of the delay.35 Moreover, New York courts consider that section 3420 ‘abrogated the common law “no prejudice” rule’ in New York,36 referring to the ‘intent of the statute, which is to prevent insurers from disclaiming coverage for inconsequential technicalities’.37 A further argument facing a policyholder who seeks to rely upon the statute is that Article VI(O) disapplies New York law insofar as it is ‘inconsistent with any provision of this Policy’. Insurers may argue that the requirement that late notice needs to prejudice the insurer is inconsistent with the language of the policy, which makes timely notice a condition precedent.38 A contrary argument is that a requirement to show prejudice is not inconsistent with the language of the policy. There is no express contractual provision in the policy which says that prejudice is not required, or which otherwise seeks to negate the statute. The effect of the statute is to negate a rule of New York common law rule that prejudice is not required.39 But it does not follow that the abrogation of this rule creates an inconsistency with the terms of the policy: the contractual provision for notice can be read consistently with New York law’s requirement that prejudice be shown.

8.18

Even if the amendment to section 3420 were otherwise applicable, it would not assist a policyholder who failed to give notice at all during the currency of the policy, and who therefore failed to satisfy one of the ingredients of an occurrence.40

8.19

The New York courts applying New York law have decided that the common law ‘no prejudice’ rule continues to apply if the statute is inapplicable to the policy in question.41 The defence of late notice may, however, be waived.42

8.20

35 See, eg Wausau Underwriters Ins Co v Old Republic Gen Ins Co 122 F Supp 3d 44, 55 (SDNY 2015); Harleysville Worcester Ins Co v Wesco Ins Co 314 F Supp 3d 534 (SDNY 2018), aff ’d, 722 F App’x 90, 94 (2d Cir 2019); Bullseye Rest Inc v James River Ins Co 387 F Supp 3d 273 (EDNY 2019). 36 Tower Ins Co of NY v Commissary Direct Inc 63 Misc 3d 1229(A), at *8 (Sup Ct 13 May 2019) (citing Conergic Corp v Dearborn Mid-W Conveyor Co 144 AD3d 516, 523 (NY App 2016); and NY Insurance Law, s 3420(a)(5)). 37 Tower Insurance v Commissary Direct 63 Misc 3d at *9 (citing Castillo v Prince Plaza, 142 AD3d 1127, 1129 (NY App Div 2016); and s 3420 Bill Jacket, L 208, ch 288)). Castillo v Prince Plaza quotes extensively from the legislative history of s 3420(a): 43 Misc 3d 335, 338–39 (NY App Div 2014). 38 Scorey et al, The Bermuda Form (above n 34) para 9.46. 39 Tower Ins v Commissary Direct 63 Misc 3d 1229(A). 40 NY Ins Law s 3420(a)(5); and see Circular Letter No 26 (2008), issued by the Insurance Department of the State of New York on 18 November 2008, and New York Comp Codes R & Regs tit 11, pt 73.1. 41 AH Prop LLC v N H Ins Co 945 NYS2d 391 (App Div 2012); Indian Harbor Ins Co v City of San Diego 972 F Supp 2d 634 (SDNY 2013), aff ’d, 586 F App’x 726 (2d Cir 2014); Ramlochan v Scottsdale Ins Co 55 NYS 3d 369 (App Div 2017); Travelers Indem Co v Northrop Grumman Corp 413 F Supp 3d 263 (SDNY 2019) (analysing ‘late notice’ issues for two separate occurrences: one for soil contamination and one for water contamination). As shown by the decisions in the Travelers v Northrop litigation on the timing of notice, particularly as there, under policies in place before the effective date of New York Insurance Law, s 3420(a), these notice and number of occurrence issues can raise complex fact issues that must be resolved by the trier of fact. See also Travelers Indem Co v Northrop Grumman Corp 416 Supp 3d 290, 298–99 (SDNY 2019) (contrasting, for the purposes of determining when notice should have been given, policy language which did not include ‘grouping language’ to that which did, and refusing to ‘group’ or aggregate exposures into a single occurrence under the former), denying in part, granting in part, motions for reconsideration No 16 Civ 8778 (LBS), 2020 WL 1469550 (SDNY 26 Mar 2020) (denying Travelers’ and Century’s motions for reconsideration, granting in part Northrop’s motion for reconsideration, rejecting summary judgment and reserving issues under the Century but not the Travelers policies for trial). 42 For example, Haslauer v N Country Adirondack Coop Ins Co 654 NYS2d 447, 448 (App Div 1997); MutRede Houses Inc v Greater NY Mut Ins Co 611 NYS2d 550, 551–52 (App Div 1994); and more generally, ch 13 below.

140

8.21

Notice of Occurrence

The current version of the Form (004) expressly makes proper notice a condition precedent of coverage, and failure to give timely notice (if proved) results in the ‘forfeiture of any rights to coverage hereunder in respect of such Occurrence’. In referring to ‘such Occurrence’, the provision makes it clear that the policyholder will still be entitled to bring a claim in respect of a different occurrence.43 Under earlier versions of the Form, the effect of a failure to give notice was stated differently: ‘Failure to provide such written notice as required herein shall result in a forfeiture of any rights to coverage hereunder.’ Insurers could argue that, under these earlier versions, a failure to give notice prevented any claim in respect of any occurrence at all, not simply the occurrence for which notification was late. This would, in our view, be an extreme interpretation and is unlikely to appeal to an arbitral tribunal: properly understood, the ‘forfeiture’ should be construed to relate to the occurrence of which notice was not given, with the words ‘in respect of such Occurrence’ implicitly understood.44 ‘Likely to Involve this Policy’

8.22

The Bermuda Form provides high-level excess general liability insurance, which swings into play when the policyholder is facing a huge, catastrophic loss. This is reflected in Article V(A), which requires notice only where an executive officer of the policyholder becomes aware of an occurrence ‘likely to involve this policy’. Accordingly, the clause focuses attention on whether the excess layer covered by the policy, and not some underlying layer, is likely to be involved.45 The clause requires actual awareness on the part of an ‘Executive Officer’. An ‘Executive Officer’ is defined elsewhere in the policy in relatively wide terms,46 with a good deal of room for potential argument at the margins.

8.23

Thus, notice to the insurers is not contractually required under the 004 Form until an appropriate individual officer or employee within the policyholder is aware that an ‘occurrence’, as the policy defines it, has actually taken place. If, for example, the occurrence has not been reported to the policyholder at all, then the notice provision cannot apply. There is, however, a separate and more difficult question as to whether the Executive Officer, if aware of the occurrence, must be aware of the character of the occurrence, viz that it is one which is likely to involve the policy. Is it necessary that the person concerned is actually aware that the occurrence is likely to involve the policy? Or is it sufficient if a

43 The first paragraph of Article V(A) refers to the obligation to give notice ‘as a condition precedent to the rights of any Insured under this Policy’. When read together with the second paragraph, it is clear that the forfeiture of rights to coverage applies only to the particular occurrence that was notified late. 44 See the decision under English law in Aspen Insurance UK Ltd v Pectel Ltd [2008] EWHC 2804 (Comm), [2009] Lloyd’s Rep IR 440, [65]. 45 See below, para 8.28. 46 See Article III(K) of the 004 Form: ‘Chairman of the Board, Chief Executive, Operating, Financial and Administrative Officers, Managing Director, and any Vice President (including, without limitation, Executive and Senior level) and any manager in the Risk Management or Law Department of the Named Insured.’ Earlier versions of the Form were less precise, and in certain respects both narrower and wider than the current Form: eg, ‘any employee of the risk management or legal department or any officer of any Insured’ (001); ‘a manager or equivalent level employee of the risk management, insurance or law department or any executive officer of the Named Insured’ (003).

Notice as a Condition of Coverage

141

reasonable executive in the position of the person in question would or should have been aware that the occurrence, objectively speaking, would be likely to involve the policy? That is, is the test objective or subjective? Although a purely objective interpretation is grammatically and syntactically possible, it is not attractive. The clause refers specifically to the (actual) awareness of particular individuals (that is, any Executive Officer), not to matters of which the ‘insured’ ought to have been aware. There is nothing in the language which suggests that the first part of the phrase is to be construed subjectively, but the second part is to be construed objectively: the natural construction of the policy language is that the Executive Officer must be aware not only of the occurrence, but also that the occurrence is likely to involve the policy. Nor is a purely objective interpretation consistent either with New York law in general or with the interpretation of the Bermuda Form generally adopted in the insurance market, as evidenced by the Notice Guidelines issued by XL for many years.47 Generally, as discussed below, New York law regards a policyholder’s notice as proper if it was reasonable. Evaluation of the policyholder’s conduct under that standard depends on a review of all of the surrounding facts and circumstances.48 This determination necessarily depends on the policyholder’s actual knowledge and understanding at the relevant time. Accordingly, a delay in giving notice may be excused if the policyholder reasonably believed in good faith that: (a) it faced no liability in connection with the underlying claims; or (b) it had no claim for coverage for the underlying claims.49

8.24

Thus, in Mighty Midgets Inc v Centennial Insurance Co,50 the insurance policy required written notice of an occurrence to be given to the insurer (Centennial) ‘as soon as practicable’. The policyholder, which had little experience in insurance matters, had promptly notified its broker orally of the relevant incident, but did not give written notice directly to Centennial because it believed that any claim would be covered by different insurers. Thereafter, the policyholder gave notice to Centennial seven and a half months after the incident. Reviewing all of the facts, the New York Court of Appeals, New York’s highest court, found that notice was given ‘as soon as practicable’. The court held that this phrase was an ‘elastic one not to be defined in a vacuum’.51 It concluded that the phrase was not necessarily synonymous with ‘immediate’ or ‘prompt’. Instead, according to the court, the phrase required consideration of ‘what was within a reasonable time in the light of the facts and circumstances of the case at hand’. Accordingly, the court held that it was relevant to consider the policyholder’s erroneous belief that Centennial’s coverage did not apply. The policyholder’s failure to notify the insurer for the first seven and a half months was held, in all the circumstances, to be reasonable. However, a policyholder should not make fine judgments about whether or not coverage applies, since New York courts have held that, where ‘coverage is unclear, reasonable insurance holders give notice’.52

8.25

47 See

below, para 8.31. Mighty Midgets v Centennial 416 NYS2d 559; Paramount Ins v Rosedale Gardens 743 NYS2d 59. 49 Reynolds Metal Co v Aetna Cas & Sur Co 696 NYS2d 563 at 567–68 (App Div 1999). See above, para 8.12. 50 416 NYS2d 559 (NY 1979). 51 Ibid at 563. 52 Pfeffer v Harleysville Grp Inc 502 F App’x 28, 30 (2d Cir 2012). 48 Eg,

142

Notice of Occurrence

8.26

In Reynolds Metal Co v Aetna Casualty & Surety Co,53 the court was concerned with two clauses that obliged the policyholder: (1) to give written notice ‘as soon as practicable’ when an occurrence takes place; and (2) to forward to the insurer, if a claim was made against the insured, every demand, notice, summons or other process received by the policyholder. Issues arose as to whether the policyholder had complied with these provisions, in the context of various claims made against the policyholder for environmental pollution. New York’s Appellate Division referred to previous case law finding an excuse for late notice when the insured believed in good faith that it had no liability in respect of the claim, and its belief was reasonable in the circumstances. The policyholder’s reasonable belief that coverage did not apply would similarly provide an excuse. The court applied these principles to failures to notify both occurrences and claims as soon as possible.54

8.27

The language of the Bermuda Form is consistent with this approach. It addresses both the questions normally addressed by a New York court in determining the reasonableness of the policyholder’s notice: the awareness of the policyholder that liability was in prospect, and the policyholder’s awareness that it was a liability for which coverage was likely to be afforded under the policy. It might be argued that this policy language is rather more generous to the policyholder than New York law normally has been, since it does not expressly refer to whether the ‘awareness’ (or absence thereof) must be reasonable or unreasonable.55 The principal objection to a subjective approach is that it would relieve an insured from the obligation to give notice in circumstances where the relevant Executive Officer should have appreciated the reasonable possibility that the occurrence would impact upon the policy, and where the executive’s view to the contrary could not reasonably have been held or would have been different if the executive had looked into the situation more carefully. However, in practice, the objective reasonableness of a policyholder’s view is likely to be relevant. Arbitral tribunals may be unlikely to accept that a large corporation with exposure to liability risk in the United States can have failed to recognise what would be, to a reasonable person, an obvious danger. Moreover, it is at least arguable that New York law, when applied to Article V, would not permit a policyholder to abuse the discretion granted to it by Article V by failing to give notice if, on the facts and circumstances as they were known to the policyholder at the time, it was clear that notice should have been given.56 On this approach, what triggers the application of 53 696 NYS2d 563 (App Div 1999); see too Turner Construction (applying New York law) 2005 WL 1022429, at *11–*14 (applying New York law). 54 There is controversy in the New York case law as to whether Reynolds was correct in holding that late notice of a claim can be excused by the insured’s good faith belief in non-liability: see Rockland Exposition, Inc v Great American Assurance Co 746 F Supp 2d 528, 540 (SDNY 2010), affd, 445 F App’x 387 (2d Cir 2011). However, this controversy does not extend to late notice of occurrence (which is the requirement of the 004 Form), where a good faith belief in non-liability, when reasonable under the circumstances, may excuse delay: Same Day Delivery Serv v Penn Star 151 F Supp 3d at 386. 55 Additionally, if the provision is construed subjectively, the burden of proof would be on the insurer to prove that the Executive Officer was aware that the occurrence was likely to involve the policy. The policyholder would therefore not have to prove either reasonableness or, for example, that the timing of its notice was the result of a deliberate determination that the claim would not reach the excess layer (see above the cases cited in nn 19 and 20). In practice, however, it is unlikely that the issue of late notice will be resolved by reference to the burden of proof; the evidence as to why notice was not given will be available to the tribunal, and this evidence will be evaluated for the purposes of the tribunal’s decision. 56 See, eg, the cases cited in which the New York courts have applied an approach to notice questions that is in part objective: eg, in considering the reaction of a reasonable person to the facts known to the policyholder

Notice as a Condition of Coverage

143

Article V is primarily subjective, in the sense that it depends on an awareness induced in particular individuals by the circumstances as they then appear. Certainly, the question will never be ‘was there in fact a likelihood of loss involving the policy?’ On the other hand, it might not be appropriately characterised as entirely subjective, in the sense of depending wholly on the honesty of the policyholder’s expectations: a policyholder who chooses to shut its eyes to an obvious or apparent risk exposes itself to a risk that its notice will be treated as late. The inquiry also understandably takes into account the layer of coverage involved. It may be appropriate to expect prompt notice to a primary insurer, which has a duty to defend, while at the same time recognising that notice to an excess insurer, whose limit attaches excess of millions of dollars, may be given later in time.57 Thus, because a Bermuda Form policy is a high-level excess policy, the policyholder is entitled to exercise reasonable judgment as to whether a particular claim or batch of claims is likely to exceed the attachment point of the policy. This reflects the nature of excess insurance, which provides cover after exhaustion of all underlying policy limits. The obligation to notify an excess insurer arises only when the policyholder determines that it faces liability sufficient to exhaust the limits underlying the excess insurer’s policy.58

8.28

Another area of potential debate concerns what is meant by ‘likely’, which is an ambiguous word.59 In our view, the word does not, in this context, mean ‘more probable than not’. If it did, a policyholder who faced major litigation with a substantial risk that damages would exceed the policy retention would be justified in withholding notice so long as it was being advised that its prospects of success in such litigation were even marginally greater than 50 per cent. That could not be right, because it would defeat one of the purposes of the notice obligation, namely to permit the insurer to participate actively in the defence of litigation which might result in loss to it. In Christiania General Insurance Corp of NY v Great American Insurance Co, the US Court of Appeals for the Second Circuit said that a provision requiring notice when it ‘appears likely’ that a claim will or may involve the policy does not require a probability – much less a certainty – that the policy at issue will be involved. All that is required is a ‘reasonable possibility’

8.29

(such as Commercial Union v International Fragrances 822 F2d at 272 and the other cases cited in n 11 above), and in considering the reasonableness of a bona fide belief held by the policyholder and advanced as the reason for late notice (see the cases cited in footnotes associated with paras 8.12 and 8.24–8.26 above). It is of course always necessary to consider whether the language and context of the clauses considered in the New York case law is sufficiently similar to the wording of Article V. For example, Article V in the Bermuda Form differs considerably from the standard in a primary policy that requires notice as soon as practicable of an occurrence that may result in a claim. 57 XL’s ‘Notice Guidelines’ to all policyholders recognise this: see para 8.31 below. 58 See, eg, Olin v INA 743 F Supp 1044; Maryland Cas Co v WR Grace & Co 128 F3d 794, 800–01 (2d Cir 1997) (applying New York law); Reynolds Metal 696 NYS2d at 569; Paramount Communications v Gibraltar Cas Co 612 NYS2d 156 (App Div 1994); Morris Park Contracting v National Union 822 NYS2d at 622; Century Indemnity v Keyspan 15 Misc 3d 1132(A). 59 The ambiguity can be appreciated if one contrasts ‘likely’ with ‘unlikely’. If ‘likely’ means, in mathematical terms, ‘with a probability P > 0.5’, then ‘unlikely’ would mean ‘with a probability P ≤ 0.5’. But in ordinary speech we would certainly not regard a probability of 0.5 as ‘unlikely’ and we would happily describe probabilities substantially below 0.5 as ‘not unlikely’. This illustrates the impossibility of correlating probability words, as used in ordinary speech, with precise measurements of probability expressed numerically.

144

Notice of Occurrence

of such happening, based on an objective assessment of the information available.60 This approach has been followed in subsequent decisions.61 Accordingly, whilst there is scope for argument on this issue,62 a tribunal may hold that an event is ‘likely’ to involve the policy where there is a substantial risk that it will do so, even if that substantial risk falls short of 50 per cent. ‘Likely’ in this context could therefore encompass probabilities below that level – occurrences which ‘might well’ or ‘easily could’ involve the policy, or in which one would say there was a ‘good chance’ or a ‘serious danger’ that they might. As with most ordinary language about probabilities and risk, the relevant threshold cannot be reduced to a mathematical probability. 8.30

In summary, therefore, the obligation to give notice under the Bermuda Form arises where a senior executive becomes aware of an occurrence whose consequences are likely (but not necessarily more likely than not) to produce a loss that would, in fact, involve the policy. Nevertheless, the policyholder is entitled to exercise its judgment as to the consequences of any known occurrence, provided it does so honestly, at least within the bounds of what it would be reasonable to conclude from the circumstances as they appear to the policyholder. The policyholder is not obliged to disclose every occurrence that might conceivably produce a loss to the policy. The XL Guidelines

8.31

This common-sense approach is confirmed and illustrated by the fact that less than a year after beginning to sell Bermuda Form liability insurance, and periodically thereafter,63 XL advised its policyholders against giving notice of all claims. Instead, XL wanted to receive notice only when a claim threatened XL’s layer. These ‘Notice Guidelines’ are worth quoting in full: When to Give Notice Notice is required to be given only when the Named Insured (i.e. an employee of the risk management insurance or law department or certain specified officers of the Named Insured) becomes aware of an Occurrence likely to involve X.L.’s policy. It thus is not necessary to give notice of every Occurrence or claim to protect the Insured’s rights under the X.L. Policy where the Insured in good faith determines its exposure to be less than its per Occurrence Retention Amount, although the Insured may at its option give precautionary notice of any

60 979 F2d 268, 276 (2d Cir 1992); Paramount Ins v Rosedale Gardens 743 NYS2d at 63. Contrast the approach taken in English insurance cases, where ‘likely’ has been held to denote at least a 50 per cent chance of a claim being made: see Layher Ltd v Lowe [2000] Lloyd’s Rep IR 510; Jacobs v Coster [2000] Lloyd’s Rep IR 506. 61 Eg, Century Indemnity v Keyspan 15 Misc 3d 1132(A); Tower Ins Co v Lin Hsin Long Co 855 NYS2d 75 (App Div 2008). 62 The policy language at issue in the New York case law is not precisely the same as the Bermuda Form policy, but there does not appear to be any significant difference. The ‘reasonable possibility’ approach has been generally applied in the New York case law without distinctions being drawn between the precise wording of the differently phrased notice clauses. 63 The earliest letter containing Notice Guidelines is dated 5 May 1987 and signed by XL’s then President, Brian M O’Hara. XL wrote similar letters on 10 January 1989, following the introduction of the 002 Form; on 7 February 1990; 6 September 1990, following the introduction of the 003 Form; and on 1 December 1995, following the introduction of the 004 Form.

Notice as a Trigger of Coverage: Legal and Practical Considerations

145

such Occurrence. If notice is not given of such an Occurrence and at a later point it is learned that there is an appreciable risk of impacting X.L.’s coverage, only then is notice required under the Policy. Prompt notice at such point will be timely, and the Insured will not be prejudiced by failure to give an earlier precautionary notice (provided, of course, that Coverage A or Coverage B, if applicable, continues in effect for the same layers). In determining exposure, X.L. requests that Insureds give more weight to the potential damages than to the risk of liability being imposed. That is, X.L. would appreciate notice of Occurrence where, even though the risk of liability is low, the potential damages are high (e.g. a seemingly frivolous class action for Personal Injuries). Of course, the mere fact that a plaintiff’s attorney adds several zeros to the prayer for relief should not be controlling in evaluating exposure. Where the risk of liability is great but the exposure small, e.g. ordinary automobile accidents (other than ‘Integrated Occurrences’), notice need not be given unless and until there develops a specific indication that the exposure is large (e.g. one of the occupants of the car had earnings of $20,000,000 per year and is forever incapable of working again due to serious injuries in the accident). It serves no one’s interest to have Insureds submit notices of Occurrences which have no reasonable probability of impacting X.L. It take time and effort for the Insureds to prepare such notices, for X.L. to review, evaluate and investigate them and for Insureds to compile information for X.L. to assist in this process and keep X.L. apprised of developments. It also makes the underwriting process more difficult and affects the result since the underwriters need to take into account the Insured’s perception that it has claims which may impact X.L.64

NOTICE AS A TRIGGER OF COVERAGE: LEGAL AND PRACTICAL CONSIDERATIONS If the policyholder becomes aware of an occurrence as defined by the policy, consideration will need to be given as to whether to serve notice of occurrence at all. The policyholder may sometimes have no choice but to do so, since the policy requires service of a notice if the occurrence is ‘likely to involve this Policy’. If the policyholder fails to do so, then coverage may be forfeited as discussed above.

8.32

In many cases, however, the policyholder will become aware of an occurrence and may reasonably form the view that the problem will never reach the high-level cover that the Bermuda Form provides. A product liability claim from an individual or a group of individuals does not necessarily become a mass tort. A factory explosion, whilst it may be serious, will not necessarily result in liability that reaches the upper layers of a company’s liability cover.

8.33

Should Notice be Served? A policyholder may decide, in these circumstances, to serve notice in order to avoid any future arguments with the insurance company about the timing of notice and consequent forfeiture of policy rights. Insurance companies who use the Bermuda Form, however, do

64 Letter

from XL Insurance Company Ltd to its policyholders dated 1 December 1995.

8.34

146

Notice of Occurrence

not encourage such purely precautionary notices.65 There are, moreover, disadvantages in taking this safe course, since the service of a notice of occurrence may lock the policyholder into the retention and limits applicable at the time when notice is served. Article II(A), the Limits of Liability provision, of the 004 Form spells out this consequence: provided, however, that for all Insureds the applicable aggregate limit of liability, per Occurrence limit of liability, per Occurrence retention, and the terms, conditions and exclusions of coverage shall be determined under the Policy as in effect at the time notice of the Occurrence or Notice of Integrated Occurrence for which coverage is asserted is first given pursuant to Article V of this Policy by any Insured.66

8.35

Accordingly, there may be sound commercial reasons why a policyholder does not want to serve notice in circumstances where the policyholder thinks that the occurrence (which for illustrative purposes we will call ‘Occurrence A’) will not reach the Bermuda Form layer. The policyholder may wish to wait and see if the problem created by Occurrence A really does threaten the layer. This may be particularly desirable in circumstances where the policyholder has already served notice in a particular year in respect of other occurrences (‘Occurrences B and C’). If Occurrences B and C are likely to exhaust the policy’s aggregate limits for that year, it is disadvantageous to serve notice of Occurrence A, as the policyholder will then likely receive no indemnity in respect of Occurrence A. Another reason for not serving notice may be because the policyholder is in the process of purchasing increased limits when extending the policy, and may want those increased limits to be available for Occurrence A in case it becomes a more significant problem.67 A policyholder in this position would, however, be well advised to discuss the matter with the insurance company, for example, at the time of the annual renewal or when increased limits are being agreed. At the time of renewal, the policyholder will normally complete application forms that will require the identification of occurrences that are likely to result in substantial loss, even if the loss is expected to be below the policy retention. These forms must, of course, be accurately and honestly completed. As a result of what is revealed on renewal, underwriters may sometimes insist that notice be given of such 65 From an early stage, as discussed above, XL issued a series of notice guidelines that were designed to make it clear to policyholders that they need not give notice of any conceivable occurrence that might impact upon the XL policy. 66 In the Limit of Liability provisions in earlier Forms, the effect of giving notice was drafted in different terms: ‘subject to the limit as stated in Item 2(b) of the Declarations in the aggregate for each annual period for all occurrences covered hereunder of which notice is first given during such annual period’. The same result as in the 004 Form was achieved, however, because the Coverage clause (Article I) of the earlier versions provided that ‘the aggregate limit, the per occurrence limit, the per occurrence retention, and the terms, conditions and exclusions of coverage shall be determined under the Policy as in effect at the time notice of the occurrence for which coverage is asserted is first given pursuant to Section (d) of Article V (Conditions) hereof’. 67 Whether this is permissible will depend upon the terms of the policy as a whole. From an early stage in the development of the Bermuda Form, XL issued a complicated endorsement entitled ‘Change of Limit and/or Retention Endorsement’. In broad terms, this was designed to lock the policyholder into the ‘old’ retention and limits applicable to occurrences of which the policyholder was aware at the time that the retention or limits were changed. Accordingly, a policyholder with this endorsement would not gain the advantage of an increased limit by delaying a notice of occurrence. Insurers have sometimes relied upon such endorsements to suggest that the policyholder has no claim at all for occurrences that were known at the time of purchase of the increased retention or limits. In our view, this interpretation is not sustainable and is inconsistent with the language and the obvious intention of such an endorsement. The relevant provision of the endorsement refers to a ‘condition precedent to the rights of any Insured pursuant to paragraph 1 of this endorsement’, and paragraph 1 refers to the increased limits.

Notice as a Trigger of Coverage: Legal and Practical Considerations

147

occurrences, or may in effect force such notice to be given by threatening to exclude the occurrence at renewal. In those circumstances the policyholder has little choice but to serve notice, even if it would otherwise be preferable and permissible to wait. Conversely, the question might arise as to whether a policyholder can serve notice of occurrence in circumstances in which the policyholder considers it unlikely that the Bermuda Form layer will ever attach. In practice, no doubt, some policyholders regularly gave such precautionary notices to avoid the possibility of a late notice defence. What would the position be, however, if the policyholder wished to change insurers and did not want to purchase Coverage B from its current Bermuda Form insurer? Could the policyholder in these circumstances protect its position by serving notice of occurrence in respect of every small claim that has been against it, simply in order to cover the possibility of these claims becoming substantially worse? Under the earlier versions of the Bermuda Form, it was open to doubt whether the policyholder was permitted to do this, as the relevant provision (Article V(d)) referred only to an occurrence or claim ‘likely to involve this Policy’. It was therefore questionable whether the policyholder was able to give a valid notice in respect of an occurrence which was not thought likely to involve the policy. Under the 004 Form, however, Article V(B) permits a ‘permissive notice’ to be served ‘at any time during the Policy Period or Discovery Period’, and it does not appear to be necessary that the occurrence needs to be ‘likely to involve this Policy’. It appears, then, that precautionary notices are valid – albeit that they are discouraged and are not necessarily in the best interests of the policyholder.

8.36

If a permissive notice is served pursuant to Article V(B) at a time when the occurrence is not ‘likely to involve the Policy’, must a further notice be served pursuant to Article V(A) when the Executive Officer becomes aware that the occurrence is ‘likely to involve the Policy’? In other words, could an insurer who had received notice under Article V(B) then take a ‘late notice’ point if no subsequent Article V(A) notice is served? A notice served pursuant to Article V(B) is clearly sufficient to constitute notice as a trigger of coverage; otherwise, the service of such a notice would be a futile exercise. We also think that it is sufficient to avoid any ‘late notice’ defence. The service of a permissive notice enables the insurer to carry out such enquiries as he or she considers necessary, and it is not obvious what useful purpose would be served by requiring a further notice. Since the effect of late notice is so drastic and since Article V does not clearly require two notices, we do not think that a tribunal would accept a late notice argument based upon a failure to serve a second notice.

8.37

Should the Notice Specify an Integrated Occurrence? As well as deciding whether to serve notice of an occurrence at all, the policyholder must decide whether to serve the notice as notice simply of an ‘Occurrence’ or as notice of an ‘Integrated Occurrence’.68 This refers to the complex policy provisions distinguishing 68 ‘Integrated Occurrence’ is often also called a ‘batch occurrence’ after the concept used in common insurance parlance and in earlier versions of the Bermuda Form – eg, Form 003 Article V(d), referring to notice of ‘batch occurrences’.

8.38

148

Notice of Occurrence

between an occurrence and an integrated occurrence. In summary, the effect of these provisions is as follows:69 (a) Where only one person or piece of property is damaged, there is (always) a single ‘ordinary’ occurrence and never an integrated occurrence. This must, however, be the exception rather than the norm in cases that result in liabilities of the magnitude with which the Bermuda Form is concerned, as it is quite unlikely that injury to just one person or item of property would be sufficiently serious to exceed the significant retention of the Bermuda Form policy. (b) Where more than one person or piece of property is damaged by the same event or defect,70 there may be: (i) one ‘ordinary’ occurrence;71 or (ii) one ‘integrated occurrence’; or (iii) a succession of ‘ordinary’ occurrences. Such injury or damage may be described as ‘related’. Unrelated injury and damage never forms part of the same occurrence – integrated or otherwise. (c) There is one ‘ordinary’ occurrence where related injury or damage commences within a single 30-day period. Thus, if there is a plant explosion in which a number of people are injured and various items of property are damaged, there will be a single occurrence, which will not be an integrated occurrence, but which will encompass all the injury and damage attributable to the same event, namely the explosion. This would remain true if, for instance, the explosion resulted in a toxic cloud of gas that settled in the vicinity of the plant for five days before dissipating;72 all relevant injury would still have commenced within 30 days. It is not necessary for such an event to be notified as an ‘integrated’ occurrence in order to aggregate all the losses for the purpose of applying the retention and limits of the policy. Aggregation is automatic. (d) Related injury which commences over a period of more than 30 days does not automatically form one single occurrence: it may be either one ‘integrated occurrence’ or many ‘single occurrences’, depending on whether the policyholder chooses to give notice of all of it as an ‘integrated occurrence’ or separate notices for each injury. The choice will typically arise in products liability situations. XL’s Notice Guidelines use the example of a truckload of contaminated frozen hamburgers consumed over a three-month period and causing two wrongful death claims more than 30 days apart. More typical still will be the product that is marketed for several years, causing (or allegedly causing) injury throughout the period, and possibly thereafter. 8.39

It will normally be in the best interests of the policyholder to have an integrated occurrence, given the high excess levels at which the Bermuda Form is used, as the policyholder normally will be able to exceed the retention only by combining the injury and damage.

69 For further discussion, see above ch 6. 70 Strictly, where injury or damage is ‘attributable directly, indirectly or allegedly to the same actual or alleged event. condition, cause, defect, hazard and/or failure to warn of such …’: Form 004, Article III(R) and (V). See generally ch 6 above. 71 The reason that there is only one occurrence in this situation is because of the ‘30-day period’ provision discussed below. 72 The example is drawn from XL’s 1995 Guidelines to policyholders on giving notice of integrated occurrences.

Notice as a Trigger of Coverage: Legal and Practical Considerations

149

If the injury or damage is not aggregated in an integrated occurrence, the injury to each individual and item of property must be treated as a separate occurrence, for which the retention must be separately exhausted. It is therefore normally in the best interests of the policyholder that notice not merely of an occurrence, but also of an integrated occurrence, is given where injury begins over a period of more than 30 days. Although it is theoretically possible for the policyholder to benefit from treating such occurrences separately, this probably would be so only in very unusual circumstances. The policyholder has what is described as an ‘option’ to give notice of an integrated occurrence or not. The limits of this ‘option’ should, however, be understood. First, it is not permissible to give notice of an integrated occurrence unless the injuries or damages in question are related. Suppose, for example, that a policyholder with a retention of US$75 million suffered one relatively minor plant explosion on 1 May, with a resulting loss of US$50 million, and another unrelated explosion on 3 June, with a resulting loss of US$50 million. No doubt, various persons suffered injury ‘commencing’ more than 30 days apart. However, because the injury does not result from the same ‘event’, it is not possible to treat the two explosions as a single integrated occurrence. An integrated occurrence is an accumulation of losses that both meet the substantive criteria for integration laid down in Article III(R) and meet the procedural criterion that notice of the occurrence as such has been duly given.

8.40

Second, it is not necessary or permissible to treat as more than one occurrence related losses that commence in a single 30-day period. In such a case aggregation is automatic, no notice of integrated occurrence is needed, and the fact that no such notice is given will not avoid the losses being aggregated.

8.41

Third, although the policyholder has a choice of whether to give notice of an integrated occurrence or not, the policy provides no intermediate choice. The policyholder can decide either to have all the injuries integrated or to have each injury treated separately, but it cannot elect to integrate some related occurrences but not others. Suppose, for example, that a product causes injury over a 10-year period from 1990 to 2000. One could not declare, in 2003, an integrated occurrence consisting of the injury from 1990 to 1997 and a separate integrated occurrence, in 2004, consisting of injury from 1998 to 2000, thereby attempting to obtain two separate limits. One must either integrate all related injury or damage completely, or not at all.

8.42

Sometimes the policyholder originally gives notice of a single occurrence (or perhaps several occurrences) and may only later realise that there is actually an integrated occurrence. For instance, the policyholder may give notice of a single large, but apparently isolated, product liability claim, which turns out to be the first of many. When that happens, the policyholder may give notice of the integrated occurrence, which will then be treated as including the earlier occurrence. Rather complex provisions then apply to determine the applicable limits and retentions.73

8.43

The provisions contained in the 004 Form relating to notice of integrated occurrence have no direct counterpart in the earlier Forms, but the origins of the current provision can

8.44

73 See

Article II(B) of the 004 Form; and ch 9 below.

150

Notice of Occurrence

clearly be seen in those Forms. The original Bermuda Form (001) simply required notice of an ‘occurrence likely to involve this Policy’ and also notice of any claim74 ‘likely to involve this Policy’. The 002 Form required a notice of occurrence or claim to identify ‘each occurrence in respect of personal injury and/or property damage from one or more units of a product as an individual occurrence or a batch occurrence’.75 The 003 Form included a definition of ‘integrated occurrence’ and also incorporated, in a somewhat cumbersome way, a lengthy definition of ‘batch occurrence’ within the notice provision.76 There was, however, no objective time limit similar to the 30-day limit set out in the 004 Form.77

74 The 004 Form contains no requirement to notify a claim, as distinct from an occurrence. In practice there appears to be a very substantial overlap between the two concepts, which perhaps explains why notice of a claim is no longer treated separately. 75 Article V(d). 76 The 002 Form included a similar definition within Article III, which contained the policy definitions. 77 Indeed, the last sentence of Article V(d) of the 003 Form is puzzling. It could, on one view, give the policyholder a complete discretion as to whether or not to include particular claims within a ‘batch occurrence’.

9 Article II: The Excess Point and Limits of Liability

Article II of the Bermuda Form, the ‘Limits of Liability’ clause, is principally concerned with two matters. First, the Article deals with the point at which coverage attaches. We shall describe this as the ‘excess point’, although people sometimes call it the ‘attachment point’ or the ‘retention’.1 Second, the Article sets out the amount of coverage provided. The Article is therefore a fundamental part of the structure of the Bermuda Form, and links with a number of ‘Declarations’ contained on the first page of the Form. Article II has been subject to redrafting through the different versions of the Form, but its substance has remained much the same. It is also linked with a number of other clauses in the Bermuda Form, and it is convenient to deal in this chapter with the Other Insurance Condition contained in all versions of the Form.2

9.01

THE EXCESS POINT The first part of Article II(A) deals with the excess point. In the 004 Form,3 the clause is as follows: Regardless of the number of Insureds under this Policy, for each layer of coverage set forth in Item 2 (a) of the Declarations the Company shall be liable only for that amount of Ultimate Net Loss for each Occurrence covered under this Policy which is in excess of the greater of: (1) the amounts indicated as the limits (including, without limitation, any reinstatements thereof, where applicable) of the underlying insurances and any self-insured retentions listed, or which should have been listed, on the present or any prior Schedule B annexed to this Policy and any other underlying insurance, as to which the Company and the Named Insured expressly agree that the insurance provided by this Policy shall: (a) be in excess in respect of such Occurrences or Claims and Ultimate Net Loss as are covered by said underlying insurances (it being understood that this Policy shall in no 1 We use the expression ‘excess point’ rather than ‘retention’, because the policy itself uses the expression ‘per occurrence retention’ but this is not necessarily (as explained below) the excess point. It is a minimum excess point. 2 Article VI(H) of the 004 Form; and Article V(i) of earlier Forms. Earlier Forms (001 and 002) also contained an ‘other insurance’ exclusion, which gave rise to difficulties in interpretation. This exclusion is discussed in Chapter 9 of the previous editions of this book. 3 See Article II(a) of earlier Forms.

9.02

152

Article II: The Excess Point and Limits of Liability way be subject to, or affected by, the terms, conditions or exclusions of said underlying insurances); and (b) apply only as if such underlying insurances were fully available (except to the extent that any aggregate limits thereof are reduced or exhausted by actual payment of claims) for all occurrences or claims covered thereunder or

(2) the per Occurrence retention amount listed in Item 2 of the Declarations (which may be satisfied only by Ultimate Net Loss as defined herein).

9.03

The commercial background to this provision is that the Bermuda Form policy operates as excess insurance.4 Unless the policyholder is prepared to assume a large self-insured retention, it is likely to have an insurance programme covering levels below the cover provided by the Bermuda Form excess insurer. The application form5 of the Bermuda Form insurer will generally require the policyholder to provide details of the insurance programme underneath the Bermuda Form policy, and the details of this programme will find its way into Schedule B thereto. Schedule B is referred to in the Limits of Liability Article and elsewhere in the Form.

9.04

The precise nature of the policyholder’s underlying coverage will depend upon market conditions at the relevant time. When use of the Bermuda Form started in the mid-1980s, a typical policyholder might purchase a low-level primary insurance policy, covering, say, US$1 million excess of US$1 million. At that time, primary covers usually covered defence costs (however large) in addition to limits of liability which were exhausted solely by payments of indemnity, so that the primary insurer would continue to pay defence costs until, in this example, the US$1 million in indemnity was paid. Above the primary layer,

4 See, eg, the ‘Other Insurance’ Condition (Article VI(H) of the 004 Form, and Article V(i) of earlier Forms) discussed later in this chapter. See also the Subrogation Condition (Article VI(I) of the 004 Form, and Article V(j) of earlier Forms) discussed in ch 11, paras 11.59–11.63 below. Article II itself refers, in (1)(a), to the Bermuda Form policy being ‘in excess’ of the other insurances referred to earlier in the clause. 5 From the time when the Bermuda Form was first introduced, application forms required the policyholder to produce a list of ‘Schedule B’ insurances. For example, the 1986 XL application form included the following question: List on Schedule B (to be attached to and form a part of the policy when issued) all insurances of the Applicant and any joint ventures to be insured. If additional insurance is secured at any time prior to the expiration of this policy, Schedule B must be updated at or prior to the effective date of such additional insurance. In 1987, XL asked the policyholder: List on Schedule B (to be attached hereon and form a part of the policy when issued) all insurances, including primary, underlying excess layers and limits excess of requested XL coverage, in force for the Applicant, all entities listed on Schedule A and any joint ventures to be insured. Be sure to include type of coverage, coverage dates, carriers, premiums and any self insurance. Subsequent to binding it is the Insured’s obligation to update Schedule B in the event of any changes in excess of the Per Occurrence Retention Amount. In its 1988–90 anniversary application form, XL asked: Update Schedule B underlying and other insurances (to be attached to and form a part of the X.L. policy) of the insured and any insured joint ventures. Please provide premiums and details of specific exclusions. Similarly, a ‘Bermuda Market Excess Liability Application’ form used in recent years states: Complete the enclosed Schedule B (to be attached to and form part of the policy when issued) listing all insurances including primary and all excess layers in force for the Applicants. Subsequent to binding, it is the Insured’s obligation to update Schedule B in the event of any changes.

The Excess Point

153

a policyholder might purchase a number of excess layers, principally from US or UK insurers. The upper layers might well be covered by underwriters writing on the Bermuda Form, with XL writing a layer below that written by ACE. Some 35 years later, market and economic conditions have changed. Policyholders may be willing or find it necessary to keep a greater self-insured retention at the ‘bottom end’, but may also be able to purchase greater vertical cover from a variety of insurance companies. The Per-occurrence Retention In broad terms, Article II provides as follows in relation to the ‘excess’ point. Article II(A)(2) makes the insurer liable only for such portion of the policyholder’s ‘Ultimate Net Loss’ as exceeds the per-occurrence retention listed in item 2 of the Declarations. This is a fixed monetary sum and represents the minimum excess point under the policy. For example, when XL started to write business in 1986, it specified a minimum excess point of US$50 million, although this was soon reduced to US$25 million.

9.05

The term ‘Ultimate Net Loss’ is defined in broad terms in Article III(AD):

9.06

‘Ultimate Net Loss’ means the total sum which the Insured shall become obligated to pay for Damages on account of Personal Injury, Property Damage and/or Advertising Liability which is, or but for the amount thereof would be, covered under this Policy less any salvages or recoveries.

A policyholder’s claim under a Bermuda Form policy starts with a computation of its Ultimate Net Loss. That concept is also critical in determining whether the per-occurrence retention has been satisfied. A number of features of the definition of Ultimate Net Loss are important in the present context. First, the concept of Ultimate Net Loss requires reference to the definition of ‘Damages’ in Article III(G). This covers judgments, settlements and defence costs. Accordingly, the total amounts which the Insured ‘shall become obligated to pay’ in respect of these matters comprise the policyholder’s Ultimate Net Loss. The definition of Ultimate Net Loss does not itself require payment to have been made in respect of these obligations. However, Article VI(F), the Loss Payable Condition, provides that liability under the policy only attaches when payments up to the excess point have been made. Policyholders are therefore likely to compute their Ultimate Net Loss by reference to payments actually made in respect of judgments, settlements and defence costs. These payments need not be made by the policyholder itself; it is sufficient if they have been made by its insurers.6

9.07

Second, the policyholder’s Ultimate Net Loss can only include sums which are covered under the Bermuda Form policy, or which would be covered under the policy ‘but for the

9.08

6 Article VI(F), the Loss Payable Condition, provides that liability under the policy shall not attach unless and until ‘the Insured’s underlying insurer(s) or the Insured’ shall have paid the greater of the relevant amounts (emphasis added). The definition of ‘Defense Costs’ in Article III(H) refers to legal costs and other expenses ‘incurred by or on behalf of the Insured’ (emphasis added). The latter provision would seem to encompass ‘Defense Costs’ incurred by an underlying insurer in connection with the policyholder’s defence of claims.

154

Article II: The Excess Point and Limits of Liability

amount thereof’. If, therefore, there is no coverage under the policy for some part of the judgments or settlements or defence costs, then those must be excluded from the calculation. For example, if the policyholder has made payments in excess of the figure identified on the Declarations page as the per-occurrence retention, but some of these would be within the ‘maintenance deductible’ pursuant to the expected or intended provisions of the occurrence definition, these payments will not count towards its Ultimate Net Loss. However, the computation of Ultimate Net Loss necessarily disregards the fact that some payments for judgments, settlement or defence costs fall below the excess point of the policy, since otherwise the excess point could never be reached. It is therefore sufficient that sums would be covered ‘but for the amount thereof’. 9.09

Third, the computation of Ultimate Net Loss looks at the position of the ‘Insured’. Under Article III(P), the Insured does not comprise simply the single entity identified as the Named Insured. It will include certain subsidiaries, as well as officers and others when acting within the scope of their duties. When there is large scale litigation in the United States, it is not unusual for the underlying claimants to sue a variety of corporate and individual defendants, each of whom qualifies as an Insured under Article III(P). Where an occurrence gives rise to payments or on behalf of a number of covered entities and individuals, it is not necessary to distinguish between them for the purposes of calculating the Ultimate Net Loss of the ‘Insured’. For example, when costs are incurred in defence of such claims, the policy and in particular the definition of ‘Defense Costs’ in Article III(H) do not require distinctions to be drawn between costs incurred ‘by or on behalf of’ the various Insureds. All covered payments can therefore be aggregated in order to calculate the Ultimate Net Loss.

9.10

The above principles as to the calculation of Ultimate Net Loss apply equally to the calculation of the Insured’s per-occurrence retention, which ‘may be satisfied only by Ultimate Net Loss as defined herein’. Thus, the per-occurrence retention is also computed by aggregating covered payments without distinguishing between each individual Insured. The per-occurrence retention does not therefore apply separately for each entity or individual covered by the policy.7

9.11

In summary, the per-occurrence retention will be satisfied where the total sum which the policyholder or its insurers have paid exceeds the amount identified in item 2 of the Declarations, provided that this total sum includes only amounts which are in principle covered by the policy. The definition of ‘Ultimate Net Loss’ does not distinguish between particular Insureds, and therefore the policyholder or Insureds may aggregate amounts of defence costs and indemnity to demonstrate that the excess point has been reached.

9.12

The per-occurrence retention referred to in Article II(A)(II) is, however, a minimum retention, because the opening words of Article II specify that liability is only for sums that exceed ‘the greater of’ the amounts identified in Article II(A)(1) and (2).8 This concept is

7 The opening words of Article II(A) are: ‘Regardless of the number of Insured’s under this Policy.’ Whilst these words are primarily directed at the limits of the policy rather than the excess point, they are also apposite in the context of the excess point. 8 The position is the same under Article II of earlier Forms, which referred to ‘the greater of’ either the underlying limits or the per-occurrence retention.

The Excess Point

155

repeated in the Loss Payable Condition, which refers to the policyholder or its underlying insurers having paid ‘the greater of the amount of any applicable underlying limits or the applicable retention set forth in Item 2 of the Declarations’. The relevant comparison is therefore between the per-occurrence retention set out in Article II(A)(2), and the limits of the underlying insurances identified in the lengthy and rather complex wording of Article II(A)(1).

9.13

The Underlying Insurance Limits The underlying insurances which are relevant for the purposes of Article II(A)(1) are identified in very broad terms. First, there are the policies listed on ‘the present or any prior Schedule B annexed to this Policy’. Schedule B accordingly lists a series of (known) insurance policies that are regarded as ‘underlying’. This list of policies in Schedule B will be based upon the information provided by the policyholder in response to questions asked in the application form as to underlying insurance, or in response to instructions on a draft Form B provided as part of the application process. Second, there are the policies which ‘should have been listed’ in Schedule B.9 If the policyholder’s list has omitted, inadvertently or otherwise, a policy which should have been listed in response to questions asked on the application form or in the instructions on the draft Form B, then it is nevertheless regarded as underlying and relevant. Third, there are the policies which comprise ‘any other underlying insurance’. It might be thought that this third category is unlikely to add materially to the prior categories, namely the policies which are listed or should have been listed ‘on the present or any prior Schedule B’. It does, however, render academic any argument, for example, that the insurer’s questions did not require a particular underlying policy to be identified by policyholder in Schedule B. Provided that the policy comprises ‘any other underlying insurance’, it counts. It would therefore potentially cover insurance which was put in place subsequent to the conclusion of the Bermuda Form policy.

9.14

The effect of Article II (A), therefore, is that, until the limits of these various policies and the retention have been exhausted, the Bermuda Form policy does not respond. Article II(A)(1) will, however, come into play only if, for some reason, the cover provided by the underlying policies exceeds the minimum fixed excess point under Article II(A)(2).

9.15

Accordingly, unlike many other excess insurance policies, the Bermuda Form policy does not simply operate with a fixed excess point. It is best regarded as ‘floating’ above a fixed minimum excess point. The fixed minimum is the per-occurrence retention that is set out in the Declarations part of the policy. But if the policyholder’s underlying programme provides cover that is greater than the per-occurrence retention, then this inures to the benefit of the Bermuda Form insurer, because the excess point is automatically raised above the per-occurrence retention.

9.16

9 Form 004 was the first time that Article II stated that the underlying insurances included not only those actually listed on Schedule B, but also those which should have been listed. For the reasons explained below, this probably does not effect a substantive change to the earlier Forms, which sought to achieve the same result through the Other Insurance Condition (Article V(i)).

156

Article II: The Excess Point and Limits of Liability

9.17

It might be thought unlikely that a situation would arise where, because of underlying insurance limits, the excess point was in fact greater than the agreed per-occurrence retention. If a policyholder were prepared to buy cover under the Bermuda Form policy attaching at (for example) US$25 million, one would not normally expect that policyholder also to have purchased underlying cover of US$30 million; the company would be wasting its premium dollars in buying double coverage. The Bermuda Form policy is, however, used by large corporations with numerous subsidiaries. It can happen that a particular subsidiary (subsidiary A) will have put in place its own insurance programme and will have purchased more vertical cover than the parent company and other subsidiaries have purchased. In this scenario, the parent company often will wish the Bermuda Form policy to attach at the point above its overall programme rather than at the point above the limit of the cover that subsidiary A has happened to purchase.

9.18

Article II(A)(1) also may be relevant and come into play where the policyholder has a coherent programme of coverage, with no apparent overlaps, but its primary insurance covers defence costs in addition to the limits of liability (which are only exhausted by payments of indemnity). If substantial defence costs are incurred and paid by the primary insurers, it is possible that the per-occurrence retention in Article II(A)(2) would be exceeded, but that there would still be some unexhausted limits so that Article II(A)(1) was not exceeded. In these circumstances, the policyholder would not be able to recover under the Bermuda Form policy until the unexhausted limits of the underlying coverage had been exhausted completely. Accordingly, the Limits of Liability provision has the effect of preventing a policyholder from ‘double dipping’; that is, claiming against two insurers for in essence the same loss. It also seeks to prevent the underlying insurer from making a claim for contribution against the Bermuda Form insurer, since there is no overlap of coverage.

9.19

In some cases, some or all of the policyholder’s underlying programme may be provided by a captive insurer. For the purposes of the operation of Article II(A)(1), there appears to be no reason to distinguish between underlying insurance provided by a captive and underlying insurance provided by an external insurer, notwithstanding that the captive may itself qualify as an Insured under the definition in Article III(P). Indeed, the captive element of the policyholder’s programme will generally, and indeed should, be listed in Schedule B. Where the captive provides cover for defence costs in addition to policy limits, the policyholder may therefore find, for the reasons discussed in the previous paragraph, that the excess point of the Bermuda Form policy is in practice higher than the per-occurrence retention listed on the Declarations page. This may come as something of a surprise to the policyholder, who may have assumed that the per-occurrence retention would apply to its business as a whole. If this result is to be avoided, the potential issue should be addressed at the time that the programme is put in place, for example by limiting the defence costs cover provided by the captive.

9.20

Article II(A)(1) refers to the ‘present or any prior Schedule B’ and therefore contemplates that the underlying policies may have been purchased over a number of years. However, the application of Article II(A)(I) clearly does not involve simply adding up the total limits of the underlying policies over a number of years. Otherwise, the excess point would automatically grow every year that the Bermuda Form coverage was in place.

The Excess Point

157

Rather, Article II(A)(1)(a) means that the relevant question is the extent to which those underlying insurances provide cover in respect of the events which have given rise to the policyholder’s loss. Hence, it is necessary to ask whether and what to what extent ‘such Occurrences, or Claims and Ultimate Net Loss’ are ‘covered by said underlying insurances’. Depending upon the nature of those events, it may be relevant to look at underlying policies which were in place over more than one year. This question as to the scope of coverage of the underlying insurance requires reference not only to the policyholder’s ‘Ultimate Net Loss’, but also ‘Occurrences or Claims’. It seems that this may be addressing the situation where the underlying coverage is broader than the Bermuda Form coverage, for example because an exclusion in the Bermuda Form coverage is not present in the underlying insurance. In such a case, the underlying coverage would not provide cover in respect of the policyholder’s ‘Ultimate Net Loss’ because that is defined by reference to the cover provided by the Bermuda Form policy. However, it might provide coverage in respect of ‘Occurrences’ or ‘Claims’, in which case it will be relevant underlying coverage for the purposes of Article II(A)(1).

9.21

If, therefore, the underlying insurances do not in fact cover the relevant occurrences, then those insurances will not be relevant to the ascertainment of the excess point under Article II(A)(1), and the fixed per-occurrence retention amount in Article II(A)(II) governs. Thus, if subsidiary A has underlying cover of US$100 million and the per-occurrence retention under the Bermuda Form (covering all subsidiaries) is US$75 million, the coverage purchased by subsidiary A will not be relevant to a claim by subsidiary B concerning an occurrence for which the US$100 million programme provides no protection. Similarly, if the underlying insurance provides for cover of US$75 million for personal injury claims and US$100 million for property damage, the property damage cover would not be relevant (to a claim under the Bermuda Form with a per-occurrence retention of US$75 million) if the claims against the policyholder were for personal injury alone.

9.22

What happens if an underlying insurer becomes insolvent and therefore fails to pay the underlying limit? The position appears to be that this does not accelerate the point at which the Bermuda Form insurer has to pay.10 The focus of Article II(A)(1) is the unexhausted limits of the underlying insurance policies, and the Bermuda Form policy attaches in excess of those limits irrespective of whether the underlying insurers have paid.11

9.23

10 This issue was litigated extensively in the United States during the late 1980s and 1990s in the wake of the insolvencies of such large commercial liability insurers in the mid-1980s as Mission Insurance Company Transit Casualty Company and Midland Insurance Company in the United States, and various London Market insurance companies, including the so-called ‘KWELM’ companies. Litigation of the issue of whether an excess insurer was obligated to ‘drop down’ in the place of an underlying insolvent insurer turned on the policy language in the limits of liability provisions at issue. For example, courts found that an excess policy that attached in excess of ‘recoverable’ insurance was obligated, in effect, to drop down in place of the underlying insolvent policy. In contrast, courts typically refused to reach that conclusion if the excess policy attached in excess of underlying ‘collectible’ insurance. For a discussion of this issue and the US law on this issue, see Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1996 and 2021 Supp) ch 13 (hereafter Masters and Stanzler, Insurance Coverage Litigation). 11 Article II(A)(1)(b) of the 004 Form. See also Article II of the 001 Form, which referred to the ‘unexhausted limits’ of the underlying insurances. In the 002 Form, Article II was redrafted so as to refer to ‘the limits (to the extent not exhausted by actual payment of claims) of the underlying insurances listed on the present and/or any prior Schedule B (hereinafter called “the underlying limits”), as to which the Company and the Named Insured

158

Article II: The Excess Point and Limits of Liability

The opening words of Article II(A)(1) refer to ‘the amounts indicated as the limits’ of the underlying insurances: this denotes the limits as stated rather than as paid. If the underlying insurers do not pay, then the policyholder will itself have to pay in order to for the policy to attach.12 Accordingly, in the case where the underlying limits do not exceed the per-occurrence retention (for example, where the policyholder’s Bermuda Form coverage is excess of US$25 million, and the underlying programme provides coverage up to US$25 million), the policyholder has to pay any shortfall in the underlying US$25 million coverage in the event of an insolvency of an underlying insurer. This seems appropriate, because the Bermuda Form policy in this example was never intended to pay before the underlying US$25 million had been paid. If the policyholder had bought no underlying insurance at all, then it would have had to bear this amount in any event. 9.24

What is the position, however, when there is an insolvent underlying insurer in circumstances where the policyholder happens to have underlying coverage greater than the minimum excess point? It would appear that, even in this situation, the policyholder would have to make up the shortfall, up to the limits of the underlying coverage, in order to reach the excess point. The policyholder cannot, therefore, simply rely upon the peroccurrence retention. Accordingly, a policyholder who has bought underlying insurance in excess of the fixed minimum retention appears to be in a worse position – in the event of an insolvency – than the policyholder who has bought no underlying insurance at all or has bought coverage only below the fixed minimum retention. The rationale is probably that the underlying limits form part of the information presented to the Bermuda Form insurer at the time that the risk is written, and they may therefore have affected the rating of the policy; that is, the Bermuda Form insurer might charge less to a policyholder with a more extensive underlying programme, since it will expect to be further away from the loss.

9.25

Where there is an insolvency, the question is how the policy of the insolvent insurer would have responded ‘as if such insurances were fully available and collectable’. There may be a straightforward answer to this question where the relevant underlying policy provides for payment of a fixed limit. It may be more complicated where, for example, the insurer’s policy provided cover for defence costs in addition to the policy limit. In such cases, it would be necessary to work out, on an ‘as if’ basis, when the policy limit would have been exhausted by payments of indemnity, thereby putting an end to the defence costs obligation.

9.26

One potentially important qualification in respect of the availability and collectability of the underlying insurance is contained in the bracketed words in Article II(A)(1)(b): ‘except to the extent that any aggregate limits thereof are reduced or exhausted by actual expressly agree that the insurance provided by this Policy … (B) apply only as if such underlying insurances were fully available and collectable for all occurrences covered thereunder’ (emphasis added). XL explained the change in a letter to policyholders dated 29 September 1988, stating that the change ‘clarifies X.L.’s difference in conditions (DIC) drop down coverage feature. The revised language more clearly states that it is not X.L.’s intent to drop down due to the financial insolvency of, or uncollectibility from, Insurers listed on Schedule B, underlying insurances. (This is only an issue if the limits scheduled are greater than the per-occurrence retention on the declaration sheet, or specified by endorsement)’. 12 See also the Loss Payable Condition (Article VI(F) of the 004 Form, and Article V(g) of earlier Forms) discussed in ch 11 below, para 11.31.

The Other Insurance Condition

159

payment of claims’. Although a policyholder may in principle have underlying coverage for the events which have given rise to the claim on the Bermuda Form coverage, such coverage may not in practice be available because the policyholder has made other claims on unrelated matters which have had the effect of reducing or exhausting the aggregate limits provided by the underlying cover. Article II(A)(1)(b) provides that, in these circumstances, the underlying coverage is not to be regarded as being fully available and collectable. This means that where there is underlying coverage greater than the per-occurrence retention, the underlying coverage can be eroded (and therefore ceasing to be relevant to the calculation of the excess point) by claims which are not covered by the Bermuda Form insurer. This may result in the per-occurrence retention becoming the relevant minimum retention even though the underlying policy limits were at one stage greater. One issue which has been litigated in insurance in the United States relates to the situation where the policyholder settles with a primary insurer for a sum which is lower than the limit of the policy. Certain policy terms enable an excess insurer to argue that this precludes any claim on the excess policy.13 Such arguments do not, however, arise in the context of the Bermuda Form wording. If the policyholder settles with the underlying insurers below limits, this has no impact on the excess point or the policyholder’s claim: the minimum per-occurrence retention is obviously unaffected, and the limits of the underlying insurance are considered on the basis as if they were they were fully available and collectible.

9.27

THE OTHER INSURANCE CONDITION The provisions of Article II serve a purpose additional to preventing a policyholder from double dipping. By making it clear that the Bermuda Form policy sits on top of other insurance, the policy is intended to prevent other insurers from making claims for contribution. This was, historically at least, an important matter to insurers writing on the Bermuda Form. Claims by other insurers for contribution give rise to a particular problem for a Bermuda Form insurer, because it may find itself being sued in court and without any recourse to arbitration. Potentially, this would result in the Form coming under the scrutiny of courts in the United States,14 something that the Bermuda Form insurers were anxious to avoid.

9.28

It is perhaps for this reason that another provision reiterates the point that the Bermuda Form sits above other insurance cover. The Other Insurance Condition, Article VI(H)

9.29

13 For example, where the excess policy states that underlying limits are exhausted only by payments by underlying insurers and not from any source. See Masters and Stanzler, Insurance Coverage Litigation (above n 10) s 13.08 [C]; Zeig v Massachusetts Bonding & Ins Co 23 F 2d 665 (2d Cir 1928); compare Ali v Federal Ins Co 719 F3d 83 (2d Cir 2013) (interpreting Zeig to apply only to first-party property insurance, not liability insurance). 14 The insurer seeking contribution would need to argue that the Bermuda Form insurer was liable to the policyholder. If this were disputed by the Bermuda Form insurer, the issue of its liability to the policyholder under the Bermuda Form would need to be determined by the court.

160

Article II: The Excess Point and Limits of Liability

of the 004 Form,15 provides that if there are other ‘valid and collectible’ insurances16 – whenever issued – available to respond to a particular loss, then the Bermuda Form policy ‘shall be in excess of and shall not contribute with such other insurance’. The relevant clause is as follows: If other valid and collectible insurance with any other insurer, whether issued prior hereto, simultaneously herewith or subsequent hereto, is available to the Insured for Ultimate Net Loss covered by this Policy, other than insurance which is expressly and specifically excess of the limits of, or quota share on the same layer as, this Policy, the insurance afforded by this Policy shall be in excess of and shall not contribute with such other insurance. Nothing herein shall be construed to make this Policy subject to the terms, conditions or limitations of other insurance. If this Policy shall be deemed or required to contribute to Ultimate Net Loss with other insurance and such contribution arises in whole or in part from the failure of the Named Insured to list such other insurance on Schedule B hereto in accordance with the instructions for such Schedule B, then the Named Insured shall indemnify the Company for the amount of any such contribution, and this Policy shall apply as if such other insurance had been so listed.

9.30

The wording of this and equivalent provisions in earlier Forms indicates that if there are other relevant insurances, including insurances not listed on Schedule B, the excess point for the policy in effect ‘moves up’ to make the policy excess of those other insurances. This appears to be intended to have two effects. First, it seeks to prevent claims under the Bermuda Form policy for any loss covered by other ‘valid and collectible’ insurance until that insurance has been exhausted. Second, the other insurers cannot make a claim against the Bermuda Form insurer for contribution. The policy states what is intended to happen if this second objective fails to be achieved – viz that the policyholder has to indemnify the insurer for the amount of any contribution that the insurer has had to make in respect of a policy that should have been listed on Schedule B, and that the policy provides coverage as if such other insurance had been listed. This may give rise to a question as to which policies should be listed on Schedule B. This is likely to be resolved by reference to the questions asked on the application form for the Bermuda Form policy, or by the instructions for completion of Schedule B which may be contained on the draft Form B provided as part of the application process and then included in the completed form which is scheduled to the policy when issued.

9.31

There is scope for debate as to the extent to which these objectives, and in particular the prevention of claims for contribution by other insurers, are in practice achieved. While the original argument for such clauses was to prevent a policyholder from receiving a double recovery, case law in the United States provides support for the argument that today other insurance clauses provide a mechanism for allocation of liability among insurers when more than one insurance policy applies to cover a claim, and do not apply to reduce an insurer’s obligations to the policyholder.17 Courts have held that these clauses apply 15 Article V(i) in earlier Forms. 16 There is an exception in Form 001 for insurance issued by ACE. This appears to have reflected the fact that, when XL started, it anticipated that its layers would be below those of ACE. In the 004 Form, the relevant clause (Article VI(H)) is expressed more generally and refers to ‘other than insurance which is expressly and specifically excess of the limits of, or quota share on the same layer as, this Policy’. See the discussion in the text below. 17 See analyses of US law on ‘other insurance clauses’ in Douglas R Richmond, Bad Faith Litigation: ‘Other Insurance’ Provisions Cutting Through the ‘Impenetrable Thicket’ of ‘Other Insurance’, New Appleman on

The Other Insurance Condition

161

only to apportion coverage among insurance policies that provide the same kind of, or ‘concurrent’, insurance, meaning insurance that is concurrent both as to the time period insured and the risks insured too.18 The Other Insurance Condition also seeks to counteract any attempt by other insurers to claim, by drafting a general provision, that the Bermuda Form policy sits below or parallel with their policy. In Bermuda Form 001 used by XL, the only exception to the Other Insurance Condition was ‘insurance provided by A.C.E. Insurance Company Ltd or, if applicable, as provided in the Coverage Gaps Endorsement’. The reference to ACE was presumably made in order to avoid any suggestion that the XL policy sat above the ACE policy, in circumstances where it was understood that ACE would be writing layers excess of XL. In the 004 Form, the provision has been broadened, but only so that the concept of ‘other insurance’ (to which the Bermuda Form policy is excess) excludes only ‘insurance which is expressly and specifically excess of the limits of, or quota share19 on the same layer as, this Policy’. Accordingly, any policy that is ‘expressly and specifically’20 excess of the Bermuda Form policy is (obviously enough) not treated as though it was beneath the policy. Similarly, if the Bermuda Form insurer writing on the 004 Form had subscribed 50 per cent of a layer, its 50 per cent share would not be treated as excess of another 50 per cent of the same layer written by another insurer.

9.32

The anxiety to prevent contribution claims by other insurers, with the consequent danger of determinations by the US courts, is also apparent in the final paragraph of the arbitration condition.21 This imposes an obligation on the policyholder to assist in obtaining the dismissal of any such claims, including an obligation to reduce any judgment or award against such other insurers ‘to the extent that the court or tribunal determines that the Company would have been liable to such insurers for indemnity or contribution pursuant to this Policy’. Accordingly, the policyholder is required, if necessary, to reduce its claim against its other insurers.22

9.33

Insurance: Current Critical Issues in Insurance Law 69 (December 2007); Susan Randall, ‘Coordinating Liability Insurance’ (1995) 1995 Wis L Rev 1339, 1353 fn 48 (‘other insurance’ clauses including in insurance policies because there is no other contractual vehicle to define the mechanism for allocating liability among insurers); Douglas R Richmond, ‘Issues and Problems in “Other Insurance” Clauses’ (1995) 22 Pepp L Rev 1373, 1301–81 (‘Inter-insurer loss allocation by way of “other insurance” clauses never permits allocation of a loss to the insured. Payment of the insured’s claim always takes priority over the allocation of the loss between concurrent insurers’); Masters and Stanzler, Insurance Coverage Litigation (above n 10) s 19.03 (discussing case law applicable to ‘other insurance clauses’ and the various kinds of ‘other insurance clauses’ included in liability insurance policies). Eg, In re Liquidation of Midland Ins Co 269 AD2d 50, 64 (NY App Div 2000) (applying ‘other insurance clauses’ to allocate liability among insurers only); Slabic v Hendrickson 556 NYS 2d 236, 237 (Sup Ct 1990) (the same in insurance company versus insurance company dispute). 18 Eg, St Paul Fire & Marine Ins Co v Vigilant Ins Co 919 F 2d 235, 241 (4th Cir 1990); Carter Wallace Inc v Admiral Ins Co 712 A 2d 1116 (1994). 19 The expression ‘quota share’ is not being used in the sense in which it is typically used in the reinsurance market. The expression seems simply to refer to the subscription of a line on the same layer. 20 See the words of Article VI(H) of the 004 Form: ‘other than insurance which is expressly and specifically excess of the limits of this Policy’. 21 Article VI(N) of the 004 Form; and Article V(o) of earlier Forms. 22 See also the Cancellation Condition, Article VI(L)(1)(d) and VI (L)(3) of the 004 Form. US courts have upheld jurisdiction under ‘long arm statutes’ for third-party contribution actions brought by other insurers (but not insured). See, eg, Tortuga Cas Co Nat’l Union Fire Ins Co of Pittsburgh, Pa 604 A2d 419 (Del 1991) (unpublished decision) (Table) (refusing appeal by third-party defendant insurers of the lower court’s rejection of the motion to quash service).

162

Article II: The Excess Point and Limits of Liability

THE POLICY LIMITS 9.34

The Bermuda Form policy contains limits applicable to: (i) each occurrence; (ii) all occurrences reported in any annual period during Coverage A; and (iii) all occurrences reported during the last year of Coverage A and the whole of Coverage B. Thus, after Article II(A) has dealt with the question of excess point, the 004 Form23 continues: and then only up to the per Occurrence limit of liability stated in Item 2 (a) of the Declarations in respect of such layer for each Occurrence covered hereunder, and further subject to the aggregate limit of liability stated in Item 2 (b) of the Declarations for all layers for all Occurrences covered hereunder of which notice is first given during each Annual Period (or during the Discovery Period with respect to the immediately preceding Annual Period or portion thereof).

A single Bermuda Form policy may cover a large number of insureds, but the limits set out in Article II are not ‘per insured’ limits. They apply ‘[r]egardless of the number of Insureds under this Policy’.24 9.35

Article II(A) thus imposes a limit of liability in respect of ‘each Occurrence covered hereunder’. If the policy contains insurance of different layers, the per-occurrence limit will apply to each layer separately. Because of the integrated occurrence or ‘batching’ provisions of the Bermuda Form, multiple injuries and claims arising from the same cause can come to be treated as a single occurrence, with the result that only one limit applies. Factual questions can of course arise as to whether particular losses are attributable directly or indirectly to ‘the same actual or alleged event, condition, cause, defect or hazard and/or failure to warn of such’.25

9.36

Article II(A) also imposes an overall limit of liability for all occurrences reported during each Annual Period. The 004 Form provides for the possibility of reinstating the aggregate limit in the course of an annual period.26

9.37

The position in relation to aggregate limits is more complex where coverage has been extended for the ‘Discovery Period’. The aggregate limit of liability applicable during the final year of Coverage A continues to apply for the Discovery Period; that is, there is one aggregate limit for the whole Discovery Period and its immediately preceding Annual Period (that is, the last Annual Period of Coverage A and the whole of the period of Coverage B).27 For example, if a policyholder purchases a policy with an aggregate limit

23 Earlier Forms contained similar wording. Article II concluded as follows: ‘irrespective of the period over which the losses, injuries, damages or liabilities occur or the number of such losses, injuries, damages or liabilities’. The 004 Form drops these words, perhaps because their subject matter is more appropriate for the definitions of ‘occurrence’ and ‘integrated occurrence’. 24 See the opening words of Article II of the 004 Form, Article II(D) of the 004 Form and Article II(b) of earlier Forms. 25 See the ‘occurrence’ definition in the 004 Form, and the ‘batching’ provisions of earlier Forms. The question of batching and occurrence integration is considered in ch 6. See too ch 1 for the historical background to batching provisions. 26 Article VI(R), the Reinstatement Condition, discussed below in ch 11, paras 11.97–11.101. 27 This appears to be the effect of the bracketed words in Article II (‘or during the Discovery Period with respect to the immediately preceding Annual Period or portion thereof’). See also the final sentence of Article VI(R)(1) (the Reinstatement Condition); Article VI(S)(3) (the Discovery Period Condition); and ch 11 below.

Joint Ventures, Partnerships and Minority Interests

163

of US$50 million and extends it for a total of five years, it will have a US$50 million aggregate limit during each of those five years.28 If, at the end of the fifth year, it decides not to continue with Coverage A, but to purchase Coverage B, the US$50 million limit will apply to the whole of the fifth year and the whole of Coverage B’s ‘Discovery Period’, taken together. Limits and Notice of Occurrence The Bermuda Form, as drafted, assumes that the policy will continue for a number of years. Over the course of time, retentions and aggregate limits and indeed other terms and conditions may change as a result of the parties’ negotiations. The question arises as to which terms apply to a particular claim. Article II provides the answer:29

9.38

provided, however, that for all Insureds the applicable aggregate limit of liability, per Occurrence limit of liability, per Occurrence retention, and the terms, conditions and exclusions of coverage shall be determined under the Policy as in effect at the time notice of the Occurrence or Notice of Integrated Occurrence for which coverage is asserted is first given pursuant to Article V of this Policy by any Insured.

Accordingly, the critical date is when notice of occurrence or notice of integrated occurrence is served. Article II(B) deals with the impact on limits where the policyholder initially gives notice of occurrence, and subsequently gives notice of an integrated occurrence that encompasses the original occurrence. In these circumstances, the applicable limits are those prevailing when notice of integrated occurrence is given. If the insurer has already paid in respect of the original occurrence, there may need to be an adjustment of limits, since that payment is treated as having been made when notice of the integrated occurrence was first given. Thus, an aggregate limit that was exhausted in an earlier year may become available again.30

9.39

JOINT VENTURES, PARTNERSHIPS AND MINORITY INTERESTS The Bermuda Form contains a modified regime, as far as limits and excess point are concerned, where the liability of an Insured arises out of the operations or existence of a joint venture. In the 004 Form, the regime is set out in Article II(C).31 Article II(C)(1) provides for a reduction of limits in these circumstances to a pro rata percentage of the

28 The 004 Form provides that, during the period of Coverage A, the aggregate limit is annually ‘reinstated’ (Article V(t)). In fact, this is dependent on the insured and insurer agreeing acceptable terms for each annual extension. There is no automatic right to a continual renewal of Coverage A at a pre-arranged premium. The position is different in relation to Coverage B. See ch 2 above. 29 In earlier Forms, the equivalent provision was at the end of Article I. 30 See generally ch 8 above for discussion of notice of occurrence. 31 In earlier Forms, the relevant provisions were, perhaps surprisingly, not contained in Article II dealing with limits of liability, but in Article III(a)(2)(A), which defined ‘Insured’.

9.40

164

Article II: The Excess Point and Limits of Liability

policy’s limits. If, for example, the policyholder’s liability arises out of a joint venture where it has a 50 per cent share, the policy’s limit of liability will be reduced to 50 per cent of the total limit provided under the policy. The relevant clause is directed at the peroccurrence limit and would seem to have no effect on the aggregate limit. Thus, if the aggregate limit is US$25 million and the insurer has paid US$10 million in respect of the liability of a 50 per cent joint venture, an additional US$15 million of the aggregate limit would be available in respect of that year’s cover. 9.41

As a quid pro quo for the reduction in limits, there is also, potentially, a reduction in the excess point (see Article III(C)(2)). As with the main provisions of Limit of Liability provision, the excess point is the higher of two potential excess points: (i) the retention amount specified in the policy, but reduced pro rata in accordance with the percentage share in the joint venture; or (ii) the limits of the underlying policies. Thus, if the per-occurrence limit in the policy is US$25 million and the liability arises in respect of a 50 per cent joint venture, the minimum per-occurrence retention is reduced to US$12.5 million. If the limits of the underlying insurances are greater than US$12.5 million, then the excess point will be higher. But Article II(C)(2)(b)32 of the Bermuda Form recognises that these underlying limits may themselves have been reduced by ‘general provisions relating to Joint Ventures’, presumably a reference to a clause which has a similar effect to the clause in the Bermuda Form which provides for a pro rata reduction in limits. Accordingly, if the underlying limits have been reduced pro rata, then it is the reduced underlying limits that are relevant in ascertaining whether the excess point has been reached.33

9.42

These provisions of the Bermuda Form providing for a reduction in limits and excess point do not, however, apply in certain circumstances. Thus, the full limits and full excess point apply if the Insured has sole responsibility for the Joint Venture or is obligated to provide insurance, such as that afforded by Bermuda Form policy, for the Joint Venture in its entirety.34 This is of importance in, for example, the petroleum industry, where it is common for a company to have only a percentage share in a venture, but to be the operator of the project.

9.43

What is the position if the Named Insured has a minority interest in a company? The Bermuda Form provides coverage only for certain subsidiaries and operations of the Named Insured. In order for a subsidiary or affiliate to qualify, the conditions set out in Article III(P)35 need to be fulfilled. For example, the subsidiary will need to have its accounts consolidated in the financial statements of the Named Insured in accordance with US Generally Accepted Accounting Principles (GAAP), or to be specifically listed in Schedule A to the Bermuda Form policy, in which case its underlying insurances will 32 Article III(a)(2)(B)(ii) in earlier Forms referred to ‘the underlying limits (as such may be reduced if the underlying insurances have been reduced by a clause having the same effect as paragraph (A))’. 33 It is doubtful whether it was necessary specifically to provide for this situation. If the underlying insurances had a reduced limit applicable to the particular occurrence, then it might have been difficult for the insurer to argue that the excess point was to be assessed by reference to the unreduced limits. The drafters of the Bermuda Form perhaps sought to avoid any doubt on this issue, particularly since an applicant’s Schedule B might have simply referred to the unreduced limit. 34 Article II(C)(3) of the 004 Form; Article III(a)(2)(C) of earlier Forms. 35 Article III(a) in earlier Forms.

Joint Ventures, Partnerships and Minority Interests

165

need to be listed in Schedule B.36 The coverage does extend, as already indicated, to joint venture arrangements. A minority interest in a subsidiary will, however, probably not be encompassed within this expression, since ‘Joint Venture’ is defined in the 004 Form to exclude joint ventures that are incorporated.37 In order to cater for minority holdings in companies which are not covered by the basic Bermuda Form, a ‘Limited Liability Entity Endorsement’ has become available in recent years.38 It has been used by the petroleum industry, where it is not unusual for minority holdings to be bought and sold with some regularity. This endorsement provides cover, on certain terms, to the company in which the minority interest is held, but subject to a reduced limit and excess point. These reductions operate in a broadly similar fashion to the reductions that apply to partnerships and joint ventures.

36 The application form will ask the insured to list subsidiaries in Schedule A; eg, ‘List on Schedule A (to be attached hereto and form part of the policy when issued) any unconsolidated subsidiary, affiliate, associated company or joint venture to be 100% insured’ (XL 1987 application form). A more recent ‘Bermuda Market Excess Liability Application’ form asks the policyholder to: ‘Complete the attached Schedule A (to be attached hereto and form part of the policy when issued) listing any other unconsolidated subsidiary, affiliate, associated company, or joint venture to be fully insured. Note changes since last Application.’ 37 Article III(S) of the 004 Form. 38 See Noble Assurance Co v Gerling-Konzern General Insurance Co [2007] EWHC 253 (Comm), [2007] 1 CLC 85, where the arbitrators’ interpretation of an LLE Endorsement led to subsequent court proceedings.

9.44

10 The Exclusions INTRODUCTION 10.01

Application of Bermuda Form policy exclusions raises a number of different issues. First, what are the facts upon which the potential application of the exclusion depends? Second, how is the exclusion to be interpreted? Is the language of the exclusion applicable to the relevant facts? How, if at all, is the interpretation of the exclusion affected by the modification of New York law set out in the governing law provision? Third, do questions of causation arise? Was the loss caused by facts that fall within the exclusion? What test of causation applies?

10.02

Before looking at the wording of the particular exclusions contained in the Bermuda Form, this chapter looks more generally at New York law in relation to (i) the interpretation of exclusions, and (ii) the tests applied in relation to causation of loss. In doing so, we consider the impact of the choice of law provision as it modifies New York law. Thereafter, we address the individual exclusions and relevant New York or, if New York law is sparse, the law of other states in the United States.

INTERPRETATION OF EXCLUSION CLAUSES: GENERAL PRINCIPLES AND THE EFFECT OF THE BERMUDA FORM’S MODIFICATION OF NEW YORK LAW 10.03

The general principles of New York law applicable to interpretation of insurance contracts, and their modification under the Bermuda Form, are discussed elsewhere.1 These general principles have engendered a number of specific doctrines that apply to the interpretation of exclusions. But to what extent are these doctrines applicable to the Bermuda Form, given the modifications made by the choice of law clause? This is a controversial question on which different views may be held. Our suggested approach is as follows.

10.04

Generally, the court will construe the terms of an insurance contract as common people understand them.2 This principle flows from the general rule that courts should construe 1 See above para 4.13 and see generally ch 4. 2 Throgs Neck Bagels Inc v GA Ins Co of NY 671 NYS2d 66, 68 (App Div 1988) (citing Gittelson v Mut Life Ins Co of NY 41 NYS2d 478, 481–82 (App Div 1943)); Lewis v Ocean Accident & Guar Corp 224 NY 18, 22 (1918); Belt Painting Corp v TIG Ins Corp 742 NYS2d 332, 334 (App Div 2002).

Interpretation of Exclusion Clauses

167

contracts to give effect to the parties’ intentions as expressed in the contract language.3 This approach is consistent with the terms of the choice of law clause in the Bermuda Form. Because New York courts construe ambiguities in an insurance policy against the insurer, as drafter of the policy,4 courts naturally so construe exclusions under New York law. Indeed, courts often single exclusions out for particular mention, emphasising that they will not be applied beyond their clear and unmistakable intent. Thus, if uncertainty exists about whether an exclusion applies, the court will adopt a narrow reading: ‘exclusions or exceptions from policy coverage must be specific and clear in order to be enforced. They are not to be extended by interpretation or implication, but are to be accorded a strict and narrow construction’.5 To ‘negate coverage by virtue of an exclusion, an insurer must establish that the exclusion is stated in clear and unmistakable language, is subject to no other reasonable interpretation, and applies in the particular case’.6

10.05

Insofar as this principle depends solely on the contra proferentem rule, the choice of law clause in the Bermuda Form would require its abandonment. However, it may be thought that there is more to the issue than that. The principle that a term drafted as an exclusion, which is ‘carved out’ as an exception to the basic obligations under a contract, should be strictly interpreted could be said to have a life of its own, independent of the identity of the party drafting the contract.7 This principle, it could be suggested, is not ‘arbitrary’.8 Rather, it reflects a rational assessment of the likely intention of the parties in the way they have structured their contract. If that is correct, then the principle that exclusions should be narrowly construed, not ‘against the drafter’ or ‘against the insurer’, but as a reflection of the relative roles of the basic obligations under a contract and terms that restrict them, remains valid under the Bermuda Form. Insurers may argue, however, that an approach which results in exclusions being narrowly construed is not consistent with the Bermuda Form’s requirements; namely, an ‘evenhanded’ approach to construction and the resolution of ambiguities ‘in the manner most consistent with the relevant provisions, stipulations, exclusions and conditions’.

10.06

As to the burden of proof under New York law, once the policyholder has shown that a valid insurance policy exists and that the policy ‘presumptively covered’ that loss, then the burden shifts to the insurance company to demonstrate that an exclusion or another

10.07

3 See above ch 4, paras 4.09–4.13. 4 Belt Painting 742 NYS2d at 334; Throgs Neck Bagels 671 NYS2d at 68 (citing Greaves v Public Serv Mut Ins Co 5 NY2d 120, 125 (1959)); Casey v General Accident Ins Co 578 NYS2d 337, 338 (App Div 1991) (citing Breed v Ins Co of N Am 413 NYS2d 351 (NY 1978)); Sturges Mfg Co v Utica Mut Ins Co 371 NYS2d 444, 448–49 (NY 1975); Thomas J Lipton Inc v Liberty Mut Ins Co 357 NYS2d 705, 708 (NY 1974). 5 Seaboard Sur Co v Gillette Co 486 NYS2d 873, 876 (NY 1984). The insurance Restatement states: ‘Exclusions are interpreted narrowly.’ American Law Institute, Restatement of the Law, Liability Insurance (St Paul, MN, American Law Institute Publishers, 2019) s 32(2). 6 Throgs Neck Bagels 671 NYS2d at 71. 7 The point is well made by Staughton LJ in the English case Youell v Bland Welch & Co Ltd [1992] 2 Lloyd’s Rep 127, 134, discussing the equivalent rules in English law. 8 See the terms of the governing law provision (Article VI(O) of the 004 Form) which prohibits any ‘presumption or arbitrary interpretation or construction in favour of either the Insured or the Company’. See the similar wording in Article V(q) of earlier Forms.

168

The Exclusions

policy provision applies to preclude coverage for the loss.9 Accordingly, the insurer must prove that an exclusion applies to the facts of a particular case.10 The question of whether a particular condition is treated as a condition of coverage or as an exclusion is for the most part a matter of substance, not of form.11 Courts look at the effect of the provision: if it acts to preclude coverage, the insurance company will bear the burden to prove its applicability. Although the rule relating to burden of proof is sometimes explained or justified on grounds which include the contra proferentem rule, we consider that the case law on burden of proof has continuing validity under the Bermuda Form, both as a positive rule of New York law and as a rational approach to the allocation of the burden of proof when a party seeks to rely on an exclusion. 10.08

Courts have described the burden of proving that an exclusion applies as a ‘heavy’ one.12 It is not clear, however, that such comments apply to the burden of proof (that is, the burden to establish relevant facts) at all. It seems, rather, to be a reference to the burden of persuading a court that the facts proved fall within the language of the exclusion. In other words, it is really a different way of presenting the principle that, if doubt exists about whether the exclusion is wide enough to cover all aspects of the loss, it will not apply. Thus, when in Seaboard Surety Co v Gillette Co,13 the New York Court of Appeals said that the insurer ‘must satisfy the burden which it bears of establishing that the exclusions or exemptions apply in the particular case and that they are subject to no other reasonable interpretation’,14 its concern was primarily with the burden of persuading the court that the terms of the exclusion apply, not with the burden of proving particular facts on the evidence. That burden, it appears, is simply the ordinary burden to establish facts on the preponderance of the evidence.

10.09

Sometimes, exclusions contain exceptions that act to preserve coverage. Exceptions to exclusions are to be interpreted broadly,15 but the application of this principle under the 9 Throgs Neck Bagels 671 NYS2d at 70. See also Great N Ins Co v Dayco Corp 637 F Supp 765, 777 (SDNY 1986). A policyholder generally has the burden to show the existence of the insurance policy and its applicability to the liability or loss, and the insurer has the burden to prove that all aspects of the loss fall entirely within the exclusion: Pan Am World Airways Inc v Aetna Cas & Sur Co 505 F2d 989, 999 (2d Cir 1974) (applying New York law), in the context of an ‘all-risks’ insurance policy. The Bermuda Form provides very wide liability coverage: it can be viewed essentially as an all-risks liability policy covering liability for personal injury, property damage and advertising liability, so that the insured bears the burden of proving a covered occurrence and the insurer has the burden of proving the applicability of an exclusion. 10 Eg, American Nat’l Fire Ins Co v Mirasco Inc 249 F Supp 2d 303, 323 (SDNY 2003), vacated in part on other grounds, Mirasco Inc v Am Nat’l Fire Ins Co 144 F Appx 171, 173 (2d Cir 2005). 11 See, eg, Stonewall Ins Co v Asbestos Claims Mgmt Corp 73 F3d 1178, 1205 (2d Cir 1995) (and the cases there cited), modified on other grounds, 85 F3d 49 (2d Cir 1996) (applying New York law). 12 See, eg, Belt Painting 742 NYS2d at 334; Roofers’ Joint Training Apprentice & Educ Comm of W NY v General Accident Ins Co of Am 713 NYS2d 615, 617 (App Div 2000) (the insurer bears ‘the heavy burden of showing that the exclusion applies in the particular case and is subject to no other reasonable interpretation’) (citing Continental Cas Co v Rapid-American Corp 593 NYS2d 966, 972–73 (NY 1993)). 13 Seaboard Surety 486 NYS2d at 876. 14 See also Westview Assocs v Guar Nat’l Ins Co 717 NYS2d 75, 78 (NY 2000); Continental Casualty v Rapid-American 593 NYS2d at 972: ‘To negate coverage by virtue of an exclusion, an insurer must establish that the exclusion is stated in clear and unmistakable language, is subject to no other reasonable interpretation, and applies in the particular case.’ 15 Bebber v CNA Ins Cos 729 NYS2d 844, 845–46 (Sup Ct 2001) (citing 68A NY Jur 2d [rev] ‘Insurance’ s 802, and Barry R Ostrager and Thomas R Newman, Handbook on Insurance Coverage Disputes, 19th edn (New York, Wolters Kluwer, 2019)) (hereafter Ostrager and Newman, Handbook on Insurance Coverage Disputes).

Causation 169

Bermuda Form is questionable. New York courts have held that the policyholder bears the burden of proving that an exception to an exclusion applies to preserve coverage.16 That principle undoubtedly would be applied under the Bermuda Form. Apart from these general principles, many New York decisions have addressed exclusions identical or similar to the exclusions contained in the Bermuda Form. For example, the pollution exclusion has given rise to extensive case law in New York, including a decision in 2003 by the New York Court of Appeals,17 and indeed by courts elsewhere.18 Such decisions often involve the use of principles of construction that the Bermuda Form would not itself countenance. Does that mean that the result is to be ignored? For reasons given elsewhere,19 we do not think those holdings should be (or can be) disregarded, at least where the decision squarely addresses the very issue that arises under the Bermuda Form. However, considerable care is needed in reasoning from such decisions by analogy. Cases dealing with causation also sometimes make use of contra proferentem reasoning. It seems to us that to ignore such decisions would be to cast the parties adrift on a featureless plain. Although one of the strands of these cases involves reasoning that would not itself be used directly under the Bermuda Form, the result is a reasonably coherent body of general principle. This general principle continues, in our view, to apply to Bermuda Form cases.

10.10

CAUSATION Questions of causation frequently arise when the issue is whether an exclusion applies. New York courts have stated the principles set out in the following paragraphs. It should be noted, however, that many of the cases concern exclusions in first-party property insurance, and arguments exist as to whether, or the extent to which, such decisions apply in the context of third-party liability insurance.20 The decisions thus must be approached with discernment. In addition, it is always necessary to consider the precise wording of the exclusion under consideration. The scope of the causation enquiry may be widened or narrowed by the particular words used. The language used in the Bermuda Form exclusions varies. In some cases, an enquiry into the cause of the policyholder’s liability is unnecessary.

16 Northville Indus Corp v National Union Fire Ins Co of Pittsburgh Pa 657 NYS2d 564, 568–69 (NY 1997); County of Fulton v US Fidelity & Guar Co 600 NYS2d 972, 974 (App Div 1993). The Restatement Liability Insurance states the general rule: ‘an exception to an exclusion narrows the application of the exclusion; the exception does not grant coverage beyond that provided in the issuing clauses’. Ibid s 32(5). 17 Belt Painting 742 NYS2d at 334. 18 A decision by the District of Columbia Court of Appeals summarises much of the voluminous case law and academic writing on the pollution exclusion. Richardson v Nationwide Mut Ins Co 826 A2d 310 (DC 2003), vacated, rehearing en banc granted, 832 A2d 752 (DC 2003), vacated due to settlement of the parties, 844 A2d 344 (DC 2004). The academic writings include articles written from both a policyholder’s and an insurer’s perspective: see the materials referred to in fn 22 of the dissenting opinion by Associate Judge Glickman in Richardson 826 A2d at 357 n 22. 19 See ch 4 above, paras 4.28–4.32. 20 In the context of environmental pollution, New York courts have sometimes applied the principles derived from third-party liability insurance to first party coverage; see, eg, Vigilant Ins Co v I.Techs Inc 696 NYS2d 596 (App Div 1998); Herald Square Loft Corp v Merrimack Mut Fire Ins Co 344 F Supp 2d 915 (SDNY 2004). But see also n 24 below.

10.11

170

The Exclusions

For example, Exclusion A excludes personal injury to an individual person or p roperty damage to specific property which takes place prior to the inception or retroactive coverage date. In relation to this exclusion, the question is simply when the injury or damage took place. Similarly, the applicability of Exclusion D(1) depends upon whether there has been damage to property owned or occupied by or rented to any Insured. Other exclusions, however, do give rise to an enquiry as to the cause of the policyholder’s liability. Many of the exclusions use the words ‘arising out of’. These words have been given a wide meaning in the New York case law.21 Thus, Exclusion C excludes liability for property damage ‘arising out of any act, error or omission in the rendering of professional services’ and Exclusion K excludes liability ‘arising out of’ the discharge of pollutants. Other exclusions are cast in apparently wider terms. As such, the war exclusion (Exclusion G) refers to personal injury and property damage ‘directly or indirectly occasioned by, happening through or in consequences of’ war, invasion and other matters. The radioactive contamination exclusion (Exclusion M) refers to liability ‘of whatsoever nature directly or indirectly caused by or contributed to by or arising from’ ionising radiations. 10.12

First, the question of causation depends upon whether the insurer is able to show that the parties contemplated that the exclusion should apply under the facts of a particular case.22 Accordingly, the language of the particular exclusion is obviously important, since some exclusions are drafted more widely than others. In Bird v St Paul Fire & Marine Insurance Co,23 a case frequently cited in later authorities,24 the question arose as to whether a first-party fire insurance policy provided coverage in the following circumstances. A fire had broken out beneath some freight cars. The cars were loaded with explosives which exploded after the fire had burned for at least 30 minutes. This explosion 21 They are wider than ‘caused by’ or the concept of proximate cause: see Burlington Ins Co v NYC Transit Auth 29 NY3d 313 (NY 2017). 22 Album Realty Corp v American Home Assur Co 592 NYS2d 657, 658 (NY 1992). For reasons given in ch 4, paras 4.23 ff, we do not think that the ‘prohibition’ in the governing law clause relating to the ‘reasonable expectation of the parties’ prevents a tribunal from seeking to ascertain the intention of the parties at the time of contracting as objectively expressed in the contract. That process could be said to involve giving effect to the ‘reasonable expectations’ of the parties; see, in England, the general statement of Steyn LJ in First Energy (UK) Ltd v Hungarian International Bank Ltd [1993] 2 Lloyd’s Rep 194 at 196. We do not think that the Bermuda Form prohibits that approach. Indeed, an interpretation of the contract to produce a result that neither party can have intended is not only counter-intuitive for both English and American lawyers, but also seems to be the antithesis of the process of interpreting a contract. The approach to ascertaining the parties’ intention under the Bermuda Form should, however, respect the language that the parties have used. This process is distinct from the application of the ‘reasonable expectations’ doctrine, which is discussed in paras 4.23 ff, or a process of interpretation that starts from a priori assumptions as to what either or both parties reasonably expected. 23 Bird v St Paul Fire & Marine Ins Co 224 NY 47 (1918) (Cardozo J). 24 Such as in Album Realty and Continental Insurance Co v Arkwright Mutual Insurance Co 102 F3d 30 (1st Cir 1996), where the US Court of Appeals for the First Circuit described Bird as ‘seminal’ and the ‘wellspring’. Bird is also referred to in passing in World Trade Center Properties LLC v Hartford Fire Insurance Co 345 F3d 154, 180–90 (2d Cir 2003), where the court addresses, under the leaseholder’s first-party property insurance programme, whether the destruction of the World Trade Center on 11 September 2001 constitutes one occurrence or two under the policy language or, where the policies did not yet exist, insurance binders for various insurance companies. The policyholder had relied upon liability insurance cases in support of its arguments for property insurance coverage. Although the court in World Trade Center did rely on some liability insurance cases in interpreting the property insurance provisions at issue there, it urged caution in using principles developed in liability insurance policies to interpret definitions used in property insurance policies. That caution also applies in reverse.

Causation 171

caused another fire, which in turn caused another, and much greater, explosion of a large quantity of dynamite and other explosives stored in the freight yard. This last explosion caused a ‘concussion of the air’ that damaged the policyholder’s vessel about 1,000 feet away. The policy covered losses by fire, and the question was therefore whether fire caused the damage to the policyholder’s vessel. The New York Court of Appeals rejected the policyholder’s claim. Judge Cardozo dealt extensively with questions of causation, drawing upon well-known (to English lawyers) English cases, and ultimately relied upon ‘the reasonable expectation and purpose’ of the ordinary business person ‘when making an ordinary business contract’, and the understanding of the need for reasonable proximity between the initial cause (here, the fire) and the loss: The problem before us is not one of philosophy. If it were, there might be no escape from the conclusion of the court below. General definitions of a proximate cause give little aid. Our guide is the reasonable expectation and purpose of the ordinary business man when making an ordinary business contract. It is his intention, expressed or fairly to be inferred, that counts. There are times when the law permits us to go far back in tracing events to causes. The inquiry for us is how far the parties to this contract intended us to go. The causes within their contemplation are the only causes that concern us. … The law solves these problems pragmatically. There is no use in arguing that distance ought not to count, if life and experience tell us that it does. The question is not what men ought to think of as a cause. The question is what they do think of as a cause. We must put ourselves in the place of the average owner whose boat or building is damaged by the concussion of a distant explosion, let us say a mile away. Some glassware in his pantry is thrown down and broken. It would probably never occur to him that, within the meaning of his policy of insurance, he had suffered loss by fire. A philosopher or a lawyer might persuade him that he had, but he would not believe it until they told him. He would expect indemnity, of course, if fire reached the thing insured. He would expect indemnity, very likely, if the fire was near at hand, if his boat or his building was within the danger zone of ordinary experience, if damage of some sort whether from ignition or from the indirect consequences of fire, might fairly be said to be within the range of normal apprehension. But a different case presents itself when the fire is at all times so remote that there is never exposure to its direct perils, and that exposure to its indirect perils comes only through the presence of extraordinary conditions. … The case comes, therefore, to this: Fire must reach the thing insured, or come within such proximity to it that damage, direct or indirect, is within the compass of reasonable probability. Then only is it the proximate cause, because then only may we suppose that it was within the contemplation of the contract. In last analysis, therefore, it is something in the minds of men, in the will of the contracting parties, and not merely in the physical bond of union between event, which solves, at least for the jurist, this problem of causation.25

Under Bird, causes that are separated temporally or spatially from the advent of loss do not qualify as proximate causes of loss.

25 224

NY at 51–55 (internal citations omitted).

10.13

172

10.14

The Exclusions

Second, where the exclusion uses language such as ‘caused by’ or ‘resulting from’ an excluded peril, the insurer must show that the excluded peril proximately caused the loss.26 Pan American World Airways Inc v Aetna Casualty & Surety Co, a seminal case in the United States on the interpretation of war-risk exclusions,27 provides a useful discussion of proximate cause in the context of exclusions in an all-risks property insurance policy. There, the insurer argued that war-risk exclusions precluded coverage for the loss of hijacked aircraft. In holding that the first-party insurance at issue there covered the loss, the court discussed the concept of proximate cause, restricting its scope to the ‘efficient physical cause of the loss’: The all risk policies exclude ‘loss or damage due to or resulting from’ the various enumerated perils, a phrase that clearly refers to the proximate cause of the loss. Remote causes of causes are not relevant to the characterization of an insurance loss. In the context of this commercial litigation, the causation inquiry stops at the efficient physical cause of the loss; it does not trace events back to their metaphysical beginnings. The word ‘due to or resulting from’ limit the inquiry to the facts immediately surrounding the loss.

Relying on the reasoning of a US Supreme Court case, the Pan American court further cautioned against taking ‘a broad view’ on causation, but rather required focus on the proximate cause, or that ‘nearest to the loss’: Thus in Queen Insurance Co v Globe & Rutger Fire Insurance Co [citations omitted], Mr. Justice Holmes wrote: ‘The common understanding is that in construing these policies we are not to take a broad view but generally are to stop our inquiries with the cause nearest the loss. This is a settled rule of construction, and if it is understood, does not deserve much criticism, since theoretically at least the parties can shape their contract as they like.’28

New York courts give especially limited scope to the causation inquiry.29 10.15

The court relied on the fact that the policy language there used ‘due to or resulting from’ and on first-party insurance case law that looked specifically to the nearest cause of loss. Using these principles, the court concluded that the hijacking of the aircraft there did not meet the common understanding of ‘war’ because, for example, the hijackers did not represent sovereign governments, or use insignia, uniforms and other indicia of stateorganised military forces.30

10.16

In reaching this conclusion, the Pan American court reviewed various authorities, including Bird and a number of English cases. It concluded that the insurer could have modified New York law’s proximate cause approach by drafting the exclusion to apply to more than the ‘efficient physical cause’ of loss under the policy language at issue there, but did not do so: These cases establish a mechanical test of proximate causation for insurance cases, a test that looks only to ‘causes nearest to the loss’. This rule is adumbrated by the maxim

26 Great Northern v Dayco 637 F Supp at 777; see also Home Ins Co v American Ins Co 537 NYS2d 516, 517 (App Div 1989). 27 Pan Am World Airways, Inc v Aetna Cas & Sur Co 505 F2d 989 (2d Cir 1974) (applying New York law). 28 Ibid (quoting Queen Ins Co v Globe & Rutger Fire Ins Co 263 US 487, 492 (1924)). 29 Eg, Pan American 505 F2d at 1006. 30 Ibid at 1015.

Causation 173 contra proferentem: if the insurer desires to have more remote causes determine the scope of exclusion, he may draft language to effectuate that desire.31

Third, New York courts have held that, where a loss results from a concurrence of two independent events, one covered and one excluded, a first-party property insurance policy will cover the loss in the absence of an explicit ‘concurrent causation’ exclusion. When an insurance policy expressly insures against direct loss and damage by one element, but excludes loss or damage caused by another element, the coverage extends to the loss even though the excluded element is a contributory cause of the loss.32

10.17

Fourth, an exclusion may be drafted so as to apply in a wider range of circumstances than indicated by the discussion above. For example, some insurance policies include language seeking to preclude coverage if a particular event is only a contributing factor rather than the proximate cause.33 In ABI Asset Corp v Twin City Fire Insurance Co,34 the court referred to such an exclusion as an ‘anti-concurrent cause clause’. The exclusion provided that ‘This policy does not insure against loss or damage caused by, resulting from, contributed to or aggravated by’ certain matters.35 The court stated that New York courts had interpreted similar clauses to exclude coverage for a loss resulting from multiple contributing causes when the insurer can demonstrate that the insurance policy plainly excluded coverage for any of the concurrent or contributing causes of the loss. The insurer must still show, however, that the excluded cause was in truth a contributing cause of the loss.36

10.18

In Album Realty Corp v American Home Assurance Co,37 the New York Court of Appeals addressed concurrent causation in a case involving water damage. A sprinkler head had frozen and ruptured, causing water to fill the sub-basement of the insured premises and causing damage to mechanical and electrical equipment located in the sub-basement and some structural damage to the building. The insurer rejected coverage under an exclusion for loss or damage caused by extremes of temperature or freezing. In the policyholder’s action to enforce coverage, the court held that the question of coverage depends on whether the parties contemplated that the exclusion would apply in a circumstance such as that presented; that is, did they expect a loss arising in this manner to be characterised as being caused by freezing?38

10.19

31 See also Cincotta v National Floor Insurer Ass’n 452 F Supp 928, 930 (EDNY 1977); Great Northern v Dayco 637 F Supp at 777. 32 Great Northern v Dayco 637 F Supp at 780. See also Mirasco 249 F Supp 2d at 325 (citing Great Northern v Dayco 637 F Supp at 779–80; and Essex House v St Paul Fire & Marine Ins Co 404 F Supp 978, 985 (SD Ohio 1975)), vacated in part on other grounds, Mirasco Inc v Am Nat’l Fire Ins Co 144 F App’x 171, 173 (2d Cir 2005) (unpublished). 33 See, eg, the discussion in Great Northern v Dayco 637 F Supp at 780 (citing Molycorp Inc v Aetna Cas & Sur Co 431 NYS2d 824 (App Div 1980)). In the Bermuda Form, see the wide wording of the radioactive contamination exclusion, Exclusion M: ‘Liability of whatsoever nature directly or indirectly caused by or contributed to by or arising from …’. 34 ABI Asset Corp v Twin City Fire Ins Co No 96 Civ 2067, 1997 WL 724568, at *1 (SDNY 18 November 1997). 35 Ibid at *2 n 3. 36 See Throgs Neck Bagels 671 NYS2d at 70, where the insurer’s reliance on the exclusion failed notwithstanding its wide ‘anti-concurrent cause’ wording because it failed to prove that the cause in the question contributed to the loss as a matter of fact. 37 Album Realty Corp v American Home Assurance Co 592 NYS2d 657 (NY 1992). 38 Ibid at 658.

174

The Exclusions

10.20

The court held that although the property damage would not have occurred in the absence of freezing, freezing was not the proximate, efficient and dominant cause of the loss. Under Bird, a reasonable business person would conclude that water was the proximate cause of the loss, not freezing, and would look no further for alternative causes. The court held that an inquiry into causation does not trace events back to their metaphysical beginnings. It thus drew a contrast between the language of the exclusion, which excluded damage ‘caused by’ freezing, and other exclusions in the policy that used broader language – for example, ‘caused by or resulting from’, ‘arising directly or indirectly from’, ‘directly or indirectly caused by’ and ‘resulting from’.39

10.21

The result of the causation inquiry in any particular case will thus depend both upon a factual enquiry, an exercise of judgment, the language of the particular policy and exclusion, and a consideration of whether the loss arises under a first-party or third-party policy. Some New York decisions show that apparently similar facts can give rise to different legal conclusions on causation. For example, in Home Insurance Co v American Insurance Co,40 the relevant exclusion sought to preclude coverage for loss caused by or resulting from electrical arcing (in lay terms, an electrical short circuit). In that case, an open drain line inadvertently introduced water and steam into a mechanical equipment room. Moisture generated by the open drain line saturated a duct system, although no water flowed directly onto the duct system. This moisture resulted in electrical arcing, causing damage. Home Insurance, which had written first-party all-risks insurance policy, argued that the exclusion did not apply because the electrical arcing was precipitated by another cause – the escape of hot water and steam from the open drain line – which was itself a peril covered under the policy. The court held, however, that the efficient or dominant cause of the loss was the short circuit in the electrical system, that the steam merely set the stage for that later event and therefore that the escape of steam was the remote, and not the proximate, cause of the loss. It held that the damage ‘would not have occurred in the absence of the electrical arcing’41 and accordingly was excluded.

10.22

By contrast, in Continental Insurance Co v Arkwright Mutual Insurance Co,42 flood water entered a building in Manhattan. The water came into contact with electrical panels, causing electrical arcing which in turn caused an immediate explosion that blew a large hole in the switching panels. The insurance company argued that a ‘flood’ exclusion applied to preclude coverage. The policyholder relied on Home Insurance v American Insurance to argue that the flood exclusion did not apply because electrical arcing, not flooding, caused the damage. The court disagreed, finding no temporal remoteness or spatial separation between the two competing causes. Relying on the ‘seminal’ judgment

39 Ibid. 40 Home Insurance Co v American Insurance Co 537 NYS2d 516 (App Div 1989). This case, as with many of the cases, involved a claim by an all-risks insurer seeking contribution from another insurer, with the latter relying upon a policy exclusion. Policyholders have argued that decisions between insurance companies like Home Insurance v American Insurance do not apply in disputes between policyholders and insurance companies. 41 This language could be equated with a ‘but for’ test, but it is clear that the court was concerned with what was the proximate, dominant or efficient cause of the loss. 42 Continental Ins Co v Arkwright Mut Ins Co 102 F3d 30 (2d Cir 1996) (applying New York law).

Causation 175

of Judge Cardozo in Bird, the court applied the ‘reasonable businessperson’ rule in finding that the flood, not the electrical arcing, was the direct cause of the loss: Given the importance placed upon temporal remoteness and spatial separation in Bird, the wellspring decision under New York law, we conclude that the district court correctly held that the legal cause of the damage to the electrical switching panels was the flooding, not electrical arcing. We therefore hold that a reasonable business person would consider that the damage sustained by the electrical switching panels in the Water Street Building, just as any other water damage to the building, was caused by flood. That is to say, as then Judge Cardozo did, since the flood waters surged onto the site of the loss, a reasonable business person would consider the damage to the electrical switching panels to have been ‘within the danger zone of ordinary experience …’.43

One area of difficulty highlighted by a large number of cases44 concerns the situation where the immediate cause of the victim’s injury is excluded (for example, an assault by an employee of the policyholder), but the claim is made against the policyholder on a dual basis; for example, vicarious liability for the assault and also negligence in appointing and failing to supervise the employee. Can the policyholder in these situations recover under the policy on the basis that its liability has arisen from a cause that is not excluded (for example, negligent appointment or failure to supervise)?

10.23

Courts in New York have held, in the context of general liability policies, that the policyholder cannot recover in this type of situation, when the policy contains the broader terms identified by the court in Album Realty (‘caused by or resulting from’, ‘arising out of’, ‘directly or indirectly caused by’ and so on). In Scottsdale Indemnity Co v Beckerman, the court considered an ‘arising out of’ exclusion. It held that the expression ‘arising out of’ is interpreted broadly to mean ‘“originating from, incident to, or having connection with”’.45 The exclusion was therefore applicable if the plaintiff in an underlying action or proceeding alleges the existence of facts clearly falling within the exclusion, and none of the causes of action that he or she asserts could exist but for the existence of the excluded

10.24

43 Ibid at 37 (internal citations omitted; footnote omitted); see Bird 224 NY at 54–55. 44 See Ruggerio v Aetna Life & Cas Co 484 NYS2d 106 (App Div 1985); New Hampshire Ins Co v Jefferson Ins Co of NY 624 NYS2d 392 (App Div 1995); US Underwriters Ins Co v Val-Blue Corp 623 NYS2d 834 (NY 1995); Mount Vernon Fire Ins Co v Creative Housing Ltd 645 NYS 433 (NY 1996); Mount Vernon Fire Ins Co v Jones No CV 95 3295, 1997 WL 37033 (EDNY 14 January 1997); US Underwriters Ins Co v Zeugma Corp No 97 Civ 8031, 1998 WL 633679 (SDNY 15 September 1998); Mount Vernon Fire Ins Co v Rennie No 97 CV 2778, 1999 WL 33113 (EDNY 22 January 1999); US Fire Ins Co v NY Marine & Gen Ins Co 706 NYS2d 377 (App Div 2000); Watkins Glen Cent Sch Dist v Nat’l Union Fire Ins Co of Pittsburgh, Pa 732 NYS2d 70 (App Div 2001); US Underwriters Ins Co v 203–11 W 145th St Realty Corp No 99 Civ 8880, 2001 WL 604060 (SDNY 31 May 2001); US Liab Ins Co v Dehkterman No 00 CV 2273, 2002 WL 31780174 (EDNY 4 October 2002); RJC Realty Holding Corp v Republic Franklin Ins Co 756 NYS2d 631 (App Div 2003), rev’d, 777 NYS2d 4 (NY 2004) (reversing Appellate Division’s decision upholding coverage for liability for sexual assault against employer on ground that employer’s liability was vicarious); Scottsdale Indem Co v Beckerman 992 NYS2d 117 (App Div 2014); Country-Wide Ins Co v Excelsior Ins Co 46 NYS3d 96 (App Div 2017); Beazley Ins Co v Ace Am Ins Co 880 F3d 64 (2d Cir 2018); Great Am Ins Co v Houlihan Lawrence Inc 449 F Supp 3d 354 (SDNY 2020). 45 992 NYS2d at 121 (quoting Maroney v NY Cent Mut Fire Ins Co 5NY3d 467, 472 (NY App Div 2005)). See, too, Burlington Ins v NYC Transit 29 NY3d at 324 (‘Furthermore, “arising out of” is not the functional equivalent of “proximately caused by”’); Lissauer v Guidone Specialty Mut Ins, 161 AD3d 974, 975–76 (NY App Div 2018) (‘arising out of’ requires ‘“only that there be some causal relations between the injury and the risk for which coverage is provided’”, quoting Mack-Cali Realty v NGM Ins Co 119 AD3d 905, 906 (NY App Div 2014)).

176

The Exclusions

activity or state of affairs.46 The test is referred to in the case law as a ‘but for’ test: if no cause of action would exist but for the excluded conduct, then the exclusion applies.47 Accordingly, the ‘but for’ test applied in these cases requires consideration of the critical operative act giving rise to the cause of action against the policyholder. 10.25

For example, in Ruggerio,48 a ‘comprehensive’ or later ‘commercial’ general liability policy (CGL) insurance policy contained a standard exclusion for liability ‘arising out of the ownership, maintenance, operation [or] use of an automobile’. A claimant recovered damages for personal injury from a cab company on the basis of (i) vicarious liability for the driver’s negligence, and (ii) the company’s own negligence in failing to ascertain whether the driver was qualified and licensed to operate a taxi, and entrusting him with a taxi when he was intoxicated. The court relied upon the automobile exclusion to reject coverage for the cab company’s claim against its insurers. It described the company’s negligence (in hiring an incompetent driver, and dispatching him when intoxicated) as ‘reasons or sub-factors’ explaining why the accident arose out of the operation of the vehicle.49

10.26

Other cases have followed this line of reasoning. In Mount Vernon Fire Insurance Co v Jones,50 the court applied the ‘but for’ test of concurrent causation under New York law to sustain an exclusion for injury arising out of the ingestion, inhalation or absorption of lead paint. It held that the exclusion precluded coverage because the personal injury action would not have arisen ‘but for’ problems with lead paint, and notwithstanding the generalised allegations against the policyholder for failure to remedy defects and hazards.51 In Bebber v CNA Insurance Cos,52 the court used the ‘but for’ test to enforce coverage in a case where a swimming pool had lifted out of the ground after it was drained for cleaning. The insurance company unsuccessfully relied upon an exclusion for damage ‘caused by water damage’, arguing that underground water pressure caused the lifting of the pool. The court held that that the damage resulted when the policyholder drained his pool. The underground water pressure was simply part of the environment within which the positive act (draining the pool) took place, causing the damage.

10.27

In many Bermuda Form cases where an exclusion is potentially applicable, this ‘but for’ test will not be satisfied. It is, for example, very common for a claim to be made against the policyholder on the basis of a number of theories of liability, some of which would be covered (for example, claims for negligence) and some of which would, if proved, be uncovered (for example, liability arising from dishonesty or other matters specified in Exclusion IV(P)). 46 992 NYS2d at 121. This test was applied in, eg, Beazley Insurance Co v Ace American Ins Co 880 F3d 64 (2d Cir 2018); and Great American v Houlihan Lawrence 449 F Supp 3d 354. 47 Beazley 880 F 3d 64, 71. 48 484 NYS2d at 106. 49 Ibid at 106–07. 50 Mount Vernon Fire v Jones 1997 WL 37033. 51 Ibid at *2. See also US Underwriters v Zeugma Corp 1998 WL 633679 at *1 (where the exclusion covered operations performed by independent contractors); Rennie No 97 CV 2778, 1999 WL 33113 (involving exposure to lead paint); US Underwriters v 203–11 West 145th Street 2001 WL 604060 (where the court held that an independent contractors exclusion applied to preclude coverage for injury sustained while performing operations for an independent contractor retained by the policyholder); Watkins Glen Cent Sch Dist v National Union Fire Ins Co of Pittsburgh, Pa 732 NYS2d 70 (App Div 2001). 52 Bebber 729 NYS2d 844. This is not a liability insurance case, but it contains discussion of the ‘but for’ concept.

The Specific Exclusions in the Bermuda Form: Introduction

177

In such cases, a cause of action would still exist but for the excluded conduct. These situations give rise to questions of allocation which are discussed in Chapter 5.53

THE SPECIFIC EXCLUSIONS IN THE BERMUDA FORM: INTRODUCTION In approaching the applicability of the exclusions contained in the Bermuda Form, the starting point is the insuring clause (Article I) of the policy, which is drafted in wide terms:

10.28

[The insurer shall] subject to the limitations, terms, conditions and exclusions below, indemnify the Insured for Ultimate Net Loss the Insured pays by reason of liability … for Damages on account of (i) Personal Injury, (ii) Property Damage, (iii) Advertising Liability …54

This is in the nature of ‘all risks’ cover against the liabilities insured. The question of whether the exclusion applies will depend, as indicated by the discussion above, upon the interplay among the facts giving rise to the claim, the existence or non-existence of coverage under the insuring clause (Article I), and the precise wording of the exclusion. The exclusions in the Bermuda Form, in many instances, are similar but not identical to the exclusions typically seen by New York courts. Most New York decisions addressing policy exclusions in general liability insurance policies are interpreting standard form policies drafted by insurance industry organisations and used over the decades since the CGL policy form first began to see wide usage in the 1950s in the United States. In part, the use of these standard form provisions results from insurance regulation in the United States which requires primary policy forms to receive approval from insurance commissions in the various states, the District of Columbia and the Territories.55 Excess, surplus lines and Bermuda Form insurers, for example, do not need, as the primary insurers do, to use the standard form language approved by insurance commissions. However, the exclusions used in their policy forms typically track the language of the underlying primary CGL policies given policyholders’ interest in having consistent coverage through the various layers of their CGL insurance programmes.

10.29

Thus, while the New York cases do not use the exact terms of the exclusions in the Bermuda Form, the exclusions in the typical CGL insurance policy interpreted by New York courts and the parallel exclusions included in the Bermuda Form use similar concepts and wording. Given the Bermuda Form’s use of New York law as its governing

10.30

53 See ch 5 above, paras 5.42–5.54. 54 Article I of the 004 Form. The earlier versions of the Form are to the same effect: the insurer agrees to indemnify the policyholder for: all sums which the Insured shall be obligated to pay by reason of liability imposed upon the Insured by law or assumed under contract or agreement for damages on account of: (1) personal injury; (2) property damage; or (3) advertising liability. 55 See, eg, discussion of insurance regulation in the United States in In re Ins Antitrust Litig 938 F2d 919 (9th Cir 1991), aff ’d in part, rev’d in part, modified sub nom Hartford Fire Ins Co v California 509 US 764 (1993); Morton Int’l Inc v General Accident Ins Co of Am 629 A2d 831 (NJ 1993).

178

The Exclusions

law, the New York cases discussed in this chapter provide the legal framework in which parallel exclusions in the Bermuda Form are interpreted. The New York case law and textbooks do need, however, to be approached with some care, bearing in mind that (i) the wording of particular exclusions may be similar but not identical to those contained in the Bermuda Form, and (ii) the reasoning in cases may not be consistent with the approach to be taken under the modified New York law requirements of the Bermuda Form’s governing law provision. Once again, the discussion in Chapter 4 applies in considering cases that may, at least in part, rely on reasoning from the ‘forbidden canons’. 10.31

The exclusions are contained in Article IV of all versions of the Form. Each exclusion is designated by a letter capitalised in the 004 Form. For the sake of simplicity, we will refer to each exclusion by its designated letter (for example, ‘Exclusion A’ or (a) rather than ‘Article IV(A)’ or IV(a)).

THE ‘PRIOR TO INCEPTION OR RETROACTIVE COVERAGE DATE’ AND ‘OTHER INSURANCE’ EXCLUSION 10.32

In the 001 and 002 Forms, this exclusion sought to preclude coverage for any liability or alleged liability for personal injury, property damage or advertising liability resulting from an occurrence where some or all of the personal injury, property damage or advertising liability resulting from such occurrence is, or but for the issuance of this Policy would be, covered by insurance issued before the Inception Date to the Insured other than insurance listed or which should have been listed on the present and/or prior Schedule B hereto.56

10.33

Chapter 9 in the previous editions of this book discusses this exclusion in greater detail.57 In short, this exclusion contains a number of obscurities. Its purpose appears, in our view, to be not so much to exclude specific matters from the scope of the coverage granted by the policy, but to reiterate that the Bermuda Form does not cover liability from occurrences taking place prior to the Inception Date of the policy in question. This purpose is much clearer in the more recent versions of the Bermuda Form (003 and 004) which have decoupled the exclusion from ‘other insurance’. The 004 Form thus now simply excludes coverage for actual or alleged injury taking place before the Inception Date or Retroactive Coverage Date. This clarification in the 003 and 004 Forms perhaps elucidates the intent behind the exclusion as it appeared in the 001 and 002 Forms.

10.34

To make this intent clear (or clearer), the 004 Form’s version of the exclusion states: This policy does not apply to actual or alleged … Personal Injury to any individual person, Property Damage to any specific property or Advertising Liability which takes place prior to the Inception Date or, if applicable, the Retroactive Coverage Date.58 56 Exclusion (a) in the 001 and 002 Forms. In the 001 and 002 Forms, this exclusion referred to ‘other insurance’. In the 003 and 004 Forms, the exclusion was redrafted to refer to ‘personal injury’ and ‘property damage’ taking place prior to the Inception Date (or a Retroactive Coverage Date, if applicable). 57 See ch 9, paras 9.18–9.32 of the 2nd edn. 58 Exclusion A in the 004 Form.

The ‘Professional Services’ Exclusion

179

Again, the provision is not in the nature of a true exclusion from coverage that would otherwise exist, but a ‘belt and braces’ clause to emphasise that the policy does not cover liability for ‘individual’ occurrences that take place prior to the relevant dates.59

THE ‘WORKERS’ COMPENSATION, ETC’ EXCLUSION This exclusion seeks to preclude coverage for:

10.35

Liability in respect of any obligation for which the Insured or any company as its insurer may be liable under any workers’ compensation, unemployment compensation or disability benefits law; provided, however, that this Exclusion B does not apply to liability of others assumed by the Insured under contract or agreement or to liability arising under the Federal Employers Liability Act, the Jones Act or, in the case of any Insured which is an authorized self-insured, the Longshoremen’s and Harbor Workers’ Compensation Act.60

Broadly speaking, this exclusion excludes coverage for certain liabilities that arise in consequence of the policyholder’s capacity as an employer. The exclusion is itself subject to two provisos. First, the exclusion contains a qualification for liabilities assumed under contract or imposed by certain US statutes like the Federal Employees’ Liability Act (FELA).61 Second, the provisions of the ‘Cross Liability’ Condition62 should be noted. That condition has the effect of providing cover when a claim is made for personal injury suffered by an employee of one policyholder for which another insured is or may be liable. In this situation, the insurance policy responds as if separate insurance policies had been issued to each policyholder. This would appear to have the consequence that an insured can recover the full extent of liability for personal injury to the employee of another insured.

10.36

THE ‘PROFESSIONAL SERVICES’ EXCLUSION This exclusion seeks to prevent recovery for liability for property damage arising from certain professional services and provides that insurance does not apply to: Liability for Property Damage arising out of any act, error or omission in the rendering of professional services, other than architectural and engineering services (which are nonetheless subject to the other exclusions herein, including, without limitation, Exclusion E below), including, but not limited to, the rendering of legal, accounting, data processing, consulting, or investment advisory services.63 59 This exclusion, both in its current and original form, is thus intimately linked with the concept of an occurrence, and the time at which that occurrence needs to take place in order for coverage to exist; see generally ch 6. Where this exclusion is relied upon, issues of allocation commonly arise because the tort will often cause injury or damage subsequent to the Inception Date or Retroactive Date as well. Allocation is considered in ch 5, paras 5.42–5.54. 60 Exclusion B in all versions of the 004 Form. See too Exclusion (b) in earlier versions of the Form. 61 Citations to the statutes identified follow: FELA, 45 USC ss 51 ff; Jones Act, 46 USC ss 30104 ff; and Longshoremen’s and Harbor Workers’ Compensation Act, 33 USC ss 901 ff. 62 Article VI(C) of the 004 Form; and Article V(c) of earlier Forms. See ch 11 for a discussion of the ‘Conditions’ of the Bermuda Form. 63 Exclusion C in the 004 Form. See too Exclusion (c) in earlier versions of the Form.

10.37

180

The Exclusions

10.38

The Bermuda Form does not define ‘professional services’ in the Definitions section of the policy, but Exclusion C itself gives a non-exhaustive list (‘including but not limited to’) of such services. In determining whether a professional service is at issue, courts look to the nature of the conduct under scrutiny rather than the title or position of those involved, as well as to the underlying complaint. The question depends upon whether the relevant individuals ‘acted with the special acumen and training of professionals’ when they engaged in the relevant acts.64

10.39

The Bermuda Form seeks to make clear that it does not provide errors and omissions coverage for professional persons such as lawyers and accountants. In truth, it would appear unlikely that property damage would result from these and similar types of professional services, as opposed to architectural and engineering services. The latter are specifically exempted from this exclusion, so that liability for property damage arising from architectural and engineering services is in principle a covered risk, subject to the other conditions of the policy including the ‘Efficacy, Loss of Use’ exclusion. Many of the features of the Bermuda Form, including the provisions for London arbitration and modified New York law, are to be found in errors and omissions policies written by, for example, XL. These policies provide coverage for ‘wrongful acts’ and therefore differ from the Bermuda Form policy, which is concerned only with personal injury, property damage and advertising liability.

THE ‘OWNED PROPERTY; CARE, CUSTODY OR CONTROL, ETC’ EXCLUSION 10.40

This exclusion seeks to preclude coverage for liability from property damage resulting in a variety of situations: D. Property Damage to: (1) (2) (3) (4)

property owned or occupied by or rented to any Insured; property loaned to any Insured; property in the care, custody or control of any Insured; or that particular part of real property or fixtures on which any Insured or any contractors or sub-contractors working directly or indirectly on behalf of any Insured are performing operations, if such Property Damage arises out of such operations;

provided, however, that paragraphs (2), (3), and (4) of this Exclusion D do not apply to liability assumed under a railway sidetrack agreement; provided further that paragraphs (1) and (3) of this Exclusion D do not apply as respects damage to property of any Insured which is an Insured solely by virtue of paragraph (4) of Definition P, where such property is not owned or occupied by, rented to, or in the care, custody or control of any Insured which is an Insured other than by virtue of paragraph (4) of Definition P.65

64 Reliance Ins Co v Nat’l Union Fire Ins Co 691 NYS2d 458 (App Div 1999); Beazley Ins Co v Ace Am Ins Co 880 F3d 64 (2d Cir 2018) (applying New York law). 65 Exclusion D in the 004 Form. Exclusion (d) in earlier versions of the Bermuda Form, whilst not identical in all versions, also concerned ‘owned property’.

The ‘Owned Property; Care, Custody or Control, etc’ Exclusion

181

The terms of the exclusion have expanded since the original (001) version of the Bermuda Form. The exclusion precludes recovery for damage to the policyholder’s own property, damage that should be insured under the policyholder’s first-party property coverage. Thus, this exclusion helps reinforce the Bermuda Form’s function as third-party liability insurance.

10.41

This exclusion includes several provisions that constitute separate exclusions in standard form CGL policies sold in the United States and by other markets like the London insurance market. Sub-paragraphs (1) and (2) typically are addressed in US cases as the ‘owned property exclusion’. Sub-paragraph (3) is sometimes called and separately addressed as the ‘care, custody and control exclusion’. Sub-paragraph (4) is sometimes called the ‘particular part exclusion’ which is now included in more recent versions of the standard CGL policies sold in the United States. We use these titles in this section below as a way to organise New York cases that have applicability to the various sub-paragraphs of this exclusion. As with any questions of insurance policy interpretation, the applicability of any decision will depend on the terms of the provision and the facts at issue.

10.42

Exclusion D(1) and (2): The ‘Owned Property Exclusion’ New York courts have limited the reach of the owned property exclusion in cases in which the policyholder’s liability resulted in part from allegations that actions were necessary to prevent or mitigate damages or future damage or injury.66 Thus, the applicability of this exclusion as a defence to coverage may turn on a finding, or existence of allegations asserting, that property belonging to someone other than the policyholder was damaged or threatened with damage. The property may be any real estate, groundwater or fixtures, or other property owned by a third party. Focusing on the purpose of liability insurance, 66 The text of the exclusion used in current versions of standard form general liability insurance policies sold in the United States, states: This insurance does not apply to: J. Damage to Property ‘Property damage’ to: (1) Property you own, rent, or occupy, including any costs or expenses incurred by you, or any other person, organization or entity, for repair, replacement, enhancement, restoration or maintenance of such property for any reason, including prevention of injury to a person or damage to another’s property; (2) Premises you sell, give away or abandon, if the ‘property damage’ arises out of any part of those premises; (3) Property loaned to you; (4) Personal property in the care, custody or control of the insured; (5) That particular part of real property on which you or any contractors or subcontractors working directly or indirectly on your behalf are performing operations, if the ‘property damage’ arises out of those operations; or (6) That particular part of any property that must be restored, repaired or replaced because ‘your work’ was incorrectly performed on it. Paragraphs (1), (3) and (4) of this exclusion do not apply to ‘property damage’ (other than damage by fire) to premises, including the contents of such premises, rented to you for a period of 7 or fewer consecutive days. A separate limit of insurance applies to Damage To Premises Rented to You as described in Section III – Limits of Insurance.

10.43

182

The Exclusions

courts have held that the owned property exclusion does not bar coverage for the costs to clean up third-party property, but rather, consistent with law on product and other business risk exclusions, is reasonably interpreted to exclude coverage for damage to the policyholder’s own property. 10.44

Insurance companies have raised this exclusion as a defence to coverage in many environmental insurance coverage cases litigated in the United States. In those cases, the policyholders have sought to recover moneys spent to clean up environmental pollution (so-called ‘cleanup costs’ under United States law) under environmental statutes like the federal ‘Superfund’ statute67 and its state counterparts. Under those statutes and common law theories like nuisance, government agencies have typically sought to force policyholders to clean up, or to recoup from the policyholder moneys spent to clean up, environmental contamination emanating from the policyholder’s property that has contaminated or otherwise threatens to damage third-party property, including natural resources like groundwater or other property off-site.68 Insurance companies have contested coverage for such costs, alleging (among other things) that costs relating to damage to the policyholder’s own property is not covered.

10.45

Courts in New York addressing the owned property exclusion have upheld coverage when the clean-up is necessary to stop or prevent future damage.69 For example, in Savoy Medical Supply Co v F&H Manufacturing Corp,70 the court held that the policyholder’s CGL insurance applied to pay for a government-mandated cleanup of the policyholder’s own property, which was necessary to prevent damage to third parties. It found that the public’s interest in a swift cleanup of the contamination militated in favour of a limitation on the reach of the owned property exclusion. It reasoned that to hold otherwise might encourage the policyholder to delay cleanup until the contamination caused damage to a third party’s property: Suffolk County did not force the cleanup of the contamination for the benefit of F&H. Instead, concerns for public health and welfare provided the initial impetus. Therefore, the Court finds that the threat to the public due to F&H’s contamination of the sanitary system placed the damage outside the confines of the owned property exclusion. To rule otherwise and find that no third party damage could have occurred would necessitate an unrealistically narrow view of the extent of potential damage, as well as the gravity of environmental pollution in general.71

10.46

Courts in New York have also held that applying the owned property exclusion to bar coverage for costs needed to prevent future damage or injury would be unreasonable.

67 In 1980, the US Congress enacted the Comprehensive Environmental Response, Compensation & Liability Act (CERCLA), 42 USC ss 9601 ff, commonly known as ‘Superfund’. CERCLA and parallel state statutes for the first time imposed joint and several retroactive liability without any required finding of fault (in other words, strict liability) on any ‘potentially responsible party’ involved in a polluted site. 68 CERCLA also specifically allows for recovery of ‘natural resource damages’, damage to wildlife or the right to enjoy clean water, fish stocks etc for recreational and related purposes. 69 For a discussion of the law in various states across the United States interpreting the owned property exclusion, see Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) (hereafter Masters and Stanzler, Insurance Coverage Litigation) ss 14.07[A], 15.11[A]. See also the discussion of coverage for damages incurred to mitigate loss in ibid, ch 10. 70 Savoy Med Supply Co v F&H Mfg Corp 776 F Supp 703 (EDNY 1991). 71 Ibid at 709.

The ‘Owned Property; Care, Custody or Control, etc’ Exclusion

183

Courts reaching this result usually conclude that a contrary ruling would encourage parties to forego cleanup until more extensive damage has taken place, and that the public policy encouraging mitigation of damages supports limits on the reach of the exclusion. In Banker’s Trust Co v Hartford Accident & Indemnity Co,72 the government ordered the policyholder to remove oil that had leaked from tanks on its property to prevent harm to an adjacent river. The court held that the owned property exclusion did not prevent coverage, despite the fact that the policyholder owned the land that was cleaned up, because the cleanup prevented damage to others’ property: [T]he work done on the property to prevent further oil seepage was as a matter of law within the coverage of the policies. First, it was done to prevent damage to the property of third parties. There was no dispute that the oil seepage was not affecting the use of the property by its owners. Second … if the preventive work had not been done, the oil would have continued to seep into the river for a substantial period of time thereafter, and [the policyholder] … would have had to spend much more to clean up the resulting damage to the river and its shores.73

In subsequent authority, the New York Appellate Division has distinguished the situation at issue in cases like Banker’s Trust. In Castle Village Owners Corp v Greater New York Mutual Insurance Co,74 a retainer wall on property owned by the policyholder collapsed, causing large quantities of debris to cascade into the adjacent property and a roadway. The debris damaged passing automobiles and surrounding property and partially blocked the Henry Hudson Parkway, a major thoroughfare. The City issued an emergency declaration and performed emergency remediation to stabilise the area preventing further collapse and damage. The City demanded payment for the emergency work and ordered Castle Village to perform further work to repair the retaining wall. The primary insurance paid for the emergency work done by the City to prevent further damage, exhausting that coverage. The excess insurer, however, refused to pay the expenses of the work to rebuild the retaining wall; citing the owned property exclusion, the trial court rejected coverage. The Appellate Division agreed, finding that, once the emergency work was completed, in effect the damage was only to Castle Village’s property (the retaining wall) and ‘an imminent, continuing danger no longer existed’ to other property.75 The court contrasted that situation to that in oil spill cases like Banker’s Trust, which rejected the applicability of the owned property exclusion because such spills ‘presented a condition hazardous to the property of others [which] were not capable of being remedied without the performance of cleanup measures on the insured’s property’.76

72 Banker’s Trust Co v Hartford Accident & Indem Co 518 F Supp 371 (SDNY), vacated due to settlement, 621 F Supp 685 (SDNY 1981). 73 518 F Supp at 373. But see RD Maidman Family LP v Scottsdale Ins Co 783 NYS2d 205, 212 (Sup Ct 2004) (the trial court in New York included ‘coverage … for remediation expenses insured by plaintiffs’ own property to prevent further damages to property of a third party are not recoverable under the circumstances presented’). 74 Castle Village Owners Corp v Greater NY Mut Ins Co 878 NYS2d 311 (App 2009). Compare RD Maidman Family LP v Scottsdale Ins Co 783 NYS2d 205, 212 (Sup Ct 2004). The Appellate Division in Castle Village refused to follow RD Maidman, noting that as an appellate court, it was not bound by the trial court decision in RD Maidman, regardless of the similarity of the facts in the two cases (878 NYS2d 311 (App Div 2009)). 75 878 NYS2d 311, 315. 76 Ibid, discussing State of NY v NY Cent Mut Fire Ins Co 147 AD2d 77 (NY App Div 1989); Don Clark, Inc v US Fidelity & Guar Co 545 NYS2d 968 (Sup Ct 1989); and Gerrish Corp v Universal U’rs Ins Co 947 F2d 1023 (2d Cir 1991) (applying New York law).

184

The Exclusions

10.47

The holding in Banker’s Trust and similar cases accords with rulings by other courts in the United States. For example, in Intel Corp v Hartford Accident & Indemnity Co,77 a case decided under California law, the US Court of Appeals for the Ninth Circuit held that, when a policyholder is covered for damage to a third party’s property, the policyholder would reasonably expect coverage for efforts to mitigate the damage, even when the hazard originates on the policyholder’s own property. Thus, the court concluded that the monies expended to abate the public danger posed by the contamination fall outside the ambit of the exclusion.78

10.48

One New York court enforced CGL insurance for cleanup costs based on the mere threat of injury to a third party. In State v Central Mutual Fire Insurance Co,79 the court held that the owned property exclusion did not preclude coverage for the costs of cleaning up a spill of home heating oil because the oil entered, or threatened to enter, the groundwater that was the property of a third party: There is no question but that the fuel oil entered the groundwater, or at the very least threatened to, and since [the] groundwater is a natural resource protected by plaintiff as trustee for its people, the oil spill caused damage to property other than that of the insured, thereby triggering defendant’s obligation under its policy and the plaintiffs entitlement to summary judgment.80

10.49

Because environmental cleanup actions typically involve contamination of groundwater beneath the policyholder’s premises, the question of whether groundwater constitutes third-party property often arises. The law on water ownership differs from state to state.81 In some states like New York, groundwater belongs to the people of the state.82 The California Court of Appeals in Aerojet-General Corp v Superior Court83 explained that ‘ground and river waters’ are public property, and state and federal governments are owners of such property ‘for purposes of insurance coverage’: In this state, all ownership of water is usufructuary; water rights decisions ‘do not speak of ownership of water, but only the right to use …’. Unquestionably, the state and federal

77 Intel Corp v Hartford Accident & Indem Co 692 F Supp 1171 (ND Cal 1988), aff ’d in part, rev’d in part on other grounds, 952 F2d 1551 (9th Cir 1991). 78 952 F2d at 1565–66. See also In re Texas E Transmission Corp 870 F Supp 1293, 1342 (ED Pa 9 July 1992); Jones Truck Lines v Transport Ins Co No Civ A 88-5723, 1989 US Dist Lexis 5092 (ED Pa 1989) (applying Missouri law); United States v Conservation Chem Co 653 F Supp 152, 200–01 (WD Mo 1986). 79 State v Central Mut Fire Ins Co 542 NYS2d 402 (App Div 1989). 80 Ibid at 403–04 (internal citations omitted). See also New Castle County v Continental Cas Co 725 F Supp 800, 816 (D Del 1989) (an owned property exclusion does not bar coverage because cleanup is designed to remediate damage to off-site property ‘and the continued threat of off-site property damage’), aff ’d in part, rev’d in part on other grounds, 933 F2d 1162 (3d Cir 1991). Cf Shell Oil Co v Winterthur Swiss Ins Co 15 Cal Rptr 2d 815, 843 (Ct App 1993) (CGL coverage available to prevent imminent damage to third-party property). 81 See generally P Flood, ‘Water Rights of the Fifty States and Territories’ in KR Wright (ed) Water Rights of the Fifty States and Territories (Denver, American Water Works Association, 1990) 31. 82 See AIU Ins Co v Superior Ct (FMC Corp) 799 P2d 1253, 1261 fn 6 (Cal 1990) (citing s 102 of the California Water Code (all water within the state is the property of people of the state)); Colonial Tanning Corp v Home Indem Co 780 F Supp 906, 926 (NDNY 1991); State v Central Mut 542 NYS2d at 403–04 (citing ss 170, 172(12), (18) of the New York Navigation Law). Cf Bausch & Lomb Inc v Utica Mut Ins Co 625 A2d 1021 (Md 1993) (the state’s interest in groundwater was found to be regulatory, not proprietary). 83 Aerojet-General Corp v Superior Ct 257 Cal Rptr 621, supplemented on denial of reh’g, 258 Cal Rptr 684 (Ct App 1989).

The ‘Owned Property; Care, Custody or Control, etc’ Exclusion

185

governments are third-party property owners for the purposes of insurance coverage. Pollution of the ground and river waters is damage to public property, as well as a direct injury to public welfare.84

However, courts have also limited this principle, potentially expanding the reach of the exclusion. For example, another California court concluded that, even though private ownership rights in water may be limited by law, groundwater and soil can be under policyholder’s care, custody or control.85 In Olds-Olympic Inc v Commercial Union Insurance Co,86 the Washington Supreme Court held that the owned property exclusion did not apply to bar coverage. Olds-Olympic sought indemnification from its primary excess and general liability insurance companies, Commercial Union Insurance Company and Fireman’s Fund Insurance Company, for costs incurred in removing contaminated soil that had allegedly polluted the groundwater near its facility. The court held that the owned property exclusion did not apply because the groundwater at issue in the case was neither owned nor within the ‘care, custody or control’ of Olds-Olympic. It noted that the plain language of ‘care, custody, or control’ did not support the application of the exclusion in this case, because groundwater was not essential to Olds-Olympic’s business; groundwater is owned by the state in Washington, and the state did not grant a permit for use of the groundwater to Olds-Olympic; and there was no evidence that Olds-Olympic had asserted control over the groundwater.

10.50

As a general proposition, the Bermuda Form has not sought a complete exclusion of coverage for pollution liability and has, indeed, provided limited pollution coverage, which may be a selling point. Other provisions in the 004 Form refer to ‘Pollution’ and may interact with the owned property exclusion. Thus, Exclusion K, the Pollution Exclusion, contains various exceptions, preserving coverage for ‘Product Pollution Liability’ and for certain discharges meant to mitigate loss or damage or discovered immediately after a discharge begins, assuming the exclusion’s specific notice provisions are met.87

10.51

In the 001–003 Forms, the owned property exclusion (like others) is, in effect, qualified by the second paragraph of the Cross Liability Condition.88 This condition enables one policyholder to recover where it is liable for causing property damage to the property of another insured. For example, a situation may arise where company A and company B are both insured under the policy (as a result of the extended definition of ‘Insured’ contained in the Bermuda Form), and company A causes damage to the property of company B. In such a case, company A would still be able to recover under the policy. In the 004 version, however, the cross-liability provision has been amended to refer only to personal injury, and accordingly the cover would appear to be more restrictive.

10.52

84 257 Cal Rptr at 629 (internal quotation marks omitted). See also Gerrish Corp v Universal Underwriters Ins Co 947 F2d 1023 (2d Cir 1991) (applying Vermont law); Conservation Chemical 653 F Supp at 200; Maryland Cas Co v Wausau Chem Corp 809 F Supp 680, 693 (WD Wis 1992). 85 Shell v Winterthur 15 Cal Rptr 2d at 844. 86 Olds-Olympic Inc v Commercial Union Ins Co 918 P2d 923 (Wash 1996). 87 Exclusion K of the 004 Form, set out and discussed below in paras 10.107–10.124. 88 Article V(c) in the 001–003 Forms.

186

The Exclusions

Exclusion D(3): Care, Custody and Control Exclusion 10.53

The exclusion in the Bermuda Form is similar to that in the standard form CGL policy used in the United States. The care, custody and control exclusion in that form precludes coverage for ‘property damage to personal property in the care, custody or control of the insured’.89 Beginning with the ISO forms approved for use as of 1 January 1986, the ISO exclusion focuses on personal property. In contrast, the exclusion in the Bermuda Form does not make that is distinction, rather referring only to ‘property’, and thus may apply to both personal and real property.90 This conclusion also seems a logical result of the policy language in the Bermuda Form. The ‘particular part’ exclusion in Exclusion D(4) specifically refers to ‘real property or fixtures’, and the failure to make that distinction in the care, custody or control exclusion seems to be a deliberate one.

10.54

Case law in the United States on the care, custody and control exclusion typically focuses on the degree of control exercised by the policyholder. Earlier versions of the exclusion used in the United States, like the exclusion now used in the Bermuda Form, did not specify whether the property involved should be real or personal property. Thus, care should be taken in looking at the language in US cases to see whether the policy exclusion in question is limited to personal property, or, like the Bermuda Form, is broader. Neither the US version of the exclusion nor that in the Bermuda Form refers to ownership of the property in question, and the exclusion’s silence on this issue could lead to factual issues requiring resolution.

10.55

Earlier cases in the United States distinguished between ‘possessory handling’ by the policyholder and ‘proprietary control’. The exclusion thus typically applied under that line of cases when the court found that the property damaged was under the policyholder’s supervision and control and was a necessary part of the policyholder’s work. Some courts found these distinctions to be unmanageable and, as one court noted, ‘having stated these general principles, the cases then proceed to ride off in all directions’.91

10.56

Few cases in New York have addressed this exclusion. One that did involve a subcontractor who scratched windows while cleaning them and was sued for the damage.92 In the ensuing dispute over insurance coverage, the court found that the policyholder did not have ‘care, custody or control’ over the windows because the subcontractor had only temporary access to a building which itself was in control of the general contractor. The court therefore followed the rule that possession or control of real property arises only when the policyholder has exclusive control and rejected application of the exclusion, finding that the property in question was in the exclusive control of someone other than

89 For example, see Exclusion (j)(4) of ISO 1986 CGL Form, reproduced in A-32 (occurrence form) and A-42 (claims-made form) of Masters and Stanzler, Insurance Coverage Litigation (above n 69). In contrast, the 1966 and 1973 versions of the exclusion referred only to ‘property’; ibid A-22 (1973), A-13 (1966). Some policies in the United States may still use the pre-1986 ISO form exclusion that refers only to ‘property’. 90 See Exclusion D of the 004 Form. 91 Royal Indem Co v Smith 173 SE2d 738, 739 (Ga App 1970). 92 Klapper v Hanover Ins Co 240 NYS2d 284 (Sup Ct 1970).

The ‘Owned Property; Care, Custody or Control, etc’ Exclusion

187

the policyholder.93 Similarly, in Broome County v Travelers Indemnity Co,94 the Appellate Division held that property damage to a vehicle on display on the policyholder’s property was not excluded from coverage under the exclusion of liability for property damage to property in the care, custody or control of an Insured or as to which the Insured is ‘for any purpose exercising physical control’. The court concluded that the Insured’s control over its building was not sufficient to trigger the exclusion because the policyholder did not control the vehicle. According to the court, the requisite care, custody or control is ‘dependent upon the insured’s rights and acts concerning the specific damaged property out of which the claim of liability arises’.95 Under the traditional rule, when the property in question is under the supervision of the policyholder and such supervision is a necessary element of the policyholder’s work, courts find that the property is under the ‘care, custody or control’ of the policyholder. Under this rule, coverage is excluded when the facts show that the policyholder exercised supervision over the property.96

10.57

Exclusion D(4): The ‘Particular Part’ Exclusion The ‘particular part’ exclusion, contained in subparagraph (4) of Exclusion D of the Bermuda Form, specifically refers to ‘real property’, and thus by its terms precludes coverage for damage to ‘that particular part’ of real property on which the policyholder was working. The policy continues to cover damage to the existing parts of other property and to loss of use of other property.97

10.58

Few courts have addressed an exclusion in these or similar terms. Of those that have, courts have drawn a line between damage to existing property, which is covered, and damage to the policyholder’s work, which is a ‘business risk’ and is not covered. In Harbor Insurance Co v Tishman Construction Co,98 the Illinois Court of Appeal refused to apply the exclusion to preclude coverage in a case involving construction defects.99 There, the insurance company claimed that an exclusion similar to the one in the Bermuda Form

10.59

93 Ibid at 285–86. 94 County of Broome v Travelers Indem Co 88 AD2d 720 (NY App Div 1982), aff ’d, 58 NY2d 753 (NY 1982). 95 88 AD2d at 720 (citing Hardware Mut Ins Co v Mason-Moore-Tracy, Inc 194 F2d 173 (2d Cir 1952)). 96 For example, Overson v US Fidelity & Guar Co 587 P2d 149 (Utah 1978). 97 In US standard form general liability policies, the ‘particular part’ exclusion has two subparts, as follows: ‘This insurance does not apply to: … (property damage) to (5) That particular part of real property on which you or any contractors or subcontractors working directly or indirectly on your behalf are performing operations, if the “property damage” arises out of those operations; or (6) That particular part of any property that must be restored, repaired or replaced because ‘“your work’” was incorrectly performed on it.’ 98 Harbor Ins Co v Tishman Constr Co 578 NE2d 1197, 1200 (Ill App Ct 1991); see Masters and Stanzler, Insurance Coverage Litigation (above n 69) A-32 (1986 occurrence form), A-42 (1986 claims-made form). 99 The exclusion quoted in Harbor included the ‘own-work’ and ‘particular-part’ exclusions: ‘This insurance does not apply to: [P]roperty damage to … [2] that particular part of any property, not on premises owned by or rented to the insured … [a] the restoration, repair or replacement of which has been made or is necessary by reasons of faulty workmanship thereon by or on behalf of the insured; … [3] to property damage to work performed by the named insured arising out of the work or any portion thereof, or out of the materials, parts or equipment furnished in connection herewith.’ Ibid at 1200. This exclusion first appeared in the US CGL form via the ‘Broad Form Comprehensive General Liability Endorsement’, issued in July 1976, reproduced in Masters and Stanzler, Insurance Coverage Litigation (above n 69) A-29.

188

The Exclusions

precluded coverage for damage to property that did not include the policyholder’s work. In upholding the insurance company’s duty to defend, the court noted that Illinois courts ‘supported the rationale behind the defect [and resulting] damage distinction holding that recovery for the costs of the removal, repair and/or replacement [of defective products or work] … was not precluded by a clause excluding damage to the insured’s own work’.100 10.60

The US District Court of the District of Illinois reached the same conclusion when it interpreted a similar exclusion in WE O’Neil Construction Co v National Union Fire Insurance Co of Pittsburgh Pa.101 In O’Neil, the court recognised that the exclusion barred ‘coverage only for damage to the defective component itself, and not to damage to the remainder of the structure’.102 The ‘particular part’ language in the clause ‘only excludes damage to that “particular part” or parts upon which the Assured has performed faulty work’.103

10.61

Courts have also found that the reference to ‘operations’ in the US exclusion refers to ongoing operations, not to the policyholder’s completed operations, as defined in the ‘products completed operations hazard’. In the Bermuda Form, the ‘particular part’ exclusion is similarly applicable only to the particular part on which the insured or its subcontractors ‘are performing operations’.

THE ‘PRODUCTS LIABILITY’ EXCLUSIONS (‘EFFICACY, LOSS OF USE, ETC’) Introduction 10.62

Article IV(E) includes a number of exclusions that typically are described collectively as ‘business risk’ exclusions, a term applied to the exclusions in CGL insurance policies that seek to limit the coverage applicable to a policyholder’s products liability.104 As a general proposition, cases hold that business-risk exclusions do not apply to preclude coverage if the policyholder’s product has caused injury or damage to a third party’s person or property.105 The particular exclusions contained within the numbered paragraphs of Exclusion E bear resemblance, to a greater or lesser degree, to exclusions that in US case law typically are called the own product, own work and ‘sistership’ exclusions.106 It is important, however, when reading case law relating to exclusions of this kind to read and consider the text of the provision at issue, rather than to rely upon labels or names.

100 578 NE2d at 1202. 101 WE O’Neil Constr Co v Nat’l Union Fire Ins Co of Pittsburgh, Pa 721 F Supp 984 (ND Ill 1989). 102 Ibid at 995. 103 Ibid (quoting Blackfield v Underwriters at Lloyd’s, London 245 Cal App 2d 271, 275–76 (1966)). 104 For example, Stonewall 73 F3d at 1210–11 (referring to the product liability exclusions as ‘business risk exclusions’). Some cases in the United States refer to one of the product liability exclusions, the own product exclusion, as the ‘business risk exclusion’ – for example, Weedo v Stone-E-Brick Inc 405 A2d 788, 795 (NJ 1979); Vari Builders Inc v United States Fidelity & Guar Co 523 A2d 549, 551 (Del Super Ct 1986). 105 See, eg, Masters and Stanzler, Insurance Coverage Litigation (above n 69) s 14.07, and cases and discussion therein. 106 See, eg, ibid.

The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)

189

This may be particularly so in the context of a dispute arising under the Bermuda Form, where a tribunal is required to apply the principles of interpretation set out in the governing law provision.107 When issues arise as to the application of Exclusion E to any particular set of facts, it is often helpful to begin the analysis by identifying, at the outset, the reason why the facts give rise to coverage pursuant to Article I of the Bermuda Form. In particular, the application of ‘business risk’ exclusions is often linked to the question of whether ‘property damage’ or bodily injury has taken place and what the nature of that damage or injury is.108 Thus, courts in New York (and in other jurisdictions in the United States) have held – when considering ‘business risk’ exclusions in CGL policy forms used in the United States – that the insurance company cannot meet its burden to show that one of these exclusions applies if the policyholder’s product or work caused damage to a third party’s person or property.109 By contrast, when the complaint against the policyholder merely alleges that its products have failed to perform in the way that was expected, courts have often denied coverage. Cases and the literature have explained that general liability insurance is not intended to guarantee a policyholder against defects in its work or in the products themselves (which are ‘business risks’), but only for injury or damage that the work or product may cause to third parties.110 Thus, courts in New York have said that general liability insurance policies are not ‘performance guarantees’.111

10.63

Exclusion E(1): Failure to Perform Exclusion Exclusion E(1) seeks to preclude coverage for the liability of the Insured arising out of the failure of any Insured’s Products or of work, including architectural or engineering services, by or on behalf of any Insured to meet any warranty or representation by any Insured as to the level of performance, quality, fitness or durability or to perform their function or serve their purpose, to the extent that such liability is for the diminished value or utility of any Insured’s Products or work by or on behalf of any Insured.112

The first part of this exclusion, with its references to ‘the level of performance’, finds parallels in the part of the exclusion often called the ‘loss of use’ or ‘failure to perform’

107 See above ch 4. 108 See the discussion of ‘property damage’ in ch 5, paras 5.58–5.64 above. 109 Masters and Stanzler, Insurance Coverage Litigation (above n 69) s 14.07. 110 See, eg, Weedo 405 A2d at 795. See also Roger C Henderson, ‘Insurance Protection for Products Liability and Completed Operations: What Every Lawyer Should Know’ (1971) 50 Nebraska Law Review 415, reprinted in (1971) 21 Law Review Digest 41 and (1971) 14 Personal Injury Commentator 322 (available at Hein Online (www.heinonline.org)). 111 See Maryland Cas Co v WR Grace & Co 794 F Supp 1206, 1226 (SDNY 1991) (citing Parkset Plumbing & Heating Corp v Reliance Ins Co 448 NYS2d 739, 740 (App Div 1982)), supplemented by No 83 Civ 7451, 1991 WL 283547 (SDNY 1991), rev’d on other grounds, 23 F3d 617 (2d Cir 1993); Zandri Constr Co v Firemen’s Ins Co of Newark 440 NYS2d 353, 355 (App Div 1981), aff ’d, 446 NYS2d 45 (NY 1981). See Laquila Constr Inc v Travelers Indem Co of Ill 66 F Supp 2d 543 (SDNY 1999), aff ’d, 216 F3d 1072 (2d Cir 2000) (unpublished table decision); Ogden v Travelers 681 F Supp 169. 112 Exclusion E(1) of the 004 Form. Exclusion (e)(1) in earlier versions of the Bermuda Form used very similar terms.

10.64

190

The Exclusions

exclusion in US cases and policy wordings or forms,113 while the second part of the exclusion, with its reference to ‘failure to perform’, finds parallels with a provision in US standard form policies that typically has been referred to as ‘design defect’ or ‘business risk’ exclusion.114 10.65

Unlike New York case law addressing the loss of use/failure to perform and design exclusions, both parts of Exclusion E(1) are modified by the words providing that coverage does not apply ‘to the extent that such liability is for the diminished value or utility of any Insured’s Products or work by or on behalf of any Insured’. Under both the terms of the exclusion, with its ‘to the extent’ provision, and under New York case law, coverage continues to apply in any case in which bodily injury or property damage has in fact taken place or if the underlying claims include allegations such as negligence and strict liability that involved true damage or injury. For example, US cases have found that the loss of use/failure to perform exclusion excludes coverage only for pure loss of use and, thus, does not apply when physical injury to property other than the policyholder’s work or product has taken place.115 In a situation that simply involves ‘diminished value or economic utility’, it is perhaps unlikely that a policyholder will actually be able to allege that ‘property damage’ (or indeed personal injury or advertising liability) has taken place as required

113 The loss of use exclusion first appeared in the standard form CGL policy adopted for use in the United States on 1 January 1973. The relevant part of this exclusion, for our purposes, provides as follows: This insurance does not apply: (m) to loss of use of tangible property which has not been physically injured or destroyed resulting from … (2) failure of the named insured’s products or work performed by or on behalf of the named insured to meet the level of performance, quality, fitness or durability warranted or represented by the named insured; but this exclusion does not apply to loss of use of other tangible property resulting from the sudden and accidental physical injury to or destruction of the named insured’s products or work performed by or on behalf of the named insured after such products or work have been put to use by any person or organization other than an insured. Quoted from the 1973 CGL Form, reproduced in Masters and Stanzler, Insurance Coverage Litigation (above n 69) A-22. The exception contained in the US exclusion does not appear in the Bermuda Form exclusion. 114 Some general liability umbrella and excess policies containing the so-called design exclusion appeared in the 1966 Form used in the United States and some general liability umbrella and excess policies. Some excess insurance policy forms continued to include the design exclusion after it was removed from the CGL policy in 1973. The exclusion typically provides: This policy shall not apply: (k) to bodily injury or property damage resulting from the failure of the named insured’s products or work completed by or for the named insured to perform the function or serve the purpose intended by the named insured, if such failure is due to a mistake or deficiency in any design, formula, plan, specifications, advertising material, or printed instructions prepared or developed by the named insured; but this exclusion does not apply to bodily injury or property damage resulting from the active malfunctioning of such products or work. Quoted from the 1996 CGL Form, reproduced in Masters and Stanzler, Insurance Coverage Litigation (above n 69) A-14. Both Weedo in New Jersey and Vari Builders in Delaware refer to the ‘business risk’ exclusion. However, the policy exclusion quoted in those opinions and actually construed by the court is not the design defect exclusion under discussion here, but the own product/work exclusion: Weedo, 405 A2d at 792; Vari Builders, 523 A2d at 550–51. 115 See, eg, Masters and Stanzler, Insurance Coverage Litigation (above n 69) s 14.07[C], [D].

The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)

191

by the policy’s coverage clause: see Chapter 5 above. Conversely, when property damage actually has taken place, the policyholder’s liability will usually go well beyond liability for the diminished value or economic utility of the insured’s products or work completed by the insured. Exclusion E(1) may have some scope for operation in circumstances involving property damage, viz when the claim against the policyholder includes, as one element, a claim for the diminished value or utility of the policyholder’s products in circumstances when the product had failed to meet a warranty of representation by the policyholder as to the level of performance of that product (or a warranty or representation as to the other matters set out in the clause). In these circumstances, a stronger argument for application of the exclusion arguably arises with regard to that part of the loss.

10.66

Whilst the wording of Exclusion E(1) differs from both the ‘loss of use’ exclusion and the ‘design defect’ exclusion, the case law relating to those exclusions may provide some general guidance as to the approach to Exclusion E(1).

10.67

Few cases in the United States address the loss of use exclusion. The few that have addressed it typically have declined to apply it when tangible property has been physically injured. In perhaps the most frequently cited case on the issue, Todd Shipyards Corp v Turbine Services Inc,116 contractors sought coverage for work in defectively repairing a turbine of a ship, causing extensive damage to the ship’s turbines. Applying Louisiana law, the US Court of Appeals for the Fourth Circuit refused to apply the loss of use exclusion ‘for the simple reason that the [ship] did sustain physical injury to its LP turbine as a result of the failure of the insured’s work product to meet the level of performance impliedly warranted by the insureds’.117 We consider that, on these facts, the same result would be reached in relation to the wording of Exclusion E(1): the contractors’ claim would not be for the ‘diminished value or utility’ of its products.

10.68

In Stonewall Insurance Co v Asbestos Management Corp,118 a federal trial court applying New York law held that the exclusion did not apply because the policyholder’s liability for asbestos property damage liability did not arise from a failure of the product to insulate, but rather from the damage it caused to claimants’ property:

10.69

[T]he exclusion is inapplicable because [the policyholder] NGC’s liability does not arise from the failure of its products to perform their fireproofing functions. Rather, the claimants allege that, in serving its intended purpose, the insulation product released asbestos fibers, causing property damage to the buildings and contents therein. In addition, the exclusion does not apply where the property, the use of which is lost, has been physically injured. NGC’s liability arises out of physical injury to property by virtue of the release and re-entrainment of asbestos fibers.119

116 Todd Shipyards Corp v Turbine Serv Inc 674 F2d 401 (5th Cir 1982) (applying Louisiana law). 117 674 F2d at 418. Alternatively, the court held that the case fell within the ‘sudden and accidental’ exception to the exclusion, a holding not relevant under the Bermuda Form version of the exclusion. 118 No 86 Civ 9671 (JSM), 1992 US Dist Lexis 7607 (SDNY 1992), aff ’d in part, rev’d in part sub nom Stonewall Ins Co v Asbestos Mgmt Corp 73 F3d 1178 (2d Cir 1995), modified on denial of reh’g, 85 F3d 49 (2d Cir 1996). 119 73 F3d at 1212. See also Int’l Hormones Inc v Safeco Ins Co of Am 394 NYS2d 260 (App Div 1977).

192

The Exclusions

In contrast, another court applied the exclusion to a policyholder that made allegedly defective insulating panels used in the construction of a building. The court found that no ‘property damage’ took place to any property other than the policyholder’s own insulating panels.120 10.70

Courts have refused to apply the ‘design defect’ exclusion when the damage may have resulted from a production or installation error, or when the product allegedly caused affirmative harm. The New York Court of Appeals addressed the ‘design defect’ exclusion, and at least some of the language found in the second part of the Bermuda Form exclusion, in Sturges Manufacturing Co v Utica Mutual Insurance Co.121 The court there, relying in part on the ‘active malfunctioning’ exception not found in the Bermuda Form exclusion, despaired of construing the exclusion: If underwriters know what this so called standard form clause means, the average insured probably does not, and this court most certainly does not. It is for this reason, of course, that exclusionary clauses, when doubtful of meaning, are construed in favor of the insured … Hence, the insurer cannot rely on exclusionary clause ‘K’ in declining to defend Sturges.122

Although Sturges may have limited utility with regard to the language in Exclusion E(1), it stands for the proposition that the insurer bears the burden to show, under the facts, that the exclusion applies. Some courts have applied the exclusion after the insurance company showed that the insured product failed due to a design defect.123 The insurance company cannot meet that burden, however, if the policyholder’s liability is due, in whole or in part, to a production or other error; not a design defect, failure to meet a warranty, or failure to perform, and nothing else. 10.71

For example, in United States Fidelity & Guaranty Co v Mayor’s Jewelers of Pompano Inc,124 a burglary took place at a jewellery store when a thief entered its building by breaking a plate glass window. Investigations revealed that ‘riot glass’ had not been installed properly by the policyholder, a glass company. The jewellery store sued the policyholder for breach of contract, negligence and breach of implied warranty. The court held that the design or business-risk exclusion, called exclusion (k) in the 1966 Form approved for use in the United States, did not bar coverage because negligent installation, not a design defect, was alleged: Exclusion (k) can be divided into two parts: the first part being what is commonly referred to as the ‘business risk’ exclusion, and the second part being an exception to the exclusion. USF&G argues that under the facts in this case, coverage is excluded because its insured’s product (the glass) failed to perform the function or serve the purpose intended by the insured and there

120 Mapes Indus Inc v US Fidelity & Guar Co 560 NW2d 814 (Neb 1997). The court also found the ‘sudden and accidental’ exception did not apply to preserve coverage because the delamination of its panels did not result from a ‘sudden and accidental’ event. Ibid at 819–20. 121 371 NYS2d 444 (NY 1975). 122 371 NYS2d at 448–49; Lowenstein Dyes & Chems Inc v Aetna Life & Cas Co 524 F Supp 574, 579 (EDNY 1981), aff ’d, 742 F2d 1437 (2d Cir 1983) (unpublished table decision). 123 For example, Pittway Corp v American Motorists Ins Co 370 NE2d 1271 (Ill App Ct 1977). 124 US Fidelity & Guar Co v Mayor’s Jewelers of Pompano Inc 384 So 2d 256 (Fla Dist Ct App 1980).

The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)

193

was no active malfunctioning of the glass to bring it within the exception. It concludes that the failure to perform had to arise from a mistake or deficiency in the plans, design or specifications. USF&G overlooks, however, the allegations in the complaint. It is not alleged that the product failed. It is alleged that the insured negligently installed the wrong type of glass. We therefore hold exclusion (k) inapplicable to the facts here.125

In United States Fidelity & Guaranty Co v Nevada Cement Co,126 the court upheld coverage when the policyholder allegedly furnished defective cement to a concrete supplier which used it in concrete supplied to a general contractor.127 After the general contractor poured the concrete, inspection tests revealed the deficiency. The deficiency caused construction delays and forced the contractor to add supportive shoring to the structure. The contractor sued the policyholder for the additional construction expenses, and the policyholder sought coverage for the alleged damage. The court held that the exclusion ‘has no application where, as here, the product’s failure to serve its intended purpose results from a “production error”, as contrasted to a “design error”’.128 In contrast, in Pittway,129 the policyholder allegedly designed defective valves for use in aerosol cans. The court noted that the policyholder’s valves were designed after the policyholder had received notice that the previous design was defective.130 Therefore, the court found that the damage resulted from a design defect, and the exclusion barred coverage.131

10.72

These cases set forth general principles that arguably can aid in the approach to Exclusion E(1) in the Bermuda Form. The ‘to the extent’ phrase in the Bermuda exclusion makes clear, consistent with the case law discussed in this section, that the policy does not apply to cover liability due to diminished value or utility of the insured’s products or work. If the policyholder’s product causes damage to a claimant’s person or property, then the exclusion does not apply to preclude coverage.

10.73

Exclusion E(2): Exclusion for Property Damage to the Insured’s Products or Work The second exclusion is what the US cases call the own product or own work exclusion.132 It seeks to preclude coverage for a policyholder’s liability as follows: This insurance shall not apply: … without limiting paragraph (1) of this Exclusion E, in respect of Property Damage to any Insured’s Products or of work, including, without limitation, architectural or engineering

125 Ibid at 258. 126 US Fidelity & Guar Co v Nevada Cement Co 561 P2d 1335 (Nev 1977). 127 Ibid at 1336. 128 Ibid at 1338. 129 370 NE2d 1271 (Ill Ct App 1977). 130 Ibid at 1276. 131 Ibid. 132 In Ogden v Travelers 681 F Supp 169, discussed para 10.77 below, the court referred to a similar clause as a ‘work product exclusion’.

10.74

194

The Exclusions

services, performed by on behalf of any Insured, if such Property Damage arises out of any portion of such products or work, or out of materials, parts or equipment furnished in connection therewith.133

10.75

The CGL policy has a separate exclusion for liability for a policyholder’s own products and one for a policyholder’s own work. In contrast, the Bermuda Form combines the two exclusions. Application of the exclusion requires analysis of the facts leading to the policyholder’s liability. New York courts, like courts generally in the United States, typically refuse to apply the exclusion when the policyholder’s product or work has caused damage to another’s property, but have used it to exclude coverage after concluding that only the policyholder’s own property or work was damaged.134

10.76

In Stonewall Insurance Co v Asbestos Claims Management Corp,135 the court considered the applicability of an ‘own product’ exclusion to asbestos property damage claims. There, the policyholder sought coverage for liability arising from the installation of its asbestos containing products into buildings, and the court held that property damage took place at the time of installation. The insurance companies argued that coverage did not apply under an exclusion for: property damage to the named insured’s products arising out of such products or any part of such products; [or] … property damage to work performed by or on behalf of the named insured arising out of the work or any portion thereof, or out of materials, parts or equipment furnished in connection thereof.136

The court found the exclusion inapplicable because the asbestos in building claims were seeking to recover for damage to property other than the policyholder’s products.137 10.77

The exclusion in the 004 Form applies ‘if such Property Damage arises out of any portion of such products or work, or out of materials, part or equipment furnished in connection therewith’. Ogden Corp v Travelers Indemnity Corp138 illustrates how this exclusion operates in a case involving property damage to the policyholder’s products. In that case,

133 Exclusion E(2) in the 004 Form. In the 001 and 002 Forms, the exclusion was similar in some respects, but drafted differently, seeking to preclude coverage: ‘on account of property damage to any portion or section of the Insured’s products or of work performed by or on behalf of the Insured, if such property damage arises out or is alleged to arise out of that portion of such products or that section of work, or out of materials, parts or equipment furnished in connection therewith’. The exclusion in the 003 Form was very similar to that in 001 and 002. 134 See, eg, decisions in Stonewall No 86 Civ 9671 (JSM), 1992 US Dist Lexis 7607, at *58–*60, aff ’d, 73 F3d 1178, 1210 (2d Cir 1995) (applying New York law); Lowville Producer’s Dairy Co-Op Inc v Am Motorists Ins Co 604 NYS2d 421 (App Div 1993); Apache Foam Prods v Continental Ins Co 528 NYS2d 449 (App Div 1988); Maryland Casualty v WR Grace 23 F3d at 627 (applying New York law). 135 73 F3d at 1211. For the decision of the trial court, see 1992 US Dist Lexis 7607. 136 73 F3d at 1210 (alterations in original). 137 Ibid at 1210. See also IJ White Corp v Columbia Cas Co 105 AD3d 531, 532 (NY App Div 2013) (CGL policies ‘do insure against property damage caused by faulty workmanship to something other than the work product’); QBE Ins Co v Adjo Contracting Corp 121 AD3d 1064 (NY App Div 2014) (CGL insurance ‘does not insure for damage to the work product itself, it insures “faulty workmanship in the work product which creates a legal liability by causing bodily injury or property damage to something other than the work product”’, quoting George A. Fuller Co v US Fidelity & Guar 200 AD2d 255, 259 (NY App Div 1994)). For a further discussion of the own product and own work exclusions, see Masters and Stanzler, Insurance Coverage Litigation (above n 69) s 14.07. 138 681 F Supp 169 (SDNY 1988).

The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)

195

defects in a vessel constructed by the policyholder caused it to sink on its maiden voyage. The court held that the ‘work-product’ exclusion – which was very similar to Exclusion E(2) in the Bermuda Form – precluded recovery by the policyholder for its liability to the purchaser for the damage to the vessel purchased. The exclusion also applies to preclude coverage for liability for property damage to ‘work … performed by or on behalf of any Insured’. This damage might well result from construction work by the policyholder, and might include architectural or engineering services. If, in this context, the policyholder’s work results in property damage to its work, then the exclusion may apply. By contrast, if the policyholder supplies and installs a defective component into a complex structure such as a rig, and the component causes damage to the structure as a whole, the exclusion will not apply.139 In that scenario, the property damage will be to the rig itself, not simply to the policyholder’s own products or work. Factual questions may arise as to what constitutes a ‘portion of work’ for the purposes of this exclusion. Exclusion D(4),140 the ‘particular part exclusion’, may also be relevant, depending upon the facts.

10.78

Exclusion E(3): The ‘Sistership Exclusion’ The third exclusion in the 004 Form excludes liability of the insured for

10.79

the costs incurred for the withdrawal, inspection, repair, recall, return, replacement or disposal of any Insured’s Products or work, including, without limitation, architectural or engineering services, or, in connection with any of the foregoing, loss of use thereof; provided, however, that this paragraph (3) shall not apply in respect of costs incurred for the withdrawal, inspection, repair, recall, return, replacement or disposal of products or work of a party other than an Insured of which the Insured’s Products or work forms a part.141

The exclusion in the 004 Form replaced the exclusion contained in earlier versions of the Bermuda Form, that sought to exclude liability: for the withdrawal, inspection, repair, replacement, or, in connection with any of the foregoing, loss of use of the Insured’s products or work completed by or for the Insured or of any property of which such products or work form a part …142

139 For a discussion of the application of the own work, own product exclusion to faulty construction, see National Union Fire Insurance Co of Pittsburgh, Pa v Turner Construction Co 117 AD3d 103 (NY App Div 2014). The court there held that the own work exclusion precluded coverage for faulty work by subcontractors hired by the insured general contractor on the ground that ‘the entire project is the general contractor’s work’; ibid at 107 (citing Firemen’s Ins Co v Nat’l Union Fire Ins Co of Pittsburgh, Pa 904 A2d 754, 760–61 (NJ App Div 2006)) (applying New Jersey and New York Law which ‘are consistent as to the issues in dispute’). This result does not pertain to policies that contain a ‘subcontractor’s exception’; where such an exception is not present, an entire building or construction may be considered the policyholder’s ‘own work’ as the exclusion precludes coverage if there is no third-party property damage. See also Exeter Bldg Corp v Scottsdale Ins Co, 913 NYS2d 733–36 (NY App Div 2010); George A Fuller Co v US Fidelity & Cas Co 613 NYS2d 152, 155 (NY App Div 1994). See also Masters and Stanzler, Insurance Coverage Litigation (above n 69) s 14.07[A][5]. For the kind of problems that can arise in a somewhat different context, see The Nukila [1997] 2 Lloyd’s Rep 146 (CA). 140 See above paras 10.58–10.61. 141 Exclusion E(3) of the 004 Form. 142 Exclusion (e)(3) of the 001–003 Forms.

10.80

196

The Exclusions

The exclusion in the earlier versions of the Bermuda Form bears similarities to the clause commonly known as the ‘sistership’ exclusion, which typically provides:143 This insurance shall … not apply: to damages claimed for the withdrawal, inspection, repair, replacement, or loss of use of the named insured’s products or work completed by or for the named insured or of any property of which such products or work form a part, if such products, work or property are withdrawn from the market or from use because of any known or suspected defect or deficiency therein.144

10.81

The sistership exclusion was intended to preclude insurance coverage for expenses incurred in withdrawing products from the market or from use because of a known defect in a ‘sister’ product.145 The colloquial name ‘sistership exclusion’ reflects that purpose, and derived from the practice in the aircraft industry of recalling planes for repairs when a plane of the same model – a sister ship – had crashed because of a design defect.146 Courts have applied the exclusion to bar coverage only for the costs incurred when a policyholder recalls a product that, it is suspected, may cause injury or damage. Thus, one court in the United States observed: The sistership clause was developed to protect insurers against liability for the cost of recalls. The clause’s name, in fact, reflects this purpose. Following an accident involving a defective airplane, the airplane manufacturer became obligated to recall the airplane’s sisterships in order to correct the common defect that caused the crash of the first airplane. Insurance companies subsequently developed the ‘sistership’ clause to make clear that, while they intended to pay for damages caused by a product that failed, they did not intend to pay for the costs of recalling products containing a similar defect that had not yet failed.147

10.82

A number of New York cases have interpreted ‘sistership’ exclusions. For example, the exclusion was held not to apply when the policyholder’s product allegedly caused damage to the property of others.148 Courts have also held that the exclusion applied to preclude coverage only for liability caused by the policyholder’s own withdrawal of its product

143 See the clause set out in Masters and Stanzler, Insurance Coverage Litigation (above n 69) s 14.07[B]. See, too, an article, containing a detailed review of the US case law on the sistership exclusion, by Jean E Maess, ‘Validity and Construction of “Sistership” Clause of Products Liability Insurance Policy Excepting from Coverage Cost of Product Recall or Withdrawal of Product from Market’ (1984) 32 American Law Reports 4th 630, superseded by Marjorie A Shields, ‘Construction and Application of “Sistership” Clause of Product Liability Insurance Policy Excepting Form Coverage Cost of Product Recall or Withdrawal of Product from Market’ (2009) 49 American Law Reports 6th 169. 144 As found in US CGL policies. See, eg, Masters and Stanzler, Insurance Coverage Litigation (above n 69) A-22 (1993 Form), A-33 (1986 occurrence form), A-43 (1986 claims-made form) (1973 CGL Form). 145 See, eg, Armstrong World Indus Inc v Aetna Cas & Sur Co 52 Cal Rptr 2d 690 (App Ct 1996). See ibid s 14.07[B]. 146 Ibid at 729. See also the seminal US case on the exclusion, Todd Shipards 674 F2d at 419. 147 Forest City Dillon Inc v Aetna Cas & Sur Co 852 F2d 168, 173 (6th Cir 1988). Stonewall 73 F3d at 1211; Imperial Cas & Indem Co v High Concrete Structures Inc 858 F2d 128, 136–37 (3d Cir 1988); Gulf Miss Marine Corp v George Engine Co 697 F2d 668, 674 (5th Cir 1983); Carey-Canada Inc v Aetna Cas & Sur Co No Civ A 84-3113, 1988 US Dist Lexis 8997, at *41–*42 (DDC 1988); Dayton Indep Sch Dist v National Gypsum Co 682 F Supp 1403, 1412 (ED Tex 1988), rev’d on other grounds sub nom WR Grace & Co v Continental Cas Co 896 F2d 865 (5th Cir 1990). 148 Eg, Stonewall 73 F3d at 1211; Maryland Cas Co v WR Grace & Co 794 F Supp 1206, 1227 (SDNY 1991); Thomas J Lipton 357 NYS2d at 707.

The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)

197

from the market, and not to withdrawal by the policyholder’s customer.149 Questions arise, however, as to the extent to which this case law applies to the Bermuda Form. The language of the 004 Form is not identical to that considered in the US cases. For the reasons already given, it is important not to approach the process of interpretation simply by attaching a label to a clause, and then reasoning from the label that has been attached.150 Furthermore, some decisions, holding that the exclusion applies only to the withdrawal or true product recall by the policyholder, appear to be based on principles of interpretation151 that are, arguably, contrary to those to be applied pursuant to the modified New York governing law provision.152 Under the case law, courts typically have not applied the sistership exclusion unless two requirements are met: (1) the withdrawal must be made by the policyholder; and (2) the withdrawal must take place before actual injury or damage takes place.153 In Stonewall, the court found, under New York and Texas law, that the sistership exclusion did not apply because the policyholder had ‘not itself withdrawn or removed its products from the market’.154 It also held that the exclusion did not apply because property other than the policyholder’s product had been damaged: Moreover, even if damage caused by asbestos precipitated a discontinuance of NGC’s [the policyholder’s] products, the exclusion would not preclude coverage for damaged property other than NGC’s products. [Citations omitted] The District Court recognized that NGC’s liability, as alleged in the complaints, arises out of damage that its products have caused to third-party property.

One of the insurance companies in Stonewall cited another version of the sistership exclusion which sought to preclude coverage for such part of any damages or expense representing the cost of inspecting, repairing, replacing, removing, recovering, withdrawing from use or loss of use of, because of any known or suspected defect or deficiency therein, any (1) goods or products or any part thereof (including any container) manufactured, sold, handled or distributed by the named insured or others trading under his name; or (2) work completed by or for the named

149 Truax & Hovey Ltd v Aetna Cas & Sur Co 504 NYS2d 934 (App Div 1986); Stonewall 73 F3d at 1211 (citing Thomas J Lipton 357 NYS2d at 707). See also Int’l Hormones Inc v Safeco Ins Co of Am 394 NYS2d 260 (App Div 1977). 150 In Thomas J Lipton, the New York Court of Appeals described as a ‘sistership exclusion’ a clause that differs from that discussed in the text above. The clause there made it a condition of the insurer’s liability that the policyholder ‘shall promptly take at his expense all reasonable steps to prevent other bodily injury or property damage from arising out of the same or similar conditions, but such expense shall not be recoverable under this policy’. 357 NYS2d at 707 n marked ‘*’. 151 See, eg, ibid at 708. 152 See Article V(q) of the 001–03 Forms; and Article VI(O) of the 004 Form. See, further, ch 4, paras 4.28–4.32 above; and paras 10.06–10.10 above. 153 For example, Maryland Casualty v Grace 23 F3d at 626; Todd Shipyards 674 F2d at 419; Gulf Mississippi Marine 697 F2d at 673–74; Thomas J Lipton 357 NYS2d at 707. See also 3 New Appleman on Insurance: Law Library Edition, Exclusion ‘n’: Recall of Products, Work or Impaired Property Exclusion – The ‘Sistership Exclusion’, s 18.03[14] (New York, LexisNexis, 2021); superseded by Marjorie A Shields, ‘Construction and Application of “Sistership” Clause of Product Liability Insurance Policy Excepting form Coverage Cost of Product Recall or Withdrawal of Product from Market’ (2009) 49 American Law Reports 6th 169. 154 Stonewall 73 F3d at 1211 (citing Thomas J Lipton 357 NYS2d at 707; Parker Prods Inc v Gulf Ins Co 486 SW2d 610, 612–15 (Tex Ct App 1972), aff’d, 498 SW2d 676 (Tex 1973)).

10.83

198

The Exclusions

insured; or (3) other property of which such goods, products or work completed are a component part or ingredient.

The insurance company argued that this version of the sistership exclusion precluded coverage because all of the costs at issue related to asbestos products that were a component of the claimant’s building. The Stonewall court rejected this position, finding that the umbrella insurance policies in question did not define the term ‘defect’ (arguably an ambiguity construed against the insurance company, and thus a holding that would be subject to examination under the Bermuda Form governing law provision). In addition, the court noted that ‘the claimants do not allege that NGC’s products failed to perform their insulating or fire retardant functions’, and thus did not involve a ‘deficiency’ in intended use.155 10.84

The exclusion applies to preclude coverage when the policyholder has withdrawn from the market products that have not failed. Thus, the sistership exclusion has been applied to preclude coverage in classic recall situations. For example, an Illinois federal court relied on the exclusion, in part, to deny coverage for the policyholder’s costs of a nationwide recall after seven people had died in the United States from ingesting Tylenol. In McNeilab Inc v North River Insurance Co,156 the insurance companies rejected coverage, asserting that the facts involved a classic recall. The court arguably ignored the seven deaths that had taken place and that the policyholder’s actions were undertaken to prevent further injury. It instead found persuasive the fact that the policyholder had rejected an opportunity to buy product recall insurance, rejecting coverage.157

10.85

However, some courts have held that the sistership exclusion bars coverage even when a third party, not the policyholder, withdraws the policyholder’s product from use.158 In Glass-Lined Pipe, only the policyholder’s product, the glass lining for pipe, was damaged. The court found no damage to any property except the policyholder’s product: ‘there was no damage caused to any other parts of the project because the [glass-lined] pipe did not function properly, no walls were taken down and no damage was done to the pipe itself’.159 The court recognised that its holding did ‘not affect other possible instances of product liability, such as those in which damage occurs … to property other than the product of the insured itself’.160 The case did not apply New York law. As discussed elsewhere,161 under New York law, covered ‘property damage’ occurs when the policyholder’s defective product is incorporated into other property.162 Law in other states does not agree with this.163

155 Ibid (citing US Fidelity & Guar Co v Wilkin Insulation Co 578 NE2d 926, 935 (Ill 1991)); Dayton Indep Sch Dist v National Gypsum Co 682 F Supp at 1412. 156 McNeilab Inc v North River Ins Co 645 F Supp 525 (DNJ 1986), aff ’d, 831 F2d 287 (3d Cir 1987) (unpublished table decision). 157 645 F Supp at 540. 158 Commercial Union Assur Co v Glass-Lined Pipe Co 372 So 2d 1305, 1309 (Ala 1979). 159 Ibid at 1308. 160 Ibid at 1309. 161 See above ch 5, paras 5.62–5.63. 162 Sturges Mfg Co v Utica Mut Ins Co 371 NYS2d 444, 446–47 (NY 1975); Thomas J Lipton 357 NYS2d at 707. 163 See the comparison of New York law, which applies the incorporation doctrine, with Illinois law, which does not, in Travelers Ins Co v Eljer Mfg Inc 757 NE2d 481 (Ill 2001). Under the incorporation doctrine,

The ‘Products Liability’ Exclusions (‘Efficacy, Loss of Use, etc’)

199

Exclusion E(4): Decline in Value Exclusion The final paragraph of Exclusion E of the 004 Form164 seeks to exclude the liability of the insured:

10.86

in respect of decline of value of real or personal property to the extent such decline in value is attributable not to physical damage or destruction thereof but to proximity to continuing operations, activities or equipment which limit the usage of such property or make occupation of such property by people less feasible or desirable.165

New York law does not appear to have addressed this or a similarly worded exclusion. The exclusion is primarily an active operations exclusion designed to preclude coverage for diminution in the value of property caused by its proximity to ongoing construction or other operations. It does not extend to completed operations and product liability claims alleging physical damage to property.

10.87

This exclusion needs to be read together with the definition of ‘Property Damage’ in Article III(AA). A number of different situations are there defined as comprising property damage. If there is ‘property damage’ that meets that definition, it is not easy to see any significant scope for the operation of Exclusion E(4). For example, the definition of property damage includes ‘loss of use of tangible property which has not been physically damaged or destroyed arising from physical damage to or destruction of other tangible property’.166 This would cover a situation in which a defective crane had collapsed and blocked access to the claimant’s restaurant business.167 The most obvious liability claim that would arise in such a situation, namely for the recovery of business lost in consequence of the blockage, would not be barred by the exclusion. This is because (i) the claim would not be for ‘decline of value of real or personal property’, but for lost profits and consequential damages, and (ii) the business would have been lost in consequence of ‘physical damage to or destruction of other tangible property’ or ‘loss of use’ within the definition of ‘property damage’ in Article III(AA). If, by contrast, a claim arose from a situation where there was no physical damage to or destruction of other tangible property, or loss of use as defined in the policy, but simply because an undesirable business was operating in proximity to the restaurant, then the requirement of ‘property damage’ would not be fulfilled and there would be no need for the insurer to invoke the exclusion. Thus, given the definition of ‘property damage’ in the 004 Form, ‘property damage’ that meets that definition is likely to preclude application of this exclusion. We are unaware of any Bermuda Form cases relying upon Exclusion E(4).

10.88

installation of a defective product causes covered property damage even under a definition of ‘property damage’ that includes, as the Bermuda Form does, ‘physical damage to tangible property’. See also Masters and Stanzler, Insurance Coverage Litigation (above n 69) ch 14; Allan D Windt Insurance Claims & Disputes: Representation of Insurance Companies and Insureds 4th edn (Minneapolis, West Group, 2011) s 11.13. 164 An exclusion in similar terms was first introduced in the 003 Form. 165 Exclusion E(4) of the 004 Form. 166 Article III(AA) of the 004 Form; and see above ch 5, paras 5.58–5.63. It may be useful in considering this exclusion to review case law addressing coverage for loss of use: eg, Masters and Stanzler, Insurance Coverage Litigation (above n 69) s 14.06, discussing the ‘incorporation doctrine’; compare ibid s 14.07[C], discussing the ‘loss of use’ or ‘failure to perform’ exclusion in ISO CGL policy forms. 167 See the example given by Judge Posner in Eljer Manufacturing Inc v Liberty Mutual Insurance Co 972 F2d 805, 810 (7th Cir 1992).

200

10.89

The Exclusions

By its express terms, the exclusion does not apply if the decline in value arises from physical damage to the property in question. Courts applying New York law have applied a ‘physical contact’ test to find ‘property damage’.168 Thus, in Lowville, while the own product exclusion precluded coverage for the policyholder’s contaminated milk, physical contact between the milk and the claimant’s site caused covered ‘property damage’ to the damaged silo. The court upheld coverage for the costs of removing the milk because that cost ‘represented the covered … damage to other property, namely [the claimant’s site], which was contaminated and rendered useless for [the claimant’s] business purpose for as long as it contained contaminated milk’.169

THE ‘ADVERTISING’ EXCLUSION170 10.90

Although advertising liability is covered by the Bermuda Form policy and is a defined term,171 exclusions qualify the cover. Advertising liability is not, in our experience, a matter that becomes the subject matter of disputes under the Bermuda Form, and we do not deal with the topic here.172

THE ‘WAR’ EXCLUSION 10.91

The war exclusion excludes coverage for: Personal Injury, Property Damage or Advertising Liability directly or indirectly occasioned by, happening through or in consequence of war, invasion, hostile action of foreign enemies, hostilities (whether war be declared or not), civil war, rebellion, revolution, insurrection, military or usurped power or confiscation or nationalization or requisition or destruction of or damage to property by or under the order of any government or public or local authority; provided, however, that this Exclusion G shall not apply to Personal Injury, Property Damage or Advertising Liability: (1) taking place in and caused by the foregoing events in the land area of the United States of America, its territories or possessions, Puerto Rico or Canada; or (2) caused by any act or acts committed by one or more persons, whether or not agents of a sovereign power, for political or terrorist purposes where (a) such person or persons are

168 For example, Eljer Mfg Inc v Liberty Mutual 972 F2d at 808–12 (applying New York law); Lowville Producers v American Motorists 604 NYS2d 421, 422; and see ch 5, paras 5.58–5.63 above. 169 Lowville Producers 604 NYS2d at 422. The case also considered loss of use as ‘property damage’. 170 Exclusion F or (f) in all versions of the Bermuda Form. 171 See Article III(A) of the 004 Form, which provides: A. ‘Advertising Liability’ means liability for Damages on account of: (1) libel, slander or defamation, (2) any infringement of copyright or of title or slogan, (3) piracy or misappropriation of ideas under an implied contract, or (4) any invasion of right or privacy, committed or alleged to have been committed in any advertisement, publicity article, broadcast or telecast and arising out of the Insured’s advertising activities. 172 For a general discussion of the topic, see Masters and Stanzler, Insurance Coverage Litigation (above n 69) ch 16; and Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 15) s 25.03.

The ‘War’ Exclusion

201

not acting on behalf of a government, governmental authority or other power (usurped or otherwise) which exercises de facto jurisdiction over part or all of the populated land area of the country in which the Personal Injury or Property Damage takes place; and (b) if such person or persons are acting as an agent or agents of any government recognized de jure by a majority of Belgium, Canada, France, Germany, Japan, the United Kingdom and the United States, such person or persons are acting secretly and not in connection with the operation of regular military or naval armed forces in the country where the Personal Injury or Property Damage takes place.173

Although both wars and conflicts over terrorist incidents have taken place since the Bermuda Form was introduced, we are not aware of any arbitration in which this clause has arisen as a defence to coverage. An exclusion of this kind is, generally speaking, more likely to be relevant to first-party losses rather than to liability claims by third parties. The clause might become relevant if, for example, a defective product resulted in injury or death to servicemen serving in a war taking place outside the United States, its territories or possessions, Puerto Rico or Canada; for example, the faulty operation of a component designed to prevent friendly-fire incidents.174 The exclusion also might come into play in lawsuits brought against a company providing services to the military in war zones. Because the US government has outsourced to private companies many functions necessary to ongoing operations in Iraq and Afghanistan, commercial policyholders in the United States have faced lawsuits alleging injuries as a result of their activities in those conflict zones.175 Although suits seeking to recover for such activities have arisen, so far the ‘war-risk exclusion’ does not appear to have arisen frequently in connection with disputes over coverage for such activities.176 Insurers also have invoked the ‘war exclusion’ to deny coverage for both first-party and third-party claims arising for widely reported cybersecurity incidents allegedly involving government or other state actors, as well as ‘civil commotion’ or ‘riot’ claims.177

10.92

Few cases have applied this exclusion, and those that have addressed the exclusion under New York law have usually addressed it in the context of first-party property insurance.178 Traditionally, courts have applied war exclusions to ‘kinetic warfare’ – attacks that the ordinary person would consider an act of war.179 As shown by the extant

10.93

173 Exclusion G in the 004 Form. See too Exclusion (g) in earlier versions of the Form. 174 As an example from theatre, in the Arthur Miller play All My Sons, the principal character supplied defective aircraft components during the Second World War, with disastrous consequences. 175 See, eg, discussion of underlying tort lawsuit by detainees at Abu Ghraib and other locations at which the policyholder operated, in CACI Int’l Inc v St Paul Fire & Marine Ins Co 566 F3d 150 (4th Cir 2009). 176 For example, CACI 566 F3d 150. Instead, in CACI, the US Court of Appeals for the Fourth Circuit rejected the policyholder’s appeal of a lower court decision denying coverage, agreeing with the trial court that the insurance policy did not cover the policyholder’s worldwide operations. Ibid at 156–57. 177 Eg, with regard to protests in the United States over, for example, the police killing of George Floyd. 178 Uniroyal Inc v Home Ins Co 707 F Supp 1368, 1389 (EDNY 1988) contains a brief discussion of a war-risk exclusion. See also Cohen v Monumental Life Ins Co 194 SE.2d 867, 867–68 (NC App 1973) (the claimant was not entitled to life insurance benefits because the insured’s death was the result of armed hostilities between the United States and Vietnam); Stanbery v Aetna Life Ins Co 98 A2d 134, 137–38 (NJ Super Law Div 1953) (exclusion eliminated coverage if the insured ‘Korean War’ veteran’s death resulted from military or naval service in time of war). 179 Before courts adopted the ‘ordinary meaning’ approach, some used the ‘technical’ approach, which essentially considers whether there has been a formal declaration of war. See, eg, Rosenau v Idaho Mut Ben Ass’n, 145 P2d 227, 229 (Idaho 1944) (‘An act of Congress is necessary to the commencement of a foreign war and is in itself a declaration … It fixes the date of the war’); West v Palmetto State Life Ins Co 25 SE2d 475, 477 (SC 1943)

202

The Exclusions

cases interpreting the ‘war-risk exclusion’ in standard form CGL insurance policies typically sold in the United States, the analysis is often complex and depends upon the facts involved. Specifically, courts have found that the typical war-risk exclusion applies to preclude coverage only when the acts of a sovereign government caused the injury or damage and not to incidents of terrorism by non-state actors. Courts have rejected coverage when faced with exclusions that specifically preclude coverage for terrorist activities or acts of entities other than sovereign governments or a government’s military forces. The Bermuda Form war exclusion, however, does preserve coverage for some acts which are committed ‘for political or terrorist purposes’, provided that the conditions set out in subparagraph (2)(a) and (b) are fulfilled. 10.94

Two cases addressing a ‘war’ exclusion under New York law provide important context for potential disputes under the Bermuda Form exclusion.180 In Pan American, the primary case on the scope of a war exclusion, two members of the Popular Front for the Liberation of Palestine (PFLP) hijacked a Pan American airliner over London, about 45 minutes after it had taken off in Amsterdam. The terrorists forced the crew to fly the plane to Cairo where the passengers were evacuated and the aircraft destroyed. The airline’s insurers disputed which insurer was liable to pay for the loss of the jetliner. One set of insurance policies excluded coverage for: 1. capture, seizure, arrest, restraint or detention or the consequences thereof or of any attempt thereat, or any taking of the property insured or damage to or destruction thereof by any Government or governmental authority or agent (whether secret or otherwise) or by any military, naval or usurped power … 2. war, invasion, civil war, revolution, rebellion, insurrection or warlike operations, whether there be a declaration of war or not … 3. strikes, riot, civil commotion.181

A second source of insurance was aviation war-risk insurance, provided by the US government, which specifically covered loss or damage ‘resulting from’ the following: war, invasion, acts of foreign enemies, hostilities (whether war be declared or not), civil war, rebellion, revolution or insurrection, military or usurped power or confiscation and/or nationalization or requisition or destruction by any government or public or local authority or by any independent unit or individual engaged in irregular warfare.182 (‘the declaration by Congress of war … was the only legal way in which this country could be placed in a state of war’); and Savage v Sun Life Assur Co of Canada, 57 F Supp 620 (WD La 1944) (holding that the war exclusion did not apply to exclude benefits under a life insurance policy issued to a soldier who died during Pearl Harbor because ‘war’ does not exist merely because of an armed attack by another nation until it is a condition recognised or accepted by the political authority of the government which is attacked, either through an actual declaration of war or other acts demonstrating such position); Universal Cable Prods, LLC v Atlantic Spec Ins Co 929 F3d 1143, 1154 (9th Cir 12 July 2019) (‘“war” has a special meaning in the insurance industry requiring hostilities between de jure and de facto governments’). See Daniel James Everett, ‘The “War” on Terrorism: Do War Exclusions Prevent Insurance Coverage for Losses Due to Acts of Terrorism?’ (2002) 54 Alabama Law Review 175, 185 (in determining whether an act constitutes ‘war’, courts consider factors such as ‘whether the combatants wore uniforms, the nature and type of weaponry used, the actual organization of the operation, the act causing the loss, whether congressional appropriations were made … and declarations by the Judge Advocate General initiating court martial jurisdiction in cases arising from the conflict’). 180 Pan American 505 F2d at 1005–22; Ennar Latex Inc v Atlantic Mut Ins Co No 94 Civ 150, 1995 US Dist Lexis 7386 (SDNY 1995). 181 505 F2d at 994. 182 Ibid at 995.

The ‘War’ Exclusion

203

However, the government insurance policy specifically disclaimed coverage to the extent that insurance was provided under any other insurance policy.183 The airline’s commercial insurers argued that their coverage did not apply because destruction of the airliner resulted from: (i) representatives of a paramilitary, quasi-governmental authority which amounted to a ‘military … or usurped power’; (ii) ‘guerrilla’ or similar warfare; (iii) an insurrection attempting to overthrow a government; or (iv) civil commotion or riot.184

10.95

On the first argument, the court acknowledged that the PFLP occupied a portion of a foreign country, operating at the sufferance of the government of Jordan, but concluded that the terrorists’ activities were ‘surely insufficient’ to constitute a ‘military … or usurped power’.185 This conclusion finds support in English law and, in fact, the US Court of Appeals for the Second Circuit analysed English authorities in reaching its conclusion that the phrase ‘military or usurped power’ has a specific meaning in the context of insurance. The court first noted that the words ‘military or usurped power’ had long appeared in insurance policies, but had received scant judicial attention, presumably because ‘the events necessary to bring them into play are extraordinary’.186 Citing Drinkwater v The Corporation of the London Assurance Co,187 an English case which, in 1767, was the first to consider the issue, the Second Circuit relied upon Mr Justice Bathurst’s conclusion that military or usurped power can ‘only mean an invasion of the Kingdom by foreign enemies … or any internal armed force in rebellion assuming the power of government, by making laws, and punishing for not obeying the laws’.188 Following this analysis of English law, the Second Circuit held that:

10.96

to constitute a military or usurped power the power must be at least that of a de facto government … [T]he PFLP was not a de facto government in the sky over London when the 747 was take[n] … the loss was not ‘due to or resulting from’ a ‘military … or usurped power’.189

With regard to whether the actions at issue constituted ‘war’, the appellate court agreed with the trial court’s conclusion that war typically involves the deployment of ‘force between government entities essentially like governments, at least de facto’.190 The Second Circuit relied on cases establishing ‘that war is a course of hostility engaged in by entities that have at least significant attributes of sovereignty … Under international law war is waged by states or state-like entities’.191 The court noted that Pan American’s loss resulted not from the actions of any state; indeed, the PFLP never claimed to act on behalf

183 Ibid. 184 Ibid at 996. 185 Ibid at 1009 (citing the conclusion of the District Court). 186 Ibid at 1010 (citing Barton v Home Ins Co 42 Mo 156, 158 (1868)). 187 Drinkwater v The Corporation of the London Assurance Co (1767) 2 Wils 363; 95 English Reports 863. There, a mob burned to the ground the policyholder’s malting house at Norwich. The policyholder’s fire insurance policy excluded coverage for fires caused by ‘any military or usurped power whatsoever’. The court held that the insurance policy did not exclude coverage because the mob did not constitute a ‘military or usurped power’. 188 Pan American 505 F2d at 1010 (quoting Drinkwater 95 Eng Rep at 863 (ellipsis in original)). 189 505 F2d at 1009. 190 Ibid. For a recent application of this approach, in the context of Israel and Hamas, see Universal Cable Productions Inc v Atlantic Specialty Ins Co 929 F3d 1143 (9th Cir 2019). 191 Ibid at 1012.

10.97

204

The Exclusions

of any government, and all of the countries within which it operated ‘uniformly opposed hijacking’.192 The court specifically rejected the argument that the PFLP’s actions constituted ‘guerrilla warfare’, holding that guerrilla groups also must have ‘at least some incidents of sovereignty before its activity can properly be styled “war”’. Although the PFLP received some governmental backing, no state had recognised it; as a result, the court found no basis to consider the group even a ‘quasi-sovereign’.193 The court also concluded that the terrorists’ actions did not constitute ‘warlike operations’, given that the commercial airliner carried no cargo destined for a theatre of war and was owned by a commercial airline, not the belligerent of a foreign country. 10.98

With regard to the third and fourth arguments, the court found no ground for equating the destruction of the jetliner with ‘insurrection’, ‘civil commotion’ or ‘riot’. On a commonly understood meaning of the term, the court concluded that an ‘insurrection’ is a ‘violent uprising by a group [of men] acting for the specific purpose of overthrowing the constituted government and seizing its powers’.194 Because the court found no such intent, it refused to apply this aspect of the exclusion to preclude coverage.195 By the same token, the hijacking and destruction of the jet liner did not qualify as a civil disturbance among citizens and thus did not constitute a ‘civil commotion’. The hijacking also did not meet the meaning of ‘riot’, which the court found to be ‘a local disturbance, normally by a mob, not a complex, traveling conspiracy of the kind in this case’.196

10.99

Almost two decades later, the US District Court for the Southern District of New York found that a war exclusion did not bar coverage in a case involving more overtones of government action than existed in Pan American. There, the policyholder sought coverage for the loss of a ship that first was seized by a paramilitary group and later was confiscated by a foreign government supporting the paramilitaries. In upholding coverage, the court in Ennar Latex Inc v Atlantic Mutual Insurance Co197 specifically distinguished between acts of terror and acts of war. It refused to apply the exclusion to preclude coverage, finding that the military group there operated separately from the foreign government. Even the after the fact, endorsement of the seizure by the foreign government did not, in the court’s view, transform the incident into an action by a foreign sovereign: ‘It is not the role of this court to recognize de facto governments.’198

10.100

The war exclusion in the Bermuda Form seeks to preclude coverage for liability ‘directly or indirectly occasioned by, happening through or in consequence of’ a number of

192 Ibid at 1013. 193 Ibid at 1013–15. 194 Ibid at 1017 (quotation marks omitted). 195 Ibid at 1015–16. 196 Ibid at 1019–20 (quotation marks omitted). 197 No 94 Civ 150, 1995 WL 325640, at *5 (SDNY 30 May 1995). 198 Ibid at *5. However, some courts have excluded coverage under a war-risk exclusion interpreted under the laws of other states; see, eg, Younis Bros & Co v Cigna Worldwide Ins Co 899 F Supp 1385 (ED Pa 1995) (applying the war-risk exclusion to preclude coverage in a case involving an insurrection in Liberia), aff ’d, 91 F3d 12 (3d Cir 1996), judgment enforced on remand, 167 F Supp 2d 743 (ED Pa 2001); TRT/FTC Communications Inc v Insurance Co of Pa 847 F Supp 28 (D Del 1993) (barring coverage for the random robbery of a facility by a man dressed in civilian clothes one day after the invasion by US military forces and six days after the country’s legislature declared a state of war), aff ’d, 9 F3d 1541 (3d Cir 1993) (unpublished table decision).

The ‘Toxic Substances’ Exclusion

205

matters, including war, invasion, civil war, rebellion and nationalisation, among other governmental actions. The main part of the exclusion uses the phrases interpreted in Pan American and the English authorities – for example, ‘military or usurped power’ – and thus must be assumed to have been written with knowledge of the import of such cases.199 The exclusion also makes clear, in referring to ‘hostilities (whether war be declared or not)’, that the exclusion will apply in situations in which a government is pursuing military action but has not issued a formal declaration of war. Thus, the exclusion would have applied to preclude coverage for liabilities due to the conflict between the United States and Vietnam, although the US government never formally declared ‘war’. The exclusion also includes exceptions preserving coverage for actions that occur in the ‘land area’ of the United States, its territories, possessions, Puerto Rico and Canada. As previously noted, an exception to the exclusion also makes clear that the policy continues to cover loss caused by certain political or terrorist activities.

10.101

THE ‘TOXIC SUBSTANCES’ EXCLUSION When the Bermuda Form was first drafted, certain products were known to have caused serious problems, both for individual claimants and indeed for the insurance industry. Accordingly, all versions of the Bermuda Form have contained specific ‘laser exclusions’ for products containing or consisting of asbestos. tobacco. dioxin. asbestiform talc. DES (diethylstilbestrol, a drug taken by pregnant women) and any intra-uterine device. The exclusion in the 004 Form seeks to preclude coverage for liability for: H. Toxic Substances Personal Injury, Property Damage or Advertising Liability arising out of the manufacture, distribution, sale, installation, removal, utilization, ingestion or inhalation of, or exposure to or existence of, as the case may be: (1) asbestos or any asbestos containing materials; provided, however, that this Exclusion H shall not apply to Property Damage arising out of asbestos not contained in the Insured’s Products as a result of explosion, hostile fire or lightening; (2) tobacco or any tobacco products (or ingredients of, or used in the manufacture or production of, such products); (3) 2.3.7.8-TCDD (2.3.7.8-tetrachlorodibenzo-p-dioxin); (4) asbestiform talc; (5) diethylstilbestrol (DES); (6) any intra-uterine device (IUD); (7) any product containing silicone which is in any form implanted or injected in the body;

199 In addition, much was written about ‘war-risk exclusions’ after the terrorist attacks of 11 September 2001. For example, Christopher Jennings, ‘Insurance Coverage of the World Trade Center: Interpretation of “War Risk” Exclusion Clauses under New York Contract Law’ (Congressional Research Service, CRS-4 18 September 2001). See also Katherine S Wan, ‘Not Petya, Not Warfare: Rethinking the Insurance War Exclusion in the Context of International Cyberattacks’ (2020) 95 Washington Law Review 1595.

10.102

206

The Exclusions

provided however, that this Exclusion H shall not apply to actual or alleged Personal Injury or Property Damage where such Personal Injury or Property Damage is not related to the asbestos, tobacco (or other consumed portion of a tobacco product), 2.3.7.8-TCCD, asbestiform talc, DES, IUD or silicone content of goods, materials or products or completed operations. The listing of materials herein shall not give rise to an inference that Personal Injury, Property Damage or Advertising Liability attributable to other materials was neither Expected nor Intended by the Insured.200

10.103

In the 004 Form, a laser exclusion for liability from ‘any product containing silicone which is in any form implanted or injected in the body’ has been added to the list set out in earlier Forms, no doubt with the extensive liability in the 1990s for alleged autoimmune disease from silicone gel breast implants in mind. The other laser exclusions included in the toxic substances exclusion in the 004 were included in earlier versions of the Bermuda Form.

10.104

The exclusion does not apply merely if the product contains asbestos or one of the other substances. Thus, to take the example of asbestos, the personal injury or damage in question must be ‘related to’ the asbestos content of the product.201 Another exception to the exclusion is set out in Exclusion H(1), namely for property damage ‘arising out of asbestos not contained in the Insured’s Products as a result of explosion, hostile fire or lightning’. Accordingly, if a manufacturing plant explodes and fibres are released from part of the building which contains asbestos within its fabric, a claim for resulting property damage can still be made. The final sentence of the exclusion emphasises that the policy may not cover liabilities for products other than those listed because of the expected and intended provision of the Bermuda Form (see Chapter 7 above).

THE ‘AIRCRAFT’ EXCLUSION 10.105

Bermuda Form 001 contained a broad exclusion for damages arising out of the design, manufacture, construction, maintenance, service, use or operation of any Aircraft, or any component part or equipment thereof, or any other Aircraft navigational or related equipment or service.202

The relevant exception in the 004 Form now contains a lengthy list of qualifications to this exclusion.203 In very broad terms, and subject to the qualifications, the exclusion is aimed at limiting coverage for an aircraft crash or hijacking.

200 Exclusion H in the 004 Form. See also Exclusion (h) in earlier Forms. 201 Once again, courts read exclusions narrowly. Thus, in one US case, the court refused to read an exclusion for ‘asbestos or similar conditions’ to exclude coverage also for other asbestos diseases, mesothelioma or bronchogenic carcinoma. See, eg, Celotex Corp v AIU Ins Co (In re Celotex Corp) 175 BR 98, 109–12 (Bankr MD Fla 1994); Carey Canada Inc v California Union Ins Co 720 F Supp 1018, 1021 (DDC 1989), aff ’d in part, rev’d in part, remanded, 940 F2d 1548 (DC Cir 1991). The court relied on medical and other evidence which showed distinctions among those diseases and other exclusions that specifically referred to all three diseases. 202 Exclusion (i) in the 001 Form. See also Exclusion (i) in the 002 and 003 Forms; and Exclusion I in the 004 Form. 203 Exclusion I in the 004 Form.

The ‘Pollution’ Exclusion

207

THE ‘WATERCRAFT’ EXCLUSION This exclusion provides that the policy does not apply to:

10.106

Liability arising out of the design, construction, maintenance, sale, manning, ownership or operation of any Watercraft, but this Exclusion J shall not apply to: (1) Watercraft or risks listed on Schedule C hereto and any additional Watercraft acquired in the ordinary course of business during the Policy Period which are of a similar type and use as the Watercraft listed on Schedule C: provided, however, that the aggregate gross tonnage of all such additional Watercraft shall not exceed 20% of the gross tonnage of Watercraft listed on Schedule C; (2) loading or unloading of any Watercraft at premises owned, leased or controlled by the Insured; (3) liability for any Personal Injury or Property Damage to third parties arising out of or allegedly arising out of Incidental Watercraft Use (provided that damage to the hull or any portion, component or equipment of the Watercraft owned, leased or chartered by the Insured or to its cargo contents shall not constitute Property Damage to third parties); (4) liability for Personal Injury, Property Damage or Advertising Liability arising out of the design, construction, maintenance or sale by the Insured of any Watercraft less than 75 feet in length; or (5) Personal Injury, Property Damage or Advertising Liability arising out of or alleged to arise out of design, manufacture, maintenance or sale by the Insured of any component part of equipment of any Watercraft.204

This exclusion, again subject to qualifications, seeks to preclude coverage for risks that would normally be covered by specialised ‘P&I’ (Protection and Indemnity) marine insurance. Shipowners have traditionally insured their liabilities by way of mutual insurance with ‘P and I Clubs’ that shipowners have joined.205

THE ‘POLLUTION’ EXCLUSION The pollution exclusion in the 004 Form applies to preclude coverage for: (1) (a) liability for Personal Injury, Property Damage or Advertising Liability arising out of the Discharge of Pollutants into or upon land or real estate, the atmosphere, or any watercourse or body of water whether above or below ground or otherwise into the environment; or (b) liability, loss, cost or expense of any Insured or others arising out of any direction or requests, whether governmental or otherwise, that any Insured or others test for, monitor, clean up, remove, contain, treat, detoxify or neutralize Pollutants; This Exclusion K applies whether or not such Discharge of such Pollutants: (i) results from the Insured’s activities or the activities of any other person or entity; (ii) is sudden, gradual, accidental, unexpected or unintended; or 204 Exclusion J in the 004 Form. See also Exclusion (j) in earlier Forms. 205 For a leading decision in England as to the impact of ‘Club rules’ in the context of insolvent shipowners and direct actions by third parties, see Firma C-Trad SA v Newcastle Protection & Indemnity Ass’n [1991] 2 AC 1.

10.107

208

The Exclusions (iii) arises out of or relates to industrial operations or the Waste or byproducts thereof.

(2) Paragraph (1) of this Exclusion K does not apply to: (a) Product Pollution Liability; or (b)

(i) liability of the Insured for Personal Injury or Property Damage caused by an intentional Discharge of Pollutants solely for the purpose of mitigating or avoiding Personal Injury or Property Damage which would be covered by this Policy; or (ii) liability of the Insured for Personal Injury or Property Damage caused by a Discharge of Pollutants which is not Expected or Intended, but only if the Insured becomes aware of the commencement of such Discharge within seven (7) days of such commencement;

provided that the Insured gives the Company written notice in accordance with Section D of Article V of this Policy of such commencement of such Discharge under subparagraphs (2) (b) (i) or (ii) of this Exclusion K within forty (40) days of such commencement. Such notice must be provided irrespective of whether notice as soon as practicable otherwise would be required pursuant to Section A of Article V of this Policy.206

10.108

The Bermuda Form 004 Policy defines ‘Discharge’, ‘Pollutants’ and ‘Waste’ as follows: I. ‘Discharge’ means discharge, emission, dispersal, migration, release or escape (or any series of such of a similar nature at the same site) but does not include any discharge, emission, dispersal, migration, release or escape to the extent that the Pollutants involved remain confined within the building or other man made structure in which they initially were located. … Y. ‘Pollutant’ means any solid, liquid, gaseous or thermal irritant, contaminant or toxic or hazardous substance or any substance which may, does, or is alleged to affect adversely the environment, property, persons or animals, including smoke, vapour, soot, fumes, acids, alkalis, chemicals and Waste. … AE. ‘Waste’ means all waste and includes, without limitation, materials to be discarded, stored pending final disposal, recycled, reconditioned or reclaimed.207

The concept of ‘discharge’ accordingly does not encompass the situation where the pollutants ‘remain confined within the building or other man made structure in which they were initially located’. This provision therefore avoids some of the difficulties that have confronted courts in the United States; for example, the situation in Belt Painting Corp,208 where the claimant had inhaled paint or solvent fumes in an office

206 Exclusion K. See also Exclusion (k) in earlier versions of the Bermuda Form. See, further, the discussion of the ‘Owned Property; Care, Custody or Control, etc’ Exclusion at paras 10.40–10.61 above. 207 Article III (I), (Y) and (AE) in the 004 Form. 208 742 NYS2d 332. For other cases refusing to apply the absolute pollution exclusion used in US policies to preclude coverage for personal injuries at the claimant’s workplace (usually indoors), see, eg, Stoney Run Co v Prudential LMI Commercial Ins Co 47 F3d 34 (2d Cir 1995) (applying New York law); Lefrak Organization Inc v Chubb Custom Ins Co 942 F Supp 949 (SDNY 1996); Roofers’ Joint Training Apprentice & Educ Comm’n of W NY v General Accident Ins Co of Am 713 NYS2d 615 (App Div 2000); Schumann v State 610 NYS2d 987 (Ct Cl 1994).

The ‘Pollution’ Exclusion

209

building in which he was performing stripping and painting work, and sued the policyholder for his injuries. Background to the Exclusion Extensive litigation in the United States relating to pollution exclusions in liability insurance policies continues, even with the revision of the CGL policy to include an ‘absolute’ or ‘total’ pollution exclusion in place of the problematic ‘sudden and accidental’, or ‘qualified’, pollution exclusion. A typical ‘absolute pollution exclusion’ used in the United States excludes: f.(1) ‘Bodily Injury’ or ‘property damage’ arising out of the actual, alleged or threatened discharge, dispersal, release or escape of pollutants: (a) At or from premises you own, rent or occupy; (b) At or from any site or location used by or for you or others for the handling, storage, disposal, processing or treatment of waste; (c) Which are at any time transported, handled, stored, treated, disposed of, or processed as waste by or for you or any person or organization for whom you may be legally responsible; or (d) At or from any site or location on which you or any contractors or subcontractors working directly or indirectly on your behalf are performing operations: (i) if the pollutants are brought on or to the site or location in connection with such operations; (ii) if the operations are to test for, monitor, clean up, remove, contain, treat, detoxify or neutralize the pollutants. (2) Any loss, cost, or expense arising out of any governmental direction or request that you test for, monitor, clean up, remove, contain, treat, detoxify or neutralize pollutants. Pollutants means any solid, liquid, gaseous or thermal irritant or contaminant, including smoke, vapor, soot, fumes, acids, alkalis, chemicals and waste. Waste includes materials to be recycled, reconditioned or reclaimed.209

Dubbed the ‘absolute’ pollution exclusion (APE), it was intended to extend the reach of the more limited ‘sudden and accidental’ pollution exclusion and, thus, further limit coverage for industrial pollution, as shown by historical insurance industry analyses.210 The Illinois Supreme Court observed that the events leading up to the insurance industry’s adoption of the exclusion are ‘well documented and relatively uncontroverted’. The exclusion was the insurance industry’s response to the ‘enormous expense and exposure [it faced] resulting from the “explosion” of environmental litigation’.211 Other courts in

209 Masters and Stanzler, Insurance Coverage Litigation (above n 69) A-32 (1986 occurrence form), A-42 (1986 claims-made form). 210 See the discussion of the regulatory background of the exclusions’ adoption by regulators in the United States in ibid ss 14.08[A] (regarding the APE) and 15.06[B] and 15.07[C] (regarding the ‘sudden and accidental’ and ‘absolute’ exclusions, respectively). 211 Am States Ins Co v Koloms 687 NE2d 72, 81 (Ill 1997).

10.109

210

The Exclusions

the United States have given the exclusions broader application.212 However, the New York Court of Appeals in Belt Painting limited the reach of the APE to ‘industrial pollution’. 10.110

Courts in different states in the United States have disagreed as to the effect of the ‘sudden and accidental’ pollution exclusion that was included in most CGL insurance policies issued between the early 1970s and 1985. That exclusion included an exception preserving coverage when the pollution was ‘sudden and accidental’. Some US courts did not, however, interpret ‘sudden and accidental’ in the way that insurers wanted, and they revised the CGL policy to use an ‘absolute’ or ‘total’ exclusion that contained no exception for ‘sudden and accidental’ discharges. The battleground is summarised in, for example, the decisions of the New York Court of Appeals in Belt Painting213 and of the California Supreme Court in MacKinnon v Truck Insurance Exchange.214 United States Case Law on the Exclusion

10.111

Courts across the United States have addressed the APE. Many of the appellate courts to address the exclusion have limited the applicability of the APE to traditional environmental pollution.215 New York’s highest court, the New York Court of Appeals, has joined other state high courts in the United States in concluding that an ‘absolute pollution exclusion’ should not apply to preclude the CGL policy’s traditional coverage for typical

212 See, eg, Deni Assocs of Fla Inc v State Farm Fire & Cas Ins Co 711 So 2d 1135 (Fla 1998); Bituminous Cas Corp v Cowen Constr Inc 55 P3d 1030 (Okla 2002) (answering a certified question from the US District Court for the Northern District of Oklahoma); Nat’l Union Fire Ins Co of Pittsburgh Pa v CBI Indus Inc 907 SW2d 517 (Tex 1995). Compare the discussion of the pollution exclusion in Board of Regents of Univ of Minn v Royal Ins Co of Am 517 NW2d 888 (Minn 1994) (limited application in the context of asbestos in building claims) with that in Wakefield Park Inc v Ram Mut Ins Co 731 NW2d 154 (Minn Ct App 207) (applied pollution more broadly regarding ‘discharge’ ‘into the atmosphere’). 213 742 NYS2d 332 (NY 2002). 214 MacKinnon v Truck Ins Exch 73 P3d 1205 (Cal 2003). For other textbook commentaries on the pollution exclusion, see Masters and Stanzler, Insurance Coverage Litigation (above n 69) ss 15.06–15.07; Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 15) ch 10. For other New York cases addressing the pollution exclusion in its different forms, see, eg, Stoney Run 47 F3d 34; Sphere Drake Ins Co v YL Realty Co 990 F Supp 240 (SDNY 1997); Garfield Slope Housing Corp v Public Serv Mut Ins Co 973 F Supp 326 (EDNY 1997); Calvert Ins Co v S & L Realty Corp 926 F Supp 44 (SDNY 1996); Westview Assocs v Guaranty Nat’l Ins Co 717 NYS2d 75 (NY 2000); Incorporated Village of Cedarhurst v Hanover Ins Co 653 NYS2d 68 (NY 1996); Continental Casualty v Rapid-American 593 NYS2d 966; Technicon Elecs Corp v American Home Assur Co 544 NYS2d 531 (NY 1989); Cepeda v Varveris 651 NYS2d 185 (App Div 1996). 215 See, eg, Bituminous Cas Corp v Advance Adhesive Tech Inc 73 F3d 335, 338 (11th Cir 1996) (applying Georgia law); Stoney Run 47 F3d at 37; Red Panther Chem Co v Insurance Co of Pa 43 F3d 514 (10th Cir 1994) (applying Mississippi law); Regent Ins Co v Holmes 835 F Supp 579, 582 (D Kan 1993); Minerva Enters Inc v Bituminous Cas Corp 851 SW2d 403, 405 (Ark 1993); Koloms 687 NE2d 72; Sullins v Allstate Ins Co 667 A2d 617 (Md 1995); Western Alliance Ins Co v Gill 686 NE2d 997 (Mass 1997); Nav-Its Inc v Selective Ins Co 869 A2d 929 (NJ 2005); West Am Ins Co v Tufco Flooring E Inc 409 SE2d 692 (NC App 1991), overruled on other grounds by Gaston County Dyeing Mach Co v Northfield Ins Co 524 SE2d 558 (NC 2000). Contra National Elec Mfrs Ass’n v Gulf Underwriters Ins Co 162 F3d 821 (4th Cir 1998) (applying DC law; called into question by Richardson v Nationwide Mut Ins Co 826 A2d 310 (DC 2003), vacated on reh’g, 832 A2d 752 (DC 2003)); American States Ins Co v Nethery 79 F3d 473 (5th Cir 1996) (applying Mississippi law); Brown v American Motorists Ins Co 930 F Supp 207 (ED Pa 1996), aff ’d, 111 F3d 125 (3d Cir 1997).

The ‘Pollution’ Exclusion

211

non-environmental products – liability or other tort liability.216 In Belt Painting Corp,217 the court found that the APE was ambiguous when applied to a claim for bodily injury arising from inhalation of paint or solvent fumes. Relying on two earlier decisions rejecting pollution exclusions in the context of asbestos bodily injury and lead paint poisoning, New York’s highest court held that ‘the purpose of the exclusion was to deal with broadly dispersed environmental pollution’.218 The court found the exclusion to be ambiguous because it ‘does not clearly and unambiguously exclude a personal injury claim from indoor exposure to a plaintiff insured’s tools of its trade’.219 It found that the ‘drifting’ of paint fumes did not meet the ‘environmental implications’ of the terms, ‘discharge, dispersal, seepage, migration, release or escape’ of ‘pollutants’ required by the clause. The court was ‘reluctant to adopt an interpretation that would infinitely enlarge the scope of the term “pollutants”, and seemingly contradict both a “common speech” understanding of the relevant terms and the reasonable expectations of a businessperson’.220 In reaching this decision, the court also reviewed the genesis of the APE, stating that it originated with insurers’ efforts to avoid potentially open ended liability for long term, gradual discharge of hazardous waste and its byproducts. Finally, the court rejected the insurer’s contention that the removal of the language ‘into or upon land, the atmosphere or any water course or below of water’ in the APE was a material difference that indicated an intent to extend the exclusion to indoor as well as outdoor pollution.

10.112

In MacKinnon,221 the California Supreme Court reached a similar result, reversing the lower court’s award of summary judgment in favour of the insurer, and holding that the APE did not clearly exclude ordinary acts of negligence involving toxic chemicals such as pesticides. In reaching its decision, the court first discussed the various reasons courts in the United States have construed the APE narrowly. Some courts have relied on the history of the pollution exclusion, which, they have found, was intended to cover only ‘traditional environmental contamination’.222 Some courts have also recognised that a broad

10.113

216 For example, the California Supreme Court in MacKinnon 73 P3d 1205; and Maryland’s high court in Clendenin Bros Inc v US Fire Ins Co 889 A2d 387 (Md 2006). 217 763 NYS2d 790 (NY 2003). 218 Ibid at 18 (citing Continental Casualty v Rapid-American 593 NYS2d 966 and Westview Associates 717 NYS2d 75). 219 Belt Painting 763 NYS2d at 795; Great Am Restoration Servs Inc v Scottsdale Ins Co 911 NYS2d 142, 146 (App Div 2010) (Scottsdale’s position that damages from asbestos are excluded under the pollution exclusion would render the specific asbestos exclusion meaningless, in violation of settled canons of construction), remanding, No 110122008, 2011 WL 11718178 (NY Sup Ct 11 Jan 2011) (compelling insurer to defend); compare Cincinnati Ins Co v Roy’s Plumbing, Inc No 16-2511, 2017 US App LEXIS 9729 (2d Cir 31 May 2017) (‘New York courts, however, limit the reach of pollution exclusions to those cases where the damages alleged are truly environmental in nature, or where the underlying complaint alleges damages resulting from what can accurately be described as the pollution of the environment’ – enforcing the APE to preclude for claims for liability for damage from sewage); see Broome Cty v Travelers Indem Co 125 AD3d 1241, 1242–43 (NY App Div 2015) (applying the APE to preclude coverage for claims for injury from silica in a first-party property insurance claim, distinguishing Belt Painting). 220 Ibid. 221 73 P3d 1205 (Cal 2003). 222 For example, Belt Painting, Koloms and MacKinnon.

212

The Exclusions

reading of the APE to cover any contaminant or irritant ‘would have absurd or otherwise unacceptable results’.223 Other courts have also recognised that the phrase ‘discharge, dispersal, release or escape’ contains terms of art describing environmental pollution and implies expulsion of a pollutant over an extended area, and not a localised toxic accident. 10.114

In analysing the case before it, the California Supreme Court relied in part on the doctrine of reasonable expectations under California law. Ultimately, however, the court relied on what it concluded was the ordinary understanding of the term ‘pollutant’, which it found sufficient to decide the case before it.224 It concluded that the exclusion did not ‘conspicuously, plainly and clearly’ apprise the policyholder that ordinary acts of negligence were not covered and found that the insurer’s broad construction of the APE would yield results that no one would consider reasonable, such as excluding a hypothetical allergic reaction to swimming pool chlorine.225

10.115

Many courts, however, have applied an APE to preclude coverage for what may be considered non-industrial pollution claims.226 In many of these cases, the courts concluded that the exclusion at issue was not ambiguous and thus applied to preclude coverage for discharges of toxic or noxious contaminants.

10.116

Given the range of decisions in the United States both upholding coverage and rejecting it in similar contexts, it is difficult to draw a bright line about how the exclusion in the Bermuda Form should be applied. Many of the decisions rejecting coverage also were decided before more recent cases like Belt Painting that have limited the exclusion. At base, the plethora of decisions on the issues may point to the necessity of analysing the applicability of the exclusion to the facts at hand in light of the parameters set forth in Belt Painting. Climate Change

10.117

Companies in a variety of industries increasingly face the prospect of lawsuits seeking to recover damages for alleged results of ‘global warming’, or in current parlance ‘climate change’, and emission of greenhouse gases (GHGs). Private plaintiffs in the United States have sued alleging that global warming has caused more ferocious hurricanes, melting polar ice caps and other environmental effects that have damaged the plaintiffs’ property.

223 MacKinnon 73 P3d at 1211. 224 Ibid at 1217–18. 225 Ibid at 1213–14. 226 For example, American States Ins Co v Nethery 79 F3d 473; Park-Ohio Indus Inc v Home Indem Co 975 F2d 1215 (6th Cir 1992); Shalimar Contractors Inc v American States Ins Co 975 F Supp 1450 (MD Ala 1997), aff ’d, 158 F3d 588 (11th Cir 1998) (unpublished table decision); Brown v American Motorists Ins Co 930 F Supp 207 (ED Pa 1996), aff ’d, 111 F3d 125 (3d Cir 1997); Board of Regents 517 NW2d 888 (Minn 1994); Landshire Fast Foods of Milwaukee Inc v Employers Mut Cas Co 676 NW2d 528 (Wis Ct App 2004); TerraMatrix Inc v US Fire Ins Co 939 P2d 483 (Colo Ct App 1997). Some courts have addressed a similar exclusion, called the ‘total pollution exclusion’ (TPE). For an example of a TPE, see Re Idle Aire Techs Corp No 08-10960 (KG), 2009 Bankr Lexis 353, at *5–*6 (Bankr D Del 2009). Cf Cook v Evanson 920 P2d 1223, 1227 (Wash Ct App 1996) (toxic sealant considered ‘pollution’) with Kent Farms Inc v Zurich Ins Co 969 P2d 109 (Wash Ct App 1998) (diesel fuel used as intended not a ‘pollutant’).

The ‘Pollution’ Exclusion

213

Decisions by federal appellate courts in the United States have allowed such lawsuits to proceed, holding that, contrary to rulings by lower courts, the plaintiffs have standing to sue companies that have emitted greenhouse gases.227 Even if the plaintiffs ultimately fail to meet their burden to show the necessary proximate cause between their damages and defendants’ conduct, these lawsuits, often brought as class actions, can magnify the potential for claims for coverage of defence costs. In a decision that may arise in disputes over insurance coverage for such lawsuits, the US Supreme Court held that emissions of carbon dioxide, one of the GHGs involved in global warming, constitute a ‘pollutant’ under the US Clean Air Act. That decision, Massachusetts v Environmental Protection Agency, interpreted the Clean Air Act, and not a definition of ‘pollutant’ in an insurance policy, referring to the ‘sweeping’ definition in that statute:

10.118

The Clean Air Act’s sweeping definition of ‘air pollutant’ includes ‘any air pollution agent or combination of such agents, including any physical, chemical … substance or matter which is emitted into or otherwise enters the ambient air …’. However, its conclusion may be cited in disputes over the applicability of a pollution exclusion to suits for damages. On its face, the definition embraces all airborne compounds of whatever stripe, and underscores that intent through the repeated use of the word ‘any’.228

The Bermuda Form Qualifications on the Exclusion The Bermuda Form exclusion contains qualifications on the pollution exclusion, albeit not the ‘sudden and accidental’ exception discussed above. The first qualification, which first appeared in the 002 Form, addressed intentional discharges of pollutants solely for the purpose of mitigating or avoiding personal injury or property damage that would be covered by the policy. The second qualification addresses the situation in which the policyholder becomes aware of the discharge of pollutants within seven days of the commencement of the discharge. This qualification is perhaps more important, and first appeared (albeit not in precisely the same terms)229 in the 001 Form. To take advantage of this qualification, however, the policyholder must give written notice within 40 days of the commencement of the discharge in accordance with Article V(D) of the policy. Article V(D) gives requirements for giving ‘notice of occurrence’, which are discussed elsewhere.230

227 Comer v Murphy Oil USA 585 F3d 855 (5th Cir 2009), reh’g granted, 598 F3d 208 (5th Cir 2010) (en banc); and the consolidated cases of Conn v Am Elec Power Co and Open Space Inst v Am Elec Power Co 582 F3d 309 (2d Cir 2009). Comer, for example, alleged that GHGs emitted by defendants caused sea levels and air and water temperatures to rise, all contributing, in turn, to increased severity of the damage caused by Hurricane Katrina on the coast of the Gulf of Mexico. 228 Massachusetts v Environmental Protection Agency 549 US 497, 528–29 (2007) (ellipsis in original). Courts in the United States have found in the context of CGL insurance that carbon dioxide does not qualify as a ‘pollutant’ as typically defined in insurance policies. See, eg, Donaldson v Urban Lead Interests, Inc 564 NW2d 728, 732 (Wis 1997) (‘unlike the nonexhaustive list of pollutants contained in the pollution exclusion clause, exhaled carbon dioxide is universally present and generally harmless in all but the most unusual instances’). 229 In particular, the 004 Form expressly requires that the discharge should be ‘not expected or intended’. 230 See above ch 8.

10.119

214

The Exclusions

Product Pollution Liability 10.120

A further qualification in the Bermuda Form is that coverage is provided for Product Pollution Liability. This is in turn defined in Article III(Z) as follows: ‘Product Pollution Liability’ means liability or alleged liability for Personal Injury or Property Damage arising out of the end use of the Insured’s Products, if such use occurs after possession of such goods or products has been relinquished to others by the Insured or by others trading under its name and if such use occurs away from premises owned, rented or controlled by the Insured; such goods or products shall be deemed include any container thereof other than an Automobile, Watercraft or Aircraft.

10.121

This provision preserves coverage in respect of pollution claims brought against policyholders when, in certain situations, their products cause (or are alleged to cause) pollution. For example, there has been a significant amount of litigation in the United States against gasoline manufacturers, distributors and retailers in relation to contamination of municipal and other water supplies alleged to have resulted from the inclusion of an additive in petroleum products.231 A claim under a Bermuda Form policy would be excluded if the pollution took place at, for example, a gasoline station that was owned, rented or controlled by the policyholder. However, a claim could be made in respect of a policyholder’s liability for pollution which occurred away from those premises; for example, if pollution was caused by the way in which the products were handled by the ultimate consumer.

10.122

The exception to the exclusion refers to the ‘end use of the Insured’s Products’. We suggest that the concept of ‘end use’ is not confined to the ultimate use by the ultimate consumer. For example, if the consumer bought a product intending to use it in a particular way (for instance, burning a petroleum product in a car engine), but before doing so a leakage resulted in the liability of the policyholder for pollution, the policyholder’s liability would ‘arise out of the end use of the Insured’s Products’. Similarly, if the policyholder supplied the products to an intermediate retailer (for example, the owner of a gasoline station which was independent of the policyholder), and an underground storage tank leaked, any liability of the policyholder would again arise out of the end use of the Insured’s Products. This is because end use, we suggest, should be judged from the perspective of the policyholder and refers to a use that takes place after the policyholder’s physical involvement with the product has ended; that is, after the policyholder has finished with its products, in the sense of having done all that it has to do to them.

10.123

This conclusion is reinforced by the definition of Insured’s Products contained in Article III(Q): ‘Insured’s Products’ means goods or products manufactured, sold, tested, handled or distributed by the Insured or others trading under its name, or tools, uninstalled equipment or abandoned or unused materials that were the subject of completed operations performed for others by the Insured.

231 In 2009, a federal jury in Manhattan awarded more than US$100 million to the City of New York in a claim against ExxonMobil for contaminating the city’s groundwater supplies with an additive known as MTBE (methyl tertiary butyl ether).

The ‘Nuclear’ Exclusion

215

This definition, with its reference to ‘completed operations’, indicates that the policy covers liability for products which attaches when the operations on the product by the Insured are completed or have come to an end. Our suggested approach to ‘end use’ is also consistent with a line of New York decisions concerning the ‘Products Hazard’ or the ‘Products Completed Operations Hazard’ or similar expressions under CGL policies.232 A policy that covers product liability or the ‘product hazard’ will cover losses from a defect in the policyholder’s product after the policyholder had relinquished control and the product had left the policyholder’s premises. ‘Completed Operations’ cover is the equivalent cover for a service provider. The essential feature of such cover is that it complements but does not overlap with ‘Premises and Operations’ coverage, by which cover is given for damage and injury occurring on or adjacent to the policyholder’s premises or during the progress of operations by the policyholder away from its premises. Thus, in Frontier Insulation Contractors Inc v Merchants Mutual Insurance Co,233 the New York Court of Appeals stated that the

10.124

insurance industry has segregated product liability hazards and the premiums charged therefore by categorizing them as either risks arising while work is in progress, or as those arising from the defective nature of a completed product that has been placed in the stream of commerce. An insured may cover the first risk by purchasing cover for ‘premises operations’. The distinct risk of loss occasioned by a defect in the insured’s product, which manifests itself only after the insured has relinquished control of the product and at a location away from the insured’s normal business premises, is covered by the purchase of separate ‘products hazard’ coverage. This class of coverage generally protects a manufacturer or seller against claims of injury due to a product defect, breach of warranty or misrepresentation.

THE ‘NUCLEAR’ AND THE ‘RADIOACTIVE CONTAMINATION (OUTSIDE THE UNITED STATES)’ EXCLUSIONS The Bermuda Form contains two exclusions limiting coverage for liability arising from nuclear materials. These exclusions have, as far as we are aware, never been litigated in Bermuda Form arbitrations and are not analysed here.

232 See, eg, the definition of ‘products – completed operations hazard’ in Masters and Stanzler, Insurance Coverage Litigation (above n 69) A-50 (1986 ISO claims-made form); see the discussion of ISO ‘products – completed operations hazard exclusion’ in ibid s 14.09. The relevant exception to the pollution exclusion in the 003 Form used the expression ‘The Products Completed Operations Hazard’. This term was not defined elsewhere in the 003 Form. 233 Frontier Insulation Contractors Inc v Merchants Mut Ins Co 667 NYS2d 982 (NY 1997). See generally Henderson. ‘Insurance Protection for Products Liability and Completed Operations’ (above n 110); Theodore A Howard, ‘Products/Completed Operations Coverage 1997: Still Learning Professor Henderson’s Lessons 25 Years Later’ (1997) 7 Coverage 33; Lorelie S Masters and Cherylyn Briggs, ‘The “Great Wall”: Inapplicability of the “Absolute” and “Total” Pollution Exclusions to Drywall and Other Product Claims’, New Appleman on Insurance: Current Critical Issues in Insurance Law (Lexis Nexis, Fall 2010) (with multi-state survey).

10.125

216

The Exclusions

THE ‘ERISA’ EXCLUSION 10.126

The ERISA exclusion seeks to preclude liability for the following: This insurance does not apply to: N. ERISA Liability arising out of any negligent act, error or omission of any Insured, or any other person for whose acts any Insured is legally liable, in the administration of any Insured’s Employee Benefits Programs, as defined below, including, without limitation, liability or alleged liability under the Employee Retirement Income Security Act of 1974, as amended, or any similar provisions of state statutory law or common law or any other law. As used in this Exclusion N, the term ‘Employee Benefits Programs’ means group life insurance, group accident or health insurance, profit sharing plans, pension plans, employee stock subscription plans, workers’ compensation, unemployment insurance, social benefits, disability benefits, and any other similar employee benefits. As used in this Exclusion N, the term ‘administration’ means any of the following acts if such acts are authorized by the Insured: (1) (2) (3) (4)

10.127

giving counsel to employees with respect to the Employee Benefits Programs; interpreting the Employer Benefits Programs; handling of records in connection with the Employee Benefits Programs; or enrolling, terminating or cancelling employees under the Employee Benefits Programs.234

This exclusion seeks to preclude coverage for liabilities arising out of any negligent act, error or omission of the Insured in the administration of any Insured’s ‘Employee Benefits Programme’ as defined in the clause.

THE ‘REPETITIVE STRESS’ EXCLUSION 10.128

At the time that the 004 Form was drafted, repetitive stress and related conditions appeared to be a likely source of substantial liability claims. With the recent focus on ergonomics and prevention, however, liability for repetitive stress injuries appears to be diminishing. The Repetitive Stress Exclusion provides: This insurance does not apply to: O. REPETITIVE STRESS Liability arising out of any repetitive motion, repetitive stress, repetitive strain or cumulative trauma disorder, including, without limitation, (i) liability or alleged liability arising from asserted improper design of goods, equipment, machinery or operations, (ii) failure to warn or properly instruct as to use of goods, equipment or machinery or conduct of operations, (iii) improper supervision of use of goods, equipment or machinery or conduct of operations,

234 Exclusion N in the 004 Form. See too Exclusion (n) in earlier versions of the Form. The term ‘ERISA’ refers to the federal statute governing employee benefits programmes, the Employee Retirement Income Security Act of 1974.

The ‘Securities, Antitrust, etc’ Exclusion

217

or (iv) without limiting the foregoing carpal tunnel syndrome arising or allegedly arising from, without limitation, use of keyboards or finger pads.235

This exclusion finds no direct counterpart in US policies and is not analysed here.

THE ‘SECURITIES, ANTITRUST, ETC’ EXCLUSION This exclusion covers a number of matters preceded by the words:

10.129

Liability arising under any statute, law, ordinance, rule or regulations, whether established pursuant to legislative, administrative, judicial, executive or other authority, of any nation or federal, state, local or other governmental or political body or subdivision thereof …236

These opening words address the legal source of the liability that the policyholder may incur. The exclusion is broadly drafted and would appear to cover most if not all legal sources of liability that could arise in relation to the matters set out in the rest of the clause. The excluded matters are then enumerated in some detail in 10 subparagraphs. One general question often arises whichever of these subparagraphs, or combination thereof, is being considered. Very frequently, the underlying claim against the policyholder will have been advanced under a number of legal theories. Indeed, the ingenuity of lawyers is such that any conceivable legal cause of action may have been alleged against the policyholder. For example, in a product liability claim, the claimant may advance claims based on strict liability, negligence, fraud and local consumer statutes which may (perhaps) come within the concept of ‘deceptive acts and practices in trade and commerce’. What happens if the policyholder is held liable for a particular sum on the basis of all these theories? What happens if, prior to trial, the policyholder settles the claim so that there are no jury findings which differentiate among those theories?

10.130

We think that the correct approach in such cases starts from the fact that the Bermuda Form, in the coverage clause, provides wide coverage and that the insurer has the burden of showing that an exclusion applies to preclude coverage for all of the allegations against the policyholder. It may also be relevant to apply the principle that, where a loss arises from two causes, one covered and one excluded, the loss will be covered.237 Some of the other issues which may arise in this context are discussed in Chapter 5 above.238

10.131

The first paragraph of the exclusion seeks to preclude coverage for:

10.132

(1) the purchase, sale or distribution of securities or offers to purchase or sell securities, or investment counseling or management, including liability under the Securities Act of 1933, the Securities Exchange Act of 1934, the Trust Indenture Act of 1939, the Public Utility Holding Company Act of 1935, the Investment Company Act of 1940, the Investment Advisers Act of 1940, and the so called ‘blue sky’ laws of the various states or other jurisdictions …

235 Exclusion

O in the 004 version of the Form. There is no equivalent exclusion in earlier versions. P in the 004 Form. See also Exclusion (o) in the 001–003 Forms. 237 See above paras 10.17–10.18. 238 See above ch 5, paras 5.42–5.54. 236 Exclusion

218

The Exclusions

An exclusion of this kind is sometimes found in Directors and Officers (D&O) insurance policies.239 10.133

The next paragraph of the exclusion seeks to preclude coverage for liability relating to: (2) antitrust or the prohibition of monopolies, activities in restraint of trade, unfair methods of competition or deceptive acts and practices in trade and commerce including, without limitation, the Sherman Act, the Clayton Act, the Robinson-Patman Act, the Federal Trade Commerce Act, the Lanham Act and the Hart-Scott-Rodino Antitrust Improvements Act …

A question arises as to whether the exclusion for ‘deceptive acts and practices in trade’ includes state consumer statutes designed to give a consumer a statutory remedy for defective products, or whether – given the language of the exclusion as a whole – the words are aimed at statutes seeking to prevent anti-competitive behaviour. We are not aware of law that illuminates this issue other than general principles of policy construction, including the canons of contract interpretation known as ejusdem generis and noscitur a sociis.240 Even if ‘deceptive acts and trade practices’ go beyond anti-competitive practices, it is doubtful whether they extend to ordinary negligence, even if a local statute treats such conduct as within the reach of its state consumer legislation. A state legislature would no doubt be entitled to define such conduct as a deceptive act or practice. We do not consider, however, that a tribunal would be bound by such definition, and is likely to consider whether the relevant acts were ‘deceptive acts and practices’ within the ordinary meaning of that expression; that is, as requiring something more than negligence. 10.134

The remainder of the exclusion covers a variety of matters, seeking to preclude coverage for: (3) fraud or breach of fiduciary duty; (4) criminal penalties; (5) the failure to pay when due any governmental tax including income, excise, property, value added and sales tax, or tariff, license fee or other governmental fee which is incidental to the conduct of business, or any assessment, fine or penalty related thereto; (6) copyright, patent or trademark infringement other than Advertising Liability with respect to titles or slogans; (7) any defect in or impairment to title to real property, including fixtures, whether or not owned by an Insured; (8) disclosure relating to, or other regulation of sales of or offers to sell, real property; (9) liability or alleged liability arising out of employee, officer or director dishonesty;241 or (10) any liability of an employee, officer or director of an Insured entity to such entity.242 239 See Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 15) s 20.05(b), which discusses the issue of causation. 240 See, eg, Holy Angels Academy v Hartford Ins Group 487 NYS2d 1005, 1006–07 (Sup Ct 1985) (applying the doctrine of ejusdem generis (meaning ‘of the same kind or class’)); Popkin v Sec Mut Ins Co of NY 367 NYS2d 492, 494–96 (App Div 1975) (applying the doctrine of noscitur a sociis (meaning ‘it is known by its associates’; ie, that a word in a list may take its character from those with which it is grouped)). 241 See the discussion of the ‘dishonesty exclusion’ in D&O insurance in Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 15) s 20.05(c). 242 See the discussion of the ‘insured versus insured’ exclusion in D&O insurance in ibid, s 20.05(d). See also Carfax Inc v Ill Nat’l Ins Co 7 NYS3d 478 (App Div 2018), denying coverage for defence costs in underlying federal antitrust lawsuit on the grounds that D&O insurance at issue did not apply under the definition of ‘Wrongful Act’ and that ‘antitrust exclusion’ (not defined in the decision) barred coverage.

11 The Conditions

Article VI of the 004 Form and Article V of earlier versions contain a large number of ‘Conditions’. These Conditions address a variety of matters, but have no unifying theme. In many cases, they link with, expand upon, or simply repeat provisions or concepts contained earlier in the Form. Hence, this chapter will frequently refer to a more detailed discussion elsewhere. In contrast to the exclusions whose origins sometimes lie in other policy forms, the Conditions in the Bermuda Form were written for the most part specifically for the Bermuda Form. Accordingly, US case law provides little guidance on most of the Conditions.

11.01

THE PREMIUM CONDITION The Condition addressing premiums states:

11.02

The premium for this Policy is a flat premium and is not subject to adjustment, except as specifically provided herein. The premium shall be paid to the Company.1

Negotiation between policyholder and insurer will usually determine the amount of the premium by reference to various factors, including the nature of the policyholder’s business, market conditions and so forth. Policies written on the Bermuda Form are usually placed through experienced insurance brokers. They will often act for many different policyholders and can therefore be expected to have a feel for what the appropriate premium pricing should be.

11.03

If, in due course, the insurer alleges that a misrepresentation affected the contract negotiation, the question of how the insurance company actually determined the premium may come into more detailed focus, because the insurer will have to show that it would have acted differently if the facts had not been misrepresented. If the insurer alleges that it would have charged a higher premium, it will need to substantiate this. This may require the insurance company to explain its premium calculations and show how it has calculated premium for similarly situated policyholders. Further discussion of misrepresentation and non-disclosure is contained in Chapter 12.

11.04

1 Article

VI(A) of the 004 Form. See also Article V(a) of earlier Forms.

220

The Conditions

11.05

The provision refers to a ‘flat’ premium that is not subject to adjustment ‘except as otherwise provided herein’. The standard policy wording does not provide for any increase or decrease in premium in the event of any contingency. Further premium may become payable by agreement; for example, if the policyholder makes a significant acquisition during an Annual Period and wishes to extend the insurance cover to the newly acquired company,2 or if it elects to purchase reinstatement cover during an Annual Period.3 If the policy is cancelled mid-year, then the cancellation is ‘on a pro rata basis’, so that the policyholder will become entitled to a pro rata return of premium.4

11.06

It is essential for the policyholder to pay the premium promptly. If the insurer does not receive premium or proof of payment within five business days of the commencement of an Annual Period, the policy is automatically cancelled.5 Although the policy will usually be placed through brokers, Article VI(A) requires payment of premium directly to the insurer rather than the broker.

11.07

Article V(a) in Forms 001–003 referred to the premium for each ‘Annual Period’. Although Article VI(A) in the 004 Form does not refer specifically to the premium for an ‘Annual Period’, the concept of the Annual Period remains a key feature of a contract on the Bermuda Form and is discussed in Chapter 2. As defined, the first ‘Annual Period’ of a Bermuda Form contract6 begins on the policy’s inception date, and further Annual Periods commence on each anniversary date of that inception date. Annual Periods continue to run in the event that Coverage B is purchased, as a result of the decision of one or other party to discontinue Coverage A.7

THE INSPECTION CONDITION 11.08

This Condition in the 004 Form provides that: The Company shall be permitted but not obligated to inspect the Insured’s property, operations, books, records and files at any time. Neither the Company’s right to make inspections nor the making thereof or a report thereon shall constitute an undertaking on behalf of or for the benefit of the Insured or others to determine or warrant that such property or operations are safe or are in compliance with any statute, law, ordinance, rule or regulation.8

The equivalent condition in the earlier versions of the Form concerned only the Insured’s property and operations. The right to inspect ‘books, records and files’, which is a common feature of reinsurance contracts, was included for the first time in the 004 Form.

2 Article III(P)(7) of the 004 Form; and Article III(a)(3) of earlier Forms. 3 Article VI(R) of the 004 Form. 4 Article VI(L)(1) of the 004 Form; and Article V(m) of earlier Forms. 5 Article VI(L)(2) of the 004 Form. Under the earlier Forms, cancellation took place 15 days after the insurance company delivered written notice of cancellation to the insured: see Article V(m)(4) of earlier Forms. 6 Article III(C) of the 004 Form; and Article III(l) and (n) of earlier Forms. 7 See the discussion of the Discovery Period Condition below, paras 11.102–11.109. 8 Article VI(B) of the 004 Form. See also Article V(b) of earlier Forms.

The Inspection Condition

221

This right to inspect is not limited to the point in time when a claim is made, but can take place ‘at any time’. A high-level catastrophe liability insurer typically has not tried to predict, by means of an inspection or otherwise, which of its policyholders will be most likely to suffer a catastrophe, or to insist on inspection as part of underwriting. Insurers, of course, may choose not to insure particular applicants with bad safety records or failures to follow best practices. By and large, however, the catastrophe insurer’s underwriting theory is likely to be motivated by a desire to spread risks and obtain a good overall premium base. Thus, the catastrophe insurer may well seek to write 100 major chemical companies, on the basis that it is improbable that more than one of them will be subject to a major catastrophe. This may be less risky than trying to select what appear to be the five ‘best’ or least risky chemical companies – a selection that will reduce the amount of premium income, but where the risk of a major catastrophe will still exist and not be substantially diminished.9 An insurer’s practice for exercising the right to inspect may also vary depending upon the coverage in question.

11.09

It is possible that, when a claim has arisen, the insurer will wish to take advantage of the inspection clause by (for example) asking an expert to inspect the policyholder’s premises. It is perhaps more likely, however, that the insurer will wish to see the policyholder’s books and records. We do not believe that the clause obliges the policyholder to allow the insurer to inspect confidential documents to which the policyholder is entitled to maintain a claim for privilege.10 Clear and express words would, in our view, be required to override the right of a policyholder to claim privilege.

11.10

The right to inspect, the making of an inspection, or the making of a report do not

11.11

constitute an undertaking on behalf of or for the benefit of the Insured or others to determine or warrant that such property or operations are safe or are in compliance with any statute, law, ordinance, rule or regulation.11

This appears to be a ‘belt and braces’ provision, which may have been included with the general aim of preventing prejudice to the insurance company from any inspection or report that it carries out, for example, by having an estoppel argument raised against it.12 The Bermuda Form contains no warranty in favour of the insurer that the policyholder’s property or operations are safe or are in compliance with any law, rule or regulation. Indeed, the purpose of a liability policy is to protect the policyholder in case things go wrong – for example, if a safety regulation is breached. This clause may also have been thought useful to avoid any claim by a third party that the insurer, because it has the

9 The point is well put in the Sherlock Holmes novel by Sir Arthur Conan Doyle, The Sign of Four (first published 1890): ‘“Winwood Reade is good upon the subject”, said Holmes. “He remarks that, while the individual man is an insoluble puzzle, in the aggregate he becomes a mathematical certainty. You can, for example, never foretell what any one man will do, but you can say with precision what an average number will be up to. Individuals vary, but percentages remain constant. So says the statistician.”’ 10 See the discussion of the Assistance and Co-operation Condition below, paras 11.17–11.22. See generally ch 16 for consideration of the question of privilege, in particular paras 16.52–16.58 for discussion of the New York case law concerning inspection and co-operation clauses and common interest privilege. 11 Article VI(B) of the 004 Form; and Article V(b) of earlier Forms. 12 For a discussion of estoppel, see ch 13 below.

222

The Conditions

means to discover deficiencies in the insured’s safety practices, might be liable for thirdparty loss. The right to inspect, in other words, is intended exclusively for the insurer’s benefit.

THE CROSS-LIABILITY CONDITION 11.12

This Condition, as set out in the 004 Form, has changed from previous versions, in that the clause no longer applies to ‘Property Damage’. In its original version, the Condition provided as follows: In the event of claims being made by reason of personal injury suffered by an employee of one Insured hereunder for which another Insured hereunder is or may be liable, then this policy shall cover such Insured against whom a claim is made or may be made in the same manner as if separate policies had been issued to each Insured hereunder. In the event of claims being made by reason of damage to property belonging to any Insured hereunder for which another Insured hereunder is or may be liable, then this Policy shall cover such Insured against whom a claim is made or may be made in the same manner as if separate policies had been issued to each Insured hereunder. Nothing contained herein shall operate to increase the Company’s limit of liability as set forth in Item 2 of ‘Declarations’.13

The 004 Form deletes the second paragraph of the above provision. 11.13

The clause in the 004 Form therefore remains applicable to the situation where personal injury is suffered ‘by an employee of one Insured hereunder for which another Insured hereunder is or may be liable’. This may happen because the Bermuda Form provides for an extended definition of the ‘Insured’,14 so that, for example, all of the subsidiaries and affiliates of the ‘Named Insured’ may be covered by the policy. The reason for the Cross-Liability Condition is, perhaps, to make it clear that one insured company can claim for personal injury caused to an employee of another co-insured company. The provision also appears, in that situation, to negate the effect of the workers’ compensation exclusion.15 This exclusion, in the 004 Form, excludes coverage for obligations ‘for which the Insured or any company as its insurer may be held liable under any workers’ compensation, unemployment compensation or disability benefits law’.16 If an insured, who does not employ the claimant, is held liable for these matters, a claim can be made against the Bermuda Form insurer. The insurer cannot, in these circumstances, argue that because of the extended definition of ‘Insured’, the workers’ compensation exclusion applies to prevent a claim by the Insured who is liable for the claim, but who is not the claimant’s employer.

11.14

The second paragraph of the Cross-Liability Condition in Forms 001–003 dealt, in a similar fashion, with the case where a claim was made ‘by reason of damage to property

13 Article

V(c) of the 001–003 Forms. See now Article VI(C) of the 004 Form. above ch 5, paras 5.7–5.10. See also ch 9, paras 9.40–9.44 above. 15 Article IV(B) of the 004 Form; and Article IV(b) of earlier Forms. See above ch 10, paras 10.35–10.36. 16 Article IV(B) of the 004 Form; and Article IV(b) of earlier Forms. 14 See

The Assistance and Co-operation Condition

223

belonging to any Insured hereunder for which another Insured hereunder is or may be liable’. The rationale for this provision appears to have been as follows. The Bermuda Form contains an ‘Owned-Property’ exclusion,17 seeking to exclude coverage for property damage to property owned by the insured. The Cross-Liability Condition, insofar as it relates to property damage, therefore enabled a claim to be made against the insurer in circumstances where one insured had damaged another insured’s property. The insurer, in that situation, could not argue that such damage was excluded. The 004 Form, however, revised this condition. Not only did the 004 Form delete the property damage paragraph of the Cross-Liability Condition, but, in addition, its ‘Owned-Property’ exclusion now refers to ‘property owned or occupied by or rented to any Insured’.18 It is also possible that the Cross-Liability Condition seeks to prevent the insurer from pursuing a subrogated claim on behalf of one insured against another insured in the circumstances described in the clause. The final paragraph makes it clear that the policy’s limits of liability are unaffected by these provisions. Again, this appears to be a ‘belt and braces’ provision, as the same ground appears to be covered in the Limit of Liability clause.19

11.15

THE NOTICE OF OCCURRENCE CONDITION20 In the 001–003 Forms, Article V(d) dealt with notice of occurrence and claim. Given its role in triggering coverage, notice of occurrence is fundamental to the Bermuda Form policy, as reflected in the fact that the provision is now designated as a separate article in the 004 Form. The notice provision is discussed separately in Chapter 8.

11.16

THE ASSISTANCE AND CO-OPERATION CONDITION The Assistance and Co-operation Condition comprises four paragraphs in the 004 Form. The first paragraph, which also appeared in earlier versions of the Bermuda Form, provides as follows: The Company shall not be called upon to assume charge of the settlement or defense of any Claim made or suit brought or proceeding instituted against an Insured, but the Company shall have the right and shall be given the opportunity to associate with the Insured or the Insured’s underlying insurers or both in the defense and control of any Claim, suit or proceeding relative to any Occurrence where the Claim or suit involves, or appears reasonably likely to involve, the Company, in which event the Insured and the Company shall cooperate in all things in the defense of such Claim.21

17 Article IV(D) of the 004 Form; and Article IV(d) of earlier Forms. See above ch 10, paras 10.43–10.52. 18 Article IV(D)(1) of the 004 Form (emphasis added). 19 Article II(D) of the 004 Form; and Article II(b) of earlier Forms. 20 Article V(d) of the 001–003 Forms; and Article V of the 004 Form. 21 Article VI(D) of the 004 Form. Article V(e) of earlier Forms contains only the text appearing in Article VI(D)(1) of the 004 Form, but not provisions that appear in Article VI(D)(2)–(4) of the 004 Form.

11.17

224

The Conditions

11.18

Litigation between policyholders and insurance companies in the United States frequently involves the question of whether the insurer has a duty to defend the policyholder, or pay the policyholder’s defence costs, in a claim against the policyholder. The insurer’s duty to defend and pay defence costs is normally wider than its duty to indemnify. New York courts have held that the insurer is obligated to defend a claim against its policyholder unless it can establish as a matter of law, looking at the complaint against the insured and the policy, that no possible factual or legal basis exists on which the insurer might eventually be obligated to indemnify the policyholder.22 Disputes relating to the duty to defend will normally concern the policyholder’s primary or first-level excess insurance companies, rather than the insurers writing a high-level catastrophe layer. The Bermuda Form Condition makes it clear, in any event, that the excess insurer writing on the Bermuda Form has no duty to step in and assume the policyholder’s defence.

11.19

If, however, the claim or suit against the policyholder will involve, or appears reasonably likely to involve,23 the Bermuda Form excess policy, the insurers are entitled to take part in the defence of the claim. If the insurer exercises the option to participate in the defence of a claim, both parties are required to co-operate ‘in all things in the defense of such claim, suit or proceeding’. It could be argued that this would include an obligation to pay the policyholder’s defence costs as they fall due. In contrast to the Appeals Condition, however, no such obligation is spelt out in the Assistance and Co-operation Condition. The insurer’s liability for defence costs generally is addressed in Chapter 5 above (in particular paragraphs 5.20–5.23). The final paragraph of Article VI(D) makes it clear that the insurer’s own costs of any claims representation do not count towards the Ultimate Net Loss under the policy and thus do not erode the policyholder’s retention.24

11.20

The Assistance and Co-operation Condition is the only contractual provision in the Bermuda Form that specifically requires the policyholder’s assistance and co-operation, and Article VI(D)(1) is limited to the situation where the Bermuda Form insurer seeks to associate in the defence and control of an underlying claim. Insurers sometimes seek to allege a more general duty by the policyholder to co-operate, for example, in relation to the provision of information to the insurer. The earlier versions of the Bermuda Form contained no express provision requiring the policyholder to provide information to the insurers, although a policyholder would have been well advised to provide

22 Villa Charlotte Bronte Inc v Commercial Union Ins Co 487 NYS2d 314, 315 (NY 1985); Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) (hereafter Masters and Stanzler, Insurance Coverage Litigation) ch 3; Barry R Ostrager and Thomas R Newman, Handbook on Insurance Coverage Disputes, 19th edn (New York, Wolters Kluwer, 2019) (hereafter Ostrager and Newman, Handbook on Insurance Coverage Disputes) ch 5. 23 See also the discussion in ch 8 above, paras 8.22–8.30, of the words ‘likely to involve this Policy’ in the Notice of Occurrence Condition. 24 That provision states: Those expenses incurred by the Company on its own behalf in connection with claims representation pursuant to this Condition D shall be at its own expense and shall not be part of Ultimate Net Loss. Article VI(D)(4) of the 004 Form. Earlier Forms do not include a parallel provision.

The Assistance and Co-operation Condition

225

information and to respond to reasonable requests for information. Yet, Article VI(D)(2) of the 004 Form now provides: The Insured shall furnish promptly all information reasonably requested by the Company with respect to any Occurrence, both with respect to any Claim against the Insured and pertaining to coverage under this Policy.25

This Condition does not, in our view, require the policyholder to disclose documents that are privileged,26 and such documents could not be ‘reasonably requested’ under the clause.27 If the policyholder breaches the obligation of co-operation, what is the consequence? Unlike other provisions of the Form, in particular the notice of occurrence article, the assistance and co-operation provision is not drafted as a condition precedent or as a provision resulting in forfeiture of cover. Accordingly, it is possible that the clause would be interpreted so that any breach would sound in damages only, and that the insurer would need to establish that it has suffered loss from the alleged breach. There is, however, a line of case law holding that a lack of co-operation can void coverage, albeit only where the insurer discharges the heavy of burden of showing (1) that the insurer acted diligently in seeking to bring about the insured’s co-operation, (2) that the efforts employed by the insurer were reasonably calculated to obtain the insurer’s co-operation, and (3) that the attitude of the insured was one of ‘willful and avowed obstruction’.28 The insurance company may waive a defence of failure to co-operate.29

25 This provision is consistent with the majority rule in the United States placing a duty to co-operate on the insured. As stated in the Restatement of the Law, Liability Insurance, an insured ‘seek[ing] liability insurance from an insurer’ has ‘a duty to cooperate with the insurer’ and thus to provide ‘reasonable assistance to the insurer: (1) In the investigation and settlement of the legal action for which the insured seeks coverage’. American Law Institute, Restatement of the Law, Liability Insurance (St Paul, MN, American Law Institute Publications, 2019) s 29 (hereafter Restatement Liability Insurance). 26 See ch 16 below, in particular paras 16.49–16.57; Masters and Stanzler, Insurance Coverage Litigation (above n 22) s 5.05; Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 22) s 2.07(b). 27 Under the Restatement Liability Insurance, an insurer does not have the right to obtain privileged information that could harm the policyholder’s interests: An insurer does not have the right to receive any information of the insured that is protected by attorneyclient privilege, work-product immunity, or a defense lawyer’s duty of confidentiality if that information could be used to benefit the insurer at the expense of the insured. Restatement Liability Insurance s 11(2); see also ibid s 14(1)(b) (stating that insurers have the duty to require defense counsel to protect privileged or confidential information from disclosure if the information could be used ‘to benefit the insurer at the expense of the insured’). 28 Thrasher v US Liab Ins Co 225 NE2d 503, 508 (NY 1967) (quoting Coleman v New Amsterdam Cas Co 160 NE 367, 369 (NY 1928); other citations omitted); US Underwriters Ins Co v 203–11 W 145th St Realty Corp No 8880, 2001 WL 604060 (SDNY 2001) 31 May (citing Garcia v Abrams 471 NYS2d 161, 163 (App Div 1984)); Continental Cas Co v Stradford 847 NYS2d 631 (App Div 2007), aff ’d as modified 871 NYS2d 607 (NY 2008) (appellate court found that the policyholder’s conduct and the insurer’s two-month delay in disclaiming coverage combined to create a question of fact, precluding summary judgment for the insurer on the issue of the policyholder’s alleged breach of the co-operation clause); see also Masters and Stanzler, Insurance Coverage Litigation (above n 22) s 5.05; Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 22) s 2.07(a); compare Restatement Liability Insurance s 30(1) (stating the majority rule in the United States that an ‘insured’s breach of the duty to cooperate relieves an insurer of its obligations under an insurance policy only if the insurer demonstrates that the failure caused or will cause prejudice to the insurer’). 29 See Luria Bros & Cos v Alliance Assurance Co 780 F2d 1082 (2d Cir 1986) (applying New York law); Vanguard Ins Co v Polchlopek 275 NYS2d 515 (NY 1966). On waiver, see generally ch 13 below.

11.21

226

11.22

The Conditions

The third paragraph of the Assistance and Co-operation Condition provides: If liabilities, losses, costs and/or expenses are in part covered by this Policy and in part not covered by this Policy, the Insured and Company shall use their best efforts to agree upon a fair and proper allocation thereof between covered and uncovered amounts, and the Insured shall cooperate with such efforts by providing all pertinent information with respect thereto.30

We suggest that the clause is not aimed at radically altering the applicable tests for whether an exclusion operates, as discussed in Chapter 10 above. That would be a surprising conclusion for a clause whose context is assistance and co-operation between policyholder and insurer, and which contains no criteria by which allocation is to be made, beyond describing it as ‘fair and proper’. Situations may arise, however, where an insurer contends that a monetary loss or expense relates both to covered and uncovered matters, and where the allocation as between those two categories is not obvious.31 In such circumstances the parties are to co-operate and use their best efforts to agree on an appropriate allocation, although in the absence of agreement the problem will need to be solved by the arbitral tribunal.

THE APPEALS CONDITION 11.23

The Appeals Condition of the 004 Form is in substance largely the same as in earlier Forms. It provides that: In the event the Insured or the Insured’s underlying insurers elect not to appeal a judgment in excess of the retention or the underlying limits, as the case may be, the Company may elect to make such appeal at its own cost and expense and shall be liable for the taxable costs and disbursements of such appeal and post-judgment interest on the judgment appealed from accruing during such an appeal. In no event, however, shall liability of the Company for Ultimate Net Loss exceed the applicable limit of liability plus the costs and expenses of such appeal.32

11.24

Accordingly, the insurer has the right to appeal a judgment against the policyholder where the judgment is ‘in excess of the underlying limits’. The provision, therefore, operates in circumstances where the Bermuda Form insurer’s policy is exposed. It may happen that the policyholder has no real economic interest in any appeal; for example, where an appeal, even if successful, would not salvage monies paid by the policyholder beneath the point where its insurance programme operates, and where the underlying claim gives rise to no exposure above the vertical limits of that programme. The policyholder may, therefore, have no real appetite for appealing an underlying judgment, as its insurers will be responsible for the loss in any event. Hence, there is a need to give the Bermuda Form insurer the right to appeal.

30 Article

VI(D)(3) of the 004 Form. above ch 5, paras 5.42–5.54 for a discussion of allocation. 32 Article VI(E) of the 004 Form. See also Article V(f) of earlier Forms. 31 See

The Loss Payable Condition

227

If the insurance company decides to pursue such an appeal, then the Appeals Condition provides that it does so at its own cost and expense. The insurance company will also be liable for the taxable costs and disbursements of any such appeal, and post-judgment interest accruing during such an appeal.33 This provision contemplates that additional sums, by way of costs or interest, may become payable to the claimant who has sued the policyholder, and the Bermuda Form insurer will have to pay these costs if it exercises its right of appeal.

11.25

If costs are incurred because the insurer exercises its right of appeal, it would appear just that these costs should not reduce the amount of the indemnity to which the policyholder would otherwise be entitled under the Bermuda Form policy. The Appeals Condition does produce this result, at least to some extent, in that it provides that the liability of the insurer for ultimate net loss should not ‘exceed the applicable limit of liability plus costs and expenses of such appeal’. However, two features of this provision are worth noting. First, in earlier versions of the Form, the relevant provision related only to the ‘per-occurrence’ limit, leaving open the possibility that the aggregate limit would be eroded. The 004 Form makes clear that the aggregate limit is not eroded. Second, it is not clear that the ‘costs and expenses of such appeal’, referred to above, include such matters as taxable costs and interest on the judgment. We think, however, that such costs and interest are included. This is because the expression ‘taxable costs and disbursements of such appeal and post-judgment interest on the judgment appealed from or interest on judgments incidental thereto’ is properly regarded as an example of the ‘costs and expenses of such appeal’ referred to above.

11.26

THE LOSS PAYABLE CONDITION The Loss Payable Condition provides as follows: Liability under this Policy with respect to any Occurrence shall not attach unless and until: (1) the Insured’s underlying insurer(s) or the Insured shall have paid the greater of the amount of any applicable underlying limits or the applicable retention set forth in Item 2 (a) of the Declarations; and (2) the Insured’s liability covered hereunder shall have been fixed and rendered certain either by final judgment against the Insured after actual trial or by settlement approved in writing by the Company, and the Insured shall have paid such liability. Any consideration paid by the Insured or the Insured’s underlying insurers other than in legal currency shall be valued at the lower of cost or market, and any element of the Insured’s profit or other benefit to the Insured shall be deducted in determining the value of such consideration. The Company may examine the underlying facts giving rise to a judgment against or settlement by the Insured to determine if, and to what extent, the basis for the Insured’s liability under such judgment or settlement is covered by this Policy.

33 This language of the 004 Form spells out rather more clearly the concept of ‘interest on judgments incidental thereto’, which is the expression used in earlier Forms.

11.27

228

The Conditions

The Insured shall make a definite demand for payment for any amount of the Ultimate Net Loss for which the Company may be liable under this Policy within twelve (12) months after the Insured shall have paid such amount. If any subsequent payments shall be made by the Insured on account of the same Occurrence or Claim, additional demands for payment shall be made similarly from time to time. Such losses shall be due and payable by the Company thirty (30) days after they are respectively paid by the Insured, demanded and proven in conformity with this Policy.34

11.28

This clause encompasses a number of concepts which are central to the question of when the policyholder’s cause of action arises and hence when a claim can properly be made under the policy. The Loss Payable Condition is therefore critical if an issue arises as to whether the policyholder’s claim is time barred by the applicable statute of limitations. Even if no limitation issue arises, it is potentially important to the period for which a policyholder can claim interest on a successful claim.

11.29

The clause refers, initially, to the ‘attachment’ of liability under the policy, and this is itself dependent on the fulfilment of the requirements set out in (1) and (2) above which concern payments up to the excess point and, in respect of payments above the excess point, the finality of judgments or the written approval of settlements by the insurer. The provision for a settlement to be ‘approved in writing by the Insurer’ can give rise to disputes when a policyholder does not reach out to seek approval in advance of the conclusion of a settlement. The fulfilment of the requirements for attachment set out in the first paragraph of the clause does not mean, however, that the insurer is required to pay the claim. The final paragraph of the clause sets out a number of further conditions which need to be satisfied before the claim is ‘due and payable’. These later requirements include the making of a demand for the losses, which must also be ‘proven in conformity with this Policy’. The insurer’s obligation to pay only arises 30 days after these later requirements have been fulfilled. Attachment to the Policy

11.30

The clause begins by referring to the ‘attachment’ of ‘Liability under this Policy’. This depends upon the fulfilment of two pre-conditions.

11.31

The first is that the underlying insurers or the insured must have paid the amount of the applicable underlying limits. This repeats the substance of the Limits of Liability provisions in Article II discussed in Chapter 9 above – namely, that the insurance attaches excess of the retention amount specified on the Declarations page or (if higher) the limits of the underlying insurances. The clause goes further, however, in making it clear that the retention amount, or the underlying limits, must have been paid. Payment by the policyholder or its underlying insurers is sufficient for this purpose. Because payment by the insured is also sufficient, there is no risk under the Bermuda Form (as there is under some excess policies) that a settlement with an underlying insurer for less than policy

34 Article

VI(F) of the 004 Form. See also Article V(g) of earlier Forms.

The Loss Payable Condition

229

limits might bar coverage.35 The Bermuda Form insurer will not ‘drop down’ to cover any shortfall; so if the policyholder has, say $100 million of underlying coverage but agrees to settle for $50 million, it will be uninsured to the extent of $50 million. But once it has itself paid that shortfall, the insurer’s ‘liability’ under the policy will attach and the policyholder can begin to make recovery under the Bermuda Form policy. The second is that the policyholder’s liability must have been crystallised either by judgment or settlement approved in writing, and that ‘the Insured shall have paid such liability’.36 These words contemplate that the judgment or settlement will not be covered by an underlying insurer, and that therefore the policyholder will have paid the amount from its own resources. This fits naturally with the Limits of Liability provisions: if the underlying insurer has met the liability for the judgment or settlement, then there would be underlying insurance which would mean that the excess point has not been reached.

11.32

The Bermuda Form does not explain or expand upon the concept of ‘attachment’; the term is not used elsewhere in the Form. It is clear, however, that attachment does not mean that the insurer has an obligation to pay the policyholder’s claim. This is addressed later in the clause, which identifies when the losses are ‘due and payable’. The concept of ‘attachment’ emphasises, first, that there is no liability for payments which fall beneath the excess point. Insofar as there are payments for judgments or settlements above the excess point, we suggest that the concept of ‘attachment’ also serves to identify the judgments or settlements that qualify for indemnity under the policy, and possibly the sequence in which they do so. Accordingly, once liability for a particular judgment or settlement has attached, it cannot later be unattached and replaced by a different judgment or settlement. This may be important in the limitation context: if a time bar has intervened, then a settlement or judgment which has attached cannot later be replaced by a subsequent settlement for which there is no time bar.

11.33

Approval of Settlements The policyholder should seek the written approval of the insurer to any settlements within the layer written by the insurer.37 The opening words of sub-paragraph (2) refer to ‘the Insured’s liability covered hereunder’,38 and thus indicate that the policyholder does not need to seek approval for settlements falling in the layers below or above the Insured’s layer. This seems unsurprising, since it would be strange to require a policyholder to obtain the consent of a high-level insurer in the tower (or indeed all insurers in the tower if the policies are all written on the Bermuda Form) to every 35 Thus, disputes raising the ‘Zeig rule’, named after the decision by the US Court of Appeals for the Second Circuit in Zeig v Massachusetts Bonding & Insurance Co 23 F2d 665 (2d Cir 1928) (applying New York law), do not typically arise under the Bermuda Form. 36 Article VI(F) of the 004 Form (emphasis added). 37 See ch 5, paras 5.34–5.40 for further discussion of approval of settlements, and paras 5.21–5.23 for discussion of the impact of the Loss Payable Condition on the recovery of defence costs. 38 Contrast the language of the definition of Ultimate Net Loss, which refers to the sum which ‘is, or but for the amount thereof, would be covered under this Policy’.

11.34

230

The Conditions

settlement concluded by the policyholder, including those within the self-insured retention, however low. Equally, a high-level insurer is unlikely to wish to incur time and expense in considering and investigating, and giving or withholding approval, to settlements which contribute to a loss which is far beneath that insurer’s layer, and may never reach it. The conclusion that the clause applies only to settlements within the insurer’s layer is reinforced by the concluding words, namely that the ‘Insured’ shall have paid such liability. These words envisage that the settlement or judgment, forming part of the policyholder’s Ultimate Net Loss for which indemnity is claimed, will be paid by the policyholder itself rather than (for example) by an insurer within the tower. If the intention had been to require approval for settlements beneath the layer written by the insurer, the clause would have referred to payment by the insured’s underlying insurer as well as the insured.39 11.35

It is usually advisable for the policyholder to seek approval before a settlement is finally concluded, since this will obviate later debates. Insurers will sometimes refrain from giving express approval, but instead state that they will nevertheless not thereafter rely upon the absence of written approval as a defence to the claim, whilst reserving all other defences to coverage. The policyholder is then in a position to conclude settlements which it considers to be reasonable, albeit that it will be open to the insurer to challenge the reasonableness of such settlements, or advance arguments other than lack of approval, at a later stage. If such confirmation can be obtained from the insurer, it will be particularly useful in the context of settlements where the policyholder is facing mass tort litigation, which may require numerous individual settlements to be negotiated and concluded simultaneously. Insurers are sometimes reluctant to give any approval for fear that doing so could be construed as an admission that the claim is covered. In our view, this concern should not be well founded. Approval of a settlement should not be taken as a promise that it will be the subject of indemnity. An insurer can perfectly well approve a settlement without conceding that the underlying claim is covered, that no exclusion applies or that any other condition of coverage has been satisfied.

11.36

Nevertheless, although prior approval is obviously the safest course, a potentially important question is whether or not the insured can seek approval subsequent to the conclusion of a settlement. Unlike the provisions in some insurance policies, the Loss Payable Condition in the Bermuda Form is not expressed as a contractual obligation by the policyholder to seek and obtain consent at a particular time. Rather, the clause is concerned with the timing of the attachment to the policy of judgments and settlements, and thereby the crystallisation of the policyholder’s potential claim under the policy. (As discussed below, further conditions need to be fulfilled before the policyholder has a cause of action and the insurer is in breach.) Thus, the policyholder does not breach the policy by failing to obtain consent prior to concluding a settlement. This serves to distinguish the Loss Payable Condition from cases where courts have found that policyholders breached a policy condition by failing to request the insurer’s consent before

39 Compare the opening words of sub-paragraph (1) of Article VI(F) which refers to payment by ‘the Insured’s underlying insurer(s) or the Insured’.

The Loss Payable Condition

231

the settlement was finalised. For example, in Vigilant Insurance Co v Bear Stearns Cos,40 the court found that the policyholder had breached a policy condition and was unable to recover in circumstances where a settlement was subject to approval by the court, but the insurers’ consent was sought only after the conclusion of the settlement, but prior to the court’s approval. The court’s finding of breach was based on its conclusion that, once the settlement had been concluded, the policyholder was ‘not free to walk away’41 from the settlement. Given the language of the Loss Payable Condition and its function as described above, there is a strong argument that prior approval is not required. The clause refers to ‘approval’ rather than consent. Whilst ‘consent’ may, certainly in some contexts, carry the connotation that it will be obtained in advance, the same is not necessarily true of an approval. It is far from unusual, in everyday life, for an approval to be sought after the event – for example, an employee’s expenses for a business trip may require approval after its conclusion but before they are payable. The clause does not refer to ‘prior approval’.

11.37

Furthermore, once it is acknowledged that insurer approval is required only for settlements which are within the insurer’s coverage layer, the difficulties with interpreting the provision as requiring prior approval become apparent. This is because there are many situations, especially in the context of a mass tort, where it will be unclear, at the time of settlement, whether an individual settlement is within or outside a particular insurer’s layer. For example, a policyholder is likely, when settling claims in a mass tort context, to be calculating the amount spent on settlement and defence costs, with a view to identifying the point in time at which its self-insured retention is exhausted and when each insurer’s layer is reached. The policyholder might seek and obtain approval for a collection of settlements based upon that calculation. Subsequently, however, once the insurer articulates and develops its defences to the claim, it may become apparent that the policyholder’s calculation as to when the excess point was reached might require review and recalculation. This may arise, for example, because a ‘maintenance deductible’ is applicable, in which case the excess point will arguably have been reached at a stage later than that initially calculated by the policyholder. There may be considerable uncertainty as to the size of the maintenance deductible, certainly at a time when the settlements are being concluded. Similarly, other defences may subsequently be raised by the insurer which have the same effect of removing particular settlements from the policyholder’s Ultimate Net Loss (for example, because they are covered by an exclusion) or reducing the recoverable amount thereof (for example, because a settlement was to a degree unreasonable in amount). Or the Ultimate Net Loss may be reduced by ‘salvages or recoveries’, and these will affect when the insurer’s layer was reached. These various possibilities may mean that settlements within the insurer’s layer may be rather different from those which the

11.38

40 Vigilant Ins Co v Bear Stearns Cos 855 NYS2d 45 (NY 2008). 41 Ibid at 49. There, the policyholder reached an agreement in principle on 12 December 2002 with the US Securities & Exchange Commission requiring payment of US$80 million and further entered into a consent agreement with the agency a few months later, acceding to entry of a final judgment against it by a federal court. Ibid at 47. The clause there sought the insurer’s consent for a settlement exceeding a threshold of US$5 million. Ibid at 48.

232

The Conditions

policyholder reasonably understood to be within the layer at the time that approval was originally sought. These possibilities would not, at least in most cases, present any real difficulties if it were permissible to seek approval after conclusion of the settlements, since it would be possible for the policyholder, after matters have become clearer and the insurer’s arguments have been articulated, to ask for approval for a group of settlements different from those which were the subject of the previous request. However, if the Loss Payable Condition were to be interpreted as requiring prior approval, then the only safe course for the policyholder would be to seek approval for any settlement that might conceivably impact a particular insurer’s layer depending upon the defences that are in due course advanced. 11.39

If the Loss Payable Condition is interpreted as not requiring the prior approval of settlements, then this may make it difficult for an insurer to rely simply upon the absence of approval as a defence to the policyholder’s claim. If an insurer were to be asked for approval after the conclusion of the settlement, the duty of good faith and fair dealing will apply, so that it is unlikely that the insurer would be permitted to withhold its approval arbitrarily or unreasonably,42 or to do so simply on the basis that this would advance its financial interests, by providing a defence to the claim based upon lack of approval.43 This means that an insurer that wished to withhold approval would need to have a substantive reason for the refusal, for example because the settlement was arguably unreasonable or collusive. In addition, in many cases where approval is or might be sought after the event, the insurer is likely to have denied liability for the claim on other grounds. In such circumstances, the policyholder may also be able to contend that the insurer’s denial of liability meant that it was no longer necessary to seek (retrospective) approval for the settlements which had been concluded.44 The Accrual of the Policyholder’s Cause of Action and the Limitation Period

11.40

The first sentence of the final paragraph sets a time limit of 12 months, from the date of payment, to make a demand for payment under the insurance policy. The policyholder will have to calculate the Ultimate Net Loss for which the demand is to be made. Additional demands can and should be made within a similar period of 12 months from

42 Arno Schefler v Livestock Cas Ins Co, 44 AD 2d 811 (App Div 1974); Matter of Prudential Prop & Cas Ins Co (King) 604 NYS2d 136 (App Div 1993); Matter of State Farm Mut Auto Ins Co v Callisto 255 AD2d 876 (App Div 1998). See above ch 5, paras 5.34–5.40. 43 Ansonia Assoc Ltd P’ship v Public Serv Mut Ins Co 257 AD2d 84, 86 (App Div 1999) (‘there is a well- established proscription against permitting an insurer to place its own financial interests above those of its insured’). 44 Bunge Corp v London & Overseas Ins Co 394 F2d 496, 497 (2d Cir 1968) (applying New York law); Isadore Rosen & Sons, Inc v Sec Mut Ins Co of NY 339 NYS2d 97, 1010 (NY 1972). See also Luria Bros v Alliance Assurance Co 780 F 2d 1082, 1091 (2d Cir 1986) (when the insurer rejects coverage, the policyholder may settle rather than proceed to trial to determine its liability); Do-Re-Knit Inc v National Union Fire Ins Co of Pittsburgh Pa 491 F Supp 1334 (EDNY 1980) (‘if an insurer declines a claim or denies liability, the insured is relieved of the duty to satisfy the conditions precedent to recovery under the policy’), citing prior authority, including Sherri v Nat’l Sur Co 243 NY 266 (NY 1926). As to the limits of this principle, see Seward Park Housing Corp. v Greater NY Mut Ins Co 43 AD3d 23 (App Div 2007). See also ch 5, paras 5.34–5.40 above.

The Loss Payable Condition

233

the date of payment. The Bermuda Form does not, however, require that the demand should be in a particular format, or contain particular information. If the policyholder fails to make a demand within the time limit, it might be argued that the claim can no longer be made. However, the obligation in the first sentence is not expressed as a condition precedent and does not contain language (such as ‘if’ or ‘unless’ or ‘until’), which is the unmistakable language of condition.45 It follows that a ‘late’ demand for payment may still be valid and effective, though the insurer would be entitled to compensation for any loss resulting from the lateness (though, apart from any loss resulting from changes in interest rates, it seems unlikely that there would be such loss). Issues may also potentially arise as to when an insured is required to make a demand in the context of continuing litigation with an uncertain outcome.46

11.41

The final sentence makes payment ‘due and payable’ within 30 days after the losses have been ‘respectively paid by the Insured, demanded and proven in conformity with this Policy’. Accordingly, a number of conditions have to be met before the insurer is required to pay. It has long been held under New York law that an insurer is not in breach of contract, by failing to pay an insurance claim, prior to the time when the loss is ‘due and payable’.47 Thus, in Continental Casualty Co v Stronghold Insurance Co, the court said:

11.42

An express contract for indemnity, however, remains a contract. Hence the parties are free, within limits of public policy, to agree upon conditions precedent to suit. Common conditions include filing of proofs and allowing the insurance company time to investigate and pay the claim. Thus, the rule has evolved in insurance cases that the cause of action accrues ‘when the loss insured against becomes due and payable’ under the policy.48

There is no suggestion in the New York case law that the conditions set out in the Loss Payable Condition (payment, demand, ‘proven in conformity’ and a waiting period of 30 days) are impermissible as a matter of New York public policy. It is difficult to see how there could be any public policy objection to the requirements either that the policyholder (or its lower layer insurer) should have paid the Ultimate Net Loss, or that a demand should be made for payment by the Bermuda Form insurer. The words ‘proven in conformity with this Policy’ impose a requirement that there should be some proof of loss furnished to the insurer. The nature of this proof is discussed below, but there can again be no public policy objection to such a provision. Indeed, the inclusion of conditions relating to proof of loss is (as the above quotation recognises) common,

45 MHR Capital Partners LP v Presstek Inc 12 NYS3d 640, 645 (NY 2009); Hahn Auto Warehouse Inc v Am Zurich Ins Co 18 NY3d 765 (NY 2012). 46 Standard form liability policies in the United States typically contain a provision which has some similarities with the Loss Payable Condition, although few cases in the United States have squarely addressed the clause. Issues as to the timing of the demand are considered in Journal Publ’g Co v Am Home Assurance Co 740 F Supp 1015 (SDNY 1990) (applying New Mexico law); Liberty Mut Ins Co v Those Certain Underwriters at Lloyd’s, London 650 F Supp 1553, 1557 (WD Pa 1987) (applying Massachusetts law). See also Masters and Stanzler, Insurance Coverage Litigation (above n 22) s 19.03[C]. 47 Ames v New York Union Ins Co 14 NY 253, 264–66 (NY 1856); Hay v Star Fire Ins Co 77 NY 235, 242 (NY 1879); Continental Cas Co v Stronghold Ins Co 77 F3d 16, 19–20 (2d Cir 1996); Myers v Cigna Prop & Cas Ins Co 935 F2d 551, 555 (SDNY 1997); Fabozzi v Lexington Ins Co 601 F3d 88, 91 (2d Cir 2010). 48 77 F3d 16, 19–20 (2d Cir 1996).

11.43

234

The Conditions

as can be seen in New York case law stretching back to the nineteenth century.49 In O’Reilly v The Guardian Mutual Life Ins Co of New York,50 the New York Court of Appeals stated the purpose of the condition (relating to proof of loss) as follows: that the insurer may be able intelligently to form some estimate of its rights and liabilities before it is obliged to pay. The condition is therefore satisfied if the policyholder provides reasonable evidence to give assurance that the event had happened upon which the liability of the insurer depends.51 Although the authorities relating to proof of loss provisions are primarily cases concerning first-party property policies or life insurance, the rationale for such clauses is equally applicable to policies covering liability to third parties and where a contractual provision for proof of loss has been included.52 11.44

In the previous editions of this book, we expressed the view that the reference to ‘respectively … proven in conformity’ with the policy was somewhat puzzling. We now consider, against the New York law background described above, that the words impose a straightforward requirement for the policyholder to provide appropriate proof of loss, and that this is a condition precedent to the loss becoming due and payable and hence to the insurer’s obligation to pay. The rationale is to enable the insurer to form an intelligent estimate of its rights and liabilities, and decide whether or not to accept or deny the claim. This conclusion is supported by other policy provisions directed towards the same end and that can be linked with the ‘proven in conformity’ requirement. Thus, the final sentence of the previous paragraph of Article VI(F) provides that the insurer ‘may examine the underlying facts giving rise to a judgment against or settlement by the Insured to determine if, and to what extent, the basis of the Insured’s liability under such judgment or settlement is covered by this Policy’. This provision contemplates an investigation by the insurer of the proof of loss to be provided by the policyholder under the last sentence of Article VI(F). The provision also explains the principal reason why the insurer would pursue such an investigation: to determine if and to what extent the basis of the Insured’s liability under such a judgment or settlement is covered by the policy. Similarly, the Assistance and Co-operation Condition53 contemplates an investigation by the insurer and imposes an obligation on the policyholder ‘to furnish promptly all information reasonably requested by the Company with respect to any Occurrence’. The Inspection Condition54 provides to the insurer a right to inspect the insured’s books,

49 In Steen v Niagara Fire Ins Co 89 NY 315 (NY 1882), the court said that ‘they are words of the under writers, intended for their protection, and qualify the general obligation to make good any loss not exceeding the sum insured or the interest of the assured in the property, by requiring preliminary proof, and a lapse even then of a fixed period before any cause of action can accrue’. A more recent decision by the Court of Appeals case giving effect to a proof of loss provision, in a limitation context, is Medical Facilities Inc v Pryke, 62 NY2d 116 (NY 1984). 50 O’Reilly v The Guardian Mutual Life Ins Co of NY 60 NY 169 (NY 1875). 51 Ibid at 172–73. This decision was applied in Wachtel v Equitable Life Assurance Society 284 NY 345 (NY 1935). 52 In one reported case, the insurers argued that the Loss Payable Condition, which concluded with the words ‘ proven in conformity with this policy’, was akin to a proof of loss provision traditionally contained in fire and property insurance policies. The judge then decided a waiver issue against the underwriters, ‘employing the Underwriter’s analogy to proof of loss provisions’: Liberty Mut Ins Co v Those Certain Underwriters at Lloyd’s, London 650 F Supp 1553 (WD Pa 1987) (applying Massachusetts law). 53 Article VI(D) of the 004 Form. 54 Ibid Article VI(B).

The Loss Payable Condition

235

records and files at any time, and such right can and is likely to be exercised in order to investigate coverage issues. Accordingly, the Bermuda Form as a whole contemplates a proof of loss process at the instigation of the insurer after the attachment of liability under the first paragraph of Article VI(F). In order to fulfil the requirements of this provision, it is not necessary for the policyholder to provide a level of proof equivalent to that which is necessary to prove its case at an arbitration hearing. Such a result would enable the insurer to benefit from wrongly disputing the policyholder’s claim and to have the use of monies for the period required to resolve the claim. In O’Reilly, the court held that the requirement in that case (for proof of death) need not be the ‘full, clear and explicit proof which would be required upon the trial of an issue upon that question’.55 Instead, what was required was ‘such reasonable evidence as the party can command at the time, to give assurance that the event has happened, upon which the liability of the insurers depends’.56 We suggest that this case provides a useful test as to what is required under the Loss Payable Condition. It is arguable that, in view of the provisions described in the preceding p aragraph,57 the standard may be more exacting, since the policyholder is under an express obligation to furnish ‘all information reasonably requested by the Insurer’, and because the insurer is entitled to examine the underlying facts to determine if and to what extent the basis of the Insured’s liability is covered. In order to meet the requirement of ‘proven in conformity’, it is not necessary for the evidence provided by the policyholder to be in a particular format. Nor is it necessary for a formal document headed ‘proof of loss’, such as might be seen in a claim under a first party policy, to be prepared and signed by the policyholder. Fulfilment of the requirement depends upon the substance of the information provided.

11.45

It is only when the requirements of the final sentence have been fulfilled, including the 30-day waiting period, that the insurer can be said to be in breach of contract by failing to pay. This is significant if an issue arises as to whether the policyholder’s claim is timebarred. In an arbitration of a Bermuda Form dispute in England, issues of limitation are governed by New York law.58 The relevant New York statute is section 213(2) of the Civil Practice Law & Rules (CPLR), which provides for a six-year limitation period from the date of the relevant breach. The limitation period under New York law starts to run at the moment when the insurer breaches its obligation to pay. Courts have held that it is at that stage that the policyholder’s cause of action against the insurer accrues.59

11.46

55 60 NY at 173. 56 Ibid. 57 In particular, the concluding words of the second paragraph of Article VI(F) (‘The insurer may examine …’), and Article VI(D). 58 See ch 3, para 3.35. 59 Micha v Merchants Mut Ins Co,463 NYS 2d 110, 111–12 (App Div 1983); Med Facilities Inc v Pryke 62 NY2d 716, 717 (NY 1984); Niagara Frontier Transp Auth v Encon Underwriting Agency, Inc 586 NYS2d 53, 54 (App Div 1992); Chester Med Diagnosis PC v Kemper Cas Ins Co 873 NYS2d 232 (Civ Ct 2008); Flat Ridge 2 Wind Energy LLC v Those Underwriter at Lloyd’s No 650508/2014, 2014 WL 3373435, at *5–*6 (NY Sup Ct 30 June 2014). See also Fabozzi v Lexington Ins Co 601 F3d 88, 91 (2d Cir 2010); and Continental Cas Co v Stronghold Ins Co 77 F3d 16, 19–20 (2d Cir 1996).

236

The Conditions

11.47

If an issue of limitation does arise, then a factual question may also arise as to exactly when the policyholder provided the appropriate proof so as to fulfil the ‘proven in conformity’ condition. In practice, and for the reasons which follow, it is likely that this question can be readily answered by reference to the date when, after investigation, the insurer denied the claim.

11.48

In many cases where there is a developing underlying loss which may reach the higher excess levels, the policyholder will have kept its insurers advised of developments following the provision of notice of occurrence. However, the insurer may well decide to await the outcome of developments before deciding to devote time and resources to the investigation of the possible claim. Once a formal demand under Article VI(F) has been made, however, the insurer is likely to begin its detailed investigation with a view to making a coverage determination. If, following such a demand, the insurer has made no requests for information, or for an inspection of books records and files relating to coverage, within a reasonable time, a tribunal may well conclude that the losses had been sufficiently proven in conformity with the policy at the time of the demand. It is more likely, however, that the demand will then trigger requests for information by the insurer, and Article VI(D)(2) provides that the policyholder should promptly respond to such requests. The insurer also may request an inspection under Article VI(B).

11.49

Whilst the process of consensual requests for and provision of information or an inspection process is continuing, the proof of loss process will be in operation. An insurer who subsequently seeks to contend that the limitation period had started to run during this process, based on the proposition (contrary to the position that the insurer took at the time) that the losses had already been proven in conformity with the policy, would face severe difficulties in that argument. This very point arose in an old case, Barnum v The Merchants’ Fire Insurance Co, decided by the New York Court of Appeals.60 The policy in that case required the submission of proofs of loss under a fire insurance policy as a pre-condition to the insurer’s obligation to pay. However, the policy included a provision that no action on the policy could be brought unless commenced within six months ‘next after the day on which any loss or damage shall occur’.61 The policyholder submitted proofs of loss on 13 June 1879, but the insurer contended that they were not adequate and needed to be amended. Correspondence then ensued, and the proofs were indeed amended, but this took some time. The insurer argued that the action had not been commenced in time. The court rejected this argument, holding that the limitation period did not start to run until the policyholder had provided the proof requested by the insurer: The defendant may, by objecting to the proofs of loss, impose upon the assured the duty of making them complete and removing, if possible, the dissatisfaction of the insurer, and if he chooses to do so, the delay is mutual and the time of limitation necessarily extended. …

60 Barnum v Merchants’ Fire Ins Co 97 NY 188, 195 (NY 1884). 61 It is not unusual for first-party policies to require proceedings to be commenced within a period shorter than the six-year limitation period for breach of contract actions under New York law, and such contractual limitations provisions are not contrary to New York public policy.

The Loss Payable Condition

237

The complaints on the part of the defendant, and efforts on the part of the plaintiffs to comply with them, continued until December. These facts are undisputed. Each party, therefore, assented to the delay, and while the negotiation was in progress, the defendant could not be called upon to pay, and consequently the cause of action did not accrue.

The decision in Barnum might now be analysed as a case of estoppel. However it is analysed, Barnum assists in explaining why many New York authorities treat the limitation period as starting to run on the date when the insurer denies liability.62 In these cases, there has usually been a continuing investigation by the insurer of the policyholder’s claim, and the denial of the claim marks the moment when the investigation has come to an end. Once the claim has been rejected, there is no longer any question of the policyholder having to provide further proof of the loss. Prior to that time, however, the process of providing such evidence is continuing, with the insurer asking for further information. This has the consequence that the insurer is thereafter unable to contend that it was obliged to pay (and that the policyholder’s cause of action had accrued) because sufficient proof had already been provided.

11.50

The practical result is, therefore, that the date of the insurer’s denial of coverage will often mark the start of the running of the limitation period in a dispute under a Bermuda Form policy. In some cases, it will be possible for one or other party to contend that the proof of loss process stopped prior to the denial. For example, the insurer’s requests for information or inspection may cease, or the policyholder may assert that it had already provided enough information and refuse to provide any more, with the consequence that for one reason or another, the consensual process of requests for and provision of information has come to an end. Furthermore, the policyholder cannot indefinitely delay the provision of the necessary proof so that the limitation period never starts to run or is extended unreasonably.63 But accrual of the cause of action is postponed for so long as the delay in submitting proof was reasonable or agreed to.64

11.51

We do not consider that the above analysis, or the case law referred to, is affected by the decision of the New York Court of Appeals in Hahn Automotive Warehouse Inc v American Zurich Insurance Co,65 which is sometimes relied upon by insurers in support of an early commencement of the running of time for limitation purposes. Hahn confirms the well-established general proposition that the statute of limitations begins to run when the cause of action accrues, and that a claim generally accrues at the time of breach. However, Hahn was not concerned with the application of that principle in the context of a claim by a policyholder for recovery of a claim for indemnity under a policy: the case concerned a claim by an insurer for retrospective premiums and similar amounts. The contracts in issue in Hahn had no language which was equivalent to Article VI(F).

11.52

62 See, eg, Niagara Frontier Transp Auth 586 NYS 2d at 54 (finding ‘[t]he breach in the instant case occurred on March 27, 1984, when the insurer denied liability under the contract’); Flat Ridge 2 Wind Energy 2014 WL 3373435, at *5–*6; see also Continental Casualty 77 F3d at 21; No Hero Enters BV v Loretta Howard Gallery Inc 20 F Supp 3d 421, 427 (SDNY 2014). 63 Continental Casualty 77 F3d at 21. 64 Aetna Cas & Sur Co v Bell 390 F2d 612 (1st Cir 1968) (applying New York law). 65 Hahn Automotive Warehouse Inc v American Zurich Ins Co 18 NY3d 765 (NY 2012).

238

The Conditions

The majority in Hahn referred to the decision in Continental Casualty in order to distinguish it, but with no suggestion that either that case or the line of authority cited therein had been wrongly decided.66 Subsequent New York case law has confirmed the continuing applicability of the approach in Continental Casualty.67 In particular, the decision in Deutsche Bank National Trust Co v Quicken Loans Inc68 shows that conditions such as those contained in the final sentence of Article VI(F) are substantive rather than ‘procedural’ conditions, because they are conditions precedent to the insurer’s performance of its obligation to pay covered losses.69 Interest and Other Matters 11.53

Since the policyholder has no cause of action for indemnity until 30 days have passed following fulfilment of the requirements in the final paragraph, it is likely that interest on a successful claim will be awarded by reference to that date.

11.54

The Loss Payable Condition contains a provision dealing with the valuation of consideration paid ‘other than in legal currency’. This may arise if, for example, the policyholder provides goods or services to the claimant in satisfaction or partial satisfaction of the claim. It may also arise in circumstances where the policyholder satisfies the claim by forgoing an existing entitlement to be paid for the supply of goods and services.

11.55

The earlier versions of the Bermuda Form (001–003) contained a final sentence that provided for the currency of payment under the policy (US dollars), and for the rate of exchange to be applied. These provisions are now contained in the currency condition of the 004 Form.70

THE REPRESENTATION CONDITION 11.56

The Representation Condition provides: The Named Insured or such other person as it shall designate in Item 5 of the Declarations shall represent and have authority to bind the Named Insured and any and all Insureds hereunder in all matters under this Policy, including, without limitation, payment of premiums, negotiation of the terms of renewal or reinstatement and the adjustment, settlement and payment of Claims. The Named Insured, by notice to the Company in writing, may designate

66 See fn 5 to the majority decision, at 18 NY3d 765, 772. The dissenting minority also referred (with approval) to Continental Casualty. 67 Superintendent of Fin Servs of NY v Guar Ins Co No 450023/13, 2013 NY Misc LEXIS 2442, at *4 (Sup Ct 10 June 2013); Flat Ridge 2 Wind Energy LLC 2014 WL 3373435, at *5–*6; Deutsche Bank Nat’l Trust Co v Quicken Loans Inc 810 Fd 861 (2d Cir 2015). 68 Deutsche Bank Nat’l Trust Co v Quicken Loans Inc 810 Fd 861 (2d Cir 2015) (applying New York law). 69 Ibid at 866–67. 70 See Article VI(M) discussed at para 11.80 below. See also the discussion of currency in ch 3, para 3.42 above, and in Lesotho Highlands Development Authority v Impregilo SpA and Others [2005] UKHL 43, [2006] 1 AC 221.

The Subrogation Condition

239

a substitute representative, which representative shall, effective as of the date such notice is received, be deemed to be designated in Item 5 of the Declarations.71

A Bermuda Form policy typically provides coverage to a number of different insured companies.72 The Bermuda Form provides that the Named Insured shall represent and has authority to bind both itself and all other insureds ‘in all matters under this Policy’. The Named Insured will often be the ultimate parent within its group. The clause avoids any potential difficulties that might otherwise arise as to whether one company within the group has authority to act for others in relation to the policy.

11.57

THE OTHER INSURANCE CONDITION73 The Other Insurance Condition reiterates that a Bermuda Form policy is excess of other insurance policies, and seeks to prevent contribution claims. This and related clauses are discussed in detail in Chapter 9 (in particular paragraphs 9.28–9.33).

11.58

THE SUBROGATION CONDITION The Subrogation Condition in the 004 Form provides as follows:

11.59

In the event of any payment hereunder, the Company shall be entitled to exercise rights of subrogation, and the Insured shall execute and deliver instruments and papers and do whatever else is necessary to secure such rights. In such case, the Company will act in concert with all other interested parties, including the Insured, concerned in the exercise of rights of recovery. The apportioning of any amounts which may be so recovered, net of expenses, shall follow the principle that any parties, including the Insured, that shall have paid an amount over and above any payment hereunder shall first be reimbursed up to the amount paid by them. The Company is then to be reimbursed out of any balance then remaining up to the amount paid by it; lastly, the parties of whose interests this coverage is in excess, including the Insured, are entitled to claim the residue, if any.74

Rights of subrogation arise, of course, when the insurer has indemnified its policyholder. In these circumstances, the insurer is, in general terms, entitled to the benefit of rights of recourse that the policyholder possesses; for example, the policyholder may possess a right against another entity that is liable to contribute to the loss suffered by the policyholder.75 A detailed discussion of principles of subrogation will be found in insurance law textbooks.76 71 Article VI(G) of the 004 Form. See also Article V(h) of earlier Forms. 72 See ch 5, paras 5.7–5.10 above. See also ch 9, paras 9.40–9.44 above. 73 Article VI(H) of the 004 Form; Article V(i) of earlier Forms. 74 Article VI(I) of the 004 Form. See also Article V(j) of earlier Forms. 75 New York courts, like other courts in the United States, follow the ‘anti-subrogation rule’, which applies principles of equity and public policy to prevent an insurer from passing the loss back to the policyholder or affiliated insured entities or persons – eg, Pennsylvania Gen Ins Co v Austin Powder Co 510 NYS2d 67 (NY 1986). For further discussion of this rule, see Masters and Stanzler, Insurance Coverage Litigation (above n 22) s 19.04. 76 See generally Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 22) s 22.06(c).

11.60

240

The Conditions

11.61

The Subrogation Condition in the Bermuda Form attempts to set out a regime that takes account of the Bermuda Form insurer’s position as a high-level excess insurer. If rights of recourse exist against a third party, a number of parties may have interests in obtaining and receiving that recovery: insurers below the Bermuda Form insurer, insurers writing at higher levels, and the policyholder in relation to any uninsured loss either by way of a self-insured retention or above the total limits of its insurance programme.

11.62

The Bermuda Form attempts to deal with these difficulties by providing that the insurer who has paid a claim under the Form will act in concert with all other interests concerned in making a recovery. A regime is then laid down for the distribution of any recoveries. The clause is not perfectly drafted,77 but the intention appears to be to create a ‘top-down’ regime, so that those who have borne the top end of the loss will receive the recovery in priority both to lower-level insurers and the policyholder in relation to any bottom-end retention.

11.63

The provisions of the Bermuda Form relating to subrogation can bind only the parties to the policy, and are therefore not binding on the policyholder’s other insurers. It is therefore possible to envisage a dispute arising between insurers as to who is entitled to the proceeds of a subrogated claim.

THE CHANGES CONDITION 11.64

The Changes Condition imposes a degree of formality in relation to any changes to the terms of the cover. It provides: Notice to or knowledge possessed by any person shall not effect waiver or change in any part of this Policy or estop the Company from asserting any right under the terms of this Policy. The terms of this Policy may not be waived or changed, except by written endorsement issued to form a part hereof and signed by the Company.78

This provision seeks to preclude the policyholder from advancing waiver arguments, or allegations of an agreed change to the cover, when the waiver or change is not reflected in a signed endorsement. The clause potentially prevents the insurer from advancing such allegations as well, but it is usually the policyholder rather than the insurer who seeks to rely upon waiver. The clause is directed towards rights under the insurance policy, and would therefore appear inapplicable to questions that concern the validity of the policy, such as misrepresentation. The question of waiver generally, including the effect of this condition, is discussed in Chapter 13.

77 In particular, the words ‘any parties, including the insured, that have paid an amount over and above any payment hereunder’ are not particularly apt to denote only payments of sums excess of the Bermuda Form policy under consideration. It is possible to read these words as referring to other underwriters who have simply paid ‘more than’ the Bermuda Form underwriters. Thus, assume a case where the Bermuda Form insurer had paid US$100 million, but underlying insurers had paid US$250 million. It might be argued that, because the underlying insurers had paid an amount ‘over and above’ the payment made by the Bermuda Form insurer, the underlying underwriters should have first call on the recoveries. We do not think, however, that such an interpretation fits easily with the clause as a whole. 78 Article VI(J) of the 004 Form. See also Article V(k) of earlier Forms.

The Assignment Condition

241

THE ASSIGNMENT CONDITION The Assignment Condition provides: ‘Assignment of interest under this Policy shall not bind the Company unless and until its consent is endorsed hereon.’79 This clause prevents ‘assignment of interest’ under a Bermuda Form policy in the absence of the insurer’s consent. Courts in the United States have addressed policy provisions that seek to prohibit assignments of rights under the policy, with the reversal in 2015 of a controversial 2003 decision by the California Supreme Court that had rejected the majority rule. Arguably, that change in California law on this issue has restored the approach and law on this issue to the situation that pertained before the 2003 California decision.

11.65

Under the ‘majority rule’, which New York courts typically have followed, policyholders are allowed to assign the proceeds of an insurance policy after a loss, in spite of a policy’s apparent prohibition against assignment. In reaching this result, courts applying the majority rule have reasoned that, after the loss, the insurance company’s potential liability is fixed, and allowing the policyholder to assign the right to the policy’s proceeds does not affect the insurer’s risk.80 Thus, in Globecon Group LLC v Hartford Fire Insurance Co,81 the US Court of Appeals for the Second Circuit explained traditional New York law on the issue:

11.66

The enforceability of a no-transfer clause in an insurance contract is limited under New York law. As a general matter, New York follows the majority rule that such a provision is valid with respect to transfers that were made prior to, but not after, the insured against loss occurred.82

States other than New York have taken a similar approach. For example, in Elat Inc v Aetna Casualty & Surety Co,83 an appellate court in New Jersey rejected the argument by insurance companies that the policyholders had violated the non-assignment clauses. There, before Elat acquired the assets of another company, groundwater at a site owned by that company became contaminated. The acquired company had assigned ‘all of its rights and claims’ under its CGL policies to Elat. The court upheld Elat’s right to these policy benefits, finding that, once a loss has taken place, the assignment is of the loss and not the policy itself, and thus is not barred by the non-assignment provision in the policy.84 79 Article VI(K) of the 004 Form; and Article V(l) of earlier Forms. 80 Travelers Indem Co v Israel 354 F2d 488, 490 (2d Cir 1965) (applying New York law) (citing Courtney v New York City Ins Co 28 Barb 116, 118 (NY Sup Ct 1858); and Carroll v Charter Oak Ins Co 38 Barb 402, 408–09 (NY Sup Ct 1862)); Texaco A/S SA v Commercial Ins Co No 90 CIV 2722 (JFK), 1995 US Dist Lexis 15818 (SDNY 26 October 1995); SR Int’l Bus Ins Co v World Trade Ctr Props LLC 375 F Supp 2d 238, 246 (SDNY 2005); Globecon Group LLC v Hartford Fire Ins Co 434 F3d 165 (2d Cir 2006). See also the discussion in Globecon Group LLC v Hartford Fire Ins Co No 03-civ-0023, 2004 US Dist Lexis 13130, at *15–*16 (SDNY 13 July 2004), aff ’d in part, vacated in part, remanded, 434 F3d 165 (2d Cir 2006). But see Renfrew Ctr v Blue Cross & Blue Shield of Central NY Inc No 94 Civ 1527 (RSP/GJD), 1997 US Dist Lexis 5088 (NDNY 1997) (upholding an anti-assignment clause in the context of health insurance, where the prohibition included the assignment of ‘your right to collect money’ for the services provided by the contract). 81 434 F3d 165 (2d Cir 2006). See also Greenhomes Am LLC v Farm Family Cas Ins Co 936 NYS 2d 829 (App Div 2012); Arrowood Indem Co v Atlantic Mut Ins Co 948 NYS 2d 581 (App Div 2012). 82 434 F3d at 170. 83 Elat Inc v Aetna Cas & Sur Co 654 A2d 503 (NJ Super Ct App Div 1995). 84 Ibid at 505–06; see also Northern Ins Co of NY v Allied Mut Ins Co 955 F 2d 1353, 1358 (9th Cir 1992).

242

The Conditions

11.67

As far as successor liability is concerned, most jurisdictions in the United States recognise four theories of imposing successor liability: express assignment; express or de facto merger; mere continuation; and fraud (that is, where the transfer of assets is for the fraudulent purpose of escaping liability for the seller’s debts). When a successor is held liable for its predecessor’s acts under these theories, the successor often seeks to access the predecessor’s insurance benefits to cover the acquired liabilities. Some courts in the United States have allowed coverage in these situations, and others have not.

11.68

Most state merger statutes expressly provide that a surviving corporation succeeds to all of the rights and liabilities of the merged corporation. For example, when a surviving corporation by merger sought coverage for a claim alleging injuries from a product manufactured by the merged entity, a US court found the insurer liable to provide coverage to the surviving company.85 According to the court: ‘Upon merger, these [insurance] rights automatically vested in the surviving corporation by operation of the merger statute.’86 The court concluded that a non-assignment clause does not preclude coverage for liabilities imposed resulting from a statutory merger.

11.69

Stock transactions present an easier coverage case because the legal identity of the policyholder remains unchanged. Liabilities from an asset sale raise more difficult questions when the policyholder’s legal identity has changed between the time of the occurrence and the time that the claim for coverage is made. However, courts have allowed rights to insurance benefits to transfer to a successor by operation of law. For example, in Northern Insurance Co v Allied Mutual Insurance Co,87 the US Court of Appeals for the Ninth Circuit in 1992 applied the majority rule, holding under California law that, when a company acquires substantially all of the assets and liabilities of another entity, the acquiring company is entitled to coverage from the predecessor company’s insurance.88 Under this rationale, the insurance thus follows the risk.

11.70

The California Supreme Court’s rejection in 2003 of the majority rule on assignment had a wide-ranging impact on this issue across the United States, and thus deserves elaboration even under policies governed by the laws of other states. Eleven years after Northern Insurance v Allied Mutual, the California Supreme Court changed California law on the issue in what was considered a landmark (and controversial) case at the time, Henkel Corp v Hartford Accident & Indemnity Co.89 The court there found that Henkel’s liability had arisen under the asset acquisition agreements at issue in that case and not by operation of law.90 The court also took a restrictive view as to whether the assignment

85 Brunswick Corp v St Paul Fire & Marine Ins Co 509 F Supp 750 (ED Pa 1981). 86 Ibid at 752–53. 87 Northern Ins Co v Allied Mut Ins Co 955 F2d 1353 (9th Cir 1992). Northern v Allied was decided more than a decade before Henkel and, given its application of a California rule now changed by the California Supreme Court in Henkel (discussed in the main text), warrants close review. It still stands as an example of the majority rule that, notwithstanding Henkel, applies in jurisdictions other than California. 88 Ibid at 1358. 89 Henkel Corp v Hartford Accident & Indem Co 62 P3d 69 (Cal 2003). 90 Ibid at 73–74. Because the contracts there contained no specific provision addressing ownership of the predecessor company’s insurance, the court concluded that Henkel had acquired all of the liabilities of the predecessor but none of its insurance (ibid at 74).

The Assignment Condition

243

had taken place prior to loss, holding that the claims ‘had not become an assignable chose in action. Those claims had not been reduced to a sum of money due or to become due under the policy’ at the time of the assignment.91 As the dissenting judge pointed out, however, the Henkel majority did not reconcile this conclusion with the formerly well-settled law, holding that coverage under occurrence policies is ‘triggered by damage or injury occurring during the policy period’ and not by the assertion of a claim or its reduction to a sum certain.92 While it was cited in assignment cases and caused consideration of how ownership of insurance was documented in corporate transactions, Henkel ultimately was not widely followed, and in 2015 was explicitly overruled by Fluor Corp v Superior Court of Orange County.93 The California Supreme Court in Fluor found that section 520 of the California Insurance Code, a state statute ‘tracing back to 1872’94 that had not been analysed in Henkel, ‘dictates a result different from that in Henkel’ and bars an insurer from refusing to honour a policyholder’s assignment of coverage for injuries that pre-date the assignment.95

11.71

Fluor’s liability insurers agreed to defend and indemnify the company in numerous asbestos bodily injury lawsuits. In later coverage litigation, one of the insurers, Hartford Accident & Indemnity Company (Hartford), sought to avoid its coverage obligations, arguing that a reverse spinoff separating Fluor into two publicly traded companies constituted a purported assignment of insurance rights that was completed without Hartford’s consent, rendering the assignment void. The lower courts agreed, rejecting Fluor’s arguments under section 520 on the ground that Henkel had definitely resolved the assignment issue against the policyholder’s position.96 The California Supreme Court, however, in an extensively reasoned and unanimous opinion written by the court’s Chief Justice, overruled Henkel, noting that the court in Henkel had not considered the ‘wealth of judicial authorities … bearing on the proper interpretation of section 520’, ‘the subsequent common law decisions of other courts, virtually all of which are at odds with our key holding in Henkel’ and the fact that Henkel ‘has not been well received’ or ‘fared better in scholarly publications or practice guides’.97 The court found the statute ambiguous with regard to whether it applied to liability policies and, after thoroughly examining the legislative history and case law from numerous jurisdictions, concluded that the statute does not require a money judgment or approved settlement before ‘a claim concerning that loss may be assigned without the insurer’s consent’.98 Thus, California now follows the majority rule, and there is no dispute that ‘after personal injury (or property damage) resulting in loss occurs within the time limits of the policy, an insurer is precluded from

11.72

91 Ibid at 75. 92 Ibid at 77–78. 93 Fluor Corp v Superior Ct of Orange Cty 354 P3d 302 (Cal 2015) (Cantil-Sakauye, CJ). 94 Ibid at 303–04. 95 Ibid. 96 The relevant language of s 520 provides that an agreement not to transfer a claim of the policyholder against its insurer ‘after a loss has happened, is void if made before the loss’. Ibid at 304. 97 Ibid at 332 fn 52. 98 Ibid at 330.

244

The Conditions

refusing to honor an insured’s assignment of the right to invoke defense or indemnification coverage regarding that loss’.99 There is also no dispute that, under California law, ‘[t]his result obtains even without consent by the insurer – and even though the dollar amount of the loss remains unknown or undetermined until established later by a judgment or approved settlement’.100 11.73

In Globecon Group LLC v Hartford Fire Insurance Co, the US Court of Appeals for the Second Circuit quoted with approval a short passage from the dissenting judgment in Henkel.101 As a Delaware court recognised in 2009 in Viking Pump, Inc v Century Indemnity Co,102 ‘New York law on this matter is in accord with the dissent in Henkel’ rather than the majority opinion.103 Years before the California Supreme Court overturned Henkel, the Delaware court concluded that the Henkel requirement that the loss be ‘reduced to a fixed amount at the time of the assignment’ did not reflect New York case law, and further found that Henkel ‘was at odds with New York’s public policy, because it could hamstring markets for the sale of corporate assets and lead to insufficient recoveries for tort plaintiffs in situations when insurance to cover the plaintiffs’ claims was bought and paid for’.104

THE CANCELLATION CONDITION 11.74

Article VI(L)(1) of the 004 Form lists four circumstances in which Coverage A105 can be cancelled ‘on a pro rata basis’.106

11.75

First, either the Named Insured or the insurance company can cancel ‘at the end of any Annual Period’ by delivering prior written notice to the other. It should be noted, however, that if Coverage A is in force, it will continue only if the parties mutually agree on terms for its extension at the end of each Annual Period (see Article VI(Q)).107 Accordingly, the right of cancellation conferred under the first paragraph of Article VI(L) would appear to add nothing to the existing legal position.

11.76

Second, the Named Insured can cancel at any stage by giving written notice. In these circumstances, a pro rata premium would be refundable.

11.77

Third, the insurer can give 90 days’ written notice to cancel, but this extends only to the cancellation of Coverage A. The policyholder has a right to purchase and continue Coverage B. This right survives termination of Coverage A other than by reason of

99 Ibid at 334. 100 Ibid. 101 434 F3d at 171. 102 Viking Pump, Inc v Century Indem Co 2 P3d 76, 105 (Del Ch 2009) (rejecting insurers’ arguments that court should adopt rulings in Henkel in a case governed by New York law). 103 Ibid at 105. 104 Ibid at 105–06. 105 The equivalent provision (Article V(m)) in earlier versions of the Bermuda Form did not distinguish between Coverage A and Coverage B. 106 Article VI(L) of the 004 Form. See also Article V(m) in earlier Forms. 107 Article VI(Q) of the 004 Form; and Article V(s) in earlier Forms.

The Currency Condition

245

non-payment of premium or because of the commencement of proceedings other than arbitration proceedings.108 Fourth, the insurer can cancel upon five days’ notice if ‘any Insured shall institute a suit or proceeding against the Company other than as provided in Condition N below (or to enforce an award arising out of such arbitration)’. This provision, which is new to the 004 Form, prescribes a draconian consequence of possible cancellation for a policyholder who decides to ignore the arbitration provisions of the Bermuda Form. Other provisions of the 004 Form make it clear that breach of the arbitration clause can result in the cancellation of the coverage under Coverage B as well as Coverage A.109

11.78

Article VI(L)(2) is also new to the 004 Form and provides for the automatic cancellation of the insurance policy if the premium or proof of payment thereof is not received within five days of the commencement of an annual period.110 Non-payment results in the cancellation of ‘the policy’, and therefore potentially affects Coverage B,111 as well as Coverage A. Cancellation for non-payment of premium is retroactive to the commencement of the Annual Period for which payment should have been made. Article VI(Q) spells out the consequences in the event that notice of occurrence or integrated occurrence has been given prior to cancellation or non-extension.

11.79

THE CURRENCY CONDITION The Currency Condition provides: The premiums and losses under this Policy are payable in the respective currency(ies) set forth in Item 6 of the Declarations. Unless otherwise specified in Item 6, such currency(ies) shall be United States dollars. If judgment is rendered, settlement is denominated or another element of Damages is stated in a currency other than in the applicable currency, payment under this Policy shall be made in the applicable currency at the rate of exchange prevailing on the date the final judgment is rendered, the amount of the settlement is agreed upon or the other element of Damages is due, respectively.112

This clause deals with two related matters. First, the currency of payment under the Bermuda Form is US dollars, unless the parties have otherwise agreed. Second, the clause provides a conversion rate113 for sums paid by the policyholder in currencies other than US dollars (or the other applicable currency if the parties have so agreed). In making its award, an arbitral tribunal should give effect to these contractual provisions. In Lesotho Highlands Development Authority v Impregilo SPA, the majority of the House of Lords held that section 48(4) of the Arbitration Act did not give a tribunal a broad

108 See Article VI(L)(3) and also Article VI(S) (the Discovery Period Condition), discussed below, paras 11.102–11.109. 109 Ibid. 110 See the Premium Condition, Article VI(A), discussed above, paras 11.02–11.07. 111 See also Article VI(S)(1). 112 Article VI(M) of the 004 Form. See also Article V(n) of earlier Forms. 113 In earlier versions of the Bermuda Form, this conversion rate appeared in the Loss Payable Condition.

11.80

246

The Conditions

discretion to award payment in any currency it thought fit, regardless of the terms of the contract.114 In the unlikely scenario of an arbitral tribunal failing to give effect to these provisions, it would not be possible to appeal against the tribunal’s decision. This is because it would amount to an error of law rather than a case where the tribunal had exceeded its powers.115

THE ARBITRATION CONDITION 11.81

The Arbitration Condition116 is the longest clause in the Conditions section and, indeed, in the entire Bermuda Form. It is a fundamental feature of the Bermuda Form aimed at keeping Bermuda Form insurers outside court systems, in particular the court system in the United States.117 The impact of the Arbitration Condition in a number of different contexts is discussed elsewhere in this book: Chapter 3 considers the legal consequences of New York law applied in a London arbitration; Chapter 14 considers issues relating to the commencement of arbitration, including the appointment of arbitrators; Chapter 15 discusses practical considerations relating to the course and conduct of a London arbitration; and Chapter 16 discusses issues of discovery and privilege. Some specific aspects of the Arbitration Condition should be noted here.

11.82

First, the Arbitration Condition gives the arbitral tribunal the power to prescribe reasonable rules and regulations governing the course and conduct of the arbitration proceeding including without limitation discovery by the parties.118

The arbitrators have such powers in any event as a result of section 34 of the English Arbitration Act 1996. The words ‘including without limitation discovery by the parties’ were added in the 004 Form, but they do not require the tribunal to order discovery. This is a matter for the tribunal’s discretion, although, in Bermuda Form arbitrations, which typically involve very substantial sums, discovery of documents will usually be ordered.119 11.83

Second, Article VI(N)(3) requires the arbitrators to publish their award within 90 days following the conclusion of the final hearing or trial. A provision of this kind is unusual.120 An extension of time can be agreed by the parties or can be granted by the court.121

114 [2005] UKHL 43, [2006] 1 AC 221. 115 Ibid, reversing the decision of the Court of Appeal in Lesotho Highlands Development Authority v Impregilo SpA and Others [2003] EWCA Civ 1159, [2003] 2 Lloyd’s Rep 497. An appeal for error of law is not possible under the 004 Form: see further ch 3, para 3.16 above and para 11.84 below. 116 Article VI(N) of the 004 Form; and Article V(o) of earlier Forms. 117 It is therefore somewhat ironic that court proceedings in the United States were commenced by the reinsurers in Noble Assurance Co v Gerling-Konzern General Insurance Co [2007] EWHC 253 (Comm), [2007] 1 CLC 85 and the insurers in C v D [2007] EWCA Civ 1282, [2008] 1 Lloyd’s Rep 239. 118 Article VI(N)(2) of the 004 Form. 119 See generally ch 16 below. 120 Michael J Mustill and Stewart C Boyd, Commercial Arbitration 2001 Companion (London, Butterworths, 2001) 334. 121 English Arbitration Act 1996, s 50.

The Arbitration Condition

247

The Arbitration Condition also makes the award:

11.84

a complete defense to any attempted appeal or litigation of such decision in the absence of fraud or collusion. Without limiting the foregoing, the parties waive any right to appeal to, and/or seek collateral review of the decision of the Board of Arbitration by, any court or other body to the fullest extent permitted by applicable law.122

In practice, there can be no appeal in England against the merits of the arbitrators’ decision, as appeals of arbitration awards can be made only for an error of English law (and even then in closely prescribed circumstances). Because a Bermuda Form policy is governed by New York law, an English court will not entertain any appeal.123 The provision waiving rights of appeal is in any event effective to preclude an appeal.124 The Arbitration Condition also purports to limit court intervention to cases of fraud or collusion, but the powers of the English court are in fact wider. If a tribunal exceeds its substantive jurisdiction, the English court can intervene under section 67 of the English Arbitration Act 1996. If, during the arbitral process, there is a ‘serious procedural irregularity’ falling short of fraud or collusion, then the English court has power to intervene under section 68 of the English Arbitration Act 1996. Both sections 67 and 68 are mandatory sections that cannot be overridden by the agreement of the parties.125 ‘Serious procedural irregularity’ is, however, quite narrowly defined and includes, for example, the failure by the tribunal to conduct the proceedings in accordance with the procedure agreed by the parties and a failure by the tribunal to deal with all the issues that were put to it. It does not appear that, in practice, Bermuda Form arbitral awards are usually the subject of challenge in the English courts, in part we believe because of the high quality of the arbitrators who are generally appointed. An English court will restrain by injunction any attempt to challenge an award elsewhere.126

11.85

Third, Article VI(N)(4) of the Arbitration Condition provides that the ‘costs of the arbitration shall be in the sole discretion of the Board, who may direct to whom and by whom and in what manner they shall be paid’. In London arbitrations, the general rule is that the arbitrators will order the unsuccessful party to bear the costs both of the successful party and of the arbitrators (including expenses, such as hiring the hearing room and transcript writers). The tribunal does, however, have a discretion and might order differently if, for example, substantial time and expense have been devoted to an issue on which the otherwise successful party has lost.127 The parties can, if they wish, make an agreement as to how the costs of the arbitration are to be dealt with, as discussed in the next paragraph.

11.86

122 Article VI(N)(3) of the 004 Form; and Article V(o) of earlier Forms. 123 See ch 3, in particular para 3.16 above. 124 Section 69(1) of the English Arbitration Act 1996 allows only for the possibility of an appeal ‘unless otherwise agreed by the parties’. 125 English Arbitration Act 1996, s 4 and Sch 1; Mustill and Boyd, Commercial Arbitration 2001 Companion (above n 120) at 351 and 354. See also ibid at 440–41, setting out paras 276 and 283 of the Report on The Arbitration Bill of the Departmental Advisory Committee on Arbitration Law. 126 C v D [2007] EWCA Civ 1282, [2008] 1 Lloyd’s Rep 239; Atlas Power Ltd. v National Transmission and Despatch Co Ltd [2018] EWHC 1052 (Comm), [2019] 1 All ER (Comm) 931. 127 The topic of costs is discussed in further detail in ch 17, paras 17.27–17.41 below.

248

The Conditions

11.87

The earlier versions of the Bermuda Form provided that the costs of the arbitration ‘shall be borne equally by the parties to such arbitration’. A provision of this kind is ineffective under English law. Indeed, it was the existing practice of insurance and other companies to include a costs-sharing provision as a standard form policy term, which led to legislation rendering such clauses ineffective.128 The current governing statutory provision is section 60 of the English Arbitration Act 1996. If the parties wish to make an effective agreement as to how the costs of the arbitration should be borne, they must do so after the dispute has arisen. The parties may wish to consider including a costssharing provision – that is, an agreement that overrides the ‘English Rule’ on costs, and provides for each party to bear its own costs and for the tribunal’s costs to be split equally – in the first procedural directions order which the tribunal enters.129 This may be a sensible agreement to make at the outset of a case, as it is a way of limiting the downside of a defeat in the arbitration, although it also of course limits the upside of a victory. If a party does not wish to make such an agreement, it should ensure that any agreed order for directions, or terms of appointment for the tribunal, does not set out the terms of an arbitration clause containing a costs-sharing provision. Otherwise, the order for directions or terms of appointment may be construed as a costs-sharing agreement concluded subsequent to the dispute.

11.88

Finally, the concluding paragraph (5) of the Arbitration Condition is one of a number of provisions of the Bermuda Form that seek to prevent the Bermuda Form insurer from becoming involved in court proceedings.130 It deals with a situation where a claim for contribution is made by another insurer. Such a claim could require a court to interpret the Bermuda Form to decide the validity of a claim for contribution. As already indicated, litigation in a US court is something that Bermuda Form insurers have sought to avoid.131 Accordingly, the Bermuda Form states that the policyholder is required

128 Mustill and Boyd, Commercial Arbitration 2001 Companion (above n 120) at 344. 129 The English Rule is to be contrasted with the ‘American Rule’ typically applied under the common law in the United States. See, eg, Campagnola v Mulholland, 556 NYS2d 239, 243 (NY 1990) (citing Alyeska Pipeline Co v Wilderness Soc’y 412 US 240, 248–52 (1975) (explaining the history of the ‘American Rule’, under which federal courts in the United States are not to award attorneys’ fees in the absence of statute expressly authorising such an award). See also Kansas v Colorado 556 US 98, 102 (2009). The New York Court of Appeals affirmed the general applicability of the American Rule except when a ‘statute, agreement or court rule’ provides to the contrary (in effect allowing the prevailing party to recover as under the ‘English Rule’). The court noted an exception, however, where a policyholder is put in a ‘defensive posture’ when its liability insurance company, which owes a duty to defend, sues to defeat coverage: It is well settled in New York that a prevailing party may not recover attorneys’ fees from the losing party except where authorized by statute, agreement or court rule. [Citations omitted] However, an insured who is ‘cast in a defensive posture by the legal steps an insurer takes in an effort to free itself from its policy obligations’, and who prevails on the merits, may recover attorneys’ fees incurred in defending against the insurer’s action. US Underwriters Ins Co v City Club Hotel LLC, 789 NYS2d 470, 473 (NY 2004). The applicability of this case may be limited under a Bermuda Form policy which includes no duty to defend, but only a duty to pay defence costs, and under the choice of law clause in the policy which applies English procedural law. 130 Article VI(N)(5) of the 004 Form; and Article V(o) of earlier Forms. See ch 9, paras 9.28–9.33 above for discussion of the ‘Other Insurance’ Condition. 131 Even before Bermuda insurers bought American insurance groups, courts in the United States had upheld jurisdiction over Bermuda insurers under states’ ‘long-arm jurisdiction’ in cases where insurers with no arbitration provisions in their policies brought the Bermuda insurers in as third-party defendants, arguing that if

The Conflicting Statutes Condition

249

to take steps to obtain a dismissal of claims for contribution or indemnity made in a US court. This provision states that the policyholder’s obligation includes an obligation by the policyholder to reduce a judgment or award against the insurers who are pursuing the contribution claim.132

THE CONFLICTING STATUTES CONDITION The Conflicting Statutes Condition appeared in the earlier versions of the Bermuda Form in the following terms:

11.89

In the event that any provision of this Policy is unenforceable by the Insured under the laws of any State or other jurisdiction wherein it is claimed that the Insured is liable for any injury covered hereby, because of non-compliance with any statute therein, then this Policy shall be enforceable by the Insured with the same effect as if it complied with such statute.133

The clause does not appear in the 004 Form, perhaps because the reason for the original inclusion of this provision, which appears to operate in favour of the policyholder, is obscure. If a need arises to enforce the insurance policy, as a result of (presumably) a dispute between the policyholder and the insurer, the relevant proceedings will take place in England in arbitration proceedings, not in the state or other jurisdiction where the underlying claim (or claims) against the policyholder is made (unless by coincidence this happens to be England). In an English arbitration, the tribunal would be likely to apply New York law to issues addressed to the validity of the insurance policy.134 It is therefore difficult to see how a situation would arise in which the policyholder was prevented from enforcing the insurance contract as a result of some provision of the law of the state or jurisdiction where the underlying claim against the policyholder was being made (unless this also happens to be New York). Accordingly, although Condition (p) appears to confer some benefit on the policyholder, we find it difficult to see much actual benefit in practice.

they were held liable to the policyholder, they had a right to seek contribution from the third-party defendant insurers. See, eg, Hoechst Celanese Corp v National Union Fire Ins Co of Pittsburgh, Pa v Centaur Ins Co No 89C-SE-35, 1991 WL 190313 (Del Super Ct 10 Sept 1991) (denying motion to dismiss a third-party complaint of National Union against an offshore insurer, finding jurisdiction under the Delaware long-arm statute on the ground that the offshore insurer had a reasonable expectation of being haled into court in Delaware and that the exercise of personal jurisdiction would not offend notions of due process), denying motion for interlocutory appeal sub nom Tortuga Cas Co v Nat’l Union Fire Ins Co of Pittsburgh, Pa 604 A2d 419 (Table) (Del 1991) (Unpublished Decision). 132 Article VI(N)(5) of the 004 Form. See also the last paragraph of Article V(o) of earlier Forms. 133 Article V(p) of the 001–003 Forms. 134 See generally, ch 3 above; Rome Convention, Article 8; Rome I Regulation, Article 11. See also Lord Collins et al (eds), Dicey, Morris & Collins: The Conflict of Laws, 15th edn (London, Sweet & Maxwell, 2012) Rules 225 and 226. The rules in the Rome Convention and the Rome I Regulation have (subject to some amendments) been retained as part of English law notwithstanding the UK’s departure from the EU: see the Law Applicable to Contractual Obligations and Non-contractual Obligations (Amendment etc) (EU Exit) Regulations 2019 (SI 2019/834). For a short summary of the position after Brexit, see Civil Procedure (generally known as the ‘White Book’) (London, Sweet & Maxwell, 2021) vol 1, para 6HJ.18.

11.90

250

The Conditions

THE LAW OF CONSTRUCTION AND INTERPRETATION CONDITION 11.91

This condition, under which the governing law of the contract is a modified form of New York law, is a key provision of the Bermuda Form.135 Its impact is discussed in Chapters 3, 4 and (in the context of exclusions) 10 above.

THE PRORATION OF LOSSES CONDITION 11.92

The Proration of Losses Condition appeared only in the original (001) Bermuda Form and provided: In the event that more than one insured, whether under this Policy or any similar such policy, of the Company incur losses due to one occurrence, or to multiple occurrences arising out of substantially similar events or conditions which give rise to substantially similar types of loss, injury, damage or liability, all insureds potentially involved shall be notified by the Company as soon as any one such insured reports such a claim to the Company and the Company is aware of such potential involvement. If an insured is known to have similar exposure to loss as determined by the Board of Directors of the Company, that insured will be instructed by the Company to file a claim as soon as practicable. The maximum amount payable to all insureds including any such insured hereunder combined shall be either (1) the total net worth (exclusive of such claims) of the Company at April 30, 1987 if the first of such claims is reported at or prior to such date or (2) subject to a minimum of $75 million, the lesser of the sum of the occurrence limits of liability of the involved named insureds under any of the Company’s policies then in existence or the total net worth of the Company at the end of the fiscal year ending prior to the reporting of the first of such claims, if such reporting occurs after April 30, 1987. The actual amount allocated to each Named Insured shall be determined by the Board of Directors of the Company taking into account such factors as they deem relevant, including, by way of illustration only, each insured’s underlying limit and exposure to loss and the number of claimants. Among other items it is expected that the allocation will be adjusted between the time the claims are reported and actually paid as more information becomes available. Disputes shall be settled by arbitration in accordance with Section (o) above. This Section (r) shall not apply if and to the extent the Company shall have obtained ‘clash’ reinsurance (as such term is commonly understood in the insurance industry) and such coverage applies to amounts payable to insureds in respect of the occurrence(s) contemplated herein.136

The clause reduced the overall exposure of the insurer to policyholders affected by a common catastrophe. Accordingly, a single, combined limit of liability was imposed as a maximum amount payable to all policyholders involved in the common catastrophe. The clause is of historical interest only, reflecting the uncertainty that existed when insurance companies like ACE and XL drafted the Bermuda Form and began underwriting risks in the mid-1980s. When introducing the 002 Form, XL advised its policyholders that

135 Article 136 Article

VI(O) of the 004 Form; and Article V(q) of earlier Forms. V(r) of the 001 Form.

The Policy Extension Condition

251

this Condition was ‘no longer required due to financial maturity of XL’.137 At the same time, the 002 Form deleted a condition that appeared in the 001 Form,138 requiring the Named Insured to purchase stock in XL.

THE LIABILITY OF THE COMPANY CONDITION The Liability of the Company Condition provides:

11.93

The Named Insured and the Insured agree that the liability and obligations of the Company hereunder shall be satisfied from the funds of the Company alone and that the individual shareholders of the Company shall have no liability hereunder to the Named Insured or the Insured.139

This condition aims to preclude imaginative claims seeking to pierce the corporate veil. The clause provides that the insurer alone, and not the insurer’s shareholders, will pay any claims. No doubt this was considered an important provision when the Bermuda Form was first drafted, because the shareholders in ACE and XL were large US companies.

THE POLICY EXTENSION CONDITION140 The Policy Extension Condition should be considered together with the Cancellation Condition,141 the Reinstatement Condition,142 the Discovery Period Condition143 and the overall structure of the Bermuda Form discussed in Chapter 2 above. In summary, the Bermuda Form contemplates that the coverage will continue from year to year, although the parties need to agree upon the precise terms of each extension each year. In the absence of agreement, the contract will terminate.144 If agreement is reached, there is an extension of the existing policy with fresh limits, rather than a ‘renewal’ in the sense of a new policy being issued for a further term. Nevertheless, the annual meetings held between insurer and policyholder to discuss extension are commonly called ‘renewal meetings’. Bermuda Form policies issued to the same policyholder often use the same policy number year

137 Letter from XL Insurance Company, Ltd, to its Policyholders dated 29 September 1988 and summarising the policy modifications. XL sent similar letters (summarising policy modifications) to its policyholders in other years, including 1987, 1990 and 1995. A similar clause contained in ACE’s Forms was removed when ACE introduced its 004 Form in 1991. 138 Article V(y) of the 001 Form. 139 Article VI(P) of the 004 Form. See too Article V(x) of the 001 Form; and Article V(r) of the 002 and 003 Forms. 140 Article VI(Q) of the 004 Form. See too Article V(s) of earlier Forms. 141 Article VI(L) of the 004 Form; and Article V(m) of earlier Forms. 142 Article VI(R) of the 004 Form; and Article V(t) of earlier Forms. 143 Article VI(S) of the 004 Form; and Article V(u) of earlier Forms. 144 See the Cancellation Condition: Article VI(L) of the 004 Form; and Article V(m) of earlier Forms. Article V(v) of earlier Forms also contained an ‘Expiration Date’ Condition, but this Condition does not appear in the 004 Form, no doubt because it is surplusage.

11.94

252

The Conditions

after year, with a two-digit extension reflecting the year in which the policy incepted. This practice may indicate whether the policy was, as some Bermuda Form policies provide, a ‘continuous renewal’.145 11.95

As with a number of provisions of the Bermuda Form, continuation of cover is dependent upon mutual agreement between policyholder and insurer on mutually acceptable terms. A policyholder might be able to complain of a breach of the requirement of good faith and fair dealing in the event that the insurer’s proposed terms for a continuation were wholly unreasonable. The resolution of such a claim would depend upon the facts and probably would succeed only in rare circumstances.

11.96

The Policy Extension Condition now spells out the consequence of cancellation in relation to occurrences or integrated occurrences of which the policyholder has given notice prior to cancellation: Where Coverage A (or Coverage B) is cancelled or not extended, such cancellation or nonextension shall not affect the rights of the Insured as respects any Occurrence or Integrated Occurrence of which notice was given in accordance with the provisions of this Policy prior to such cancellation or non-extension and shall not limit whatever rights the Insured otherwise would have under this Policy as respects actual or alleged Personal Injury, Property Damage or Advertising Liability included in such Occurrence or Integrated Occurrence taking place subsequent to such cancellation or non-extension.146

THE REINSTATEMENT CONDITION 11.97

The Reinstatement Condition in the 004 Form is set out in two numbered paragraphs. The first, Article VI(R)(1), is relatively brief and is substantially the same as an equivalent clause in the 001 and 002 Forms.147 The second, Article VI(R)(2), sets forth a complicated regime enabling a policyholder to reinstate the aggregate limit of liability in the event of impairment of the limits.148 These two subparagraphs of Article VI(R) need to be considered separately.

11.98

Article VI(R)(1) of the 004 Form provides as follows: At the time of each annual Policy extension of Coverage A, the aggregate limit of liability set forth in Item 2 (b) of the Declarations shall, unless otherwise agreed in writing between the Named Insured and the Company, automatically be reinstated with respect to covered Occurrences of which notice is first given during the following Annual Period. There shall be no separate premium charged for this automatic reinstatement in addition to that provided for in Condition Q above. There shall be no reinstatement of the aggregate limit of liability, unless otherwise agreed in writing by the Company, as respects Coverage B, and the remaining amount, if any, of the aggregate limit for the final Annual Period under Coverage A shall apply as respects the Discovery Period.149

145 See,

eg, XL Employment Practices Liability Form XLEPL005. VI(Q) of the 004 Form. 147 Article V(t) of earlier Forms. 148 This regime also appears in the 003 Form. 149 Article VI(R) of the 004 Form; and Article V(t) of earlier Forms. 146 Article

The Reinstatement Condition

253

The clause thus provides for the automatic reinstatement of the aggregate limit each time that the policy is extended for an Annual Period, and that no separate premium shall be charged for this automatic reinstatement. The clause perhaps promises more than it delivers in the sense that each annual extension is subject to agreement on the terms and conditions of the extension: see the Policy Extension Condition,150 as well as Article VI(R)(1) which applies ‘unless otherwise agreed in writing’ between the policyholder and the insurance company. Accordingly, the Bermuda Form does not really confer an automatic right to a reinstatement of the aggregate limit. All will depend upon whether the policyholder and the insurer can agree upon mutually acceptable terms during the renewal discussions. The clause will, however, form the background to the negotiations, so that a policyholder can reasonably expect that, if the cover is extended, the aggregate limit will be automatically reinstated. Aggregate limits have been discussed in Chapters 2 and 9 above. In short, the aggregate limit is refreshed on each extension, but only with respect to occurrences first reported in the new Annual Period that is about to commence. Accordingly, if the policyholder has a US$50 million aggregate limit in year 1 and reports a number of occurrences in year 1, those year 1 occurrences will all be subject to that US$50 million annual aggregate limit. If the policy is extended for year 2, with an unchanged aggregate limit of US$50 million, the new aggregate limit will only apply to occurrences reported in year 2. It is not unusual for the parties to change the aggregate limit from one year to another, but the same principles will apply.

11.99

If the parties do not agree to extend Coverage A, but the policyholder elects to buy Coverage B, there is (unless the parties agree otherwise) no reinstatement available in the discovery period. Accordingly, the aggregate limit agreed for the last year of Coverage A will apply both to that last year and also to the whole of the Discovery Period.

11.100

The regime thus described can be modified to some extent, at the election of the policyholder, under Article VI(R)(2). The clause is lengthy and appears complex, but its broad effect is to enable the policyholder to buy an increased aggregate limit during an Annual Period. The policyholder may wish to do this if it appears that the existing aggregate limit is likely to be impaired by an occurrence of which notice has been given to the insurance company. The clause contains provisions that prevent the policyholder from using the increased limits to cover occurrences that are known at the time of the election to purchase the reinstatement cover.151 Accordingly, the clause is aimed at providing increased protection to a policyholder who has given notice of one or more occurrences and wants to purchase more cover in order to guard against the unknown. The reinstatement premium is in the discretion of the insurance company, but cannot exceed 125 per cent of the premium for the Annual Period in which the reinstatement takes place.

11.101

150 Article VI(Q) of the 004 Form, discussed at paras 11.94–11.96 above. 151 See the similar issue which arises when a policyholder purchases increased limits, as discussed in ch 8, in particular para 8.35 above.

254

The Conditions

THE DISCOVERY PERIOD CONDITION 11.102

This clause is part of the basic structure of the Bermuda Form152 and allows a policyholder to elect to obtain coverage (Coverage B) during an extended ‘Discovery Period’. ‘Discovery Period’ is a defined term,153 referring to the period commencing upon the expiration of Coverage A and ending when the policyholder decides not to continue Coverage B or the insurance policy is otherwise cancelled. The first paragraph of Article VI(S) sets out the policyholder’s basic entitlement to purchase Coverage B: In the event of Termination of Coverage A, other than by reason of cancellation for nonpayment of premium or due to institution of a proceeding other than as contemplated by Condition N, the Named Insured may elect, prior to the Termination Date of such Coverage A, to secure Coverage B for the following Annual Period for such Insureds as the Named Insured shall designate, by giving the Company written notice of such election and paying to the Company the annual premium set forth in the attached Schedule D no later than the date of the commencement of such Annual Period.154

11.103

A policyholder may wish to purchase Coverage B in the following situation. The policyholder may have purchased Coverage A for five years. That coverage would provide cover for occurrences that took place during that period, and that were also reported during that period.155 If the policyholder then decided to stop buying Coverage A from its existing insurers and to buy it from new insurers, it would face a potential exposure to claims arising from occurrences that have already taken place during the five-year period, but that have not yet been reported.

11.104

In this situation, the policyholder may have a choice. The new insurers may be prepared to provide cover for occurrences that have already taken place, but have not yet been reported. Such coverage, for what is typically known as ‘IBNR’ claims (that is, claims ‘incurred but not reported’), is the essence of the ‘retroactive coverage’ that insurers writing on the Bermuda Form are sometimes prepared to write. The policyholder may therefore choose to protect its exposed ‘tail’, or exposure to possible liability, by purchasing retroactive coverage from its new insurers. Alternatively, it may decide to purchase that protection by electing for the Discovery Period coverage that the existing Bermuda Form insurers have provided under Article VI(S). Thus, Article VI(S) helps ensure that the policyholder is always covered (subject of course to the other terms of the Bermuda Form) for occurrences that take place during the period that Coverage A has been purchased, whether reported during that period or not.

11.105

To take advantage of the extended Discovery Period, however, the policyholder must elect to purchase it prior to the termination of Coverage A. Accordingly, if Coverage A is about to expire, the policyholder will need to make that election before the expiration of Coverage A. Similarly, if the policyholder has elected to have Coverage B for a year,

152 See

above ch 2. Article III(J) of the 004 Form; and Article III(m) or (o) in earlier Forms. 154 Article VI(S) of the 004 Form. See also Article V(u) of earlier Forms. 155 See above chs 2 and 6. 153 See

The Discovery Period Condition

255

then if it wishes to extend Coverage B for a further year or years, it will have to make this election before Coverage B expires. Thus, Article VI(S)(2) provides: In the event that the Named Insured elects to secure Coverage B pursuant to paragraph (1) above, the Named Insured may elect to continue such Coverage B for any number of additional Annual Periods by giving the Company written notice of each election for a subsequent Annual Period and paying to the Company the corresponding annual premium set forth in the attached Schedule D no later than the end of the Annual Period for which such Coverage B was previously elected. If the Named Insured shall fail to elect Coverage B for any Annual Period, it may not elect Coverage B for any subsequent Annual Period.

Accordingly, if Coverage B is discontinued, it cannot be revived at a later date. The policyholder must also pay the premium no later than the date when the previous 11.106 Annual Period starts and the new one commences. Failure to pay the premium in time results in the automatic cancellation of the policy.156 By making its election, the policyholder can keep the Discovery Period coverage in effect for as long as the policyholder wants it. The insurance policy contains a built-in sliding scale premium, which reduces as the years go by. Schedule D provides that the annual premium for the first year of Coverage B is 30 per cent of the premium for the last year of Coverage A. By the fifth and additional years of Coverage B, the premium is only 9 per cent. This reflects the fact that, the further the distance of time from the happening of the occurrence, the less likely it is that an occurrence likely to affect the policy will be reported and therefore threaten the insurers. The policyholder will therefore have to make an economic choice between bearing the risk of delayed reporting of occurrences itself, purchasing retroactive coverage from its new insurers, or electing to continue with Coverage B from its existing Bermuda Form insurers.

11.107

Article VI(S) also contemplates that, when the policyholder first elects to purchase Coverage B, the policyholder will decide to continue coverage, but only in relation to particular policyholders. In this situation, the policyholder might expect to pay a lower premium for Coverage B than a straightforward percentage of the premium payable for the last period of Coverage A; that is, to reflect the fact that the number of insured companies protected during Coverage B is lower than the number protected during Coverage A. The Bermuda Form does not, however, provide for any calculation of a reduced premium. In many cases, no doubt, this will be the subject of negotiations with the insurer. If not, however, it would appear that the policyholder must pay the full percentage stipulated in Schedule D of its policy.

11.108

The final paragraph of Article VI(S) provides that:

11.109

For the purpose of application of retentions and limits of liability, notice of an Occurrence given during the Discovery Period shall be deemed to have been given during the final Annual

156 Article

VI(L)(2) of the 004 Form.

256

The Conditions

Period in the Policy Period. The aggregate limit of liability shall not be reinstated for the Discovery Period.157

Accordingly, the retentions and limits in place during the final year of Coverage A apply to occurrences reported in the Discovery Period. There is, however, no reinstatement of the aggregate limit. The aggregate limit in the final year of Coverage A therefore applies for that year as well as the whole of the Discovery Period. If part of that limit has already been used by claims reported in the final year of Coverage A, only the balance is available for occurrences reported during the Discovery Period.158

THE EXPIRATION DATE CONDITION 11.110

Bermuda Forms 001–003 contain Article V(v), an Expiration Date Condition, which provides: Except as otherwise provided in Section (w) of this Article V, Coverage A shall expire upon cancellation thereof or at the end of an annual period if not extended. Coverage B shall expire upon termination of the Discovery Period.

This provision duplicated other provisions and does not appear in the 004 Form.159

THE FORMER SUBSIDIARIES, AFFILIATES AND ASSOCIATED COMPANIES CONDITION 11.111

The Former Subsidiaries, Affiliates and Associated Companies Condition provides as follows: If any subsidiary, affiliate or associated company of the Named Insured which is an Insured hereunder shall cease to be such a subsidiary, affiliate or associated company of the Named Insured, then at such time Coverage A shall automatically terminate as to such former subsidiary, affiliate or associated company. Coverage A shall continue with respect to the Named Insured and any other entity which remains an Insured for its own liability, if any, arising out of its prior ownership of or affiliation or association with the former subsidiary, affiliate or associated company. At such time of such automatic termination of coverage, Coverage B shall, unless the Named Insured otherwise specifies, automatically incept as to such former subsidiary, affiliate or associated company and continue in force for the balance of the Annual Period, such former subsidiary, affiliate or associated company may, with written consent received by the Company from the Named Insured, elect to extend Coverage B beyond the end of the Annual Period on such terms and conditions, for such period, subject to such limits and for such additional premium as may be agreed with the Company.160

157 See also the final sentence of Article V(u) in earlier Forms. The position is perhaps the same under earlier Forms, but the wording of those Forms is less clear and detailed than that in the 004 Form. 158 See Article VI(R)(1) of the 004 Form. 159 See the Policy Extension Condition, discussed at paras 11.94–11.96 above. 160 Article VI(T) of the 004 Form. See too Article V(w) of earlier Forms.

The Notice Condition

257

Under this provision, if a subsidiary, affiliate or associated company of the Named Insured ceases to be such a subsidiary, affiliate or associated company – for example, through a sale or divestment – Coverage A automatically ceases with respect to that company. The former group company will automatically be granted Coverage B coverage – that is, for occurrences that take place during the currency of Coverage A, but are reported later – until the end of the Annual Period during which it ceased to be a subsidiary, affiliate or associated company. Before the end of that Annual Period, the former group company can elect to extend Coverage B, upon terms agreed with the insurer. As with a number of provisions of the Bermuda Form, where continuation of cover is to be the subject of agreement between policyholder and insurer, the issue could arise as to whether the terms offered by the insurer are consistent with the principle of good faith and fair dealing.

THE NOTICE CONDITION The Notice Condition is a crucial provision because a number of significant clauses in the Bermuda Form – for example, the notice of occurrence Article, the pollution exclusion, the Cancellation Condition and the Reinstatement Condition – depend upon notice. Notice is also crucial to the occurrence reported trigger of coverage used in the Bermuda Form.161 The Notice Condition provides:

11.112

All notices under any provision of this Policy shall be in writing and given by hand, prepaid express courier, airmail or telecopier properly addressed to the appropriate party and will be deemed as having been effected only upon actual receipt. Notice to any Insured may be given to the Named Insured at the address shown in Item 1(b) of the Declarations or to such other person as the Named Insured shall designate in Item 5 of the Declarations.162

Accordingly, the Condition provides not simply for the manner of notice, but (perhaps unusually) for the notice to be deemed effected ‘only upon actual receipt’. This provision appears to be directed at nullifying the rule of New York law that a properly mailed letter is presumed to have been received.163 If any issue arises as to this, it will be for the party seeking to prove receipt of the notice to do so by evidence, and proof that a notice was properly sent will be of merely evidential weight. Earlier versions of the Form deemed notice of occurrence to be received once the policyholder sent notice in accordance with the policy provisions.164 Policyholders are well advised to take some care of notice provisions, because honest mistakes can easily be made. For instance, policyholders sometimes assume that email correspondence with some particular employee of the insurance company, or 161 See the discussion of the occurrence reported trigger in the Bermuda Form in ch 2, paras 2.02–2.08 above, and generally ch 8 below. 162 Article VI(U) of the 004 Form; see also Article V(y) of the 003 Form, when the Notice Condition was first introduced. 163 See, eg, De Feo v Merchant 454 NYS2d 576 (City Ct 1982); New York Presbyterian Hosp v Allstate Ins Co 814 NYS2d 687 (App Div 2006). 164 Article V(d) of the 001–003 Forms (‘Notice shall be deemed to be received if sent by prepaid mail properly addressed to the address in the Declarations’).

11.113

258

The Conditions

correspondence with lawyers, constitutes good notice. But in its unmodified state, the Bermuda Form does not so provide. Unless the Form has been amended, both parties are well advised to ensure that the specified formalities (however antiquarian they may now appear) are scrupulously followed.

THE HEADINGS CONDITION 11.114

This is a standard (though irrational) provision, stating that headings are for convenience only and do not affect the application or interpretation of the policy.165

165 Article

VI(V) of the 004 Form; and Article V(x) or (z) of earlier Forms.

12 Misrepresentation and Non-disclosure INTRODUCTION1 In the early years of Bermuda Form disputes, it was not unusual for an insurer to raise issues of misrepresentation and non-disclosure. This may perhaps have been a consequence of the involvement of English lawyers and the prominent role that arguments as to avoidance played in English insurance litigation. Nowadays, however, disputes tend to relate to coverage issues, and it is rare for issues of misrepresentation and non-disclosure to be raised.

12.01

There are no doubt different reasons for this decline in enthusiasm for avoidance arguments. Purchasers of Bermuda Form policies will usually be well-run companies with a professional risk management department, as well as experienced brokers. Policyholders can therefore reasonably be expected to provide accurate answers to the questions raised on the insurers’ application forms. Also, avoidance of a Bermuda Form policy for misrepresentation under New York law is by no means easy. An insurer who seeks a remedy for misrepresentation would require convincing evidence in order to prove the matters which New York law requires. Avoidance of a Bermuda Form policy for non-disclosure is even harder. It would in practice require the insurer to prove dishonesty against the policyholder – an argument that would be unlikely to succeed if a case of misrepresentation were to fail. In addition, if there were significant delays in the insurer seeking rescission, a case for avoidance would likely be met by arguments as to waiver or limitation. Misrepresentation and non-disclosure arguments may therefore not have provided fertile ground on which to resist a policyholder’s claim, except in very clear cases.

12.02

New York law on misrepresentation is governed by statute, viz section 3105(a) of New York’s Insurance Law. The relevant question, when misrepresentation is alleged, is whether the insurer would have written the insurance policy on the same terms if the policyholder had presented the facts accurately. This inquiry typically raises questions of fact and so constitutes a ‘jury’ question under New York law.2

12.03

1 In previous editions, this chapter has begun with a comparison between English law and New York law, since both systems derived from the common law. There were significant differences in their respective approaches to misrepresentation and non-disclosure, even prior to the radical alteration of English law effected by the Insurance Act 2015. Since the principles developed by English common law and earlier statutes no longer apply in England, we do not consider that a detailed comparison between New York law and English law, whether prior or subsequent to the Insurance Act 2015, is required. 2 Eg, Leamy v Berkshire Life Ins Co 383 NYS2d 564, 564 (NY 1976).

260

Misrepresentation and Non-disclosure

12.04

The burden of proof remains on the insurer throughout. Section 3105 of New York’s Insurance Law requires the insurer to show that the misrepresentation was ‘material’.3 As a starting point, an insurer will need to call the underwriter or underwriters who wrote the policy to give convincing evidence that they would not have written the policy on the same terms. The New York case law shows that unsupported conclusory statements from the underwriter are generally insufficient to sustain the insurer’s burden of proof.4 Accordingly, the insurer will likely need to back up the underwriters’ evidence with corroborative internal documentation. The New York statute contains a provision making the ‘practice of the insurer’ with regard to ‘similar risks’ relevant on the issue of misrepresentation.5 The issue of what a ‘prudent underwriter’ would have done therefore plays no role in the drama. Expert evidence, as to what a prudent underwriter would or would not have done, is not critical and may be of little assistance.6

12.05

New York law imposes no general duty to disclose, and treats only contracts of marine insurance and analogous contracts as uberrimae fidei for these purposes.7 Thus, New York law does not impose on an applicant for other types of insurance a duty to disclose information about which the insurer did not inquire, as long as the applicant has no affirmative duty to know that the information was material to the insurer. As one federal court stated in applying New York law to a commercial (non-marine) insurance dispute: An applicant is entitled to remain silent on matters concerning which he is not questioned. His insurance policy may be voided for concealment only when he conceals matters material to the risk and he does so in bad faith with intent to deceive the insurer.8

12.06

A policyholder can rely upon the doctrines of waiver or estoppel in response to a case of misrepresentation or non-disclosure. New York law draws a distinction between waiver and estoppel. The concept of waiver under New York law focuses on the conduct and knowledge of the insurer. Waiver does not require the insurer to have full knowledge of the relevant facts. A showing that the insurer had constructive notice of the facts will

3 The statute states: ‘No misrepresentation shall avoid any contract of insurance or defeat recovery thereunder unless such misrepresentation is material. No misrepresentation shall be deemed material unless knowledge by the insurer of the facts misrepresented would have led to a refusal by the insurer to make such contract.’ New York Insurance Law, s 3105(b). 4 See below the discussion in paras 12.23–12.27. 5 On this point, the New York statute states: ‘In determining the question of materiality, evidence of the practice of the insurer which made such contract with respect to the acceptance or rejection of similar risks shall be admissible.’ New York Insurance Law, s 3105(c). 6 It is potentially relevant, in that expert evidence as to what a prudent underwriter would or would not have done may provide some evidence to undermine or bolster what the actual underwriter says that he or she would have done. An expert may also be able to assist an arbitral tribunal on matters of general market background which may be beyond the tribunal’s expertise and which may bear on materiality issues. An arbitral tribunal in a Bermuda Form arbitration would have a discretion as to whether or not to admit such evidence, and is unlikely to exclude it if one party says that it wishes to call it. 7 See, eg, Puritan Ins Co v Eagle SS Co 779 F2d 866, 870–71 (2d Cir 1985) (applying New York law); Alaz Sportswear v Pub Mut Ins Co 600 NYS2d 63, 64 (NY 1993). 8 First Fin Ins Co v Allstate Interior Demolition Corp 193 F3d 109, 118 (2d Cir 1999) (applying New York law, quoting Boyd v Otsego Mut Fire Ins Co 510 NYS2d 371, 372 (App Div 1986)).

Introduction

261

suffice.9 The concept of estoppel focuses in New York law on representations made to the policyholder and any prejudicial reliance thereon. These principles, including the approach taken to section 3105, have been spelt out in the New York case law. Various aspects of this case law may seem surprising to an English lawyer or insurer familiar with English decisions in this area. A very large number of the cases involve applications, usually by the insurer, for summary judgment, particularly in the field of life insurance. Unlike English law reports, which are full of reasoned judgments after trial in commercial and other cases involving misrepresentation and nondisclosure, there is a relative paucity of such judgments in New York. There are a number of reasons for this. It may not be the first instinct of the New York lawyer to reach for a misrepresentation defence, save perhaps in life insurance cases where the facts may make the existence of the defence fairly obvious. Perhaps more importantly, however, the majority of trials in the United States in this area, as in other civil cases, involve jury trials. Accordingly, the judge in such cases does not deliver a reasoned judgment at the end of the case; the judge’s role is simply to direct the jury as to the applicable law.

12.07

In directing the jury, the judge will have the assistance of New York Pattern Jury Instructions, which contain a specimen instruction to the jury on the issue of materiality under section 3105(b).10 This specimen instruction to the jury, which is discussed further below, is not binding on a judge, but in practice judges in New York courts are likely to use or at least draw upon this pattern jury instruction. Sometimes, however, the case proceeds without a jury and is then known as a bench trial. A judge typically will deliver after a bench trial a written decision that will contain a more detailed analysis of the law than the usually short decision (if any) issued as a result of a jury verdict or a motion for summary judgment in a non-commercial case, such as one involving life insurance or health insurance.11 The written decisions in commercial cases typically will include more extensive reasoning on the issue of materiality. Lawyers and tribunals may find such judgments more valuable than the brief decisions written as a result of a motion for summary judgment or even a jury decision in a non-commercial case.

12.08

This is not to say that the summary judgment cases are not valuable sources of New York law. Some care needs to be taken, however, when approaching and using them. Under

12.09

9 See generally, Securities & Exch Comm’n v Credit Bancorp Ltd 147 F Supp 2d 238, 255–56 (SDNY 2001). See also New York v AMRO Realty Corp 936 F2d 1420, 1431 (2d Cir 1991) (applying New York law). See further ch 13 below. A waiver is therefore more likely to be found under New York law as compared to English law, which requires knowledge of the right to rescind and the facts giving rise to that right. 10 Committee on Pattern Jury Instructions, Association of New York Supreme Court Justices, New York Pattern Jury Instructions – Civil PJ14 s 4:75. The pattern instruction is based on New York Insurance Law s 3105; and Bronx Sav Bank v Weigandt 1 NY 2d 545 (NY 1956); Langer v Metro Life Ins Co 290 NY 601 (NY 1943); Sommer v Guardian Life Ins Co of Am 28 NY 508, 509–10 (NY 1938). Comments state that the first three subdivisions of s 3105 are applicable to all types of insurance, citing inter alia, Designcraft Jewel Indus Inc v St Paul Fire & Marine Ins Co 59 AD2d 857, 859 (NY App Div 1977), aff ’d, 6 NY2d 796 (NY 1978), but that s 3105(d) applies only to life, health, and ‘accident’ insurance. In the state of New York, the Supreme Court is a trial-level court of general jurisdiction. 11 See, eg, the fully reasoned decision of Judge Block in Home Insurance Co of Illinois (New Hampshire) v Spectrum Information Technologies Inc 930 F Supp 825 (EDNY 1996).

262

Misrepresentation and Non-disclosure

procedure in the United States, including that in the state of New York, a party may move for summary judgment in a case in which the facts are so clear that there is no ‘genuine issue of material fact’ that the judge could properly submit to the jury.12 Summary judgment procedure in the United States then parallels the very similar summary judgment application in England, where one party says that the position is so clear that no trial is needed. When an insurer moves for summary judgment on a misrepresentation defence under New York law, the insurer bears the burden of establishing that the misrepresentation was material’ as a matter of law’; that is, that the facts are so clear that no factual question should be presented to the finder of fact to decide.13 Courts pursue a similar analysis where an insurer moves for a directed verdict during trial or for judgment after a verdict. Again, the insurer will try to show that the misrepresentation was material as a matter of law, and that the court should direct a verdict in its favour on its misrepresentation defence, because there was only one way that the case could properly have been decided. It follows that the cases decided on motions for summary judgment are generally those where the issues are not well balanced, but instead those where the facts clearly favour one party over the other. Cases involving genuine factual questions are not decided on summary judgment, but by juries, often with no written decision seeing the light of day.

NON-DISCLOSURE UNDER NEW YORK LAW 12.10

Under New York law, an ordinary contract of insurance, such as the Bermuda Form excess liability policy, is not a contract of the utmost good faith, and the policyholder is not under a duty to volunteer all conceivable material facts to the insurer when the insurer does not request them. New York law does recognise a sub-category of insurance contracts that require uberrimae fidei, or utmost good faith, namely marine insurance contracts and reinsurance contracts.14 But in ordinary contracts of insurance, rescission for non-disclosure depends on the insurer pleading and proving (by clear and convincing evidence) ‘fraudulent concealment’ of the material facts, that is, where the non-disclosure is done ‘in bad faith, with intent to mislead the insurer’.15

12.11

Accordingly, the elements of actionable non-disclosure under New York law are (i) an insurance applicant’s failure to reveal facts (ii) that are material to the risk presented to the insurer (iii) when the applicant has a duty to reveal the facts. The duty to disclose facts exists only when (i) disclosure is plainly and directly requested in application questions or (ii) non-disclosure would be tantamount to fraudulent concealment.16 Thus, under

12 Eg, Fed R Civ P 56; NY Civ Prac Law & R (CPLR), s 3212. 13 See, eg, Chicago Ins Co v Halcond 49 F Supp 2d 312, 318 (SDNY 1999); Kroski v Long Island Sav Bank 689 NYS2d 92, 93–94 (App Div 1999). 14 Vella v Equitable Life Assur Soc’y 887 F2d 388, 392 (2d Cir 1989) (applying New York law); Home Insurance v Spectrum 930 F Supp at 836. For the position in reinsurance contracts, see Barry R Ostrager and Thomas R Newman, Handbook on Insurance Coverage Disputes, 19th edn (New York, Wolters Kluwer, 2019) s 15.04[a]; Christiania Gen Ins Corp v Great Am Ins Co 979 F2d 268, 278 (2d Cir 1992). 15 Home Insurance v Spectrum 930 F Supp at 840. 16 Vella v Equitable Life 887 F2d at 391–92; Home Insurance v Spectrum 930 F Supp at 835–37.

Misrepresentation under New York Law

263

New York law, non-disclosure arises, in practice, only in the absence of specific questions on the insurer’s application. If the insurer asked a specific question and received an incorrect answer, then the insurer’s defence is misrepresentation, not non-disclosure.

MISREPRESENTATION UNDER NEW YORK LAW The Statutory Requirements The relevant statute is now section 3105 of the New York Insurance Law, which provides as follows:

12.12

(a) A representation is a statement as to past or present fact, made to the insurer by, or by the authority of, the applicant for insurance or the prospective policyholder, at or before the making of the insurance contract as an inducement to the making thereof. A misrepresentation is a false representation, and the facts misrepresented are those facts which make the representation false. (b) No misrepresentation shall avoid any contract of insurance or defeat recovery there under unless such misrepresentation was material. No misrepresentation shall be deemed material unless knowledge by the insurer of the facts misrepresented would have led to a refusal by the insurer to make such contract. (c) In determining the question of materiality, evidence of the practice of the insurer which made such contract with respect to the acceptance or rejection of similar risks shall be admissible.

Subsection (d) deals with a specific issue relating to life insurance, which is not relevant for the present purposes. The various features of this statutory definition, as developed in New York case law, will now be discussed in greater detail. Statement of Past or Present Fact The matter misrepresented must be a past or present fact. Issues can arise as to whether a statement is one of fact or opinion. For example, where applicants for life insurance represent that they are in good health, the New York courts interpret that as a statement that policyholders believe themselves to be in good health.17 This distinction between fact and opinion is important, because a statement of opinion is not a misrepresentation merely because the opinion is wrong. In order to establish a misrepresentation, the insurer must show either that the applicant did not actually hold the opinion or that it was obviously unjustifiable. A striking illustration of this principle is Chicago Insurance Co v Halcond.18 In that case, the policyholder was an anaesthetist who answered ‘no’ in response to a question as to whether any facts or circumstances had occurred in the past year ‘that might give rise to a claim or suit’. In fact, the policyholder had served as nurse/anaesthetist during two surgical procedures that had gone catastrophically wrong,

17 See, 18 49

eg, Kroski 689 NYS2d at 93–94; Bronx Sav Bank v Weigandt 154 NYS2d 878. F Supp 2d 312 (SDNY 1999).

12.13

264

Misrepresentation and Non-disclosure

leaving the patients seriously injured. The insurers’ application for summary judgment failed because the relevant question contained a subjective element – namely whether any of the facts and circumstances ‘might’ lead to a claim or suit. The court held that the anaesthetist might be able to justify his answers at trial, and there would be no misrepresentation if the policyholder justifiably believed that his answer to the application question was accurate.19 The Application Form 12.14

It is not unusual for questions on the insurer’s application form to be unclear. Where questions on an insurer’s policy application form are ambiguous, and the question arises as to whether they were answered accurately, the policyholder is entitled to the most favourable construction of them.20 Questions on insurance applications are ambiguous if they are capable of more than one interpretation when viewed objectively by a reasonably intelligent person. The question is whether the application questions are ‘so plain and intelligible that any applicant can readily comprehend them. If any ambiguity exists, the construction will obtain most favourable to the policyholder’.21 This principle parallels the New York law construing ambiguities in policy exclusions in a manner favourable to the policyholder.22 But whereas Article VI(O) of the 004 Form (or Article V(q) of earlier Forms) modifies the latter rule, that Article does not apply to questions on the application form in the context of a misrepresentation defence. It is well established that the burden of proving all elements of a misrepresentation or non-disclosure defence rests squarely on the insurer.23

12.15

If questions on the application form are interpreted in the manner most favourable to the policyholder, then it may be that a similar approach would be taken to the policyholder’s answers. An ambiguous or partial answer may also call for further enquiry by the insurer. An insurer who has been provided with partial information that was ‘sufficiently indicative of something more to be tantamount to notice’ of the information subsequently alleged to have been misrepresented is not entitled to rescind.24 Materiality

12.16

New York law does not distinguish between ‘materiality’ and ‘inducement’. It is not concerned with what the hypothetical underwriter would have done, but only with what

19 Ibid at 316–17. 20 Vella v Equitable Life 887 F2d at 391–92; Bifulco v Great N Ins Co No 99-CV-0119E(M) 2001 WL 877335, at *3 (WDNY 3 July 2001); Home Insurance v Spectrum 930 F Supp at 837–38. 21 Vella v Equitable Life 887 F2d at 392, citing Halpin v Ins Co of N Am 120 NY 73, 78 (NY 1890). See also Home Insurance v Spectrum 930 F Supp at 837–38; Bifulco 2001 WL 877335, at *3. 22 See above ch 10, in particular paras 10.03–10.10. 23 Home Insurance v Spectrum 930 F Supp at 835. 24 Cherkes v Postal Life Ins Co 138 NYS2d 788, 790 (App Div 1955), aff ’d, 309 NY 964 (1956). See also Glatt v Union Cent Life Ins Co No 92 Civ 1227 (SWK) 1994 WL 329985 (SDNY 11 July 1994).

Misrepresentation under New York Law

265

the particular insurer would have done. Assuming the existence of a false representation, the insurer bears the burden of showing that the misrepresentation was material. Section 3105 of New York’s Insurance Law expressly provides that no misrepresentation shall be deemed material ‘unless knowledge of the facts misrepresented would have led to a refusal by the insurer to make such contract’.25 This includes the scenario where the insurer can show that, had it known the facts, it would have refused to write the policy on the terms that it did.26 The cases where the insurer would have been prepared to write the risk, but would have insisted upon a higher premium or an exclusion, are cases where it is able to establish that it would have refused ‘to make such contract’ within the meaning of section 3105; that is, the contract that it did make. It is therefore not sufficient if all the insurer can do is show that it would have asked or negotiated for a higher premium, unless it can show that it would have declined the risk unless the higher premium was paid. The legislative background to the present section 3105 is explained in the decision of the New York Appellate Division in Giuliani v Metropolitan Life Insurance Co.27 Section 3105 is the successor to section 149 of the Insurance Law, which became effective on 1 January 1940, and the critical section of the present day statute (section 3105(b)) is identical to section 149(2) of the previous statute. The 1940 statute itself replaced section 59 of the New York Insurance Law 1909, pursuant to which (as explained in Giuliani) the tendency of the courts was to determine that every misrepresentation, except the most trivial ones, was material, and thus avoided the policy.28

The Appellate Division in Giuliani distinguished the wording of the earlier statute (section 59), as set forth in an earlier decision of the New York Court of Appeals,29 making clear that the revision of the statute rejected the application of a ‘prudent’ or reasonable underwriter standard in misrepresentation cases: Attention to this status of the law and its apparent unfairness to the assured has been drawn by the dissenting opinion of Judge Finch in Geer v Union Mutual Life Ins Co 273 NY 261, 272, 7 NE2d 125, 130, and by Judge Rippey in his dissenting opinion, in Glickman v New York Life Ins Co 291 NY 45, 52 50 NE2d 538, 540, 148 ALR 454. The examination of the tentative draft of the proposed Insurance Law Revision (1937) shows that in regard to what is now subdivision 2 of section 149, the comment (pp 143 and 144) was made ‘The rule proposed is in accord with the able dissenting opinion by Judge Finch, in which he relies upon the decision of the Privy Council of England’. Reference in such comment was to Geer v Mutual Life Insurance 273 NY 261, 7 NE 2d 125, and Mutual Life Ins Co of New York v Ontario Metal Products Co Ltd [1925] LRAC 344. This comment further states that under subsection 4 as then proposed (and similar to that later enacted as subdivision 3 of section 149)30 ‘proof of what a prudent insurer would have done is merely evidence to show what the insurer in question would have done, and is not the conclusive test’. Considering the background

25 NY

Insurance Law s 3105(b). Insurance v Spectrum 930 F Supp at 841. 27 Giuliani v Metro Life Ins Co 56 NYS2d 475 (App Div 1945). 28 Ibid at 479. 29 Geer v Union Mut Life Ins Co 273 NY 261 (NY 1937). 30 Now New York Insurance Law, s 3105(c). 26 Home

12.17

266

Misrepresentation and Non-disclosure

of the enactment of section 149, as briefly stated herein, it is only fair to conclude that the Legislature in enacting such section intended that, except in cases where it could be said as a matter of law that the concealment of the facts concealed or the misrepresentations and the facts misrepresented were so serious that their very seriousness would establish their materiality as a matter of law, the question of materiality as to misrepresentations and concealments was a question of fact for a jury.31

12.18

Accordingly, leaving aside those cases where materiality can be treated as a ‘matter of law’ – that is, those cases where summary judgment is appropriate because there is no sufficient issue of fact to be properly left to a jury – the question of materiality is a jury question. It involves a straightforward issue of fact: would the underwriter have refused to issue the policy in question. Hence the most recent New York Pattern Jury Instructions provide in relevant part, as follows: An insurance company has the right to ask questions concerning the health of an applicant in order to decide whether it wishes to accept the risk of insuring the applicant. If the answer to any question is not true, and if with knowledge of the truth the company would have refused to issue the policy, the company is obligated only to return the premiums paid and is not liable under the policy. This rule applies even though the Insured person died from a cause that had no connection to the fact misrepresented. … The factual questions that you must decide are whether [the policyholder’s] statement that ((he, she) had not been treated for coronary artery disease) was true, and if you find that it was not, whether knowledge of the true facts would have led [the insurer] to refuse to issue the policy. On both those questions, [the insurer] has the burden of proof. … If you find that the statement was not true, you will then consider whether knowledge of the true facts would have led [the insurer] to refuse to issue the policy that it issued on [the policyholder’s] life. If you find that knowledge that [the policyholder] had (been treated for coronary artery disease) would have led [the insurer] to refuse to issue the policy, you will find for the plaintiff in the amount of the premiums paid on the policy … If you find that knowledge that [the policyholder] had (been treated for coronary artery disease) would not have led [the insurer] to refuse to issue the policy, you will find for the plaintiff in the face amount of the policy.32

12.19

There are numerous cases where the courts have asked this straightforward question, namely whether the insurer would have refused to issue the policy.33 One oddity in this 31 56 NYS2d at 479. 32 New York Pattern Jury Instructions – Civil PJ1 s 4:75 (above n 10). New York’s Pattern Jury Instructions refer to a claim under a life insurance policy, and use the illustration of a policyholder of a life insurance policy who has told the insurer incorrectly that he or she has not been treated for coronary heart disease. The relevant Pattern Jury Instructions have been in this form for many years, but may be revised by the judge to fit the case at hand. The notes to the Pattern Jury Instructions indicate that the pattern instruction to the jury should be amended where the insurance company ‘maintains that it would not have issued the policy at the premium charged’. 33 See also First Financial Ins Co v Allstate 193 F3d at 118 (‘such misrepresentation is material under New York law only if it is shown that the insurer would not have written the insurance policy had the facts at issue been disclosed’) (citing Amrep Corp v Am Home Assur Co 440 NYS2d 244, 247 (App Div 1981); and Testa v Utica Fire Ins Co 610 NYS2d 85 (App Div 1994)); Christiania v Great Am Ins 979 F2d at 278 (applying New York law) (‘A fact is material so as to void ab initio an insurance contract if, had it been revealed, the

Misrepresentation under New York Law

267

area of the law, however, is the citation in some cases,34 particularly summary judgment cases, of a passage from the majority judgment in Geer v Mutual Life Insurance Co35 which appears to apply a different test from the straightforward ‘would have led to a refusal’ criterion set out in section 3015(b). Taken out of context, this passage suggests that it is sufficient for the insurer to show only that it ‘might’ otherwise have taken different action.36 Geer was decided in 1937, prior to the enactment of the 1940 Insurance Law and its revision in section 3105. We consider, however, that the law of New York is represented by section 3105, not (insofar as it poses a different test) by the majority decision in Geer. The New York state authorities that cite Geer as an authority37 do so without apparently being conscious of any contradiction between the test set out in the majority judgment and the wording of section 149 of the 1940 statute (or section 3105 of the current statute), and without insurer or reinsurer would either not have issued the policy or would have only at a higher premium’); Zilkha v Mut Life Ins Co of NY 732 NYS2d 51, 52 (App Div 2001) (‘A misrepresentation is material if the insurer would not have issued the policy had it known the facts misrepresented’) (citing Penn Mut Life Ins Co v Remling 702 NYS2d 375, 376–77 (App Div 2000)); Carpinone v Mut of Omaha Ins Co 697 NYS2d 381, 383 (App Div 1999) (‘The insurer’s proof must establish that it would not have issued the same policies if the correct information had been disclosed in the applications’); Campese v National Grange Mut Ins Co 689 NYS2d 313, 314 (App Div 1999) (‘To meet its burden, defendant had to submit proof concerning its underwriting practices with respect to applicants with similar circumstances, establishing that it would have denied the application had it contained accurate information’); Feldman v Friedman 661 NYS2d 9, 10 (App Div 1997) (‘A fact is material so as to avoid ab initio an insurance contract if, had it been revealed, the insurer or reinsurer would either not have issued the policy or would have only at a higher premium’); Cutrone v Am Gen Life Ins Co of NY 606 NYS2d 491, 492 (App Div 1993) (‘To meet that burden [summary judgment], defendant was required to adduce proof concerning its underwriting practices with respect to applicants with similar conditions, establishing that it would have rejected the application if the information had been truthful’); Tennenbaum v Ins Corp of Ireland 579 NYS2d 351, 352 (App Div 1992) (‘An innocently made factual misrepresentation may serve to void an insurance contract if knowledge by the insurer of the facts misrepresented would have led to a refusal by the insurer to make such contract’) (citing New Insurance Law, s 3105(b)); Morales v Castlepoint Ins Co 4 NYS 3d 297 (App Div 2015). 34 See, eg, Vander Veer v Continental Cas Co 356 NYS2d 13 (NY 1974); Leamy 383 NYS2d 564; Process Plants Corp v Beneficial Nat’l Life Ins Co 385 NYS2d 308 (App Div 1976), aff ’d, 397 NYS2d 1007 (NY 1977); Barrett v State Mut Life Assur Co 396 NYS2d 848 (App Div 1977), aff ’d, 407 NYS2d 478 (NY 1978). 35 Geer 273 NY 261. 36 See the majority judgment in ibid at 266–67 (citing Jenkins v John Hancock Mut Life Ins Co 257 NY 289 (1931) and Keck v Metro Life Ins Co 264 NYS 892 (App Div 1933), aff ’d, 264 NY 422 (1934)): No method has been devised by which the processes of the human mind can be charted and the force of inducement mechanically measured. The materiality of a representation may then depend upon the idiosyncrasies or the individuality of the person who acts upon the representation, and often must be determined as a question of fact by the trier of the facts. Nevertheless, at times, departure in a representation from an accurate statement of the truth may be so slight that we may confidently say that the difference could not affect the decision of any reasonable person. Then as a matter of law the misrepresentation is not material. On the other hand, where an applicant for insurance has notice that before the insurance company will act upon the application, it demands that specified information shall be furnished for the purpose of enabling it to determine whether the risk should be accepted, any untrue representation, however innocent, which either by affirmation of an untruth or suppression of the truth, substantially thwarts the purpose for which the information is demanded and induces action which the insurance company might otherwise not have taken, is material as a matter of law. The question in such case is not whether the insurance company might perhaps have decided to issue the policy even if it had been apprised of the truth, the question is whether failure to state the truth where there was a duty to speak prevented the insurance company from exercising its choice of whether to accept or reject the application upon a disclosure of all the facts which might reasonably affect its choice.

37 See,

eg, Process Plants 385 NYS2d at 311; Vander Veer 356 NYS2d at 52–53.

12.20

268

Misrepresentation and Non-disclosure

reference to the legislative history of the statute or the Giuliani decision. It may be that Geer can be harmonised with the statute on the basis that the majority in Geer were concerned with the question of whether the untrue statement induced the insurer to make the same contract. That question raises a causal enquiry as to the effect of the untruth upon the insurer’s decision to insure, and that in turn can only be answered in the case of concealed information by asking whether, on the balance of probabilities, the insurer would not have granted the insurance on the same terms but for the misrepresentation or concealment. That is in substance the same question as that envisioned in section 3105(b)’s formulation of the materiality standard. 12.21

Furthermore, the citation of Geer must not be taken out of context. In the summary judgment cases, the insurer is saying that the facts are so clear that the misrepresentation is material ‘as a matter of law’ and that no disputed factual issues should go to the jury. The case law developed in that context shows that the insurer must establish a number of matters to obtain summary judgment. First, the insurer must show that it would not have written the risk if it had known the facts. The affidavit evidence supporting the insurer’s motion for summary judgment must satisfy the materiality standard set forth in section 3105.38 Second, conclusory statements by the underwriter to this effect are not sufficient; instead, or in addition, the insurer must support its motion with evidence of its underwriting practices, such as underwriting manuals or other corroborating evidence.39 Third, the evidence on materiality must be clear and substantially uncontradicted.40 When the insurer meets these conditions, then the court grants summary judgment, avoiding a jury trial. It is in this context that the quotation from Geer sometimes appears. However, this is not because the legal test for materiality under New York law is anything other than as set out in the statute. Indeed, the requirements for obtaining summary judgment are stringent, and an insurer will not obtain a summary judgment if it merely offers statements by its own underwriters that it might not have issued the policy. Instead, it is because if those conditions are satisfied, the court is able to say that the representation was material as a matter of law; that is, that there is no genuine issue of material fact that should go to a jury. If the conditions are not satisfied, then the case is set for trial by the trier of fact.

12.22

At all events, it is unlikely that a tribunal will feel inclined to use the majority decision in Geer as the standard for materiality given the following circumstances: (a) Geer purports to apply the earlier 1909 Insurance Law rather than either section 149 or the current section 3105; (b) the judgment in Giuliani shows not only that New York statutory law was changed after (and because of) the Geer decision, but also that the verdict in Giuliani was upheld after the court had instructed the jury under section 149 of the

38 See Berger v Manhattan Life Ins Co 805 F Supp 1097, 1103 (SDNY 1992) (for a case where the court denied the insurer’s summary judgment motion for lack of supporting evidence. The court relied on the fact (amongst other reasons) that the underwriter did not unequivocally state that he would not have issued the exact policy had he known the truth at the time of the application). 39 See, eg, Wittner v IDS Ins Co of NY 466 NYS2d 480 (App Div 1983). 40 See Berger 805 F Supp at 1097; Bifulco 2001 WL 877335, at *3 (The evidence on materiality may be contradicted by some relatively small matter, which will prevent a successful summary judgment application); see, eg, in Bifulco the fact that following the loss, the policy was renewed.

Misrepresentation under New York Law

269

1939 statute; and (c) numerous New York Appellate Division cases decided since Geer41 have disregarded the Geer test and have applied instead a ‘would not have’ test, as do the pattern jury instructions. Discharging the Burden of Proof The burden of proving all aspects of misrepresentation, in particular materiality, is upon the insurer who asserts it.42 Section 3105(c) of the New York Insurance Law provides

12.23

evidence of the practice of the insurer which made such contract with respect to the acceptance or rejection of similar risks shall be admissible.

In the context of applications by the insurer for summary judgment, where the insurer is seeking to prove materiality as a matter of law, the approach of the New York courts has not simply been to admit evidence of the insurers’ practices, but to require such evidence if the insurer is to discharge its burden. Accordingly, if the insurer’s evidence comprises only conclusory statements by underwriters as to what they would have done, unsupported by underwriting manuals or other documentary evidence relating to the insurer’s practices, summary judgment will not be granted. Accordingly, unless the insurer is able to renew the summary judgment application with appropriate evidence, the case will be sent for trial for findings of fact to be made.43 Thus, the materiality of an applicant’s misrepresentation is ‘ordinarily a factual question unless the insurer proffers clear and substantially contradicted evidence concerning materiality, in which event the matter is one of law for the court to determine’.44 In order for an insurer to establish materiality as a matter of law, ‘it was required to present documentation concerning its underwriting practices, such as its underwriting manuals, rules or bulletins which pertain to insuring similar risks’.45 A controversial question is whether the summary judgment requirement (that is, for documentation pertaining to similar risks) is applicable when an issue of materiality comes to be resolved at the substantive hearing on the merits in a Bermuda Form arbitration. If underwriters provide persuasive evidence that they would not have written

41 Zilkha v Mut Life Ins Co 732 NYS2d 51, 52 (App Div 2001); Carpinone 697 NYS2d at 383; Campese 689 NYS2d at 314; Meagher v Exec Life Ins Co 607 NYS2d 361 (App Div 1994); Cutrone 606 NYS2d at 492; Tennenbaum 579 NYS2d at 352–53; Sonkin Assocs v Columbian Mut Life Ins Co 541 NYS2d 611, 612 (App Div 1989). See also the cases in n 33 above. 42 Home Insurance v Spectrum 930 F Supp at 835. In Ashline v Genesee Patrons Cooperative Insurance Co 638 NYS2d 217 (App Div 1996), the court referred in passing to the need for the insurer to establish misrepresentation by ‘clear and convincing evidence’. 43 See, eg, Wittner 466 NYS2d 480; Berger 805 F Supp at 1102; Feldman 661 NYS2d 9; McDaniels v Am Bankers Ins Co 643 NYS2d 846, 847 (App Div 1996); Nationwide Mut Fire Ins Co v Pascarella 993 F Supp 134 (SDNY 1998); First Financial v Allstate 193 F3d at 119; Campese 689 NYS2d at 314; Carpinone 697 NYS2d at 383; Curanovic v NY Cent Mut Fire Ins Co 762 NYS2d 148 (App Div 2003); Parmar v Hermitage Ins Co 800 NYS2d 726 (App Div 2005); Lenhard v Genesee Patrons Co-op Ins Co 818 NYS2d 644 (App Div 2006); Schirmer v Penkert 840 NYS2d 796 (App Div 2007). 44 Carpinone 679 NYS2d at 383. 45 Ibid.

12.24

270

Misrepresentation and Non-disclosure

the particular risk if they had not been misled, but are unable to buttress the evidence with an underwriting manual or other documentary evidence, is a tribunal required, as a matter of New York law, to reject the insurer’s case on materiality? In other words, is it impermissible, as a matter New York law, for a tribunal to uphold an insurer’s arguments on materiality, in circumstances where no manual or documentary evidence has been produced to buttress the evidence of the underwriter? 12.25

An argument can certainly be advanced that it is not permissible. The principal cases in which the courts have referred to the need for documentary evidence are, however, summary judgment cases. As indicated above, if the insurer cannot obtain summary judgment, then the question of materiality is a question of fact to be determined subsequently. This question of fact would need to be determined on the basis of all the evidence tendered. Thus, there are a number of cases where, if the facts are considered in detail, it appears that the insurer has succeeded in establishing materiality without providing documentary evidence to buttress the testimony of the underwriter.46 In some cases, courts have suggested that, at trial, documentary evidence should be presented if available.47 In other cases, courts have indicated that the absence of corroborating evidence was a matter for the jury to consider when evaluating the evidence as a whole.48 The absence of documentary evidence is therefore not a matter that entitles the policyholder automatically to obtain summary judgment in its favour.49

12.26

In Home Insurance v Spectrum,50 the court, delivering judgment after a bench trial, referred to cases identifying the need for an insurer to produce documentary evidence of its practices. It then rejected a misrepresentation defence asserted under directors’ and officers’ insurance policies when the insurers failed to establish their underwriting practices with regard to similarly situated policyholders. There, the insurers

46 See, eg, Greene v United Mut Life Ins Co 238 NYS2d 809 (Sup Ct 1963), aff ’d, 258 NYS2d 323 (App Div 1965) (where the court distinguished between the position of a medical officer with sole responsibility for deciding upon insurance applications, and the position of a giant life insurance company with offices around the country and whose officers were ‘of necessity for uniformity subject to rules and manuals of their company’); L Smirlock Realty Corp v Title Guar Co 421 NYS2d 232 (App Div 1979), aff ’d as modified, 437 NYS2d 57 (NY 1981) (where the court considered that no insurer with knowledge of the relevant facts would have issued the policy); Equitable Life Assur Soc’y of US v O’Neil 413 NYS2d 714 (App Div 1979) and Ris v National Union Fire Ins Co of Pittsburgh, Pa No 86 Civ 9718, 1989 WL 76199 (SDNY 6 July1989) (where the court was prepared to grant summary judgment in favour of the insurer in an obvious case, even in the absence of the insurer’s rules or practices); Meagher v Exec Life Ins Co 607 NYS2d 361 (App Div 1994) (where the court, in upholding the jury’s verdict, referred only to the testimony of the underwriter). 47 Brown v Metro Life Ins Co 343 NYS2d 443, 446 (App Div 1973); Lindenbaum v Equitable Life Assur Soc’y of US 174 NYS2d 421, 423 (App Div 1958). 48 Olezeski v Fingers Lakes-Seneca Co-op Ins Co 629 NYS2d 873, 874 (App Div 1995) (‘we are unwilling to say that it was irrational for the jury to discount the probity of [the insurer’s vice president’s] unsupported assertion’); Winnick 494 NYS2d at 510 (‘A jury having the right and the responsibility to determine the credibility of the evidence could properly have made the determination that the underwriter’s conclusion, unsupported by any other evidence, failed to satisfy defendant’s burden of proof’). 49 See, eg, Campese 689 NYS2d at 314, where the court’s order, following dismissal of the insurer’s summary judgment application, was to reinstate the complaint. Compare Tuminelli v First Unum Life Ins Co 648 NYS2d 967 (App Div 1996), where, in a very short judgment, the court did grant summary judgment for the policyholder, presumably on the basis that the policyholder’s case was sufficiently strong. 50 930 F Supp 825 (EDNY 1996).

Misrepresentation under New York Law

271

argued that the policyholder had made material misrepresentations in various insurance applications by failing to disclose an inquiry by the US Securities and Exchange Commission (SEC) in response to questions seeking information about ‘actions’, ‘litigation’ and ‘proceedings’ in which the policyholder or its officers and directors were involved. After the bench trial, the court concluded that the alleged misrepresentations were not material to the insurers’ decisions to sell the insurance because: (i) there was a dearth of evidence – for example, evidence regarding the insurers’ underwriting of other policyholders’ insurance – to support the insurers’ assertion that disclosure of the SEC inquiry would have been material to their underwriting decisions; (ii) evidence regarding the underwriting history and course of dealing between the insurers and the policyholder undercut the insurers’ argument as to the materiality of the SEC inquiry; and (iii) the policyholder had disclosed the facts underlying the SEC inquiry to the insurers, even though the policyholder did not disclose the inquiry.51 We doubt whether this case establishes a rule of New York law that an insurer cannot succeed in proving materiality unless it provides evidence of its practice with respect to similar risks. The court’s conclusions appear to have been based on its review of all the evidence52 and, in any event, was decided by a New York federal court, not by the state’s highest court, the New York Court of Appeals.53 We therefore consider that a tribunal is likely to treat the question of materiality simply as a question of fact. The decision in Home Insurance Co of Illinois v Spectrum does, however, illustrate that an insurer who fails to provide documentary support for its case on materiality may find it difficult to discharge its burden of proof. Even if there is no principle of law that corroborating documentary evidence needs to be produced, the New York case law does at least demonstrate that corroborating evidence is important if the insurer is to discharge its legal burden. Accordingly, a Bermuda Form insurer who relies upon a misrepresentation defence must expect, as part of the discovery process, to give the opposing policyholder fairly extensive documentation. This disclosure may include the insurer’s underwriting manuals (if they exist), and also documents relating to risks of policyholders in a similar position to the opposing policyholder. The tribunal will no doubt seek to keep such discovery within appropriate bounds, but in principle section 3105(c) of the New York Insurance Law requires such discovery so as to enable the policyholder to test the insurers’ case.54 It might be suggested that section 3105(c) need

51 Ibid at 841–43. 52 See, in particular, Home Insurance v Spectrum 930 F Supp at 841 (‘Based on all the circumstances, the Court concludes’); ibid at 843, where, after a review of the evidence, the court concluded that Aetna and Home had not met their burden of establishing the materiality of the SEC Inquiry to their underwriting decision. Other cases where the insurer’s case on materiality has failed in the absence of corroborating documentary evidence are also explicable on the basis that the evidence tendered was insufficient to persuade the jury, rather than that it was insufficient to meet a legal requirement; see, eg, Olezeski 629 NYS2d 873; Zielinski v Associated Mut Ins Co 629 NYS2d 894 (App Div 1995);Winnick 494 NYS2d 509. 53 See the discussion of the US court system in the Appendix, paras A.12–A.21 below. 54 Contrast the position in English High Court litigation, where the court is sometimes reluctant to order disclosure of documents in order to avoid having to go through the process of trying more cases than the one before it; see Marc Rich & Co v Portman [1996] 1 Lloyd’s Rep 430, 441–42 (Longmore J). The decision was not disturbed on appeal: [1997] 1 Lloyd’s Rep 225 (CA).

12.27

272

Misrepresentation and Non-disclosure

not be applied by a London arbitration tribunal, since the admissibility of evidence is a matter within the discretion of the tribunal under section 34 of the English Arbitration Act 1996. Even if a tribunal accepted this proposition as a matter of principle, both the statute and the New York case law are likely to guide the tribunal’s exercise of its discretion during the discovery process. If the insurer does not provide the relevant documentation, then the tribunal is entitled to infer that the insurer’s underwriting practices do not support its misrepresentation defence. Anti-avoidance Provision of the Statute 12.28

Section 3105(b) contains an ‘anti-avoidance’ provision; that is, a provision designed to stop an insurer from evading the statute’s materiality requirements by drafting methods, such as drafting into the policy an obligation to give materially accurate answers as a condition precedent to cover. Thus, the statute provides that: ‘No misrepresentation shall avoid any contract of insurance or defeat recovery thereunder unless such misrepresentation was material.’55

REMEDY FOR MISREPRESENTATION 12.29

Misrepresentation affects the validity of the contract, and the appropriate remedy is to rescind the policy ab initio (from the beginning). An insurer can assert this entitlement by declaring that the contract has been rescinded and suing to rescind the contract, or by making an affirmative demand for rescission in defence of the suit. In such circumstances, the insurer will generally be expected to tender the premium back prior to trial. Alternatively, it seems that the insurer can defend the claim on the basis of the misrepresentation, in which case the insurer is not required to tender the premium prior to trial.56 When the insurer relies upon misrepresentation, the defence is often referred to as ‘fraud in the procurement’.

12.30

When an insurer successfully defends a claim based upon misrepresentation, does it mean that the policy remains otherwise intact, thereby enabling the insurer to retain the premiums after the successful defence? This would be the position if a defence based upon misrepresentation was in the nature of a coverage defence. The cases indicate, however, that this is not the nature of defence based on misrepresentation. In Process Plants Corp v Beneficial National Life Insurance Co,57 the court described it as a ‘defense sounding in equity to rescind, i.e. set aside an insurance contract for material misrepresentation’.

55 New York Insurance Law, s 3105(b) (emphasis added). 56 Berger 805 F Supp at 1110; Perry v Metro Life Ins Co 153 NYS 459 (App Div 1915). 57 385 NYS2d 308; cf Mooney v Nationwide Mut Ins Co 577 NYS2d 506 (App Div 1991) (where rescission was barred by statute, but the insurer was permitted to advance an affirmative defence of fraud. In such a case, the ‘defeat recovery thereunder’ language in s 3105(b) would also presumably have the effect of imposing a requirement of materiality).

Remedy for Misrepresentation

273

Other cases show that the consequence of a successful defence is that the policy is void ab initio.58 This legal consequence equates to a rescission of the insurance policy. Thus, where the defence succeeds, the claim for payment under the policy will fail, but the policyholder will be entitled to a return of the premium paid.59 It is also clear that the defence of misrepresentation is not simply a matter that relates to the scope of coverage, but goes to the question of whether the policy was ever validly entered into.60 Thus, as illustrated by the cases discussed below, the defence can be waived, for example by conduct which affirms the continued existence of the policy. Since a defence based on misrepresentation establishes the initial invalidity of the insurance policy, the insurer would not therefore be in a position to assert the validity of the policy for the purposes of seeking to retain premiums. Although the law relating to misrepresentation is now grounded in section 3105, the origins of the defence and the remedy of rescission of the contract lie in the common law. In this context, the law relating to the avoidance of insurance contracts for misrepresentation arises from a more general principle that any misrepresentation which induces a contract (whether or not a contract of insurance) renders the contract voidable.61 It is because the contract is voidable (and therefore void ab initio if the misrepresentee validly elects to rescind) that the contract is capable of affirmation. As Professor Corbin states:

12.31

As is the case with any kind of voidable transaction, the aggrieved party may affirm the contract and thereby ratify it. The ratification may be express, or it may occur by actions inconsistent with disaffirmance after acquisition of facts that give notice that a misrepresentation has been made; for example an insurance company’s acceptance of premiums after learning of a misrepresentation precludes it from avoiding the policy. Whether particular conduct constitutes ratification is often a question of fact.62

Rescission is the natural and expected remedy for a defence that goes to contract formation and not to contract interpretation. A misrepresentation defence goes to the existence of the contract, and not simply to its meaning, and thus is not an ordinary coverage or contract defence, but, rather, seeks to put the parties back in the position they would

58 See the cases in n 64 below; and In re World Com Inc Sec Litig 354 F Supp 2d 455, 465 (SDNY 2005) (where the court held that, until the issue of rescission is adjudicated by the court, the contract of insurance remains in effect and accordingly a duty to pay defence costs remains enforceable). In Federal Insurance Co v Kozlowski 792 NYS2d 397, 401–02 (App Div 2005), the court held that notice of rescission could not, without court sanction, serve to suspend even temporarily obligations which have accrued under the policy. If the rescission was justified, then ‘the policy will be rendered void from its inception irrespective of the point in the life of the policy that a liability claim may have arisen’. Under English law, rescission is not the act of the court, but the act of the party that declares the rescission. See Brotherton v Aseguradora Colseguros SA [2003] EWCA Civ 705, [2003] 2 All ER (Comm) 298, [27]. 59 See, eg, Kantor v Nationwide Life Ins Co 227 NYS2d 703 (Sup Ct 1962) and the cases in n 65 below. 60 SEC v Credit Bancorp 147 F Supp 2d at 257–58. 61 See generally Joseph M Perillo, Corbin on Contracts: Avoidance and Reformation (rev ed) (Newark, NJ, Matthew Bender & Co, 2002). See also CNA Reinsurance of London v Home Ins Co No 85 Civ 5681, 1990 WL 3231 (SDNY 10 January 1990), where the court held that the general contract principles relating to waiver/ affirmation should ‘apply with equal force in the insurance context’. 62 Perillo, Corbin on Contracts (above n 61) para 28.23 (footnotes omitted).

12.32

274

Misrepresentation and Non-disclosure

have been in but for the alleged misrepresentation or non-disclosure.63 Thus, numerous cases proceed on the basis that the effect of a misrepresentation is to avoid the contract ab initio; that is, that the insurer is entitled to rescind.64 Similarly, since rescission operates ab initio, numerous cases have required the insurer to return the policyholder’s premiums.65 Indeed, as already seen, the pattern jury instructions expressly require the jury to award return of premiums to an unsuccessful claimant in the event that the insurer’s misrepresentation defence succeeds. 12.33

It is the line of authority concerning waiver – that is, the circumstance where the insurer is held to have waived its right to rely upon a misrepresentation – that shows, perhaps most clearly, that misrepresentation is not a coverage defence, but, instead, goes to the validity of the insurance policy. The next chapter deals with waiver in more detail, but some of the cases should be noted in the present discussion.

12.34

In McNaught v Equitable Life Assurance Society of United States,66 it was (unusually) the policyholder who wanted to rescind the contract and receive a return of premiums. But the policyholder was concerned that if he stopped paying premiums and his claim for rescission failed, he would lose all benefit of the contract, because the insurer would exercise its entitlement under the contract to treat the premiums paid as forfeit for nonpayment. Accordingly, the policyholder attempted to pay further premiums without prejudice to his claim to rescind the contract. The court held that his act of paying the premiums, even though accompanied by a reservation of rights, affirmed the contract. The judgment of the Appellate Division sets out, in this context, principles of English law and equity as to rescission, and indeed cites an English House of Lords case. According to the Appellate Division: it seems elementary enough that a contract cannot be both rescinded and affirmed at the same time, and that ‘the effect of payment of the premium after discovery of the alleged fraud was … to continue the contract in force’.67

63 See generally American Law Institute, Restatement of the Law (Third) Restitution s 1, cmt a (St Paul, MN, American Law Institute Publishers, 2011). 64 See, eg, Christiania v Great American 979 F2d at 278 (‘A fact is material so as to void ab initio an insurance contract if, had it been revealed, the insurer or reinsurer would either not have issued the policy or would have only at a higher premium’); SEC 147 F Supp 2d at 253 (‘An insurance policy will be void where it is proven that the insured fraudulently concealed a material fact in applying for coverage’); American Int’l Specialty Lines Ins Co v Towers Fin Corp No 94 Civ 2727, 1997 WL 90627, at *7 (SDNY 12 September 1997) (‘under New York law … an insurance policy issued in reliance on material misrepresentations is void from its inception’) (citing cases including Rep Ins Co v Masters Mates & Pilots Pension Plan 77 F3d 48, 52 (2d Cir 1996); In re Union Indem Ins Co of NY 611 NYS2d 506, 511 (App Div 1994), aff ’d, 651 NYS2d 383 (1996)); Feldman 661 NYS2d at 10 (‘a fact is material so as to avoid ab initio an insurance contract if, had it been revealed, the insurer or reinsurer would either not have issued the policy or would have only at a higher premium’); Process Plants 385 NYS2d at 311. 65 See, eg, First Financial v Allstate 193 F3d at 111; Friedman v Prudential Life Ins Co 589 F Supp 1017, 1027–28 (SDNY 1984); Bifulco 2001 WL 877335, at *2; Carpinone 679 NYS2d at 383; Myers v Equitable Life Assur Soc’y of US 401 NYS2d 325, 327 (App Div 1978). A ruling requiring the return of premiums can have significant financial consequences if a substantial amount of time has passed since the time the premiums were paid. In such a case, an insurer ordered to disgorge payment of policy premiums may also be required to pay interest on the amount of the premiums in question. 66 McNaught v Equitable Life Assur Soc’y of US 121 NYS 447 (App Div 1910). 67 Ibid at 451.

Remedy for Misrepresentation

275

As McNaught shows, the party asserting misrepresentation must not, by conduct inconsistent with avoidance, have affirmed the contract.68 Zeldman v Mutual Life Insurance Co of New York69 was a more typical case where the insurer was seeking to avoid cover, and the policyholder argued that the insurer had waived this defence. The Appellate Division found no waiver on the facts of the case, but explained the general principles applicable to waiver:

12.35

The general rule applicable to waiver is that the delivery of a policy or the receipt of premiums with knowledge of a then existing breach of conditions as to the health of the insured, or his treatment, will give rise to a waiver, or more properly, an estoppel … Upon receipt of knowledge during the lifetime of the insured of breach of condition, the insurer, if it desires to do so, must promptly exercise its election to void the policy.70

The nature of the remedy for misrepresentation arose squarely in Securities & Exchange Commission v Credit Bancorp Ltd.71 The insurers in that case had, with knowledge of facts misrepresented, retained the policyholder’s premiums, negotiated endorsements, redrafted various marketing and business documents relating to the insurance, and annually renewed the policyholder’s insurance policies. The policyholder alleged that these facts amounted to a waiver of the insurers’ right to rely on the misrepresentations. The insurers countered, relying on the principle (discussed further in the following chapter) that defences relating to the existence of coverage cannot be waived because a court cannot create coverage where none exists. Whilst accepting such a principle, the court held that misrepresentation was not a defence to which this principle applied because, with misrepresentation, ‘the issue is not the coverage sought, but whether the policy was ever validly entered into’.72

12.36

In CNA Reinsurance of London v Home Insurance Co,73 London reinsurers sought to avoid coverage for an alleged misrepresentation, and the reinsured argued that the reinsurer had waived the defence. The court agreed, explaining that an ‘insurer’s renewal of an insurance policy with knowledge of the insured’s prior misrepresentation or breach operates as a waiver of any claim by the insurer based upon the misrepresentation or breach’.74

12.37

68 Although McNaught was decided in the early part of the twentieth century, the New York Appellate Division has relied upon the decision in recent times in Continental Insurance Co v Helmsley Enterprises Inc 622 NYS2d 20, 20–21 (App Div 1995). For examples of non-insurance cases applying similar principles, see, eg, ESPN Inc v Office of Comm’r of Baseball 76 F Supp 2d 383 (SDNY 1999); Barrier Sys Inc v AFC Enterprises Inc 694 NYS2d 440 (App Div 1999). 69 Zeldman v Mut Life Ins Co of NY 53 NYS2d 792 (App Div 1945). 70 To similar effect, see Scalia v Equitable Life Assur Soc’y of the US 673 NYS2d 730 (App Div 1998), where premiums were accepted after the insurer gained sufficient knowledge of the misrepresentation. Contrast Belesi v Conn Mut Life Ins Co 707 NYS2d 663 (App Div 2000), where no premiums were accepted and the court held that there had been no waiver. See also ch 13 below for a discussion of the differences between waiver and estoppel. 71 147 F Supp 2d 238. See too US Life Ins Co in City of NY v Blumenfeld 938 NYS 2d 84 (App Div 2012) (‘An insurer’s failure to rescind a policy promptly after obtaining sufficient knowledge of alleged misrepresentations by an insured constitutes ratification of the policy … Moreover, an insurer that accepts premiums after learning of facts that it believes entitles it to rescind the policy has waived the right to rescind’). 72 Ibid at 257–58. 73 No 85 Civ 5681, 1990 WL 3231 (SDNY 1990). 74 Ibid at *9.

276

Misrepresentation and Non-disclosure

Return of Premiums 12.38

If the insurer does validly avoid a contract on the Bermuda Form, a question arises as to the amount of the premium that it is required to return. As discussed earlier,75 the Bermuda Form does not, as other insurance policy forms do, expire at the end of an annual policy period; instead, the Bermuda Form is extended and therefore continues at the end of each ‘Annual Period’, until the policy is cancelled by one of the parties. For example, a Bermuda Form policy may have run for a large number of years, with an original placement and a number of extensions. The relevant misrepresentation may have been made only at, for example, one of the renewals. Does the insurer have to return all the premiums received or only the premium received in relation to the renewal tainted by the misrepresentation?

12.39

The general principle is that rescission ‘must be as to the whole contract and not merely as to part’.76 New York law sometimes permits partial rescission, but only where a part of the contract can properly be regarded as separate and distinct.77 The Bermuda Form contract is, however, a continuous contract. If the misrepresentation were made at the inception of the contract, then it seems logical that all premiums paid as from inception, including subsequent extensions, should be returned. This is because the later extensions cannot have an existence separate from an original contract tainted by misrepresentation. The position is, perhaps, not so straightforward if the only misrepresentation is made in year 5 of the policy. It could be argued that the insurer would be entitled to retain the premiums up until year 5, and that the extension in year 5, together with all subsequent extensions, can be treated separately. The consequence would be that all premiums received for year 5 and later years would need to be returned. Again, given the continuous nature of the contract, it seems impossible to pluck one year (in this example, year 5) from the entirety of the contract and treat that single year as severable from all the rest. Accordingly, in this example, we do not consider that the only premium returnable by insurer would be the premium received for year 5. Limitation

12.40

Section 213 of the Civil Practice Law & Rules (CPLR) governs limitation periods under New York law. Section 213(1) provides for a six-year limitation period where no period is specifically prescribed by law. Section 213(8) governs cases where fraud is alleged and provides: an action based upon fraud; the time within which the action must be commenced shall be the greater of six years from the date the cause of action accrued or two years from the time the plaintiff … could with reasonable diligence have discovered it.

75 See above, eg, ch 2, para 2.09; and ch 11, paras 11.94–11.96. 76 McNaught 121 NYS at 449. 77 NY Marine & General Ins Co v Tradeline (LLC) No 98 CIV 7840; 2000 WL 739567, at *4 (SDNY 7 June 2000), aff ’d in part, rev’d in part, 266 F3d 112 (2d Cir 2001).

Remedy for Misrepresentation

277

Where an insurer alleges misrepresentation in the inducement of a contract, the six-year limitation period for rescission begins to run on the date when the contract was concluded.78 If the substance of the allegation is fraud, then the insurer will have the benefit of the additional two-year period running from the date when it could with reasonable diligence have discovered the fraud.79

12.41

It is often the case under New York law that the insurer will take the initiative by commencing an action seeking rescission of the contract, not least because it may be under a continuing duty to defend the policyholder until there is a judicial determination as to whether it is entitled to rescind. In In re Worldcom Inc Securities Litigation, the court held that until ‘the issue of rescission is adjudicated, a contract of insurance remains in effect and the duty to pay defense costs is enforceable’.80 In Federal Insurance Co v Tyco International Ltd, the court held that after claims had been asserted, ‘a rescission by notice cannot, without legal sanction, have retroactive effect and serve to suspend, even temporarily, obligations that – absent a basis for rescission – have accrued under the policy’.81

12.42

The six-year limitation period will therefore clearly apply where it is the insurer who commences the action seeking rescission. Since it is essential that the insurer does seek to rescind where misrepresentation is alleged, it would seem that the statute would also be applicable in the case where the policyholder commences an arbitration and the insurer seeks to defend the claim by rescinding for misrepresentation.82 The six-year limitation period means that, where an Bermuda Form policy has run for many years, there may be limited scope for a Bermuda Form insurer to rely upon misrepresentations made when the contract was policy was initially purchased.

12.43

78 35 Park Ave v Campagna 48 NY 2d 813 (NY 1979); Bowes & Co v American Druggists Ins Co 467 NYS 2d 202 (App Div 1983); Goldberg v Manufacturesrs Life Ins Co 672 NYS 2d 39 (App Div 1998); Mills v Everest Reinsurance Co 410 F Supp 2d 243, 253 (SDNY 2006); Certain Underwriters at Lloyd’s v Milberg LLP 2009 WL 3241489 (SDNY 30 September 2009); AXA Equitable Life Ins Co. v Malen 986 NYS 2d 132 (App Div 2014); Fin Guar Ins Co v Putnam Advisory Co LLC No 2020 WL 264146 (SDNY 17 January 2020). 79 For the relevance of the substance of the allegation, see Mills v Everest (above n 78) at 253. See also Kaufman v Cohen 760 NYS 2d 157 (App Div 2003). 80 In re World Com Inc Sec Litig 354 F Supp 2d at 465. 81 Federal v Tyco 92 NYS2d at 401–402. 82 See by analogy cases concerning reformation (in English terminology, rectification) of a contract where s 213(6) CPLR provides for a six-year period for an action based upon by mistake. In Wallace v 600 Partners Co 86 NY 2d 543, 547 (NY 1995), the tenant’s defence based upon a claim for reformation was held to be time barred. See also Commerce & Indus Ins Co v Gun Hill Mgmt Inc No 10147910, 2010 WL 4530221 (Sup Ct 2010 (where the defendant policyholder’s answer and counterclaim seeking reformation was held to be time barred)).

13 Waiver and Estoppel and Reservations of Rights INTRODUCTION 13.01

New York law provides two related common law defences – waiver and estoppel – that may operate to preclude an insurance company from advancing a defence to a policyholder’s claim. Waiver is a voluntary and intentional relinquishment of a known right. In Albert J Schiff Associates Inc v Flack,1 the New York Court of Appeals said that the doctrine evolved because of courts’ disfavor of forfeitures of the insured’s coverage which would otherwise result where an insured breached a policy condition, as for instance, failure to give timely notice of a loss or failure to co-operate with the insured.2

Courts accordingly find waiver where there is direct or circumstantial proof that the insurer intended to abandon the defence.3 By contrast, equitable estoppel arises where the insurer has taken some action upon which the policyholder has relied to its detriment; for example, where an insurer, without asserting policy defences or reserving the right to do so, undertakes the defence of the case against the policyholder, in reliance on which the policyholder suffers the detriment of losing the right to control its own defence.4 13.02

The approach of New York law is similar in many respects to that of English law, which has drawn similar distinctions between waiver (in the sense of election or affirmation) and estoppel.5 Perhaps the most significant difference is that, under New York law, unlike English law, waiver does not require actual knowledge of the facts entitling the insurer to take the relevant course of conduct (for example, to rescind the insurance policy for misrepresentation). Constructive knowledge is sufficient.6

13.03

In addition to these common law principles, section 3420(d) of the New York Insurance Law provides a statutory requirement that the insurer disclaim coverage within a

1 Albert J Schiff Associates Inc v Flack 435 NYS2d 972 (NY 1980). 2 Ibid at 975. 3 Ibid. 4 Ibid. 5 See, eg, Motor Oil Hellas (Corinth) Refineries SA v Shipping Corp of India (The Kanchenjunga) [1990] 1 Lloyd’s Rep 391 (HL). The English cases, particularly the older ones, do not always use the terminology rigidly. For example, an estoppel is sometimes referred to as a waiver. 6 Luria Bros & Co v Alliance Assur Co 780 F2d 1082 (2d Cir 1986) (applying New York law).

Introduction

279

reasonable time in cases involving death or bodily injury arising out of a motor vehicle accident or any other type of accident occurring in New York.7 A disclaimer is an outright denial of coverage.8 There is a well-established distinction under New York law between a letter which reserves rights and a letter which denies or disclaims liability.9 A failure by the insurer to disclaim within a reasonable time will preclude effective disclaimer or denial.10 Section 3420(d) only applies to a liability policy issued or delivered in New York. The New York Court of Appeals has, however, given a wide interpretation to the meaning of ‘issued or delivered’ in the context of section 3420. The section therefore applies where both insureds and risks are located in New York.11 This particular requirement of the statute may well be fulfilled in the case of many Bermuda Form policies, since they often cover a large US company with a substantial business presence in New York which creates risks in that state.12 However, the section may well, for other reasons, be inapplicable to many, if not all, Bermuda Form claims. First, claims under Bermuda Form policies will frequently involve claims in respect of property damage, as to which the section also has no application.13 Second, Bermuda Form policyholders typically operate throughout the United States or in many countries, and the claim for indemnity for a personal injury claim is unlikely to concern, to any significant degree, an accident occurring in New York. It is unclear whether or how the section would apply in the context of a mass tort giving rise to the integration of injuries to numerous people across the United States, but

7 Section

3420(d) states:

If under a liability policy issued or delivered in this state, an insurer shall disclaim liability or deny coverage for death or bodily injury arising out of a motor vehicle accident or any type of accident occurring within this state, it shall give written notice as soon as is reasonably possible of such disclaimer of liability or denial of coverage to the insured and the injured person or any other claimant. Courts have considered this statutory requirement in addressing arguments that an insurer’s disclaimer was not timely, drawing distinctions between the defences to coverage at issue. The cases cannot always be reconciled; however, the decisions place a judicial gloss on the statutory standard. Courts have applied this standard to preclude an insurer from relying on a disclaimer of coverage of certain defences to coverage; see, eg, Roman Catholic Diocese of Brooklyn v National Union Fire Ins Co of Pittsburgh, Pa 21 NY3d 139, 146–47 (NY 2013) (finding s 3420(d)(2) applies only when the disclaimer raises an exclusion and refusing to apply it to preclude coverage defences where the insurer failed to give the insured notice of self-insured retention (SIR) requirements); QBE Ins Corp v Adjo Contracting Corp 121 AD3d 1064, 1081 (NY App 2014) (estopping insurers from denying coverage based on late notice on the ground that the disclaimer was untimely); Dynatec Consulting, Inc v Burlington Ins Co No 655241, 2019 WL 199768 (Sup Ct 7 January 2019), aff ’d as modified, 184 AD3d 475 (NY App Div 2020) (‘disclaimer pursuant to section 3420(d) is necessary when the denial of coverage is based on a policy exclusion’, citing Worcester Ins Co v Bettenhauser 95 NY2d 185, 188 (NY 2000) (preventing the insurer from relying on exclusion on the ground that the disclaimer was untimely); see also County of Niagara v Netherlands Ins Co 839 F App’x 581, (2d Cir 8 Dec 2020) (unpublished decision) (upholding the insurer’s disclaimer, rejecting the policyholder argument that the fact that the insurer improperly addressed disclaimer should lead to waiver). 8 Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (New York, Wolters Kluwer, 1999 and 2021 Supp) (hereafter Masters and Stanzler, Insurance Coverage Litigation) s 3.06; Barry R Ostrager and Thomas R Newman, Handbook on Insurance Coverage Disputes, 19th edn (New York, Wolters Kluwer, 2019) (hereafter Ostrager and Newman, Handbook on Insurance Coverage Disputes) s 2.05. 9 Allstate Ins Co v Gross 27 NY 2d 263 (NY 1970); Hartford Ins Co v County of Nassau 46 NY 2d 1028 (NY 1979). 10 Hartford Ins Co v County of Nassau 46 NY 2d 1028, 1029 (NY 1979). 11 Carlson v American Int’l Grp Inc 30 NY 3d 288 (NY 2017). 12 Ibid at 306; Vista Eng’g Corp v Everest Indem Ins Co 170 AD 3d 915 (NY App Div 2019). 13 Keyspan Gas E Corp v Munich Reinsurance Am Inc 23 NY 3d 583 (NY 2014).

13.04

280

Waiver and Estoppel and Reservations of Rights

where only a proportion were in New York.14 Third, there are arguments as to whether this section ‘pertains to regulation’, so that it falls outside the scope of Article VI(O), the Bermuda Form governing law clause.15 13.05

When considering the New York case law in the context of a Bermuda Form dispute, it is therefore necessary to distinguish between cases applying section 3420(d) (or its predecessor section 167, subd 8 of the Insurance Law), and cases applying common law principles of waiver and estoppel.16

13.06

It may also be useful to bear in mind, when reading New York cases, that they typically concern policies where, unlike the Bermuda Form, the insurer has a duty to defend. Frequently, an insurer will advise the policyholder that it will provide a defence subject to a reservation of its right to deny coverage, and will therefore not disclaim coverage at that stage, but may only do so later. Issues of waiver and estoppel will often therefore fall to be resolved by reference to the terms in which the insurer has expressed its initial reservations of rights, or its ultimate disclaimer, or both. Questions may arise as to whether an insurer has waived reliance upon defences which were not identified with specificity in its disclaimer letter, and this may involve consideration of the terms of any further reservation of rights contained in the insurer’s disclaimer letter.

WAIVER General Principles 13.07

The court in State of New York v AMRO Realty Corp explained the distinction in New York law between a direct and an implied waiver: waiver may be found where there is direct or circumstantial proof that the insurer intended to abandon the defence … [An] implied waiver exists when there is an intention to waive unexpressed but clearly to be inferred from circumstances.17

14 The decision in TIG Insurance Co v Town of Cheektowaga 142 F Supp 2d 343, 364 (WDNY 2001) indicates that the statute may apply in relation to one aspect of the policyholder’s claim (in respect of personal injuries) even though it is inapplicable to another (property damage). However, where numerous injuries or alleged injuries have been aggregated so as to reach an attachment point, it is not easy to see how they could be disaggregated for the purposes of the application of the statutory requirement to disclaim within a reasonable time. 15 See the discussion of this issue in the context of another aspect of s 3420 in ch 8 above, para 8.17. 16 For more detailed consideration of the insurer’s obligation to give a timely disclaimer of coverage, see Masters and Stanzler, Insurance Coverage Litigation (above n 8) s 3.06; Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 8) s 2.05. In the case of an excess insurer, the duty to disclaim does not arise until the primary insurance coverage has been exhausted or the retention met: see Gardner v Ryder Truck Rental Inc 690 NYS2d 614, 616 (App Div 1999); AllCity Ins Co v Sioukas 378 NYS2d 711, 712 (App Div 1976), aff ’d, 393 NYS2d 993 (NY 1977). Although s 3420(d) may have limited application in the context of Bermuda Form disputes, a failure to disclaim coverage within a reasonable time may warrant estoppel on common law grounds: see Progressive Cas Inc Co v Featherstone Foods Inc No 18 Civ 7923 (AKH), 2019 WL 8333518 (SDNY 1 August 2019); United Spec Ins Co v Premier Contracting of NY, Inc 375 F Supp 3d 389, 398 (SDNY 2019); see also Frazier v Royal Ins Co 110 F Supp 2d 110 (NDNY 2000), also discussed in para 13.24 below. Disclaimer on one ground but not another may bring the doctrine of waiver as a matter of law into operation, as discussed in paras 13.16–13.18 below. 17 State of NY v AMRO Realty Corp 936 F 2d 1420, 1431 (2d Cir 1991).

Waiver

281

Thus, the waiver doctrine under New York law looks solely to acts by the insurance company and does not require the policyholder to show prejudice. The relevant conduct of the insurance company may consist of affirmative acts or conduct (such as the renewal of the insurance policy despite knowledge of a misrepresentation defence), or omissions such as a failure to reserve rights or disclaim coverage. Often, when the policyholder can properly allege waiver, the insurer’s conduct consists of a combination of affirmative acts and omissions. Waiver should not, however, be ‘lightly presumed’.18 Courts have held that a policyholder cannot use the waiver doctrine to create coverage that does not otherwise exist under the insurance policy. Thus, the New York Court of Appeals in Albert J Schiff Associates Inc v Flack19 held that the waiver doctrine did not apply because the policyholder’s claim fell outside the scope of coverage.20 Accordingly, the waiver doctrine may be of no application to many Bermuda Form disputes, where the relevant issue is the scope of coverage. However, the doctrine is potentially applicable to defences such as late notice,21 breach of the duty to co-operate22 and failure to obtain approval to settlements.23 None of these defences relate to the scope of coverage; rather, they seek to nullify the coverage provided by the policy.

13.08

Courts have also applied the waiver doctrine to preclude an insurance company’s reliance upon a defence of misrepresentation. As the court in Securities & Exchange Commission v Credit Bancorp Ltd24 held, a misrepresentation or non-disclosure defence does not relate to the scope of coverage, but rather to contract formation. Thus, the proper question is: was the insurance policy ever validly entered into?25

13.09

In order to establish a waiver, the policyholder must show under New York law that the insurance company had actual or constructive notice of the facts supporting the defence. That constructive notice is sufficient is established by, amongst other cases, State of

13.10

18 Gilbert Frank Corp v Federal Ins Co 525 NYS2d 793 (NY 1988); Chicago Ins Co v Kreitzer & Vogelman 265 F Supp 2d 335, 343 (SDNY 2003). 19 435 NYS2d 972 (NY 1980). 20 Ibid at 975. Juliano v Health Maint Org of NJ 221 F3d 279 (2d Cir 2000) (applying New York law); compare Lauder v First Unum Life Ins Co 284 F3d 375, 382 (2d Cir 2002) (applying New York law). Similarly, a failure timely to disclaim will not create coverage that is not otherwise provided for in the policy: TIG Ins Co v Town of Cheektowaga 142 F Supp 2d 343, 362 (WDNY 2001). 21 See, eg, Mutual Redevelopment Houses Inc v Greater NY Mut Ins Co 611 NYS2d 550, 551–52 (App Div 1994) (late notice defence was waived by the insurance company’s failure to assert it in the disclaimer letter, notwithstanding the letter’s inclusion of statement purporting to reserve ‘all other rights under the applicable policy provisions’); Haslauer v North Country Adirondack Coop Ins Co 654 NYS2d 447, 448 (App Div 1997) (late notice defence was waived by the insurer’s failure to assert it in the disclaimer letter, notwithstanding having mentioned it in a prior reservation of rights letter as a ‘possible justification’ for disclaimer); State of NY v AMRO Realty Corp 936 F 2d 1420 (2d Cir 1991). 22 Albert J Schiff Associates v Flack 51 NY 2d at 698. 23 MBIA Inc v Federal Ins Co 652 F3d 152, 169 (2d Cir 2011) (‘By an insurer’s unreasonable delay, silence or conduct, it can either waive a consent requirement or acquiesce in a settlement’). The principle that a policyholder need not seek the insurer’s consent to a settlement after a denial of liability (see above ch 5, paras 5.36–5.37) has sometimes been expressed on the basis of waiver: see Insurance Co of N Am v Lindsey 372 NYS2d 164, 166 (Sup Ct 1975). 24 Securities & Exchange Comm’n v Credit Bancorp Ltd 147 F Supp 2d 238, 258 (SDNY 2001). See above ch 12. 25 As discussed in ch 12, paras 12.29–12.37 above, misrepresentation goes to the very validity of the insurance policy.

282

Waiver and Estoppel and Reservations of Rights

New York v AMRO Realty Corp,26 Luria Bros & Co v Alliance Assurance Co27 and Securities & Exchange Commission v Credit Bancorp Ltd.28 If, for example, the question of waiver arises in the context of a misrepresentation issue, the question will involve consideration of when the insurance company first received sufficient indications of a misrepresentation that would have led a prudent person to inquire further about the matter.29 13.11

If the insurance company had such knowledge, then the question arises as to whether the conclusion is justified that the insurer intended to abandon or not insist upon the particular defence later relied upon. Evidence that the insurer subjectively intended to abandon the particular defence would no doubt be admissible. However, the policyholder more usually seeks to establish waiver by looking at the objective conduct of the insurer, and it is not necessary to establish subjective intent. This can be seen in the cases, discussed below, where the New York courts have found a waiver as a matter of law.30 Thus, courts have treated affirmative conduct by insurance companies that is incompatible with a defence of misrepresentation or fraudulent non-disclosure as establishing a waiver, whatever the insurer’s subjective intentions. Similarly, where liability has been denied on a specified ground, the insurer has been prevented from shifting its disclaimer to another ground known at the time of the original disclaimer. Waiver of Defences Concerning the Validity of the Policy

13.12

In CNA Reinsurance of London Ltd v Home Insurance Co,31 the court held that ‘an insurer’s renewal of an insurance policy with knowledge of the policyholder’s prior misrepresentation or breach operates as a waiver of any claim by the insurer that the policyholder’s insurance is void due to a misrepresentation at the time of purchase or similar breach’. The court set out the following statement of New York law in explaining that an insurance company’s renewal of an insurance policy after it had knowledge of facts supporting misrepresentation justified a finding of waiver: A ratification arises ‘when a party to a voidable contract accepts benefits flowing from the contract, or remains silent, or acquiesces in [the] contract for any considerable length of time after he has opportunity to annul or void the contract’. A party who voluntarily renews a contract despite a belief that he has been fraudulently induced to enter into the agreement is deemed to have waived any claim based on the alleged fraud.32

26 936 F2d 1420, 1431 (2d Cir 1991) (applying New York law). 27 780 F2d 1082, 1090–91 (2d Cir 1986) (applying New York law). 28 147 F Supp 2d 238, 256 (SDNY 2001). 29 147 F Supp 2d at 256; see also City of Utica v Genesee Mgmt Inc 934 F Supp 510, 521 (NDNY 1996); First Penn Banking & Trust Co v US Life Ins Co 421 F2d 959, 963 (3d Cir 1969). 30 See below paras 13.16–13.18. Whilst the cases generally take an objective approach to waiver, some cases contain language suggestive of a requirement of subjective intent – eg, Sirignano v Chicago Ins Co 192 F Supp 2d 199 (SDNY 2002). 31 CNA Reins of London Ltd v Home Ins Co No 85-CIV 5681, 1990 WL 3231, at *9 (SDNY 1990); see also SEC v Credit Bancorp 147 F Supp 2d at 255–56. 32 1990 WL 3231, at *9 (internal citations omitted).

Waiver

283

Similarly, in SEC v Credit Bancorp, the insurers had retained premiums, negotiated endorsements, redrafted marketing and business documents, and engaged in annual re-signings of the insurance policies. The court, applying New York law, rejected their claim for rescission, finding that ‘they failed to rescind the Policies and thereby ratified the existing coverage and the additional policies which were sold after grounds for rescission were established’.33 It held that a finding of ratification will defeat even a valid claim of misrepresentation where the party seeking to avoid the contract does not take prompt action after discovery of the alleged false statement. When determining ratification, the key factors are ‘whether [the] party silently acquiesced in the contract or rather promptly interposed his objections upon discovering the basis for the claim of rescission’.34

13.13

The insurance company’s acceptance of premiums under the insurance policy after it has knowledge of facts relating to misrepresentation or non-disclosure has long been held to preclude rescission.35 Thus, assuming the requisite knowledge, the insurer will be treated as having ratified or affirmed the insurance policy and therefore to have waived any right to avoid the policy once the insurer accepts the benefits flowing from the contract. In addition, as CNA Reinsurance and SEC v Credit Bancorp show, significant delay by the insurer in taking the point on misrepresentation will be treated as acquiescence. However, courts may conclude that waiver is not established if the insurer’s act is doubtful or equivocal.36

13.14

What happens if an insurance company, with sufficient knowledge to take a misrepresentation point, decides to reserve its rights but at the same time or subsequently extends or renews its insurance policy, or takes the benefit of some other provision of the policy, such as inspecting the books and records of the policyholder? Will the insurance company’s reservation of rights protect it from the consequence that would normally follow from conduct inconsistent with a defence, such as misrepresentation, based upon the invalidity of the policy? The point arose, in the reverse context, in McNaught v Equitable Life Insurance Co.37 There, the policyholder paid premiums (so as to avoid a forfeiture) under a reservation of rights, whilst at the same time alleging that the contract should be treated as rescinded. The New York Appellate Division held that the policyholder’s conduct amounted to a waiver. To prevent this consequence from arising, the insurance company and the policyholder should agree that the payment of premiums does not constitute an affirmation of the contract. The issue has also arisen, in a different context, in England and Australia. The principle is neatly captured in the pithy statement of

13.15

33 147 F Supp 2d at 257–58. 34 Ibid at 256. 35 See the cases cited in Scalia v Equitable Life Assur Soc’y of the US 673 NYS2d 730 (App Div 1998); Continental Ins Co v Helmsley Enters Inc 622 NYS2d 20, 20–21 (App Div 1995). 36 See Chicago Ins Co v Kreitzler & Vogelman 265 F Supp 2d 335, 343–44 (SDNY 2003); Horne v Radiological Health Servs PC 371 NYS2d 948, 961 (Sup Ct 1975), aff ’d, 379 NYS2d (App Div 1976). Horne was not an insurance case, but involved the enforcement of a non-compete clause in an employment contract. The plaintiff’s former employer was alleged to have waived the right to enforce the non-compete clause because the employer did not enforce non-compete clauses with respect to several other employees. 371 NYS2d at 959–60. The court reasoned that ‘[t]he intent [to waive] must be clearly established and cannot be inferred from a doubtful or equivocal act’. Ibid at 961. 37 McNaught v Equitable Life Ins Co 121 NYS 447 (App Div 1910).

284

Waiver and Estoppel and Reservations of Rights

expression of Long Innes J in Haynes v Hurst:38 ‘A man, having eaten his cake, does not still have it, even though he professed to eat it without prejudice.’39 A similar sentiment is to be found in a number of English landlord and tenant cases.40 Waiver Based on Statements in a Reservation of Rights or Disclaimer 13.16

As previously discussed, New York statutory law requires an insurance company to disclaim coverage in writing, and this must be done promptly and with specificity.41 Even if the statute is inapplicable (as may be the case in most Bermuda Form disputes), where an insurance company intends to deny liability under the policy, it will invariably issue a letter to this effect. The fact that there is no statutory requirement to disclaim does not render the waiver doctrine inapplicable.42 Disclaimer letters are frequently preceded by reservation of rights letters. A number of New York cases have considered the waiver doctrine in circumstances where an insurance company has issued either a reservation of rights letter or a disclaimer letter in which a particular defence was not identified. Courts thus have applied the waiver doctrine as a matter of law to reject a defence that the insurance company did not identify in its reservation of rights or disclaimer of coverage.

13.17

Thus, New York law establishes that an insurer is deemed, as a matter of law, to have intended to waive a defence to coverage (i) where other defences are asserted, and (ii) where the insurer possessed sufficient knowledge (actual or constructive) of the circumstances regarding the unasserted defence.43 Thus, in TIG Insurance Co v Town of Cheektowaga,44 the court held that one of the insurance companies, TIG, had waived its right to disclaim based on a late notice defence. It found that TIG’s reservation of rights letter specified only scope of coverage issues and that the insurer had at that time already received information sufficient to alert a reasonable insurer as to the availability of the late notice defence. 38 Haynes v Hurst 27 New South Wales State Reports 480. 39 Ibid at 489. 40 See Matthews v Smallwood [1910] 1 Ch 777, 786–87 (Parker, J); approved by the Court of Appeal in Oak Property v Chapman [1947] 1 KB 886, 898; Central Estates (Belgravia) Ltd v Woolgar [1972] 1 WLR 1048, 1051–52 (Lord Denning MR), 1054 (Buckley LJ). Although a landlord and tenant case, Matthews v Smallwood is nonetheless treated as containing general principles relevant to waiver and it is referred to, for example, in John Birds et al (eds) MacGillivray on Insurance Law, 18th edn (London, Sweet & Maxwell, 2018) paras 10–100 and 10-113. The receipt of premiums, for example, is analogous to the receipt of rent in the cases referred to, and is treated by the English courts as an unequivocal act consistent only with an intention to continue the insurance contract. See Cia Tirrena v Grand Union [1991] 2 Lloyd’s Rep 143. 41 See above para 13.03; General Accident Ins Grp v Cirucci 414 NYS2d 512, 514 (NY 1979). 42 MCI LLC v Rutgers Cas Ins Co No 06 Civ. 4412(THK), 2007 US Dist Lexis 59241, at *23–*29 (SDNY 2007). See also cases interpreting s 3420(d) of the New York Insurance Law: eg, Roman Catholic Diocese 21 NY3d at 146–47 (refusing to apply s 3420(d) to preclude coverage defence). New York courts may estop an insurer from disclaiming coverage after finding failure to comply with s 3420(d): see, eg, QBE Ins v Adjo Contracting 121 AD3d at 1081; Progressive Casualty v Featherstone Foods 2019 WL 8333518, at *3 (granting the insured’s motion for summary judgment and dismissing the insurer’s late notice defence for failure to comply with s 3420(d)). 43 AMRO Realty Corp 936 F2d at 1431; Luria Bros 780 F2d at 1090–91. See also TIG v Town of Cheektowaga 142 F Supp 2d at 362 (applying same principle to defences specified in reservation of rights letter); MCI LLC v Rutgers Cas Co No 06 Civ. 4412(THK), 2007 US Dist Lexis 59241 at *32–*37 (SDNY 2007). 44 TIG Ins Co v Town of Cheektowaga 142 F Supp 2d 343 (WDNY 2001).

Estoppel

285

This principle requires consideration of the terms, if any, in which an insurance company expresses its reservation of rights or disclaimer. Where the disclaimer is accompanied by a reservation of the right to advance other defences, consideration will also need to be given to the precise terms in which the reservation of the right is expressed. In State of New York v AMRO Realty Corp,45 the insurance company’s disclaimer included a general statement that ‘we reserve our rights to rely on additional reasons for disclaimer should they become apparent in the future’. The policyholder argued that the insurer had waived its right to rely on any defence not specifically identified in its reservation of rights, while the insurer contended that its general statement preserved its right to assert any defences not specifically raised in its written reservation. The court struck a middle ground, holding that the insurer’s general statement preserved its right to assert only those defences that were unknown at the time of the reservation. By contrast, in Lugo v AIG Life Insurance Co,46 the insurance company ended its disclaimer with a general statement that ‘issuance of the denial is not to be interpreted as a waiver of any and all other rights and defenses that AIG Life Insurance Company may have under the policy provisions, all of which are hereby expressly reserved’. The court held that this kind of general statement protected an insurer that had not identified defences in its disclaimer, even though they were then known to it.

13.18

ESTOPPEL Estoppel arises when an insurance company acts in a manner inconsistent with a lack of coverage, and the policyholder reasonably relies on those actions to its detriment.47 The distinction between waiver and estoppel was clearly drawn by the New York Court of Appeals in Albert J Schiff Associates Inc v Flack.48 The court illustrated the operation of estoppel by the example of an insurance company, that without asserting policy defenses or reserving the privilege to do so, undertakes the defense of the case, in reliance on which the insured suffers the detriment of losing the right to control its own defense.49

In such circumstances, courts will not hear the insurer say that coverage does not exist, even though it actually does not.50

45 936 F2d 1420, 1431 (2d Cir 1991). See too Mut Redevelopment Houses Inc v Greater NY Mut Ins Co 611 NYS 2d 550 (App Div 1994). 46 Lugo v AIG Life Ins Co 852 F Supp 187 (SDNY 1994). Other cases where the reservation of rights was sufficiently broad to protect the insurer include: Mount Vernon Fire Ins Co v William Monier No 95 Civ 0645, 1996 WL 447747 (SDNY 1996); Heiser v Union Cent Life Ins Co No 94CV179, 1995 WL 355612 (NDNY 1995); Home Décor Furniture & Lighting Inc v United Nat’l Group No 05CV02005, 2006 WL 3694554 (EDNY 2006). 47 See, eg, Masters and Stanzler, Insurance Coverage Litigation (above n 8) s 12.02; Ostrager and Newman, Handbook on Insurance Coverage Disputes (above n 8) s 2.06[b]. 48 51 NY2d 692 (1980). 49 Ibid at 698. 50 See also Marino v New York Tel Co No 88 Civ 5817, 1992 WL 212184, at *8 (SDNY 1992) (quoting Lone Star Indus Inc v Liberty Mut Ins Co 689 F Supp 329, 333 (SDNY 1988)).

13.19

286

Waiver and Estoppel and Reservations of Rights

13.20

Thus, estoppel requires words or conduct by one party on which the other relies to its prejudice, in contrast to waiver, where no showing of prejudice is necessary. Unlike waiver, however, the operation of an estoppel does not depend upon actual or constructive knowledge on the part of the person whose words or conduct give rise to the prejudicial reliance. A further difference is that the impact of an estoppel may in effect widen coverage, whereas, as discussed earlier in this chapter, the doctrine of waiver does not expand coverage.

13.21

Different formulations of the relevant principles can be found in the authorities. The basic principle can be found in the decision of the US Supreme Court in Insurance Co v Wilkinson: It is in precisely such cases that courts of law in modern times have introduced the doctrine of equitable estoppels, or, as it is sometimes called, estoppels in pais. The principle is that where one party has by his representations or his conduct induced the other party to a transaction to give him an advantage which it would be against equity and good conscience for him to assert, he would not in a court of justice be permitted to avail himself of that advantage. And although the cases to which this principle is to be applied are not as well defined as could be wished, the general doctrine is well understood and is applied by courts of law as well as equity where the technical advantage thus obtained is set up and relied on to defeat the ends of justice or establish a dishonest claim.51

13.22

In Nassau Trust Co v Montrose Concrete Products Corp,52 the New York Court of Appeals said that an estoppel rests upon the word or deed of one party upon which another rightfully relies and so relying changes his position to his injury: It is imposed by law in the interest of fairness to prevent the enforcement of rights which would work fraud or injustice upon the person against whom enforcement is sought and who, in justifiable reliance upon the opposing party’s words or conduct, has been misled into acting upon the belief that such enforcement would not be sought.

13.23

In Intelligent Digital Systems LLC v Beazley Insurance Co, the court referred to prior authority and said: The doctrine of equitable estoppel precludes a party at law and in equity from denying or asserting the contrary of any material fact which he has induced another to believe and to act on a particular manner. In this way, equitable estoppel prevents one from denying his own expressed or implied admission which has in good faith been accepted and acted upon by another. Under New York law, equitable estoppel requires a showing of (1) an act constituting a concealment of facts or a false misrepresentation; (2) an intention or expectation that such acts will be relied upon; (3) actual or constructive knowledge of the true facts by the wrongdoers; (4) reliance upon the misrepresentations which causes the innocent party to change its position to its substantial detriment. There does not have to be an actual attempt to mislead; instead, it is sufficient that the party being estopped knew or had reason to believe that their acts or inaction might prejudice the party asserting the estoppel. Moreover, to support an equitable

51 80

US 222, 233 (1872). Trust Co v Montrose Concrete Prods Corp 56 NY2d 175 (NY 1982).

52 Nassau

The ‘Changes’ Condition in the Bermuda Form

287

estoppel argument, a party must show that its detrimental reliance was reasonable under the circumstances.53

The application of estoppel is illustrated by Frazier v Royal Insurance Co.54 There, homeowners, as judgment creditors of their builder, sought coverage from the builder’s CGL insurer for property damage to their home. The insurance company claimed that the insurance policy did not apply, because the homeowners sued the builder under a breach of warranty theory that was not covered under the policy. The court granted the policyholder’s motion for summary judgment, holding that the insurance company was estopped from denying coverage, regardless of whether coverage properly existed or not. The court found that the insurer knew of its ground for disclaimer, but waited more than six months to disclaim coverage to its policyholder (the builder) and more than 21 months to the homeowners. Rather than disclaim coverage, the insurance company exchanged engineering reports, informed the homeowners that it was investigating the matter and offered to settle the claim. These actions led the plaintiff homeowners to rely upon the insurer’s actions confirming coverage for their losses. The court held that the plaintiffs had suffered prejudice during the extended course of correspondence, inspections, exchange of reports, settlement negotiations, arbitration and litigation, during which time they incurred costs and expenses. Relying on representations by the insurer, the plaintiffs also failed to take additional steps such as securing the builder’s assets. The court concluded that the insurance company could not ‘turn a blind eye and keep the plaintiffs in the dark about its position for almost two years and expect to avoid responsibility’.55

13.24

THE ‘CHANGES’ CONDITION IN THE BERMUDA FORM Article VI(J) of the 004 Form is a ‘Changes’ or ‘Non-waiver’ Condition, which provides that: Notice to or knowledge possessed by any person shall not effect waiver or change in any part of this Policy or estop the Company from asserting any right under the terms of this Policy. The terms of this Policy may not be waived or changed, except by written endorsement issued to form a part hereof, signed by the Company.56

Similar clauses have appeared in policies of insurance, particularly liability insurance, for a very long time.57 Decisions concerned with such clauses have frequently involved local insurance agents engaged by insurance companies, and the relevant clauses have specifically referred to notice to agents. Companies writing on the Bermuda Form do not operate through the agents such as those considered in many of the cases.

53 Intelligent Digital Sys LLC v Beazley Ins Co 906 F Supp 2d 80, 94–95 (EDNY 2012) (internal citation omitted). 54 Frazier v Royal Ins Co 110 F Supp 2d 110 (NDNY 2000). 55 Ibid at 116. 56 Article VI(J) of the 004 Form. See also Article V(k) of earlier Forms. 57 Drennan v Sun Indem Co of NY 271 NY 182 (NY 1936), in which even earlier cases were reviewed.

13.25

288

Waiver and Estoppel and Reservations of Rights

13.26

The clause is directed towards the ‘terms of this Policy’, and therefore should have no application where the issue is whether the insurer has waived, or is estopped from asserting, a misrepresentation or non-disclosure defence. Such defences are not concerned with the terms of the insurance policy, but its validity. The New York case law, relating to such clauses, generally addresses situations where the insurer invokes a policy defence, such as late notice, rather than defences that relate to the validity of the insurance policy.

13.27

New York courts have recognised that such clauses are in principle valid and effective,58 but impose limits on the circumstances in which they operate. It is therefore open to the policyholder to prove that the requirement for a written endorsement was itself waived. This will depend upon what was agreed with or conveyed to the policyholder by words or conduct, and the actual or apparent authority of the representative of the insurance company whose words or conduct are relied upon. This will in turn depend upon the rank and authority of the agent, and the subject matter with reference to which he or she assumes to act.59 If, for example, the policyholder is relying upon the words or conduct of a senior executive of the insurance company, a case of waiver may be easier to establish than where (as is often the situation in the case law on this topic) the words or conduct of a local agent has been in issue.

13.28

Where the policyholder is able to invoke an agreement, the clause is unlikely to assist the insurer.60 Similarly, where the policyholder is able to invoke an estoppel, rather than simply waiver, the clause is unlikely to be effective. The clause refers only to ‘notice to or knowledge possessed by any person’. As already explained, the operation of estoppel does not depend upon (or even require) notice to the insurer, but instead requires words or conduct upon which the other party relies to its detriment.

13.29

Courts in New York have held that such clauses are not effective where the elements of an estoppel are present. Thus, in Drennan v Sun Indemnity Co of New York,61 the New York Court of Appeals explained that a course of conduct by an insurer can effect a waiver notwithstanding such a clause: It is true that we have also held that the insurance company may itself waive any contractual condition or restriction in a policy, even the condition that a waiver, extension or change must be in writing. Such waiver may be established by a course of conduct which gives rise to an

58 Bazar v Great Am Indem Co 306 NY 481 (1954); Nothhelfer v American Sur Co of NY 302 NY 910 (1951); Drennan v Sun Indem Co of NY 271 NY 182 (1936); Smith v Zurich Gen Accidental & Liab Ins Co 105 NYS2d 713 (App Div 1951), aff ’d, 303 NY 948 (1952); Mighty Midgets Inc v Centennial Ins Co 403 NYS2d 768, 276 (App Div 1978) (Cutrone J, dissent), aff ’d, 416 NYS2d 559 (NY 1979); General Accident Fire & Life Assur Co v Bongiorno 161 NYS2d 551 (Sup Ct 1957), aff ’d, 177 NYS2d 1019 (App Div 1958), aff ’d, 201 NYS2d 778 (NY 1960). For the decision of the New York Court of Appeals in Mighty Midgets, see 416 NYS2d 559. See also, in the context of leases, NL Indus Inc v Painewebber Inc 720 F Supp 293 (SDNY 1989); Excel Graphics Techs Inc v CFG/ AGSCB 75 Ninth Avenue LLC 767 NYS2d 99 (App Div 2003). 59 Drennan v Sun Indem Co of NY 271 NY 182 (NY 1936); Biloz v Tioga Cty Patrons’ Fire Relief Ass’n 21 NYS2d 643, 648–49 (Sup Ct 1940); Reese v Schneider 201 NYS2d 857 (Sup Ct 1960). See too Thomas J Sette Constr Corp v National Fire Ins Co of Hartford 196 NYS2d 144 (Sup Ct 1959); Trend Export Funding Corp v Foreign Credit Ins Ass’n 670 F Supp 480, 484 (SDNY 1987); Collins v Isaksen 633 NYS2d 539, 540–41 (App Div 1995). 60 Reese v Schneider 201 NYS2d at 859. 61 271 NY 182 (NY 1936) (internal citations omitted).

The ‘Changes’ Condition in the Bermuda Form

289

estoppel or by word or act of a duly authorized agent. That does not mean that the provisions of a policy limiting the powers of an agent, or restricting the manner in which the policy may be extended or its provisions waived, are not given their intended effect. Such provisions are intended to be effectual unless or until the parties agree otherwise and the company can be bound by an agreement or waiver only where such agreement is made by an agent who is authorized to make it.62

Similarly, in Merchants Mutual Casualty Co v Wildman,63 the dissenting judge (Christ J) referred to the non-waiver provisions of the insurance policies and stated the relevant principles, emphasising that an estoppel may arise if the insurer’s conduct has caused the policyholder to rely to its detriment:

13.30

Plaintiff is entitled to enforcement of the provisions of both Condition 23 and Condition 5 unless its conduct has been such as to estop it from such enforcement. To give rise to an estoppel, however, plaintiff’s conduct must have misled defendant to change his position to his prejudice. In order to succeed, one who invokes the doctrine of estoppel against another must first show that, in reliance on the other’s conduct, he has changed his position and would be defrauded or otherwise prejudiced unless the other person be held to be estopped.64

There may be other arguments which could not effectively be met by reliance on the Changes Condition. For example where the issue concerns the absence of insurer approval to the policyholder’s settlements, the policyholder may rely upon various theories, such as acquiescence or breach of the duty of good faith and fair dealing. These arguments depend upon facts which are not confined to ‘notice to or knowledge possessed by’ the insurer.

62 Ibid at 187. 63 Merchants Mut Cas Co v Wildman 209 NYS2d 242 (App Div 1960). The judgment of the majority is brief, and refers to both waiver and estoppel. 64 Ibid at 245–46 (internal citations omitted).

13.31

14 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators INTRODUCTION 14.01

This chapter considers the practical aspects of initiating a Bermuda Form dispute in London. In Chapter 15 we consider the practical aspects of the arbitration once it is underway, but this chapter focuses on the first stages of the arbitral process. Who should be appointed as legal representatives? How should the arbitration be commenced? Whom should a party appoint as its party-appointed arbitrator, and how should the third arbitrator, who is the chair of the Bermuda Form arbitral panel,1 be selected? On what grounds can objections be made to an appointment?

LEGAL REPRESENTATION 14.02

When a company, either policyholder or insurer, becomes involved in a potential claim under the Bermuda Form, legal advice is likely to be required well before arbitration proceedings are actually commenced. The commencement of arbitration proceedings is often a last resort when the process of correspondence and negotiation between policyholder and insurer has failed to resolve the claim. A typical large corporate policyholder based in the United States will have a number of lawyers to whom it is accustomed to turn for advice. In addition to its in-house legal team, a corporation involved in a significant liability problem will invariably be represented by one or more firms of US lawyers handling the underlying claims against the corporation. The corporation also may seek advice on insurance coverage issues, and this advice may or may not be available within the law firms defending the underlying claims.

14.03

The policyholder’s initial correspondence and meetings with the insurance company or its representatives will therefore nearly always be handled by its US lawyers. This is appropriate, since the law governing the Bermuda Form contract is New York law, and the policyholder is unlikely to engage English lawyers to advise on coverage issues and

1 Section 16(5) of the English Arbitration Act 1996 provides that, where there are three arbitrators, the third arbitrator appointed by the other two arbitrators is the chair.

Legal Representation

291

correspondence and negotiations with insurers in the early stages of a potential claim, when the claim may never crystallise into a dispute that requires arbitration. Similarly, the insurance company is likely to be receiving advice from its US lawyers on the claim. A large insurer writing on the Bermuda Form will usually have a number of US law firms to which it turns for advice on potential claims. It will also have experienced claims managers and (often) in-house lawyers. If the parties conclude that they cannot resolve the claim amicably without arbitration, the dynamic may then change. It is at that stage that the claim acquires, or is more clearly seen to acquire, an international element. The arbitration proceedings will not be held in the United States. The majority of Bermuda Form insurance policies require arbitration of disputes in London. Some Bermuda Form policies require arbitration in Bermuda, a country whose legal system is much closer to that of England than that of the United States. This chapter assumes that the arbitration will be held in London rather than Bermuda, but what is said is also broadly applicable to a Bermuda-based arbitration.

14.04

If it appears that arbitration proceedings may be required, each party will consider engaging English lawyers to represent its interests. As discussed further below, selection of the tribunal (called ‘Board of Arbitration’ in the 004 Form) is a crucial strategic decision and each party will also need to consider whether to appoint an English or US arbitrator, or even an arbitrator from another country.2 The composition of the tribunal will also affect the party’s decision as to who should represent it, and will probably have a greater impact than the fact that London is the seat of the arbitration. An arbitration taking place in London before a tribunal composed entirely of US lawyers3 gives rise to a dynamic that differs from an arbitration where one or more English lawyers are appointed to the tribunal. In the latter case, a party may wish to have English lawyers representing them. The question specifically may be whether the policyholder should hire a barrister (an English lawyer trained to conduct trials) or a Queen’s Counsel (QC).4 Another factor that tends, rightly or wrongly, to influence a party’s choice is how the other party has chosen to be represented. If the majority of the tribunal is English, and one side has engaged English counsel, there is inevitably a pressure on the other party to engage English counsel as well.

14.05

2 It is uncommon for a party to appoint an arbitrator from continental Europe for a Bermuda Form case. But it is not unknown for arbitrators from other common law jurisdictions, such as Canada, Australia or South Africa, to be appointed. 3 This would be unusual in Bermuda Form arbitrations. Insurance companies such as ACE (now Chubb) and XL (now Axa XL) have usually appointed senior English barristers (QCs) or retired judges as their arbitrators. Even where both policyholder and insurer have appointed US lawyers, the fact that the arbitration is to be held in London, with English law governing the arbitration procedure, often results in the appointment of an English chair. 4 ‘QC’ stands for ‘Queen’s Counsel’, which is a rank conferred on selected senior lawyers, who receive ‘letters patent’ granted by the Queen. The senior lawyers so chosen have, with relatively few exceptions, been barristers; ie, the branch of the English legal profession that specialises in trial (and arbitration) advocacy. Approximately 10 per cent of English barristers are qualified as QCs. In 1995 the system was changed so that solicitors as well as barristers could be appointed as a QC, and a number of solicitors – including some who practise frequently or exclusively in arbitration rather than litigation – are now appointed each year. Prior to 2005, QCs were appointed by the Lord Chancellor on the advice of the government. Since 2005, the process of selecting QCs has been the responsibility of a selection panel independent of the government and the governing bodies of the legal profession after an extensive process including application, confidential assessments from judges, colleagues at the Bar and clients, and a comprehensive interview process.

292 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

14.06

Accordingly, a party has a choice. Is the company to be represented in the hearing exclusively by its US lawyers? Should it instruct English lawyers instead of the US lawyers? Or should the company use both US and English lawyers? If so, should the company’s US insurance coverage lawyers run the case, with an English lawyer assisting? If the company chooses to instruct English lawyers, should these be solicitors or barristers (or both), the English legal profession being divided into these two branches? If the company uses both its US and English lawyers, which lawyers will determine the strategy or be the ultimate decision-makers? As an additional wrinkle, should the company hire a QC? Should the company hire both an English solicitor firm and a QC or other barristers? The company may choose to have its US lawyers be the ultimate decisionmakers about strategy on the case, as the company may feel that the US lawyers may have a better understanding of the company’s business, the American legal system and the historical development of the underlying liability claims. However, the company may decide (or its American counsel may advise) that the lead counsel role in the final trial or hearing be entrusted to an English trial lawyer, typically a barrister. If the company decides to follow this latter course, the English counsel’s role in strategy may well increase as issues relating to trial strategy arise. A good working relationship between the company’s American and English lawyers is the ideal, and is a goal that the client can encourage by making the roles of the lawyers clear at the time that it engages them. If the company’s English lawyers are to handle the trial, other strategic questions arise. For example, how are issues of New York law going to be handled at the hearing? Should a New York insurance coverage lawyer become part of the team to advise and argue points of substantive New York law?

14.07

A party’s decision on legal representation is to a large extent a matter of taste and tactics. There are no rules that apply and no real principles which guide. Our experience is that there is no universal approach to these questions. Very frequently, policyholders and insurers have used a combination of US lawyers and English barristers. (The concept of ‘US lawyers’ is increasingly broad, with many US firms having offices in London with substantial litigation and arbitration expertise.) Sometimes, however, the US lawyers have handled matters exclusively. Sometimes clients have followed what might be regarded as the traditional route of instructing English solicitors and barristers, with a US insurance coverage lawyer assisting them and arguing relevant points of New York (or other US)5 law. For the reasons that follow, we think that the combination of US lawyers and English barristers can work very well. The US lawyers can carry out the day-to-day pre-trial work, with advice being given as necessary by English barristers. Advocacy at the interlocutory and substantive hearings could be conducted either by the English barrister or a US trial attorney, or a combination of both. There may, however, be much to be said in a particular case for one of the other options.

14.08

Should the fact that the tribunal has English members, or that English lawyers have been engaged by the other party, lead a party to appoint English lawyers? It would be regrettable if a US client or its US lawyers were to feel compelled to instruct English lawyers for fear of being ‘home-towned’ by a predominantly English panel in circumstances

5 On

occasion, points of law relevant to the underlying cases or liability may be argued.

Legal Representation

293

where the opposing party has engaged English lawyers. Many English arbitrators are sympathetic to this view. Indeed, it should be remembered that many people in London – lawyers, arbitrators, arbitral bodies and trade organisations – seek to promote London as an international arbitration centre. These efforts have been very successful. English arbitral law was modernised in the shape of the 1996 Arbitration Act, which aimed amongst other things to make arbitration law more accessible to potential overseas users and to demonstrate that English arbitral procedure was up to date. The concept of an international arbitration centre ought in our view to carry with it the connotation not only that London is a place for arbitrations between parties based in different jurisdictions, but also that it is a place where foreign lawyers, and their clients, should feel comfortable in conducting arbitrations.6 Apart from the obvious fact that New York substantive law governs the interpretation of the Bermuda Form, other factors support the decision to use US lawyers for much of the trial preparation work typically performed by English or other solicitors, certainly in the early stages of the case preparation. If the US lawyers are already familiar with the facts of the case and the documentation (for example, as a result of having advised in the context of the underlying litigation or in relation to claims against other insurers), they may find it easier to draft the pleadings than a newly instructed English lawyer who must learn the case from scratch. (However, if English lawyers are to be involved, it will be sensible to obtain their views on the pleadings, particularly as to how the pleadings will be viewed by any English members of the tribunal.) The client’s documents may be located principally in the United States, and it may make sense for the document production phase to be conducted by lawyers there. The witnesses, many of them of them at least, are likely to reside in the United States, and the preparation of their witness statements (which is now an essential phase in London arbitration) may be more conveniently, and more efficiently, done by the client’s US lawyers. The case may involve detailed, technical evidence, and the client’s in-house and perhaps external experts may be more accessible to the US lawyers. More generally, a great deal of the effort in arbitration, as in litigation, comes at the interlocutory stages. A considerable amount of time and money is spent on document production, correspondence, and the preparation of pleadings and witness statements. For most of the time, the lawyer is therefore either dealing with his or her own client, or is dealing with the opposing firm of attorneys and interacting with them. The tribunal itself will often have only a relatively small involvement in the whole process that takes place prior to the actual hearing, although the work carried out by the lawyers will obviously be shaped by strategic decisions about the hearing that is going to take place. The majority of the preparatory work is unlikely to differ greatly from the work that an American attorney typically carries out in litigation and arbitration at home.

14.09

In addition to these logistical considerations, other human factors come into play. The client may be used to dealing with a particular American firm, or its lawyers. The client’s executives or in-house counsel may have a long history of dealing with the US attorneys, who may in turn have an in-depth substantive knowledge of the industry and

14.10

6 On the subject of US lawyers conducting arbitrations in London, see generally Jeff Dasteel and Richard Jacobs, ‘American Werewolves in London’ (2002) 18 Arbitration International 165.

294 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

the client – the client’s personnel, procedures, practices and business generally – that an English lawyer would be unable to match. The US attorneys may have been involved in defending the policyholder in the underlying litigation, and may have acquired a deep understanding of technical issues that often arise when defending the kind of underlying claim (whether ‘boom’ or ‘batch’) that gives rise to a Bermuda Form insurance claim. The technical issues involved in the underlying claim often repeat themselves, in a modified form, in the insurance claim; for example, if the insurer relies upon an ‘expected or intended’ defence. The client may have confidence in the American attorney’s advocacy skills from past successes, and there may also be a compatibility of style and a long-term personal relationship and trust between them. The client is likely to want its American attorneys to influence the strategy and handling of the case in any event. 14.11

Accordingly, there is no reason in principle why US lawyers should not conduct a Bermuda Form arbitration in London. A good American trial attorney will have the skills and resources necessary to carry out both the pre-hearing work and the advocacy at the hearing.

14.12

Nevertheless, we think that in many cases it will be sensible and advisable to employ an English lawyer as part of the team. This is not simply because a number of English lawyers have acquired considerable experience in handling and trying Bermuda Form arbitrations. In addition, a Bermuda Form arbitration tribunal will often include one or more English barristers or retired judges, and an experienced English trial lawyer will typically have a better feel as to how the English members of the panel will expect the case to be handled and presented, and what will sway their ultimate decision. An English lawyer should also have an understanding of the instincts of the English members of the panel. Those instincts will necessarily have been shaped by both the substantive law and arbitration practice with which those arbitrators have experience. An insight into the tribunal’s procedural instincts may be important, as it may inform a party’s decisions as to which procedural applications to make and which to resist. An insight into the tribunal’s ‘substantive law’ instincts may be even more important, because those instincts may be entirely inappropriate when the substantive law of New York in certain circumstances takes, as it does, a different substantive approach than English law. The English members of the tribunal may subconsciously start from the notion that New York law is likely to be much the same as English law. Whilst this may be true on many questions, there may be critical issues where this misconception needs to be identified and corrected. An English lawyer is more likely to understand these fault lines between New York law and English law, with the help of the client’s American counsel, than would an US lawyer acting alone.

14.13

Ideally, the English lawyer will have experience of handling insurance disputes and be familiar with arbitration practice. Both aspects are important. Arbitrations (including Bermuda Form arbitrations) can give rise to difficult procedural issues,7 although the

7 Document production can often be a problematic area in Bermuda Form arbitrations, and questions can arise as to the volume of documents to be disclosed and questions of privilege. This topic is addressed s eparately in ch 16.

The Commencement of Arbitration

295

resolution of these issues is very unlikely to require intimate knowledge of some aspect of arbitration procedure and practice that is outside the experience of an experienced English trial lawyer specialising in commercial law. It should be noted, in this context, that English law relating to arbitration is now much more accessible as a result of the Arbitration Act 1996. Before that statute, the principles of arbitration procedure had to be sought in part in earlier legislation, but principally in the diffuse case law that had developed over decades.8 The 1996 Act is a ‘user-friendly’ statute, and the report of the Departmental Advisory Committee, which preceded the Act, serves as an excellent commentary on the statute itself.9 An English lawyer should also be able to identify the less tangible cultural differences between English and US arbitration procedure that would not emerge from a scrutiny of textbooks and authorities. One of the keys to success in any arbitration and litigation is for an attorney to appear credible in the eyes of the tribunal. A misconceived application, a tactic that does not work or a strategy that annoys the tribunal can all result in a loss of credibility in the eyes of the tribunal, or, worse still, make the tribunal mistrust or be suspicious of what the lawyer is saying or doing. Credibility, once lost, can be difficult to regain. Suspicion, once created, can take a long time to dispel. An English lawyer assisting the US counsel ought to be able to advise as to what works and does not work in an English arbitration, as well as what is likely to appeal to the tribunal and what is likely to go down badly.

14.14

THE COMMENCEMENT OF ARBITRATION Appointment Procedures and Default Appointments English arbitration law provides that the parties are free to agree on the procedure for appointing the arbitrator or arbitrators, including the procedure for appointing any chair or umpire.10 The arbitration provision of the Bermuda Form11 sets out the agreed procedure governing the conduct of the arbitration in some detail. A party may commence an arbitration under a Bermuda Form policy by giving notice to the other party invoking the arbitration provision in the policy. The Appendix to this chapter contains a pro forma notice. The commencement of a Bermuda Form arbitration raises a number of issues, which are discussed further in this section.

8 It was not until the publication in 1982 of Michael J Mustill and Stewart C Boyd, Commercial Arbitration (London, Butterworths, 1982) that a comprehensive work on modern English arbitration law and practice in commercial arbitrations was available. 9 This is reproduced in many textbooks; see, eg, Michael J Mustill and Stewart C Boyd, Commercial Arbitration 2001 Companion (London, Butterworths, 2001) Appendix 1. There is now a substantial body of case law on the 1996 Act; see, eg, Robert M Merkin, Arbitration Law (Looseleaf, London, Informa, 2010 and Supps), which contains a detailed and up-to-date discussion. See the Bibliography for a description of other textbooks on arbitration law. 10 Arbitration Act 1996, s 16(1). Since London is the seat of the arbitration, the English Arbitration Act applies: see above ch 3, in particular para 3.12. Under s 16(5) of the Arbitration Act 1996, the third arbitrator in a Bermuda Form case is the chair rather than an ‘umpire’. 11 Condition (o) in the original 001 Form; and Condition N in later Forms.

14.15

296 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

14.16

The procedure for initiating arbitration in Form 004 is as follows:12 Any party may, in the event of such a dispute, controversy or claim, notify the other party or parties to such dispute, controversy or claim of its desire to arbitrate the matter, and, at the time of such notification the party desiring arbitration shall notify any other party or parties of the name of the arbitrator selected by it. The other party who has been so notified shall within (30) calendar days thereafter select an arbitrator and notify the party desiring arbitration of the name of such second arbitrator … The two parties, chosen as above provided, shall within (30) calendar days after the appointment of the second arbitrator choose a third arbitrator.13

14.17

Accordingly, the notice beginning the arbitration needs to identify the ‘matter’ which is to be arbitrated. It is advisable for the notice to identify the ‘matter’ in reasonably wide terms, so as to avoid any later contention that a particular claim or issue is outside the jurisdiction of the tribunal because it falls outside the scope of the subject matter of the original appointment. The notice also needs to specify the person chosen as arbitrator, as shown in the pro forma notice set out at the end of this chapter.14 The tactical and other considerations which relate to this choice of arbitrator are discussed below.

14.18

The arbitration provision also contains a mechanism for default appointments, namely an application to the High Court of Justice of England and Wales.15 Any such application would be made to the Commercial Court in London. We have not heard of a case where one of the parties has refused to nominate its arbitrator, leading to a default appointment by the court. Given the sums at stake in Bermuda Form arbitrations, it is unlikely that either side will want any part of the proceedings to go by default.

14.19

There have, however, been cases where the party-appointed arbitrators have not been able to agree, and in which the court has been asked to make an appointment. The 004 Form provides for appointment of the third arbitrator in the following terms: The two arbitrators, chosen as above provided, shall within thirty (30) calendar days after the appointment of the second arbitrator choose a third arbitrator. In the event of the failure of the first two arbitrators to agree on a third arbitrator within said thirty (30) calendar day period, either of the parties may within a period of thirty (30) calendar days thereafter, after notice to the other party or parties, apply to a judge of the High Court of Justice of England and Wales for the appointment of a third arbitrator and in such case the person so appointed shall be deemed and shall act as the third arbitrator. Upon acceptance of the appointment by said third arbitrator, the Board of Arbitration for the controversy in question shall be deemed fixed.

14.20

There is often some jockeying for position in relation to the third arbitral appointment. The process of agreement of the third arbitrator and the possibility of an application to the court give rise to a number of issues. 12 Earlier forms contained a requirement of ‘claim’ or ‘demand’ which had been denied or not met for a period of 20 days – a requirement that was fairly easy to satisfy in practice. Although the 004 Form contains no such requirement, so in practice the dispute will almost invariably be preceded by such correspondence. 13 Article VI(N) of the 004 Form. 14 There is nothing to prevent the parties from agreeing to carry out the early stages of the arbitration w ithout actually appointing their arbitrators; for example, exchanging pleadings and agreeing on a timetable for appointment thereafter. 15 Arbitration Act 1996, ss 17 and 18.

The Commencement of Arbitration

297

First, there is the question of whether the parties can or should express their prefer- 14.21 ences to the party-appointed arbitrators. It is a firm rule of English arbitral practice that arbitrators are independent of the parties, and one aspect of that is that there should not be ex parte communications between an appointor and ‘its’ arbitrator. How far that rule can be relaxed when discussing potential candidates for the third- arbitrator position is a matter on which opinions sometimes differ.16 In our view, however, under the modern practice, it is generally desirable either to avoid such communications entirely or that the parties and the tribunal should clearly specify the ‘ground rules’ and apply them evenhandedly. For example, the parties may agree that each side can discuss the identity of the chair with its appointee. What must certainly be avoided is any one-sided difference in treatment. It is often positively desirable for the parties to be able to explain to their appointees the likely issues in the case and (in general terms) to express their views as to the characteristics that might be appropriate for a third arbitrator. In general, however, this is better done transparently in correspondence than in private conversations with the appointees, although parties may be willing to agree ground rules which permit such conversations. Second, should the parties seek to agree the identity of the chair themselves? Parties sometimes make efforts to do so. This might seem to be a good idea in theory, on the basis that it will lead to a chair in which both parties have confidence. In practice, however, it may lead to a considerable amount of time being taken up on discussions which ultimately prove fruitless. As more Bermuda Form disputes have been the subject of decision by tribunals, the ‘track record’ of particular arbitrators (which may be known to the parties’ advisers) may lead one or other party to consider a particular arbitrator as an undesirable choice as chair. This is in our view a most regrettable development, and it has made the process or seeking to agree a chair more difficult than in the past. It is therefore in our view preferable for the parties to leave the choice of chair to the two party-appointed arbitrators, as provided for by the terms of the Bermuda Form itself. If each party has confidence in its chosen arbitrator to decide the merits of the dispute, then it could reasonably be expected that the arbitrator would make a sensible choice as chair.

14.22

Third, how should the two party-appointed arbitrators approach the task of appointing the chair? This is ultimately a matter for the two arbitrators, but there are various ways in which they may seek the parties’ assistance in identifying an appropriate candidate. One procedure that some tribunals adopt is as follows. The tribunal selects three or five candidates acceptable to it. Those names are put forward to the parties, which rank them (without consulting each other) from 1 (most preferred) to 5 (least preferred). The preferences are then combined, and the candidate with the best combined preference is selected. If two candidates are tied, then the one chosen is the one with the least difference between the parties in terms of preference. This is a workable system which tends to avoid the most divisive candidates and yields a clear answer. In our experience it usually works smoothly, and better than systems which invite parties to volunteer candidates or rank each other’s preferences, while still avoiding the risk that the arbitrators might inadvertently select a

14.23

16 See

below, para 14.35.

298 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

person to whom one party or the other violently objects. But, unless the parties agree, the ultimate decision on appointment lies with the two party-appointed arbitrators, and it is often undesirable to canvass names (rather than general indications of qualification) because canvassing tends to turn the appointment process into a battle in which there can be ‘winners’ and ‘losers’, which gets the arbitration off to quite the wrong start. 14.24

Fourth, what happens if the procedure fails? Under section 18 of the Arbitration Act 1996, the parties are ‘free to agree’ what is to happen if the procedure for appointing a chair fails, and the court has a power to appoint if they have not agreed. The Bermuda Form is a curiosity in this sense: it does contain an agreed procedure, but under that agreed procedure, if the appointment is not made by the arbitrators, either party may ‘apply to a judge of the High Court of England for the appointment of the third arbitrator’. Is such an application to be made under section 18 of the Act? Or is it an application outside the Act, by agreement of the parties, which just happens to make a ‘High Court judge’ the appointing authority? On that interpretation, the application can be made entirely informally, by letter, to anyone who is a ‘High Court judge’. In some cases this is what has happened.

14.25

We are highly doubtful that the latter is a proper procedure. High Court judges are public officials, and they ought generally to exercise not some private jurisdiction, even one that they enjoy by virtue of their status, but only their official jurisdiction in accordance with the High Court’s jurisdiction, which is not a matter of private agreement. The Arbitration Act 1996 does indeed make provision for High Court judges to act occasionally as ‘judge arbitrators’.17 But it is notable that it was thought necessary to enact special statutory provisions to allow this, and it can only happen with the consent of the Lord Chief Justice. For a High Court judge to accept a role as a ‘privately selected appointing authority’ is constitutionally questionable. To which judge should such an application be made? Is the petitioner free to select anyone who happens to be a High Court judge, such as a judge of the Family Division, or a criminal specialist? Is the decision taken a judicial one, subject (to the extent permitted by the Arbitration Act 1996) to appeal? Or is it some other beast, subject to no or some other form of judicial control?

14.26

In our view, the proper conclusion is that the reference to the parties’ ability to apply to ‘a judge of the High Court’ should be taken not as an effective agreement which constitutes any randomly chosen High Court judge (of whom there are more than 100) as an agreed ‘appointing authority’, but simply as a reference to the High Court’s statutory power to appoint an arbitrator. It follows that the application should be made under section 18 of the Arbitration Act 1996 and in accordance with the relevant rules of court. It should be determined by a judge, sitting judicially and in a public capacity. A High Court judge who is approached informally to appoint – in practice, usually, the judge in charge of the Commercial Court18 – ought to decline to do so, but instead require a formal application, properly made (with the appropriate application fee paid), to the court.

17 Arbitration Act 1996, s 93. 18 An individual judge of the Commercial Court will be appointed as the ‘judge in charge’ and will usually occupy that role for one or two years.

The Commencement of Arbitration

299

The practice of informal approach is sometimes said to be supported by an old decision under the Arbitration Act 1950. In that case,19 faced with very narrow statutory provisions and an arbitration clause which made the ‘High Court’ (not any individual judge) responsible for appointment in cases of disagreement, Donaldson J was prepared (on an application at which only one party appeared) to say that the parties could confer jurisdiction to appoint on the court. But it seems doubtful if that conclusion is required in the light of the broader terms of the 1996 Act, which contains sufficient statutory authority, nor that Donaldson J intended to sanction informal grant of power to individual judges or its exercise, effectively, extra-judicially.

14.27

We do not think that this case provides adequate support for any use of ‘informal approaches’ to High Court judges. The language of the arbitration clause in the Bermuda Form does not make the court an ‘appointing authority’;20 rather, the clause says that the party can ‘apply to a judge of the High Court’. The language is therefore concerned with the making of an application. As such, it mirrors the language under section 18(2) of the 1996 Act: ‘… any party to the arbitration agreement may (upon notice to the other parties) apply to the High Court’. This indicates that what is required is an ordinary application under section 18. Furthermore, there is nothing in the clause which suggests that the way in which a party must ‘apply to the court’ is any different from other applications under the 1996 Act. The expression ‘apply to the court’ is used many times in the Act.21 All such applications are made by issue of an Arbitration Claim Form in accordance with the procedures in the Part 62 of the Civil Procedure Rules (CPR). There is no reason why a different procedure should apply in the context of the appointment of the third arbitrator. Indeed, CPR 62.2(1) defines an ‘arbitration claim’ as including not only any application under the 1996 Act, but also ‘(d) any other application affecting – (i) arbitration proceedings (whether started or not); or (ii) an arbitration agreement’. Even if the provisions of the Bermuda Form do not (contrary to our view) engage section 18 in the event that the arbitrators do not agree on the third, the procedure under Part 62 of the CPR is therefore nevertheless applicable.

14.28

There have undoubtedly been cases (both formal applications and, we believe, informal ones, made contrary to the practice we suggest is proper) for appointment of a third arbitrator where the party-appointed arbitrators have been unable to agree. The cases are not systematically reported, but the results of three are given in the following decisions.

14.29

In XL Insurance Ltd v Toyota Motor Sales USA Inc,22 the insurers’ appointee had proposed a number of candidates, namely a retired US federal judge, a former justice of the Supreme Court of Canada and a retired English Law Lord. The policyholder’s appointee had proposed a partner in a US law firm, and a well-known Canadian arbitrator

14.30

19 Medov Lines SpA v Traelandsfos A/S [1969] 2 Lloyd’s Rep 225. 20 It is not unusual for individuals holding particular positions to be identified in arbitration clauses as an ‘appointing authority’; for example, the President of the London Court of International Arbitration or the President of the Law Society. For the reasons given in the text, we do not consider that the provision in the Bermuda Form, enabling a party to apply to a judge of the High Court, is to be equated with such clauses. 21 See, eg, ss 9(1), 16(3), 21(5), 24(1), 25(3), 28(2), 56(2), 63(4), 67(1) and 68(1). 22 Judgment of Aikens J delivered on 14 July 1999. This decision is not reported.

300 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

and arbitration practitioner. Unfortunately, each side’s suggestions were unacceptable to the other side, and the matter came before the English Commercial Court. The judge considered that the court had to look at all the circumstances of the case before deciding whom it should appoint. The judge thought that the relevant factors in that case were: (i) the choice of London as a neutral seat; (ii) the choice of English law as the curial law of the arbitration; (iii) the choice of New York law as the substantive law of the contract, but with modifications which were chosen ‘in conjunction with a London arbitration venue in order to ensure that they could be considered dispassionately and independently by the tribunal that is appointed’; and (iv) the fact that the arbitration clause left it to the two party-appointed arbitrators to agree on the identity of the third. The court considered that its choice lay between the retired US federal judge and the retired English Law Lord. In the particular circumstances of that case, the court chose the retired Law Lord. 14.31

In Guidant LLC v Swiss Re International SE,23 the main issue was whether preference should be given to an arbitrator who was involved in related arbitrations. The court decided that it should not appoint that candidate. Instead, it appointed (from a number of ‘impeccably qualified’ candidates) a recently retired High Court judge, apparently on the grounds of his ready availability. In proceedings which led ultimately to the Halliburton dispute,24 a well-known London arbitration specialist had been appointed.

14.32

Although those cases might suggest that the English court has a preference for appointing English lawyers or retired judges, it should not be assumed that it will do so. Naturally, familiar names may be likely to be thought safe appointments. But the court is likely to be concerned with the legitimate needs of the particular case and, as Guidant suggests, with the desirability of finding a truly neutral arbitrator who can enjoy the confidence of both parties and who has (importantly) the time to devote to the arbitration in order to achieve a prompt resolution. This means that the court is unlikely to be attracted to appointing anyone whom one party has previously ‘advocated for’, and is likely to be looking for a truly neutral candidate. The particular circumstances of each case will demand close attention. Declarations and Hypothetical Questions

14.33

An interesting question arises as to the extent to which either the policyholder or insurer can commence arbitration proceedings to clarify the applicability of the policy when there is uncertainty about whether the policyholder’s liability will ever reach the attachment point of the excess policy involved. For example, the policy may apply only in excess of US$100 million, and the factual position may be that, although the claims against the policyholder may exceed US$100 million, there is considerable uncertainty as to whether

23 Guidant LLC v Swiss Re International SE [2016] EWHC 1201 (Comm), [2016] 1 CLC 767. 24 Halliburton Co v Chubb Bermuda Insurance Ltd (Formerly Ace Bermuda Insurance Ltd) [2020] UKSC 48 [2020] 3 WLR 1474. For a discussion of a US perspective on this litigation, see Lorelie S Masters, John Jay Range and Paul Moura, The Bermuda Form and Arbitration of Disputes in London’ (2018) 73(1) Dispute Resolution Journal s IV.B (available at www.arbitrationlaw.com/books/dispute-resolution-journal).

The Commencement of Arbitration

301

this will actually happen and, if so, when. Can such a dispute be referred to arbitration so as to obtain the tribunal’s ruling as to whether the insurance policy applies if the policyholder’s liability ever exceeds the policy’s attachment point?25 There is an argument that the tribunal has no jurisdiction to decide such a ‘hypothetical’ dispute.26 It is, however, a weak one. The Form has expanded the subject matter of the arbitration from ‘any dispute arising under this Policy’ (as the original versions had it) to ‘any dispute, controversy or claim arising out of or relating to this Policy or the breach, termination or invalidity thereof’.27 This expansion of jurisdiction makes it more difficult for a party to argue that a particular issue is not covered by the arbitration clause, and thus reinforces the exclusion of the courts from Bermuda Form cases.28 We consider, therefore, that the hypothetical nature of the dispute is a matter which is relevant not to jurisdiction, but to the tribunal’s discretion to decide whether a remedy should be granted. The question of whether the tribunal should exercise its discretion to grant a declaration is an issue that is procedural rather than substantive, and is therefore governed by English law (as the law of the arbitration procedure) rather than New York law.29 Whatever might have been its reluctance in the past, English law is now willing to make declarations of rights: ‘There must, in general, be a real and present dispute between the parties before the court as to the existence or extent of a legal right between them. However, the claimant does not need to have a present cause of action against the defendant.’30 So long as the debate is not merely academic but has some real bearing on the parties’ practical position, we think that a tribunal can and should entertain a reference that seeks declaratory relief. In circumstances where the underlying claims are a very long way from, and may never reach, the excess point of the excess policy in question, the tribunal may be reluctant to see substantial sums of money spent, against the opposition of one party to the reference, in deciding what may turn out to be an academic question. The tribunal may therefore consider putting substantive hearings, and any expensive interlocutory processes such as disclosure, on ice until it becomes clearer whether there is a real or only a theoretical dispute. Such a course of action would not be inconsistent with the general duty of the arbitral tribunal set out in section 33 of the English Arbitration Act 1996, which provides that the tribunal shall ‘adopt procedures suitable to the circumstances of the particular case, avoiding unnecessary delay or expense, so as to provide a fair means for

25 A request for such a ruling is similar to a policyholder’s request in an American court proceeding for a declaratory judgment. Both US federal statutes (28 USC s 2201) and parallel statutes in various US jurisdictions allow for such declarations of rights and obligations under contracts. 26 See, eg, Re Clay (Clay v Booth) [1919] 1 Ch 66; and EI Dupont de Nemours & Co v IC Agnew and Others [1987] 2 Lloyd’s Rep 585. Under the Forms prior to the 004 Form, there was also a question whether such relief could constitute a ‘claim’ or ‘demand’ under the policy. We think that it could. 27 Article VI(N). 28 The English courts now adopt a liberal approach to the scope of arbitration clauses: Fiona Trust & Holding Corp v Privalov [2007] UKHL 40 [2007] 4 All ER 951; Ace Capital Ltd v CMS Energy Corporation [2008] EWHC 1843. 29 See above ch 3, in particular para 3.41; and L Collins et al (eds) Dicey, Morris & Collins: The Conflict of Laws, 15th edn (London, Sweet & Maxwell, 2012) paras 7-011–7-014. 30 Rolls Royce plc v Unite the Union [2009] EWCA Civ 387, [2010] 1 WLR 318, at [120] (Aikens LJ), where the relevant principles are usefully summarised.

14.34

302 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

the resolution of the matters falling to be determined’. In other cases, however, it may be crucial to one party for there to be an early decision on whether there is coverage or whether a defence advanced by the insurer is effective.

THE SELECTION OF AN ARBITRATOR Overview 14.35

Decisions as to the composition of the arbitral tribunal can be of great importance in arbitrations. Lawyers and their clients will often devote substantial time and thought to this question. This decision may involve detailed investigations into the proposed arbitrator, and an attempt to forecast how the case will play out if X is appointed rather than Y. It has become increasingly common for clients to want to meet with potential arbitrators before appointing them, although both clients and possible appointees should exercise the greatest care when doing this. One international arbitrator31 has prescribed a number of rules which should govern any such meetings, for example: the meeting should be limited to 30 minutes, the participants should not discuss the merits of the case, the meeting should not take place at the offices of the client or its lawyers, and there should be no ‘hospitality’ involved except for tea or coffee. It is unlikely that an English court would criticise an arbitrator who followed these ‘rules’. Nevertheless, the English courts discourage unilateral communications between an arbitrator and the appointing party.32 Accordingly, if discussions are to take place between appointor and appointee as to the identity of the third arbitrator, then it is appropriate and necessary for the parties to agree that such discussions are permissible.33

14.36

As with the choice of legal representation, the appointment of an arbitrator is essentially a matter of tactics rather than principle. It has been the practice of insurance companies

31 See Nigel Blackaby et al, Redfern and Hunter on International Arbitration, 6th edn (Oxford, OUP, 2015) para 4.73 (hereafter Redfern and Hunter on International Arbitration). 32 In Makers UK Ltd v The Mayor and Burgesses of the London Borough of Camden [2008] EWHC 1836 (TCC), [2008] BLR 470, the court considered that it was ‘unexceptionable’ for a party to contact an adjudicator (who resolves, on an interim basis, disputes under building contracts) to check availability and conflicts, but generally discouraged unilateral contact. In Norbrook Laboratories Ltd v Tank [2006] EWHC 1055 (Comm), [2006] ArbLR 50, the sole arbitrator had made direct unilateral conduct with the parties after the arbitration was underway, and the court said that such contact ‘is generally to be deprecated’: see paras [132]–[136]. 33 The International Bar Association (IBA) ‘Guidelines on Conflicts of Interest in International Arbitration’ consider it permissible for an arbitrator to have ‘initial contact with a party or an affiliate of a party (or their counsel) prior to appointment, if this contact is limited to the arbitrator’s availability and qualifications to serve or to the names of possible candidates for a chairperson, and did not address the merits or procedural aspects of the dispute, other than to provide the arbitrator with a basic understanding of the case’: see ‘Green List’, para 4.4.1. A similar approach is taken in paras 5.1 and 5.2 of the 1987 IBA Rules of Ethics for International Arbitrators. The authors of Redfern and Hunter on International Arbitration ((above n 31) para 4.40) state: ‘It is usual for each of the arbitrators to consult with the party who nominated them as to the identity of the presiding arbitrator and by this means ensure that the appointee is acceptable to all concerned. In this context, the party-nominated arbitrators should inform each other that this procedure is being followed.’ Consultation by the party-appointed arbitrators in relation to the third arbitrator is expressly permitted under the rules of the Insurance and Reinsurance Arbitration Society) (UK): ARIAS Arbitration Rule 6.7.

The Selection of an Arbitrator

303

such as ACE (now Chubb) and XL (now Axa XL) to appoint English QCs or English retired judges as their arbitrators, and to avoid appointment of US lawyers or jurists. The thinking behind this is probably the insurers’ desire to have the insurance policy interpreted in a ‘black-letter’ fashion by a lawyer who comes to the policy without any perceived preconceptions that might be derived from extensive involvement in insurance disputes or the legal system in the United States. In addition, insurers will expect that an English QC will not have an inherent sympathy for one side or the other, since – unlike many US attorneys – English QCs specialising in insurance law will generally have acted for both policyholders and insurers. It is not the practice or custom in England for a barrister to represent exclusively policyholders or insurers, while American legal ethics rules may preclude lawyers from representing both insurers (whether on insurance or other issues) and policyholders. Since the object of the insurers in seeking dispute resolution outside the US court system, and seeking to modify New York law in the governing law provision, was to insulate the resolution of the claim from the approach taken by courts in the United States, the appointment of an English lawyer or retired judge as arbitrator is a logical extension of these developments. Another down-to-earth practical reason lies behind the approach taken by insurers. Many of the insurance companies who write on the Bermuda Form have now fought a number of cases on various issues arising under the Form, and they now have some knowledge about what particular arbitrators think about particular issues. Appointments of this sort may raise issues of impartiality, which are considered below.

14.37

Potential Grounds for Objection The most difficult questions concern the impartiality of arbitrators. It is, naturally, a requirement of arbitration that arbitrators must be impartial, and the Arbitration Act 1996 empowers the court to remove them if there are ‘justifiable doubts as to [an arbitrator’s] impartiality’.34 The common law test of bias applied in Locabail35 was whether there was a ‘real danger’ of bias on the part of the relevant member of the tribunal in question, in the sense that the member might unfairly regard (or have regarded) with favour, or disfavour, the issue under consideration.36 This test was somewhat modified in the light of the statutory enactment in England of the European Convention on Human Rights. The question is now ‘whether the fair-minded and informed observer, having considered the facts, would conclude that there was a real possibility that the tribunal was biased’.37

34 Arbitration Act 1996, s 24(1); see also s 33 for the duty to act ‘impartially’ between the parties. If issues as to impartiality arise, an English court would decide them under English law, since this is the curial law of the arbitration; see ch 3 above, in particular para 3.16. 35 Locabail (UK) Ltd v Bayfield Properties Ltd [2000] QB 451. 36 This test was formulated in R v Gough [1993] AC 646. 37 Porter v Magill [2001] UKHL 67, [2002] 2 AC 357; Helow (AP) v Secretary of State for the Home Department [2008] UKHL 62, [2008] 1 WLR 2416. The fair-minded observer is not ‘unduly sensitive or suspicious’ (at [2]. The need to modify the Gough test in the light of European case law was recognised in Re Medicaments and Related Classes of Goods (No 2) [2001] 1 WLR 700. In Halliburton Co v Chubb Bermuda Insurance Ltd (Formerly Ace Bermuda Insurance Ltd) [2020] UKSC 48, [2020] 3 WLR 1474, the Supreme Court reaffirmed the applicability of this test in the context of a Bermuda Form arbitration.

14.38

304 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

This modification has been described as modest,38 and guidance provided in Locabail is still applicable after the modification.39 The court will adopt the ‘fair-minded and informed observer’ test in approaching any application under section 24.40 In practice, the court is also often influenced by international norms, in particular by those in the International Bar Association (IBA) ‘Guidelines on Conflicts of Interest in International Arbitration’.41 14.39

This is not the place for a comprehensive discussion of the principles; there are many factors which may arise in any arbitration (such as financial interests, partnership in firms and so forth). However, there are two related questions which have been particularly prominent in Bermuda Form cases.

14.40

The first arises where a single large loss to an insured, or sometimes a number of insureds, spawns a number of references to arbitration. Is it legitimate for a single arbitrator to be appointed in more than one reference? And if so, what disclosure should be given? The doubts about this practice stem from several sources. First, if one party has appointed an arbitrator repeatedly, he or she may be, or appear to be, beholden to that party. Second, because arbitrations are confidential, there is a risk that the arbitrator may learn information in one case which will be known to one party and not the other. And there is a similar risk that the party may acquire information about the arbitrator’s views which will be known to one party and not the other. In Halliburton,42 such a case reached the Supreme Court. A well-known English arbitrator had been appointed by the court to chair one arbitration concerning the loss of the Deepwater Horizon oil platform in the Gulf of Mexico, and had also been appointed by insurers (including the insurers who were party to the case he was chairing) in other references arising out of the same facts, without making full disclosure in the case he was chairing. Although, on the particular facts of that case, the arbitrator in question was held to be acceptable, that can only really be regarded as a matter of considerable good fortune as far as the arbitrator and the insurers in the arbitration in question were concerned. The basic principle that the case establishes is clear: an arbitrator in such a situation must disclose the multiple appointments to the parties in each arbitration, and it would be prudent to assume that in the absence of such disclosure the modern rule would be that the arbitrator could not continue to act.43

14.41

Two supplementary points need to be made. First, although disclosure is required, it is not a panacea; there may be circumstances in which the conflict is so profound that it 38 See Porter v Magill, ibid, at [103] (Lord Hope). 39 Amec Capital Projects Ltd v Whitefriars City Estates Ltd [2004] EWCA Civ 1418, [2005] 1 All ER 723, [18]; El-Farargy v El- Farargy [2007] EWCA Civ 1149, [2007] 3 FCR 711, [26]. 40 Halliburton (above n 37) at [52]. In Laker Airways v FLS Aerospace [1999] 2 Lloyd’s Rep 45, 48, Rix J equated the (then) common law test with the statutory definition. In Norbrook Laboratories v Tank [2006] EWHC 1055 (Comm), [2006] ArbLR 50, Colman J applied the common law test in the context of an application under s 24; see, in particular, paras [145], [155]–[156]. See too the approach taken by Morison J in ASM Shipping Ltd of India v TTMI Ltd of England [2005] EWHC 2238 (Comm) [2006] 2 All ER (Comm) 122 on an application under s 68 of the Arbitration Act 1996. 41 https://www.ibanet.org/MediaHandler?id=e2fe5e72-eb14-4bba-b10d-d33dafee8918. The IBA Guidelines, published in 2014, are referred to extensively in Halliburton. However, they are not principles which bind the English court; see W Ltd v M SDN BHD [2016] EWHC 422 (Comm), [2017] 1 All ER (Comm) 981. 42 Halliburton (above n 37). 43 Ibid at [81], [137]–[138] (Lord Hodge).

The Selection of an Arbitrator

305

would be impossible for the arbitrator to act even with disclosure.44 Second, although the court in Halliburton made it clear that a failure to disclose was material from which ‘reasonable doubts’ might be inferred, it did not go so far as to say that if there is disclosure, the party to whom it is made will necessarily be able to make good an objection. That will depend on the nature and extent of the issues which might arise – for example, the similarity of the issues, any overlap of witnesses, the number of arbitrations and so forth. No doubt, in practice, parties and arbitrators would be well advised to take the safe course and avoid overlapping appointments to which objection is taken. But it should not be supposed that there is a legal rule to that effect: an arbitrator might disclose the overlap, face an objection and resist it.45 The question would then be whether given the disclosure, the overlap creates a real possibility of bias in the mind of the fair-minded and informed observer. Second, Halliburton dealt with a case where the overlap was not merely in the existence of multiple arbitrations concerning the same type of insurance contract, but multiple arbitrations arising from the same underlying event; in other words, it was quite specific. As such, it does not directly address the issues that arise where an arbitrator is repeatedly appointed in unrelated disputes involving the Bermuda Form. In general, in court proceedings, it could not be regarded as problematic that a judge had previously expressed views on the same or a similar legal issue. Thus, in Locabail (UK) Ltd v Bayfield Properties Ltd,46 the English Court of Appeal held that it was not a valid objection to a judge or arbitrator that the individual had previously expressed a view or decided an issue in a manner favourable or adverse to one side or the other.47

14.42

But arbitrations are a bit trickier, because whereas the information about a judge’s previous decisions or judicial philosophy is equally available to any litigant with access to a law library, arbitration decisions are confidential; one party therefore knows about an arbitrator’s views, where the other may not. The views previously expressed by an arbitrator will not be equally available to both parties. Any previous awards are confidential,48 and arbitrators would possibly breach confidence if they advised the parties of views expressed in an earlier arbitration to which one or both of them was not a party. Accordingly, a party is not in a position to know whether an arbitrator’s views expressed on a previous occasion (perhaps even in a dissenting opinion) were expressed so as to reveal an entrenched mind, nor (in the less extreme case) what the arbitrator’s

14.43

44 Ibid [108] (Lord Hodge), [170] (Lady Arden). 45 It may be the proper course for the arbitrator to adopt in relation to a challenge which he or she considers to be unsustainable, ie, leaving it to the court to decide, in the event that a challenge is made. See the stance taken by the arbitrator in Laker Airways v FLS Aerospace [1999] 2 Lloyd’s Rep 45. See too Redfern and Hunter on International Arbitration (above n 31) para 4.151. 46 Locabail (UK) Ltd v Bayfield Properties Ltd [2000] QB 451. 47 See in particular ibid [25] of the judgment of the Court of Appeal: ‘The mere fact that a judge, earlier in the same case or in a previous case, had commented adversely on a party or witness, or found the evidence of a party or witness to be unreliable, would not without more found a sustainable objection.’ It would seem to follow that judges or arbitrators can continue to act even if they have previously expressed views, favourable to one party, as to the interpretation of a contract. 48 Confidentiality is discussed in ch 15, paras 15.23–15.27 and (in the context of discovery) ch 16, paras 16.93–16.94.

306 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

chain of reasoning was so as to enable it to be the subject of intellectual attack. What can a party do, however, if it learns that the opposing party has appointed an arbitrator who has previously expressed views on a key issue in the case? It could be argued that there is an unfairness because there is a lack of ‘equality of arms’49 in circumstances where one party has private knowledge of the fact that its chosen arbitrator has decided a particular point in its favour on one or more occasions, and this is unknown to the other party. An application for the arbitrator to stand down is, however, unlikely to be met with success, at least in the absence of evidence that the arbitrator has previously expressed trenchant views on the subject.50 An application for the production of the earlier awards (redacted so as to avoid revealing the names of the parties) is perhaps more attractive. Production of the awards will enable a party to see the thought processes of the tribunal and thereby enable it to address the views of the arbitrator. However, the production of awards, even if in a redacted form, would probably breach confidentiality and would therefore require either the consent of both parties to the previous decision or an order from the subsequent tribunal. A tribunal may think it appropriate to make such an order where the previous decision is closely related to the dispute being arbitrated. This would seem to be particularly appropriate if, as further discussed above, the arbitrator has previously been party to a decision in respect of the same problem giving rise to the liability of the policyholder, for example, on a different layer of the policyholder’s programme. In Emmott v Michael Wilson and Partners Ltd,51 Thomas LJ drew attention to various problems associated with the confidentiality of awards in the insurance and reinsurance market, but his remarks do not seem to have been the spur to any change of practice. 14.44

As framed in the previous paragraph, this issue is not really one of bias so much as one of equality of arms. However, in addition, there is at least a perception that arbitrators who receive regular, perhaps lucrative, appointments from one particular source may be more disinclined to bite the hand that feeds them than strict impartiality might dictate. Halliburton does not deal with such cases, though it does at least show that the courts are alive to them. It would, we think, be unrealistic to suppose that an arbitrator could be disqualified based simply on some history of previous appointment, or some previous decisions. But that does not mean that the issue is irrelevant. It is likely to be a question of degree.52 We think that the English courts would pay regard to the position expressed in the IBA Guidelines, which regard it as an ‘orange list’ item (ie, an item in relation to

49 An expression used in the jurisprudence relating to the European Convention on Human Rights, which now forms part of English law and means that there is a fair balance between the opportunities afforded to the parties involved in litigation. The application of the European Convention is not affected by Brexit, the UK’s departure from the EU. 50 This will be difficult to show, not only because the previous award or awards may be unavailable, but also because they may be awards of a panel of three and it could be said that the award represents the views of the panel, not the particular individual. The counter-argument is that by agreeing to the award, the individual must be taken to accept that the views set out therein are his or her own. 51 Emmott v Michael Wilson and Partners Ltd [2008] EWCA Civ 184, [2008] 2 All ER (Comm) 193, [131]. 52 See Cofely Ltd v Bingham [2016] EWHC 240 (Comm), [2016] 2 All ER (Comm) 129, where the court removed an arbitrator for apparent bias. He had received multiple appointments from the same source and had failed to answer reasonable questions which had been asked on that issue.

The Selection of an Arbitrator

307

which disclosure should be considered and which may at least raise doubts) if an arbitrator has been appointed by a party or an affiliate of a party twice or more in the previous three-year period.53 Whether that would be disqualifying would then no doubt depend on the particular facts disclosed, including the circumstances and frequency of appointment, the similarity of the issues to be considered, and the significance of the appointments in the arbitrator’s practice generally. The standard to be applied will be that of the fairminded and objective observer, informed of those circumstances, and considering whether there is a ‘real possibility’ that the frequent appointments may create bias. Considerations for Choosing a Party-Appointed Arbitrator If the choice for one of the parties, say, the insurer, is likely to be an English barrister or retired judge, should the other party, in this example the policyholder, follow suit by appointing an English lawyer (whether barrister, retired judge or solicitor)? The policyholder should certainly consider this option, particularly if the proposed arbitrator has considerable experience in international arbitrations and, in particular, Bermuda Form arbitrations. Such an appointment may have the advantage of providing a solid counterweight to the insurer’s appointee. It may also possibly produce the result that the two English arbitrators decide to appoint a distinguished US lawyer or retired judge as chair or umpire. This may be perceived to be a desirable consequence from the policyholder’s point of view, although the choice of the chair, under the arbitration clause in the Bermuda Form, depends upon the views of the two arbitrators, not their appointors.54

14.45

A policyholder’s natural first-choice arbitrator, however, is likely to be a US lawyer, whether a practising lawyer or a retired judge. We think that there is a great deal to be said for this approach. The ultimate battle between the parties will concern substantive issues of New York insurance law and an understanding of the US tort law and judicial systems, and the policyholder will often be well served by appointing an arbitrator who is familiar with the law, practice and legal system in the United States, including, if possible, the applicable principles of New York insurance law. The policyholder’s legal team will undoubtedly include US lawyers. They will be seeking to persuade the tribunal as to what the relevant New York law principles are. Their task can be expected to be somewhat easier if at least one member of the tribunal understands the legal and practical context on which their client’s case is based. If the tribunal consists exclusively of English or non-US lawyers, the possibility for a misunderstanding of applicable principles or complexities of the US tort law system is obviously much greater. Indeed, the policyholder may find itself having to debate before the tribunal issues of US law and practice that would be second nature to a tribunal composed of US lawyers. For example, an exclusively English tribunal

14.46

53 IBA Guidelines (above n 41) para 3.1.3. 54 See the discussion earlier in this chapter about the selection of the chair at paras 14.19–14.32. In practice, if the parties agree upon the choice of a chair, the two arbitrators will almost certainly give effect to that agreement.

308 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

may not have a good feel for the weight to be accorded to federal court decisions applying New York law, and insurers have been known to argue that a decision of the lowest-level New York State court should automatically be applied in preference to a decision by the US Court of Appeals for the Second Circuit, applying New York law.55 A US appointee may also better understand other aspects of the approach taken by the policyholder’s US lawyers, for example, in relation to discovery or expert witnesses or other procedural matters. Issues will also often arise regarding the handling and outcome of the underlying claims litigation against the policyholder in the US, and a US lawyer may readily see the reasons behind a decision that may at first sight seem surprising to an English lawyer. The underlying claims litigation may well involve class actions or multi-district litigation,56 and it may be advantageous to the policyholder to have a panel member who is familiar with such proceedings and the pressures which they exert. 14.47

In addition, as we have said, a key part of the presentation of any case is for the lawyer to establish, and not lose, credibility with the tribunal. If English attorneys are acting before an English panel, the chances are that the advocate, whether a solicitor or barrister, will be known by at least one member of the panel. Indeed, it is not uncommon for the English arbitrator and the barrister who is acting for one of the parties to be from the same chambers. Under English law as it currently stands, this presents neither a conflict of interest nor a ground of challenge for bias.57 Unlike the relationship of partners in American law firms and English solicitors’ firms, each barrister is an independent contractor. It will, however, mean that the US lawyer acting for the other party will need to understand and deal with the personal relationship between the barrister and the arbitrator. The problem may be even greater if an US lawyer were to appear, without the benefit of English counsel, before a panel of English barristers who may never have heard

55 See ch 7 for discussion of the cases relating to the issues of the policyholder’s expectation or intent, and whether New York law applies a subjective or objective test. 56 The US Code allows for ‘multi-district litigation’, a special federal legal procedure under which a panel of federal court judges can decide to consolidate and where necessary transfer to one court multiple lawsuits raising common questions of fact that are pending in federal courts around the country, for the purposes of pre-trial proceedings and discovery. See 28 USC §§ 1407, 2122. The objective is to speed the handling of complex cases and promote efficient pretrial proceedings. 57 In Laker Airways v FLS Aerospace [1999] 2 Lloyd’s Rep 45, the English Commercial Court rejected an application to remove an arbitrator for bias on the basis that he was in the same chambers as the barrister for one of the parties. The court’s judgment referred to other decisions where a similar challenge had been rejected by the Paris Court of Appeal and the London Court of International Arbitration (LCIA). The Laker Airways case appears to have been an ad hoc reference that was not conducted under the rules of any arbitral institution. In Smith v Kvaerner Cementation Foundations Ltd [2006] EWCA Civ 242 [2007] 1 WLR 370 [17], the Court of Appeal referred to Laker Airways with approval. See too Watts v Watts [2015] EWCA Civ 1297 [28]; Zuma’s Choice Pet Products v Azumi [2017] EWCA Civ 2133 [37]. It is now increasingly common, and probably represents good practice, for an arbitrator in an international arbitration to disclose the fact that counsel is a member of the same chambers; see too the IBA Guidelines on Conflicts of Interest in International Arbitration (October 2014), pt II, para 3.3.2. There is, however, an increasing hostility in some quarters to the notion that a barrister should be able to appear before an arbitrator from the same chambers. In ICSID Case No ARB/05/25 (available at http://icsid. worldbank.org/ICSID), an ICSID tribunal ruled that a QC was not entitled to participate in an arbitration where the chair was a member of the same chambers. Whilst that case may have turned upon the fact that the QC’s participation had not been disclosed until very late in the day, it may foreshadow a more general approach, at least in certain types of international arbitration; see Redfern & Hunter on International Arbitration (above n 31) paras 4.127–4.130.

Appendix: Notice to Commence Arbitration

309

or dealt with that attorney or the attorney’s firm. This problem can be solved or certainly mitigated by the appointment of a US lawyer to the tribunal. If a US lawyer is to be appointed, should it be a practising lawyer or a retired judge? 14.48 In many cases, either a distinguished lawyer in practice or a retired judge will be a good choice. Care should, however, be taken not to appoint a person who is likely to be perceived as inherently hostile to one side in the dispute, as this may result in a perception that the arbitrator is not objective and the chair of the tribunal may then accord less weight to that arbitrator’s views. A distinguished retired judge may therefore be an appropriate choice, since his or her views (even if perceived initially to be ‘pro- policyholder’) ought to carry significant weight both with the other arbitrator and more significantly with the chair. We do not believe that a US lawyer cannot be fair simply because of the nature of the lawyer’s practice; to say so would endorse arguments that the English courts viewed with scepticism in Halliburton, for example. However, there is no point in appointing someone who, in practice or known temperament, a party believes will not be listened to by the other members of the tribunal. In our view, many distinguished practising insurance attorneys would make an excellent choice as arbitrator. If a retired judge is to be appointed, he or she is likely to have some basic familiarity with US insurance law, but, while such expertise could be useful, we do not think that it is essential for the appointee to have been an insurance coverage practitioner either in practice or when on the bench. Appointment of the Chair The arbitration clause provides that the two party-appointed arbitrators ‘shall … within (30) calendar days after the appointment of the second arbitrator choose a third arbitrator’.58 The third arbitrator is then the chair of the tribunal (section 16(5) of the Arbitration Act 1996). Appointment of a chair is ultimately a matter for the two arbitrators. The appointment of the chair gives rise to various issues which are discussed in paragraphs 14.19–14.32 above.

APPENDIX: NOTICE TO COMMENCE ARBITRATION Re: Policy CD 123456 To Whom It May Concern: AB Engineering Inc refers to its letter dated 1 October 2020 in which AB Engineering demanded that CD Insurance Co Ltd pay $75,000,000 under Policy No CD 123456 in respect of the loss referred to in that letter. In the light of CD’s letter dated 15 October 2020 refusing the claim, we hereby give you notice that AB Engineering Inc appoints

58 Condition

N of the 004 Form; and Condition (o) of earlier Forms.

14.49

310 Commencing a Bermuda Form Arbitration and Appointing Attorneys and Arbitrators

Mr Robert Henry of 1201 Avenell Street, New York, New York, as its arbitrator in respect of its claim for indemnity for the loss referred to above. Please advise me as to the name of your appointed arbitrator within the time limit set out in Article VI(N) of the above policy. Yours faithfully (Legal counsel, AB Engineering Inc)

15 The Course and Conduct of a Bermuda Form Arbitration in London INTRODUCTION This chapter considers how the arbitration will proceed in practice, following the initial appointments described in Chapter 14. These appointments may well result in a mixture of US and English lawyers being engaged, either as the legal team for one or both of the parties or on the panel itself. We identify certain problems that might be encountered because of cultural and procedural differences between English and US arbitration procedures that might surprise American lawyers or English lawyers who are asked to assist them. Things that are intuitive to an American lawyer may be counter-intuitive to English lawyers, and vice versa. We also consider the different stages of a Bermuda Form arbitration, and the steps that can be taken to challenge an award with which a party is dissatisfied. Again, the focus of this chapter is on a Bermuda Form arbitration conducted in London. Much of what we say is applicable to a Bermuda Form arbitration conducted in Bermuda, particularly if the panel is composed of a majority of English lawyers, but the statutory background in Bermuda is different and will need to be taken into consideration in those arbitrations conducted in Bermuda.

15.01

THE OVERALL SHAPE OF THE ARBITRATION London arbitrations will vary considerably in the way in which they are conducted. Section 34 of the English Arbitration Act 1996 gives the tribunal discretion as to all procedural and evidential matters, and the tribunal is free to adopt procedures it thinks are suitable for the determination of the case. Thus, the tribunal has discretion to decide when and where any part of the proceedings is to be held;1 whether any and if so what form of written statements of claim and defence are to be used, when these should be supplied and the extent to which such statements can later be amended;2 whether any and if so which documents or classes of documents should be disclosed between and produced by the parties and at what stage;3 whether any and if so what questions should

1 English

Arbitration Act 1996, s 34(2)(a). s 34(2)(c). 3 Ibid, s 34(2)(a), 34(2)(d); and see below ch 16. 2 Ibid,

15.02

312

The Course and Conduct of a Bermuda Form Arbitration in London

be put to and answered by the respective parties, and when and in what form this should be done;4 whether to apply strict rules of evidence (or any other rules) as to the admissibility, relevance or weight of any material sought to be tendered on any matters of fact or opinion;5 whether the tribunal should itself take the initiative in ascertaining the facts and the law;6 and whether and to what extent there should be oral or written evidence or submissions.7 15.03

The wide-ranging statutory discretion given to arbitrators gives rise to an important point. Bermuda Form arbitrations, and international arbitrations generally, should not mimic the procedures adopted in the English (or indeed American) courts. Although the tribunal may look at the way in which an English court would answer a particular procedural problem, the tribunal should not start from the assumption that English court practice and rules should apply unless there is a good case not to do so. The parties to a Bermuda Form arbitration have deliberately decided not to go to an English or any other court, and many arbitrators (rightly in our view) view with scepticism – if not distaste – any submissions which are based on the premise that English or other court procedures should be the default position for international arbitrations.

15.04

As with many other aspects of the 1996 Act, the discretion as to procedural and evidential matters is subject to the principle of party autonomy, viz ‘subject to the right of the parties to agree any matter’.8 In practice, an arbitral tribunal will expect a high degree of co-operation between the parties in relation to procedural matters. Whilst it is unlikely that the parties will be able to agree on everything, they should try to agree on as much as they can. Each party would be well advised not to use the interlocutory stages as a means of scoring points against the opposite party in the hope that this will sway the tribunal’s ultimate decision on the merits. In our view, the interlocutory matters referred to the tribunal for decision should, if possible, be kept to a minimum. The tribunal will usually not wish to see, as a matter of routine, the correspondence between the attorneys dealing with discovery of documents.

15.05

The flexibility of section 34 of the 1996 Act is illustrated by the way in which New York law is now adduced before tribunals sitting in Bermuda Form cases. In the past, some arbitrators and parties thought that New York law should be adduced through the testimony of expert witnesses who could be cross-examined. This would replicate the position in English court proceedings, where ‘foreign’ law is treated as a question of fact that must be strictly proved in evidence. Even prior to the 1996 Act, this approach was viewed with disfavour by some arbitrators, who considered that it would be far more efficient for New York law simply to be argued in legal submissions from the parties. The 1996 Act makes it clear that arbitrators are not bound by the strict rules of evidence, and accordingly arbitrators need not require that New York law be adduced as evidence rather than by way of submission. Use of legal submissions on New York law has now become the norm

4 Ibid,

s 34(2)(e). s 34(2)(f). 6 Ibid, s 34(2)(g). 7 Ibid, s 34(2)(h). 8 Ibid, s 34(1). 5 Ibid,

The Overall Shape of the Arbitration

313

in Bermuda Form and indeed other arbitrations, although occasionally a panel agrees that New York law should be adduced by way of expert evidence. Although Bermuda Form arbitrations do vary in terms of the detail, they tend to follow a fairly standard course. The parties will exchange statements of case. A discovery or ‘disclosure’ or ‘document production’ stage will follow. In England, however, ‘discovery’ connotes the disclosure of documents, whereas in the United States, it usually involves not only disclosure or production of documents, but also oral discovery through depositions, as well as answers to interrogatories and requests for admissions. English practice does not include the taking of depositions.9 Following disclosure of documents, the parties will exchange witness statements and if necessary expert evidence. The witness statements are of critical importance as they constitute ‘direct testimony’ (or ‘evidence in chief’ as it is called in England) in English arbitrations. A trial (or ‘hearing’ as it is called in England) will then be held where some or all of the witnesses who submitted statements appear to give oral evidence. The oral evidence will essentially consist of the witness answering questions in cross-examination, since the direct testimony will have been provided in the witness statements. Prior to the oral hearing, the parties will have put in written arguments or trial briefs. Closing arguments at the end of the hearing are now commonly put in writing and supplemented orally.

15.06

All of this may seem very familiar to an American attorney. But what is missing, apart from the witness giving direct testimony, is the possibility of obtaining oral deposition testimony from witnesses in advance of the hearing. Apart from cutting out a large amount of work and expense, this may make, and often does make, for a significantly different trial than that put on in an American court. The US attorney going into trial in an American courtroom will usually have had a chance to cross-examine the opposing party’s important witnesses in advance. Moreover, except in the unusual situation in which the lawyer decides to hold back a critical piece of information for cross-examination at trial, the lawyer will know roughly what the witness is expected to say on all of the issues and facts. The cross-examining lawyer will almost certainly have found out which witnesses can speak to the topics of significance. American arbitral tribunals are often in a position to order depositions.10 Where depositions have been obtained, the result is that the oral examination of witnesses in an American arbitration or trial can often be very brief.

15.07

Consider, in contrast, the position in an English arbitration. The parties will have exchanged witness statements (usually simultaneously but sometimes sequentially) and perhaps supplementary witness statements. These statements are often the product of work carefully carried out between the witness and the lawyers for the party tendering the witness. Cupboards with skeletons will generally have been kept firmly locked. The evidence of that witness will be directed towards the topics in which the party tendering the witness is interested. The impact of this approach was strikingly illustrated in a

15.08

9 Depositions are discussed in greater detail in ch 16 below, paras 16.96–16.102. 10 In Commerce and Industry Insurance Co of Canada v Lloyd’s Underwriters [2002] 1 WLR 1323, the English court declined to assist a New York arbitral process by compelling witnesses resident in England to provide depositions. The court concluded that the depositions were in the nature of oral discovery rather than for the purposes of providing evidence at the substantive hearing.

314

The Course and Conduct of a Bermuda Form Arbitration in London

piece of litigation where three different parties (the plaintiff, defendant and a third party) approached a witness to provide a witness statement, and the witness willingly obliged, providing a statement for each of the three parties. Each of his three statements was honest and accurate. But each was ‘spun’ by the party taking the statement in order to advance the case which it was presenting, and therefore addressed the issues which that party wanted to address. Accordingly, in English practice, the pre-trial ‘agenda’ as to what the witness will discuss is set by the party calling the witness. The opposing party will not, in advance of the trial, have had the chance to ask the witness about the topics on its agenda. As a result, the process of cross-examination at an English arbitration differs from that which would take place if there had been oral depositions beforehand. The cross-examiner may be interested in exploring what the witness knows about a topic, for example, with a view to finding out what was going on behind the scenes. He or she will seek to use the cross-examination as a means of bringing out the whole of the story relevant to the client’s case, instead of the partial picture that may emerge from the witness statements.

THE ‘PLEADINGS’ STAGE 15.09

There has always been flexibility as to the way in which arbitrations are conducted, a flexibility that is reinforced by section 34 of the 1996 Act. In theory, there is no reason why arbitration should require the production of written pleadings or statements of case at the outset of the dispute. Nevertheless, over the course of several decades, the basic procedures that apply in English High Court proceedings have come to be widely adopted in English arbitration. This is partly because the arbitral process has been shaped by lawyers acting as counsel and frequently sitting as arbitrators. It is also partly because some of the High Court procedures, such as the production of written pleadings, are seen as beneficial in terms of crystallising the issues in dispute. It is partly because some mechanism for defining the issues in documentary form is so evidently necessary to rational decision making that, albeit under different names (whether ‘memorials’, ‘pleadings’ or the settling of ‘terms of reference’), it forms a common part of most arbitral regimes. The result has been that it is unusual, if not unheard of, for the parties or the arbitrators to dispense with, or substantially modify, the stage of written pleadings. In Bermuda Form arbitrations, written pleadings are the norm, and they should be designed to enable both the parties and the tribunal to understand the issues in dispute. But how detailed should these pleadings be?

15.10

What will perhaps surprise many US attorneys involved in London arbitrations (not only Bermuda Form arbitrations) is the detail into which pleadings often descend. The concept of ‘notice’ pleading used in many US courts, which involves a very brief description of the case to be advanced, is generally not followed in London. English lawyers tend to start from the assumption that pleadings should be detailed. There is a tendency for some lawyers to use them as an opportunity to argue the case in detail or to highlight the supposed strengths of their case or weaknesses in that of the opposing party. As a corollary, a pleading that contains insufficient detail is likely to be the subject of a request for ‘further particulars’ or ‘further information’. A lengthy pleading may also result from

The ‘Pleadings’ Stage

315

a defensive approach by the lawyers to the case. They may wish to ensure that every possible point is pleaded so as to avoid a later argument that an unpleaded argument is being advanced and may be precluded. The lawyers may therefore think that it is better to put every conceivable argument forward on the pleadings, on the basis that it is easier to drop a point than to apply to amend the pleadings to include a new one. If anything, this tendency towards voluminous pleadings has been increasing. The practice in international arbitration generally tends to prolixity, with memorial-style pleadings which sometimes descend in great detail into evidence and legal argument, and Bermuda Form arbitrations are not immune from that pressure. The result of all this is that the pleadings in many commercial arbitrations can easily fill a lever-arch file (binder), and often several such files. These files may contain a statement of the claimant’s case, defence and counterclaims, a reply and defence to counterclaims, and perhaps a reply to that defence. Requests for further particulars of each of these documents may have been requested and provided. London arbitration pleadings often contain even more detail than the particularity required for fraud claims under the Federal Rules of Civil Procedure and related state rules of procedure used in US courts. The detail approaches (and may even exceed) the responses to interrogatories in which a party is asked to set out all facts on which it relies in support of its case. These extensive pleadings are not found in many American arbitrations.

15.11

What might then also surprise US attorneys is that the lever-arch binders containing the pleadings will rarely be referred to at the ultimate substantive hearing. They might occasionally be dipped into if an issue arises as to whether a party is straying beyond the case that was initially pleaded. But advocates tend to think that these so-called ‘pleading’ points do not go down well with arbitrators. By the time of the ultimate hearing, the shape of the parties’ cases will have become apparent from the witness statements, experts’ reports and written arguments filed in preparation for the hearing. Almost inevitably, the parties’ respective cases will have developed from the position adopted by the parties at the outset of the proceedings. Arbitrators are generally not interested in the route that the parties have travelled to get to where they are. Rightly, arbitrators are more interested in what the parties’ cases are, not what they were. This is not to say that a significant alteration in a party’s case cannot influence a tribunal. For example, it might reveal that a witness has undergone an important alteration in his or her recollection of some critical event. More frequently, however, a change in case will simply reflect the fact that the party’s thinking has developed as the case has proceeded, perhaps in the light of discovery or witness statements, or perhaps simply because a fresh view has been taken.

15.12

So, it might be asked, what purpose do these elaborate pleadings serve? Could they not be replaced by the concept of American ‘notice’ pleading, with each party’s case being allowed to develop as the case runs along? Why not simply see what is said in the witness statements and briefs for the hearing, and require the parties to respond to that? We think that there is much to be said for the view that pleadings in international arbitrations need not follow, and should not be required to follow, the pleadings that have been required or have become common in English court proceedings. Indeed, the Commercial Court in England has reacted against the prolixity of pleadings by imposing

15.13

316

The Course and Conduct of a Bermuda Form Arbitration in London

a 25-page limit. Permission is required for longer documents and is (in principle, at least) granted only if there are ‘good reasons’ to do so.11 We think that a tribunal appointed in a Bermuda Form arbitration should recognise the international character of the dispute, the parties and their legal representatives, and should be flexible as to the way in which a case is pleaded. There is no reason why a party’s American attorneys should not draft the pleadings in the way in which they consider it appropriate, bearing in mind that the essential function of the pleadings is not to argue the case or to set out the evidence in detail, but to identify the issues. It is sometimes helpful to the tribunal for key documents to be annexed to the pleading, so that the arbitrators can refer to these documents when they are reading into the case at the interlocutory stages. We think that defences or replies that consist of a series of ‘admissions’, ‘non-admissions’ and ‘denials’, and that leave the case to be advanced as opaque, are best avoided. Whilst such pleadings used to be common in English High Court litigation, they are not consistent with the spirit of international arbitration and are falling into disfavour with many international arbitrators. It is now common, for example, for a defence to identify the principal responses in an introductory section, followed by responses to particular paragraphs. Whilst extensive citation of case law is not required, it is permissible and may be useful to refer to any key New York law authorities which establish an important point of principle which is relevant to a party’s case. 15.14

Although it rarely happens, the parties may consider limiting the length of pleadings; for example, the parties might agree that the statement of case and defence should not exceed 20 pages, and the reply should not exceed five or ten pages. If these short pleadings genuinely give rise to difficulties in understanding the case to be advanced by one or other party, then a request for further information or clarification could be made, but this should not be seen as an opportunity to impose a burdensome series of questions on the other party with a view to highlighting the weaknesses in its case. Indeed, the latter tactic often backfires, because it simply identifies to the other party the areas on which it needs to focus in order to strengthen its case.

15.15

Questions sometimes arise in the interlocutory stages, or during the course of the substantive hearing, as to whether or not a particular argument strays beyond the case that a party has pleaded. Provided that the pleadings contain the core of the case advanced, an arbitrator is not likely to be unduly concerned at alterations to a party’s case which are within the spirit of what has been previously set out in writing. If the argument is genuinely not covered by previous pleading, a party will need to ask the tribunal for permission to amend to raise the new point. The tribunal has a discretion about the extent of amendments to the parties’ original statements of case.12 Generally, arbitrators are likely to allow an amendment provided that the opposing party can deal with the new argument in a satisfactory manner. An amendment made well before the substantive

11 The recommendations in paras 44–50 of the Report and Recommendations of the Commercial Court Long Trials Working Party (December 2007) have been substantially adopted in the Admiralty and Commercial Court Guide, 10th edn (2017) s C1.2. 12 English Arbitration Act 1996, s 34(2)(c).

The First Order for Directions

317

hearing is likely to be allowed. An amendment made at the substantive hearing may be allowed if it does not really take the other party by surprise and the existing witnesses can deal with it. If the amendment is so significant that it requires an adjournment of the substantive hearing to allow the parties to deal with the point, then the tribunal may well be reluctant to allow it.

THE FIRST ORDER FOR DIRECTIONS Neither the Bermuda Form itself nor the English Arbitration Act 1996 prescribes directions for any of the stages of the arbitral process. Accordingly, the procedures to be used in a particular arbitration will be set out in ‘directions’ that are either agreed by the parties or ordered by the arbitrators. Directions will include a schedule for exchange of pleadings, for document production, and for exchange of witness statements and expert reports. In addition, the directions may address both procedural and evidential matters relevant to the substantive hearing. The procedural matters may include, for example, the allocation of time at the substantive hearing. The parties may also agree on evidential matters, for example, making it clear that New York law is a matter for submission not evidence, or that significantly diminished weight is to be accorded to the evidence of a witness who is not tendered for cross-examination.

15.16

At an early stage, after the appointment of the tribunal, it is likely that one or other party will ask the tribunal to make a first order for directions. The initiative is likely to be taken by the claimant (whether policyholder or insurer) since usually the claimant wants to drive the arbitration forward. The parties should, as already indicated, co-operate on agreeing directions. They may be able to agree on the basic timetable necessary to take the matter through to a final substantive hearing, the key events being disclosure of documents, witness statements, experts’ reports (if any) and the substantive hearing itself.

15.17

The parties may also find that, at an early stage of the proceedings, the tribunal wishes to formalise the terms upon which it is appointed and draws up terms of appointment for signature by the parties. These terms will likely deal with matters such as the remuneration of the arbitrators, including cancellation fees, payments of deposits and the tribunal’s immunity from suit. These terms may also address procedural matters which might otherwise be addressed in the first order for directions.

15.18

Not infrequently, one party wishes a more accelerated timetable than the other. In this event, it will be necessary to obtain the tribunal’s decision. It is the duty of an arbitral tribunal to ‘adopt procedures suitable to the circumstances of the particular case, avoiding unnecessary delay or expense, so as to provide a fair means for the resolution of the matters falling to be determined’.13 A tribunal is likely to be sympathetic to the party that wants an early resolution, provided that an early resolution is consistent with fairness

15.19

13 Ibid,

s 33(1)(b).

318

The Course and Conduct of a Bermuda Form Arbitration in London

for all parties. In practice, a key factor is often the availability of the tribunal. If the tribunal is to take an early decision on timetabling or any other interlocutory matters, it will want to have a reasonable idea as to the issues in the substantive case. Accordingly, it will usually be important for the statement of case and defence to have been served before the tribunal can sensibly rule on any disputed issue relating to the future conduct of the case. 15.20

The conduct of the case will often be discussed at an initial interlocutory hearing attended by the parties. It is by no means essential for this hearing to involve a physical meeting of all parties and the tribunal. Even before the COVID-19 pandemic in 2020, it was not unusual for a hearing to take place by telephone or video conference call. ‘Remote’ hearings, which involve all participants attending using a video conference platform, are now increasingly common both in the English Commercial Court and arbitration. If there is good co-operation between the parties, then there is no reason why the initial hearing and directions order cannot be dealt with in a telephone or by video conference. This should be possible even if the initial hearing raises matters which are difficult and contentious.

15.21

The initial directions need not try to deal with all procedural and evidential matters that might ultimately need to be resolved. It may be better to deal with the framework of the arbitration at the first hearing, leaving particular issues to be the subject of later directions orders in the event that the parties cannot agree on them. The parties may agree to delegate certain future interlocutory decisions to the chair of the panel, so that these decisions can be dealt with swiftly and without having to convene all the members of the tribunal. This may be appropriate, for example, in relation to any disputes as to the scope of discovery. The chair of the panel will usually, however, reserve the right to involve the other panel members in the event that the subject matter of the application requires their involvement; for example, one party may apply to adjourn a hearing date which has previously been set, or may request that the substantive hearing should address certain issues but not others.

15.22

It is necessary for the first procedural order for directions to address only essential matters such as those relating to the service of pleadings, disclosure of documents, and other important scheduling matters. Parties also sometimes use this procedural order as the vehicle for agreeing or recording other matters such as the venue for the arbitration, the applicable law, the confidentiality of the arbitral process and the manner in which later documents are to be served. If the parties are agreed that each side shall bear its own costs of the arbitration, a signed procedural order is a convenient way of giving effect to this agreement.14 A specimen first directions order (which includes a timetable leading to a hearing) is set out at the end of this chapter. The specimen includes matters which are sometimes covered in the tribunal’s terms of appointment.15

14 The provisions to this effect in the early versions of the Bermuda Form are not effective under English law. A valid agreement for each party to bear its own costs must be concluded after the dispute has arisen; see Arbitration Act 1996, s 60. See also ch 11, paras 11.86–11.87 above and ch 17, para 17.31 below. 15 See above para 15.18.

Confidentiality

319

CONFIDENTIALITY Parties sometimes wish to include a confidentiality provision in the first directions order. The confidentiality of the arbitral proceedings, and documents exchanged within those proceedings, may be particularly important in the context of a Bermuda Form arbitration. For example, litigation of the underlying claims against the policyholder may be continuing in the United States or elsewhere, and the policyholder may wish to avoid documents exchanged in the arbitration becoming discoverable in underlying litigation against it. A policyholder may also seek to resist, on the grounds of the confidentiality of the arbitration process, an insurer’s request for materials, such as the pleadings or transcripts or award, relating to a dispute in a layer beneath the insurer’s. The insurer may also wish to maintain confidentiality. For example, the insurance company may be facing claims from a number of different policyholders involved in the same underlying problem, and the insurer may not want other policyholders to become aware of arguments successfully deployed in arbitration against the company. Indeed, looking at the position more generally, the fact that all of the insurer’s Bermuda Form disputes are arbitrated rather than litigated means that no body of precedent is built up that a policyholder can rely upon in its claims against insurers. Insurers can legitimately re-argue points – for example, points of policy interpretation – that have been decided against them on more than one occasion in the hope of persuading a particular arbitral tribunal of the merits of their position.

15.23

English law protects the confidentiality of arbitration proceedings, and, indeed, confidentiality is a term implied into arbitration agreements as a matter of law.16 Accordingly, neither party should disclose or use for any purpose (other than the arbitration itself) any documents prepared for and used in the arbitration, or disclosed or produced in the course of the arbitration, or transcripts or notes of the evidence in the arbitration or the award, nor should it disclose what evidence has been given by any witness in the arbitration. The obligation is not limited to information which is inherently confidential, such as trade secrets.17 There are exceptions to this general principle. However, English common law does not clearly define where the boundaries lie, and the topic was considered too difficult to be the subject of legislation and codification in the English Arbitration Act 1996.18 The content of the obligation may depend upon the context in which it arises and on the nature of the information or documents at issue, and the limits of the obligation are still in the process of development on a case by case basis.19 The current authorities indicate that disclosure is permissible in the following cases: the

15.24

16 Ali Shipping Corp v Shipyard Trogir [1999] 1 WLR 314; Michael Wilson & Partners Ltd v Emmott [2008] EWCA Civ 184, [2008] 1 Lloyd’s Rep 616. 17 Michael Wilson & Partners Ltd v Emmott, [105]. 18 See paras 11–17 of the report of the Departmental Advisory Committee (February 1996), reproduced in Michael J Mustill and Stewart C Boyd, Commercial Arbitration 2001 Companion (London, Butterworths, 2001) 397–99. See also ibid at 112–13; Associated Electric and Gas Insurance Services Ltd v European Reinsurance Co of Zurich [2003] UKPC 11, [2003] 1 WLR 1041; Department Of Economics Policy & Development of the City of Moscow v Bankers Trust Co [2003] EWHC 1377 (Comm), [2003] 1 WLR 2885 (Cooke J), [2004] EWCA Civ 314, [2005] QB 207 (CA). There is now a significant amount of academic writing on the subject of confidentiality. 19 Michael Wilson & Partners Ltd v Emmott, [107] (Collins LJ) and [129(vi)] (Thomas LJ).

320

The Course and Conduct of a Bermuda Form Arbitration in London

consent (express or implied) of the party that produced the document in the arbitration; an order or leave of the court; where it is reasonably necessary for the protection of the legitimate interests of an arbitrating party; where the interests of justice require disclosure; and perhaps where the public interest requires disclosure.20 15.25

A dispute between the parties to a pending Bermuda Form arbitration as to the existence or scope of their confidentiality obligations probably falls within the scope of the arbitration clause. Accordingly, the dispute should be referred to the tribunal rather than the English court.21 If a confidentiality provision is to be included in the directions order, the parties may wish to make it clear that disputes as to confidentiality are to be resolved by the tribunal.

15.26

In addition to including a confidentiality provision in the directions order, the parties may also draft and sign an express confidentiality agreement which binds themselves and indeed the arbitrators. Such agreements can have unforeseen consequences. In Associated Electric & Gas Insurance Services Ltd v European Reinsurance Co of Zurich,22 a party (unsuccessfully) relied on a confidentiality agreement in order to preclude an arbitral tribunal from knowing about, and giving effect to, a previous award between the same parties relating to the same contract. The advantage of confidentiality agreements, however, is that they make express that which would otherwise have to be implied.

15.27

The existence of an express or implied confidentiality agreement does not prevent the disclosure, in subsequent English court or arbitration proceedings, of documents produced or generated during the course of an earlier arbitration. Confidentiality is a factor to be taken into account, but it is not decisive.23 The extent to which an express or implied confidentiality agreement is effective to prevent discovery in the United States is, however, a matter which depends upon the law applied by a particular court in the United States.24

DISCOVERY OF DOCUMENTS 15.28

Whether one uses the traditional common law term ‘discovery’, the modern English jargon of ‘disclosure’ or the arbitration community’s preferred term ‘document production’, the 20 Ibid, [107] (Collins LJ). 21 This was the tentative conclusion of Collins LJ in ibid at [84], [110], and the firm view of Thomas LJ at [116]–[124], [129(ix)]. If the arbitration has concluded, then the English court can determine issues relating to confidentiality, for example, an application to use confidential arbitration documents for the purposes of other proceedings; see Westwood Shipping Lines v Universal Schiffahrtsgesellschaft MbH [2012] EWHC 3837 (Comm) [6]. 22 [2003] 1 WLR 1041. 23 Dolling-Baker v Merrett [1990] 1 WLR 1205; Michael Wilson & Partners v Emmott at [72], [73], [127]. See further ch 16 below, paras 16.93–16.94. 24 For examples of cases where US courts have upheld the confidentiality of private alternative dispute resolution proceedings, see, eg, United States v Gullo 672 F Supp 99 (WDNY 1987) (the government failed to make adequate showing of the need for disclosure of statements made during a confidential dispute resolution process); Wright v Brocket 571 NYS2d 660, 663 (Sup Ct 1991) (‘[A]ll statements and documents involving [an alternative dispute resolution proceeding] are confidential and may not be used in any proceeding for any purpose’); Williams v State 770 SW2d 948, 949 (Tex App Ct 1989) (‘[W]e cannot consider any evidence from the

Preliminary Issues or ‘Bifurcation’

321

identification and exchange of relevant documents by each party to the other gives rise to a number of problems in Bermuda Form arbitrations. In particular, questions can arise about the extent to which the policyholder must produce documents generated in the underlying litigation, which may involve thousands, even tens or hundreds of t housands of cases. Where this litigation is mass tort or multi-site environmental litigation, the production of all underlying materials could be extremely burdensome. The problems have been exacerbated by the exponential proliferation of electronic documents, and the constant march of technology to handle them. Issues also arise as to the extent of privilege, and whether a policyholder should waive privilege. Discovery is a topic that merits separate discussion, and it is covered in Chapter 16. In the present context, however, it should be noted that discovery is an area where co-operation between the parties is essential and will be insisted upon by the tribunal. It is possible that the tribunal will be called upon to rule on specific questions, but the parties should work to reach agreement as far as this is possible, and the tribunal should not routinely be copied into the parties’ correspondence relating to discovery. It is often sensible for the parties not only to identify the documents that they want from the other party, but also to state in broad terms the nature of the discovery that they are themselves providing. In particular, a policyholder may be able to avoid later criticisms of its discovery if it tells the insurer, at an early stage, the categories of documents that it intends to disclose. If the insurer does not take issue with those categories, or if the tribunal endorses the policyholder’s approach, it will be difficult for the insurer to criticise the policyholder at a later stage.

15.29

A regrettable feature of English litigation and arbitration practice is the tendency, which often comes as a surprise to US lawyers, for late requests to be made for discovery, including requests made during the substantive hearing itself. In the United States, absent sufficient cause demonstrated for the delay, the court or tribunal likely will deny a request made late in the proceedings as a delaying tactic or as a request that the requesting party could, and should, have made much earlier. Late requests often result from the fact that gaps in the documentation become apparent when one or other party carries out the process of ‘getting the case up’ in detail in the weeks before the substantive hearing. Late requests of this kind should be avoided, although they may be legitimate if the case has taken a particular turn that was not anticipated at an earlier stage or if the statement of a particular witness has suddenly brought a particular matter into focus. If the parties agree on the scope of discovery at an early stage, then the scope for later disagreement and late applications may be minimised.

15.30

PRELIMINARY ISSUES OR ‘BIFURCATION’ The parties should consider, during the interlocutory stages, whether it is appropriate for the tribunal to determine certain issues before others.25 In England, these tend dispute resolution procedure that appellant and complainant participated in … as disclosures made in an ADR procedure are confidential, and not subject to disclosure’). 25 Section 47 of the English Arbitration Act 1996 gives the tribunal the power to make more than one award at different times on different aspects of the matters to be determined.

15.31

322

The Course and Conduct of a Bermuda Form Arbitration in London

to be known as ‘preliminary issues’. In American practice, this process may involve motions for summary judgment or a motion to bifurcate or sever certain issues. Whether or not resolution of preliminary issues is appropriate depends upon the facts of each case. It is sometimes possible to identify a key issue of interpretation that will resolve the dispute one way or the other. More frequently, however, issues of interpretation are intimately connected with factual questions, and a party may well find that its interests are best served by having one substantive hearing that resolves all matters in dispute at one time.

WITNESS STATEMENTS 15.32

Direct evidence is presented in a party’s initial witness statements, which, as explained below, are usually exchanged simultaneously by the parties. Parties often agree to allow for the service of supplemental witness statements that respond to points raised in the opposing party’s initial witness statements. It is possible that a party will find it necessary to obtain a statement from an additional witness in order to respond to a point made in the statements served by the other side. It is therefore sensible to make it clear in the procedural order for directions that ‘supplemental’ statements are not confined to statements from witnesses who provided evidence in the initial exchange.26

15.33

The preparation of witness statements now forms a key part of the process of preparing cases for hearing, whether in arbitration or English High Court proceedings. Generally speaking, there is no examination in chief, or what in the United States is called direct examination of witnesses, in London arbitration or English civil litigation. Some regret this development. Examination in chief gave the witnesses the chance to tell their stories. Some regarded the process of getting the witness to tell his or her story, by means of questions that were not ‘leading’, as the most important task of the advocate.27 Instead, witnesses must now tell their story in writing and, at the hearing, they are then immediately pitched in to face cross-examination by the opposing advocate, who inevitably has an alternative agenda to pursue. A curious byproduct of this approach is that it is the respondent who, through cross-examination of the claimant’s witnesses, makes the first detailed presentation.

15.34

Lawyers have recognised that considerable care is required in relation to the preparation of witness statements. Regrettably, this has led to various problems, including: the over-lawyering of witness evidence; witness statements of excessive length, containing a recitation of documents which the tribunal could read for itself; and generally a tendency to lose sight of the function of a witness statement, which should be for the witness to give factual evidence in the witness’ own words.28 Tribunals do, of course, understand

26 This is expressly provided for in Article 4(6) of the IBA Rules on the Taking of Evidence in International Arbitration 2010. The parties may agree that the IBA Rules should apply generally in the arbitration, including in relation to statements of witnesses of fact (see Article 4) and expert witnesses (see Article 5). 27 See, eg, George Keeton, Harris’s Hints on Advocacy: the Conduct of Cases, Civil and Criminal, 18th edn (London, Stevens & Sons, 1943). Direct testimony remains an important feature of criminal trials in England. 28 See, by analogy, the Admiralty and Commercial Court Guide, 10th edn (2017), section H1.1.

Witness Statements

323

that witness statements are usually the product of a joint effort between witness and the party’s lawyer.29 When it comes to critical questions, however, tribunals can usually detect the lawyers’ carefully crafted phrases, especially when put into the mouth of a witness for whom the language of the courtroom and the brief is completely unnatural. This, coupled with the absence of examination in chief, often results in more attention and weight being given to what witnesses say under cross-examination than in their witness statements. Although witness statements were originally an innovation introduced in the Commercial Court, perceived problems with witness statements in commercial litigation in England led to a review of the process and a report in December 2019.30 Paragraph 44 of that report contains guidance which may be of some value in the context of arbitrations conducted in England, notwithstanding that arbitration practice and procedures do not and should not mimic those of English High Court litigation. The guidance may, however, assist the parties as a means of enhancing the utility of witness statements for a tribunal, and the weight which the tribunal may attach to them. This guidance is that:

15.35

(i) a witness statement should be confined to the evidence that the witness would give if properly examined in chief; it must be the evidence a witness would give if asked open (non-leading) questions about his or her recollection of events by a competent advocate with an understanding of the issues in the case; (ii) a witness statement must use the witness’ own words, based on his or her own recollection, with revisions limited to aiding brevity and clarity without changing meaning or emphasis; the evidence which a witness is able to give must in no circumstances be altered, distorted or spun in order to seek to help the case of the party calling the witness; (iii) the content of a witness statement must be regulated by the parameters of the relevant issues and by the relevant evidential rules; (iv) the focus of a witness statement must be on its utility to the trial judge in presenting accurately the witness’ own recollection, not as a tool for internal purposes or presentation to the other side; (v) lawyer assistance and input into the preparation of a witness statement is useful if not essential, but must be provided with conspicuous care with the above principles in mind and conscious of the risk of corrupting memory through the process. In drafting witness statements, the parties and their advisers will need to give thought to the points they need the witness to bring out. A well-prepared witness statement should not be a lengthy description of the chronological sequence of events seen from the witness’ perspective. Decisions will need to be made as to which were the important events and

29 This is recognised in Article 4(3) of the IBA Rules on the Taking of Evidence in International Arbitration 2010, which states that that it is ‘not improper’ for a party and its lawyers to interview witnesses and potential witnesses and discuss their testimony. 30 Witness Evidence Working Group, Factual Witness Evidence in Trials before the Business and Property Courts (2019), https://www.judiciary.uk/wp-content/uploads/2019/12/Witness-statement-working-group-FinalReport-.pdf.

15.36

324

The Course and Conduct of a Bermuda Form Arbitration in London

what the witness can say about them. It may therefore be useful for the statement to identify, at the outset, the particular topics that are to be addressed. 15.37

Witness statements cannot be drafted and exchanged unless the lawyers have a clear understanding of the legal case advanced by each side. Because there is no oral evidence in chief or direct evidence at the hearing, the witness statements will need to cover all the points that a party needs a witness to prove in support of its legal case. The lawyers will need to think about the points the other side are asserting and therefore intending to prove. If the opposing party’s case as set out in written pleadings is to be disputed in some material respects, the witness evidence will need to deal with that.

15.38

‘Hostages to fortune’ will, of course, need to be avoided. Some lawyers prefer to take a minimalist approach to witness statements: do not say something unless it needs to be said. That approach is consistent with a focus on the testimony that a witness would be asked to give if asked questions by way of direct examination. It may be sensible for witnesses to avoid, unless necessary, committing themselves to a description of what their thought processes were some time ago: they are unlikely to remember clearly, very often it does not matter, and claims to recollection may provide an opening to the other side in cross-examination. If, however, some important matter is omitted from a witness statement and then emerges in oral evidence, it will likely be the subject of adverse comment from the opposing party. The witness who, in the course of his or her evidence, suddenly remembers a crucial telephone conversation is unlikely to impress.

15.39

The practice in England is for witness statements to be drafted as a narrative description. Some US attorneys will be more familiar with statements that consist of written questions being posed and answered. Although this is unusual in England, there is no reason why this approach should not be taken in a Bermuda Form arbitration. It is common, although not essential, for a witness to attach any key documents cited in the witness statement. This practice serves as a useful discipline, since it should force the parties and the witness to look carefully at the documents that are attached. If the number of documents referenced in the witness statement is voluminous, it may be sensible to limit the number of exhibits. Tribunals (and indeed the parties) find it helpful if, in due course, the statements are cross-referenced to the pages of the trial bundles, and this is now done as a matter of course.

15.40

The usual practice is for witness statements to be simultaneously exchanged, with each side thereby revealing its factual evidence at the same time, although sequential exchange is sometimes ordered (for example, on a particular issue). As previously stated, arbitral procedural directions will often provide for the exchange of supplementary statements. English arbitrators will, however, expect the initial round of witness statements to address not only the facts necessary for a party’s affirmative case, but also the factual issues raised by and relevant to the opposing party’s case. In other words, responding to the opposing factual case is not simply a matter for supplemental statements, but should be included in the statements which are the subject of the initial exchange. However, lawyers may exercise judgment as to the detail in which a party needs to address the opposing case, particularly if the case to be advanced is not clear. For example, one party may have raised an estoppel plea in very general terms. The opposing party may legitimately decide to

Expert Evidence

325

address that plea very lightly in its first round of witness statements, and wait to see how the case is put before deciding whether to serve more detailed evidence in a supplemental witness statement. Even if no specific direction exists, a party will in practice be allowed to serve a supplementary statement, provided that the opposing party has sufficient time to review and if necessary respond to it. Supplementary statements provide an opportunity to deal with any unanticipated points arising from the other side’s statements. They also provide an opportunity to deal with any points that may not have been adequately covered in the first round of witness statements. In a Bermuda Form arbitration, it is likely that the policyholder, rather than the insurer, will adduce the preponderance of factual evidence. Indeed, unless the issue is misrepresentation, the insurer may be able to adduce no factual evidence at all, but may confine itself to questioning the policyholder’s witnesses in order to establish that the policy did not cover the loss.

15.41

EXPERT EVIDENCE Under English court procedure, the function of expert evidence is essentially to educate the court on matters of genuine technical expertise on which the court could not form a view without specialist instruction. In R v Bonython,31 the court defined the circumstances in which expert evidence was permissible in the following terms:

15.42

(a) whether the subject matter of the opinion is such that a person without instruction or experience in the area of knowledge or human experience would be able to form a sound judgment on the matter without the assistance of witnesses possessing special knowledge or experience in the area and (b) whether the subject matter of the opinion forms part of a body of knowledge or experience which is sufficiently organised or recognised to be accepted as a reliable body of knowledge or experience, a special acquaintance with which the witness would render his opinion of assistance to the court.

An arbitral tribunal is, of course, not bound by the approach taken in English court procedure, but it is likely to take a broadly similar attitude. The parties will often be able to agree upon the need for expert evidence and the nature of the experts to be called. For example, an issue on ‘expected/intended’ may give rise to the need to tender expert scientific evidence in order to explain technical matters. If the parties agree that experts should be called, the tribunal must, as with other procedural matters which are agreed between the parties, give effect to the parties’ agreement.32 In the event of disagreement, the 31 R v Bonython (1984) 38 SASR 45, 46 (South Australian State Reports). R v Bonython has been cited with approval in a number of cases, including by the UK Supreme Court in Kennedy v Cordia (Services) LLP [2016] UKSC 6, [43]. The case contains a valuable discussion as to the role of experts: see [38]–[61]. See also Red Sea Tankers Ltd v Papachristidis and Others [1997] 2 Lloyd’s Rep 547, 597–98; Secretary of State for Trade & Industry v Baker and Others (No 5); Re Barings plc and Others (No 5) (Jonathan Parker J) [1999] 1 BCLC 433, 489–95. There are obvious similarities to the Daubert standard applied by federal courts in the United States (Daubert v Merrell Dow Pharmaceuticals Inc 509 US 579 (1993)), but without the same need for the court to act as ‘gatekeeper’ to prevent potentially misleading evidence going before a jury that may be poorly equipped to assess it. English courts are more inclined to treat factors going to the reliability or cogency of the expert opinion as matters for assessment by the trier of fact than admissibility as such. 32 See s 34(1) of the English Arbitration Act 1996.

15.43

326

The Course and Conduct of a Bermuda Form Arbitration in London

parties should expect the tribunal to want to understand why expert evidence is considered necessary for the fair resolution of the case. The procedural order for directions should provide for a mechanism by which the parties identify, at an early stage (for example, after service of the claimant’s reply), the disciplines of the experts that they intend to call, and a broad outline of the issues that they will address. It should also provide for the resolution of any dispute as to whether expert evidence in particular disciplines, or indeed generally, is appropriate. 15.44

Many Bermuda Form cases involve technical issues, and a party’s factual witnesses may well need to address and explain complex scientific matters. For example, if expected or intended issues arise in a pharmaceutical case, the policyholder will likely wish to adduce factual evidence from highly skilled individuals who worked for the company at the material time, and who will explain the technical development of the product and their understanding and expectations at that time. Such evidence may have an ‘expert’ flavour to it, but this is inevitable and unobjectionable; the evidence remains factual evidence about the witnesses’ thought processes and impressions, and is not offered as evidence of opinion as such. It may be sensible for the procedural order for directions expressly to state that this is permissible.

15.45

If there is a dispute as to whether or not expert evidence is appropriate at all or as to whether it is appropriate on a particular topic, the arbitrators will have to decide that dispute. The arbitrators are entitled to rule that such evidence is not to be tendered, for example, because they feel that it is irrelevant or peripheral. They may, for example, want to be satisfied that the evidence sought to be tendered does not violate the ‘parol evidence’ exclusion in the governing law clause.33 It is fair to say, however, that arbitrators will often be reluctant to prevent a party adducing the expert evidence that it says is necessary in order to support its case, and where an arguable case can be made to this effect. Directions as to expert evidence are frequently sought and given at a time when the arbitrators are still feeling their way into the dispute. Whilst a tribunal may query the need for expert evidence on a particular issue, it may not feel sufficiently confident of ruling it out altogether, particularly since the shape of the ultimate arguments will not be clear at an early stage. In order to deter unnecessary expert evidence, the tribunal may well give warnings as to the consequences in costs if the evidence proves unnecessary. It may also give a direction which limits the amount of time that expert evidence can occupy at the hearing.

15.46

The practice in England is, generally speaking, for experts’ reports to be exchanged simultaneously. It sometimes happens that sequential reports are ordered. This may well happen if the parties dispute the necessity for expert evidence. The tribunal may feel that the need for expert evidence is uncertain and that the party who says it is necessary should therefore serve its report on the other side before receiving any report addressing the expert issues. The other side would then have the opportunity of considering that report as part of the process of deciding whether expert evidence in response is necessary.

33 Article

VI(O) of the 004 Form; and Article V(q) of the 001–003 Forms. See also ch 3 above.

Expert Evidence

327

If expert reports are ordered to be exchanged, under the practice in England, the report should address the technical issues raised by both sides in their respective cases. In other words, the claimant’s expert report will not simply address the issues that the claimant has set out to prove; it should also address the issues that the respondent has raised.34 This is in contrast to some US practice, where the expert’s comments on the other side’s technical case are dealt with only in a response report. Although the form of the expert report is subject to the directions agreed between the parties or ordered by the tribunal, English arbitrators are likely to be familiar with The Ikarian Reefer,35 a case that sets out the form an expert report should take for an English High Court proceeding. This form of report sets out the questions on which the expert was asked to opine, the expert’s opinions, the bases for the opinions, and matters relied on in reaching those opinions. The principles derived from that decision, which are now embodied in the practice of the English Commercial Court,36 are designed to ensure that, as far as possible, the expert genuinely gives his or her technical views and does not simply act as an advocate for a party. The tribunal may wish to adopt these court principles in the arbitration.37 Because there are no depositions of experts in advance of the hearing, the expert report is expected to be detailed.

15.47

After the expert reports have been exchanged, English counsel or one of the English arbitrators may suggest that the experts meet to determine which points can be agreed and which remain in dispute. This manner of narrowing the expert issues is gaining in popularity in England. The expert meeting process makes many American lawyers nervous, because of the apparent loss of control over what may or may not be agreed among the experts. Although such meetings are not ordered by arbitration tribunals as a matter of course, the possibility that such a meeting may be ordered should be considered when selecting experts.

15.48

Recent revisions to the Federal Rules of Civil Procedure in the United States have narrowed the gap, noted in earlier editions of this book, between English and American law on what is discoverable from a testifying expert. Before revisions to the Federal Rule of Civil Procedure 26(b)(4) in the United States in 2010, once designated as a testifying expert, the expert lost the attorney work product protection that shielded the communications between the expert and the attorneys, and the expert’s work product up to that point. In December 2010, amendments to Rule 26 went into effect, making clear that work product protections extend to drafts of expert reports, regardless of the form in which the draft is recorded.38 The revised rule also protects, subject

15.49

34 Unlike American pleadings, English-style pleadings are usually sufficiently detailed to make reasonably apparent the matters subject to expert evidence. Accordingly, both sides will from the beginning of the case have a reasonable idea of the issues to which expert evidence is likely to be directed. 35 The Ikarian Reefer [1993] 2 Lloyd’s Rep 68. 36 See The Admiralty and Commercial Courts Guide, 10th edn (2017) Section H.2 and Appendix 8; Civil Procedure Rules, Part 35 Practice Direction. 37 Or the parties may wish to incorporate Article 5 of the IBA Rules on the Taking of Evidence in International Arbitration 2010. This requires a statement of independence and an ‘affirmation of his or her genuine belief in the opinions expressed in the Expert Report’. 38 Fed R Civ P 26(b)(4)(B) (rev’d effective 1 December 2010). The Advisory Committee Notes for the 2010 amendments to Rule 26(b)(4) explain the unintended consequences and costs from the 1993 version of the Rule which compelled the 2010 revisions.

328

The Course and Conduct of a Bermuda Form Arbitration in London

to three exceptions,39 disclosures to and communications between attorneys and the expert, regardless of whether those communications are written, electronic or otherwise. Similarly, New York procedural rules protect attorney–expert communications and related attorney work product and mental impressions.40 15.50

English law, by contrast, has always been protective of the privilege that attaches to communications with experts. Litigation privilege is not lost once an expert is designated to testify. Generally speaking, prior drafts and communications with the sponsoring attorney need not be produced, and only the theories finally presented in the expert report need be revealed. However, an expert can be asked in cross-examination to state the basis on which his or her opinion has been formed, and this may to some extent open up evidence or discussions that might have been privileged if the expert had not been called to testify.41

PREPARATION FOR THE SUBSTANTIVE HEARING 15.51

In the months before the substantive hearing date, the parties should give consideration as to how the documentation will be presented to the tribunal. The tribunal will have been provided with copies of the pleadings as they are served, and it may also be given the witness statements and expert’s reports. They will not, and should not, be given copies of all the documents copied or inspected as part of the discovery process. Nor will they expect to see all the correspondence exchanged between the parties relating to such issues.

15.52

The parties will need to co-operate on the preparation of trial exhibits which are organised into ‘hearing bundles’. Again, English and American practice diverge.

15.53

Typically, Americans are accustomed to working from exhibits to affidavits, witness statements or depositions. English lawyers and arbitrators like to work from documents that have been put together into suitably arranged ‘bundles’, that is, ring binders or (in modern practice) electronic document databases. This will usually include a chronological run of the important documents in the case, irrespective of which party or witness originally produced a particular document. Ideally, the key documents will also be presented, for the sake of convenience, in a ‘core’ bundle. It is not necessary or

39 The exceptions concern: information about compensation for the expert’s study or testimony; facts or data that the party’s attorney provided and that the expert considered in forming the opinions to be expressed; and assumptions that the party’s attorney provided and that the expert relied on in forming the opinions to be expressed. 40 New York Civil Practice Law and Rules (CPLR) § 3101(a)(1). These protections can be overcome only on a showing of substantial need. Ibid, § 3101(d)(2). 41 See General Mediterranean Holdings v Patel [2000] 1 WLR 272, 294–95. English High Court procedure now requires an expert to state the substance of all material instructions, whether written or oral, on the basis of which the report was written. The court has the power to order disclosure of these instructions if there are reasonable grounds to consider that the expert’s statement of instructions is inaccurate or incomplete: see Civil Procedure Rules 35.10. This is in effect a statutory inroad into privileged communications, and arbitrators do not have the same power. See further ch 16 below.

Preparation for the Substantive Hearing

329

appropriate to include every disclosed document in the bundles. It may be preferable to have a number of different chronological bundles dealing with different topics; for example, documents relating to the placement of the risk (if relevant) may best be kept separate from the policyholder’s documents relating to the development of the product that gave rise to the underlying case and the insurance claim. Regrettably, it has long been the practice in England for many more documents to be copied into hearing bundles than is necessary. Some lawyers simply copy the entire production of both parties and bundle it. This should not happen. The parties should attempt to keep the bundled documentation to the necessary minimum, and avoid adding documents which are not actually going to be required. Lawyers and tribunals are now increasingly familiar with electronic bundles, particularly as a result of remote hearings becoming commonplace during the COVID-19 pandemic. The parties should obviously discuss with the tribunal members whether they wish to have hard copy or electronic bundles of documents, or a mixture of the two. The parties may also wish to clarify and agree upon the evidential status of the documents which are contained in the hearing bundles. The practice in England is that either party can rely upon the documents as evidence in the case. The documents do not therefore have to be ‘proved’ by a witness in order to acquire the status of constituting evidence in the case. It will also be assumed that the documents in the hearing bundles are authentic. Accordingly, if one party challenges the authenticity of a document (something that would be very unusual in a Bermuda Form arbitration), it is preferable if not essential that the challenge is made known as early as possible; the order for directions may prescribe a deadline for such objections to be communicated.

15.54

The English practice of producing a common bundle of documents for use at the hearing enables each side to see, in advance of the hearing, the documents that the other side considers to be important. Each side will try to work out why the other side has decided to insert particular documents into the bundle, and the witnesses will therefore get some forewarning about the documents on which they are likely to be cross-examined. Previously, it was not unusual for parties to add documents to the bundle as the hearing progressed. This may be reasonable in circumstances where a new point has arisen or a witness has said something unexpected. However, it can lead to unfairness because a party and the witness had no forewarning that particular documents were considered relevant. It is therefore sensible for the procedural order for directions to contain provision for the circumstances in which a party can add additional documents to the common bundles, so as to ensure that these are added in sufficient time for the parties and their witnesses to consider them in advance of cross-examination. This can be subject to the overriding discretion of the tribunal to allow documents to be added at a later stage – for example, if a particular issue comes into focus and there are material documents that relate to it. In the United States, by contrast, file production of common bundles is far from usual. Instead, American cross-examiners will select in advance the documents from discovery that they wish to use in cross-examination, but often will produce these documents (assuming that they have not been used for other purposes at the trial) only in the course of the crossexamination itself. Obviously, this adds to the possibility of surprising the witness with a line of cross-examination.

15.55

330

The Course and Conduct of a Bermuda Form Arbitration in London

15.56

Both the tribunal and the parties themselves will find it useful if the witness statements and expert reports are cross-referenced to the main hearing bundle, so that the tribunal can quickly locate the relevant document in its chronological sequence.

15.57

Prior to the hearing, the parties will usually be asked to exchange written arguments. Within living memory, there was no practice or requirement in English courts or arbitration procedures for the parties to produce a written argument prior to the case; everything was done orally. This changed when the English Court of Appeal introduced a requirement for ‘skeleton arguments’. These were supposed to be a crisp written summary of the points to be advanced. As time has passed and computers have eased the task of producing documentation, ‘skeleton arguments’ have grown into written documents that are often as fleshy as many full-scale American trial briefs. For a Bermuda Form arbitration, each party should expect to produce a document that sets out its case in detail, since the tribunal will rely upon this document not only in reading into the case at the beginning but also in writing its award at the end. There will often be more time available to prepare the opening brief than a written closing. It should not be assumed that post-hearing written submissions will be permitted. Although practice varies, many tribunals prefer to complete the entire hearing process, including closing argument, in full at the hearing itself.

15.58

Arbitrators can be expected to pre-read a reasonable amount of material prior to the substantive hearing, and the parties are expected to co-operate in providing a reading list of useful material, including the important cases and key documents. As part of the pre-hearing directions, it is now increasingly common in London for the arbitrators to lay down time limits. Opening statements may be limited to, for example, one hour per party. The remainder of the time may, for example, be divided equally, so that it is up to each party to decide whether to use its allotted time in cross-examination, re-examination or applications to the tribunal.

THE SUBSTANTIVE HEARING 15.59

The parties and the tribunal will have discussed, at one of the directions hearings, the length of time required for the substantive hearing. It is important for the parties to set aside enough time to ensure that the hearing is completed in one sitting. If more time is needed, it may be difficult to reconvene the tribunal and the parties’ representatives for some months. The current tendency in international arbitration is for the oral phase of hearings to be shorter than perhaps used to be the case. One essential purpose of the oral phase is to enable the witnesses to be cross-examined. The other is to enable the tribunal to hear oral argument and, critically, to raise any questions of its own. It would be unusual for a Bermuda Form arbitration to last more than three weeks, and many arbitrations will be far shorter.

15.60

As in the United States, the hearing will commence with the parties’ opening statements. These are unlikely to last more than a day in total, and an hour per party may be sufficient.

The Substantive Hearing

331

Visual aids such as slides and electronic presentations can be used in argument, although this is by no means standard or typical practice in London. (In practice, the quality of the presentation is more important than its form: most Bermuda Form t ribunals consist of experienced lawyers and judges accustomed to bench trials and appeals, and they tend to fight shy of being treated as if they were juries.) The witnesses will then be called. The parties will be asked whether they require the witnesses to be sworn, and this is nearly always dispensed with in London arbitration, though tribunals often ask the witness to confirm that they will tell the truth before they begin to give evidence. In the United States, all witnesses are sworn as a matter of course. Unless either party specifically objects, witnesses are allowed to attend the hearings and hear the evidence of other witnesses, even if they have not yet given their testimony. It is now increasingly common for witnesses of lesser significance to give their evidence via video link, a practice that usually works well.

15.61

Prior to being called, a party may want to try to ‘prepare’ the witness for the cross- 15.62 examination that is in store, and indeed the witness will usually wish to have a good idea of what he or she might be asked. The party’s lawyers may be subject to professional or ethical rules in this connection.42 In any event, the experienced arbitrators who generally sit in Bermuda Form arbitrations are likely to identify a witness who is giving pre-rehearsed answers. In US practice, the claimant will usually call all its evidence, both technical and expert, before the respondent calls any evidence. In London, the practice has developed for the factual evidence on both sides to be given first, and for the experts to give their evidence after all of the factual evidence has been tendered. The idea behind London practice is that the experts should have heard all the facts before finalising their opinions.

15.63

Since witness statements have been served, there is usually no or minimal examination in chief. If issues have arisen that the witness has not addressed in the written evidence originally served, the witness will usually be expected to produce a further short written statement. After the witness has confirmed the accuracy of his or her written evidence (and the parties can even agree that this should be taken as having been done), the witness is then available for cross-examination. For reasons already given, cross-examination can often be a lengthier process than in the United States. American lawyers should also bear in mind the English rule of practice that a witness should be given the opportunity to deal with any significant areas where his or her evidence is challenged. In this context, there is an important divergence between the English and American approaches to crossexamination. Take a simple case where the issue is whether A met B on a particular occasion in New York and agreed something, and A’s case is that a meeting did take place with other people but B did not attend it. An English lawyer cross-examining B would

15.64

42 See, eg, Rule rC9.4 of the Code of Conduct of the Bar of England & Wales: ‘you must not rehearse, practice with, or coach a witness in respect of their evidence’. A party’s US lawyers may not be so constrained; indeed, they may regard it as normal preparation and as part of their duty to their client.

332

The Course and Conduct of a Bermuda Form Arbitration in London

be duty bound to put to B that he never attended the meeting in question, so as to ensure that B had a chance to deal with the allegation. An American attorney may approach the cross-examination differently, declining to ask B about the meeting at all. Instead, after B has finished giving evidence, the American attorney might intend to call another witness to show that B was in a meeting in London on the day in question, thereby proving B’s story to be untrue. Then, in closing submissions, the American attorney will submit to the tribunal that B lied in testifying about the meeting. Such an approach would likely evince a strong adverse reaction from English members of a tribunal, on the basis that B was never given an opportunity to respond about being in London on the day in question and that there was no challenge to the evidence concerning the whereabouts of B. As a further consequence of this approach, an advocate is expected not to ‘pull any punches’ when questioning the witness; if the allegation is that the witness is lying, that is something which should be put directly to the witness.43 15.65

A rule of practice in England is that, once the oral evidence has begun, a witness cannot discuss the case or the evidence with anyone until the witness finishes giving evidence. Accordingly, the lawyers have no opportunity to influence what the witness says once the evidence has started. This rule (known as the ‘Delaware rule’) is adopted in some, but by no means the majority, of jurisdictions in the United States. This rule has its most striking impact in relation to re-examination. Re-examination is permitted, but the re-examiner cannot talk to the witness about the topics that are the subject of re-examination. This makes re-examination difficult from both the re-examiner’s and the witness’ point of view. The re-examiner may not know how the witness is going to respond to any question. In addition, the lawyer cannot, or at least should not, put leading questions to the witness in re-examination; English panel members are likely to ignore answers elicited in response to leading questions in re-examination. Accordingly, the witness will have to be sufficiently astute to work out the points that the re-examiner wishes to hear. Difficulties are compounded by the fact that, following the end of cross-examination, witnesses inevitably relax in the face of gentler questioning from their own side. Hence, re-examination can be, and very often is, counter-productive. It should be avoided unless necessary. As often as not, answers helpful to the re-examiner are not elicited, or the witness gives an answer that helps the other side.

15.66

After all the evidence has been called, the tribunal will sometimes allow time for the parties to put their final submissions in writing. A short final oral hearing will follow. If the timetable is tight, however, the tribunal may expect any written submissions to be prepared as the case goes along, and this may necessitate some members of the legal team missing the latter stages of the hearing in order to work on final submissions. In order to avoid prolix submissions, some arbitrators will limit the amount of written material that can be put in.

15.67

It is not usual in London for the oral hearing to be followed by post-hearing briefs. Time will usually have been set aside for closing arguments, sometimes after a break

43 Again, this is a matter of professional obligation (Code of Conduct of the Bar of England and Wales, Rule rC7.3), as well as a matter of practice following the so-called ‘rule in Browne v Dunn’ (1893) 6 R 67.

The Award and Post-Award Events

333

following the conclusion of the evidence, and the lawyers will have been expected to put any written arguments to the tribunal by the time of oral closings. American lawyers accustomed to relatively short closing arguments in American courts may be surprised at the length and depth of the closing arguments encouraged by English arbitrators. The benefit of the longer arguments is that the arbitrators frequently test the arguments of the parties through detailed questioning and reading of principal cases cited so that the advocate has the opportunity to make sure that the members of the tribunal understand his or her points clearly. It is important to allow sufficient time to address the tribunal at the end of the hearing, particularly bearing in mind that some of the panel members may have less familiarity with New York law than others. Parties often spend considerable amounts of time in effect arguing their cases with witnesses in cross-examination. That time may be better spent in explaining their case to the tribunal at the end of the hearing. There is also a different approach in England and America to the citation of authority. An American brief will usually contain much more extensive citations than English written or oral arguments, including perhaps a summary of the facts of each case. Very frequently, an American judge or arbitral tribunal will not read all of the cases cited, focusing only on the cases principally relied upon, but will assume that what is said about the cases is accurate unless disputed by the other side. In England, however, arbitrators are not generally impressed by copious citations. They wish to ascertain what the legal principle is. It is important, in closing submissions (whether oral or written), to direct the tribunal’s attention to the passages in the relevant judgments or decisions that really matter and that establish the principle for which the attorney contends. A detailed d iscussion of key decisions is not unusual.

15.68

THE AWARD AND POST-AWARD EVENTS Unless the parties agree otherwise, the tribunal’s award must contain the reasons for the award.44 In practice, the parties can expect to receive a detailed and often lengthy award. The arbitrators must deal with all the issues that were put to them. If they fail to do so, the award may be open to challenge under section 68 of the Arbitration Act 1996. This does not mean that the arbitrators need to deal with every argument on every issue raised, but they must deal with essential issues and can do so concisely.45

15.69

After the award has been delivered, it is sometimes possible to begin court proceedings to challenge the award or aspects of it. However, the circumstances in which challenges are permitted are closely circumscribed. A challenge for error of law is a non-starter in a Bermuda Form arbitration because English courts will review only errors of English law,

15.70

44 English Arbitration Act 1996, s 52(4). 45 Van der Giessen-de-Noord Shipbuilding Division BV v Imtech Marine & Offshore BV [2008] EWHC 2904 (Comm), [2009] 1 Lloyd’s Rep 273; Secretary of State for the Home Department v Raytheon Systems Ltd [2014] EWHC 4375 (TCC) [33], summarising the authorities.

334

The Course and Conduct of a Bermuda Form Arbitration in London

not foreign law.46 An award may be open to challenge if there has been a serious irregularity in the arbitral process, such as a failure by the tribunal to deal with all the issues that were put to it.47 With an experienced tribunal likely to consist of distinguished lawyers, it is unlikely that such an irregularity will take place, and there is no reported case of any challenge on these grounds, successful or otherwise, to a Bermuda Form arbitration award. It is not possible to challenge an award on the basis that the tribunal failed to address or overlooked aspects of the evidence.48 A challenge can also be made if the tribunal exceeds its substantive jurisdiction,49 but questions of jurisdiction are unlikely to arise given the wide terms of the Bermuda Form arbitration clause. Challenges to a tribunal’s substantive decision are sometimes dressed up as jurisdictional challenges, and the English court will generally be astute to prevent this. There is, however, no history of this happening in the context of Bermuda Form cases, and indeed there is currently no reported case of any challenge on jurisdictional grounds. Yet, arbitrators do have the power to correct certain errors in their award without the need for court application or intervention. For example, a tribunal can correct an error in the calculation of the sum due by way of principal or interest under section 57 of the 1996 Act, which gives the tribunal the power to correct an award so as to remove any clerical mistake or error arising from an accidental slip or omission.

46 The power to review for errors of law is contained in s 69 of the English Arbitration Act 1996. Even if the error is an error of English law, it is difficult to obtain the permission to appeal that is a prerequisite of any challenge. See further ch 3 above, in particular para 3.16. An allegation that the tribunal failed properly to apply New York law would not enable a challenge to be made under s 68 of the Act, in the absence of an a llegation (amounting to impropriety) that the tribunal consciously disregarded the provisions of New York law: A v B [2010] EWHC 1626 (Comm), [2010] 2 CLC 1, [25]–[31] (where the court rejected an application based upon the tribunal’s f ailure to apply Spanish law). 47 English Arbitration Act 1996, s 68. 48 UMS Holding Ltd v Great Station Properties SA [2017] EWHC 2398 (Comm), [2018] 1 All ER (Comm) 856, [28]. 49 English Arbitration Act 1996, s 67.

Appendix 1: First Order for Directions

335

APPENDIX 1: FIRST ORDER FOR DIRECTIONS IN THE MATTER OF THE ARBITRATION ACT 1996 AND IN THE MATTER OF AN ARBITRATION BETWEEN: AB ENGINEERING INC Claimant -andCD INSURANCE CO LTD Respondent FIRST ORDER FOR DIRECTIONS 1. The Parties: a. The Claimant in this arbitration is AB Engineering Inc, represented in this arbitration by [insert names and addresses] b. The Respondent in this arbitration is CD Insurance Co Ltd, represented in this arbitration by [insert names and addresses] 2. The Arbitrators [insert names and addresses] 3. The Arbitration Agreement The parties confirm the existence and validity of their arbitration agreement contained in Condition (N) (‘the Arbitration Agreement’) of the CD Insurance Co Ltd policy in issue, Policy No CD 123456. It is agreed that the arbitration agreement’s reference to the provisions of ‘the English Arbitration Acts of 1950, 1975 and 1979 and/or any statutory modifications or amendments thereto’ signifies for this arbitration the English Arbitration Act of 1996. 4. Appointment of Board a. The parties confirm their acceptance that the Board of Arbitration composed of [insert names] has been validly established in accordance with Article VI(N) of their Agreement. b. Each arbitrator is and remains unbiased and independent of the parties. c. The parties and the arbitrators agree that the Chair of the Board of Arbitration may, upon application of the parties, receive, hear and decide upon procedural questions as may arise in the course of these proceedings prior to the substantive hearing (such as discovery or scheduling) alone and without the involvement of the other two arbitrators, in which case the Chair shall be deemed to act with the full authority of the Board of Arbitration; provided, however, that where either of the parties or the Chair considers any such matter better suited for consideration by the full Board of Arbitration, then the full Board shall duly consider such matter.

336

The Course and Conduct of a Bermuda Form Arbitration in London

5. Immunity from Suit a. No arbitrator shall be required to be a party or witness in any judicial or other proceedings arising out of the arbitration; and b. No arbitrator shall be liable to any party in respect of any act or omission in connection with any matter related to the arbitration. 6. Communications a. The parties shall not engage in any oral or written communications with any member of the Board ex parte in connection with the subject matter of the arbitration, save in respect of routine administrative matters. b. Written pleadings, submissions, written evidence and other formal documents which are the subject of any procedural order shall normally be delivered to the Board by [courier, email or as specified]. Correspondence between the parties and the Board shall normally be sent by email. All communications to the Board shall be delivered individually to each member of the Board. c. The Parties shall send copies of correspondence between them to the Board only if it pertains to a matter in which the Board is required to take some action or be apprised of some relevant event. 7. Applicable Law Pursuant to Article VI(N) of the Agreement, the juridical seat of the arbitration is London, England. Accordingly, the law governing the arbitration itself is English law, regardless of whether meetings and hearings take place elsewhere in the interest of saving costs or convenience. 8. Confidentiality All pleadings, documents and testamentary evidence submitted in the arbitration and the deliberations of the Board of Arbitration, and the contents of the award itself, shall remain confidential in perpetuity, in accordance with and subject to the terms set out in Appendix 2.50 An arbitrator shall not participate in, or give any information for the purpose of assistance in, any proceedings relating to the arbitration or the award unless compelled to do so by a court of competent jurisdiction. PROCEDURAL TIMETABLE 9. Pleadings The Claimant’s Points of Claim were served on [ ]. The Respondent’s Points of Defence were served on [ ]. The Claimant shall file its Reply to the Points of Defence by [ ]. 10. Document Production a. The parties and the Board shall use, as a guideline, the IBA Rules on the Taking of Evidence in International Commercial Arbitration (‘the IBA Rules’).51 50 See Appendix 2 to this chapter in the text below. 51 The IBA Rules were originally issued in 1999. A revision was approved by the IBA in May 2010. Relatively modest amendments to the 2010 IBA Rules were approved on 17 December 2020. See ch 16 below.

Appendix 1: First Order for Directions

337

b. On [ ], each party shall submit to the other party copies of all documents a vailable to them on which they rely, including public documents and those in the public domain; and make originals available for inspection within two days thereafter. c. On [ ], each party shall submit to the other party a request for production in accordance with Article 3.3 of the IBA Rules. d. On [ ], the parties shall produce to each other copy documents pursuant to the requests served and Article 3.4 of the IBA Rules and make originals of the same available for inspection within two days thereafter; or alternatively, state any objections to production in accordance with Article 3.5 of the IBA Rules. e. The parties shall submit to each other responses to objections to produce documents on or before [ ], and make any application to the Board for disclosure of such documents on that date. f. Copies of the documents produced by the parties need not be produced to the Board. If an application for disclosure is made, the parties shall produce to the Board only such documents as are necessary for the resolution of the issues. g. The Board is to resolve any outstanding disputes on document production at a hearing on [ ]. No later than two clear days before this hearing, the parties shall submit to the Board a Redfern schedule summarising the outstanding requests and objections to requests to produce. h. Documents produced by the parties shall be listed in whatever manner may be convenient, whether electronically or otherwise. 11. Witnesses of Fact a. Signed witness statements of fact are to be exchanged on [ ]. b. Signed supplemental witness statements of fact are to be exchanged on [ ]. Each party shall be permitted to serve statements from witnesses whose statements were not originally exchanged, provided they are responsive to issues raised by the opposing party’s witness statements. c. It shall be permissible for witnesses with scientific or other expertise to address matters of a technical nature in their witness statements, and to express opinions thereon, provided that this is necessary to do so in order for them to give their factual evidence relevant to the issues in dispute. d. Unless otherwise ordered, witness statements are to stand as the evidence in chief of the witness at the substantive hearing. e. On or before [ ], each party shall give notice of those witnesses, whose statements have been served by the other side, whom it is requesting be made available in person for cross-examination at the substantive hearing. Unless there is good reason not to do so, a party must make available for cross-examination witnesses whose statements have been served and whose presence at the hearing has been requested by the other side. If a witness is not made available despite a request to do so, the Board may determine what (if any) weight will be given to that witness’ evidence. f. To reduce costs and promote efficiency and convenience, witnesses may, subject to any contrary direction, be presented at the hearing by way of video presentation instead of appearing live. The parties will liaise with each other to ensure that such evidence is timetabled conveniently. Where a witness is to give evidence

338

The Course and Conduct of a Bermuda Form Arbitration in London

by video, both parties will be entitled to have a representative present at the location where the witness is being videoed. 12. Expert Witnesses a. On or before [ ], the parties shall exchange memoranda concerning the scope of their proposed expert evidence, including (a) the number of expert witnesses that each intends to call; (b) their disciplines; and (c) the subject matter of such testimony. The parties shall be deemed to have reserved the right to call an equal number of experts from the same disciplines to testify about the same issues as its opposing party (whether or not originally identified in their respective memoranda), without prejudice to the right to make objections to the proposals of the opposing party. Each party reserves the right to apply to the Board to amend their memoranda, including any of items (a)–(c) in this paragraph, based on information in statements of fact witnesses. b. In the event that a party objects to expert evidence being called by its opposing party in a particular discipline or in relation to a particular issue, that party shall object (‘the Objection’) in writing by [ ], setting out the basis of such Objection. The party wishing to call the disputed expert evidence shall respond to the objection by [ ]. c. The parties shall endeavour to settle any Objection by agreement between themselves. If no agreement is reached, the parties shall make written submissions to the Board in order to enable the dispute to be resolved by no later than [ ]. d. Signed reports of experts are to be confined to one expert for the Claimant and one for the Respondents in each of the areas of expertise so designated and agreed between the parties, or as determined by the Board in the event of a dispute. e. Reports to be exchanged on [ ]. f. By [ ], the experts in each discipline shall meet to discuss the expert issues, with a view to identifying those issues on which they agree and those on which they disagree. g. By [ ], the experts in each discipline shall produce joint memoranda, identifying the issues on which they agree and those on which they disagree. h. Any supplemental or reply reports of expert witnesses shall be exchanged on [ ]. 13. The Substantive Hearing a. The substantive hearing shall take place commencing on [ ] for three weeks. All issues of liability and quantum shall be addressed at the substantive hearing. All closing arguments shall made within the three-week hearing. The timetable should allow time for the preparation (and consideration by the Board) of written closing arguments, as well as one day for each party to make oral closing submissions. b. Agreed paginated bundles of documents shall be submitted to the Board on [ ]. c. The parties are to serve written opening submissions, not exceeding [ ] words, on [ ].

Appendix 1: First Order for Directions

339

d. The parties shall agree a timetable for the hearing by [ ]. Subject to contrary order, the time shall be divided equally. Opening statements shall be limited to one hour per party. e. Following the conclusion of the substantive hearing, the parties will be permitted time to prepare written closing submissions, with oral closing arguments to be presented on [ ] (one day per party). 14. General Matters a. The Board shall determine any issue as to the substantive internal law of the State of New York on the basis of the written submissions and oral argument of the parties and of the legal authorities submitted by the parties to the Board, without either party calling expert evidence on the law of New York. b. The parties agree that the Board may direct any party to do all such things during the arbitral proceedings as may reasonably be needed to enable an Award to be made properly, fairly and efficiently. c. Where documents are required pursuant to this order to be served or exchanged by the parties, the latest time for service or exchange shall be 9 pm London time (4 pm New York time) on the specified date. 15. Permission to Apply The parties have liberty to apply to vary or supplement these directions. Signed and Executed as follows [Counsel for the Claimant] [Counsel for the Respondent] [The arbitrators] [ ]

340

The Course and Conduct of a Bermuda Form Arbitration in London

APPENDIX 2: PROTECTIVE ORDER IN THE MATTER OF THE ARBITRATION ACT 1996 AND IN THE MATTER OF AN ARBITRATION BETWEEN: AB ENGINEERING INC Claimant -andCD INSURANCE CO LTD Respondent PROTECTIVE ORDER WHEREAS the parties have agreed to have any dispute arising under CD Policy CD 123456 resolved in an ad hoc private arbitration, which is being conducted in England and under English law, and is thereby subject to English law protecting the confidentiality of the proceedings; and WHEREAS the parties intend that the dispute resolution proceedings and all documents, information and evidence in connection therewith be kept confidential; NOW THEREFORE the Board issues this Protective Order to govern the confidentiality of these arbitration proceedings and to give effect to the intention and agreement of the parties, as expressed by their countersignatures hereon. Accordingly, it is hereby ordered that: 1. All documents, including documents exchanged and written submissions made by the parties (including without limitation all witness statements, briefs, correspondence, pleadings, transcripts and all other written materials exchanged during, produced in or engendered by or in the course of this arbitration) shall be kept confidential and used solely for the purposes of this arbitration. 2. All proceedings, including all hearings, held in connection with this arbitration, and any transcripts of such proceedings, shall be kept confidential and used solely for the purposes of this arbitration. 3. This Order shall not prevent disclosure of documents or information to the extent permitted by the English common law rules protecting the confidentiality of arbitration.

Appendix 2: Protective Order

341

4. For the avoidance of doubt: (a) The parties may disclose the documents or information particularised under paragraphs 1 and 2 above to employees of or counsel for the parties, prospective or actual fact and expert witnesses, non-testifying expert consultants and to the Board. (b) The Respondent may disclose such documents or information to its reinsurers solely to the extent necessary to comply with any reinsurance agreements and/or if required in any proceeding for the purpose of reinsurance recovery. 5. This Order shall not apply to documents or materials which are in the public domain or were obtained prior to or independently of this arbitration. 6. Nothing in these terms shall prevent disclosure of documents or information by a party to its respective auditors and/or regulators or to the extent that such disclosure is required by court order or other applicable law. 7. The Board shall have jurisdiction to resolve any dispute relating to this Protective Order including its scope. Signed and Executed as follows: [] [Counsel for the Claimant] [Counsel for the Respondent] [The arbitrators]

16 Discovery, Privilege and Waiver of Privilege GENERAL PRINCIPLES 16.01

Many disputes concerning the Bermuda Form involve some debate about questions of fact. The parties therefore often wish to obtain orders from the arbitral tribunal requiring the production of documents or other evidence in advance of the hearing, in accordance with the pre-trial discovery or disclosure procedures of common law jurisdictions.1 This can produce various questions. Should disclosure be required? If so, how broad should it be? How, procedurally, should it be carried out? Can a party withhold documents on grounds of privilege or confidentiality? Questions such as this often arise in practice.

16.02

Since discovery is a procedural matter, the tribunal’s powers are determined by English law as the curial law of the arbitration.2 This gives a broad, though not unlimited, discretion to the tribunal as to whether discovery needs to be given at all and, if so, to what extent. Section 34 of the English Arbitration Act 1996 provides: (1) It shall be for the tribunal to decide all procedural and evidential matters, subject to the right of the parties to agree any matter. (2) Procedural and evidential matters include … (d) whether any and if so which documents or classes of documents should be disclosed between and produced by the parties and at what stage.

16.03

An arbitral tribunal, in exercising that discretion, is not bound to follow the practices customary in litigation, in England or elsewhere.3 Indeed, many arbitrators who are familiar with international arbitration react adversely to any assumption that rules for

1 The traditional expression, still used in the United States and Canada, is ‘discovery’. English courts now use the expression ‘disclosure’. Some arbitration practitioners dislike the use of either expression on the grounds that each is redolent of an atmosphere of adversarial formalism out of place in international arbitration, and prefer terms such as ‘document production’. The choice of expression hardly matters, but it is important to remember that international arbitration will not normally or necessarily follow court procedures in such matters. Our use of terms such as ‘discovery’ or ‘disclosure’ should be understood with that important qualification. 2 See above ch 3, in particular paras 3.12–3.15. 3 N Blackaby et al, Redfern and Hunter on International Arbitration, 6th edn (Oxford, OUP, 2015) paras 6.94 and generally paras 6.92–6.119. At fn 73 to para 6.94, the authors state that subject ‘to any mandatory rules of the lex arbitri, or agreement of the parties, the process known as “discovery” has no place in international arbitration’. For the current practice relating to disclosure in England, see below, paras 16.10–16.13.

General Principles

343

English civil litigation should be carried over into an international arbitration. In our view, this adverse reaction is justified, and counsel should be wary of citing English civil procedure rules as anything more than a helpful example of how a particular problem might be addressed. It is now increasingly common for tribunals in international arbitrations to adopt, either in whole or with adaptations, rules created by the International Bar Association (IBA): the IBA Rules on the Taking of Evidence in International Commercial Arbitration (hereinafter ‘the IBA Rules’). The IBA Rules were originally issued in 1999 and, according to one international arbitration treatise, became ‘almost universally recognised as the international standard for an effective, pragmatic, and relatively economical document production regime’.4 A revision was approved by the IBA in 2010 and some relatively modest changes were approved in 2020. The broad scheme set out in Article 3 of the IBA Rules is that document disclosure starts with the production by each party of the documents on which it intends to rely. Each party may then also submit a ‘Request to Produce’ to the opposing party, and a party can either produce the documents requested or make a reasoned objection. The rules contain provisions for the resolution of disputes about disclosure.

16.04

The IBA Rules provide for a more focused and narrower disclosure regime than prevails under US or (at least until recently) English court practice, reflecting a view that arbitral justice and efficiency is best served by a rifle shot, rather than a buckshot, approach to disclosure.5 This approach can be seen in the requirement in Article 3 that the ‘Request to Produce’ should contain either a description of each requested document, or a description of a ‘narrow and specific requested category’ of documents. Since a typical request is for a category of documents rather than an individual document, the ‘narrow and specific’ requirement is an important limitation on what can be requested. The ‘Request to Produce’ must also contain a statement as to how the documents requested are ‘relevant to the case and material to its outcome’. Relevance alone is therefore insufficient. A further requirement is that the request must contain ‘a statement that the Documents requested are not in the possession, custody or control of the requesting Party or a statement of the reasons why it would be unreasonably burdensome for the requesting Party to produce such Documents’.6 Accordingly, the IBA Rules contemplate that requests will generally not be made for documents which have already been produced to the requesting party. This may be significant in Bermuda Form cases, where the policyholder will in many cases have given extensive documentation to the insurer prior to the commencement of the proceedings in the context of the insurer’s coverage investigation.

16.05

In a case where the IBA Rules are adopted, the drafter of the ‘Request to Produce’ will therefore need to draft the request with some precision. It is appropriate to avoid wide

16.06

4 Ibid para 6.95. 5 W Craig, W Park and J Paulsson, International Chamber of Commerce Arbitration, 3rd edn (New York, Oceana Publications, 2000) para 26.01 at 452–53. 6 Article 3(c)(i). The IBA Rules are available on the IBA website, which includes a version which compares the 2010 rules with the 2020 revision: see https://www.ibanet.org/MediaHandler?id=3E6FF22261EB-4ED6-9A3D-67D642629539.

344

Discovery, Privilege and Waiver of Privilege

and sweeping formulations, which might be adopted in the context of court proceedings; such as ‘all documents concerning, relating to or connected with … and including but not limited to …’. Whether a particular request is ‘IBA Rules-compliant’ will depend upon the issues in the case, and the formulation of the request. Some requests, which appear at first sight to be too broad, may nevertheless be justified and permissible – for example, a request for all the correspondence of a particular custodian, concerning a specific issue, during a limited date range. In practice, it is not unusual to see widely worded ‘Requests to Produce’, with the drafter taking the view that if there is an objection to production, the request can be narrowed down at a later stage. 16.07

The opposing party, having received the Request to Produce, is then entitled to respond within a period of time that will have been provided for in the order for directions. Privilege is a ground for objection,7 and it is common for some requests to be objected to on that basis. Other common objections are that the request is not for a narrow and specific category of documents, or that the documents are not relevant and material to the outcome of the case, or that there is an ‘unreasonable burden to produce the requested documents’,8 or ‘considerations of procedural economy, proportionality, fairness … that the Arbitral Tribunal determines to be compelling’.9 Frequently, some or all of these objections will be relied upon in combination in response to a specific request. For example, a policyholder may have already carried out an extensive exercise in order to provide wide-ranging discovery in relation to the underlying litigation under US discovery rules, and made this discovery available to the insurer either prior to or within the arbitration proceedings. In such a case, a policyholder may legitimately respond to further requests by arguing that the documents requested are not ‘material to the outcome of the case’ in view of the extensive disclosure already provided, that it is unreasonably burdensome to be required to repeat the exercise already carried out in the context of the underlying cases, and that there are compelling reasons of procedural economy why it is unnecessary to incur the expense of repeating the exercise.

16.08

Even though the requested party may consider the request to be too broad, it is often sensible to take a pragmatic approach to the request. Thus, rather than simply relying upon a wholesale objection to a particular request, a party will often set out the objections, but add that, nevertheless, that party will carry out a particular search (for example, the documents of a particular custodian or custodians within a specific timeframe) and will disclose documents which are revealed by that search. There are many advantages to taking this pragmatic approach. The requesting party and the tribunal will see that the requested party is making a reasonable effort to produce relevant and material documents. It may be that the requesting party will acknowledge that the original request was too wide, and will be content with the narrower search and production proposed by the requested party. Furthermore, the pragmatic approach means that the extent of the search is fashioned and focused by the requested party, rather than being shaped by the terms of the request. Of course, there may still remain a dispute as to whether the

7 Article

9(2)(b). 9(2)(c). 9 Article 9(2)(g). 8 Article

General Principles

345

proposed search and production is wide enough, and the tribunal may have to resolve that dispute. However, if the tribunal considers that the approach taken is reasonable, then it may be disinclined to require more to be done, at least at that stage; the tribunal may wish to see what the proposed search produces, giving the requesting party the ability to revive a wider request at a later stage. It has now become very common in Bermuda Form arbitrations for the parties to agree to apply the IBA Rules, perhaps with some variation. Nevertheless, disputes can arise as to whether the IBA Rules should be adopted at all, or with modifications. Any opposition to the application of the IBA Rules is likely to come from the insurer, who is invariably the party who is seeking (rather than providing) the significant disclosure in the arbitration, and who may wish to apply disclosure procedures, closer to those which might apply in English court litigation, on the basis that disclosure will then be more wide-ranging. Sometimes, this dispute can be resolved by the parties agreeing that the IBA Rules should apply as a ‘guide’ or ‘guideline’, thereby leaving the tribunal with a discretion as to whether or not to apply the IBA Rules to resolve a dispute about disclosure of a particular category of documents. In practice, however, this may not make very much difference in the event of a dispute as to whether a particular document or category of documents should be disclosed, which requires resolution by the tribunal. If the tribunal considers that the documents sought are potentially important documents, then it is likely to order disclosure whether the IBA Rules are being fully applied or applied only as a ‘guide’.

16.09

In deciding whether or not to adopt or adapt the IBA Rules, a tribunal may bear in mind the extent to which the policyholder (upon whom the more substantial burden of document production usually falls) has already made documents available to the insurer and the insurer has carried out inspection. For example, if there has been little or no prior production or inspection, a tribunal may consider that a Request to Produce should follow a more general production by the parties; for example, a production of documents which includes documents that adversely impact each party’s case. The concept of an order for disclosure of documents ‘adversely impacting’ a party’s case would reflect the approach to disclosure taken, at least until recently, by the Civil Procedure Rules10 in English civil litigation. The requirements of such an approach, given the term ‘standard disclosure’, was that the party giving disclosure should make a ‘reasonable and proportionate’ search for, and disclose:

16.10

(a) the documents on which he relies; and (b) the documents which – (i) adversely affect its own case; (ii) adversely affect another party’s case; or (iii) support another party’s case; and (c) the documents which he is required to disclose by a relevant practice direction.11

This concept is narrower than that which formerly applied in English litigation. Until 1999, documentary discovery in English litigation was wide, requiring disclosure of all documents in a party’s possession, custody or power which were even remotely relevant to the proceedings, including those that were only relevant because they might lead

10 Part

31 and Practice Direction. Procedure Rules 31.6.

11 Civil

16.11

346

Discovery, Privilege and Waiver of Privilege

to a ‘train of inquiry’ which might elicit useful material for the opposing party.12 It is also narrower than the approach taken by courts in the United States.13 Moreover, the ‘standard disclosure’ approach introduced in 1999 also laid greater stress on the requirement of ‘proportionality’, ie, the need for pre-trial preparation, and its associated costs, to be reasonably related to the amount at stake and the importance of the issue being addressed. 16.12

However, recent developments make it less likely that a tribunal will make orders which are akin to ‘standard disclosure’. Not only has the adoption of the IBA Rules become generally accepted in international arbitration, but the English court have also started to move away from the ‘standard disclosure’ model. This modification and restriction of the approach to disclosure in English litigation has resulted from concerns as to the cost and volume of providing disclosure, particularly as a consequence of the proliferation of electronic documents amongst numerous custodians. For a trial two-year period beginning on 1 January 2019, subsequently extended until December 2022, disclosure in proceedings in the Business and Property Courts of England and Wales has been carried out pursuant to a new ‘pilot’ regime which is contained in CPR Practice Direction 51U. The Business and Property Courts include the English Commercial Court which deals with the most significant commercial cases, including major insurance disputes. This new pilot, which may well become permanent and may also spread to other English courts, has borrowed significantly from the successful approach taken in the IBA Rules. Under the pilot, parties are required to select appropriately from a menu of different ‘models’ for disclosure, with any dispute as to the appropriate model to be resolved by the court. One such model, known as Model C, is ‘Request-led search-based disclosure’. This provides for a party to give disclosure of particular documents or narrow classes of documents, relating to a particular issue for disclosure, by reference to requests. This model is therefore very similar to that provided by the IBA Rules. Model C can be regarded as being, at least to some degree, the default model, because if either party is proposing that there should be disclosure akin to ‘standard’ disclosure (Model D) or an even wider search-based disclosure including ‘train of enquiry’ documents (Model E), then the parties should be ready to explain to the court why Disclosure Model C is not sufficient.

16.13

The approach to disclosure taken in litigation in the Business and Property Courts therefore lends some support to a party who is arguing for the application of the IBA Rules, albeit that in an appropriate case a tribunal may decide that a different approach should be taken.

12 Compagnie Financière et Commerciale du Pacifique v Peruvian Guano Co (1882) 11 QBD 55. 13 See, eg, Fed R Civ P 26(b) and parallel rules of civil procedure in the various states, the District of Columbia and US territories. Until 2015, this rule required production of all material ‘reasonably calculated to lead to the discovery of admissible evidence’. The procedural rules in many of the states and other jurisdictions in the United States (meaning the District of Columbia and the US territories) still apply that standard. However, Fed R Civ P 26(b)(1) (as amended 2015) now applies a ‘relevance and proportionality’ standard: Parties may obtain discovery regarding any non-privileged matter that is relevant to any party’s claim or defense and proportional to the needs of the case, considering the importance of the issues at stake in the action, the amount in controversy, the parties’ relative access to relevant information, the parties’ resources, the importance of the discovery in resolving the issues, and whether the burden or expense of the proposed discovery of ways its likely benefit.

Discovery in Arbitration in Practice

347

The overall result is that orders for documentary disclosure in a Bermuda Form arbitration will almost certainly be narrower than those which US attorneys, in particular, are accustomed to seeing in litigation, even under the 2015 amendment of the discovery standard in the Federal Rules of Civil Procedure. If an issue as to the scope of disclosure arises, then (even in a case where the IBA Rules are not being applied) there is likely to be close attention paid to the potential relevance and materiality of any document or class of documents to the issues in dispute, and also whether the requested documents are already in the possession of the requesting party. Document requests will need to be drafted carefully and are more likely to be effective if targeted at either specific documents or narrow categories of documents that go to prove or disprove particular issues in dispute. Arbitrators tend to be unimpressed by requests which are so broadly drawn that they can be characterised as ‘fishing’.

16.14

DISCOVERY IN ARBITRATION IN PRACTICE14 Very frequently, the parties are able to agree on the approach to be taken, and an arbitral tribunal will usually encourage such agreement, whilst recognising that specific issues may need to be referred to the tribunal after the initial exchange of documents has taken place. For some time, however, arbitrators have sought to control this process so that it does not become too burdensome, and the IBA Rules are generally seen as assisting towards this end. It is also in the interests of both parties to agree upon the scope of mutual discovery, not least because agreement will prevent disruptive applications for disclosure made in the later stages of the arbitral process, or even during the substantive hearing itself. In the initial stages of proceedings, the atmosphere may be more conducive to co-operation and agreement than during the substantive hearing. The tribunal will expect a high degree of co-operation, but will usually not want to be copied in on the correspondence generated during the discovery process. If an issue arises which the tribunal needs to resolve, then the parties will need to select and give relevant correspondence to the tribunal.

16.15

The first procedural order will invariably identify the broad procedures and timetable which are to govern the disclosure process. In order to give clarity to disclosure disputes that require resolution, tribunals commonly require the parties to prepare a ‘Redfern schedule’15 or chart which identifies the documents sought, identifies the contentions of each party in relation to the documents, and contains a blank column for the tribunal’s decision. If the IBA Rules are being applied, there will typically be a Redfern schedule which sets out in tabular form: (1) the number of each request; (2) the documents or category of documents requested; (3) the relevance and materiality of the document or category requested, according to the requesting party; (4) the responses or objections to the request; (5) the requesting party’s reply; and (6) a blank column for the tribunal’s decision. The first procedural order should spell out the manner in which disclosure disputes are to be resolved – for example, that they should be resolved following an application on paper, unless either party or the tribunal requests a hearing. In a three-person tribunal,

16.16

14 See also ch 15, in particular paras 15.28–15.30 above. 15 The name ‘Redfern schedule’ is now very widely used; see Redfern and Hunter on International Arbitration (above n 3) paras 6.100–6.102.

348

Discovery, Privilege and Waiver of Privilege

the parties may agree that any disputes on discovery can be decided by the chair, perhaps with the qualification that either party or the chair may request the involvement of the other two arbitrators. Such agreement tends to facilitate a more rapid and less expensive decision-making process on disclosure issues. 16.17

The willingness of the tribunal to order extensive discovery will depend upon the nature of the dispute. If the dispute is sufficiently substantial, as will probably be the case in Bermuda Form arbitrations, the parties’ lawyers and the arbitrator are likely to countenance reasonably extensive discovery. A tribunal will want to be satisfied that the extent of discovery ordered is proportional both to the amounts in dispute and perhaps more importantly to the likely usefulness of the documents sought, bearing in mind any difficulties involved in locating the documents or producing them.

16.18

Historically, when disclosure principally involved the production of documents from a party’s hard copy files, the manner of production in English arbitration was substantially different from American practice.16 However, such differences have now narrowed as a result of various factors: the disclosure process is now largely focused on documents created and held electronically; the processes and systems for identifying and handling such disclosure are similar, including the use of disclosure platforms and companies which specialise in operating them; there are now increasing numbers of US law firms with offices in the United Kingdom, and US lawyers with experience of international arbitration in London, as well as London law firms with a presence in the United States. There are no rules which govern the manner of production, and arbitration tribunals will be sympathetic to any reasonable suggestion which ensures a well-organised and usable document production, and will be unsympathetic to any approach that relieves one party of any burden by placing it entirely on the shoulders of the other. To avoid surprises, it is best to agree on the manner of production with opposing counsel in advance.

16.19

In Bermuda Form cases there has often been extensive production of documents in underlying litigation between the policyholder and plaintiffs, and such disclosure is often available via document depositories or, nowadays, in electronic form. In those circumstances, it is usually convenient to give disclosure of those documents by providing access to the document depository or collection. But some care is needed, on three counts. First, the disclosure in the underlying cases may well include disclosure of documents of no real relevance to the arbitration. In those circumstances, the recipient of the disclosure may feel entitled to expect some assistance in sifting the potentially interesting from the entirely irrelevant, while the disclosing party legitimately objects to being required to list or disclose in some different form, at vast expense, all that has been previously disclosed. A sensible approach can usually be found, for instance, by providing indexes or explanations of how the documents have been arranged. Second, there are likely to be some documents relevant to the insurance dispute which were not relevant to the underlying litigation, and a proper search for these documents needs to be carried out. Third, care needs to be taken that the standard applied to claims of privilege in the underlying litigation has not resulted in documents being withheld there on grounds of privilege which are 16 Under US practice, parties typically produced documents either organised in response to an opponent’s document request or in the order in which responsive documents were found in the client’s files. In England, it was common for documents to be produced in chronological order.

Discovery in Arbitration in Practice

349

disclosable in the arbitration.17 Normally, however, that can be done without completely reviewing every document that was withheld, by explaining the standard that was applied, and if necessary referring to the privilege log that will normally have been prepared in the underlying litigation. Nevertheless, applications are sometimes made for an order that a party’s English lawyers should review all documents withheld from production in underlying US litigation, in order to verify that the documents are also privileged under English law principles. We suggest that such an order is not necessary as a matter of course. The majority of documents withheld from production in the underlying litigation are likely to be confidential documents generated for the purposes of defending the underlying claims, and (as explained below) such documents will be covered by ‘litigation privilege’ under English law. In relation to other documents, the standard applied to the claim for privilege can be identified, and a review of particular documents or categories of documents can be carried out if it appears that privilege has been claimed in the United States on a wider basis than would be permissible in England. American lawyers are accustomed in virtually all US courts and in many domestic American arbitration proceedings to testing the adequacy of an opposing party’s production by taking the deposition of the custodian of records for the opposing party. This procedure is not a part of English arbitrations. Instead, a party will have to base any plea that relevant documents have not been produced on apparent gaps in the existing production or other facts, for example, statements made in witness statements. A tribunal will not usually, at the pre-trial stage, be prepared to go behind a properly considered assertion that a document does not exist or cannot, after being searched for, be found.18 Instead, the question is generally left to the substantive hearing. If, at that stage, the missing documents emerge, or are shown to have existed all along, adverse inferences may, depending on the particular circumstances and any explanation that is given, be drawn.

16.20

Perhaps less familiar to American lawyers is the way in which privilege issues are handled in London. In the United States, it is common practice for the parties to produce ‘privilege logs’, that is, detailed schedules identifying for each document withheld as privileged, its authors, recipients, date, topic and privilege claimed. These logs then enable a party to make, if so advised, a detailed challenge to the claim for privilege. They are usually expensive, and difficult and burdensome to create. The court practice in England, however, is to claim privilege not on a document-by-document basis, but instead by category of documents, providing only a generic description of the documents that have been withheld from production (for example, ‘attorney–client communications concerning the matter in litigation’).

16.21

17 See further below, paras 16.23 ff. 18 See, in the context of English court practice, West London Pipeline & Storage Ltd v Total UK Ltd [2008] EWHC 1729 (Comm), [2008] 2 CLC 258, [63]–[86], in particular [86(4)(d)]: ‘In cases where the issue is whether the documents exist … the existence of the documents is likely to be an issue at the trial and there is a particular risk of a court at an interlocutory stage impinging on that issue.’ In addition, courts and arbitration tribunals will generally not question a statement from a reputable lawyer that, following a search, no further documents within a particular category exist; see Jaffray and Others v Society of Lloyd’s [2002] EWCA Civ 1101, (2002) 146 SJLB 214, [527]–[529]. See too, in relation to redaction of documents, Lincoln National Life Insurance Co v Sun Life Assurance Co of Canada [2004] EWHC 343 (Comm), [2004] 1 Lloyd’s Rep 737, [102] (‘arbitrators and judges habitually trust what they are told on procedural matters by the parties’ legal representatives’). If a relevant document or category of documents is known to have existed in the past but to have been lost or destroyed, it is good practice to explain that fact. A party or the tribunal may require formal written confirmation to that effect.

350

Discovery, Privilege and Waiver of Privilege

English courts consider that to require a detailed description of the documents might itself require disclosure of information that is privileged.19 A similar approach will usually be taken, and will be regarded as acceptable, in Bermuda Form arbitrations. If the IBA Rules are being followed, then the claim for privilege is likely to be made in generic terms in response to the particular request. A listing of privileged documents in response to the request is not expected or required. The tribunal trusts that the attorneys for each party have properly decided which documents are privileged or whether particular documents fall within or without the generic description of privilege. Some American lawyers will find this procedure uncomfortable, because it will not easily allow a specific test as to a specific document withheld from production. Although the parties may agree to proceed more like the American practice (for example, for each to agree to provide privilege logs), it is unlikely that an English panel will order such a procedure over the objection of one of the parties. Nor, without express agreement, will a tribunal undertake itself to make an ex parte examination of documents which are allegedly privileged or irrelevant in order to verify the claim. If a tribunal considers that the question of privilege has not been properly approached, it is likely to order a party to reconsider the relevant documents with the appropriate principles in mind, and may perhaps require the provision of a witness statement to explain the position.20 If a dispute remains and requires particular documents to be inspected, a tribunal can order that this review should be carried out by a third party who can then report back to the tribunal, but without disclosing the contents of any documents for which a claim for privilege can properly be maintained. 16.22

The duty to disclose documents whose production has been ordered is a continuing one. Thus, if, after disclosure has been given, additional disclosable documents come to light, they must be promptly disclosed. This situation is not expressly addressed in the IBA Rules, and it therefore may be appropriate (in cases where the IBA Rules are adopted) to make clear that there is a continuing duty to disclose any documents in categories which the requested party has agreed to produce, or where production has been ordered by the tribunal.

LEGAL PROFESSIONAL PRIVILEGE Applicable Law 16.23

Since questions of disclosure are procedural, and generally subject to English law, it might be thought that a logical consequence of this is that questions of privilege21 are also governed by English law. What happens, however, if there are differences between 19 See Derby & Co Ltd v Weldon (No 7) [1990] 1 WLR 1156, 1175–80. In Tchenguiz Imerman v Imerman [2012] EWHC 4047 (Fam), [2014] 1 FLR 232, the court ordered a listing of privileged documents, but accepted that where there was a large number of documents falling within a particular category, it was sufficient to list the category and date ranges. See the discussion in C Hollander, Documentary Evidence, 14th edn (London, Sweet & Maxwell, 2021) paras 15.06–15.09. 20 For the position in the English courts when the court is not satisfied as to the claim for privilege, see West London Pipeline & Storage Ltd v Total UK Ltd [2008] EWHC 1729 (Comm), [2008] 2 CLC 258, [86]; WH Holding Ltd and West Ham United Football Club Ltd. v E20 Stadium LLP [2018] EWCA Civ 2652, [29]–[41]. 21 In this section we principally address legal professional privilege, ie (in American terminology) attorney– client privilege and attorney work product doctrine. The discussion, however, is generally applicable to other

Legal Professional Privilege

351

the law as to privilege in England when compared to other laws that may have a bearing on whether a document should be privileged? For example, a party might wish to argue for the application of the law of its domicile or residence, or the law governing the relationship between a party and its lawyer, or the law most closely connected with the communication or document, or perhaps even international law or general principles of law. How should such arguments be addressed?22 There are, as one would expect, divergences between English law in relation to legal professional privilege, and the law as applied by different states in the United States to attorney–client privilege and the attorney work product doctrine.23 It appears that, in certain respects, English law is more solicitous of such privilege than the law prevailing in certain jurisdictions in the United States.24 Situations therefore arise where English law would recognise a privilege that would not be recognised by certain US courts. Conversely, there are situations where a privilege is not recognised as such in England, but is recognised by a foreign system of law.25 The potential for conflicts between applicable laws arises not only as between English law and the laws of different states in the United States. Civil law systems may become relevant, as the claimant in a Bermuda Form arbitration could well be a European company, for example, a company whose US parent is the named insured under the policy, or the claim may include liability arising out of legal proceedings in Europe. Section 34(2)(d) of the English Arbitration Act 1996 gives the tribunal a discretion as to ‘whether any and

fundamental privileges (notably the privilege against self-incrimination and public interest immunity). It would also probably apply to the privilege English law attaches to settlement negotiations (so-called ‘without prejudice privilege’). That might, however, be more open to question, since ‘without prejudice’ privilege might not be thought to have the same ‘fundamental’ quality as the other privileges, albeit that strongly positive assertions that the foundation of the privilege is a matter of public policy can be found; see, eg, Unilever plc v Procter & Gamble [2000] 1 WLR 2436, 2448–49; Savings & Investment Bank v Finken [2003] EWCA Civ 1630, [2004] 1 All ER 1125, 1138; Motorola Solutions Inc v Hytera Communications Corp Ltd [2021] EWCA Civ 11, [2021] 2 WLR 679, [47]. 22 See R Mosk and T Ginsburg, ‘Evidentiary Privileges in International Arbitration’ (2001) 50 International and Comparative Law Quarterly 345. This article not only contains a valuable comparative analysis of law relating to privilege in different countries, but also addresses in detail the question of how to resolve the conflicts issues which arise. See too J Rubenstein and B Guerrina, ‘The Attorney-Client Privilege and International Arbitration’ (2001) 18 Journal of International Arbitration 587; F von Schlabrendorff and A Sheppard, ‘Conflict of Legal Principles in International Arbitration’ in G Aksen et al (eds), Liber Amicorum in Honour of Robert Briner (Paris, ICC, 2005) 743; G Burn and Z Skelton, ‘The Problem with Legal Privilege in International Arbitration’ (2006) 72 Arbitration 124; K-P Berger, ‘Evidentiary Privileges: Best Practice Standards versus/and Arbitral Discretion’ (2006) 22 Arbitration International 501. 23 Mosk and Ginsburg (above n 22) at 351–53. 24 See Nederlandse Reassurantie Groep Holding NV v Bacon & Woodrow (A Firm) and Others [1995] 1 All ER 976, 986–87, where Colman J in the English Commercial Court said that English law does not adopt a principle of waiver of privilege which is as wide as that indicated by the reasoning in Hearn v Rhay, 68 FRD 574 (1975). See also Paragon Finance v Freshfields [1999] 1 WLR 1183, 1193G and British American Tobacco Investments Ltd v United States of America [2004] EWCA Civ 1064, where the application of English law principles relating to waiver of privilege led to a result different from that reached in the US District Court for the District of Columbia. Another example is the balancing exercise that US courts sometimes carry out in deciding whether the interests of justice outweigh the need for attorney work-product privilege; see Hoechst Celanese Corp v National Union Fire Ins Co of Pittsburgh, Pa 623 A2d 1118 (Del Super Ct 1992). 25 See, eg, Mosk and Ginsburg (above n 22) at 357 and 361, discussing accountant–client privilege and business-secrets privilege; The RBS Rights Issue Litigation [2016] EWHC 3161 (Ch), [2017] 1 WLR 1991, in particular [96] and [139], discussing the application of legal advice privilege to employees.

16.24

352

Discovery, Privilege and Waiver of Privilege

if so which documents or classes of documents should be disclosed between and produced by the parties and at what stage’. But neither that section nor any other provision of the Act specifically addresses the question of privilege or the law to be applied in relation thereto. 16.25

An arbitral tribunal in a London arbitration, particularly a tribunal with a majority of English lawyers, tends to approach the question of privilege by applying English law as the law of the forum where the arbitration is taking place. This is in accordance with the well-established practice of the English courts in relation to proceedings taking place in England. The English case law was reviewed in detail by Hildyard J in The RBS Rights Issue Litigation.26 He held that an English court applies the lex fori to issues of privilege, although this could be ‘mitigated in appropriate circumstances by the Court’s discretion to decline to order inspection’.27 Amongst the authorities reviewed, and supporting the application of the lex fori, was Bourns v Raychem,28 where the Court of Appeal held that privilege is not lost in England because privilege could not be claimed for documents in another country: To suggest otherwise would mean that a court, when deciding whether to uphold a claim for privilege, would need to be informed as to whether privilege could be claimed in all the countries of the world.29

The attitude of the English court is therefore that rules of privilege are matters for the court in which the protection is relied upon.30 English law is therefore applied whether or not the foreign rule would require disclosure or protect from disclosure.31 16.26

It might be argued that arbitration tribunals can, under section 34(2)(d) of the English Arbitration Act 1996, take a more liberal approach to privilege, and are not necessarily required to apply the law of the forum when resolving a dispute about privilege.32 For example, it could be suggested that a tribunal can order disclosure of documents that are clearly privileged under English law, on the basis that they are not privileged under some other law. Indeed, it has even sometimes been suggested that the tribunal’s power, under section 34, to dispense with rules of evidence permits the tribunal to choose to ignore claims to privilege entirely.

16.27

We doubt whether such arguments would succeed, and we do not think that they are correct. Section 34 of the English Arbitration Act 1996 provides that the tribunal shall

26 The RBS Rights Issue Litigation (above n 25) (Ch), [140]–[174]. 27 Ibid [140] and [175]–[187]. 28 Bourns v Raychem [1999] 3 All ER 154. See also Re Duncan [1968] P 306, 311 (‘The nationality of the foreign lawyer is as irrelevant as his address for this purpose’); Lawrence v Campbell (1859) 4 Drewry 485. 29 Bourns v Raychem (above n 28) at 167. This statement was applied in British American Tobacco Investments Ltd v United States of America [2004] EWCA Civ 1064. The US District Court in the District of Columbia had held that privilege had been waived by British American Tobacco. The English Court of Appeal, applying English law principles, reached the opposite conclusion. 30 RBS Rights Issue Litigation (above n 26) [163]. 31 Ibid [170]. 32 Cf the liberal provisions of s 46(3) of the English Arbitration Act, which concern the law applicable to the substance of the dispute.

Legal Professional Privilege

353

decide all procedural and evidential matters, including whether any documents should be disclosed. However, it is submitted that privilege, properly so called, would be an exception to this principle. Privilege is not a matter of discretion; it is a fundamental rule of law.33 Thus, it has been held that: Legal professional privilege is … much more than an ordinary rule of evidence, limited in its application to the facts of a particular case. It is a fundamental condition on which the administration of justice as a whole rests.34

In R (Morgan Grenfell Ltd) v Special Commissioner, privilege was described as ‘a fundamental human right long established in the common law’.35 The European Court of Human Rights held that privilege is part of the right of privacy guaranteed by Article 8 of the European Convention on Human Rights.36 In the absence of express words or necessary implication, a statute will not be regarded as abrogating such rights,37 and section 34 of the English Arbitration Act 1996 does not contain an express directive or purport, by necessary implication, to do this. Accordingly, in an arbitration subject to English curial and procedural law, the tribunal will be unlikely to order disclosure of documents that are, under English law, privileged, in the absence of some express or clearly implied agreement between the parties that the tribunal should have that power.38

16.28

But what is the position if the privilege would not be recognised in England, but exists under some other potentially relevant law? For example, what if a US policyholder is asked to produce documents subject to a privilege under the applicable law in the United States? In such a case, the US company could argue that the applicable law should not be the law where the arbitration proceedings happen to be taking place, but, instead, should be the law of the relevant state in the United States where the document was created or (in the case of a communication between lawyer and client) the law that governed the relationship between the client and the lawyer.

16.29

33 R v Derby Magistrates’ Court ex parte B [1996] 1 AC 487. The facts in R v Derby Magistrates’ Court were exceptional and striking. A defendant charged with murder sought disclosure from another man, who had first confessed to the murder, then changed his story and been acquitted. The documents sought related to the instructions that the second man had initially given to his solicitors. The documents were highly relevant to the defendant’s defence. Yet the House of Lords upheld the privilege and refused to indulge in any exercise of balancing the interests of justice against the claim for privilege. In Paragon Finance v Freshfields (above n 24) 1193, Lord Bingham said, in commenting on the decision in R v Derby Magistrates’ Court: ‘This authority is important, not only for its clear restatement of principle, but also as illustrating in graphic terms the all but absolute nature of this privilege in the absence of waiver. If ever there was a case in which the interests of justice militated in favour of disclosure that surely was it.’ See too B v Auckland District Law Society [2003] UKPC 38, [2003] 2 AC 736. 34 R v Derby Magistrates’ Court ex parte B (above n 33) 507. See also R (Morgan Grenfell Ltd) v Special Commissioner [2003] 1 AC 563; General Mediterranean Holdings SA v Patel [2000] 1 WLR 272. 35 [2002] UKHL 21, [7], [2003] 1 AC 563. 36 Ibid, citing Campbell v United Kingdom (1992) 15 EHRR 137; Foxley v United Kingdom (2000) 31 EHRR 637. 37 [2002] UKHL 21, [7]; R (on the Application of Kelly) v Warley Magistrates’ Court [2007] EWHC 1836 (Admin) [2008] 1 WLR 2001. See also General Mediterranean Holdings SA v Patel (above n 34) for a review of the authorities. 38 Cf Mosk and Ginsburg (above n 22) at 376–77: ‘If privileges are considered to be procedural, the arbitrators would not need to defer to them unless they are mandatory in arbitration under the procedural law of the local forum or the parties have agreed to the application of a certain procedural law.’ For the reasons set out above, it seems that respect for privilege probably would, under English law, be regarded as mandatory.

354

Discovery, Privilege and Waiver of Privilege

16.30

It is possible that the tribunal would take a narrow view and hold that, as the parties have agreed to submit their disputes to resolution in London, the extent of that right should be determined by the law of the agreed forum alone. However, arguably, a tribunal is not compelled to take this approach and has some flexibility as to which law to apply.39 The tribunal could thus take the view that the question of privilege should be determined by reference to a law other than English law; for example, the law most closely connected to the communication or document.

16.31

Even if the tribunal does decide to apply English law, it does not follow that production of the documents would be compelled. The tribunal might properly decline to order production of documents where a party had a reasonable expectation of privilege under the relevant local law, even if English law would not recognise the claim. But this matter goes to the tribunal’s discretion rather than to the question of whether discovery is in principle available.40 In The RBS Rights Issue Litigation,41 the court decided that whilst English law applied to determine issues of privilege, disclosure or inspection of a document could be prevented as a matter of the court’s discretion. In an appropriate case, an English court might therefore, by the exercise of its discretion to decline to order production, give precedence to the privilege that exists under the foreign law, notwithstanding that the document is not privileged under English law. Nevertheless, the court said that it was ‘likely to lean heavily in favour of disclosure unless compelling grounds are provided to suggest that the public interest and private entitlement to trial on all disclosable evidence must yield to some exceptional concern’.42 The discretion of an arbitral tribunal is not constrained by the approach that would be taken by an English court, and a tribunal may therefore apply a more flexible approach. In a case where the IBA Rules on the Taking of Evidence in International Arbitration43 are applied or treated as a guide, there can be no doubt that a flexible approach can be taken. English Rules Applicable to Legal Professional Privilege

16.32

English law relating to privilege and waiver of privilege is a substantial topic that receives detailed discussion in textbooks concerned with discovery.44 Privilege interrelates with the law relating to confidentiality, and the interrelationship is itself a complex subject that

39 See the arguments advanced by Mosk and Ginsburg (above n 22) at 374–75. In Morris v Banque Arab & International D’Investissement SA [2001] 1 BCLC 263, Neuberger J held that a court might refuse to order inspection of documents, where provision of the documents would be an offence under a foreign law. 40 See paras 16.93–16.94 below, and the cases there cited, for further discussion of the tribunal’s discretion. 41 [2016] EWHC 3161 (Ch), [2017] 1 WLR 1991 at [140] and [175]–[187]. 42 Ibid [183]. 43 Article 9 of the IBA Rules identifies a number of matters which a tribunal ‘may take into account’ in considering issues of ‘legal impediment or privilege under the legal or ethical rules determined by the Arbitral Tribunal to be applicable’. These include ‘the expectation of the Parties and their advisors at the time the legal impediment or privilege is said to have arisen’. 44 See in particular Hollander, Documentary Evidence (above n 19) chs 12–19; P Matthews and H Malek, Disclosure, 5th edn (London, Sweet & Maxwell, 2017) ch 11; H Malek (ed), Phipson on Evidence, 19th edn (London, Sweet & Maxwell, 2018) chs 23–26; B Thanki, The Law of Privilege, 3rd edn (Oxford, OUP, 2018); C Passmore, Privilege, 4th edn (London, Sweet & Maxwell, 2020).

Legal Professional Privilege

355

has merited academic treatment.45 The purpose of this section is to outline the applicable law and to identify and discuss particular difficulties which can arise in the context of a Bermuda Form arbitration. A distinction is normally drawn between two categories of privilege that exist under English law, although there is often an overlap between them. The terminology used by the English courts is not necessarily consistent, but the two categories of privilege generally recognised in English practice can be described as ‘legal advice privilege’ (sometimes called legal professional privilege) and ‘litigation privilege’.46

16.33

Legal advice privilege extends to all communications between the client and the client’s legal adviser for the purpose of obtaining or giving legal advice. This purpose must be the dominant purpose.47 It exists whether litigation is anticipated or not and is similar to the privilege commonly called the ‘attorney-client privilege’ in the United States. The privilege promotes the public interest requiring full and frank exchange of confidence between lawyer and client to enable the latter to receive necessary legal advice.48 Originally, it related only to communications where legal proceedings were in being or in contemplation, but it is now clear that legal advice or legal professional privilege extends beyond legal advice given in regard to litigation, though the privilege is wider where litigation is imminent or current than in non-contentious matters. Legal advisers in this context include in-house lawyers.49

16.34

In Balabel v Air India,50 the relevant transaction involved the grant of a lease, and legal advice was given in that context. The issue which arose was whether privilege applied to certain documents that formed part of the communications between the lawyer and client, but that did not themselves convey legal advice; for example, documents that essentially recorded information or transactions. The Court of Appeal held that privilege applied, applying the test of whether ‘the communication or other document was made confidentially for the purposes of legal advice’.51 It held that those purposes had to be construed broadly, and that privilege obviously attached to a document conveying legal advice from lawyer to client and to a specific request from the client for such advice. But the court concluded that it did not follow that all other communications between

16.35

45 See P Matthews, ‘Breach of Confidence and Legal Privilege’ (1981) 1 Legal Studies 77; N Andrews, ‘The Influence of Equity upon the Doctrine of Legal Professional Privilege’ (1989) 105 Law Quarterly Review 608; A Newbold, ‘Inadvertent Disclosure in Civil Proceedings’ (1991) 107 Law Quarterly Review 99; J Auburn, Legal Professional Privilege: Law and Theory (Oxford, Hart Publishing, 2000) ch 12; ISTIL Group Inc v Zahoor [2003] 2 All ER 252, 267. 46 The distinction between legal advice privilege and litigation privilege was first drawn in Anderson v Bank of British Columbia (1876) 2 Ch D 644. The policy reasons for privilege and the historical background are traced in the judgment of the House of Lords in Three Rivers District Council v Governor and Company of the Bank of England (No 6) [2004] UKHL 48; [2005] 1 AC 610, [23]–[35] and [85] ff. See too Winterthur Swiss Insurance Co v AG (Manchester) (In Liquidation) [2006] EWHC 839 (Comm); and Director of the Serious Fraud Office v Eurasian Natural Resources Corp Ltd [2018] EWCA Civ 2006, [2019] 1 WLR 791, [61]–[66]. 47 Civil Aviation Authority v R (on the Application of Jet2.Com Ltd) [2020] EWCA Civ 35, [2020] QB 1027, [96]. 48 Balabel v Air India [1988] 1 Ch 317, 324; Ventouris v Mountain [1991] 1 WLR 607, 611; Three Rivers District Council (above n 46) [34]; Winterthur (above n 46) [69]. 49 The numerous authorities in this respect are reviewed in PJSC Tatneft v Bogolyubov [2020] EWHC 2437 (Comm). 50 Balabel v Air India (above n 48). 51 Ibid at 330.

356

Discovery, Privilege and Waiver of Privilege

the lawyer and client lacked privilege. It said that, in most lawyer–client relationships, especially where a transaction involves protracted dealings, advice may be required or appropriate on matters great or small at various stages. There will be a continuum of communication and meetings between the lawyer and the client: Where information is passed by the solicitor or client to the other as part of the continuum aimed at keeping both informed so that advice may be sought and given as required, privilege will attach. A letter from the client containing information may end with such words as ‘please advise me what I should do’. But even if it does not, there will usually be implied in the relationship an overall expectation that the solicitor will at each stage, whether asked specifically or not, tender appropriate advice. Moreover, legal advice is not confined to telling the client the law; it must include advice as to what should prudently and sensibly be done in the relevant legal context.52

16.36

The general approach of the Court of Appeal in Balabel was endorsed by the House of Lords in Three Rivers District Council and Others v Bank of England (No 6).53 Lord Carswell said that all communications between a solicitor and the client relating to a transaction in which the solicitor has been instructed for the purpose of obtaining legal advice will be privileged, notwithstanding that they do not contain advice on matters of law or construction, provided that they are directly related to the performance by the solicitor of the solicitor’s professional duty as legal adviser of the client.54 If, as in Balabel, genuine legal advice, properly so called, is being given, it will be wrong to draw too pedantic a distinction between different documents relating to the transaction in respect of which advice is being given. But where a lawyer undertakes to give advice which is not strictly legal at all, the privilege will not apply. If a solicitor becomes the client’s ‘man of business’, responsible for advising the client on all matters of business, including investment policy, finance policy and other business matters, then the advice may ‘lack a relevant legal context’.55 In Three Rivers (No 6), the Bank of England had taken advice from lawyers as to how it should present its case at an inquiry, and the advice was held to be privileged because it had been given in a relevant legal context.

16.37

Legal advice privilege relates only to communications between the lawyer and the client. Legal advice privilege cannot be claimed for documents other than those passing between the client and its legal advisers.56 Documents, such as internal memoranda, prepared with 52 Ibid. 53 Three Rivers (No 6) (above n 46). 54 Ibid, at [111]. 55 Ibid, at [38] (per Lord Scott of Foscote). 56 Three Rivers District Council v Bank of England (No 5) [2003] EWCA Civ 474, [2003] QB 1556. In the case of corporate clients, interesting questions can arise as to who is the ‘client’ for these purposes. In Three Rivers (No 5), the Court of Appeal had taken a restrictive view as to who, on the facts, was the solicitor’s ‘client’. In Three Rivers (No 6) (above n 46), the House of Lords declined to reconsider the question because it did not arise for decision. The decision of the Court of Appeal continues to represent English law on the issue: The RBS Rights Issue Litigation (above n 25) paras [54]–[60]; Director of the Serious Fraud Office (above n 46) paras [123]–[130], where the Court of Appeal indicated that this was an issue that will have to be considered again by the Supreme Court. Accordingly, where the issue concerns a corporation and its employees, the ‘client’ consists only of those employees authorised to seek and receive legal advice from the lawyer, and does not extend to information provided by employees and ex-employees for the purposes of being placed before a lawyer; ibid, [91]. A different approach is taken in the United States; ibid, paras [96] and [139].

Legal Professional Privilege

357

a view to obtaining legal advice are not privileged under this head of privilege, although they may come within the category protected by litigation privilege. Nor are communications between a lawyer and a third party protected by legal advice privilege, though they too may be protected by litigation privilege. Legal advice privilege is thus concerned with documents passing between the client and lawyer, with the specific purpose of advising in a ‘relevant legal context’. It relates to a narrow class of documents. By contrast, the second category of privilege, ‘litigation privilege’, may include documents that pass between the client, or its attorney, and a third party. A classic example is a draft witness statement that the lawyer obtains from a potential witness in the litigation. Or it may include documents prepared with a view to obtaining legal advice on pending litigation, but which are not actually sent to the legal adviser; for example, a first draft memorandum prepared by the client setting out an account of events on which advice is sought.

16.38

Litigation privilege covers documents which come into existence either for the purpose of obtaining or giving advice in regard to litigation or of obtaining evidence to be used in litigation, or for obtaining information which might lead to the obtaining of evidence.57 This purpose must be the dominant purpose. Litigation privilege will cover legal advice given so as to head off, avoid, or settle actual or reasonably contemplated proceedings.58 The rationale for this privilege, as explained in the early cases, is that communications made for the purposes of defending or bringing the action ‘are in fact the brief59 in the action, and ought to be protected’60 or ‘as you have no right to see your adversary’s brief, you have no right to see that which comes into existence merely as materials for the brief’.61 A more recent rationale is that the preparation of a case is inextricably linked to the advice to the client on whether to fight or to settle and, if so, on what terms.62 In Three Rivers District Council v Bank of England (No 5),63 the Court of Appeal said that the need for the client to make a ‘clean breast of it’ to its legal advisers was paramount when litigation either exists or is contemplated. It is in the interests of the state which provides the court system and its judges at taxpayers’ expense that legal advisers should be able to encourage strong cases and discourage weak ones. Hence, the rationale rests on principles of access to justice, the proper administration of justice, a fair trial and equality of arms.64

16.39

57 See Wheeler v Le Marchant (1881) 17 Ch D 675, 681, 684–85. 58 WH Holding Ltd and West Ham United Football Club Ltd. v E20 Stadium LLP [2018] EWCA Civ 2652, [12]–[18], [27]. The privilege did not apply in that case because the relevant documents were purely concerned with discussion of a commercial settlement. 59 In this context, ‘brief’ refers to the document and its enclosures sent by the solicitor instructing the barrister to appear on the client’s behalf. In current US terminology, of course, ‘brief’ now refers to the written argument that is presented to the court or arbitral tribunal before the case starts (in the case of an opening brief) or after it finishes (in the case of a closing brief). 60 Wheeler v Le Marchant (1881) 17 Ch D 675, 684–85. 61 Anderson v Bank of British Columbia (1876) 2 Ch D 644, 656. 62 ISTIl Group Inc v Zahoor [2003] 2 All ER 252, 264. 63 Three Rivers (No 5) (above n 56) [2003] EWCA Civ 474, [26], [2003] QB 1556, 1578. 64 Winterthur (above n 46) [68].

358

Discovery, Privilege and Waiver of Privilege

16.40

It is not unusual for documents to come into existence for more than one purpose. For example, after an accident, a report may be commissioned into the reasons why the accident had taken place. In such situations, a claim for privilege will arise only if the document came into existence with the dominant purpose of submission to the party’s legal advisers in anticipation of litigation. Litigation does not have to be the sole purpose. Thus, in the leading case of Waugh v British Railways Board,65 a claim for privilege failed in respect of a report prepared where the litigation purpose was of equal rank and weight with the purpose of railway operation and safety. It is therefore important to consider whether advice was taken in contemplation of litigation or for some other purpose, since litigation privilege is wider than legal advice privilege. It is not necessary that litigation is inevitable. Where advice is sought with a view to ascertaining whether there is an arguable claim or defence, it being anticipated that (if there is) the claim or defence will be asserted and resisted in litigation which is likely to ensue, then litigation privilege will apply to that advice and to communications (for instance with potential witnesses or experts) relating to it.66

16.41

The client owns the privilege and may waive it. If the client chooses not to waive privilege, it is not permissible for an adverse inference to be drawn.67 The assertion of privilege does not enable the tribunal to conclude that information is being hidden or that adverse legal advice had been received.

16.42

English courts place the burden of establishing that a communication is privileged on the party claiming privilege,68 and it is likely that a Bermuda Form arbitration tribunal would take the same approach.

16.43

Unless there is a waiver of privilege, there is no rule of law which allows the English court to override a claim to privilege simply because it would be fair to do so or because the documents are relevant.69

WAIVER OF PRIVILEGE The Problems in Outline 16.44

A Bermuda Form arbitration will typically relate to a claim by the policyholder to recover sums paid to third parties to resolve the policyholder’s legal liability to them. The liability may have resulted from a court decision against the policyholder or a settlement of the underlying claim. If the underlying claim became a mass tort action, numerous actions may have proceeded to judgment or settlement. In these circumstances, there will inevitably be documentation, often a large volume of documentation, that relates to the 65 Waugh v British Railways Board [1980] AC 521. 66 Re Highgrade Traders Ltd [1984] BCLC 151; Director of the Serious Fraud Office (above n 46) [102]–[107]. 67 Wentworth v Lloyd (1864) 10 HLC 589. 68 West London Pipeline & Storage Ltd v Total UK Ltd [2008] EWHC 1729 (Comm), [2008] 2 CLC 258 [86(1)]. 69 Digicel (St Lucia) Ltd v Cable and Wireless Plc [2009] EWHC 1437 (Ch), [53]; Farm Assist Ltd (In Liquidation) v Secretary of State for Environment Food & Rural Affairs [2008] EWHC 3079 (TCC), [2009] BLR 80, [54].

Waiver of Privilege

359

underlying claim, including in all likelihood advices and views expressed by the policyholder’s internal and external counsel. These advices and views will often be relevant to the issues arising in a Bermuda Form arbitration. For example, a question may arise under the occurrence definition as to whether the policyholder expected or intended damage or allegations and, if so, to what degree.70 If the claim has been settled, there may be a question as to whether the settlement was reasonable. Or the insurer may contend that the overall settlement amount was reasonable only because the policyholder needed to settle claims that were not covered or were excluded by the terms of the policy. Alternatively, the insurer may raise an issue of misrepresentation, alleging that the policyholder misrepresented facts about its knowledge or understanding of the underlying claims in response to questions on the insurer’s application form. For example, the policyholder may be asked to identify claims expecting to cost in excess of US$1 million. If the policyholder fails to refer, in response to the insurer’s questions, to a claim that ultimately costs more than that, the question may arise as to how the policyholder viewed the claim at the time of the application.

16.45

These and similar issues give rise to questions concerning waiver of privilege. Advice from internal and external counsel on the merits of claims is, for the reasons set out above, privileged by reason of either legal advice privilege or litigation privilege, and usually both. But the insurer may be anxious to see these advices, as they may provide information that will bolster the insurer’s defences to coverage. Equally, the policyholder may want to produce some of these advices, as they may assist in rebutting the insurer’s case. For example, they may show that, although a plaintiff may ultimately have won a case against the policyholder, the policyholder had received legal advice showing that the policyholder never anticipated the amount of damages ultimately recovered by the plaintiff, or showing that a problem which ultimately developed into a mass tort was initially seen as confined to a particular location or set of circumstances. The policyholder may, however, be wary of producing privileged documents, for fear that they will lead to orders for production of these privileged documents in the underlying cases.

16.46

The starting point in considering these issues is that the policyholder is not required to produce its privileged documents as a prerequisite to recovering from its liability insurer merely because they are relevant. English law does not permit any balance to be struck between the interests protected by privilege and other interests. A person for whose benefit documents are privileged has a right to assert and maintain that privilege. That right is subject to a number of identifiable and established exceptions. However, unless one of those exceptions applies, a party is entitled to assert and maintain privilege. That is so even where the privileged documents are relevant (even directly relevant) to an issue in the case. There is no general exception permitting privilege to be set aside because the privileged material is relevant, or permitting privilege to be ‘balanced’ against other interests.71

16.47

70 See above ch 7. 71 See in particular R v Derby Magistrates’ Court ex parte B (above n 33), especially at 503, 507, 508, 509, 511–12. See also Paragon Finance v Freshfields (above n 24); Three Rivers (No 6) (above n 46) [25].

360

16.48

Discovery, Privilege and Waiver of Privilege

The insurer may, however, assert that the policyholder should disclose its privileged documents, arguing either that the insurer shares a common interest with the policyholder or that, by commencing proceedings or placing certain matters ‘at issue’ in the proceedings, the policyholder has waived privilege. But, for the reasons explained in the following paragraphs, these arguments are unlikely to succeed. Common Interest Privilege

16.49

Applications by insurers for the policyholder’s privileged documents, based upon ‘common interest’ arguments, are sometimes (but by no means invariably) made. Such applications give rise to questions as to the law to be applied in resolving the application, as well as issues relating to the exercise of the tribunal’s discretion. In part, this is because there are, or at least may be, some differences between (i) the approach of English law to common interest privilege, as reflected in a relatively small number of decisions of the English courts, and (ii) the approach under New York law and indeed the laws of other states in the United States.

16.50

Issues of applicable law and discretion also arise for a variety of other reasons. The policyholder’s privileged documents are likely to have been created in the United States and in the context of relationships (for example, between US attorneys and clients) which are far removed from English law. There will probably have been an expectation that the documents, when created, were privileged and confidential to the policyholder. The privileged documents (for example, advice as to the merits of the underlying claims or the defence strategy to be pursued) will have been created in the context of ongoing US litigation by third party claimants, and this litigation may even be continuing at the time of the Bermuda Form arbitration. Questions may arise as to what would have been, or might still be, the impact, in the context of US litigation, of the production of privileged documents by the policyholder to its insurers, and specifically whether this might result in their becoming available to the underlying plaintiffs.

16.51

These issues would not arise in a straightforward case where the insurer had exercised its discretion under Article VI.D(1) of the Bermuda Form to associate in the defence of the underlying claims. If there had been a joint defence, then ordinarily the insurer will have seen the relevant advices on the merits and strategy. If the insurer had for some reason not seen them, then the policyholder would generally be unable to resist an order for production of privileged documents created in the context of that joint defence.72 But in a Bermuda Form case, it is unlikely that the insurer will have participated in the defence of the underlying claims. In a typical case, the insurer will have made an early reservation of rights, the policyholder will have conducted the defence, and the insurer subsequently denied liability, leading to arbitration.

16.52

If an application is made for production of privileged documents, the first issue to be addressed should be whether or not the express or implied terms of the Bermuda Form 72 Under English law, where there is a joint retainer of an attorney, neither client can assert privilege against the other: see para 16.62 below. The position is similar under New York law; see paras 16.55–16.56 below.

Waiver of Privilege

361

policy entitle the insurer to see the policyholder’s privileged documents, or conversely permit the policyholder to decline to produce its privileged documents to an insurer who requests them. A number of clauses are potentially relevant in this context: Article VI.B (Inspection) and Article IV.D(2) (Co-operation), and Article VI.F (Loss Payable), which permits the insurer to ‘examine the underlying facts giving rise to a judgment or settlement by the Insured to determine if, and to what extent, the basis of the Insured’s liability under such judgment or settlement is covered by this Policy’. Any issue as to whether there is a contractual entitlement on the insurer’s part to privileged documents must necessarily be resolved by reference to New York law, which governs the parties’ substantive rights. The general trend of the US case law is that co-operation and inspection of records clauses do not permit insurance companies access to otherwise privileged materials.73 New York law follows this approach. In Gulf Insurance Co v Transatlantic Reinsurance Co,74 the court said that: ‘Access to records provisions in standard reinsurance agreements, no matter how broadly phrased, are not intended to act as a per se waiver of the attorney-client or attorney-work product privileges. To hold otherwise would render these privileges meaningless.’ That decision was applied by the Appellate Division in JP Morgan Chase & Co v Indian Harbor Insurance Co.75 Accordingly, the generally worded clauses of the Bermuda Form will not be sufficient, under New York law, to entitle the insurer to the policyholder’s privileged documents. There appears to be no injustice in this approach. If an insurer wishes to have an entitlement to these documents, then it should seek to include specific wording which puts the policyholder on notice that its privileged documents will be disclosable.

16.53

If the express terms are insufficient to create the entitlement, then it follows that any argument based upon an implied term would equally fail.76 There are, certainly, no New York authorities which provide support for an argument that a policyholder has impliedly agreed to make privileged documents available to its insurer. Indeed, it does not seem that the argument has ever been advanced. An implied term would also be inconsistent with the authorities discussed in the following paragraph.

16.54

In addition to rejecting arguments based upon generally worded express terms, New York courts have also repeatedly rejected attempts by insurers and reinsurers to invoke a ‘common interest’ doctrine in order to obtain production of privileged materials, based on the parties’ shared interest in seeing the defeat of an underlying claim. In order to obtain such documents, the insurer must show something more than simply a

16.55

73 The US authorities are reviewed in AB Moore and AJ Wheeler, One Man’s Shield, Another Man’s Sword: Contesting Privilege in Coverage Litigation 22 Mealey’s Litigation Report: Insurance No 27 (15 May 2008). See too BR Ostrager and TR Newman, Handbook of Insurance Coverage Disputes, 19th edn (Austin, Wolters Kluwer, 2019) s 2.07[b]. 74 Gulf Ins Co v Transatlantic Reins Co 788 NYS2d 44, 45–46 (App Div 2004). 75 JP Morgan Chase & Co v Indian Harbor Ins Co 947 NYS2d 17 (App Div 2012). 76 In D Scorey, R Geddes and C Harris, The Bermuda Form: Interpretation and Dispute Resolution of Excess Liability Insurance, 2nd edn (Oxford, OUP, 2012) paras 19.81–19.85, the authors articulate the possible implied term argument which insurers might advance, whilst also stating that the thrust of the policyholder’s response is ‘also forceful’. The authors do not, however, discuss the New York law cases which have rejected arguments based on co-operation or inspection of records clauses, and also rejected arguments based on common interest privilege.

362

Discovery, Privilege and Waiver of Privilege

relationship of insured and insurer, for example, joint representation or a joint defence effort and strategy.77 In American Re-Insurance Co v United States Fidelity & Guaranty Co,78 the New York Appellate Division said: As a general rule, ‘there is no automatic waiver of the attorney-client privilege merely because [the parties] have a “common interest” in the outcome of a particular issue’. Indeed, ‘the fact that an attorney’s services for a client benefitted a third party does not establish an attorneyclient relationship’. The clearest indication of common interest is dual representation. Common interest also extends to a situation where there is a joint defense or strategy, but separate representation. Here, there is neither dual representation nor a joint defense or strategy … Moreover the parties’ interests in the present action are indisputably adverse, and the mere fact that they shared an interest in the eventual outcome of the underlying coverage litigation is not sufficient to create a common interest so as to defeat USF&G’s claimed privileges. Indeed, the Reinsurers were ‘not responsible for the defense [of the underlying lawsuit], although [they] indirectly benefitted from it’. Accordingly, we find the common interest doctrine inapplicable.

16.56

It is not only in the situation where the insurer is seeking the policyholder’s privileged documents that the New York courts have declined to accept that there is a common interest, absent joint representation or a joint defence effort. The same approach has been applied where a third party claimant seeks access to the policyholder’s privileged documents in a situation where these have been shared by the policyholder with an insurer who is not conducting a joint defence. A decision that the policyholder does not share a common interest with the insurer will expose the policyholder to the risk that a third party (in particular a claimant in litigation against the policyholder) will contend that any privilege in the disclosed documents has been waived. In In re Pfizer Inc Securities Litigation,79 the third party claimant sought correspondence between Pfizer and its insurers relating to insurance coverage for certain litigation. The court held (applying a decision of the US Court of Appeals for the District of Columbia Circuit)80 that disclosure of attorney–client communications to an insurer, where there was no joint defence strategy, waived privilege in the communications. The court held that there was no privilege which generally protected communications between policyholder and insurer. However, it also held that disclosure of attorney-work product documents to the insurer did not automatically waive privilege in those documents. This was not, however, because the documents were protected by ‘common interest’ privilege, but rather because New York courts generally find a waiver only if the disclosure substantially increased the opportunity for potential adversaries to obtain the information.

16.57

The approach in In re Pfizer Inc Securities Litigation, resulting in disclosure or potential disclosure of documents to third parties, has been taken in a number of subsequent 77 North River Ins Co v Philadelphia Reins Corp 797 F Supp 363, 367 (DNJ 1992), rev’d on other grounds, 52 F3d 1194 (3d Cir 1995) (although it is a New Jersey case, the decision in North River v Philadelphia has been applied in subsequent New York case law); International Ins Co v Newmont Mining Co 800 F Supp 1195, 1196–97 (SDNY 1992); North River Ins Co v Columbia Cas Co No 90 Civ 2518, 1995 US Dist Lexis 53 (SDNY 1995); Gulf Ins Co v Transatlantic Reins Co 788 NYS2d 44, 46 (App Div 2004); American Re-Ins Co v US Fidelity & Guar Co 837 NYS2d 616 (App Div 2007); Egiazaryan v Zalmayev 290 FRD 421, 434 (SDNY 2013); Granite State Ins Co v R&Q Reins Co No 654494/2013, 2015 NY Misc Lexis 2616, at *3–*4 (Sup Ct 21 July 2015). 78 American Re-Ins Co v US Fidelity & Guar Co (above n 77). 79 In re Pfizer Inc Sec Litig No 90 Civ 1260, 1993 WL 561125 (SDNY 23 December 1993). 80 Linde Thomson Langworthy Kohn & Van Dyke PC v Resolution Trust Corp 5 F3d 1508 (DC Cir 1993).

Waiver of Privilege

363

cases in New York and elsewhere.81 Whether or not disclosure to an insurer will result in a waiver that can be relied upon by third parties will depend upon the law applied by the particular court in the United States that becomes seised of the issue as and when any application is made by the third party. Where a policyholder is facing litigation in numerous jurisdictions in the United States, a policyholder will likely (and, in our view, justifiably) be unwilling voluntarily to disclose privileged documents to an insurer (unless there is a joint defence), not least because of concerns about waiver and the impact of disclosure on the underlying proceedings. Against this New York legal background, a policyholder who is faced with an application by a Bermuda Form insurer for disclosure of privileged documents will seek to persuade the tribunal that it should apply New York law to the issue, both as a matter of principle or discretion. In contrast, an insurer will likely argue that the issue should be approached purely as a matter of English law, arguing that the English authorities (discussed further below) adopt a more expansive approach to common interest privilege than New York law.

16.58

We think that there is much to be said for the argument that New York law should be applied as a matter of principle to claims for production of documents based upon common interest arguments, and that this remains the case even if English law is taken as the starting point for consideration of the common interest issue. This is, first, because the question of whether there is a sufficiently close ‘community of interest’82 between insured and Bermuda Form insurer, so as to entitle the insurer to see the insured’s privileged documents, should be determined by reference to the applicable law of that relationship – ie, New York law. The contrary approach, namely to look exclusively to English law to determine the rights and obligations between the parties in relation to privileged documents, would seem artificial if not wrong in principle. Second, it is clearly established (under English law) that the alleged joint interest in the privileged documents must exist at the time when the documents were created.83 In most cases, the relevant privileged documents will have been created for the purposes of defending the underlying litigation and at a time long before the arbitration proceedings between policyholder and insurer had commenced. Accordingly, at the time that the documents were created, the only relationship between insured and insurer was the relationship created by the contract (governed by New York law), rather than any further relationship created by the commencement of arbitration proceedings.

16.59

There is also, in our view, much to be said for a tribunal generally declining to order the production of privileged documents as a matter of discretion, for a variety of reasons. The documents relating to the defence of underlying litigation conducted by

16.60

81 Go Medical Indus Pty Ltd v CR Bard Inc No 3:95 MC 522, 1998 WL 1632525 (D Conn 1998); Aiena v Olsen 194 FRD 134 (SDNY 2000); SR Int’l Business Ins Co v World Trade Center Props LLC No 01 CIV 9291 (JSM) 2002 WL 1334821 (SDNY 2002); Calabro v Stone 225 FRD 96 (SDNY 2004); cf Kingsway Financial Services Inc v PriceWaterhouseCoopers LLP No 03-Civ-5560 (RMG) (HBP) 2008 WL 4452134 (SDNY 2008), where privilege was upheld in relation to communications in furtherance of a joint defence. See generally D Greenwald and ML Slachetka, Protecting Confidential Legal Information: A Handbook for Analyzing Issues under the Attorney-Client Privilege and the Work Product Doctrine (Chicago, Jenner & Block, 2019) 231–35. 82 See the expression used by Moore-Bick J in Commercial Union Assurance Co plc v Mander [1996] 2 Lloyd’s Rep 640, 644. 83 Ibid, at 646.

364

Discovery, Privilege and Waiver of Privilege

the policyholder’s attorneys, and the communications between the attorney and the policyholder in relation to those proceedings, will have been created in the expectation that the documents will be and remain privileged. There are also likely to be good reasons why a policyholder will not want to share privileged documents with an insurer who has reserved rights, is not conducting or participating in a joint defence, and whose interest may potentially be adverse to the policyholder; in particular, the risk (discussed above) that disclosure may constitute a waiver entitling third party claimants to see sensitive material which may be prejudicial to the policyholder’s defence. 16.61

Since New York law gives no help to an insurer who seeks disclosure of the policyholder’s privileged documents, an insurer may seek to persuade the tribunal that the issue of common interest privilege should be resolved solely by reference to English law on the basis that it is the law of the arbitration. It is therefore necessary to consider English law in this area in order to see whether it leads to a result different from New York law, including whether it compels an analysis which pays no regard to New York law. For the reasons that follow, we do not think that English law leads to either such conclusion.

16.62

Under English law, the expression ‘common interest privilege’ can be used to describe a number of related concepts. Where two clients jointly instruct a single lawyer, there is a joint retainer.84 The communications between the lawyer and each of the clients are privileged against outsiders, but not as between the clients. In those circumstances, therefore, the concept of ‘common interest’ privilege has two features: negatively, it operates as a ‘shield’ that preserves privilege despite disclosure between those with the common interest; and positively, it can be said to operate as a ‘sword’ and to preclude the assertion of privilege between the joint clients. The same principle applies even if there is no joint retainer, but where persons have a joint interest in the subject matter of the communication. This can apply, for example, in the context of company and shareholder, or trustee and beneficiary.85 Neither ‘joint retainer’ nor ‘joint interest’ arise, however, simply because of the existence of a relationship between liability insurer and policyholder.

16.63

It is now well established that ‘common interest’ privilege can arise under English law, certainly to protect privileged communications from disclosure to third parties, in situations outside joint retainer and joint privilege. The origin of the line of authority, whereby confidential documents which have been disclosed to another person remain privileged against third parties, is the decision of the Court of Appeal in Buttes Gas & Oil Co v Hammer (No 3).86 Thus, privilege has been extended to those who are not in fact jointly instructing a single lawyer, but might have done so. In those circumstances – for instance, if several people with the same interest exchange information between themselves – such disclosure will not normally result in loss of the privilege. What sort of interest will suffice? Some of the tests proposed are so broad that they provide rather little concrete assistance.87 It has sometimes been suggested that the question is whether 84 See Hollander (above n 44) at 19-01; Thanki (above n 44) at 6-02. 85 Hollander (above n 44) at 19-02; Thanki (above n 44)at 6-07–6-13. 86 Buttes Gas and Oil Co v Hammer (No 3) [1981] QB 223; Commercial Union Assurance Co plc v Mander (above n 82) 645. 87 For example, Formica v Export Credit Guarantee Department [1995] 1 Lloyd’s Rep 692.

Waiver of Privilege

365

the parties could share a common solicitor if they chose to do so.88 But, as Hollander points out,89 this seems unnecessarily restrictive, since there may be a common interest sufficient to justify some confidential exchange of documents even where there are known conflicts which would be too profound to permit dual representation. Such is the case, for instance, where co-defendants have both a common interest in defeating the claim, but also conflicting interests (if that objective fails) in pinning primary responsibility on each other. The tendency in this context has been to expand the range of relationships in which common interest can be used as a shield, thereby permitting disclosure between willing parties without risk of the loss of privilege. There is now no doubt that, from the perspective of English law, the relationship between insurer and insured, and reinsurer and reinsured, can give rise to a common interest sufficient to justify disclosure without loss of privilege.90 However, the need to identify relationships where common interest can be used as a shield has been diminished, and perhaps made redundant,91 by the establishment of the principle that where the document is shown to a limited class of people under conditions of confidentiality, a party does not lose its entitlement to privilege.92

16.64

But does the expansion of the circumstances where a communication can be protected automatically also expand the range of circumstances in which ‘common interest’ privilege can be invoked in order to obtain another party’s otherwise privileged documents – ie, in which common interest is used as a ‘sword’? Does it always follow that if mutual disclosure would be possible, it is mandatory?

16.65

The decision in Commercial Union Assurance Co plc v Mander makes clear that it does not. In that case, which contains a detailed discussion of prior English authority, Moore-Bick J said:

16.66

[I]t is not enough that the person seeking disclosure of confidential documents can show that he has an interest in the subject matter which would be sufficient to give rise to common interest privilege if the documents had been disclosed to him: he must be able to establish a right to obtain access to them by means of a common interest in their subject matter which existed at the time the advice was sought or the documents were obtained.93

Thus, whether privileged documents may be safely disclosed as a matter of voluntary decision is one question. Whether they must be disclosed is quite another. It is therefore potentially misleading to refer to common interest privilege being used as a ‘sword’. That shorthand tends to suggest that if the circumstances are such that common 88 The Good Luck [1992] 2 Lloyd’s Rep 540, 542. 89 Hollander (above n 44) at 19-07. 90 Guinness Peat Properties Ltd v Fitzroy Robinson Partnership [1987] 1 WLR 1027; Svenska Handesbanken v Sun Alliance and London Insurance PLC [1995] 2 Lloyds Rep 84; Winterthur (above n 46) [80]. Contrast the position in New York law, discussed in paras 16.55–16.56 of the text. 91 Hollander (above n 44) at 19-11. 92 See below, para 16.90. 93 See Commercial Union Assurance Co plc v Mander (above n 82) 648 (and 644–45), disapproving Formica v Export Credit Guarantee Department (above n 87) to the extent that it might have assumed symmetry between the two questions. In James-Bowen v Commissioner of Police of Metropolis [2018] UKSC 40, [2018] 1 WLR 4021, the Supreme Court recognised (at [39]–[45]) that there was no such symmetry.

366

Discovery, Privilege and Waiver of Privilege

interest privilege would protect documents given by A to B from disclosure to third parties, then it follows that A is not entitled to assert privilege in documents which had never been provided to B. Accordingly, talking in terms of common interest privilege as a ‘sword’ can be said to be misplaced.94 Hollander says that although situations in which B has rights of access to A’s privileged documents have been described as using common interest privilege as a ‘sword’, the phrase has now fallen out of fashion in view of the disparate nature of the cases.95 16.67

Whether or not it is correct to refer to using common interest privilege as a ‘sword’, there is no doubt that in certain circumstances a party may have rights of access to another party’s privileged documents. The leading textbooks contain a detailed discussion of the relatively small handful of English cases96 in which the courts have referred to common interest privilege in the context of granting B rights of access to A’s privileged documents.97 Hollander suggests that an entitlement to access privileged documents arises ‘in circumstances where there is a relationship between the parties close to joint privilege or where one party has a contractual right to obtain documents from the other’.98 The relevant question is therefore whether the claim for privilege is ‘consistent with the underlying relationship’.99 If not, then the document should be disclosed. Thanki suggests that the key question is whether the litigating parties have a ‘joint interest’ in the confidential document.100 It is clear that something more than a shared interest in the outcome of the litigation is required.101

16.68

In our view, none of the English cases provides a foundation for the proposition that a Bermuda Form insurer is entitled to see the policyholder’s privileged documents. It is clear from the New York authorities discussed above that neither the terms of the policy nor the relationship of insured and insurer creates any entitlement on the part of the insurer to access the policyholder’s documents. A claim for privilege is therefore wholly consistent with the underlying relationship. Indeed, the assertion by a liability insurer (who is not conducting a joint defence or strategy) of an entitlement to the privileged documents is inconsistent with the relationship between the parties. Thus, if (at the time that the documents were created) the insurer had asked to see the insured’s privileged documents, the insured would be entitled to refuse.102 It must also follow that a Bermuda Form liability insurer cannot establish, simply by virtue of his or her status as a liability insurer, a joint interest in the policyholder’s confidential documents.

94 Thanki (above n 44) at 6-55. In Winterthur (above n 46), counsel accepted that common interest privilege can be used as a sword (at [59]). 95 Hollander (above n 44) at 19-12. 96 CIA Barca de Panama v George Wimpey & Co [1980] 1 Lloyd’s Rep 598; Formica v Export Credit Guarantee Department (above n 87); Commercial Union Assurance Co plc v Mander (above n 82); Winterthur (above n 46). See too Accident Exchange v McLean [2018] EWHC 23 (Comm). 97 The invocation of ‘common interest’ privilege as a ‘sword’ is discussed in detail in Hollander (above n 44) at 19-07–19-17; and Thanki (above n 44) at 6.41–6.59. 98 Hollander (above n 44) at para 19-12. 99 Ibid, para 19-16. The equivalent statement in Malek (above n 44) was approved in James-Bowen v Commissioner of Police (above n 93) at [41]. 100 Thanki (above n 44) at para 6.46. 101 James-Bowen v Commissioner of Police (above n 93) [42]. 102 See the test identified by Hollander (above n 44) at 19-14.

Waiver of Privilege

367

Furthermore, none of the English ‘common interest’ privilege cases has considered issues arising from the potential application of different laws. The three insurance cases (ECGD, Mander and Winterthur) all concerned insurance contracts governed by English law. The rights and obligations of the parties were therefore considered in an English context.103 In a purely English context, there is no doubt that there is a common interest sufficient to justify disclosure without loss of privilege as between a liability insurer and a policyholder. However, as the previous discussion of the New York cases shows, this is not the position under New York law in the absence of a joint defence strategy, and it cannot be assumed to be the position under all of the different laws where litigation against the policyholder may take place. This is important because one of the lynchpins underlying the English courts’ approach to common interest privilege is that if the documents had been shared with the person asserting ‘common interest’, the documents would remain privileged. It is impossible to see that an English court would countenance disclosure of privileged material on the basis of ‘common interest’ if the effect of disclosure would result in third parties being able to obtain disclosure of the documents.104

16.69

We consider that those English cases where disclosure is required are best seen as resting on some express or implied obligation of co-operation or to permit access. Such obligations readily flow, as a matter of necessary implication, from an agreement jointly to instruct a lawyer, or in some of the other cases where common interest has been held to have such consequences, such as between partners105 or as between a company and its shareholders.106 But not every situation of common interest will give rise to such an implication.

16.70

Whether or not an insurer will be entitled to see the policyholder’s privileged documents therefore depends on the terms of the policy and the extent to which it confers such rights.107 Moreover, such rights cannot subsist where the insurer denies the existence of the interest.108 Nor, indeed, does common interest operate as a ‘sword’, even where a lawyer is jointly instructed, once a conflict of interest arises.109

16.71

The most detailed discussion in the English cases, in the insurance context, is the decision of Moore-Bick J in Commercial Union Assurance Co plc v Mander, where a reinsurer sought an insurer’s advices and other privileged documents. The judge considered that there was sufficient community of interest between the reinsurer and insurer in that case,

16.72

103 For example, Aikens J’s reference in Winterthur (above n 46) [115] to the insurance contract being one of ‘utmost good faith’ is an English concept, which is not applicable to a New York law insurance contract. 104 Commercial Union Assurance Co plc v Mander (above n 82) 646, where Moore-Bick J said that the insurers in that case could not withhold documents from the reinsurers and that it ‘would follow, of course, that documents of that kind would be subject to common interest privilege in the hands of the reinsurer if communicated in confidence to him’. 105 Re Pickering (1884) 25 Ch D 247. 106 Re Hydrosan Ltd [1991] BCLC 418. 107 Brown v Guardian Royal Exchange Assurance plc [1994] 2 Lloyd’s Rep 325, 329; Commercial Union Assurance Co plc v Mander (above n 82) 648. 108 Commercial Union Assurance Co plc v Mander (above n 82) 648 (purported avoidance of policy defeated a claim to be entitled to see the documents, even though the documents were created prior to the avoidance). It is doubtful whether this principle would apply where the documents were created at a time when the insurer had simply reserved rights and there was a subsequent denial. However, once the insurer had denied coverage for the claim, it is difficult to see how the insurer could assert any community of interest thereafter. 109 TSB Bank plc v Robert Irving & Burns [2000] 2 All ER 826.

368

Discovery, Privilege and Waiver of Privilege

but this was only because of the ‘follow the settlements’ clause in the reinsurance, pursuant to which the reinsurer was bound to accept the insurer’s handling of the original claim, provided that the insurer had acted in good faith and in a business-like manner.110 The case does not suggest that a reinsurer or insurer is entitled to the privileged documents of the reinsured or insured, simply by virtue of his position as reinsurer or insurer and therefore its general interest in the outcome of the underlying claim. The ‘follow the settlements’ clause created the necessary community of interest, notwithstanding that the reinsurer may only have had a limited right to inspect the insurer’s documents under the terms of the policy.111 16.73

The focus in Mander on the specific terms of the contract is consistent with the approach taken in other cases. In Brown v Guardian Royal Exchange Assurance plc,112 the liability insurer successfully obtained the insured’s documents as a result of clause 8 of the relevant policy. The case was not decided on the basis of common interest privilege, but Neill LJ made it clear that the ‘fact that insurers fund the cost of legal advice and representation and have a common interest in the defeat of a claim against their insured does not necessarily mean that they are entitled to see all the documents passing between the insured and his solicitors’.113 In Winterthur Swiss Insurance v AG (Manchester) Ltd, Aikens J held (at [80]) that the relationship between insured and insurer ‘can’ give rise to a common interest. He did not say that it automatically did so.114 He went on to hold that the terms of the subrogation clause in the policy entitled the insurer to certain privileged documents of the insured.115 The terms of the contract, including the subrogation clause, and the documents related to the contract were also relied upon by Aikens J in accepting the argument that common interest entitled the insurer to see the privileged documents.116

16.74

In the light of those authorities, the relevant question seems to be: does the Bermuda Form expressly or by necessary implication grant the insurance company a right to see privileged documents in circumstances where they have not jointly appointed the lawyers in question and are denying liability? In our view, no version of the Form goes so far. Early versions granted insurers the right to ‘associate’ in the defence of suits against the policyholder. If the insurance company chose to exercise that right, it would be entitled to access privileged information which resulted from the joint defence. But in most arbitrations, the right has not been exercised. The 004 Form gives a similar right, with similar consequences: the insurance company has the right to associate in the defence, ‘in which event’ it shall be afforded full co-operation.117 If it has not exercised its right, that entitlement does not arise. Nor do we consider that the policyholder’s obligation to ‘furnish promptly all information reasonably requested by the Company’118 110 Commercial Union Assurance Co plc v Mander (above n 82) 646. 111 Ibid at 643. 112 Brown v Guardian Royal Exchange Assurance plc (above n 107). 113 Ibid at 329. 114 Winterthur (above n 46) [80], [113]. 115 Ibid at [103]–[111]. 116 Ibid at [113]–[117]. See also the explanation by Hollander (who appeared as successful counsel in the case) in Documentary Evidence (above n 44) at para 19-13. See too Thanki (above n 44) at 6.59. 117 Article VI(D)(1). 118 Article VI(D)(2). See also ch 11 above, para 11.20.

Waiver of Privilege

369

entitles the insurance company to demand access to privileged documents, much less to do so on a broad or unrestricted basis; ‘information’ is not the same thing as ‘documents’, nor would it be reasonable to require wholesale disclosure of legally privileged material.119 New York law, as the substantive law of the contract, governs the question of whether these provisions of the Bermuda Form entitle the insurers to have access to privileged documentation. As explained above, New York courts have rejected attempts by insurers to obtain privileged documents by reliance upon generally worded co-operation or inspection clauses, or upon their interest in seeing the defeat of the underlying claim. Nor, in our view, is there any analogy between a Bermuda Form insurer and a reinsurer (such as the reinsurer in Mander) who has agreed to ‘follow the settlements’ of a reinsured. There is no equivalent to a ‘follow the settlements’ clause in the Bermuda Form policy. On the contrary, Article VI.F provides that liability in respect of settlements does not attach unless and until they have been approved in writing. It goes on to provide for an entitlement on the insurer to ‘examine the underlying facts giving rise to … settlement by the Insured’ in order to determine whether there is coverage. It is also open to a Bermuda Form insurer to challenge the reasonableness of a policyholder’s settlements; the insurer is not bound by those settlements simply because the policyholder has acted in good faith and in a business-like manner.

16.75

Implied Waiver The alternative argument upon which privileged documents may be demanded is based on implied waiver of privilege. It is broadly along the following lines: that the policyholder has impliedly waived privilege by commencing proceedings under its liability policy, since the issues (for example, the expected or intended issue, or questions as to the reasonableness of settlements) necessarily require the tribunal to consider privileged communications in order to fairly resolve the issues raised. Support for such a proposition at one time appeared (conceivably) to be found in certain widely worded passages in the judgment of the Court of Appeal in Lillicrap v Nalder & Sons120 and in a number of cases subsequent to that decision.121

16.76

It is now clear, however, that any such argument is wrong. Waiver is not implied when a policyholder brings a claim for coverage that raises issues to which privileged documents are relevant. In Paragon Finance Plc v Freshfields,122 the Court of Appeal made it clear that the commencement of proceedings operated as a waiver only in a narrow class of case (claims by clients against their solicitors) and that the waiver was itself narrow.

16.77

119 It is conceivable that this provision might, in appropriate circumstances, justify a demand for some particular item of legally privileged information – for instance, a summary of the advice the policyholder is receiving with regard to liability and quantum where consent for a settlement is sought and the insurer is accepting that there is coverage. 120 Lillicrap v Nalder & Sons [1993] 1 WLR 94. 121 See, eg, Kershaw v Whelan [1996] 1 WLR 358; and Hayes v Dowding [1996] PNLR 578, both of which were discussed in Paragon Finance v Freshfields (above n 24). 122 Paragon Finance v Freshfields (above n 24).

370

Discovery, Privilege and Waiver of Privilege

In that case, solicitors were sued for alleged negligence in relation to the creation of a scheme backed by insurance. The insurers had denied liability and the plaintiffs had settled with them. The defendant solicitors sought disclosure of documents relating to their former client’s pursuit and settlement of insurance claims, including privileged documents created after new solicitors had been instructed. The Court of Appeal rejected the application. The court reasoned that the client cannot open up the confidential relationship between solicitor and client and at the same time deprive the solicitor of the means of defending the claim by insisting upon the enforcement of that confidential relationship. But this principle has no application outside actions against the solicitor by the client,123 and even then only applies to documents created in that very relationship. It therefore cannot be transplanted into the sphere of claims by a policyholder against its liability insurer.124 16.78

Accordingly, although English law recognises implied waiver in certain cases, it does so on a narrow basis. The fact that a party makes allegations in respect of which privileged communications are relevant does not, without more, provide a basis for any implied waiver. English law does not hold that, by pleading confidential communications ‘into relevance’, the privilege is lost, and cases decided on that basis (including Hayes v Dowding,125 which rested in part on an analysis of US cases)126 were expressly disapproved by the Court of Appeal in Paragon Finance.

16.79

In order to avoid these difficulties, an insurer may argue that the tribunal should apply the ‘at issue’ doctrine of waiver as applied in the United States. It is unlikely that an English tribunal could be persuaded to take such an approach or that it could properly do so. There is no basis for inviting an English tribunal to apply a doctrine applied by other forums, but rejected by the law of the seat of the arbitration. That is particularly so where the effect of the doctrine would be to remove a privilege which is recognised by the law of the forum, and which that law would regard as not having been waived.127 Moreover, it is difficult to see any logical reason why any law other than English law should determine whether or not there has been a waiver and (if so) what other law could sensibly be applied. The argument for ‘at issue’ waiver would necessarily be reliant on conduct that had taken place in the course of English (arbitration) proceedings. It would follow that English law should determine whether or not that conduct was sufficient to amount to a waiver of privilege. The only other candidate would be New York law. But New York law applies to the parties’ substantive rights and relationship, and the ‘at issue’ doctrine appears to be a matter of New York procedure rather than substance.128

123 See ibid at 1191, and also 1193, where the Court of Appeal indicated that US authority to the contrary did not represent English law. See also Farm Assist Ltd (In Liquidation) v Secretary of State for Environment Food & Rural Affairs [2008] EWHC 3079 (TCC); Digicel (above n 69). 124 Even when an implied waiver does exist in cases brought by a client against its solicitor, the waiver extends only to communications of which they both were aware, not to communications which the solicitor had never seen. 125 Hayes v Dowding (above n 121). 126 Moore and Wheeler (above n 73) and Ostrager and Newman (above n 73) summarise US case law on this topic. 127 See above paras, 16.26–16.28. 128 See above ch 3, paras 3.30–3.45 in particular para 3.45.

Waiver of Privilege

371

Voluntary Waiver by the Policyholder Yet, for the reasons already indicated, the policyholder may consider that it is in its own interests to adduce evidence relating to matters otherwise privileged. For example, the policyholder may wish to call its in-house lawyer or an external lawyer to give evidence as to the legal advice which they gave in relation to an underlying claim at a particular point in time. The policyholder may also wish to rely upon documents that would otherwise be privileged in order to support the evidence called. If the policyholder decides to do so, then there may be a waiver of privilege. Again, a number of questions arise. Which law governs the question of waiver of privilege? What conduct gives rise to a waiver of privilege? And, if there is a waiver, what is the extent of the waiver that has taken place?

16.80

Since the waiver arises or potentially arises from matters arising within the conduct of the (English) arbitration, the question of waiver should be governed by English law. Waiver arises under English law when a party deploys in court material which would otherwise be privileged.129 The opposite party and the court must then have an opportunity of satisfying themselves that what the party has chosen to release from privilege represents the whole of the material relevant to the issue in question.130 A party who makes a partial waiver of privileged material is therefore not entitled to cherry-pick from the material so as to disclose and deploy the part of the material which suits him or her, but withhold the parts which do not.131

16.81

The mere fact that a party refers to a privileged document does not in itself amount to a waiver of privilege. The case law has drawn a distinction between a reference to the ‘effect’ of a privileged document and reliance on its ‘content’.132 It is only if there is a reference to the contents of the legal advice that there may have been a waiver.133 The question is therefore whether the substance or gist or content of the legal advice has been disclosed, or simply its effect.134 The distinction is not always easy to draw. It has been said that the fuller the information about the legal advice, the greater the risk that waiver will have occurred, and that ultimately the question is whether fairness requires that the full advice be made available.135 However, ‘fairness’ is not the touchstone by which it is determined whether there has been a waiver of privilege.136

16.82

129 This has been described as ‘express’ waiver, in contrast to ‘collateral waiver’ and ‘implied waiver’; Glenn v Watson [2016] EWHC 3259 (Ch), [2017] 4 WLR 48, [14]. Collateral waiver arises when there is an express waiver of privilege, and fairness requires the disclosure of additional material relating to the transaction in question; see below, para 16.82. Implied waiver most commonly arises where parties sue their solicitors; Glenn v Watson [16] and see above, paras 16.76–16.78. 130 Nea Karteria Maritime Co Ltd v Great Lakes Steamship Corp (No 2) [1981] Com LR 138 (Mustill J), cited in many subsequent authorities, eg, Dunlop Slazenger International Ltd v Joe Bloggs Sports Ltd [2003] EWCA Civ 901, [11]. 131 Digicel (above n 69) [14]. 132 Dunlop Slazenger v Joe Bloggs (above n 130) [11]. 133 Digicel (above n 69) [27]; MAC Hotels Ltd v Rider Levett Bucknall UK Ltd [2010] EWHC 767 (TCC), paras [60]–[61]. 134 Brennan v Sunderland City Council [2009] ICR 479, [64]. 135 Ibid at [65]–[67]. 136 Digicel (above n 69) [41].

372

Discovery, Privilege and Waiver of Privilege

16.83

English law therefore takes a fairly narrow approach to waiver, which is not easily established.137 Witnesses in English arbitration proceedings may therefore be able to give evidence about matters which would risk a finding of waiver in proceedings in the United States. For example, if there is an issue as to the reasonableness of settlements, lawyer witnesses may be able to give evidence as to the way in which the settlements were concluded, and their general approach to settlement and their view of reasonableness, without deploying the contents of any legal advice and thereby waiving privilege. Equally, however, it may be necessary for ‘difficult and invidious choices’138 to be made by a party which needs to prove or rebut a particular point, and could more easily do so by relying on the contents of privileged documents. Where there is no waiver, a court or tribunal will ‘simply have to do the best that it can to come to what it hopes will be the right conclusion on all the evidence presented’.139

16.84

Under English law, a party who waives privilege in relation to one communication is not held to have waived privilege in relation to all related communications. In other words, English law does not apply a broad subject-matter waiver as some US jurisdictions do. Equally, however, English law refuses to allow a party to waive privilege in such a partial and selective manner that unfairness may result.140 Accordingly, when there has been a waiver of privilege, a party waives in relation to what the courts have described as the specific ‘transaction’ in question.141 For example, if there is an issue as to what was said by witness X to Y in a particular conversation, a party may wish to put in evidence a document that records what witness X reported to its lawyer on the day of the conversation. In such a case, the waiver would extend to the conversation itself, that being the relevant ‘transaction’ for these purposes. The waiver would not extend to all privileged communications concerning the matters discussed in the course of the conversation between X and Y. Nor would the waiver open up the confidentiality of later conversations between X and the lawyer, so as to require disclosure of all draft witness statements made by X.142 Difficulties can of course arise in identifying the precise limits of the ‘transaction’ in respect of which the waiver has taken place, but the tendency of English law is to draw the boundaries fairly narrowly.143

16.85

Questions also arise as to when the waiver takes place. For example, a party may disclose some privileged documents as part of the discovery process. The party may then include reference to those documents, or to privileged advice generally, in the witness statements served in the run-up to the substantive hearing. At what stage can the other party ask for discovery of all documents relating to the ‘transaction’ in question?

16.86

It appears that the party must wait until the privileged material is deployed to the arbitrators before it may seek production of the privileged communications. This will usually be at the stage when a party actually calls the witness in question or relies upon the 137 Ibid [66]. 138 MAC Hotels (above n 133) [43]. 139 Digicel (above n 69) [53]. 140 Paragon Finance v Freshfields (above n 24) 1188. 141 Civil Aviation Authority v R (above n 47) [111]–[114]. 142 General Accident Corp v Tanter [1984] 1 WLR 100. 143 Fulham Leisure Holdings Ltd v Nicholson Graham & Jones [2006] EWHC 158 (Ch), [2006] 2 All ER 599, [14–18], which contain a review of the case law. See too Civil Aviation Authority v R (above n 47).

Waiver of Privilege

373

privileged documents as part of its case. Normally, this will happen at the substantive hearing, but it is possible that privileged material will be deployed at an interlocutory hearing.144 Until that time, the party has not gone beyond the point of no return, and the privilege remains. Privilege will, of course, have been waived, vis-a-vis the other party, in the documents and the contents of the witness statement that have actually been given to the other party. But there is still an opportunity for the party to reconsider the position and decide whether it does wish to deploy the relevant material in evidence before the arbitrators.145 This raises the spectre that further extensive discovery will be required in the middle of a hearing, with attendant disruption and additional expense. An arbitral tribunal will be keen to avoid this and to ensure that documents are available well in advance of a hearing. It will be apparent from the foregoing discussion that a party who intends to waive privilege would be well advised to consider the possible ramifications of its waiver well in advance of the hearing itself. The question will almost certainly arise both at the disclosure stage and also at the time when witness statements are being prepared for exchange. A party should try to avoid a situation where, shortly before or in the middle of a hearing, an application for extensive further discovery is made because of a waiver of privilege. To this end, the party should consider identifying with as much precision as possible the scope of the waiver, and if possible seek agreement on the scope with the other side, so as to avoid an argument that the waiver extends further than the party making it intends. A party may also be well advised to give disclosure, well in advance of the hearing, of documents that relate to the transaction in respect of which the waiver of privilege is being made. Again, the party might seek to reach agreement on this issue with the other side – for example, by agreeing that it would be sufficient to disclose all relevant privileged documents from the files of the in-house legal department, but not all the documents from the files of external lawyers.

16.87

If privilege is to be waived in the course of an English arbitration, then a party will need to consider whether such waiver will produce any further consequences in other litigation. For example, the policyholder may be involved in continuing litigation with claimants in the United States. The policyholder may well wish to avoid giving the claimants an argument that the policyholder’s waiver in the arbitration proceedings allows them access to the documents for use in the litigation of the underlying claims. Whether this consequence would follow will, of course, depend upon the laws that apply in the courts of the United States where the issue is raised.146

16.88

Production of Privileged Documents in Previous Proceedings Conversely, the policyholder may be asked to produce in the arbitration documents as to which there may have been a waiver, or partial waiver, of privilege in US proceedings 144 Deployment at an interlocutory hearing is sufficient, even if there is no intention to deploy the evidence at the substantive hearing: Berezovsky v Abramovich [2011] EWHC 1143 (Comm), [13]–[15]. 145 See the discussion in Hollander (above n 44) at para 23-18; Matthews and Malek (above n 44) at paras 16-23–16-27. 146 See the discussion as to the absence of ‘common interest’ privilege protection where the insurer is not participating in the defence or defence strategy at paras 16.53–16.57 above.

16.89

374

Discovery, Privilege and Waiver of Privilege

to which the policyholder was a party. For example, a US court may have compelled the policyholder to produce documents that would otherwise be regarded as privileged, subject to a ‘protective order’. In an effort to protect the policyholder’s privilege, the protective order may seek to limit the use of such documents and thus seek to preserve privilege at least to some extent. The question may then arise in the London arbitration as to the extent to which the insurer can take advantage of this waiver in order to obtain documents that it could not otherwise see. 16.90

The position under English law appears to be as follows. English law regards confidentiality as a necessary precondition to a claim for privilege. If a document enters the public domain, the necessary confidentiality is lost.147 But where the document is shown to a limited class of people under conditions of confidentiality, a party does not lose its entitlement to privilege (except, of course, vis-a-vis those to whom it is shown). The two early cases on this issue, NRG v Bacon & Woodrow148 and City of Gotha v Sothebys,149 both concern disclosure to third parties (and, in the NRG case, disclosure of some documents to the party seeking the others). In each case the court held that, although there was no privilege as against a person to whom the documents had already been disclosed, the privilege is not lost against third parties in circumstances where the disclosure was coupled with obligations of confidentiality. A similar conclusion was reached in B v Auckland District Law Society.150 Documents were provided to counsel appointed by the Law Society to investigate a complaint, on the basis that privilege was not waived and the documents were not to be copied. It was held that privilege had not been waived, since disclosure had been made only for a limited purpose.

16.91

An interesting decision in this context is Bourns v Raychem.151 In that case, B had litigated against R in England, and there had been an order for costs between them. In the course of the ‘taxation’ (that is, assessment) of costs, B had produced privileged material to R. R sought to use those documents in the course of litigation in the United States between the parties and sought an order from the English court allowing such use. The Court of Appeal noted that it was possible to waive privilege for a specific purpose and in a specific context without waiving it for any other purpose or in any other context. Documents disclosed on taxation … are disclosed for the purposes of that taxation and, perhaps absent special circumstances, the privilege is only waived for the purpose for which the documents are disclosed.152

The Court of Appeal accordingly granted an injunction to prevent the use of the documents in the US litigation. It did not strictly matter that there was no order made for production of the documents, but instead that they were disclosed in response to an informal request from R (although, as a matter of background, the Court of Appeal held that if an order had been sought, it would probably have been made). 147 Barings plc v Coopers & Lybrand [2000] 1 WLR 2353; Harris v The Society of Lloyd’s [2008] EWHC 1433 (Comm), [2009] Lloyd’s Rep IR 119, [12]–[13]; Hollander (above n 44) at paras 13-01, 24-02. 148 NRG v Bacon & Woodrow [1995] 1 All ER 976. 149 City of Gotha v Sothebys [1998] 1 WLR 114. 150 B v Auckland District Law Society (above n 33). See also USP Strategies plc v London General Holdings [2004] EWHC 373 (Ch). 151 Bourns v Raychem (above n 28). 152 Ibid at 162.

Other Forms of Discovery

375

It must therefore be strongly arguable that if documents have been disclosed under court order for the purposes of specific litigation, and subject to rules protecting their confidentiality and preventing further dissemination, then they remain privileged. Under English law, in those circumstances, the privilege is not lost, either by waiver or by loss of confidentiality.

16.92

CONFIDENTIAL DOCUMENTS The fact that a document is ‘confidential’ is not equivalent, under English procedural law, to a claim for privilege. Thus, confidentiality does not present any absolute bar to an order for disclosure. Instead, whether documents may be withheld in discovery on grounds of confidentiality depends on the application of the court’s discretion on a case-by-case basis.153 That principle applies equally to documents originating in a confidential arbitration: the confidentiality is recognised and respected in English law, but it is not regarded as absolute, and it is not in itself decisive where disclosure is sought.154

16.93

If a tribunal is asked to order disclosure of confidential (but unprivileged) documents, then it will need to strike a balance between breaking the confidence and the interests of justice. How the balance is struck depends naturally on a number of factors. Duties of confidentiality owed to third parties naturally attract greater protection than mere ‘commercial confidentiality’ protecting the interests of one of the parties. A frequently cited textbook on the topic of disclosure suggests that refusal to order disclosure is the ‘exception rather than the rule’, at least in a situation where the interests of a third party are not affected.155 Much necessarily depends, however, on how important the documents are. And an English court will look for ways to preserve confidentiality at least to some degree, short of complete refusal of any disclosure. For example, disclosure may be made to a limited category of named individuals (such as the lead lawyers in the case) who form a ‘confidentiality club’. In principle there is no reason why a different approach should not be taken in arbitration. The arbitrators have (as set out above) a discretion conferred by statute. The main differences are factual. In particular, the fact that the arbitral proceedings are themselves private, so that disclosure within the proceedings does not mean disclosure to the public in general and carries no serious risk of that happening, needs to be considered.

16.94

OTHER FORMS OF DISCOVERY As has been explained above, an arbitration tribunal has ample power to order parties to disclose their own documents. Discovery in litigation often ranges wider than this. 153 For example, Science Research Council v Nasse [1980] AC 1028. 154 Dolling-Baker v Merrett [1990] 1 WLR 1205; Michael Wilson & Partners v Emmott [2008] EWCA Civ 184, [2008] 1 Lloyd’s Rep 616, [72]–[73], [127]. 155 Hollander (above n 44) at para 10-01.

16.95

376

Discovery, Privilege and Waiver of Privilege

It may extend to documents held by third parties, to discovery of oral evidence by deposition, either of parties or of third parties, and to discovery by interrogatory. In arbitration, however, the availability of pre-trial measures such as these tends to be highly restricted. Oral Discovery: Depositions 16.96

England has no extant tradition of oral discovery, that is, discovery by deposition. Depositions are occasionally used in English litigation as a way of obtaining for trial evidence which might otherwise not be available (for instance, the evidence of a witness who is abroad or who is seriously ill).156 But there is no general practice of requiring or permitting depositions to be taken before trial. In English practice, the written witness statement and expert report serves some of the same functions as a deposition, in that it gives the opposing party advance notice of the evidence a witness is expected to give. But since they are carefully (and sometimes, as many believe, too artfully) prepared by lawyers behind closed doors, witness statements offer no opportunity to test lines of potential cross-examination, to demonstrate inconsistency or to elicit evidence useful to the opposing party’s case. An American lawyer may well think them a poor substitute for the oral deposition.157

16.97

Be that as it may, it is simply not the practice of London arbitration tribunals to require depositions. Indeed, the extent to which they could do so is doubtful. It is probable that a tribunal could, in practical terms, achieve this so far as a party and witnesses who could be identified with that party (for example, current employees) are concerned.158 But it would certainly be an exceptional course for arbitrators in England to order even this. So far as third parties are concerned, arbitrators have no power to compel attendance, whether for depositions or for the hearing itself. It is possible to apply to the English court under section 44 of the Arbitration Act 1996, with the tribunal’s permission (or by agreement), for an order that a witness attend for the purpose of being deposed.159 Such applications are unusual, and the limited case law concerns applications for depositions in the context of arbitrations conducted overseas.160 In the context of an English arbitration, a court is more likely to use its power simply to compel ‘the attendance before the tribunal of a witness in order to give oral testimony’ rather than to order depositions. An order to this effect can be obtained from the English court under section 43 of the 1996 Act, with the tribunal’s permission (or by agreement).161 However, this power only applies to witnesses within the United Kingdom.

156 Depositions can be ordered pursuant to CPR 34.8. 157 See further ch 15, in particular paras 15.7–15.8 and 15.32–15.41 above. 158 The tribunal may decide ‘whether any and if so what questions should be put to and answered by the respective parties and when and in what form this should be done’ (emphasis added): English Arbitration Act 1996, s 34(2)(e). 159 A and B v C, D and E (Taking Evidence for a Foreign Seated Arbitration) [2020] EWCA Civ 409, [2020] 1 WLR 3504. 160 Ibid; Commerce and Industry Insurance Co of Canada v Lloyd’s Underwriters [2002] 1 WLR 1323. 161 English Arbitration Act 1996, s 43 (emphasis added).

Other Forms of Discovery

377

Indeed, securing the attendance or evidence of witnesses outside the United Kingdom at a London arbitration raises some surprisingly tricky problems. The English court cannot use its compulsory process where a witness is not in the jurisdiction. Nor, probably, is an English court entitled to use a letter of request to a foreign court under the Hague Convention on the taking of evidence abroad in civil or commercial matters,162 since Article 1 of the Hague Convention provides that a letter of request ‘shall not be used to obtain evidence which is not intended for use in judicial proceedings, commenced or contemplated’. It seems unlikely that use in a private arbitration tribunal qualifies as use in ‘judicial proceedings’.163

16.98

Nevertheless, it is likely that a combination of section 44 of the Arbitration Act, CPR 34.13 and if necessary the inherent jurisdiction of the court164 would be sufficient to enable an English court to request a foreign court to assist in obtaining evidence for an arbitration. In Silver Dry Bulk Co v Homer Hulbert Maritime Co Ltd, the court was prepared to assume that a power existed under section 44 to make a letter of request for an overseas witness to produce a specific document, although the judge declined to do so as a matter of discretion.165 In Dtek Trading SA v Morozov, the judge considered that a letter of request could be made for a very specific document pursuant to section 44.166 The possibility of an application to the English court under section 44 to make a letter of

16.99

162 847 UNTS 231. 163 There appears to be no authority directly in point; see GB Born, International Commercial Arbitration, 3rd edn (Dordrecht, Kluwer Law International, 2021) 2599–2600. But see, by analogy, Commerce and Industry Insurance Co of Canada v Lloyd’s Underwriters (above n 160) 1326. In that case, the Evidence (Proceedings in Other Jurisdictions Act) 1975, which gives effect to the Convention, and refers to requests from a ‘court or tribunal’, was held to be inapplicable to a request by an arbitral tribunal. This decision was followed in Benhurst Finance Ltd v Colliac [2018] EWHC 2188 (QB). Until recently, US courts interpreting the expression ‘foreign or international tribunal’ in 28 USC § 1782 have generally concluded that it does not cover a private arbitral tribunal: National Broadcasting Co v Bear Stearns & Co 165 F3d 184 (2d Cir 1999); In re Application of Republic of Kazakhstan 168 F3d 880 (5th Cir 1999); In re Application of Medway Power Ltd 985 F Supp 402 (SDNY 1997). Following the decision of the Supreme Court in Intel Corp v Advanced Micro Devices Inc 542 US 241 (2004), some US courts concluded that § 1782 does cover private arbitration tribunals: In re Oxus Gold plc No Misc 06-82, 2006 WL 2927615 (DNJ 11 October 2006); Re Roz Trading Ltd 469 F Supp 2d 1221, 1222 (ND Ga 2006); In re Hallmark Capital Corp 534 F Supp 2d 951 (D Minn 2007); In re Chevron Corp 709 F Supp 2d 283 (SDNY 6 May 2010). Other courts have taken the opposite view: In re Arbitration in London, England, between Norfolk S Ry and General Sec Ins Co and ACE Bermuda Ltd 626 F Supp 2d 882 (ND Ill 2009), where the underlying case was a Bermuda Form arbitration in London; El Paso Corp v La Comision Ejecutiva HidroElecctrica Del Rio Lempa 341 F App’x 41 (5th Cir 6 August 2009); In re Application of Operadora DB Mexico, SA DE CV No 6:09-cv-383-Orl-22GJK, 2009 WL 2423138 (MD Fla 4 August 2009). Many of the authorities after Intel are referred to in the decision of the US Court of Appeals for the Second Circuit in In re Application and Petition of Hanwei Guo 965 F3d 96 (2d Cir 2020). The court there reaffirmed its earlier holding in National Broadcasting Co v Bear Stearns that § 1782 does not apply to international private commercial arbitration. The court applied a multi-factor test to determine whether an arbitral body constitutes a private body or a state-sponsored body for the purposes of § 1782. There can be no doubt that a Bermuda Form tribunal would be regarded as a private body. See generally J Fellas, ‘Using Section 1782 in International Arbitration’ (2007) 23 Arbitration International 379; L Schaner and B Scarborough, ‘US Discovery in Aid of International Arbitration and Litigation: The Expanded Role of 28 USC § 1782’ (2008); Austrian Arbitration Yearbook 299; Redfern and Hunter on International Arbitration (above n 3) paras 7.39–7.44. 164 For the inherent jurisdiction to issue letters of request where there is a lacuna in the procedural rules, see Panayiotou v Sony Music Entertainment (UK) Ltd [1994] Ch 142. 165 Silver Dry Bulk Co v Homer Hulbert Maritime Co Ltd [2017] EWHC 44 (Comm), [2017] 1 All ER (Comm) 791, [44] (Males J). 166 Dtek Trading SA v Morozov [2017] EWHC 94 (Comm), [62] (Sara Cockerill QC).

378

Discovery, Privilege and Waiver of Privilege

request, whether for the attendance of witnesses or the production of specific documents, is also recognised by a number of commentators.167 16.100 However, such requests are procedurally cumbersome, although less so as a result of the decision of the Court of Appeal in A and B v C, D and E (Taking Evidence for a Foreign Seated Arbitration).168 In that case, which overruled earlier first instance authority, the Court of Appeal held that section 44 (2)(a) of the Arbitration Act enables the court to order that evidence can be taken from a witness who is not a party to the arbitration. For that purpose, it is permissible to join the proposed witness as a respondent to the application. Since an application under section 44 against a non-party witness is permissible, it follows that permission can, if necessary, be obtained to serve out of the jurisdiction pursuant to CPR 62.5(1)(b).169 An application under section 44 does, however, require (except in the case of urgency) notice to the other party or parties to the arbitration and either the permission of the tribunal or the agreement in writing of the other parties.170 Yet, even if the foreign court were prepared to answer such a request (either under the Hague Convention or as a matter of discretion), it is not likely that the English courts would be willing to make such a request for the purpose of obtaining oral discovery. English courts are themselves reluctant to order pre-trial depositions, even pursuant to a letter of request from a foreign court,171 and are unlikely to find it attractive to seek assistance they would not themselves be willing to give. For all these reasons, we do not believe that an English court would, even if it could, assist an arbitral tribunal in England to conduct compulsory depositions of non-party witnesses, whether in England or elsewhere. 16.101 It is possible that the parties or the tribunal might, under certain circumstances, be able to obtain the assistance of a foreign court directly to carry out depositions. Whether that is the case will depend on the applicable law of the court which is being asked to assist. Federal courts in the United States could, perhaps, provide that assistance under 28 USC § 1782, but the position is far from certain.172 Some courts have been willing to use powers under § 7 of the Federal Arbitration Act.173 There are, however, a number

167 S Cockerill, The Law and Practice of Compelled Evidence in Civil Proceedings (Oxford, OUP, 2011) paras 1.19, 5.40–5.41; D St J Sutton et al, Russell on Arbitration, 24th edn (London, Sweet & Maxwell, 2015) para 7-198, commenting that the power is ‘rarely invoked’; RM Merkin, Arbitration Law (Looseleaf, London, Informa, 2010 and Issue 87 (2021)) para 15.8, commenting that: ‘No mention is made of this possibility in the 1996 Act itself, but it was expressly referred to in the July 1995 Arbitration Bill: its exclusion from the 1996 Act was on the ground that the possibility was covered by s 44(2)(a) without having to be spelt out.’ 168 A and B v C, D and E (above n 159). 169 See the reasoning of Foxton J in A and B v C, D and E (Taking Evidence for a Foreign Seated Arbitration) [2020] EWHC 258 (Comm) [2020] Bus LR 426, paras [31]–[33]. This analysis remains correct, even though Foxton J’s decision was overruled on other grounds: A and B v C, D and E (above n 159). 170 Section 44(4). 171 Rio Tinto Corp v Westinghouse Corp [1978] AC 547, 608–10; Commerce and Industry Insurance Co of Canada v Lloyd’s Underwriters (above n 160) 1329. The application was, however, granted in A and B v C, D and E (above n 159). 172 See the cases cited above at n 163. 173 9 USC § 7. This section authorises arbitrators to ‘summon any person in writing to attend before them as a witness’ and empowers the US District Court for the district in which the arbitrators are sitting ‘to compel the attendance of such person or persons before said arbitrator or arbitrators, or punish said person or persons for contempt’ in the same way as a witness disobeying a subpoena could be punished.

Other Forms of Discovery

379

of reasons why this is not likely to be a satisfactory solution. The main problem is territorial: the provision seems inapt unless the arbitrators are ‘sitting’ in the United States. Even if the arbitrators, by moving the venue of a particular hearing,174 could arrange matters so that they ‘sit’ in the United States, a tribunal is unlikely to do this if a party, for good reason, objects. Many insurance companies using the Bermuda Form are anxious to avoid conduct in the United States, for jurisdictional reasons; that is one of the reasons to elect arbitration elsewhere. Those companies are likely to object to participating in proceedings in the United States, and a tribunal is unlikely to override those objections unless the circumstances are highly exceptional.175 In any case, it is doubtful whether the power under the Federal Arbitration Act to compel witnesses to attend ‘before [the arbitrators]’ extends to a power to order discovery.176 In summary, therefore, a Bermuda Form arbitration is very unlikely indeed to involve oral discovery, either from parties or, a fortiori, from non-parties. There may, moreover, be some difficulty in obtaining oral testimony at the hearing by compulsory process where the witness in question is not resident in the United Kingdom.

16.102

Documentary Discovery from Non-parties There are also difficulties in obtaining discovery or documentary evidence from non-parties. Where the non-party is resident in the United Kingdom, a witness summons 174 Without the agreement of the parties, they could not move the juridical seat, which is fixed by agreement in England. They can, however, decide to hold the hearings elsewhere: English Arbitration Act 1996, s 34(2)(a). Whether that would be sufficient to enable powers to be invoked under 9 USC § 7 is questionable. 175 This assumes that Article VI(N)’s stipulation for arbitration ‘in London, England’, which undoubtedly fixes the juridical seat of the arbitration, does not also fix, by agreement, a venue for the hearings, since such agreement would preclude the exercise of any discretion under s 34. The better view is probably that the arbitrators retain a discretion to fix a venue for hearings other than London, since Article VI(N)(2) permits the tribunal to fix ‘a reasonable time and place for the hearing’. It could be said, however, especially in those cases where the insurer is anxious to avoid any possible entanglement with the United States, that the ‘reasonable place’ must be a place in London. 176 There is conflicting authority on this point. Contrast, eg, Life Receivables Trust v Syndicate 102 at Lloyd’s of London 549 F3d 210 (2d Cir 2008) (applying New York law) (s 7 of the Federal Arbitration Act, by its clear text, confers limited power to compel production of documents only from a witness testifying in the arbitration) with Meadows Indemnity Co v Nutmeg Insurance Co 157 FRD 42 (MD Tenn 1994) (power to order evidence at a hearing implies a power to order discovery in advance). Federal appellate courts in the United States are split on the issue of whether s 7 authorises arbitrators to issue subpoenas for pre-hearing discovery from non-parties who do not testify. Compare Life Receivables 549 F2d at 212 (Second Circuit – precluding such discovery) with In the Matter of an Arbitration between Sec Life Ins Co of Am 228 F3d 856, 970–71 (8th Cir 2000) (Eighth Circuit – allowing ‘production of relevant documents for review by a party prior to the hearing’). If a US court were to permit, or be asked to permit, US witnesses to be deposed for the purposes of London arbitration proceedings, this might lead to applications to the English court by one party to enjoin the other party from deposing potential witnesses. An anti-suit injunction of this kind was granted in Omega Group Holdings Ltd and Others v Kozeny and Others [2002] CLC 132 (Mr Peter Gross QC, sitting as a Deputy High Court Judge) to restrain a party from deposing witnesses in the United States in relation to proceedings that were continuing in the English High Court. The injunction applied to witnesses who intended to give oral evidence at trial. An anti-suit injunction to restrain depositions was also granted in Benfield Holdings Ltd v Richardson [2007] EWHC 171 (QB), even though the depositions were in aid of proceedings in New York, where there was no good reason for seeking depositions in advance of a trial to take place in London. An anti-suit injunction was, however, refused in Dreymoor Fertilisers Overseas Pte Ltd v Eurochem Trading GmbH [2018] EWHC 2267 (Comm), [2018] 2 Lloyd’s Rep 536.

16.103

380

Discovery, Privilege and Waiver of Privilege

may be used to compel production of specific documents at the hearing.177 But the Arbitration Act 1996 contains no mechanism for the tribunal to order discovery, as such, from non-parties. The tribunal may try to encourage such disclosure – for example, by directing a party (or both parties) to make a request to a non-party.178 Such encouragement may be more effective in a case where the non-party is an associated company of the party to the arbitration (for example, the parent of a captive); for example, because a party and its associated company may wish to avoid the possibility of creating a bad impression with the tribunal if documents held by the associated company are not disclosed, at least without good reason. If a party seeks to invoke the powers of the English court in order to obtain documents from a third party, it will need to obtain the permission of the tribunal or the agreement in writing of the other parties. Even then, the English court will not order ‘discovery’ from a third party, but will consider compelling production only where documents have been identified with specificity.179 The English court cannot order pre-arbitration disclosure in the same way that it can order disclosure prior to the commencement of English litigation.180 16.104 Difficulties analogous to those that affect oral discovery will arise if the person with the relevant documents is outside the United Kingdom.181 Some US courts may be willing to make orders pursuant to 28 USC § 1782.182 If the US court is willing to entertain the application, it may wish to take into account the views of the tribunal as to the scope of production that should be ordered. If a very extensive request for production is made, this may have the potential to disrupt the arbitration. The tribunal may again be in a position to encourage disclosure of documents which it considers appropriate; for example, by indicating to the party seeking disclosure that it would not be prepared to delay the arbitration in order to accommodate a very wide-ranging application under § 1782, but at the same time indicating that the arbitration would accommodate a more limited application. Discovery by Interrogatory 16.105 Another form of discovery traditional in the common law is discovery by interrogatory, that is, by written questions being put to a party and answered in writing. In English High Court proceedings, this form of discovery has now been swallowed up by the concept of ‘further information’, which may relate either to evidential matters or to matters such as the clarification of a party’s pleaded case or argument.183 177 Arbitration Act 1996, s 43(1). 178 Article 3(9) and (10) of the IBA Rules contemplates that documents may be requested from third parties. 179 BNP Paribas v Deloitte and Touche LLP [2003] EWHC 2874 (Comm), [2004] 1 Lloyd’s Rep 233; Assimina Maritime Ltd v Pakistan Shipping Corp [2004] EWHC 3005 (Comm), [2005] 1 Lloyd’s Rep 525; Tajik Aluminium Plant v Hydro Aluminium [2005] EWCA Civ 1218 [2006] 1 WLR 767. On the distinction between a (legitimate) summons to produce particular documents and an (illegitimate) discovery exercise, see also In re Asbestos Insurance Coverage Cases [1985] 1 WLR 331, 337–38; Panayiotou v Sony Music Entertainment (UK) Ltd [1994] Ch 142, 153. 180 Edo Corporation v Ultra Electronics Ltd [2009] EWHC 682 [2009] 2 Lloyd’s Rep 349. 181 See above, paras 16.98–16.100. 182 See the cases cited above at n 163. 183 Civil Procedure Rules, Part 18.

Other Forms of Discovery

381

Arbitrators do have power to order the provision of information by a party; the power is a broad one to order ‘questions [to] be put to and answered by the respective parties’.184 Such powers are, however, normally exercised with rather a light touch. It is not unusual for questions to be put at an early stage in the proceedings with a view to clarifying either party’s case, but they do not normally go much further than that. Questions may, again, sometimes be put as an adjunct to the documentary disclosure process (for instance, to ascertain what has become of documents which probably once existed but have disappeared). Such questions, however, are not really designed, at least directly, to elicit evidence for use at the hearing. It is customarily thought better to wait for the exchange of witness statements which will normally deal with all relevant evidential matters. It would only be in rather unusual circumstances, and normally after exchange of witness statements, that more searching questions might be sanctioned; for instance, if no statement was being tendered from an important witness on a particular issue, so that it was likely to be difficult to investigate that issue properly at the hearing. Where the issues have been dealt with in the witness statements, a tribunal will not generally countenance the use of requests for information as a way of anticipating cross-examination or submission on that evidence.

184 English

Arbitration Act 1996, s 34(2)(e).

16.106

17 Interest and Costs INTEREST Issues Relating to Interest 17.01

If the policyholder’s claim for indemnity under the policy succeeds,1 questions will arise as to whether interest should be awarded upon the principal sum, the rate of interest, and the date from which interest should run. Since the principal sum covered by a Bermuda Form policy will generally be many millions of dollars, and a significant amount of time can be taken in the resolution of a contested claim, substantial sums of money can turn on the outcome of these issues relating to interest. In Bermuda Form arbitrations, there is usually no dispute that a successful claimant who obtains a monetary award should be entitled, in principle, to an award of interest to compensate for the non-payment by the losing party. Issues more commonly arise as to the rate of interest and the date from which it should run. There are, principally, two statutory provisions which are potentially relevant to issues concerning interest: section 49 of the English Arbitration Act 1996 and sections 5001–04 of the New York Civil Practice Law and Rules (CPLR). Section 49 of the English Arbitration Act 1996

17.02

The natural starting point for any discussion of interest is the English statutory regime applicable to interest, which is contained in section 49 of the English Arbitration Act 1996. This provides as follows: (1) The parties are free to agree on the powers of the tribunal as regards the award of interest. (2) Unless otherwise agreed by the parties the following provisions apply. (3) The tribunal may award simple or compound interest from such dates, at such rates and with such rests as it considers meets the justice of the case— (a) on the whole or part of any amount awarded by the tribunal, in respect of any period up to the date of the award;

1 Apart from claims for costs, claims for monetary sums in Bermuda Form arbitrations are generally made by the policyholder rather than the insurer. This chapter therefore assumes that the claim for interest is a claim advanced by the policyholder whose claim for indemnity has succeeded. The principles relating to interest discussed in this chapter would apply equally in the event of a claim by an insurer – for example, for repayment of moneys wrongly paid or indeed for costs.

Interest

383

(b) on the whole or part of any amount claimed in the arbitration and outstanding at the commencement of the arbitral proceedings but paid before the award was made, in respect of any period up to the date of payment. (4) The tribunal may award simple or compound interest from the date of the award (or any later date) until payment, at such rates and with such rests as it considers meets the justice of the case, on the outstanding amount of any award (including any award of interest under subsection (3) and any award as to costs). (5) References in this section to an amount awarded by the tribunal include an amount payable in consequence of a declaratory award by the tribunal. (6) The above provisions do not affect any other power of the tribunal to award interest.

Section 49(3) therefore confers a very general discretion upon a tribunal in relation to the award of interest. The approach taken by the English courts in relation to the statutory discretion to award interest under the Senior Courts Act 1981 is that interest is regarded as compensatory rather than punitive. Accordingly, the successful party is awarded interest at the rate which a natural or legal person, with the general attributes of the claimant, would have had to pay in order to borrow the amount recovered from the date when it should have been paid.2 In exercising its discretion, the court may take into account delay by the claimant in the prosecution of the claim.3 The practice of the English Commercial Court has, in the past, been to award interest at 1 per cent above the Base Rate, there being a rebuttable presumption that such is the cost of sterling borrowing for a substantial corporate borrower.4 However, there is no longer a presumption that this is the appropriate rate of interest, and the Commercial Court may well use some other reference point such as LIBOR.5 A similar approach to dollar claims has in the past led to a rebuttable presumption that US prime is the appropriate rate, being the rate charged to the most creditworthy customers.6 Again, however, this will not necessarily be the rate which is awarded.7

17.03

A Bermuda Form tribunal is, of course, under no obligation to follow English court practice in exercising its discretion as to the award of interest.8 Section 49(3) gives

17.04

2 Hugh Beale (ed), Chitty on Contracts, 33rd edn (London, Sweet & Maxwell, 2018) vol 1, para 26-285; Tate and Lyle Food & Distribution Ltd v Greater London Council [1982] 1 WLR 149; Kazakhstan Kagazy plc v Zhunus and Others [2018] EWHC 369 (Comm), [23]. 3 See Robert M Merkin, Arbitration Law (Looseleaf, London, Informa, 2010 and Issue 87 (2021)) para 18.65. 4 National Westminster Bank plc v Rabobank Nederland [2007] EWHC 1742 (Comm), [2008] 1 All ER (Comm) 243, [49]–[58]. 5 The Commercial Court Guide, 10th edn (2017), para J.14; Kazakhstan Kagazy v Zhunus (above n 2) [29]–[31]; Assetco plc v Grant Thornton UK LLP [2019] EWHC 592 (Comm), [2019] 1 Costs LR 197, [10]–[27]. See generally James Edelman, McGregor on Damages, 21st edn (London, Sweet & Maxwell, 2021), paras 19-109–19-121. LIBOR itself will be phased out by the end of 2021. 6 The US prime rate has been said to be the nearest equivalent of Base Rate plus 1 per cent: Kuwait Airways Corp v Kuwait Ins Co SAK [2000] 1 Lloyd’s Rep IR 678, 692–93; Hellenic Industrial Development Bank SA v Atkin [2002] EWHC 1405 (Comm), [2003] Lloyd’s Rep IR 365, [19]–[20]. See Edelman (above n 5) para 19-120. 7 Triple Point Technology Inc v PTT Public Co Ltd [2018] EWHC 45 (TCC), reviewing prior authorities (reversed in part on other grounds). 8 English legislation does provide for a very high rate of interest under the Late Payment of Commercial Debts (Interest) Act 1988. This Act would normally be inapplicable to claims made in a Bermuda Form arbitration: a claim by a policyholder would not be a ‘qualifying debt’ under the statute, and in any event there would not a sufficient connection with England to make the statute applicable; see Martrade Shipping & Transport GmbH v United Enterprises Corp [2014] EWHC 1884 (Comm), [2015] 1 WLR 1. Section 13A of the Insurance Act 2015 provides a remedy for late payment of an insurance claim, but this provision would not apply to a contract governed by New York law.

384

Interest and Costs

an arbitration tribunal a very broad discretion in relation to interest, including as to the rates to be applied.9 Nevertheless, our experience is that Bermuda Form tribunals are often attracted to the compensatory approach applied in English court practice.10 This in turns leads to the selection of the US prime rate as the applicable rate, at least as the starting point. If this approach is taken, then the decision as to whether to award simple or compound interest is also, pursuant to section 49(3), a matter for the discretion of the tribunal. If a tribunal is to award compound interest on prime rates, it may require evidence that companies in the position of the claimant would normally have to pay compound interest on its borrowings. The tribunal may not wish to award compound interest for some other reason – for example, because the policyholder delayed in pursuing its rights or because of some other conduct which made it reasonable for the insurer to defend the claim. 17.05

The foregoing discussion assumes that the question of interest is simply a matter for the tribunal’s discretion pursuant to section 49(3). This, however, is a point which is controversial. An issue which has come into focus in recent years is whether Bermuda Form tribunals should award interest at the 9 per cent (simple interest) rate which is mandated by section 5004 of the New York CPLR. Since this point now arises in many arbitrations, this chapter addresses the issue in some detail. Sections 5001–04 of the New York CPLR

17.06

As stated in section 101, the CPLR governs ‘the procedure in civil judicial proceedings in all courts of the state and before all judges, except where the procedure is regulated by inconsistent statute’. Section 5001(a) CPLR provides that interest shall be recovered upon a sum awarded because of a breach of performance of a contract … except that in an action of an equitable nature, interest and the rate and date from which it shall be computed shall be in the court’s discretion.

Accordingly, there is a right to such interest, except in actions of an equitable nature where the court has a discretion. Under section 5001(b), interest ‘shall be computed from the earliest ascertainable date the cause of action existed’. Section 5001(c) provides for interest up to the date when the verdict was rendered or the report or decision was made. Section 5002 provides for interest from the verdict or decision or judgment until the date of entry of the final judgment. Section 5003 provides for interest from the date of entry of the judgment. Section 5004 provides for the rate of interest, namely ‘at the rate of nine per centum per annum, except where otherwise provided by statute’. 17.07

US prime rates have been substantially below the 9 per cent rate for some years. An award of interest at the 9 per cent simple interest rate may well be more advantageous 9 See Lesotho Highlands Development Authority v Impregilo SpA [2005] UKHL 43, [2006] 1 AC 221. The discretion would enable a tribunal to compensate, via its interest award, for a decline in the value of the currency of its award (per Lord Phillips at [43]). 10 The Departmental Advisory Committee (DAC) report on the Arbitration Bill (which became the English Arbitration Act 1996) envisaged that interest would be awarded on a compensatory rather than a punitive basis: see para 237 in the context of the power to award compound interest.

Interest

385

to a policyholder even when compared to compound interest at US prime. This has led policyholders to argue that 9 per cent should be awarded by tribunals, either as a matter of right or alternatively as a matter of discretion. There is no doubt that if section 49(3) of the English Arbitration Act 1996 applies, a tribunal has a discretion to award interest at 9 per cent. But a tribunal might be reluctant to do so, since this would result in the successful party receiving, at a time of low interest rates, a sum greater than that required to compensate it for the loss suffered by being out of its money for the period taken to resolve the claim. Hence, policyholders have argued that there is a substantive right, as a matter of New York law, to interest at the 9 per cent rate. It is argued that since the parties have chosen New York law as the governing law of the contract, the tribunal should give effect to the substantive right. In other words, it is argued that since the entitlement to interest (and the rate of interest) is a matter on which the parties have agreed, the award of interest should reflect the parties’ agreement and it is not a matter for the tribunal’s discretion under section 49(3). Thus, it is said that the discretionary power of the tribunal to award interest under section 49(3) of the English Arbitration Act 1996 can be displaced by the agreement of the parties; see section 49(2), which provides that the following provisions in section 49, including section 49(3), apply unless ‘otherwise agreed by the parties’. Section 49(3) is not a mandatory provision of the Act; see section 4 of the 1996 Act. The parties have agreed, by Article V(O) of the Bermuda Form, that the policy is governed and construed in accordance with the internal laws of the State of New York. Those internal laws of the State of New York include the law as set out in sections 5001–04 CPLR, which give a substantive right to interest. As a matter of New York law, the right to interest contained in the CPLR is regarded as a substantive right rather than simply a matter of procedure.

17.08

As discussed below, the US case law provides substantial support for the proposition that both the entitlement to interest and the 9 per cent rate of interest, under the CPLR, are substantive rights as a matter of New York law. A tribunal is certainly entitled to take this feature of New York law into account in exercising its discretion, and there can be no doubt that a tribunal has a discretion to award 9 per cent. However, it is one thing to say that there is a discretion to award interest at 9 per cent, but quite another to conclude that a tribunal is bound to do so.

17.09

We addressed this argument in the second edition of this book, where we indicated that it was not straightforward or compelling. It would lead to the result, which might seem odd, that a Bermuda Form tribunal was bound to award interest at 9 per cent despite the discretion contained in section 49(3), whereas a US arbitral tribunal would not be bound to award the 9 per cent rate if appointed pursuant to arbitration rules which gave a discretion.11 In view of the decision of the Supreme Court in Enka Insaat Ve Sanayi As v OOO Insurance Company Chubb discussed below,12 we consider that it can no longer

17.10

11 Coastal Caisson Corp v EE Cruz No 05 Civ 7462 (DLC), 2007 WL 2285936 (SDNY 2007) (where the American Arbitration Association (AAA) arbitration rules gave the arbitrators a discretion to award interest at such rate and from such date as the arbitrators deemed appropriate), vacated on other grounds, 346 F App’x 717 (2d Cir 2009). 12 Enka Insaat Ve Sanayi As v OOO Insurance Company Chubb [2020] UKSC 38, [2020] 1 WLR 4117.

386

Interest and Costs

realistically be argued that the tribunal is bound to award interest at 9 per cent on the basis that the discretionary power under section 49(3) has been displaced by the agreement of the parties in a Bermuda Form contract. Since that power has not been displaced, it remains open to the tribunal to exercise its discretion under section 49(3) as it considers meets the justice of the case. 17.11

The issue of displacement of the section 49(3) discretion had previously been addressed by the House of Lords (the predecessor of the Supreme Court) in Lesotho Highlands Development Authority v Impregilo SpA.13 The relevant contract was expressly governed by the law of Lesotho, and this was reiterated in the International Chamber of Commerce (ICC) Terms of Reference to which the parties had agreed.14 Under Lesotho law, interest could only be awarded if there was culpable default, and the amount of interest could not exceed the amount of the principal sum.15 The arbitrators had not applied these principles, but had instead approached interest on the basis that they had a discretion pursuant to section 49(3) of the Arbitration Act 1996. Lesotho Highlands then argued that the arbitrators had awarded interest in circumstances which were not permitted under Lesotho law.16 The award on this aspect was successfully challenged at first instance and in the Court of Appeal, but the House of Lords held that the award was not open to challenge. The challenge failed because the House of Lords held that the requirement of ‘substantial injustice’, which was a prerequisite to a challenge under section 68 of the Arbitration Act 1996, had not been fulfilled. Lord Steyn went on, however, to identify what he regarded as other formidable difficulties which confronted the challenge. He said that only an agreement in writing as defined by the 1996 Act could qualify as an agreement to the contrary under section 49. An agreement that law of Lesotho applied was not an agreement to the contrary in writing.17 Accordingly, the discretionary power under section 49(3) remained and was available to the tribunal.

17.12

The reasoning of Lord Steyn was subjected to close analysis in Enka. That case was not concerned with the power of a tribunal to award interest, but with more general questions as to the applicable law of an arbitration agreement.18 However, the arguments advanced were wide-ranging. These included submissions as to the correctness of Lord Steyn’s dictum in Lesotho and the impact of section 4(5) of the English Arbitration Act 1996, which had not been referred to in Lord Steyn’s judgment in Lesotho. Section 4(5) reads as follows: The choice of a law other than the law of England and Wales or Northern Ireland as the applicable law in respect of a matter provided for by a non-mandatory provision of this Part is equivalent to an agreement making provision about that matter.

13 Lesotho Highlands (above n 9) reversing the Court of Appeal [2003] EWCA Civ 1159, [2003] 2 Lloyds Rep 497, and Morison J [2002] EWHC 2435 (Comm), [2003] 1 All ER (Comm) 22. 14 Lesotho Highlands (above n 9) [2005] UKHL 43, [8]. 15 See judgment of Morison J at ibid [25(7)(1)]. 16 See ibid at [14]. 17 Ibid at [37], [39]. 18 See above ch 3, paras 3.06–3.11.

Interest

387

In relation to the Lesotho case and interest, the majority19 of the Supreme Court said:

17.13

[89] We do not think it credible that Lord Steyn in the Lesotho case intended to endorse such an interpretation of section 4(5), and to do so without giving any reasons or even mentioning that provision of the Act at all. The likely reason why no reference was made to section 4(5) is that it was not relevant to the power to award interest. The Court of Appeal in the Lesotho case characterised the power to award interest under section 49(3) of the 1996 Act as discretionary and procedural – a characterisation which Lord Steyn seems to have endorsed when referring to the reasoning of the Court of Appeal in para 38 of his speech. The fact that section 49(3) was treated by both the Court of Appeal and the House of Lords in the Lesotho case as procedural in nature was later relied on by the Court of Appeal in Maher v Groupama Grand Est [2009] EWCA Civ 1191; [2010] 1 WLR 1564, para 38, to support a similar characterisation of the power of a court to award interest under section 35A of the Senior Courts Act 1981. Because section 49(3) is procedural, the choice of the law of Lesotho to govern substantive contractual rights was not in respect of a matter provided for by section 49(3) and therefore did not engage section 4(5). As it was not in doubt that the curial law governing the arbitration process was English law, to disapply section 49(3) would accordingly have required a specific agreement (in writing), as Lord Steyn observed. Whether or not Lesotho law contained any equivalent procedural power was in these circumstances not relevant. Even if it did, the law of Lesotho concerning that matter could not amount to an agreement to the contrary. [90] This is, we think, how Lord Steyn’s dictum should be understood. But whether this was what was meant or not, we are satisfied that section 4(5) does not require a specific agreement to disapply a non-mandatory provision of the Act.

Accordingly, the Supreme Court considered that the power of a tribunal to award interest under section 49(3) was, from the perspective of English law, a procedural power. Under English conflict of laws rules, it is for English law to decide whether a particular foreign rule is ‘procedural in the English sense’.20 The balance of English authority prior to Enka favoured the view that the rate of interest, in the context of a claim for damages for nonpayment of a debt, is a matter of procedure and is therefore governed by English law.21 This has now been confirmed by the Supreme Court in Enka.

17.14

Since the power was a procedural and since there was no doubt that the arbitration procedure was governed in the Lesotho case by English law (ie, the curial law was English law), there would need to be a specific agreement in writing to displace the power in section 49(3). An ordinary agreement that the law of Lesotho governed the contract (ie, the parties’ substantive rights under the contract) would not be sufficient. Nor was it relevant to enquire whether Lesotho law contained a power equivalent to that under section 49(3).

17.15

This reasoning is, in our view, directly applicable to a claim for interest under the Bermuda Form. There can be no doubt that English law governs the arbitration procedure in a Bermuda Form arbitration. Indeed, the concluding words of Article V(O) of the Bermuda Form specifically provide that ‘as respects arbitration procedure pursuant

17.16

19 Lord Burrows and Lord Sales dissented on the principal issue in Enka, but did not comment specifically on Lesotho. They did, however, agree with the approach of the majority as to the effect of s 4(5) of the Act; see Lord Burrows at [249]. 20 Lord Collins et al (eds), Dicey, Morris and Collins: The Conflict of Laws, 15th edn (London, Sweet & Maxwell, 2012), para 7-002 and see generally paras 7-001–7-004 (hereinafter Dicey, Morris and Collins). 21 Ibid, Rule 20(2) and (3), and commentary at paras 7-098–7-106.

388

Interest and Costs

to Condition N, the internal laws of England and Wales shall apply’. In those circumstances, the procedural power under section 49(3) would require a specific agreement in order to be displaced. It would not be displaced simply by the parties’ agreement that (modified) New York law was the governing law of the Bermuda Form; ie, that it is the law that governs their substantive contractual rights. It follows that, in Bermuda Form arbitrations, the tribunal’s discretion under section 49(3) remains fully effective and is not displaced by any provisions of New York law concerning the 9 per cent interest rate. 17.17

The remaining question is whether a tribunal should, as a matter of discretion under section 49(3), award interest at the 9 per cent rate. One of the arguments in favour of the tribunal so exercising its discretion is that the New York CPLR does indeed create – as a matter of New York law – a substantive right (not merely a procedural right) to interest at 9 per cent. When cases are before the New York courts, it is usually unnecessary for the court to consider whether it is applying interest under the CPLR as a substantive rather than a procedural right. Nevertheless, the US case law, in a number of different contexts, does indicate that the right to interest and the 9 per cent rate are, under New York law, substantive rather than procedural rights.

17.18

One such context is the case law where New York courts have been determining, by the application of New York conflicts principles, rights which are governed by laws other than New York law. If the CPLR interest provisions were procedural, then one might expect the New York courts to apply the CPLR to such cases. Indeed, the CPLR is expressed to apply to ‘civil proceedings in all courts of the state’.22 However, in cases where the New York courts apply the law of another state (for example, because it is the proper law of the contract), courts have held that pre-judgment interest is determined by the rule of the jurisdiction whose law determines liability.23 In such cases, New York courts applying New York conflicts rules therefore do not apply the interest provisions of the CPLR. This suggests that the CPLR provisions are matters of substantive New York law rather than New York procedure.

17.19

In Coastal Power International Ltd v Transcontinental Capital Corp,24 the federal court gave two reasons for applying the CPLR interest provisions. One reason was that the contract contained a New York governing law clause. This suggests that the relevant rights are substantive.

17.20

This analysis and conclusion is supported by the decision of the federal court in Granite Ridge Energy LLC v Allianz Global Risk US Insurance Co,25 where there is a 22 NY CPLR s 101. 23 Entron Inc v Affiliated FM Ins Co 749 F2d 127, 131–32 (2d Cir 1984); Patch v Stanley Works 448 F2d 483, 494 fn 18 (2d Cir 1971). 24 Coastal Power Int’l Ltd v Transcontinental Cap Corp 10 F Supp 2d 345, 371 (SDNY 1998). See also Commonwealth Associates v Letsos 40 F Supp 2d 170, 177 fn 42 (SDNY 1999); In re Cassandra Group 338 BR 583, 599–600 (Bankr SDNY 2006) (applying the CPLR where avoidance of a transfer was ‘predicated on New York substantive law’ – the court also relied on Second Circuit case law allowing recovery of pre-judgment interest for the nature of the tortious conduct at issue); Granite Ridge Energy LLC v Allianz Global Risk US Ins Co 979 F Supp 2d 385, 392–93 (SDNY 2013); Nature’s Plus Nordic A/S v Natural Organics, Inc 108 F Supp 3d 52, 55–56 (EDNY 2015), aff ’d, 646 F App’x 25, at *31 (2d Cir 15 Apr 2016) (summary order; not selected for publication) (holding an award of pre-judgment interest to be ‘mandatory’ under the facts of the case, citing s 5001 of the New York CPLR, among other authorities). 25 Granite Ridge Energy LLC v Allianz Global Risk US Ins Co (above n 24).

Interest

389

full discussion of the relevant principles. That case concerned a claim made under insurance policies which were expressly governed by New York law. The policyholder argued that interest should be awarded based upon New York substantive law. The insurer contended that New York law only governed the interpretation of the policy terms, not which law would govern an award of interest; accordingly, the law of New Hampshire should be applied, because that dispute had its centre of gravity and most significant contacts there. The court accepted the policyholder’s argument and applied New York law, which the parties had expressly chosen. Its choice of law analysis was based upon the principle, contained in the Restatement (Second) of Conflict of Laws,26 that where parties have validly chosen a state’s law to govern their contractual rights and duties, the same state’s substantive law will govern the measure of recovery for breach of contract. Accordingly, the court held that ‘New York substantive law is applicable to the determination of pre-judgment interest because it is the same as the law the parties chose to govern the interpretation of the insurance policies’. The other reason given in Coastal Power was that the court was sitting in a diversity action. In diversity actions, the right to pre-judgment interest and the rate of interest under the CPLR are regarded as substantive, and are often described as such. Federal courts in diversity actions apply the relevant sections of the CPLR because they apply the law of the state that would be controlling in an action upon the same claim by the same parties in a state court.27 The federal interest rate is therefore not applied in relation to pre-judgment interest. However, it can be argued that the approach of the federal courts applying New York law in diversity actions does not greatly assist on the question of whether the CPLR right to interest at 9 per cent creates a substantive right as a matter of New York law. Federal courts apply the CPLR, as New York law, in relation to pre-judgment interest, but not in relation to post-judgment interest.28 Furthermore, in US federal practice, a federal court has to decide whether a relevant state law is ‘substantive’ for the purposes of Erie principles; that is, the question of whether or not to apply a relevant state law.29 This depends upon whether the state law applicable to the issue or issues of the suit would significantly affect the outcome of the suit.30 The issue of whether a law is ‘substantive’ in that context is different from the question of whether a law is procedural or substantive for conflict of laws purposes.31 26 American Law Institute, Restatement (Second) of Conflict of Laws (St Paul, MN, American Law Institute Publishers, 1971 & Supps) ss 187, 207 cmt e. 27 Commonwealth Associates (above n 24); Oy Saimaa Lines Logistics Ltd v Mozaica-NY Inc 193 FRD 87 (EDNY 2000); Wechsler v Hunt Health Sys Ltd 330 F Supp 2d 383 (SDNY 2004); Sony Fin Servs LLC v Multi Video Group Ltd No 03 Civ 1730 LAK GWG, 2005 WL 91310, at *27 (SDNY 2005); Northrop Corp v Triad Int’l M arketing SA No 03 Civ 1730 LAKGWG, 842 F2d 1154 (9th Cir 1995); Nature’s Plus Nordic A/S v Natural Organics Inc (above n 24) at 55–56. Compare Oy Saimaa with North River Ins Co v ACE Am Reins Co 361 F3d 134, 143–45 (2d Cir 2004) (applying s 5001(a) of the New York CPLR) (pre-judgment interest on a partial award was recoverable only if the sum was included in the final judgment or the party had reserved its rights on the issue). 28 Northrop v Triad (above n 27); Commonwealth Associates (above n 24). 29 See further Appendix, para A.15. The ‘Erie doctrine’ specifically applies to federal cases in the United States whose jurisdiction is based on complete diversity of citizenship between all plaintiffs on the one hand and all defendants on the other. 30 Guaranty Trust Co of NY v York 326 US 99 (1995). 31 For the different meanings of ‘substantive’ in the context of the Erie doctrine, and for the purposes of conflict of laws, see, eg, Maryland Cas Co v Williams 377 F3d 389, 393 and fn 10; Yohannon v Keene Corp 924 F2d 1255, 1264–69 (3d Cir 1991).

17.21

390

Interest and Costs

17.22

There are also some US cases in the context of the interest rate that can be awarded by arbitrators.32 It can be argued that these cases provide some support for the proposition that arbitrators are not bound to apply the New York CPLR interest provisions and that, accordingly, those provisions are not part of the substantive law of New York. However, these cases do not appear to contain any detailed analysis of the characterisation of the relevant right under New York law.

17.23

The discretion under section 49(3) ultimately depends, under the terms of that section, upon the tribunal’s view as to what is required in order to meet ‘the justice of the case’. It may well be that a tribunal will not derive much assistance, in approaching the exercise of that discretion, from an analysis of the legal classification of section 5004 as a matter of either New York law or English law. In short, some tribunals may take the view that since the substantive law of the contract is New York, it is appropriate to look to New York law as to the approach to interest,33 and to do so irrespective of the correct classification of that law (procedural or substantive) as a matter of either New York or English law. Other tribunals may take the view that since the seat of the arbitration is in England, the compensatory approach to interest applied by the English courts should be applied by analogy. In our view, neither approach can be criticised. Commencement Date for Award of Interest

17.24

If, as suggested above, section 49(3) of the English Arbitration Act is applicable to a Bermuda Form policy, then the tribunal has a discretion as to the date from which interest should run. In relation to claims by the policyholder for indemnity under the policy, Article VI(F), the Loss Payable Condition, provides that losses are due and payable 30 days ‘after they are respectively paid by the Insured, demanded and proven in conformity with the Policy’. This clause, which is discussed in detail in Chapter 11, is critical in relation to the determination of when the policyholder’s cause of action arises, and therefore when the insurer is in breach by failing to pay.34 This in turn will impact upon the date from which a tribunal will determine that interest should run. Once each of the various requirements of the Loss Payable Condition has been satisfied, the insurer is in breach and interest should in principle start to run.

17.25

English court practice in relation to the award of interest sometimes results in the postponement of the commencement date to the time when a reasonable investigation of the claim ought to have been completed by the insurer.35 This ground is largely covered by the requirement in Article VI(F) that the losses shall be ‘proven in conformity with the Policy’. This means that the insurer’s obligation to pay only arises when the policyholder

32 Shamah v Schweiger 21 F Supp 2d 208 (EDNY 1998); Coastal Caisson v EE Cruz/NAB 2007 WL 2285936. 33 GB Born, International Commercial Arbitration, 3rd edn (Dordrecht, Kluwer Law International, 2021), 3361–62: ‘arbitrators have in practice generally looked to the substantive law governing the parties’ underlying claims for standards regarding interest’. 34 See above ch 11, paras 11.40–11.52. 35 J Birds et al (eds) MacGillivray on Insurance Law, 18th edn (London, Sweet & Maxwell, 2018) para 21-076; Merkin (above n 3) para 18.65.

Costs

391

has provided satisfactory proof of its losses, and in practice this will require the policyholder to provide some documentation to the insurer as part of the latter’s investigation of the claim.36 Section 5001(b) of the New York CPLR provides for the running of interest ‘from the earliest ascertainable date the cause of action existed’. If it is correct that, under New York law, the right to interest pursuant to section 5001(b) is a substantive right,37 then there is a powerful case for saying that the tribunal, even if exercising a statutory discretion under section 49(3), should give effect to that right. The English conflicts rule is that the right to interest is a matter of substance which is determined by the proper law of the contract (New York law), in contrast to the rate of interest, which is a procedural matter and therefore governed by English law as the law of the forum.38 Interest to the Date of the Award and Beyond The tribunal should be asked to award interest both in respect of the period up to the date of the award and from the date of the award until payment. These two periods are addressed separately in section 49(3)(a) and 49(4) of the English Arbitration Act 1996. They are usually dealt with separately in the award itself, even if the same rate of interest is awarded as a matter of discretion. If the award does not provide for interest to run on the award itself, then the English court cannot fill that deficiency if and when judgment is entered in terms of the award.39 Accordingly, it is essential to ask the tribunal to address the question of interest from the date of the award.

17.26

COSTS Timing of the Tribunal’s Decisions as to Costs A Bermuda Form arbitration will typically involve one or more directions hearings prior to the substantive hearing at which the dispute between the parties is resolved. If ‘preliminary issues’ are ordered (that is, there is a ‘bifurcation’ of the hearing), then the substantive hearing will in effect be divided into two or more hearings. After any of these hearings, and certainly after the conclusion of the substantive hearing, one or other party may make an application for the tribunal to make orders in respect of the costs of the proceedings. There are essentially two categories of costs which will need to be allocated: (i) the arbitrators’ fees and expenses; and (ii) the legal or other costs of the parties.40

17.27

When procedural matters are resolved at directions hearings, it is typically the case that the tribunal will order the costs to be ‘costs in the arbitration’. The general principle in section 61 of the English Arbitration Act 1996 is that ‘costs should follow the event except

17.28

36 See

above the discussion of the Loss Payable Condition in ch 11, in particular paras 11.42–11.52. the discussion above in paras 17.17–17.22. 38 Dicey, Morris and Collins (above n 20), Rule 20 (2); and paras 17.13–17.14 above. 39 Walker v Rome [1999] 2 All ER (Comm) 961. 40 Arbitration Act 1996, s 59(1)(a) and (c). 37 See

392

Interest and Costs

where it appears to the tribunal that in the circumstances this is not appropriate in relation to the whole or part of the costs’. At the stage when procedural matters are being resolved, the relevant ‘event’ is likely to be viewed as the ultimate outcome of the case, which by definition is not known at the time of the directions hearings. Furthermore, directions hearings tend to involve a mixture of contested and uncontested matters, as well as providing an opportunity for the tribunal to shape the preparation of the case. This makes ‘costs in the arbitration’ the appropriate order in most situations. 17.29

The situation might possibly be viewed differently if a directions hearing were arranged in order to deal with a particular issue, the parties incurred substantial costs in relation to that issue, and there was a clear winner on that issue. In that situation, the tribunal might consider it appropriate to award the winning party the costs relating to the hearing on that issue. Even then, however, the tribunal might consider it appropriate to reserve its decision until the outcome of the dispute was known, so that the impact of the issue could be seen in the context of the case as a whole. The position is similar where ‘preliminary issues’ have been decided. If there is a clear winner, the tribunal may decide to make an order for costs in favour of that party, but equally it may wait to see how the case is ultimately resolved.

17.30

When the case proceeds to the substantive hearing and an award on the merits, a tribunal will typically not deal with questions of costs in the award itself, but instead will leave the allocation of costs for further argument after the parties have considered the result. However, a tribunal is not obliged to do this, and it is sensible to raise the matter with the tribunal prior to the conclusion of the substantive hearing, so as to ascertain the tribunal’s intended approach. The parties may wish to make a specific agreement that questions of costs should be reserved for later decision by the tribunal. If questions of costs are to be reserved for later decision, a tribunal may consider it useful for each party to provide, prior to the award, a summary of their respective costs. This may assist the tribunal in its later determinations of potential arguments by the paying party that the receiving party’s costs were unreasonably high. The Tribunal’s Discretion as to Costs

17.31

Versions of the Bermuda Form prior to the 004 Form contained a provision that each party should bear its own costs of an arbitration. Section 60 of the English Arbitration Act 1996 provides that an agreement under which a party is to pay the whole or part of the costs of the arbitration in any event is only valid if made after the dispute in question has arisen.41 Accordingly, in the absence of such later agreement, the tribunal’s discretion to award costs will remain.

17.32

The governing principle under section 61(2) of the English Arbitration Act 1996 is that: Unless the parties otherwise agree, the tribunal shall award costs on the general principle that costs should follow the event except where it appears to the tribunal that in the circumstances this is not appropriate in relation to the whole or part of the costs.

41 See

above ch 11, paras 11.86–11.87.

Costs

393

The expression ‘costs should follow the event’ is essentially the same formulation as that contained in the English court rules which existed at the time of the English Arbitration Act 1996.42 The general principle in current English court practice is that the ‘unsuccessful’ party will be ordered to pay the costs of the ‘successful’ party, and for this purpose the successful party is identified by reference to the outcome of the litigation as a whole.43 This is essentially the same as the principle that costs should follow the event. The basic rule is therefore that the loser should pay, and accordingly it is necessary to identify the winning party. This is straightforward in a case where the claim under the policy has succeeded in full or has been dismissed. In a case where the policyholder has succeeded in recovering part of its claim, but not the full amount, the policyholder is still likely to be regarded as the winning party, unless the amount awarded is so small that the insurer can be regarded as the effective winner.44

17.33

Under section 61(2) of the English Arbitration Act 1996, the principle that costs follow the event is the starting point for the tribunal’s decision. The tribunal has a discretion to take a different approach if it considers it inappropriate to order that costs should follow the event.

17.34

It is not unusual – indeed, it is very common – for the overall winner to be unsuccessful on some of the issues raised in arbitration. This can lead to an argument that the successful party should be deprived of the costs relating to those issues and should pay the costs which the otherwise unsuccessful party incurred in relation to those issues. The approach taken in English court practice, which may well commend itself to a Bermuda Form tribunal, is that there is no automatic rule requiring reduction of a successful party’s costs if it loses on one or more issues. English courts have recognised that in complex litigation, any winning party is likely to fail on one or more issues in the case and that it is a fortunate litigant who wins on every point. In such circumstances, an English court may consider it inappropriate to make separate orders for costs in respect of points which the winning party has lost, unless those points were unreasonably taken.45 If, however, a very significant amount of time and expense is incurred in relation to a particular issue upon which the overall winner was unsuccessful, then this may well be reflected in the allocation of costs.46

17.35

If the tribunal considers it appropriate to make allowance for the costs incurred on issues on which the winning party has failed, it is likely to approach the matter on a broad basis – for example, by ordering that the successful party should recover only a percentage of its costs. The tribunal, and indeed the parties, may consider this more sensible than making an order which requires the costs of particular issues to be deducted from the winning party’s costs or paid by the winning party to the losing party, since this would involve a detailed enquiry at a subsequent stage into what costs were indeed incurred in relation to such issues.

17.36

42 Rules of the Supreme Court, Order 62 Rule 3. 43 Civil Procedure Rules 44.2(2)(a); HLB Kidsons v Lloyd’s Underwriters [2007] EWHC 2699 (Comm), [2008] 3 Costs LR 427, [10], summarising the authorities; Multiplex Constructions (UK) Ltd v Cleveland Bridge UK Ltd [2008] EWHC 2280 (TCC), [2009] 1 Costs LR 55, [72]. 44 See Merkin (above n 3) para 18.80 and, generally on the question of costs, paras 18.74–18.116. 45 HLB Kidsons (above n 43) [11]. 46 See, eg, Kastor Navigation Co Ltd v Axa Global Risks (UK) Ltd [2004] EWCA Civ 277 [2004] 2 Lloyd’s Rep 119.

394

17.37

Interest and Costs

The tribunal will also need to consider whether any percentage reduction in costs recoverable by the winning party should also be reflected in the allocation of the tribunal’s fees and expenses. The tribunal may, depending upon the circumstances, take the view that the entirety of its costs should be allocated to the losing party, even if that party has prevailed on some issues. Determination of the Quantum of Recoverable Costs

17.38

If a party has a costs order in its favour, then (subject to any order of the tribunal to the contrary) the determination of the quantum of its recovery pursuant to that order will be on the basis that it can recover a reasonable amount in respect of all costs reasonably incurred.47 The burden of proving reasonableness is on the receiving party, again subject to any order of the tribunal to the contrary. Thus, any doubt as to whether costs were reasonably incurred or were reasonable in amount is to be resolved in favour of the paying party.48 A tribunal is sometimes asked to order that this burden of proof should be reversed, thereby placing the onus on the paying party to demonstrate unreasonableness. In English court practice, this is known as an award of ‘indemnity’ costs, where it is very much the exception rather than the rule. The critical requirement for such an award of ‘indemnity costs’ is that there should be some conduct or circumstance that takes the case ‘out of the norm’.49 Such orders would be rare in a Bermuda Form arbitration.

17.39

In principle, the determination of the amount of recoverable costs is a matter for the tribunal.50 In theory, it would be open to a tribunal to appoint an expert in costs matters in order to assist with the determination.51 In practice, neither the parties nor the tribunal is likely to wish to do this, since it would involve an unnecessary increase in the costs of the arbitration. It is now common for the parties to agree that the question of the amount of recoverable costs should be delegated to one member of the panel, usually the chairman.

17.40

The tribunal has a discretion to order that interest should be awarded on the costs incurred by a party which are the subject of the award, both in respect of the period prior to and subsequent to the award.52

17.41

As far as the tribunal’s own fees and expenses are concerned, the English Arbitration Act 1996 contains procedures by which their reasonableness can be assessed.53 In practice, however, rates will have been agreed with the arbitrators at the outset of the arbitration, and it is most improbable that any dispute would arise.

47 Arbitration Act 1996, s 63(5)(a). 48 Ibid, s 63(5)(b). 49 For the principles in English court practice, see National Westminster Bank plc v Rabobank Nederland [2007] EWHC 1742 (Comm), [2008] 1 All ER (Comm) 243; Hosking v Apax Partners LLP [2018] EWHC 2732 (Ch), [2019] 1 WLR 3347. 50 Arbitration Act 1996, s 63(3). 51 Ibid, s 37. 52 Ibid, s 49(3)(a) and (4). 53 Ibid, s 64.

Appendix ENGLISH AND NEW YORK LAW: SOURCES AND LEGAL APPROACH Because the Bermuda Form incorporates elements of New York law and elements of English law, questions under both systems of law are likely to arise in the course of any arbitration. It is common practice to have multinational tribunals (usually a mix of English and American lawyers) and multinational teams of counsel representing the parties (again, usually a mix of English and American lawyers). Communication is smoother if each can take a certain amount of very basic knowledge about the other system for granted. Moreover, it is only human for arbitrators to look at issues, albeit governed by an unfamiliar system of law, from the perspective of their own legal background; this may influence their approach, perhaps unconsciously. It is therefore sometimes useful to consider, if only as a matter of background, how a particular issue of, say, insurance law would be approached in England, even though New York law will apply.

A.01

This Appendix does not attempt to discuss in any detail the particular issues of procedure and substance which arise under the Form; they are dealt with in the body of this book. This Appendix instead aims to provide a much more basic introduction to the materials and fundamental approaches to legal argument in England and New York – the sort of knowledge that any detailed discussion of an issue tends to take for granted.

A.02

ENGLISH LAW FOR NEW YORK (AND OTHER UNITED STATES) LAWYERS The United Kingdom has three separate legal systems: the law of England and Wales,1 the law of Scotland, and the law of Northern Ireland. One should therefore avoid saying, in connection with any civil case, ‘UK law’, for there is not – as a general matter – any such thing.2 Scotland, for example, retains a distinct system of courts, a distinct legal profession and distinct legal principles applicable to matters such as contractual

1 Since Welsh devolution, it is arguable that the systems in England and Wales should be treated separately, but historically and for most practical purposes they remain the same. 2 Of course, the United Kingdom has a system of constitutional law that could properly be described as ‘UK’ law, and some aspects of the law are identical in England and Scotland, but in relation to civil obligations, the expression should be avoided.

A.03

396

Appendix

obligations and arbitration.3 What follows, therefore, is a description of the law of England and Wales (called, for the sake of brevity, English law), which is the law that will be applied in a Bermuda Form case, to the extent that the law of any part of the United Kingdom applies. The Profession and the Courts A.04

English courts are organised into three tiers. (i) Civil cases are decided initially by judges of first instance, sitting without a jury. Numerically, most cases are decided in county courts, but the judgments of these courts are generally neither reported nor cited as authority. More substantial cases are decided in the High Court, which is, for practical purposes, the lowest level of the English senior judiciary. It has a high standard and a good reputation. Judges tend to specialise, at least to some degree, in particular areas of law and are appointed from senior practitioners. Most cases involving insurance and arbitration are decided by judges of the Commercial Court, a ‘branch’ of the High Court staffed by specialists in commercial transactions. (ii) Appeal from decisions of the High Court lies, generally, with the Court of Appeal.4 Appeals are decided by (normally three) senior judges promoted from the ranks of the High Court judges. Appeals are screened to eliminate cases or arguments with no real prospect of success, and there may be further restrictions.5 (iii) From the Court of Appeal, appeal now lies to the Supreme Court, which on 1 October 2009 replaced the House of Lords as the final tier of appeal in the United Kingdom. Prior to October 2009, appeals from the Court of Appeal were heard by the House of Lords. Although formally an appeal to one of the houses of the legislature, these appeals were in practice heard and decided by appellate committees of five Lords of Appeal in Ordinary (colloquially ‘Law Lords’) who were promoted to their position from the Court of Appeal (or from prominent judicial positions in Scotland and Northern Ireland). Upon its creation in 2009, the new Justices of the Supreme Court essentially comprised the Law Lords who were then in office. The Supreme Court, like the US Supreme Court, screens potential appeals and chooses those it will hear largely on the basis of perceptions of their importance and interest.6

A.05

Practitioners in England have traditionally been divided into two separate professions: solicitors7 and barristers. The traditional division of functions was that solicitors dealt 3 In International Commercial Arbitration, Scotland applies the UNCITRAL Model Law: Law Reform (Miscellaneous Provisions) (Scotland) Act 1990. See generally Lord Hope of Craighead, ‘Arbitration’ in Stair Memorial Encyclopaedia, The Laws of Scotland (Reissue, Edinburgh, Law Society of Scotland, 1999). The English Arbitration Act 1996 does not extend to Scotland (s 108). 4 American lawyers should note the singular: Appeal not Appeals. 5 See generally Inco Europe Ltd v First Choice Distribution [2000] 1 WLR 586. 6 In order to bring a case before the Supreme Court, it is necessary to obtain permission to appeal. The Court of Appeal can grant permission to appeal from its own judgment, but usually leaves it to the Supreme Court to decide. As one appeal judge remarked in the days when appeals were heard by the House of Lords: ‘Their Lordships prefer to dine à la carte.’ Applications to the House of Lords for permission were known as ‘petitioning’ for ‘leave’ or ‘permission’ to appeal. The words ‘pet[ition] dis[missed]’ sometimes added to an English citation are equivalent to ‘cert[iorari] denied’ and (like that indication) tell one nothing about the correctness of the decision. 7 Before the mid-nineteenth century, there were both solicitors (who practised in chancery proceedings) and attorneys (who practised in the common law courts), but the former were more ‘respectable’ than the latter, and when the courts were fused, it was this title that survived.

English Law for New York (And Other United States) Lawyers

397

directly with lay clients, handled most transactional work, and dealt with the mechanics of litigation and all aspects of the preparation of a case for trial (such as discovery and the investigation of the facts). Barristers were always instructed by the solicitor (not directly by the client), had a legal monopoly on advocacy in court and professed special expertise in some preparatory steps (especially the drafting of formal pleadings). Many of the legal rules which underpinned this distinction have now been removed, but it is still generally true that barristers are specialised advocates, who take their instructions from another professional (usually a solicitor or a foreign lawyer). Some barristers are appointed as Queen’s Counsel,8 a mark of professional distinction. Barristers often work in ad hoc teams of ‘leading’ and ‘junior’9 counsel. Barristers have not, in the past, formed partnerships with other barristers;10 each barrister is a sole practitioner, though barristers tend to share office facilities and support staff in affiliations known as ‘chambers’.11 English legal rules do not restrict who may appear as a lawyer or advocate in an arbitration in England. Clients are free to use whatever combination of foreign and English lawyers they think best.12 If for any reason court proceedings take place (for instance, concerning the appointment of arbitrators), it is in practice necessary to instruct English solicitors to act in those proceedings, and common to instruct one or more barristers as well.

A.06

Sources of Law There are two primary sources of law in England, at least so far as the law of obligations is concerned: statutes and case law.13

A.07

Statutes are cited by reference to their short title and the year they were passed; for example, the Arbitration Act 1996. So far as the Bermuda Form is concerned, the main area of English law – the law governing arbitration – is primarily statutory, and the

A.08

8 Also referred to as ‘silks’ (after the gowns they are entitled to wear in court), ‘leading counsel’ or ‘leaders’. 9 ‘Junior’ in this context means only that the barrister has not been appointed Queen’s Counsel. 10 Professional rules of conduct have, in the past, forbidden it; Code of Conduct for the Bar, 8th edn (2004), para 205. These rules have now been revised. It is likely, however, that many barristers will wish to preserve their status as sole practitioners rather than forming partnerships with other barristers, despite the relaxation of the professional rules. 11 Because they are not partners, it is permissible (and, indeed, common practice) for different members of the same ‘set’ of chambers to appear on opposite sides in the same case, or for one member of the set to be an arbitrator and another to act as advocate. See Laker Airways Inc v FLS Aerospace Ltd [1999] 2 Lloyd’s Rep 45 (Rix J); and ch 14, para 14.47 above. These are reasons why many barristers will not wish to form partnerships with other barristers in their chambers. However, the approach of the English courts in Laker Airways is not without its critics in certain parts of what might be called the international arbitration community. The ability of one member of a set of chambers to appear as advocate before another as arbitrator has increasingly been called into question in some international arbitrations. 12 A barrister can accept instructions from a foreign lawyer in connection with an arbitration (though not in court litigation). 13 The European Union also produces law which, whether in the form of its foundational treaties, legislation, or case law, was – prior to the United Kingdom’s departure from the European Union – binding in England. But the field with which the Bermuda Form is concerned was, even prior the United Kingdom’s departure, little touched by this.

398

Appendix

English Arbitration Act 1996 comes close (if not going quite all the way) to codifying the relevant principles. In most areas of civil and commercial law, however, the statutes are not codifications, but exist to deal with particular points. Marine insurance law was largely codified in 1906;14 many, but not all, of the statutory provisions relating to marine insurance are regarded as succinct and accurate general statements of the law applicable to insurance in general.15 Certain important aspects of the 1906 Act and the common law (principally concerning non-disclosure of material facts) were significantly altered by the Insurance Act 2015. A.09

Although England does not have a written constitution or judicial review of statutes in the strict sense, some principles are recognised as being of constitutional significance. In particular, the Human Rights Act 1998 requires that all legislation (including, for instance, the English Arbitration Act 1996) should be interpreted if possible to be consistent with the human rights guaranteed by the European Convention on Human Rights. This could become important in ensuring, for example, that the procedures adopted by an arbitration tribunal do not infringe the rights of due process under Article 6 of the Convention, since the English Arbitration Act 1996 will also be subject to such interpretation. These principles are unaffected by the United Kingdom’s departure from the European Union.

A.10

Much of English law – whether it be decisions in areas governed exclusively by the common law or decisions applying and construing statutes – is therefore generated in case law, through the system of precedent.16 Judgments of first instance by judges in the High Court are not formally binding on any other judge, though in practice there is a strong tendency to follow them. Judgments of the Court of Appeal are strictly binding on all first-instance judges and on the Court of Appeal itself (though there are, of course, various ways in which precedents can be avoided or restricted, and a few circumstances in which they may be formally and directly ignored or overruled by the Court of Appeal). Judgments of the House of Lords and the Supreme Court are binding upon the Court of Appeal. The House of Lords could, but rarely did, depart from its own previous judgments.17 English courts accord very strong persuasive force to judgments of the Privy Council.18 They will consider judgments from other common law jurisdictions, though as guides or sources of inspiration rather than constraints.

A.11

English law has an active tradition of treatise and monograph writing by practitioners and first-rate scholars,19 and the leading scholarly monographs are often referred to in arguments and judgments (and they are often influential even where they have not been explicitly referred to). A brief guide to some of the leading works relevant to the Bermuda Form can be found in the Bibliography. 14 Marine Insurance Act 1906. 15 See Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co Ltd [1995] 1 AC 501, 518. 16 See generally Halsbury’s Laws of England, 5th edn (London, LexisNexis Butterworths, 2020) vol 11, paras 25–40; R Cross and J Harris, Precedent in English Law, 4th edn (Oxford, Clarendon Press, 1991) ch 3. 17 Practice Statement (Judicial Precedent) [1966] 1 WLR 1234. The Supreme Court will no doubt take a similar approach to its own decisions and those of the House of Lords; para 6.3.4 of UKSC Practice Direction 6 requires a party, in its written case to the Supreme Court, to state whether it is requesting the Supreme Court to depart from one of its own decisions or a decision of the House of Lords. 18 A court that hears appeals from a number of Commonwealth jurisdictions. 19 Writing or editing treatises ranks far higher on the priorities of (most) English legal academics than it does in the American legal academy.

New York Law for English Lawyers

399

NEW YORK LAW FOR ENGLISH LAWYERS The Court System In New York, as throughout the United States, there are two systems of courts. On the one hand are New York’s state courts, with three levels.20 At the lowest level for the present purposes are 11 trial courts with limited jurisdiction21 and the Supreme Court,22 which (despite its name) is New York’s trial court of general jurisdiction. The Supreme Court has state-wide jurisdiction without regard to the amount in controversy. Appeals from New York trial courts generally go to the Appellate Division, which sits in four departments.23 Further appeals go to the New York Court of Appeals, which is the highest court in New York State.24

A.12

Alongside this system is the federal court system. In that system the lowest court is the US District Court (of which, for the geographical area of New York State, there are four: the Southern District, the Northern District, the Western District and the Eastern District).25 As the district court dealing with cases in Manhattan, the Southern District naturally has great experience in commercial disputes. Appeals from the US District Courts go to the US Court of Appeals for the Second Circuit (which also hears appeals from federal district courts in the states of Vermont and Connecticut).26 In theory, further appeal would be to the US Supreme Court, if that court has mandatory jurisdiction or gives permission by granting a writ of certiorari.27 As with second-tier appeals in England, for most of the cases it decides, the US Supreme Court grants permission to appeal on the basis of the importance of the case from the perspective of the public in general, as set forth in the US Constitution and in federal statutes defining the court’s jurisdiction.

A.13

The federal courts are not restricted to dealing with questions of federal law; they also hear ordinary civil cases which meet the criteria for federal jurisdiction.28 In insurance cases, the federal courts’ basis for jurisdiction is commonly known as ‘diversity of

A.14

20 Some states in the United States have only two levels of courts: a court of general jurisdiction and an appellate court, often called the state’s Supreme Court. Other states, like New York, have three levels of courts (trial, intermediate appellate and high court). 21 NY Const art 6, §§ 1–17. 22 Ibid, § 7. 23 Appeals from the City Court of New York and the District Courts of Nassau and Suffolk Counties go to the Appellate Terms of the First and Second Departments; thereafter, appeals may be taken to the Appellate Division. 24 If a case before the New York Court of Appeals raises a point of federal statutory or constitutional law and is not subject to the court’s mandatory jurisdiction, a party might petition the US Supreme Court to hear the matter further (28 USC § 1257). However, the US Supreme Court is unlikely to grant a petition for writ of certiorari in cases concerning common law obligations, like insurance contract disputes, as such disputes with very limited exceptions defined by federal statute are considered matters of state law. In such cases the New York Court of Appeals as the state’s highest court has the final word. 25 28 USC § 112. 26 28 USC § 41. 27 28 USC § 1254. 28 28 USC § 1332. Those criteria require: (i) that the case involves more than US$75,000, exclusive of costs and interest; and (ii) that there is ‘diversity of citizenship’, that is, all of the plaintiffs are domiciled in a different state or states from all of the defendants. Corporations, for this purpose, are considered citizens of both the state of their incorporation and the state where their principal place of business is located. Ibid § 1332(c)(1). Diversity jurisdiction is not available, however, if any of the defendants reside in the state in which suit is brought; Wisconsin Dep’t of Corrections v Schacht 524 US 381, 388 (1988).

400

Appendix

citizenship’ jurisdiction or simply ‘diversity jurisdiction’. Because New York is a major commercial centre, many of its insurance coverage disputes (especially large commercial disputes) are dealt with in the federal courts. As a result, the opinions of the US Court of Appeals for the Second Circuit and the US District Courts in New York, especially in the Southern District (Manhattan and Westchester Counties) and the Eastern District (Brooklyn and Long Island), often deal with issues of New York insurance law, and those courts are recognised to have considerable expertise in the field. A.15

Although two systems of courts operate in New York, there are not two systems of common law. There is, in general, no ‘federal’ common law applicable to insurance coverage disputes where a federal court has jurisdiction by virtue of diversity of citizenship. Unless federal legislation or federal constitutional law applies, federal courts exercising diversity jurisdiction in such cases are required to apply the substantive law of the state in which they sit.29 This will normally be the case in insurance disputes. This naturally means that a federal court sitting in New York and deciding a case falling within its diversity jurisdiction will typically apply the substantive law of New York.30

A.16

When applying New York State law, federal courts will follow the statements of New York law made by the New York Court of Appeals, which is the ultimate arbiter of New York State law. Where the New York Court of Appeals has not spoken on an issue, federal courts often will attempt to predict how the New York Court of Appeals would decide the issue. The federal courts will not necessarily follow decisions made by the trial and lower appellate courts31 within the New York State system if they conclude that the New York Court of Appeals would reach a different result or a superior court in the federal system in New York has spoken authoritatively on the issue. The US Court of Appeals for the Second Circuit can, in an appropriate case, certify a question of New York law to the New York Court of Appeals, thereby obtaining an authoritative decision on the point.32 This is, however, unusual. Normally, federal courts decide questions that have been left open by the New York Court of Appeals, using their own legal skill and experience in New York law. Sources of Law

A.17

In practice the main sources of law relevant to issues arising under the Bermuda Form are New York statutes and relevant case law.33 New York does have a body of statutory law 29 As a result of the famous decision in Erie Railroad v Tompkins 304 US 64 (1938), called the ‘Erie Doctrine’. See also Green v Santa Fe Indus, Inc 576 F Supp 269 (SDNY 1983). 30 The substantive law of New York in such cases will include New York’s conflict of laws rules. Application of those rules in some cases will result in the application of the law of another state to contractual disputes, like insurance coverage disputes, as the lex causae. 31 The First and Second Departments of the New York Appellate Division, lower appellate courts in the New York system, are considered particularly persuasive in New York and in the United States generally. 32 See Rooney v Tyson 674 NYS2d 616 (NY 1998). State supreme courts in other states, if provided for by statute, can also certify an issue of New York law to the New York Court of Appeals. One important example is the Delaware Supreme Court’s certification of insurance issues to the New York Court of Appeals in In re Viking Pump 27 NY3d 244 (NY 2016) (answering question certified by Delaware Supreme Court, 146 A3d 1046 (Del 2015)). 33 Other potential sources of law in New York, such as the US and New York State Constitutions and federal legislation, are not in practice of much, if any, importance in this field of private law, at least insofar as substance is concerned. Such legislation (eg, the Federal Arbitration Act) may sometimes affect procedural issues, such as reference to arbitration.

New York Law for English Lawyers

401

dealing with insurance. Much is regulatory, but some provisions (such as those relating to misrepresentation and to warranty) do alter the common law rules, and are to be applied. Case law is, of course, produced by both the state and the federal systems. Within the state system, decisions by the state trial courts, called the supreme court, are of rather little value as precedent: they do not bind other judges and most are never reported. Appellate Division opinions are binding upon judges sitting within the Department concerned, but are not binding on judges in the other three Departments (and it is not unknown for an issue to be the subject of a division between the Departments until it is resolved by the Court of Appeals). Judgments of the Court of Appeals are binding on judges of all New York State courts, although the Court of Appeals may overrule its own prior jurisprudence.

A.18

Alongside this jurisprudence runs that created by the federal courts in New York, including the Second Circuit. Within that circuit, judgments of the federal district courts are not strictly binding, but are commonly relied upon as persuasive. Decisions by the Second Circuit are binding on all district courts within that Circuit34 and on the Second Circuit Court of Appeals itself.35 As explained above, federal courts are bound to follow any decision of the New York Court of Appeals on a point of New York law.36

A.19

There is, therefore, a multiplicity of different systems of stare decisis within New York: one for each of the US Circuit Courts and one for each of the appellate divisions of the New York State courts. The New York Court of Appeals sits (so far as New York State law is concerned) at the apex of all of the systems; therefore, if the New York Court of Appeals has not resolved an issue, it is not possible unequivocally to state what New York law is.37 Yet there is naturally a great deal of cross-fertilisation. In particular, federal district courts and the US Court of Appeals will naturally pay careful attention to, though they will not necessarily follow, relevant precedents from the intermediate appellate courts of New York State. In turn, the appellate courts of New York, including the Court of Appeals, are often influenced by federal decisions, especially those of the US Second Circuit,38 although they are not bound by them. Appointments to the federal bench are generally regarded as more prestigious than appointments to the lower echelons of the state bench (where judges are often elected), and federal court decisions have correspondingly higher status. It would certainly be wrong to assume that a decision of, say, the New York Supreme Court or the Appellate Division should necessarily be regarded as a more accurate or better statement of New York law than one of the Second Circuit or a federal district court in New York. 34 Decisions made by a Court of Appeals in one circuit do not bind those in other circuits, though they may be persuasive. As a result, conflicts of authority do sometimes emerge over issues of federal law; one function of the US Supreme Court is to resolve these conflicts. 35 Unless, rarely, overruled by a judgment of the Court of Appeals sitting en banc. 36 Occasionally, a federal court will not follow principles of state law, even sometimes those determined by the state’s highest court. Such decisions are subject to reversal and are easily distinguished. 37 American lawyers are much more used to conflicts in authority than English lawyers: ‘Federal case law below the level of the United States Supreme Court is often conflicting, and the same is true in lower state courts’. See P Atiyah and R Summers, Form and Substance in Anglo-American Law (Oxford, Clarendon Press, 1987) 66. 38 See, eg, Home Ins Co v American Ins Co 537 NYS2d 516, 517 (NY 1989) (citing Pan Am World Airways, Inc v Aetna Cas & Sur Co 505 F2d 989 (2d Cir 1974) (applying New York law)).

A.20

402

A.21

Appendix

Apart from New York decisions (whether federal or state decisions), courts may also take into account decisions from other jurisdictions within the United States or decisions from other common law countries (for instance, from England), particularly where local authority is lacking or ambiguous.

CULTURAL DIFFERENCES A.22

Although both American and English law are common law systems and the raw material on which they work (cases and statutes) is similar, our experience shows that the two systems differ significantly in terms of style and legal culture. Identifying these differences is inevitably somewhat a matter of impression, especially since lawyers within each tradition themselves have some diversity of approach. The following are our impressions.

A.23

In general, it is easier for English lawyers to take a more formal approach to legal reasoning and precedent than American lawyers.39 The English judiciary is smaller, more cohesive and (on the whole) faced with a more cohesive body of law than the American judiciary. Because the hierarchy of English precedent is clearer, there are fewer precedents to manage and there are no concerns about ‘sovereign’ courts in a federal system, so it is relatively easy for English judges to produce a coherent and consistent body of case law. English judges are correspondingly more likely to follow precedents by which they are not, strictly, bound,40 and to spend more time and effort analysing the language of previous decisions – often with long quotations41 – and reconciling and systematising them. One English commentator has said that the ‘text of earlier opinions is sometimes subjected to almost the same level of minute verbal analysis which English judges customarily adopt when interpreting statutes … American judges proceed differently’.42 Conversely, a common perception in England is that, on the whole, English judges resist arguments based on ‘public policy’, at least when they are overtly made, and are inclined to deprecate what they perceive to be excessive reliance by American courts on such arguments.43

A.24

The variety of American precedents from the various states and from the federal system presents American judges with a range of ways to resolve a particular problem. They feel correspondingly less inclined to undertake the (often impossible) task of synthesising all that has gone before, and more interested in the policy arguments and broad principles

39 See generally, Atiyah and Summers, Form and Substance in Anglo-American Law (above n 37). This is not to say that American law is always less formal than English law; in some areas (such as the law of evidence, where jury trials require rather strict application of rules of evidence), American law is more formal. See below n 42. 40 A judge at the trial court level should follow an earlier precedent, even if not strictly binding, ‘unless he is convinced that the judgment was wrong’: Police Authority for Huddersfield v Watson [1947] KB 842, 848. 41 For criticism of this tendency, see, eg, F Reynolds, ‘Reconsider the Contract Textbooks’ (2003) 119 Law Quarterly Review 177, 180: ‘There is a danger that the common law … may throttle itself by a mountain of paper.’ 42 Atiyah and Summers, Form and Substance in Anglo-American Law (above n 37) 125. 43 See, eg, the approach taken by the English Court of Appeal in Pilkington United Kingdom Ltd v CGU plc [2004] EWCA Civ 23, [2004] Lloyd’s Rep IR 891 in discussing American authorities on the meaning of ‘physical damage to tangible property’.

Cultural Differences

403

underlying existing decisions and relevant statutes or regulations (if any). Where they feel that an earlier decision neatly encapsulates a principle or rule in a particularly telling phrase, they may well quote that phrase, but the quotations are normally short, and they are presented rather as a summary of the rule that is to be applied than as material requiring exhaustive linguistic and critical analysis. It has been suggested that precedent has ‘lower mandatory finality in America than in England’.44 Another significant difference arises from the jury system. Since American trials, both civil and criminal, are usually conducted before a jury,45 American judges do not make detailed findings of fact in cases tried to a jury. In those insurance coverage cases tried by a jury, judges typically do not relate those findings to the law.46 On the whole, in those jury cases, the judge’s role is to arrive at fairly broad formulations of the applicable legal rules, to be applied in asking (for the purposes of summary judgment) whether an issue is worthy of trial or in directing a jury. Within this system, the need to frame rules in ordinary or lay terms, so that a jury can understand them, and the possibility of leaving some issues to the jury as broad questions of fact mean that some issues may be left to the jury in fairly wide terms. Oftentimes, judgments resulting from a jury trial are not memorialised in any written decision that may be cited in precedent in later cases. In contrast, English trial judges – who are their own fact finders – can use much more complicated rules, and English appellate courts, which often have a much more comprehensive record of the detailed facts found by the trial judge, can further elaborate these rules. Depending on the circumstances of the case, caseloads in American courts and other factors, the approach in our experience followed by a typical English judge may permit more intricate legal reasoning than the American system generally encourages, at least in relation to some issues.47 A third difference in culture arises out of the sort of case in which law is made. On the whole, in England, insurance and arbitration law is hammered out in cases involving commercial parties, often large organisations. Partly for that reason, there has historically appeared to be little need to accord special protection to the insured. There are, of course, smaller insurance disputes, but these are generally decided in county courts and therefore come into view as precedents only if there is an appeal. In contrast, although American courts naturally decide a good number of large commercial cases, many more ‘ordinary’ insurance disputes involving individuals produce reported decisions – as shown by the many decisions by the New York State courts on issues arising under health and life insurance policies.

44 Atiyah and Summers, Form and Substance in Anglo-American Law (above n 37) 126. 45 The Seventh Amendment to the US Constitution guarantees the right to a jury trial, and disputed factual issues are to be resolved by a jury or other trier of fact. Parties may agree to ‘bench trials’. 46 Insurance coverage cases are also tried in bench trials; eg, Sec & Exch Comm’n v Credit Bancorp, Ltd 147 F Supp 2d 238 (SDNY 2001). In those cases the American judgment naturally resembles more closely the English trial judgment, with a detailed explanation of how the law has been applied to the facts. 47 The difference in approach used by the typical English judge also means that much less attention is paid to rules of evidence in civil cases in England, since the fact-finding process is limited to legal professionals, and may be subject to closer supervision by appellate courts. Thus, in civil cases, rules of evidence (as such) have largely withered in England, being replaced by procedural rules about how evidence should be adduced to ensure fairness and principles about how evidence is to be assessed.

A.25

A.26

404

A.27

Appendix

Another difference is that England is disproportionately, through the London insurance market, an insurance exporter. Insofar as there may be a subconscious tendency to perceive the needs ‘of the market’ as the needs ‘of the London insurance market’, this may have contributed to the tendency – which is certainly marked – for English law to be remarkably tender to insurers, especially compared to American law, where the tendency is, or is perceived in England, to be rather the other way. In part because of the use of ‘boilerplate’ language and the executory nature of the insurance transaction, American courts have developed rules of policy interpretation that often refer to the insured as an economically weak consumer faced with a sophisticated and wealthy corporation, and that seek to avoid exploitation of consumers.48 English courts, by contrast, are more inclined to see the insurer as being at a marked disadvantage compared to the insured, who is generally perceived to be much better informed regarding the facts of the claim,49 and to fear exploitation of the insurer. English courts are correspondingly more inclined to impose or to uphold ‘protections’ designed to circumvent any possibility of careless or unscrupulous practices by the insured.

48 A fact on which English judges sometimes remark with disapproval; see, eg, Yorkshire Water Services Ltd v Sun Alliance & London Insurance plc [1997] 2 Lloyd’s Rep 21, 28. Naturally they regard their own approach as entirely neutral, but an American court (or policyholder) might well detect an opposite bias. 49 It is a remarkable fact that the only significant area where any positive duty on contracting parties to disclose any facts should have survived (and indeed expanded) during the nineteenth century is in insurance law, where its survival can only be explained on the basis that insurers are felt to require special protection; Carter v Boehm (1766) 3 Burr 1905. The rationale (systematic imbalance in information) is not limited to insurance contracts, but the rule has been. On the development of the rule, see R Hasson, ‘The Doctrine of Uberrima Fides in Insurance Law: A Critical Evaluation’ (1969) 32 Modern Law Review 615. The decision in Carter v Boehm and its progeny were eventually abrogated by the Insurance Act 2015.

Bibliography This bibliography lists a selection of standard works which may be found useful in researching points of English arbitration law, New York insurance law and English insurance law. American Law Institute, Restatement of the Law, Conflict of Laws (Second) (American Law Institute Publishers, 1971 and Supps). Compiled by the prestigious American Law Institute (ALI), composed of leading scholars, academics, judges, in-house counsel and practitioners throughout the United States. The Restatement on Conflict of Laws (Second) is considered to be an authoritative exposition of the law on conflicts and choice of law in the United States. The Foreword to the Restatement (Second) makes clear that, unlike other Restatements, such as the Restatement of the Law Contracts (Second), it supersedes the Restatement of the Law, Conflict of Laws (First), given the wide-ranging changes ‘jettisoning … a multiplicity of rigid rules in favour of standards of greater flexibility’ which had taken place since the time of the Restatement (First). The ALI drafted the Restatement (Second) over a period of more than 12 years, and circulated and revised three ‘official drafts’ before publishing this volume. The ALI is currently working on the Restatement of the Law, Conflicts of Law (Third), but there does not appear to be a firm timetable for completion. American Law Institute, Restatement of the Law, Contracts (Second) (American Law Institute Publishers, 1981 and Supps). Also compiled by the ALI, the Restatement of the Law, Contracts (Second) is considered to be an authoritative exposition of the law of contracts in the United States. Building on and refining the work of the influential Restatement of the Law, Contracts (First), the Restatement (Second) was begun in 1962 and completed in 1979 after 14 ‘tentative drafts’ were prepared and revised by the Reporters and Committee of Advisors of the ALI. American Law Institute, Restatement of the Law, Liability Insurance (American Law Institute Publishers, 2019). Also compiled by the ALI, this is the first Restatement on insurance. It covers the law of contracts in the context of liability insurance coverage and the management of insured liabilities. John A Appleman et al, Insurance Law and Practice (West Publishing Co, 1941; LexisNexis, 1998 and Supps). The original of this 22 (plus)-volume treatise on insurance law in the United States has been comprehensively updated by law professor and adviser to the Restatement Liability Insurance Jeffrey E Thomas and others under his editorship. It is published, among other titles, as New Appleman on Insurance Law Library Edition (LexisNexis Matthew Bender, 2009 and Supps). It remains a general resource on issues arising under personal lines, property casualty and other insurance policies. Unlike many insurance treatises, it includes discussions of taxation, regulatory and related issues pertaining to the organisation and regulation of insurance in the United States. John Birds et al, MacGillivray on Insurance Law, 14th edn (Sweet & Maxwell, 2018 and Supps). One of the standard works on English (non-marine) insurance law. Nigel Blackaby et al, Redfern and Hunter on International Arbitration, 6th edn (OUP, 2015). This resource provides a balanced account of the law and practice of international commercial arbitration in general.

406

Bibliography

Joseph Chitty, Chitty on Contracts (HG Beale (ed)), 33rd edn (Sweet & Maxwell, 2018 and Supps). The leading English professional treatise on general contract law. Malcolm A Clarke, Law of Insurance Contracts, 6th edn (Informa, 2009 and Supps). One of the standard works on English insurance law. It contains some (but by no means comprehensive) discussion of American case law and can provide a useful comparative view on some topics. Arthur L Corbin, Corbin on Contracts (Joseph M Perillo et al (eds)) (LexisNexis, 2013 and Supps). With Williston (below), one of the pre-eminent treatises on contract law in the United States. This multi-volume treatise is consulted and cited frequently by judges, professors and practitioners. A member of the prestigious American Law Institute, Corbin was Advisor on the influential Restatement of the Law of Contracts (First), Reporter on the Restatement of Remedies and Consultant to the Restatement on the Law of Contracts (Second) (above). Stephen A Cozen et al, Insuring Real Property (LexisNexis Matthew Bender, 1989 and Supps). A multi-volume looseleaf, updated annually. Stephen Cozen, the General Editor of this work, is a well-known insurance law practitioner in the United States who typically represents insurance companies. Insuring Real Property is a comprehensive treatment of the issues arising under first party property insurance policies. Mitchell F Dolin and Ethan M Posner, ‘Understanding the Bermuda Excess Liability Form’ (1998) 1(4) Journal of Insurance Coverage 68–84. An article that summarises issues that arise under various versions of the Bermuda Form. E Allen Farnsworth, Farnsworth on Contracts, 4th edn (Wolters Kluwer, 2020 and Supps). A useful treatise on contract law by a distinguished academic. English lawyers especially will find it helpful as its approach and layout are closer to an English textbook than the multi-volume treatises such as Corbin and Williston. Jack P Gibson, Maureen C McLendon, Richard J Scislowski, and W Jeffrey Woodward, Commercial Liability Insurance (International Risk Management Institute, digital access only at www.irmi. com). Includes commentary and legal case citations tied to specific provisions of the standard ‘comprehensive’, and later ‘commercial’, general liability (CGL) forms used in the United States since the 1950s. It is a particularly useful resource for its reproduction of the various versions of the CGL forms used in the United States over the years since the first widely circulated CGL form was issued in 1955. Charles Hollander, Documentary Evidence, 14th edn (Sweet & Maxwell, 2021). A specialised text that is often useful for its discussion of documentary evidence in English law, especially privilege. Eric M Holmes et al, Holmes’ Appleman on Insurance, 2nd edn (LexisNexis, 1996 and Supps). A multi-volume treatise on insurance law in the United States (comparable to Couch), addressing a multitude of issues under property and liability insurance. It can be a useful research tool, but its size tends to diminish its focus: it is useful as a quarry for general background on an issue of US insurance law, but of limited value as a guide to specific points of law. The original edition (by Appleman) remains sometimes useful. See also Appleman (above). Richard P Lewis and Nicholas M Insua, Business Income Insurance Disputes, 2nd edn (Wolters Kluwer, 2020 and Supps). This treatise covers the basics of first party property insurance and case law, and is frequently updated. It provides comprehensive coverage of the COVID-19 pandemic insurance litigation and coverage. Lorelie S Masters, ‘ACE and XL: A New ‘Batch’ of Coverage Issues’, Coverage 9(1) at 24 (Insurance Coverage Litigation Committee, Section of Litigation, American Bar Association, January/ February 1999). Written by a US practitioner, this is one of the first and only articles analysing issues arising under Forms 001–003 of the Bermuda Form. Lorelie S Masters and Jordan S Stanzler, Insurance Coverage Litigation, 2nd edn (Wolters Kluwer, 1999 and Supps). Deals with insurance coverage law throughout the United States (including New York) with an emphasis on commercial disputes and liability insurance.

Bibliography

407

Paul Matthews and Hodge M Malek, Disclosure, 5th edn (Sweet & Maxwell, 2017 and Supps). A specialised text relating to disclosure issues and useful for its discussion of privilege. Robert M Merkin, Arbitration Law (Looseleaf, Informa, 2010 and Supps). A comprehensive and generally up-to-date treatise on English arbitration law, but without the incisiveness of Mustill and Boyd (below). Robert M Merkin and Louis Flannery, Merkin and Flannery on the Arbitration Act 1996, 6th edn (Informa, 2020). This provides a useful section-by-section commentary on the English Arbitration Act. Susan Miller and Philip Lefebvre, Miller’s Standard Insurance Policies Annotated (Thomson Reuters, 1986 and Supps). A looseleaf, updated annually. It includes more than 232,800 annotations on cases organised and tied to specific provisions of the standard forms used in the United States. Its reproductions of the various versions of the standard CGL policy forms in use in the United States since the 1950s are particularly useful. Michael J Mustill and Stewart C Boyd, Commercial Arbitration, 2nd edn (Butterworths, 1989) and Commercial Arbitration: 2001 Companion (2001). This was the leading text on English arbitration law. There has been no new edition for some time, but a third edition is planned for 2021. The second edition does not take account of the English Arbitration Act 1996, but the 2001 Supplement deals specifically with it. Merkin, Arbitration Law offers more comprehensive coverage of cases, and although now somewhat out of date, the commentary by Lord Mustill and Stewart Boyd QC in the 2001 Supplement still carries weight. Barry R Ostrager and Thomas R Newman, Handbook on Insurance Coverage Disputes, 19th edn (Wolters Kluwer, 2019). Used by judges and practitioners throughout the United States, this treatise now spans two volumes and covers law insurance coverage issues throughout the United States, including New York. Sidney L Phipson, Phipson on Evidence, 19th edn (Hodge M Malek (ed)) (Sweet & Maxwell, 2018 and Supps). The standard English treatise on the law of evidence and privilege. Lee R Russ and Thomas F Segalla, Couch on Insurance, 3d edn (Thomson Reuters, 1995 and Supps). With Appleman (above), one of the most commonly used treatises on insurance law. Like Appleman, it can be very useful as a research quarry. Francis Russell, Russell on Arbitration, 25th edn (David St John Sutton, Judith Gill and Matthew Gearing (eds)) (Sweet & Maxwell, 2021). A text which was once the standard treatise on English arbitration law, but which was superseded by Mustill and Boyd, Commercial Arbitration and later works. The work has been entirely rewritten and is a well-regarded resource. David Scorey, Richard Geddes and Chris Harris, The Bermuda Form: Interpretation and Dispute Resolution of Excess Liability Insurance, 2nd edn (OUP, 2018). This comprehensive work contains a detailed discussion of the Bermuda Form by authors with considerable experience of acting for insurance companies in Bermuda Form and other cases. Since the views expressed by the present authors sometimes diverge from those expressed by Scorey, Geddes and Harris, the latter is essential reading on potentially controversial issues. The first edition was the joint winner, with the second edition of the present work, of the annual book prize awarded by the British Insurance Law Association. Sir Geoffrey Vos et al, Civil Procedure 2021 (Sweet & Maxwell, 2021). Called ‘The White Book’, this contains the English rules of civil procedure and a brief commentary on the, and is the standard work for English practitioners on any question of litigation procedure (including applications to court in relation to arbitration). Volume 2 contains the text of the 1996 Act and a brief commentary, as well as the procedural rules relating to proceedings concerning arbitration in the English High Court. Samuel Williston, A Treatise on the Law of Contracts, 4th edn (Richard A Lord et al (eds)) (Thomson Reuters, 1990 and Supps). One of the pre-eminent treatises on contract law in the

408

Bibliography

United States, often referred to as Williston on Contracts. This multi-volume treatise was first written by Samuel Williston, a member of ALI and Chief Reporter for the Restatement of the Law, Contracts (First), promulgated and published by ALI in 1932 for the first time. Allan D Windt, Insurance Claims and Disputes: Representation of Insurance Companies and Insureds, 6th edn (Thomson Reuters, 2013 and Supps). Another widely used resource on insurance law in the United States, English practitioners may find its one-volume format familiar and convenient. However, its treatment of the issues is less focused on issues of excess general liability coverage than other US insurance law treatises cited in big-case coverage litigation that has moved through US courts in the past two and a half decades and that is of most relevance to most of the issues arising under the Bermuda Form.

Index accrual of a cause of action 11.40–11.52 ACE Insurance Co Ltd, creation of 1.01, 1.06, 1.16–1.20, 9.32 advertising liability coverage clause 5.04, 5.12, 5.55, 5.65–5.66, 5.67 definition 5.65 exclusions 10.90 occurrence definition 6.06–6.14 aggregate limits 8.04, 9.35, 9.37–9.40, 11.97–11.101, 11.109 aggregation 6.23–6.46 causal links 6.33, 6.35 common causes 1.18, 6.27 disaggregation 6.36–6.46 effects of aggregation or disaggregation 6.37–6.46 integrated occurrences 1.32–1.33, 6.24–6.25, 6.27, 6.38–6.46 post-expiration injury 6.38–6.46 pre-inception injury 6.38–6.46 trigger of coverage, notice as 8.35, 8.40–8.41 aircraft exclusion 10.105 allegations of injury or damage 5.57, 6.11, 6.18–6.20, 7.04–7.05, 7.32–7.47, 7.56, 7.60–7.61, 7.66 allocation of payments 5.42–5.54 United States decisions 1.09–1.15, 5.48, 5.51–5.54 Annual Periods 1.25, 9.36–9.37 inception date 2.19 notice of occurrences 8.03, 8.08 period of cover 2.09–2.14, 2.19 Premium Condition 11.05–11.07 Reinstatement Condition 11.98–11.101 anti-suit injunctions 3.11 appeals Arbitration Condition 11.84 Appeals Condition 11.23–11.26 English law/practice 1.28, 11.84, A.04, A.10 errors of law 1.28, 11.84 law, points of 3.16 New York law/practice A.12–A.14, A.16, A.18, A.20 applicable law 3.04, 3.20–3.28 see also choice of law issues; governing law

appointment and selection of arbitrators 14.15–14.32, 14.35–14.49 barristers 14.36, 14.45, 14.47 chair 14.22–14.24, 14.45, 14.49 considerations for choice of arbitrator 14.45–14.48 default appointments 14.18 English retired judges 3.02, 14.12, 14.32, 14.36, 14.45 High Court, applications to 14.18, 14.24–14.32 impartiality 14.37, 14.38–14.44, 14.47 meetings before appointment 14.35 objection, potential grounds for 14.37, 14.38–14.44 overlapping appointments 14.40–14.42 Queen’s Counsel 14.36 solicitors 14.45, 14.47 tactics 14.36 unilateral communications 14.35 United States lawyers 14.30, 14.33, 14.36, 14.45–14.48 United States retired judges 14.46, 14.48 views, knowledge of arbitrators’ 14.40, 14.43, 14.48 arbitration agreements anti-suit injunctions 3.11 closely connected system of law, the most 3.09 confidentiality 15.24–15.25 enforcement of arbitration agreements 3.11 express choice of law 3.09 governing law 3.04, 3.06–3.11 implied choice of law 3.09 seat, choice of 3.09–3.10 separability of arbitration clauses 3.06, 3.09–3.10 termination of contract, agreements as surviving 3.06 arbitration awards appeals 1.28 challenges under Arbitration Act 1996 15.69–15.70 costs 17.30 errors, correction of 1.28, 15.70 jurisdictional challenges 15.70 reasons 15.69 serious irregularities 15.70

410

Index

Arbitration Condition 11.81–11.88 arbitrators see also appointment and selection of arbitrators fees and expenses 17.27, 17.37, 17.41 removal 3.16 asbestos asbestos-containing building materials (ACBM), costs of removal of 1.05 CGL policies 1.02 exposure/manifestation debate 1.07, 1.11, 5.59 injury-in-fact test 5.60 long-tail exposure 1.22 mass tort claims 1.05, 1.30 pollution exclusion 10.111 products liability exclusions 10.69, 10.76, 10.83 property damage 1.05, 5.59–5.60, 5.62, 5.64 toxic substances exclusion 10.102, 10.104 Assignment Condition 11.65–11.73 Assistance and Co-operation Condition 11.17–11.22 associated companies and affiliates 5.07–5.09, 11.111, 16.103 attachment to the policy 5.24, 11.29, 11.30–11.33, 11.44 aviation war-risk insurance 10.94–10.98 awards see arbitration awards AXA XL 1.20 barristers A.05–A.06 arbitrator, selection of 14.36, 14.45, 14.47 engaging English barristers 14.05–14.07, 14.12 Queen’s Counsel (QCs) 14.05–14.06, 14.36, A.05 solicitors, division of functions with A.05 basic structure of Bermuda Form 2.01–2.24 batch occurrences see integrated or batch occurrences Bermuda, arbitration in 1.01, 3.01, 3.03, 14.04, 15.01 bifurcation 15.31, 17.27, 17.29 bodily injury see personal injury burden of proof allocation of payments 5.50–5.51 delay 8.11–8.12 exclusions 10.07 expected or intended injury 7.14, 7.15–7.18 misrepresentation and non-disclosure 3.32, 12.04, 12.09, 12.14, 12.16, 12.23–12.27 business risk exclusions 10.62–10.64, 10.71 cancellation Cancellation Condition 11.74–11.79 Coverage A 11.74–11.79 Coverage B 11.77–11.79 Discovery Period Condition 11.102, 11.106 Named Insured, by 11.75–11.76

non-payment of premiums 11.79 notice 11.75–11.79, 11.112 period of cover 2.09–2.10 Policy Extension Condition 11.94, 11.96 pro rata premium, refund of 11.74, 11.76 captive insurers 9.19 care, custody or control etc exclusion 10.40–10.42, 10.53–10.57 case law see also precedent English law/practice A.07, A.10 New York law/practice A.17–A.18 causation aggregation 6.33, 6.35 concurrent causation exclusion 10.17–10.19 efficient or dominant cause 10.20–10.21 exclusions 10.01–10.02, 10.10, 10.11–10.27 expected or intended injury 7.33 occurrence definition 6.05, 6.09–6.11, 6.20–6.21 personal injury 5.17 property damage 5.17 CGL policies see Comprehensive General Liability (CGL) policies Changes/Non-waiver Condition 13.25–13.31 choice of law issues 3.01–3.46 express choice of law 3.01–3.03, 3.09, 3.18–3.20 implied choice of law 3.09 Chubb Corp 1.20 claims made insurance 1.23–1.26, 2.03, 5.03, 6.01, 6.12 climate change 10.117–10.118 commencing a Bermuda Form arbitration 14.01–14.49 appointment procedures 14.15–14.32 legal representation 14.02–14.14 notice to commence arbitration Appendix to chapter 14 selection of an arbitrator 14.35–14.49 common interest privilege 16.49–16.75 conflict of interests 16.71 co-operation 16.70 discretion 16.49–16.51, 16.58, 16.60 express or implied terms of Bermuda Form 16.52–16.55, 16.74 joint representation or joint defence 16.55–16.56, 16.59, 16.62 joint retainer/joint interests 16.62, 16.67 shield, common interest privilege used as a 16.62, 16.64 sword, common interest privilege used as a 16.62, 16.65–16.67, 16.71 third parties, disclosure to 16.63 companies associated companies and affiliates 5.07–5.09, 11.111, 16.103 mergers and acquisitions 5.10, 11.67–11.68

Index officers, directors and employees acting within scope of duties 5.09 subsidiaries 5.07–5.09, 9.17, 9.43, 11.111 Comprehensive Environmental Recovery, Compensation, and Liability Act (CERCLA) (US) 1.05 Comprehensive General Liability (CGL) policies allocation of payments 5.52–5.53 asbestos 1.02 continuous trigger 1.07–1.08 expected or intended injury 7.01, 7.09, 7.31 interpretation 1.16, 4.25, 4.30 standard terms 1.02, 1.10 conditions 11.01–11.114 Appeals Condition 11.23–11.26 Arbitration Condition 11.81–11.88 Assignment Condition 11.65–11.73 Assistance and Co-operation Condition 11.17–11.22 Cancellation Condition 11.74–11.79, 11.94, 11.96, 11.112 Changes Condition/Non-waiver Condition 11.64, 13.25–13.41 Conflicting Statutes Condition 11.89–11.90 Cross-Liability Condition 10.52, 11.12–11.15 Currency Condition 11.55, 11.80 Discovery Period Condition 9.37, 11.94, 11.102–11.109 Expiration Date Condition 11.110 Former subsidiaries, affiliates and associated companies Condition 11.111 Heading Condition 11.114 Inspection Condition 11.08–11.11 Law of Construction and Interpretation Condition 11.91 Liability of the Company Condition 11.93 Loss Payable Condition 11.27–11.55 Notice Condition 11.112–11.113 Notice of Occurrence Condition 8.02, 8.10–8.31, 11.16, 11.21, 11.112 Other Insurance Condition 11.58 Policy Extension Condition 11.94–11.96 Premium Condition 11.02–11.07 Proration of Losses Condition 11.92 Reinstatement Condition 11.94, 11.97–11.101 Representation Condition 11.56–11.57 Subrogation Condition 11.59–11.63 confidentiality 15.23–15.27, 16.93–16.94 arbitrator, selection of 14.40–14.41, 14.43 disclosure/discovery 15.24–15.27, 16.93–16.94 discretion 16.93 implied term in arbitration agreements 15.24 legal professional privilege 16.32 protective orders Appendix 2 to chapter 15 waiver of privilege 16.90–16.92

411

continuous policy, Bermuda Form as a 1.25, 12.39 contra proferentem rule 4.07, 4.27, 4.28–4.29, 4.31 ambiguities 4.22 applicable law 3.27 causation 10.10 exclusions 10.06–10.07, 10.10 contribution 9.28, 9.31, 9.33, 11.58, 11.87 co-operation Assistance and Co-operation Condition 11.17–11.22, 11.44 course and conduct of a Bermuda Form arbitration 15.04, 15.17, 15.20 directions 15.17, 15.20 failure to co-operate, waiver of defence of 11.21 costs 17.27–17.41 see also defence costs arbitrators’ fees and expenses 17.27, 17.37, 17.41 discretion of tribunal 17.31–17.37, 17.40 follow the event 17.28, 17.33–17.34 indemnity costs 17.38 percentage deductions 17.36–17.37 preliminary issues/bifurcation 17.27, 17.29 reasonably incurred costs 17.38–17.41 successful parties 17.29, 17.33–17.37 timing of decisions 17.27–17.30 course and conduct of a Bermuda Form arbitration in London 15.01–15.70 awards and post-award events 15.69–15.70 bifurcation 15.31 bundles 15.52–15.56 closing arguments 15.06 confidentiality 15.23–15.27, Appendix 2 to chapter 15 co-operation 15.04, 15.17, 15.20 cross-examination 15.07–15.08 cultural differences between English and US arbitration 15.01, 15.06–15.07, 15.31, 15.52–15.57, A.22–A.27 directions, first order for 15.16–15.22, 15.23, 15.25, Appendix 1 to chapter 15 disclosure/discovery 15.02, 15.04, 15.06–15.07, 15.21, 15.28–15.30, 15.51–15.56 electronic document databases 15.53 evidence 15.02, 15.04–15.08, 15.16, 15.52–15.53 exhibits 15.52–15.53 expert evidence 15.05–15.06, 15.42–15.50, 15.51, 15.56 foreign law as question of fact to be proved in evidence 3.14, 15.05 pleadings stage 15.09–15.15 preliminary issues or bifurcation 15.31 preparation for substantive hearing 15.51–15.58 procedure 15.02–15.04, 15.16, 15.18 remote hearings 15.20, 15.53 skeleton arguments 15.57

412

Index

statements of case, exchange of 15.06 substantive hearings 15.59–15.68 witness statements, exchange of 15.08, 15.32–15.41, 15.51, 15.56 written arguments, exchange of 15.57 written statements of claims and defence 15.02 Coverage A 1.26, 2.09–2.14, 5.01, 5.67 all occurrences reported in any period during Coverage A 2.15–2.16 Cancellation Condition 11.74–11.79 commencement 2.10 Coverage B 2.10–2.11 Discovery Period Condition 11.102–11.105 expiry 11.102–11.105 extension 2.09, 2.11, 2.14 mutual agreement, extension by 2.11 notice of occurrences 2.07, 8.03 period of cover 2.09–2.14 policy limits 9.34, 9.37 reinstatement 2.13, 11.100 renewal meetings 2.09 revised premiums or terms 2.11, 2.14 Coverage B 1.26, 2.09–2.15, 5.01, 5.67 all occurrences reported in any period during Coverage B 2.15–2.16 Cancellation Condition 11.74–11.79 Coverage A 2.10–2.11 Discovery Period Condition 11.102–11.103, 11.105–11.106 election to obtain 11.102–11.103, 11.105–11.106 extension or obtained at option of policyholder 2.11 fixed terms 2.11 notice of occurrences 2.07 period of cover 2.09–2.14 policy limits 9.34, 9.37 Reinstatement Condition 11.100 single aggregate limit 2.13 coverage clause 5.01–5.67 allocation of payments 5.42–5.54 definitions 5.02 Insured, definition of 5.07–5.10 judgments 5.19–5.23 liability, insurance against 5.03–5.06 mitigation of damages 5.41 personal injury 5.04, 5.12, 5.55, 5.56–5.57, 5.67 property damage 5.04, 5.12, 5.55, 5.58–5.64, 5.67 settlements by the policyholder 5.24–5.40 ultimate net loss/all sums insurance for damages 5.11–5.18 cross-examination 15.07–15.08, 15.35, 15.50, 15.55, 15.59, 15.62, 15.64–15.65 Cross-Liability Condition 10.52, 11.12–11.15

cultural differences between English and US arbitration A.22–A.27, 14.14, 15.01, 15.06–15.13, 15.31, 15.52–15.57 curial law 3.04, 3.10, 3.16–3.17, 14.30, 17.13, 17.15 currency 3.43, 11.55, 11.80 damages see also punitive damages allocation of payments 5.43–5.44 causal connection 5.17 consequential loss 5.16 defence costs 5.20–5.24 definition 5.04, 5.13, 5.10–5.24, 9.07 judgments 5.19, 9.06–9.09 settlements 5.24, 9.06–9.09 defence costs 5.20–5.24 allocation of payments 5.48, 5.50–5.51 excess point 9.04, 9.06, 9.18–9.19 demand, making a 11.29, 11.41–11.42, 11.48–11.49 depositions 15.06–15.07, 16.20, 16.96–16.102, 16.104 DES (diethylstilbestrol), miscarriages caused by 1.05, 1.30, 10.102 directions, first order for 15.16–15.22, Appendix 1 to chapter 15 directors and officers officers, directors and employees acting within scope of duties 5.09 disaggregation 6.36–6.46 disclaimers 13.03, 13.06, 13.16–13.18, 13.24 disclosure see disclosure/discovery; misrepresentation and non-disclosure disclosure/discovery 15.02, 15.04, 15.51–15.56, 16.01–16.106 associated companies, disclosure from 16.103 authenticity, challenges to 15.54 bundles 15.52–15.56 confidentiality 15.24–15.27, 16.93–16.94 continuing duty 16.22 co-operation 15.29, 16.15 depositions, taking 15.06–15.07 electronic documents 15.28, 16.18–16.19 foreign courts, assistance in obtaining evidence from 16.98–16.102 Hague Convention, letters of request under 16.98, 16.100 IBA Rules on the Taking of Evidence in International Commercial Arbitration 16.04–16.16, 16.21–16.22, 16.31 interrogatory, discovery by 16.105–16.106 late requests 15.30 legal professional privilege 3.46, 15.28, 16.07, 16.19, 16.21, 16.23–16.92 non-parties, documentary discovery from 16.103–16.104

Index objections 16.07 oral discovery (depositions) 16.96–16.102, 16.104 pilot scheme in Business and Property Courts of England and Wales 16.12–16.13 practice, in 16.15–16.22 proportionality 16.11, 16.17 Redfern schedule 16.16 Requests to Produce 16.04–16.10, 16.12 scope of discovery 16.07–16.08, 16.11–16.17 searches 16.19–16.20 standard disclosure 16.11–16.12 third parties 16.63 train of enquiry documents 16.11–16.12 witnesses outside UK, securing attendance of 16.98–16.102 discovery see disclosure/discovery diversity actions 17.21, A.14–A.15 employment Employee Benefits Programs 10.126–10.127 ERISA exclusion 10.126–10.127 officers, directors and employees acting within scope of duties 5.09 workers’ compensation 1.05, 10.35–10.36 endorsements 1.19, 2.24 English law/practice A.01–A.11 aggregation 6.33 appeals 1.28, 11.84, A.04, A.10 appointment of arbitrators 14.15, 14.18, 14.24–14.32, 14.35, 14.47 barristers A.05–A.06, 14.05–14.07, 14.12, 14.36, 14.45, 14.47 Business and Property Courts of England and Wales, pilot scheme on disclosure/ discovery in 16.12–16.13 case law 15.68, A.07, A.10 common interest privilege 16.49–16.50, 16.58–16.69 confidentiality 15.24, 16.93 costs 3.45, 17.35, 17.38 courts A.04–A.06, A.10 cultural differences with US A.22–A.27, 14.14, 15.01, 15.06–15.13, 15.31, 15.52–15.57 curial law, as 3.10, 14.30 disclosure/discovery 16.02–16.03, 16.05, 16.11–16.12, 16.18–16.20, 16.93, 16.98–16.101 estoppel 3.34 experts 15.42–15.43, 15.47–15.48, 15.50 interest 17.03–17.04 judiciary A.04 legal approach A.01–A.02 legal profession 14.03, 14.05–14.09, 14.12–14.14, A.04–A.06, A.12–A.21

413

legal professional privilege 3.46, 16.23–16.25, 16.29–16.43 disclosure/discovery 16.19 waiver 16.79, 16.81, 16.83–16.84, 16.91–16.92 New York lawyers, for A.03–A.11 Queen’s Counsel (QCs) 14.05–14.06, 14.36, A.05 remedies 3.40 retired judges 3.02, 14.12, 14.32, 14.36, 14.45 separate legal systems A.03 skeleton arguments 15.57 solicitors 14.06–14.07, 14.45, 14.47, A.05–A.06 sources A.01–A.02, A.07–A.11 statutes A.07–A.09 substantive hearings 15.61–15.68 Supreme Court A.04, A.10 treatises and monograph writing A.11 waiver 13.02 witnesses outside UK, securing attendance of 16.98–16.101 statements, exchange of 15.08, 15.35, 15.39 environmental claims see also pollution exclusion aggregation 1.31 causation 10.117 CERCLA 1.05 clean-up costs 7.09, 10.43–10.51 climate change 10.117–10.118 limited coverage 1.09 owned property exclusion 10.43–10.51 ERISA exclusion 10.126–10.127 estoppel 13.01–13.06, 13.19–13.24 abuse of process 3.34 affirmation 3.35 change of position 13.21 detrimental reliance 13.01, 13.19–13.20, 13.23, 13.28–13.31 disclaimers 13.24 election 3.35 issue estoppel 3.34 Loss Payable Condition 11.50 misrepresentation and non-disclosure 12.06, 13.23, 13.26 procedure and substance, division between 3.34–3.35 waiver 3.35, 12.06, 13.01–13.02, 13.20 words or conduct 13.20–13.23 evidence see also disclosure/discovery; expert evidence; witnesses course and conduct of arbitration 15.02, 15.04–15.08, 15.16, 15.52–15.53 directions 15.16 discretion 15.02, 15.04 exhibits 15.38, 15.52–15.53 foreign courts, assistance in obtaining evidence from 16.98–16.102

414 Hague Convention, letters of request under 16.98, 16.100 misrepresentation and non-disclosure 12.02, 12.04, 12.23–12.27 oral evidence 15.02 parol/extrinsic evidence 3.26–3.27, 3.33, 4.25–4.26, 4.27, 7.51 party autonomy 15.04 procedure 3.14 relevance 15.02 settlements 5.30 weight 15.02 written evidence 15.02 examination in chief 15.33–15.34, 15.64 excess point 2.17–2.18, 9.01, 9.02–9.27 aggregation 6.26 Declarations 9.05, 9.08 defence costs 9.04, 9.06 excess layers 9.04 joint ventures 9.40, 9.42 Limits of Liability 9.03–9.04 Loss Payable Condition 9.07, 11.29, 11.32–11.33, 11.38 market conditions 9.04 minimum excess point 2.17, 9.05, 9.15–9.17, 9.24–9.25 per-occurrence retention 9.05–9.13 policy limits 9.34 Schedule B, insurance policies listed in 2.17–2.18, 9.03 Ultimate Net Loss 9.02, 9.05–9.13, 9.20 underlying insurance limits 9.14–9.27 exclusions 2.21, 10.01–10.134 advertising exclusion 10.90 aircraft exclusion 10.105 all risks 10.28 allocation of payments 5.42, 5.44 burden of proof 10.07 business risk exclusions 10.62–10.63 care, custody or control etc exclusion 10.40–10.42, 10.53–10.57 causation 10.01–10.02, 10.10, 10.11–10.27 CGL policies 10.25, 10.29–10.30, 10.42 contra proferentem rule 10.06–10.07, 10.10 ERISA exclusion 10.126–10.127 exceptions preserving coverage 10.09 expected or intended injury 1.34–1.37, 7.15–7.18 facts falling within exclusion 10.01 interpretation 10.01–10.10, 10.30, 10.42 language 1.30 nuclear and radioactive contamination (outside the US) exclusions 10.125 own product exclusions 10.75 own work exclusions 10.75

Index owned property exclusion 10.40–10.52 particular part exclusion 10.58–10.61 pollution exclusion 10.10, 10.107–10.124 prior to inception or retroactive coverage date and other insurance exclusion 10.32–10.34 professional services exclusion 10.37–10.39 repetitive stress exclusion 10.128 securities, antitrust etc exclusion 10.129–10.134 specific exclusions 10.28–10.31 toxic substances exclusion 10.102–10.104 war exclusion 10.91–10.101 watercraft exclusion 10.106 workers’ compensation etc exclusion 10.35–10.36 EXEL Ltd 1.20 expected or intended injury or damage 1.34–1.37, 7.01–7.49 actual injury or damage 7.03–7.05, 7.32–7.47 allegations of injury or damage 7.04–7.05, 7.32–7.47 burden of proof 7.14, 7.15–7.18 causation 7.33 claims history 7.36 coverage clause 5.02 deliberate wrongdoing 7.13 fortuitous losses and related doctrines 7.06, 7.07–7.11 historical experience of a level or rate of actual or alleged injury or damage 7.36–7.38 imputation of knowledge to entity 7.31 integrated or batch occurrences 7.03, 7.35, 7.38 known losses or risks 7.10 maintenance deductible 7.01–7.06, 7.12, 7.32–7.33, 7.41, 7.47, 7.50–7.70 nature of expectation or intent 7.05, 7.33 negligence 7.13, 7.24, 7.28 punitive damages 5.43 standard of proof 7.18 subjective standard 7.14, 7.18, 7.19–7.30, 7.39 technical issues 15.43–15.44 timing of determination 7.05, 7.22, 7.48–7.49 what must be expected or intended 7.14, 7.32–7.47 whose expectation or intention is relevant 7.14, 7.31 expert evidence 15.42–15.50 Expiration Date Condition 11.110 exposure theory 1.07–1.08, 1.11, 5.59 extension of policies 11.94–11.96 federal court system (United States) A.13–A.16, A.18–A.21, A.24 fees and expenses of arbitrators 17.27, 17.37, 17.41 foreign law 3.14, 15.05

Index Former Subsidiaries, Affiliates and Associated Companies Condition 11.111 fortuitous losses and related doctrines 7.06, 7.07–7.11 fraud allocation of payments 5.44–5.45 concealment of material facts 7.08, 12.10–12.11, 12.41 exclusion 10.129–10.134 further particulars/further information 15.10–15.11, 15.14 good faith fair dealing 5.38–5.39, 11.39, 11.95, 11.111, 13.31 notice of occurrences 8.12 settlements 5.38–5.39 uberrimae fidei 12.05, 12.10 governing law 3.04–3.30 applicable substantive law 3.04, 3.20–3.28 arbitration agreements, law governing 3.04, 3.06–3.11 curial law 3.04, 3.16–3.17 internal procedural law of the arbitration 3.04, 3.12–3.15 law governing choice of law 3.04, 3.18–3.19 legal rules governing the dispute 3.04 underlying claims 3.04, 3.29–3.30 Hague Convention on the Taking of Evidence Abroad 16.98, 16.100 Heading Condition 11.114 human rights due process A.09 European Convention on Human Rights (ECHR) 14.38 Human Rights Act 1998 A.09 legal professional privilege 16.28 private and family life, right to respect for 16.28 IBA Guidelines on Conflicts of Interest 14.38, 14.44 IBA Rules on the Taking of Evidence in International Commercial Arbitration 16.04–16.16, 16.21–16.22, 16.31 impartiality of arbitrators 14.37, 14.38–14.44, 14.47 inception date 2.04, 2.10, 2.19 independence of arbitrators 14.21 insolvency 9.23–9.25 Inspection Condition 11.08–11.11, 11.44, 11.48–11.49 Insurance Services Office (ISO) (US) 1.02, 1.23, 10.53

415

Insured, definition of 5.07–5.10, 11.12 integrated or batch occurrences 2.05–2.06 aggregation 1.32–1.33, 6.24–6.25, 6.27, 6.38–6.46 limits of liability 2.16 maintenance deductible 7.66 notice of occurrences 8.04, 8.09 policy limits 9.35, 9.39 trigger of coverage, notice as 8.38–8.44 interest 17.01–17.26 9 per cent rate 17.05–17.10, 17.16–17.17, 17.21 applicable law 17.12–17.20 choice of law 17.20 commencement date 17.24–17.25 compensatory, as 17.03–17.04, 17.23 compound interest 3.44, 17.04, 17.07 conflict of laws 17.14, 17.18, 17.25 date of award, to the 17.26 delay in prosecution of the claim 17.03 discretion 17.03–17.24 diversity actions 17.21 Loss Payable Condition 11.28, 11.53–11.55, 17.24 pre-judgment interest 17.18, 17.20–17.21 procedure and substance, division between 3.44 punitive, as compensatory rather than 17.03–17.04 rate 17.03–17.26 simple interest 3.44, 17.04–17.05, 17.07 substantive right to interest 17.08–17.09, 17.16–17.22, 17.25 internal procedural law of the arbitration 3.04, 3.12–3.15 interpretation of insurance contracts 4.01–4.31, 10.01–10.10 aids to interpretation 4.09 ambiguities 4.09, 4.13, 4.22, 4.27 analogy, reasoning by 4.31–4.32 applicable law 3.25–3.28 arbitrariness 4.27 basic approach 4.09–4.13, 10.01–10.10 canons of construction, modification of 3.25 context 4.12, 4.25 contra proferentem rule 3.27, 4.22, 4.27, 4.28–4.29, 4.31 drafting, legal background to 4.25 even-handed interpretation 4.10, 4.13 forbidden grounds in Bermuda Form 4.21–4.27, 4.30–4.32 fruit of the poisonous tree 4.28–4.32 intention of parties 4.09, 4.11, 4.13, 4.24–4.26, 4.29 literal interpretation 3.27 objectivity 4.24

416 parol/extrinsic evidence 3.26–3.27, 4.18–4.19, 4.25–4.26, 4.27 plain or natural meaning 4.09–4.11 presumptions 4.27 principles of construction 4.21–4.27 public policy 3.27–3.28 reasonable expectations 3.26, 4.23–4.24, 4.27 rules of construction 4.09–4.10 standard terms 4.22 subjectivity 4.13, 4.25 two-stage approach 4.09–4.10 whole, interpretation of contracts as a 4.24 interrogatory, discovery by 16.105–16.106 intra-uterine device claims 1.30, 10.102 joint ventures 5.09, 9.40–9.44 JP Morgan 1.01, 1.16–1.17 judgments 5.19–5.23 allocation of payments 5.43–5.44 coverage clause 5.19–5.23 damages 5.19, 9.06–9.09 Loss Payable Condition 5.19, 11.32–11.33, 11.36, 11.44 key features of the Bermuda Form 1.21–1.37 aggregation/stacking of claims 1.31–1.33 broadening insurance coverage 1.27 claims made policies 1.23–1.25 detailed policy language 1.29–1.30 dispute resolution mechanism (London arbitration) 1.09, 1.27–1.28 drafting 1.29 end-points 1.25 expected or intended injury 1.34–1.37 maintenance deductible 1.34–1.37 occurrence forms 1.09, 1.22–1.26 known loss/risk 7.07–7.11 late notice defence 8.05, 8.10–8.21, 8.37, 13.08, 13.17 Law of Construction and Interpretation Condition 11.91 legal and economic origins of the Bermuda form 1.01–1.37 1980s, collapse of excess casualty insurance in 1.01–1.02, 1.06, 1.09 ACE Insurance Co Ltd, creation of 1.01, 1.06, 1.16–1.20 alternative market, creation of an 1.01 key features of the Bermuda form 1.09, 1.21–1.37 policy forms and liability problems 1.01–1.06 United States legal decisions on insurance coverage issues 1.07–1.15 XL, creation of 1.01, 1.06, 1.16–1.20

Index legal professional privilege 16.07, 16.21, 16.23–16.43 see also waiver of privilege adverse inferences 16.41 applicable law 16.23–16.32 disclosure/discovery 3.46, 16.07, 16.19, 16.21, 16.23–16.92 discretion 16.31 human right, as 16.28 IBA Rules on the Taking of Evidence in International Commercial Arbitration 16.21, 16.31 Inspection Condition 11.10 legal advice privilege 16.33–16.40 litigation privilege 16.19, 16.33, 16.37–16.40 procedure and substance, division between 3.46 public interest 16.34 legal representation 14.02–14.14 advocacy work 14.07, 14.10–14.11 barristers, engaging English 14.05–14.07, 14.12 English lawyers 14.03, 14.05–14.09, 14.12–14.14, A.04–A.06, A.12–A.21 QCs, engaging 14.05–14.06 solicitors, engaging English 14.06–14.07 trial preparation work 14.09–14.11 United States lawyers 14.02–14.03, 14.06–14.14 letters of request 16.98–16.100 liability, insurance against 5.03–5.06 liability insurance crisis during the 1980s 1.01–1.02, 1.06, 1.09, 1.21, 1.24, 1.27, 1.30 Liability of the Company Condition 11.93 likely to involve the policy, claims which are 1.32, 8.05, 8.06, 8.22–8.37 limitation periods Foreign Limitation Periods Act 1984 3.36 Loss Payable Condition 11.28, 11.33, 11.40–11.52 misrepresentation and non-disclosure 12.40–12.43 procedure and substance, division between 3.36 limits of liability 2.15–2.16, 9.34–9.39 see also excess point all occurrences reported in any period during Coverage A 2.15–2.16 amount of coverage provided 9.01 Declarations 9.01 each occurrence, application to 2.15–2.16 excess point 9.01, 9.03–9.04 joint ventures 9.40–9.44 last year of Coverage A and whole of Coverage B, all occurrences reported in 2.15–2.16 reduction in limits 9.40–9.42 Reinstatement Condition 11.97–11.98 Loss Payable Condition 11.27–11.55 accrual of a cause of action 11.40–11.52 additional demands 11.40 Assistance and Co-operation Condition 11.44

Index attachment to the policy 11.29, 11.30–11.33, 11.44 demand, making a 11.29, 11.40–11.42, 11.48–11.49 estoppel 11.50 excess point 9.07, 11.29, 11.32–11.33, 11.38 finality of judgments 11.29 Inspection Condition 11.44, 11.48–11.49 interest and other matters 11.28, 11.53–11.55, 17.24 judgments 11.32–11.33, 11.36, 11.44 limitation periods 11.28, 11.33, 11.40–11.52 Limits of Liability 11.31–11.32 obligation to pay arises, time when 11.29 proof of loss 11.44–11.45, 11.49 retention amounts 11.31 settlements 5.24, 5.34, 11.29, 11.32–11.39 Ultimate Net Loss 11.34, 11.38, 11.40, 11.43 underlying insurance 11.31–11.32 maintenance deductible 7.50–7.70 actual injury or damage 7.47, 7.56, 7.66 aggregation 7.67 alleged injury or damage 7.47, 7.56, 7.60–7.61, 7.66 balance between policyholders and insurers 1.37 business expenses 7.65 expected or intended injury 7.01–7.06, 7.12, 7.32–7.33, 7.41, 7.50–7.70 fortuity 7.64 fundamental difference in nature 7.12 historical experience 7.50, 7.54, 7.58, 7.67–7.69 integrated or batch occurrences 7.50, 7.66 mass torts 7.63–7.64 noise level routine claims 1.36–1.37, 7.51 notice of integrated occurrences 7.66 pharmaceuticals 7.52, 7.59 reasons for maintenance deductible 7.52 spike in claims 1.36–1.37 term, use of 7.51 vastly greater in order of magnitude 7.12, 7.54–7.69 wording, development of 7.02 manifestation theory 1.07, 1.10 Marsh & McLennan 1.01, 1.16–1.18 mass tort claims Agent Orange in Vietnam, use of 1.05, 5.27, 5.30 asbestos 1.05, 1.30 CERCLA 1.05 DES (diethylstilbestrol), miscarriages caused by 1.05, 1.30, 10.102 intra-uterine device claims 1.30, 10.102 maintenance deductible 7.63–7.64

417

notice of occurrences 8.09 occurrence definition 6.05 settlements 5.33, 11.35, 11.38 mergers and acquisitions 5.10, 11.67–11.68 military or usurped power 10.95–10.96, 10.100 misrepresentation and non-disclosure 12.01–12.43 affirmation 13.14–13.15 application forms 12.11, 12.14–12.15 avoidance 12.01–12.02, 12.28, 12.32–12.35, 12.38, 13.14 burden of proof 12.04, 12.09, 12.14, 12.16, 12.23–12.27 Changes Condition/Non-waiver Condition 13.26 continuous contract, Bermuda Form as a 12.39 English law 12.01, 12.07 estoppel 12.06, 13.23, 13.26 evidence 12.02, 12.04, 12.23–12.27 experts 12.04 formation of contracts 13.09 fortuitous losses and related doctrines 7.08 fraud 12.10–12.11, 12.29, 12.41, 13.12 inducement 12.16, 12.20, 12.41 interpretation 12.14–12.15 knowledge 13.14–13.15 legal professional privilege, waiver of 16.45 life insurance 12.07 limitation periods 12.40–12.43 material misrepresentation 3.32, 7.08, 12.04, 12.08–12.11, 12.16–12.28 opinion, statements of 12.13 premiums acceptance 13.14 amount 12.38–12.39 rescission 12.29–12.30, 12.34, 12.38–12.39 return 12.29–12.30, 12.34, 12.38–12.39 remedies 12.29–12.43 rescission 3.39, 3.41 ab initio 12.29–12.32 affirmation of contracts 12.31, 12.34 affirmative demands 12.29 common law 12.31 declarations 12.29 delays in seeking 12.02 formation of contracts 12.32 fraudulent concealment of material facts 12.10 limitation periods 12.41–12.43 partial 12.39 premiums, return of 12.29–12.30, 12.34, 12.38–12.39 retrospectivity 12.42 voidable contracts 12.31 waiver 12.02, 12.30, 12.33, 12.35–12.37, 13.13 specific questions in applications 12.11 statements of past or present fact 12.13

418

Index

subjectivity 12.13 summary judgments, application for 12.07, 12.08–12.09, 12.13, 12.18, 12.21–12.25 uberrimae fidei 12.05, 12.10 waiver 12.02, 12.06, 13.09–13.15, 13.26 modification of the New York law proviso, interpretation of 4.14–4.20 ambiguity 4.05 canons of construction 4.03, 4.05–4.07 choice of law 4.02, 4.04 context 4.15, 4.17–4.18 contra proferentem rule 4.07 dictionaries 4.20 even-handed interpretation 4.05–4.07, 4.15 exclusions 10.01–10.10, 10.42 fruit of the poisonous tree 4.28–4.32 intention of parties 4.14–4.15 parol or extrinsic evidence 4.18–4.19 plain or natural meaning 4.14 positive approach 4.03–4.06 punitive damages 4.02 structure 4.03–4.08 surrounding circumstances 4.18–4.20 whole, interpretation of contract as a 4.16 Named Insured Insured, definition of 5.07–5.10 minority interests 9.43–9.44 Representation Condition 11.56–11.57 New York law/practice A.12–A.21 aggregation 6.33–6.34 allocation of payments 1.13, 5.48, 5.51–5.54 case law A.17–A.18 causation 10.11, 10.14–10.24 choice of law 3.01–3.03, 3.18–3.19 court system A.12–A.18 diversity jurisdiction A.14–A.15 federal court system A.13–A.16, A.18–A.21, A.24 fortuitous losses and related doctrines 7.07–7.11 Insurance Law 8.15–8.21, 12.03–12.04, 12.12, 12.16–12.23, 12.27–12.28, 12.32, 13.03, 13.05 interest 17.01, 17.05–17.25 misrepresentation and non-disclosure see misrepresentation and non-disclosure New York Pattern Jury Instructions 12.08, 12.18, 12.32 notice of occurrences 8.05, 8.10, 8.13–8.21, 8.27 property damage 5.61–5.63 public policy 3.24 punitive damages 3.23 remedies 3.39, 3.41–3.42 rescission 3.39, 3.41, 13.13 settlements 5.25–5.31, 5.33, 5.36–5.40 sources A.01–A.02, A.17–A.21

state courts A.12, A.15–A.16, A.18, A.21, A.24 statutes A.17 waiver 13.01–13.17 non-disclosure see misrepresentation and non-disclosure notice of occurrences 8.01–8.44 aggregate limit 8.04, 8.34–8.35, 9.38–9.39 as soon as practicable 8.01, 8.05, 8.12, 8.25–8.26 condition of coverage, as 8.02, 8.10–8.31 Coverage A 2.07 Coverage B 2.07 coverage clause 5.67, 8.03 currency of policy (Coverage A), during 8.03 delay 8.05, 8.10–8.21, 8.24 drafting 8.09 early notice 2.08 excess policies 8.11–8.12, 8.28 extended reporting period (Coverage B), during 8.03 fixing limits and retentions applied to claims 8.02 functions 8.02 good faith 8.12 integrated or batch occurrences 2.08, 7.20, 7.66, 8.01, 8.04, 8.09, 8.38–8.44, 9.38–9.39 knowledge of occurrence 8.05, 8.11, 8.24, 8.27, 8.30 late notice defence 8.05, 8.10–8.21, 8.37, 13.08, 13.17 likely to involve the policy, claims which are 1.32, 8.05, 8.06, 8.22–8.37 mass torts 8.09 method of giving notice 8.07–8.09 Named Insured, from 8.05 objectivity 8.11, 8.23, 8.29 occurrence reported policy, Bermuda Form as 8.03 permissive notice 8.01 policy limits 9.38–9.39 Policy Period, during 8.03 pollution 8.06, 10.119 prejudice, relevance of 8.14–8.21 primary coverage 8.11, 8.14, 8.28 promptly, requirement to be given 8.05 reasonableness 8.11–8.12, 8.24–8.27, 8.30 retentions and limits, relation to 8.04 scope of notice 8.09 subjectivity 8.11, 8.23, 8.27 timely notice 8.06, 8.10–8.21 trigger of coverage, notice as 8.02–8.04, 8.32–8.44 underwriters, to 8.08 writing, in 8.06–8.08, 8.25–8.26 XL Guidelines 8.24, 8.31 nuclear and radioactive contamination (outside the US) exclusions 10.125

Index occurrence see integrated or batch occurrences; notice of occurrences; occurrence definition occurrence definition 6.01–6.47 actual or alleged injuries 6.11, 6.18–6.20, 7.03 aggregation 6.23–6.46 allocation of payments 5.42 basic definition 6.06–6.22 bodily injury 6.01, 6.05 causal links 6.05, 6.09–6.11, 6.20–6.21 CGL policies 6.01, 6.12 claims made policies 6.01, 6.12 commencement of event or conditions 6.13–6.14, 6.17–6.21 common factors 6.08 continuous, intermittent or repeated exposure 6.10, 6.17 coverage clause 5.02, 5.67 events and products 6.09–6.11 expected or intended injury 6.47, 7.01, 7.03, 7.12, 7.15, 7.40, 7.44–7.45 extent of the occurrence 6.21–6.22 functions of definition 6.03 inception date 2.04, 2.19 insured perils 6.03 personal injury 6.01, 6.05, 6.07–6.11, 6.16, 6.19, 6.22, 6.47, 7.03 product liability 6.07, 6.11, 6.15 property damage 6.01, 6.07–6.11, 6.47, 7.03 settlements 5.25 temporal factors 6.03, 6.12–6.22 triggers 6.01 types of case 6.04 occurrence integration see integrated or batch occurrences officers, directors and employees acting within scope of duties 5.09 Other Insurance Condition 9.28–9.33, 11.58 owned property exclusion 10.40–10.52 parol/extrinsic evidence 3.26–3.27, 3.33, 4.25–4.26, 4.27, 7.51 particular part exclusion 10.58–10.61 per-occurrence limit 9.05–9.13, 9.16–9.17, 9.19, 9.22–9.26, 9.35, 9.40 period of cover 2.09–2.14 Annual Periods 2.09–2.10, 2.14 cancellation 2.09–2.10 Coverage A 2.09–2.14 Coverage B 2.09–2.14 renewal meetings 2.09 temporal limitations 2.09–2.12 personal injury see also expected or intended injury or damage actual damage or injury 6.11, 6.18–6.20, 7.04–7.05, 7.32–7.47, 7.56, 7.66

419

alleged injuries 5.57, 6.11, 6.18–6.20, 7.04–7.05, 7.32–7.47, 7.56, 7.60–7.61, 7.66 causal connection 5.17 consequential loss 5.15 coverage clause 5.04, 5.12, 5.55, 5.56–5.57, 5.67 definition 5.56 expected or intended injury 6.47, 7.04–7.05, 7.32–7.47 injury-in-fact trigger 1.14, 5.57 occurrence definition 6.01, 6.05, 6.07–6.11, 6.16, 6.18–6.20, 6.22, 6.47 progressive injuries or illnesses 5.57 specific person or property, actual or alleged injury or damage to a 7.34 pharmaceuticals aggregation 6.27–6.31, 6.38–6.46 expected or intended injury 6.29, 15.44 injuries not qualifying as occurrences 6.38–6.46 maintenance deductible 7.52, 7.59 product liability 7.41–7.42 warn, failure to 7.41–7.42 pleadings stage 15.09–15.15 amendments 15.15 citations 15.13 cultural differences between English and US arbitration 15.09–15.13 detail, amount of 15.10–15.11 further particulars/further information 15.10–15.11, 15.14 limits 15.13–15.14 purpose 15.13 Policy Extension Condition 11.94–11.96 policy limits see limits of liability; excess point political or terrorist purposes 10.93–10.96, 10.101 pollution exclusion 10.10, 10.107–10.124 absolute pollution exclusion (APE) 10.109–10.115 asbestos 10.111 background 10.109 CGL policies 10.109–10.111, 10.124 climate change 10.117–10.118 discharge, meaning of 10.108 Exclusion K 10.107 hazardous waste 10.112 industrial pollution 10.109 interpretation 10.10, 10.112 negligence 10.113–10.114 non-industrial pollution claims 10.115 Notice Condition 11.112 notice of occurrence 10.119 paint or solvent fumes, inhalation of 10.111–10.112 personal injury, property damage or advertising liability 10.107, 10.119 pollutant, meaning of 10.108 product pollution liability 10.120–10.124

420

Index

qualifications on the exclusion in Bermuda Form 10.119 reasonable expectations doctrine 10.114 sudden and accidental pollution exclusion 10.109–10.110, 10.119 waste, definition of 10.108 post-expiration injury 6.38–6.46 precedent confidentiality 15.23 cultural differences A.23–A.26 English law A.10, A.23–A.26 jury trials A.25 large commercial cases A.26 New York law/practice A.18–A.21, A.23–A.26 pre-inception injury 6.38–6.46 preliminary issues/bifurcation 15.31, 17.27, 17.29 premiums amount 11.02–11.05 Annual Period 11.05–11.07 flat premiums 11.05 increase or decrease in amount 11.05 misrepresentation 11.04, 13.14 non-disclosure 11.04 non-payment 11.79 Premium Condition 11.02–11.07 pro rata premium, refund of 11.74, 11.76 prompt payment 11.06 Reinstatement Condition 11.101 rescission 12.29–12.30, 12.34, 12.38–12.39 preparation for hearings 14.09–14.11, 15.51–15.58 private and family life, right to respect for 16.28 privilege see legal professional privilege privity of contract 1.05 procedure see also procedure and substance, division between appointment of arbitrators 14.15–14.32 Arbitration Act 1996 3.12–3.14 co-operation 3.13 costs 15.22, 17.28 course and conduct of a Bermuda Form arbitration 15.01–15.70 default powers and rules 3.13 delay 3.13 directions 15.16, 15.18 discretion 15.02, 15.04 evidential matters, tribunal’s power to decide 3.14 interest 17.08, 17.13–17.18, 17.21, 17.25 internal procedural law of the arbitration 3.04, 3.12–3.15 party autonomy 15.04 procedure and substance, division between 3.31–3.46 burden of proof 3.32 costs and attorneys’ fees 3.45

currency 3.43 damages 3.37–3.38 estoppel 3.34–3.35 interest 3.44 limitation of actions 3.36 remedies 3.37–3.42 product liability 5.04–5.06 aggregation 6.27–6.31, 6.35 expected or intended injury 7.34, 7.41–7.43, 7.48 incorporation doctrine 5.62 Insured’s Products 10.120–10.124 modifications to products 7.43 occurrence definition 6.07, 6.11, 6.15 warn, failure to 7.41–7.42 professional services exclusion 10.37–10.39 property damage 5.04, 5.12, 5.55, 5.58–5.64 see also expected or intended injury or damage asbestos 5.59–5.60, 5.62, 5.64 continuous injurious process 5.64 coverage clause 5.04, 5.12, 5.55, 5.58–5.64, 5.67 decline in value exclusion 10.86–10.89 definition 5.17, 5.58, 5.62 expected or intended injury 6.47, 7.04–7.05, 7.32–7.47 incorporation doctrine 5.62 injury-in-fact test 1.14, 5.60 occurrence definition 6.01, 6.07–6.11, 6.18–6.20, 6.47 owned property exclusion 10.40–10.49 products liability exclusions 10.66, 10.69, 10.76–10.77, 10.86–10.89 tangible property, physical damage or destruction of 5.58–5.63 timing 5.59, 5.64 Proration of Losses Condition 11.92 protective orders Appendix 2 (ch 15) 16.89 punitive damages applicable law 3.23–3.24 enforcement of arbitration awards 3.24 expected or intended injury 5.43 limited coverage 1.09 maintenance deductible 7.57 vicarious liability 3.23 Queen’s Counsel 14.05–14.06, 14.36, A.05 reasonable expectations 3.26, 4.23–4.24, 4.27, 10.13, 10.114, 16.45 reasonableness of settlements 5.25, 5.30–5.32, 5.36, 5.38–5.40, 11.35 re-examination 15.65 Reinstatement Condition 11.94, 11.97–11.101

Index remedies 3.37–3.42 see also damages; rescission anti-suit injunctions 3.11 applicable law 3.40–3.41 hypothetical or advisory remedies 3.42 misrepresentation and non-disclosure 12.29–12.43 negative declarations 3.39, 3.42 New York law 3.39, 3.41–3.42 procedure and substance, division between 3.37–3.42 specific performance 3.40 remote hearings 15.20, 15.53 repetitive stress exclusion 10.128 Representation Condition 11.56–11.57 rescission ab initio 12.29–12.32 affirmation of contracts 12.31, 12.34 affirmative demands 12.29 common law 12.31 declarations 12.29 delays in seeking 12.02 formation of contracts 12.32 fraudulent concealment of material facts 12.10 limitation periods 12.02, 12.41–12.43 misrepresentation and non-disclosure 3.39, 3.41, 12.29–12.34, 13.14 partial rescission 12.39 premiums, return of 12.29–12.30, 12.34, 12.38–12.39 retrospectivity 12.42 voidable contracts 12.31 waiver 12.02, 12.30, 12.33, 12.35–12.37, 13.13 reservation of rights 13.15–13.18, 16.60 retroactive coverage date 1.05, 2.10, 2.20, 10.32–10.34, 11.103–11.109, 12.42 Rome I Regulation 3.40 Schedules to Bermuda Form 2.23–2.24 additional insureds 2.23 Coverage B, calculation of premiums applicable to 2.24 Schedule A 9.43 Schedule B 2.17–2.18, 9.03, 9.30, 9.43 underlying insurance 2.23 securities, antitrust etc exclusion 10.129–10.134 selection of an arbitrator see appointment and selection of arbitrators separability of arbitration clauses 3.06, 3.09–3.10 settlements 5.24–5.40 approval 5.24, 5.34–5.40, 11.29, 11.32–11.39, 13.08 attachment point, reaching the 5.24 consent 5.24, 5.34, 5.36, 5.39–5.40, 11.37 costs 5.21, 11.38 coverage clause 5.24–5.40

421

damages 5.24, 9.06–9.09 defend, breach of duty to 5.37 denial of liability 5.37 dispute over whether case should be fought 5.39 English law 5.29, 5.39–5.40 evidence 5.30 expert evidence 5.30 good faith and fair dealing 5.38–5.39, 11.39 layer, within insured’s 11.34, 11.38 legal liability to pay 5.25–5.33 legal professional privilege, waiver of 16.87 liability policies 5.24 Loss Payable Condition 5.24, 5.34, 11.29, 11.32–11.39 low-level settlements 5.34, 9.27 mass tort claims 5.33, 11.35, 11.38 New York law 5.25–5.31, 5.33, 5.36–5.40 notification 5.35 policyholders, by 5.24–5.40 public policy 5.31 reasonableness 5.25, 5.30–5.32, 5.36, 5.38–5.40, 11.35, 16.45 writing 11.29, 11.32–11.35 sistership exclusions 10.62, 10.79–10.85 skeleton arguments 15.57 solicitors (English law/practice) 14.06–14.07, 14.45, 14.47, A.05–A.06 sources of law A.01–A.02, A.07–A.11, A.17–A.21 stacking see aggregation state courts A.12, A.15–A.16, A.18, A.21, A.24 statements of case, exchange of 15.06 statutes as a source of law A.07–A.09, A.17 subrogation 11.59–11.63, 16.73 subsidiaries 5.07–5.09, 9.17, 9.43, 11.111 substantive hearings 15.59–15.68 citation of authorities 15.68 closing arguments 15.67 costs 17.27, 17.30 cross-examination 15.59, 15.62, 15.65 duration of hearings 15.59 examination in chief 15.64 experts 15.63 final submissions 15.66 one sitting, time to conclude hearing in 15.59 opening statements 15.60 oral phase 15.59–15.68 post-hearing briefs 15.67 preparation 15.51–15.58 re-examination 15.65 remote hearings 15.20, 15.53 substance and procedure, division between 3.31–3.46 video links 15.61 visual aids 15.60 witnesses 15.61–15.65

422

Index

successor liability 11.67–11.68 Supreme Court England and Wales A.04, A.10 United States A.13 terrorist or political purposes 10.93–10.96, 10.101 toxic substances exclusion 10.102–10.104 trigger of coverage, notice as 8.02–8.04, 8.32–8.44 aggregation 8.35, 8.40–8.41 integrated or batch occurrences 8.38–8.44 knowledge of occurrence 8.32–8.33 late notice defence 8.37 likely to involve the policy, claims which are 8.32–8.37 Notice of Occurrence Condition 11.16 occurrence reported form 2.02–2.03, 6.01 precautionary notices 8.34–8.36 whether notice should be served 8.34–8.35 uberrimae fidei 12.05, 12.10 Ultimate Net Loss/all sums insurance for damages 5.11–5.18 causal connection 5.17 computation 9.08–9.09 coverage clause 5.11–5.18 damages 5.12–5.17, 9.07 defence costs 5.20–5.23, 9.06–9.09 excess point 9.02, 9.05–9.13, 9.20 Insured, identification of the 9.09 Loss Payable Condition 11.34, 11.38, 11.40, 11.43 per-occurrence retention 9.05–9.13 umbrella policies 1.03–1.05, 5.03 underlying insurance captive insurers 9.19 defence costs 9.18–9.19 excess point 9.14–9.27 exclusions 9.21 insolvency 9.23–9.25 limits 9.15–9.25 Loss Payable Condition 11.31–11.32 minimum fixed excess point 9.15–9.17, 9.24–9.25 parent companies and subsidiaries 9.17 per-occurrence retention 9.16–9.17, 9.19, 9.22–9.26 scope of coverage 9.21 settlement for lower sum than policy limit 9.27 vicarious liability 3.23 video links 15.61 waiver 13.01–13.18 see also waiver of privilege Changes Condition/Non-waiver Condition 11.64, 13.25–13.31

co-operate, breach of duty to 11.21, 13.08 definition 13.01 disclaimers 13.03, 13.06, 13.16–13.18 estoppel 3.35, 12.06, 13.01–13.02, 13.20 general principles 13.07–13.11 implied waiver 13.07 knowledge 13.02, 13.10, 13.12, 13.14–13.17 late notice defence 8.20, 13.08, 13.17 misrepresentation and non-disclosure 12.02, 12.06, 12.30, 12.33, 12.35–12.37, 13.09–13.15 settlements, failure to provide approval for 13.08 validity of the policy, waiver of defences concerning the 13.12–13.15 waiver of privilege 16.44–16.92 see also common interest privilege confidentiality 16.90–16.92 content of privileged documents, reliance on 16.82–16.83 fairness 16.82 implied waiver 16.76–16.79 partial waiver 16.81, 16.89 production previous proceedings, in 16.89–16.92 time for seeking 16.85–16.87 protective orders 16.89 references to privileged documents 16.82 scope of waiver 16.87 subject matter waiver 16.84 voluntary waiver by policyholders 16.80–16.88 warn, failure to 6.28–6.30, 6.33, 7.41–7.42 watercraft 2.23, 10.106 witness statements, exchange of 15.08, 15.32–15.41, 15.51, 15.56 bundles, cross-referencing to 15.56 Commercial Court, guidance from the English 15.35 cross-examination 15.33, 15.38 examination in chief, absence of 15.33–15.34 exhibits 15.39 interrogatory, discovery by 16.106 minimalist approach 15.38 narrative description, drafting as a 15.39 preparation 15.33–15.36 substantive hearings 15.64 supplemental statements 15.32, 15.40–15.41 witnesses 15.61–15.65 see also witness statements, exchange of cross-examination 15.07–15.08, 15.35, 15.50, 15.55, 15.59, 15.62, 15.64–15.65 Delaware rule 15.65 depositions 15.06–15.07, 16.20, 16.96–16.102, 16.104

Index discussing case before finishing evidence, prohibition on 15.65 examination in chief 15.33–15.34, 15.64 outside UK, securing attendance of witnesses 16.98–16.102 preparation 15.62 re-examination 15.65

substantive hearings 15.61–15.65 sworn witnesses 15.61 video links 15.61 workers’ compensation 1.05, 10.35–10.36 XL Insurance Ltd, creation of 1.01, 1.06, 1.16–1.20, 9.32

423

424