Research Handbook on Marine Insurance Law (Research Handbooks in Private and Commercial Law series) 1803926678, 9781803926674

Citation preview

RESEARCH HANDBOOK ON MARINE INSURANCE LAW

RESEARCH HANDBOOKS IN PRIVATE AND COMMERCIAL LAW The Research Handbooks in Private and Commercial Law series is a forum for Research Handbooks covering both the traditional private law topics, such as torts, contracts, equity and unjust enrichment, as well as more commercial topics such as the sale of goods, corporate restructuring, commercial contracts and taxation, among others. Reflecting the approach of the wider Elgar Research Handbooks programme they are unrivalled in their blend of critical, substantive analysis and synthesis of contemporary research. Each Research Handbook stands alone as an invaluable source of reference for all scholars interested in private and commercial law. Whether used as an information resource on key topics or as a platform for advanced study, volumes in this series will become definitive scholarly reference works in the field. For a full list of Edward Elgar published titles, including the titles in this series, visit our website at www​.e​-elgar​.com​.

Research Handbook on Marine Insurance Law Edited by

Özlem Gürses Professor of Commercial Law, King’s College London, UK

RESEARCH HANDBOOKS IN PRIVATE AND COMMERCIAL LAW

Cheltenham, UK • Northampton, MA, USA

© Editor and Contributing Authors Severally 2024

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Control Number: 2023952104 This book is available electronically in the Law subject collection http://dx.doi.org/10.4337/9781803926681

ISBN 978 1 80392 667 4 (cased) ISBN 978 1 80392 668 1 (eBook)

EEP BoX

This book is dedicated by the editor to the memory of Greg Pynt, for his friendship and contribution to insurance law education.

Contents

List of contributorsix Introduction to Research Handbook on Marine Insurance Law1 Özlem Gürses PART I

 

1

Marine insurance law as a treasure trove of legal principles Satoshi Nakaide

2

Protection and indemnity insurance: is it truly insurance? Rhidian Thomas

PART II

6 26

 

3

Has the principle of ‘utmost good faith’ in marine insurance law become like the Cheshire Cat – now you see it and now you don’t? Richard Aikens

4

Co-insurance and rights of subrogation post-Gard Marine And Energy v China National Chartering Company Ltd63 Kyriaki Noussia and Yavuz Can Aslan

5

Scuttling, fortuity and marine perils – from the mortgagee’s and the cargo owner’s points of view Özlem Gürses

42

87

PART III   6

To what extent should marine cargo insurance be construed to include cover for financial loss? John Dunt

7

Total losses under marine policies Peter MacDonald Eggers KC

8

Forwarding cargo to destination: a review of Rules F and G, York– Antwerp Rules Richard Sarll

108 130

vii

163

viii  Research handbook on marine insurance law PART IV   9

Protection & indemnity clubs and arbitration clauses Rob Merkin

10

Taxonomizing third-party rights of direct action against marine liability insurers204 Paul Myburgh

11

Choice-of-law issues in marine insurance cases in the United States Michael Sturley

PART V

183

223

 

12

Insurance requirements in Incoterms® 2020 Margarida Lima Rego

245

13

The use of insurance documents in international trade: enabling digitalisation Miriam Goldby

260

PART VI   14

Marine insurance fraud and emerging technology Gary Meggitt

275

15

The role of insurance in regulating cyber risks in the shipping industry Feng Wang

306

16

Legal and regulatory issues for sustainable marine insurance: the Poseidon Principles for Marine Insurance Livashnee Naidoo

17

The German Pandemic Exclusion Clause and its write-back clauses Dieter Schwampe

333 353

Index372

Contributors

Richard Aikens (The Rt Hon Sir) was a barrister specialising in commercial law before going to the High Court Bench in 1999, where he sat in the Commercial Court. He was a Lord Justice of Appeal from 2008 to 2015. He now works as an international arbitrator and is Visiting Professor at King’s College, London and Queen Mary, University of London, where he teaches aspects of English contract law, maritime law and arbitration law. He is one of the authors of Bills of Lading (3rd ed. 2021) and an editor of Dicey, Morris & Collins On the Conflict of Laws (16th ed. 2022), and contributes to legal journals. He was President of the British Insurance Law Association from 2012 to 2014. Yavuz Can Aslan is Research Assistant at Kadir Has University’s Faculty of Law, specialising in Commercial Law. Yavuz is studying a PhD in Private Law at Istanbul University. He holds an LLM in Maritime Law from both University College London and Queen Mary University of London. He also earned a Master of Arts in Private Law from Marmara University and an LLB from Galatasaray University. Yavuz has assisted with courses in Maritime Law, Insurance Law, International Sales Law and International Law of the Sea, and is committed to advancing legal knowledge through his research and teaching endeavours. He has been called to the Bar of Istanbul since 2014. John Dunt is Visiting Fellow at the Institute of Maritime Law, University of Southampton. He is author of Marine Cargo Insurance (2nd edn. Informa Law from Routledge, 2016) (winner of the 2010 British Insurance Law Association prize) and editor of International Cargo Insurance (Informa, 2012). John was a partner at Clyde & Co LLP from 1977 to 2007, specialising in marine insurance, and was ex officio legal advisor to the joint Cargo Committee from December 1999 to April 2007. He was involved in drafting the wordings used in marine cargo insurance policies worldwide to control terrorism risks and was legal advisor to the Joint Cargo Committee Working Party responsible for updating and revising the Institute Cargo clauses 2009. John has written widely on marine cargo insurance legal issues and regularly contributes articles and case notes to LMCLQ and JIML. He is currently revising and updating the third edition of Marine Cargo Insurance. Miriam Goldby is Professor of Shipping, Insurance and Commercial Law and Director of Research at the Centre for Commercial Law Studies, Queen Mary University of London. She was previously director of the Centre’s Insurance, Shipping and Aviation Law Institute (2019–22). She has published extensively in the fields of shipping, insurance and financial law. She participated in the work of UNCITRAL WG IV which led to the adoption of the Model Law on Electronic Transferable Records in 2017. She also undertook a part-time secondment to the Law Commission of England and Wales to work on the Electronic Trade Documents project between November 2020 and March 2022. She is a member of the British Insurance Law Association (BILA) Committee and editor of the BILA journal. Özlem Gürses is Professor of Law at King’s College London. He is the author of Reinsuring Clauses (2010), Marine Insurance Law (3rd ed 2023), The Law of Compulsory Motor Vehicle ix

x  Research handbook on marine insurance law Insurance (2019, Informa) and numerous articles on insurance and reinsurance law. Özlem sits in the Presidential Council of International Association of Insurance Law (AIDA) and the British Insurance Law Association Committee. She chairs the Reinsurance Working Party of AIDA. Özlem frequently presents her research at international and national conferences, and she teaches insurance, marine insurance and reinsurance law at higher education institutions at home and abroad, including Singapore, Sweden, Germany, Italy and China. Margarida Lima Rego is Full Professor and Dean of NOVA School of Law, NOVA University, Lisbon, where she has lectured regularly since 2005. She is the holder of the Jean Monnet Module in EU Insurance Law: Challenges in the SDG Era. Margarida’s main areas of research and practice are civil and commercial law, with a special focus on insurance law. She has been a member of the Board of Appeal of the European Supervisory Authorities since December 2021. Margarida is an active researcher at CEDIS, the School’s R&D Unit, where she founded and coordinates the NOVA Knowledge Centre for Data-Driven Law. Current positions include President of AIDA Portugal and member of the Executive Committee and Scientific Committee of AIDA Europe, of the Scientific Committee of CILA and of the Presidential Council of AIDA World. Peter MacDonald Eggers KC is a barrister at 7 King’s Bench Walk, London specialising in all aspects of commercial law, especially insurance and reinsurance. Peter acts as an arbitrator and was a Deputy Judge of the High Court from 2017 to 2023. Peter has appeared in leading insurance/reinsurance cases. Peter teaches at UCL and is a contributing editor of Chitty on Contracts, co-author of Good Faith and Insurance Contracts and Carver on Charterparties and author of Deceit: The Lie of the Law and The Vitiation of Contractual Consent. Peter is a member of the BILA Committee. Gary Meggitt is Associate Professor, Faculty of Law, The University of Hong Kong. Gary was admitted as a solicitor in England and Wales and spent 12 years in practice with several leading UK law firms. He was also called to the Bar in England and Wales. After leaving private practice, Gary taught at BPP Law School in London and was Course Director of its Full-Time Bar Vocational Course (BVC). He joined the law faculty at the University of Hong Kong in 2007 and teaches civil litigation, commercial dispute resolution, professional conduct and insurance law. He is the author of Wilkinson’s Professional Conduct of Lawyers in Hong Kong (6th Desk Edition, LexisNexis, 2022) and Mediation and ADR Confidentiality in Hong Kong (Wildy, Simmonds & Hill Publishing, 2019). Gary is a member of the Hong Kong Law Society’s Insurance Law Committee and also serves as Convener for Head IV (Accounts & Professional Conduct) of the Overseas Lawyers Qualification Examination. Rob Merkin (KC hon, LLD) is part-time Research Professor in Commercial Law at the University of Reading, Honorary Professor at the University of Auckland and special counsel to New Zealand solicitors Duncan Cotterill. Although semi-retired from academic life, he continues to teach insurance law at a number of universities. He is co-editor of the Lloyd’s Law Reports, editor of the Journal of Business Law and author or co-author of a number of texts on insurance law including Colinvaux’s Law of Insurance (UK), Sutton on Insurance Law (Australia), New Zealand Insurance Law, Marine Insurance: A Legal History and The Law of Motor Insurance. He has also authored texts on arbitration and is on the arbitration panels of international arbitral bodies in Singapore, Hong Kong, Japan, China and New Zealand.

Contributors  xi Paul Myburgh is currently Professor of Law at the Auckland University of Technology (AUT) Law School and Adjunct Research Professor in the Centre for Maritime Law (CML) at the National University of Singapore (NUS). He was formerly Deputy Director of the CML and Associate Professor at NUS Law School. Paul previously held other faculty positions in Aotearoa New Zealand and South Africa. He also held visiting teaching positions and research fellowships at the TC Beirne School of Law, University of Queensland; the Nordisk Institutt for Sjoerett, University of Oslo; and City University of Hong Kong (as KH Koo Foundation Visiting Research Fellow at the Hong Kong Centre for Maritime and Transport Law). Paul is Chief Editor of the CML CMI Database on Judicial Decisions on International Conventions. He is the New Zealand correspondent for Lloyd’s Maritime and Commercial Law Quarterly, and sits on the editorial boards of the Journal for International Maritime Law and the Maritime Business Review. Livashnee Naidoo is Lecturer in Commercial Law at the School of Law, University of Glasgow. Dr Naidoo holds degrees from the University of Southampton (PhD, Commonwealth Scholar), the University of Cape Town (LLM in Shipping Law with distinction) and the University of KwaZulu-Natal (LLB cum laude). Prior to joining academia, Dr Naidoo practised as an attorney and notary of the High Court of South Africa in the areas of shipping and marine insurance law. Her research interests are primarily in insurance law (with a particular focus on commercial and marine insurance) and shipping law. She is a member of the British Insurance Law Association (BILA) Committee and is an Assistant Honorary Journal Editor of the BILA Journal. She has held visiting research and teaching positions at the University of Cambridge and the University of Cape Town. She lectures and publishes in various areas of insurance law and theory and has presented her research both nationally and internationally. Satoshi Nakaide (LLM (London), Dipl. (Cambridge), Dr (Waseda)) is Professor at Waseda University, Japan. He is also the Dean of the Research Institute of Business Administration. Satoshi worked for Tokio Marine Insurance Group for 28 years, including as the head of the legal department, before becoming a professor. His primary focus is the study of insurance law from a comparative standpoint of continental law and common law. He has authored and co-edited more than ten books on insurance law, earning three academic prizes, and has written numerous articles and book chapters on insurance law. Satoshi frequently travels to present his research. He currently serves as Chairman of the Marine Insurance WG of AIDA and as President of the International Academy of Financial Consumers (IAFICO). Kyriaki Noussia is a Greek Advocate (Solicitor and Barrister), Arbitrator and Mediator and an academic (Associate Professor in Law, School of Law, University of Reading). Her main expertise spans insurance, reinsurance, AI (regulatory aspects, data ethics), environmental law and dispute resolution. Prior to joining the University of Reading she held the position of senior lecturer at the University of Exeter, UK and has also worked as a lawyer, and in arbitration. She has an LLB from the University of Athens, Greece, an LLM from the University of Essex, UK and a PhD from the University of Southampton, UK. Richard Sarll is a barrister at 7 King’s Bench Walk and specialises in disputes relating to commercial and admiralty law. He is particularly well known for his work in shipping, commodities, insurance and reinsurance, energy and shipbuilding disputes. His contribution to this volume concerns the law of general average, a field in which he has built up considerable expertise. In 2012, Richard sat the associateship examinations of the Association of Average

xii  Research handbook on marine insurance law Adjusters, which he passed with distinction. In 2013, he joined a panel of British average adjusters that was convened to consider and report to the CMI upon the proposals which resulted in the York–Antwerp Rules 2016. Subsequently, he co-edited Lowndes & Rudolf, General Average and York–Antwerp Rules, 15th edition, with Richard Cornah and Joseph Shead. Currently, he sits on the GA sub-committee of the Association of Average Adjusters, a panel which occasionally considers topical GA issues. His reported cases include The Longchamp, a decision of the Supreme Court upon the principle of substituted expenses in general average. Richard is also a co-author of Carver on Charterparties, in which he authored the chapter on the termination of charters. Dieter Schwampe is a German lawyer with more than 38 years of practice. He specialises in shipping law and marine insurance law. He is an advisor to the German Insurance Association for hull and machinery insurance and a Professor of Law at Hamburg University, where he teaches marine insurance law. He is also President of the German Maritime Law Association, Vice-President of the Comité Maritime International (CMI) and a member of the Salvage Forum of the International Union of Marine Insurance (IUMI). He has published numerous articles and books on subjects of shipping law and marine insurance law, including commentaries on the standard German hull and machinery insurance conditions (DTV Hull Clauses and DTV-ADS). He is also the publisher of and main contributor to the 3rd edition of the main German legal commentary on transport insurance (Thume/Schwampe, Transportversicherung). Michael Sturley holds the Fannie Coplin Regents Chair in Law at the University of Texas Law School, where he teaches inter alia maritime law and commercial law courses and co-directs the Supreme Court Clinic. He received his undergraduate education at Yale, and has law degrees from Yale and Oxford. Prior to joining the Texas Law faculty, Professor Sturley was associated with Sullivan & Cromwell in New York. He also served as a law clerk to Justice Lewis F. Powell, Jr., at the Supreme Court of the United States, and to Judge Amalya L. Kearse of the United States Court of Appeals for the Second Circuit in New York. He has held visiting professorships at Queen Mary and Westfield College (now Queen Mary University London) and the National University of Singapore. Professor Sturley is a Titulary Member of the Comité Maritime International; a proctor member of the Maritime Law Association of the United States; the Senior Advisor on the US Delegation to Working Group III (Transport Law) of the United Nations Commission on International Trade Law (UNCITRAL); a member of the UNCITRAL Experts’ Group on Transport Law; and a life member of the American Law Institute. Rhidian Thomas is Professor Emeritus of Maritime Law and Founder Director of the Institute of International Shipping and Trade Law at Swansea University, Wales, UK. He is Editor-in-Chief of the Journal of International Maritime Law, a member of the Comité Maritime International (CMI) and the International Standing Committee on Marine Insurance, and an Honorary Member of the Croatian Maritime Law Association. His principal teaching and research interests are in the fields of Admiralty, maritime and shipping law, marine insurance law, international trade law and commercial dispute resolution. He has written, edited and contributed to many books and published widely in academic and professional journals on topics within his fields of interest. He is a frequent speaker at conferences and seminars, and also acts as an expert witness and consultant.

Contributors  xiii Feng Wang is Lecturer in Law and a member of the Institute of Maritime Law at Southampton Law School. His research interests are primarily in the fields of insurance law, marine insurance law, maritime law and law and technology. Apart from teaching undergraduate and postgraduate courses at Southampton, he is also actively involved in delivering training to both academic and industrial partners. Feng has published a number of book chapters and international journal articles. His monograph Illegality in Marine Insurance Law was published by Informa from Routledge in 2016. His journal article ‘Blockchain Bills of Lading and their Future Regulation’ was published by Lloyd’s Maritime and Commercial Law Quarterly in 2021. In addition, Feng was invited by the PRC Ministry of Transport to assist on the Reform of Chinese Maritime Law in 2018.

Introduction to Research Handbook on Marine Insurance Law Özlem Gürses

English marine insurance principles are codified under the Marine Insurance Act 1906 (MIA 1906), which reflects the customs of commerce and practice applied over centuries prior to the drafting of the Act. The common law cases that replicated such practices were codified in the Act, which did not aim either to dismiss or to modify the common law but attempted to bring certainty to marine insurance law principles. The Act did not aim to deprive judges of their discretion in the interpretation of the law, which, as a result, has continuously been developed through practices to date. After the adoption of the Insurance Act 2015, the marine insurance law principles are to be found in three different resources: the MIA 1906, the common law cases and the Insurance Act 2015. Inevitably, rapid developments in the use of technology in society have introduced new issues to be dealt with. A new risk will require a new type of insurance for it to be actively employed by people and businesses. The peculiarities and dynamics of the insurance and (non)availability of insurance products for emerging risks have profoundly impacted upon the regulatory nature of insurance. It is rarely the case that the insurance market is soft, with more insurance cover offers than demand. After a very brief period which coincided with the time at which the Insurance Act 2015 came into force, the insurance market returned to its hard self. The closure of the Suez Canal due to its accidental blockage by a container ship in 2021 illustrated the dependency of global supply chain management on the shipping industry. The COVID-19 pandemic has affected every person in the world, irrespective of their health or financial position. The losses suffered due to the pandemic emphasised the crucial role of the insurance industry not only in personal finances but also in the sustainability of national economies. The war between Russia and Ukraine has revisited many of the issues pertinent during the centuries in which marine insurance principles first evolved. With all the above-mentioned considerations in mind, this Research Handbook aims to present a broad coverage of current marine insurance-related issues. It includes discussions on the concept of marine insurance; some major general principles of marine insurance, including duty of good faith, total losses, insurers’ subrogation, choice of law, issues surrounding the direct right of action against insurers and general average; the insurance provisions in Incoterms® Rules; digitalisation of insurance documents; developments in the insurance law and practice in light of ESGs; and cyber and pandemic risks. These discussions are organised into six parts. Part I considers the concept of marine insurance and where marine insurance principles are to be placed in legal systems. Satoshi Nakaide examines the distinct legal significance of marine insurance law compared to general insurance law based on a comparative analysis of common law and continental law. While some may perceive it as relevant only to marine specialists, Satoshi argues that it embodies a unique legal principle for non-mass risk insurance 1

2  Research handbook on marine insurance law and that the study of marine insurance law is essential not only for marine experts but also for insurance lawyers handling other types of insurance, particularly those involving large risks. Rhidian Thomas contributes to this part with an analysis of the relationship between Protection and Indemnity (P&I) Clubs and their members. Thomas questions if such a relationship is truly insurance. He presents the issue as one of substance and not form, and examines it from the standpoints of the definition of insurance, the nature of the contract between P&I clubs and their members and the terms and conditions of the cover. Part II considers some controversial general principles of marine insurance law. Richard Aikens questions if, after the amendments made by the Insurance Act 2015, there remains some legally enforceable ‘principle’ or ‘obligation’ of utmost good faith between the parties to the contract. Aikens further examines, in the case of its existence, if such obligation operates both before and after the contract’s formation, and what legal consequences are attached to failure to observe this. In particular, he argues that the remedy of damages for breach should now be available. In the absence of such legal principle or obligation, Aikens asks if there is any point of retaining a statutory reference to utmost good faith at all; hence his allusion to the Cheshire Cat, which appeared and disappeared at will. Kyriaki Noussia and Yavuz Aslan examine the availability for insurers to subrogate into the assured’s rights to claim damages against the third party where both are co-assureds under the same policy. As well as addressing the English court cases on this matter, Noussia and Aslan take the matter a step further and include detailed assessments of BIMCO’s BARECON 89, cll 12 and 13 and BARECON 2017 cl17(a). Özlem Gürses addresses the innocent mortgagee’s position towards the insurers who either agreed to co-insure the mortgagee’s interest together with that of the shipowner, or issued a mortgagees’ interest insurance (MII) under a separate policy. The majority decision in Samuel v Dumas [1924] AC 431 appears to be the commonly accepted position in English law. The market has developed a solution with a clause inserted in MII policies; however, an innocent co-assured’s disadvantageous position where the shipowner wilfully gives damage to the insured vessel remains intact. Part III focuses on insurance coverage issues, types of loss which can be claimed under an insurance policy and general average losses. John Dunt examines the extent to which marine cargo insurance should be construed to include cover for financial loss unrelated to loss of or damage to cargo. The matter is particularly acute today in view of the expansion of cargo insurance to include storage risks for static cargo. After considering cases from both sides of the Atlantic, which establish that ‘clear words’ are needed to extend cargo insurance to pure financial loss, Dunt examines the application of this rule in a number of important recent English cases. Peter MacDonald Eggers KC presents a detailed analysis of total losses in marine insurance in reference to each element of actual and constructive total losses as regulated under the MIA 1906. MacDonald Eggers notes that the law of total losses under marine insurance policies has been developed with a remarkable degree of maturity and stability to develop solutions which take into account the respective commercial positions of the insurer and the assured, but some challenging issues still remain. Richard Sarll examines forwarding expenses incurred when cargo is transhipped onto a substitute vessel for carriage to destination, rather than being carried there aboard the vessel onto which it was originally loaded. Sarll examines whether such expenses can be claimed as general average and discusses the matter in the light of the York Antwerp Rules G and F. Part IV discusses a number of contemporary matters in relation to conflict of laws and marine insurance. Robert Merkin examines the enforcement of arbitration clauses in the rules

Introduction  3 of P&I Clubs as highlighted by the complex and ongoing Prestige litigation. The context is that a third party makes a direct claim against the shipowner’s P&I Club in a foreign court under that country’s cut-through legislation, to be met by an application by the Club in England for an anti-suit injunction to preclude a breach of the arbitration clause and of the pay to be paid clause in the Club’s rules. Merkin’s chapter deals also with the complex issues that arise where the third party is a state, giving rise to issues of sovereign immunity. Paul Myburgh addresses third party victims’ rights of direct action (TRDAs) against insurers through a comparative conflict of laws lens. Analysing how Courts have taxonomised the ‘peculiar world’ of TRDAs brought against P&I Clubs, Myburgh emphasises the lack of uniformity in this respect and discusses if there is a way forward to unify or at least better harmonise outcomes and proposes a solution to it. Michael Sturley focuses on the choice between federal law and state law to govern a marine insurance contract. Sturley addresses which body of US law governs when US law applies. Sturley approaches the matter from three different angles: (1) the ‘horizontal’ choice between US law and the law of some other country; (2) if US law governs, the ‘vertical’ choice between maritime law, which is federal, and the law of a US state with an interest in the resolution of the dispute; and (3) if US state law governs, the ‘horizontal’ choice among the possibly relevant states. Part V examines the insurance obligation under Incoterms® Rules and the digitalisation of insurance documents in international trade. Margarida Lima Rego focuses on insurance obligations as set out in CIF and CIP clauses under Incoterms® 2010 and 2020. Lima Rego addresses the concept of the passing of risk in sale of goods, the history and legal nature of the Incoterms® Rules and the implications of the amendments introduced by Incoterms® 2020 with respect to the insurance requirements. Miriam Goldby, while addressing the need for harnessing the new technologies to digitalise cargo insurance documentation, acknowledges the challenges inherent in the current legal framework governing this documentation and its use in transactions. Goldby proposes how such challenges could be overcome and how the UK Electronic Trade Documents Act supports this. Part VI includes new technologies, emerging risks and how the marine insurance industry can tackle them. Gary Meggitt’s chapter analyses the issues arising from the encounter between marine insurance fraud and emerging technology. Meggitt’s assessment of the matter includes the nature and development of ‘InsurTech’, and addresses how current UK law and regulation deal with marine insurance fraud. It also discusses how the former could be employed in the insurers’ fight against fraud, which is currently thriving. Feng Wang explores the role of insurance as an effective risk management measure that can be taken against cyber risk in the shipping industry. Wang addresses the challenges faced in tackling cyber risks through insurance and presents proposals to provide a balanced solution for the parties involved in cyber insurance. Wang highlights the significance of obtaining trustworthy data from clients, the fragmentation of cyber insurance policies and the cultivation of cyber awareness in the maritime industry. Livashnee Naidoo’s focus is environmental, social and governance risks (‘ESG risks’) in marine insurance and she questions whether the Poseidon Principles for Marine Insurance (the ‘PPMI’) is an effective blueprint for the marine insurance industry to mitigate climate change. Naidoo examines the objectives of the PPMI and how it relates to the specificities and the vision of the marine insurance market. She identifies the primary legal and regulatory concerns relating to the PPMI and assesses the regulatory dimension of it. Dieter Schwampe analyses the pandemic-related clauses that the German Insurance Association released in 2021. The

4  Research handbook on marine insurance law Clause for the Exclusion of Loss/Damage due to a Dangerous Communicable Disease for the Use in Marine Insurance (‘Pandemic Exclusion Clause’) was designed for use in all segments of the transport insurance market. The clause comes with various write-back possibilities. Schwampe examines closely the meaning of each wording in this clause, and how this clause would operate under the general principles of marine insurance law.

PART I  

1. Marine insurance law as a treasure trove of legal principles1 Satoshi Nakaide

1. INTRODUCTION Marine insurance is indispensable for the maritime and trade industries, and a sound understanding of marine insurance law is essential for those who handle such insurance. Marine insurance claims can result in substantial losses, and the policies involved can be intricate, giving rise to a multitude of legal issues. Given this, the study of marine insurance law is imperative for practical purposes. Nevertheless, there is some question among insurance lawyers regarding the relevance and value of studying marine insurance law. Marine insurance is closely linked with international maritime law, trade and shipping. Furthermore, it is becoming increasingly sophisticated as new technologies are developed and the trade and maritime industries evolve. Consequently, marine insurance law is becoming a more specialised field, leading to the perception that it is only relevant to maritime and trade law professionals, with less relevance for scholars in the broader field of insurance law. Throughout history, the insurance industry has evolved to include various types of commercial insurance; marine insurance emerged in the late fourteenth century, followed by fire insurance in the seventeenth century and modern life insurance and various types of casualty insurance in the eighteenth century. Since then, the insurance industry has exploded in volume worldwide, but marine insurance currently accounts for a very small share of all insurance by volume. Indeed, in Japan, marine insurance accounts for around 3 per cent of non-life insurance premiums, including land cargo transportation insurance.2 Accompanying the growth of the insurance industry on personal lines has been growth in insurance disputes between sellers and consumers, and in response, many countries have revised their insurance laws to enhance consumer protection. In contrast, governments have treated marine insurance law as an area of contractual freedom. Marine insurance law tends to be handled mainly by specialists in the field, which has raised concerns about the decline in scholarly research in this area, particularly internationally.3 The United Kingdom (UK) stands out as a hub for global marine insurance; for instance, London is a frequent site of marine insurance litigation or arbitration. Indeed, abundant scholarly research on marine insurance has made it a well-established field of study in the UK. The significance of the UK’s role in the marine insurance market can largely be attributed to

This work was supported by JSPS KAKENHI Grant Number 19H01430. The General Insurance Association of Japan, Fact Book 2021–2022. The direct premium of marine insurance including land-transported cargo written in Fiscal Year 2021 was 3.20 per cent. 3 For example, in Japan there used to be many researchers in marine insurance law, but now there are very few. 1 2

6

Marine insurance law as a treasure trove of legal principles  7 London’s broader insurance industry. In contrast, in other countries the proportion of marine insurance has decreased as a share of the broader industry, and it appears that there is less and less scholarly interest in marine insurance law in these jurisdictions. Among the major countries involved in marine insurance, Germany has played a significant role. Its Commercial Code includes provisions concerning marine insurance, and extensive research on marine insurance law in Germany has had a significant impact on continental legal scholars. However, the 2008 Insurance Contract Act (VVG: Versicherungsvertaragesgesetz) excluded marine insurance (VVG section 209), and Germany’s Commercial Code abolished all marine insurance provisions.4 Nonetheless, since marine insurance policies are still contracts, the Civil Code applies. The absence of specific marine insurance provisions in law may raise questions about the future of legal research on marine insurance in Germany. However, it is worth noting that general insurance clauses on marine insurance are still of practical interest in Germany and have been the subject of detailed books by practitioners. Japan’s Commercial Code of 1899 contains provisions on insurance contracts in the chapter on commercial transactions and provisions on marine insurance contracts in the chapter on maritime commerce, respectively. These provisions have remained unchanged for more than a century. In 2008, provisions on insurance contracts were fundamentally revised in light of current practices and were made independent from the Commercial Code and enacted as the Insurance Act,5 which came into force in 2010. The Insurance Act was enacted as a comprehensive law on insurance contracts of any type including marine insurance contracts. The next task was to revise the maritime commercial law section of the Commercial Code, but the question was whether special provisions on marine insurance were still necessary given the Insurance Act. Clearly, marine insurance law addresses risks associated with maritime activities. However, the questions are more fundamental. What are marine insurance laws and how do they differ from general insurance laws? Even if they have differences, do they possess any special significance to be contained in the Commercial Code? In this chapter, I examine the significance of marine insurance law in the study of insurance law, focusing on the Commercial Code of Japan’s 2018 amendments to its marine insurance provisions in comparison and contrast with the Insurance Act. For this purpose, I first provide an overview of the general structure of Japanese insurance law including marine insurance. Next, I discuss the amendments to the Commercial Code, analysing the important marine insurance provisions. Based on these discussions, I consider the importance of studying marine insurance law.

2.

JAPANESE LAW APPLICABLE TO MARINE INSURANCE

Japan is a country of civil law guided by statutory law, and court judgments are not regarded as law.6 The most important laws relating to insurance are the Insurance Business Act (IBA),

Robert Koch, Insurance Law in Germany (Wolters Kluwer, 2018), p.191. Law No. 56 of 2008. 6 For English literature on Japan’s legal system, see Hiroshi Oda, Japanese Law (4th ed. Oxford University Press, 2021). 4 5

8  Research handbook on marine insurance law enacted in 1939, and the Insurance Act of 2008.7 The IBA is a regulatory law that regulates all aspects of insurance including licensing, sales, accounting and government supervision. The marine insurance business is conducted in Japan under the licence of the Ministry. The Insurance Act is a law on insurance contracts including marine insurance as a special law of the Civil Code. In regard to the marine insurance contracts, the Commercial Code’s maritime provisions on insurance also apply, as explained above. The IBA is a legislation with enforcement orders, enforcement regulations and notifications from supervisory agencies that specify detailed rules, and these regulations are revised almost every year to respond to changes in the market and business activities. The Insurance Act and the Commercial Code, meanwhile, serve as basic laws and provide fundamental rules that are intended to remain in place for an extended period, forming an essential legal framework of principles for the law and remaining stable over time. Looking at the marine insurance business, the IBA, under Article 186, Paragraph 1, prohibits direct overseas insurance; instead, assets located in Japan can only be insured by companies located in Japan or with established branches there.8 However, overseas hull insurance, overseas cargo insurance and certain other specific types of insurance are exceptions to this prohibition.9 Therefore, for overseas ocean-going vessels and cargoes, Japanese shipowners and shippers/consignees may insure their property either with a Japanese company or directly with a foreign company. For the insurance of domestic hull, and cargo transported only in Japan, policies and certificates written in Japanese are used. These contracts are governed by Japanese law, and their general conditions are in line with those of land-based insurance; some companies use the same insurance policy form for marine and land transportation. In the case of overseas hull insurance, Japanese policies written in Japanese,10 Japanese policies translated into English or the Institute Time Clauses (ITC 1983) in London11 are used for Japanese shipowners,12 and for cargo for export or import, the Institute Cargo Clauses (ICC) in the 2009 or 1983 versions13 are widely used. When Japanese insurance companies use the ITC or ICC, they add ‘Law and Jurisdiction Clauses’ stating that English law and practices govern payment responsibility and

7 For a short explanation of law on insurance in Japan, see Satoshi Nakaide, ‘Marine Insurance Law in Japan: A Structure Based on a Combination of Civil Law and English Marine Policy Wordings’, The Journal of Business Law, issue 5 (2015), 416. 8 IBA, Article 186, Paragraph 1. 9 Article 19 of the IBA’s Enforcement Regulations. 10 This is used where the policyholder and assured are all Japanese. 11 ITC 1/10/83. The author of this chapter understands that Japanese insurers are not using ITC 1995 version or International Hull Clauses (01/11/03). 12 The Japanese shipowner is allowed to insure its vessel navigating overseas with a foreign insurer. A foreign insurer may use other policy forms, such as the Nordic Plan. 13 In Japan, the ICC（1/10/1983）was used for a long time after the 2009 revision of the ICC, but recently the 2009 Form (1/1/09) has been in common use.

Marine insurance law as a treasure trove of legal principles  9 settlement whereas Japanese law governs the validity and legality of the insurance contract.14 This clause creates a hybrid contract combining English law and Japanese law.15

3.

REVISION OF INSURANCE LAW IN JAPAN WITH RESPECT TO MARINE INSURANCE CONTRACTS

As noted earlier, Japan’s Commercial Code was promulgated in 1899, notably modelled after Germany’s Commercial Code. The 1899 Commercial Code contained provisions on insurance in its third chapter and on marine insurance in its fourth chapter. The former was modernised as the Insurance Act 2008, superseding the provisions of the Commercial Code. Insurance contracts can also be executed by mutual aid organisations, such as mutual insurance societies. The new Insurance Act also applies to such insurance as well as to P&I insurance by the mutual club. The Act includes unilateral mandatory provisions from the perspective of consumer protection which invalidate any changes from the rules of the Insurance Act that are disadvantageous to policyholders and assured.16 However, for insurance contracts that cover business risks, such as marine insurance, freedom of contract is permitted and any changes from the Insurance Act are allowed.17 After the enactment of the Insurance Act in 2008, the government launched a revision of the marine insurance provisions in the Commercial Code. The first question was whether to abolish the provisions on marine insurance in the Commercial Code or not. Looking to Germany, as seen before, the 2008 amendment to the Insurance Contract Act (VVG) fundamentally revised the law, resulting in the exclusion of marine insurance and reinsurance from its application. In Japan, unlike Germany, the Insurance Act also applies to marine insurance contracts. Therefore, there is at least a legal basis for the contract of marine insurance. Furthermore, all provisions concerning marine insurance in the Commercial Code are optional rules and marine insurance is an insurance contract between professionals, and it is extremely rare that marine insurance disputes are resolved by a court; for that reason, many did not recognise a practical need for any statutory provisions. Some even expressed concern that provisions in the Code for marine insurance might be unduly restricting marine insurance practices.

Its wording differs between cargo and hull insurance and also varies among insurers. As to the jurisdiction, hull policy states the jurisdiction of Tokyo district court, while cargo policy does not provide for jurisdiction and leaves it to the courts of the State in which the action is brought. As an example, Tokio Marine and Nichido Fire Insurance Co’s wordings for cargo marine insurance, revised in 2019, are: (1) Subject to the below provisions, this insurance contract is governed by Japanese law. (2) Notwithstanding anything contained herein or attached to the contrary, English law and practice shall apply to only the interpretation of policy terms, liability and settlement of any and all insurance claims. (3) For the avoidance of doubt, matters relating to the existence and validity of the insurance contract and the duty of disclosure and any remedy available in case of breach of that duty are to be addressed by reference to Japanese law, as per clause 1 above. 15 Nakaide, p.421. 16 Japanese insurance law, like continental law, distinguishes between the concepts of policyholder and assured. The former is the contracting party who enters into an insurance contract and has contractual obligations and rights, while the latter is the person in a position to suffer loss due to an accident in the case of non-life insurance, or the person whose life is insured in the case of life insurance. 17 As its exception, some provisions in the Act such as the existence of insurable interest and the statute of limitations are regarded as mandatory to marine insurance as well. 14

10  Research handbook on marine insurance law The recommendations of a private study group on marine insurance legislation, which was established by the General Insurance Institute of Japan, had an important influence on the revision of the Commercial Code.18 The study group consisted of legal scholars and insurance experts who extensively examined the state of the law and practice on marine insurance contracts. Their recommendations to revise the law were published and proposed that articles on marine insurance in the Commercial Code should be retained and modernised to conform to current practice.19 The group based its recommendations on a comprehensive set of reasons, including the following: first, many of the major shipping countries have laws that govern marine insurance contracts, and if Japan had no such provisions, the rest of the world would have no information on Japan’s laws on them; second, the provisions on marine insurance in the Commercial Code are important in that they provide for rules on corporate insurance which are different from the Insurance Act, which is focused on consumer protection. Of course, the Commercial Code provisions on marine insurance do not directly apply to other types of corporate insurance, but it is significant that there are points of reference. Additional recommendations of the study group included: (a)

It is desirable to have provisions in harmony with English law and practice, as that is the global standard in marine insurance. (b) Because English Institute clauses are widely used in overseas cargo and global hull insurance and Japanese insurers add the English Law Jurisdiction Clause, provisions should be made in the Commercial Code to enhance the legal stability of such hybrid jurisdiction and governing law contracts. The Study Group also recommended specific revisions to each of the provisions on marine insurance in the Commercial Code to make them consistent with current practice. In the government’s revision of the Commercial Code, the provisions on marine insurance were retained and modernised as necessary. However, due to the Commercial Code’s nature as a code of general principles, the number of articles was minimised. The study group’s recommendation to revise the law by referring to English law and aligning Japanese law with English law, as well as establishing a provision regarding the governing law, was not adopted. In the writer’s view, this was because marine insurance is less formal than other categories and does not require the same level of strict governance as under the Insurance Act. The government also considered that hybrid contracts were governed by provisions of legislation other than the Commercial Code. As a result, the 2018 revised Commercial Code has only 16 articles pertaining to marine insurance, down from the previous 28:20 ● Definitions: Article 815 ● Insurer’s Liability to Compensate: Article 816, 817 18 The head of the group is Ochiai Seiichi, Emeritus Professor of Tokyo University. The group had 17 members, including the writer of this chapter. 19 Study Group of Marine Insurance Legislation, ‘Study Report: Recommendations for the state of marine insurance legislation in Japan: based on standard marine insurance practice’, Songaihoken-kenkyu, Vol. 75-4(2014), p.311 (in Japanese). 20 This chapter relies on the Japanese government’s translation of the Commercial Code: see Commercial Code (Tentative translation) Act No. 48 of March 9, 1899. https://​www​.ja​paneselawt​ ranslation​.go​.jp/​ja/​laws/​view/​4020

Marine insurance law as a treasure trove of legal principles  11 ● ● ● ● ● ● ● ● ● ● ● ● ●

Insured Value of Ship Insurance: Article 818 Insured Value of Cargo Insurance: Article 819 Obligation of Disclosure: Article 820 Information Contained in Document to Be Delivered upon Conclusion of Policy: Article 821 Change of Voyage: Article 822 Significant Increase of Risk: Article 823 Change of Ship: Article 824 Provisional Insurance: Article 825 Exemption of Insurer from Liability: Article 826 Liability to Compensate Where Cargo Is Damaged: Article 827 Liability to Compensate Where Cargo Is Sold Due to Force Majeure: Article 828 Cancellation by Reason of Nondisclosure: Article 829 Application Mutatis Mutandis to Mutual Insurance: Article 830

4.

DIFFERENCES BETWEEN MARINE INSURANCE LAW AND OTHER INSURANCE LAW

Next, I consider the differences by referring to the provisions of the Japanese Commercial Code. 4.1

Duty of Disclosure

The global trend in insurance contract law is to reconsider rules from a consumer protection perspective. In many countries, the review of insurance contract law has progressed from the perspective of enhancing consumer protection. Among various rules and principles in insurance law, the duty of disclosure is unique to insurance contracts. Traditionally, this requirement meant that the policyholder21 had to disclose important facts voluntarily even if not asked, and the position under the Japanese Commercial Code was the same.22 The Insurance Act changed it to an obligation to disclose important facts about risks that the insurer requested to be disclosed (Article 4). This means that the prospecting policyholder does not need to disclose important facts if it is not asked to disclose by the insurer. This duty is a duty to answer questions on important matters properly. For insurance of personal lines, normally, insurers require policyholders to complete questionnaires to disclose facts. If the questions are honestly answered, there is no violation of duty.23 This rule is a unilateral mandatory and any provision in the policy disadvantageous to the policyholder than the provision of the Act is null and void.

21 Hereinafter policyholder means the contractual party to insurance and/or the assured under Japanese law. 22 Satoshi Nakaide, Pre-contractual Duties under the Japanese Insurance Law, contained in Yong Qiang Han and Greg Pynt eds, Carter v Boehm and Pre-Contractual Duties in Insurance Law: A Global Perspective after 250 Years (Hart Publishing, 2018), p.293. 23 Article 28(1) provides that an insurer may cancel a non-life insurance policy when a policyholder or an insured has, intentionally or by gross negligence, failed to disclose facts or disclosed false facts regarding matters to be disclosed. The scope is limited to intentional or grossly negligent acts only. Mere

12  Research handbook on marine insurance law However, the subject-matter in marine insurance and their location are diverse, and it is not possible for insurers to ask every possible relevant question; clearly, the new rule does not fit marine insurance. Because there is no unilateral mandatory effect in marine insurance contracts, it is possible to agree on disclosure obligations different from those in the Insurance Act. However, in practice, it is difficult for marine insurers to agree with customers on an obligation stringent than that of the Insurance Act. Marine insurers are left to explain to customers why this is the case. Obviously, this process incurs additional transaction costs with a high possibility of the customer disagreeing with the insurer and the contract being lost to the competitor. It must also be noted that the duty to disclose is an obligation applicable before the contract concludes, and there are practical difficulties and uncertainty in agreeing on a duty different from the default rule. Considering this situation, the study group as well as insurers proposed maintaining the Commercial Code’s principles of voluntary disclosure. This proposal was supported by the government, and in contrast to the Insurance Act, the new Commercial Code stipulates that for marine insurance, the disclosure of material facts is the obligation of the policyholder (Article 820). Although this provision is discretionary, it is significant in that it is now the default rule for marine insurance. The duty of disclosure gives the insurer the right to cancel the contract in the case of a breach.24 Although the right is strong, Japanese insurers seldom take advantage of this remedy. Insurers may use this remedy where there is a high suspicion of deliberate loss or damage caused by the assured and there is also a breach of duty of disclosure. The marine insurance business is based on long-term customer relationships, and the termination of a particular policy by the insurer has serious consequences that could end the business altogether. The theoretical ground for the duty of disclosure is sometimes explained from the utmost good faith. However, in the writer’s view, in the case of marine insurance, the duty of disclosure is better to be explained from the underwriting efficiency in setting proper conditions including premium rate. Then, a question does arise of whether the voluntary disclosure obligation is specific to marine insurance. In response, I argue later in this chapter that it can apply to other large-scale corporate insurance as well.

negligence does not constitute a breach of duty, which differs from UK law and many countries in the EU. 24 Article 28(1) gives the insurer a right to cancel the contract. The cancellation is only for the future and does not have a retroactive effect (Article 31(1)). However, the insurer is not liable to compensate for the damage arising from an insured event that occurred before the policy is cancelled, except for damage that occurs independently of the facts not disclosed (Article 31(2)). This means that if the contract was once valid, the insurer is allowed to retain the premium for the period before the cancellation. There must be some causal link between the accident covered by insurance and the fact of non-disclosure. As an illustration, if the insured, a cancer patient, lies in response to the insurer’s questions, the insurer can cancel the policy for breach of duty of disclosure. However, if the insured dies in a traffic accident before the cancellation, the insurer must pay the claim as far as there is no causal link between cancer and the traffic accident.

Marine insurance law as a treasure trove of legal principles  13 4.2

Insurable Interest

While in the UK the Marine Insurance Act 1906 (hereinafter referred to as the MIA) provides for insurable interest, including its definition,25 in Japan, the term ‘insurable interest’ itself does not exist in the legal text. It is a concept used in academic studies as well as in insurance practice. However, the Insurance Act stipulates that for non-life insurance contracts, ‘only benefits that can be estimated in monetary terms can be the purpose of the insurance contract’. It is construed that a concept equivalent to insurable interest is required by this provision. So far, there is no academic consensus on how to understand and define the concept of insurable interest in Japan; in particular, there is controversy over the concept of interest in liability and expense insurance. Nonetheless, the tendency is to recognise insurable interest in a flexible and broad manner. For life insurance, insurable interest is not required, but life insurance on death requires the consent of the insured person, which is the same as in many continental laws. The Commercial Code also does not include any provisions on insurable interest, although Article 815(1) recognises marine insurance as ‘insurance that insures the subject-matter of the insurance or the interest of the insured person in the subject matter’. Thus, the concept of insurable interest is perceived in law. The parties to marine insurance policies may have multiple types of insurable interest depending on the situation. For example, hull insurance can involve parties such as the registered owner of the vessel (especially with respect to flags of convenience), bareboat charterer, lessees, time charterer, management companies, crewing agencies and financiers, who might all have varying interest for losses that can depend on their relationship with the vessel. Parties liable for civil obligations in particular vary according to different countries’ laws. In international trade, goods are transferred from the exporter to the importer, but the transfer of risk bearer and transfer of ownership does not necessarily occur simultaneously. In addition, prior to payment, the bank can have a right to the cargo. Insurable interests are based on the complex relationship between interested parties. Particularly, in cargo insurance, if we analyse the legal nature of insurable interest, we recognise interests that are contingent or defeasible in nature.26 Thus, marine insurance deals with many types and instances of insurable interest. Hence, a detailed study of insurable interest in marine insurance is important for understanding insurable interest and applying the concept to other insurance fields. Specifically, the research will be useful for insurance on a joint venture involving multiple parties. 4.3

Provisional Insurance

Provisional insurance is an option when some of the important elements of a contract are not yet finalised; this system is primarily used with marine cargo insurance, both for individual shipments and for comprehensive agreements known as open cover. By recognising the validity of such contracts, the parties involved enjoy their respective contractual rights and obligations. For instance, if an accident occurs before the insurer is notified of the details, the insurer is still responsible for providing coverage in accordance with the agreed-upon terms and conditions, and if the insurer receives notice of the confirmed facts after the risk has ended 25 26

MIA ss.4–15. MIA s.7.

14  Research handbook on marine insurance law without an accident, the policy will not be void as retroactive insurance, and the insurer is entitled to receive its premiums.27 Article 6 of Japan’s Insurance Act specifies the information that must be included in an insurance policy, and similarly, the Commercial Code outlines the information that must be included in marine insurance policies. For marine insurance contracts, the following information should be stated in the policy: insurance period, insured property, insurable value, insured amount, insurance premiums or the payment method thereof, name of the ship or place of shipment, port of loading, port of discharge or place of arrival of the cargo, and so on. The Insurance Act does not contain any provisions which allow for forming policies without finalising certain terms. Moreover, it includes provisions that render retroactive insurance contracts invalid (Article 5). As a result, there is legal uncertainty regarding provisional insurance. The previous Commercial Code on marine insurance did address provisional insurance, and the revised Code has ensured the legal stability of such contracts with the expansion of its provisions to include an open cover and the establishment of requirements for forming provisional insurance contracts under Article 825. This article also requires the assured to promptly notify the insurer of finalised terms and specifies that intentional or grossly negligent delays in providing such notification will invalidate the contract. While provisional insurance is frequently used in marine cargo insurance for import and export purposes, there is also a need for it in other types of corporate insurance policies. Therefore, the legal provisions regarding provisional insurance have broader significance beyond marine insurance. 4.4

Insurance Period

Most casualty insurance policies have a fixed duration, such as one year, but marine insurance frequently incorporates voyage insurance policies under which the insured period depends on the length of the voyage. The Japanese Insurance Act requires the insurance period as an item to be included in the document that must be provided at the time of the conclusion of the insurance contract, usually the insurance policy.28 There is no provision showing a definition and any rule on the insurance period in the Insurance Act; rather, it is considered something the contracting parties would agree on and build into the policy. On the other hand, the Commercial Code on marine insurance did address the insured period for voyage and cargo insurance as well as cargo profit insurance. These articles were all abolished in the 2018 revision. During the discussions on revising the Commercial Code, the study group proposed retaining the voyage policy provision, and there was strong support from the industry as well. However, this proposal was rejected for the following reasons. Firstly, marine insurance need not always take the form of voyage insurance; there are instances, such as a time policy, where the insurance period is fixed at a specific date without invoking the concept of a voyage. Secondly, certain land-based insurance policies, like construction insurance, determine the coverage period based on the construction timeline, which may vary depending on the construction’s status. Thus, the existence of a voyage policy in marine insurance doesn’t alter the

There are various types such as floating policy, obligatory open cover and facultative obligatory cover. See Özlem Gürses, Marine Insurance Law (3rd ed. Routledge, 2023), pp.18–21. 28 Insurance Act, Article 6. 27

Marine insurance law as a treasure trove of legal principles  15 legal nature of the insurance period. It was concluded that the insurance period is a matter to be specified in the contract, and there is no need to establish special provisions in the law. The reason the marine insurance industry argued for retaining the provisions regarding the insured period was to contrast the default of insurance coverage continuing regardless of the length of time. In particular, in marine insurance, there is an important aspect of voyage policy not just showing the period, such as treating a voyage as a unit of coverage. This leads to several important effects: First, under the standard time policy, if it is terminated during the insurance period, the unexpired premium calculated on the duration remaining on the policy is refundable. In voyage insurance, since the policy covers one voyage as a unit, the pro-rata reimbursement method is not appropriate. Provisions on voyage insurance in the Commercial Code supported this reasoning. Second is the change of risk. Because voyage policy covers a voyage as a unit, a change to a voyage means a change from the original risk. Third, loss of voyage means a loss of the insured interest, allowing for a claim for it. Fourth, by using the concept of a voyage policy, the land transport part of multimodal transport will also be recognised as part of the marine insurance contract and is governed by the Commercial Code. Although marine insurance is a type of insurance contract that covers risks in navigation, through the concept of a voyage policy, the scope of marine insurance contracts will be expanded to include the land transportation portion as well.29 For these reasons, the concept of voyage policy has various importance for insurance law. Historically, the concept of voyage policy dates back to the early days of marine insurance, where insurance was typically offered on a per-voyage basis. From the perspective of the insurance period, a voyage policy can be seen as one of the ways of setting an insurance period. However, what is important to note is that the traditional approach in marine insurance has been to ensure ‘voyage’, which is linked to legal theories such as alteration of risk, insurable event, loss recognition and return of premium. The author finds it a missed opportunity that the proposal of the insurance industry was rejected by the Japanese government – as mentioned above – and provisions on voyage policy were all abolished. Nevertheless, the Commercial Code did not entirely remove references to voyage insurance. Article 821 (‘Matters to be included in documents to be delivered at the time of conclusion of a contract’) states that one of the items that must be included in a Ship Insurance Policy is ‘navigation area (If a Ship Insurance Policy is concluded for a single voyage, the port of departure and the port of arrival (including a port of call if any is specified)’. The expression ‘Policy is concluded for a single voyage’ indicates that a method of voyage insurance is recognised in the Commercial Code. The author submits that arguments by the industry were not just a matter of insurance period, but were also related to cargo insurance covering ‘voyage’.30 Their attempt, however, proved fruitless, with the government’s refusal to maintain or create a provision for voyage policy. The revisions to the Code emphasised modernising existing provisions based on the minimum necessary changes. While the significance of insuring ‘voyage’ is a particularly important area of study in marine insurance, it is clear that its significance is not limited to this area alone.

MIA section 2 ‘Mixed sea and land risks’ makes it clear. The question is: what does the cargo insurance cover? Does it cover the cargo or the voyage of the cargo or both? 29 30

16  Research handbook on marine insurance law 4.5

Coverage and Exclusion: Principles and Exceptions

4.5.1 Principles The Japanese Commercial Code provides in principle that marine insurance shall broadly cover accidents related to navigation. Article 816, ‘Insurer’s Liability to Compensate’, states: ‘Unless otherwise provided for in this Chapter or in a Marine Insurance Policy, an insurer is liable to compensate for any and all loss or damage arising from accidents relating to voyages that occur to the insured property during the insurance period.’ Article 817(1) states: ‘An insurer is liable to compensate for the amount to be paid by the assured to engage in marine salvage or share a general average.’ This approach is known as the comprehensive liability principle in marine insurance, which is in line with continental law. The marine insurance policies from the fourteenth century, when marine insurance was created, covered a wide range of perils related to navigation.31 It was indeed a comprehensive coverage of risks. Over time, transportation has become increasingly intricate, the items to be insured more sophisticated and insurance itself more complex. Marine insurance, which comprehensively covers risks associated with navigation, now requires specific clarification on the circumstances under which risks are not covered. As a result, various exclusions have been added, further complicating the structure of coverage. While other insurance policies also feature a range of exclusions, they differ in several key ways. For instance, fire insurance solely covers certain perils, such as fire and explosions, meaning that the exclusions are originally limited to those on these risks. Consequently, the types and number of exclusions are limited in nature and in number. Adding to the above, risks covered by marine insurance are very diverse and are different in each vessel and cargo in question. This makes it necessary to tailor the coverage to each risk for the cover. Various exclusions are necessary to control the risk for the insurer to cover. Marine insurance policies can become detailed and complex, which makes it reasonable that legislation does not attempt to incorporate detailed coverage rules into law but instead provides merely the basic principles that serve as the backbone for the laws. 4.5.2 Exclusions In Japan, the various exclusions stated in the previous Commercial Code on general insurance were simplified in the Insurance Act 2008 and the Act addresses only two exclusions: ‘loss or damage arising from the intention or gross negligence of the policyholder or the assured (in the case of a liability insurance policy, their intention)’32 and ‘loss or damage arising from a war or any other social disturbance’ (Article 17). Regarding what is covered, the Insurance Act defines policies of general insurance as contractual agreements to compensate for damage caused by accidents (Article 2). From this definition, general insurance covers only ‘accidental events’. All other matters are left to be determined in the policy.

Dr Kimura displays the photocopy and its translation of the Insurance Policy of Pisa of 11 July 1385. The policy states that it covers all risks, accidents and misfortune of any type of God, sea and human. Eiichi Kimura, Lloyd’s Insurance Policy seisei-shi (History of the development of Lloyd’s SG Policy) (in Japanese) (Kaibundo, 1979), p.4. 32 This means that liability insurance does not exclude gross negligence, thus safeguarding victims from the gross negligence of the assured. There is no statutory definition of gross negligence in Japanese law. 31

Marine insurance law as a treasure trove of legal principles  17 In contrast with the Insurance Act, Article 826 of the Commercial Code adds the exclusions of:33 ‘(i) loss or damage arising from the nature of or a defect in the insured property, or from ordinary wear and tear thereof’; ‘(iv) in the case of a Ship Insurance Policy, loss or damage arising from the failure to satisfy the requirements at the time of departure of the Ship’;34 (v) ‘in the case of a Cargo Insurance Policy, loss or damage arising from the insufficient packaging of cargo’. During the legislature’s discussions on revising the Commercial Code, cargo insurers argued that delay should be explicitly stated as an excluded risk in marine insurance, but this did not happen: the Japanese standard cargo policy as well as the UK’s ICC exclude delay from its coverage in cargo insurance. Section 55 2(b) of the UK’s MIA 1906 also explicitly excludes delay. English lawyers might wonder why Japan did not include delay in its marine insurance exclusion provisions. The author’s view is that it is because of the complex nature of the concept of ‘delay’ in shipping and its relationship to marine insurance loss and damage coverage. For instance, Christmas products might lose their commercial value if a vendor does not receive them before Christmas, but the loss in value does not arise from damage to the insured property, and therefore, it is not covered. Rust of machinery and food spoilage caused by delay are also exempt from liability because there could have been inherent vice and/or nature of the cargo that rusted or the food that spoiled. Additionally, it is difficult to establish legally what length of delay should be covered versus excluded and who should be held responsible for the delay. A delay is a complex event, and the Commercial Code expects to stand for long periods of time. Therefore, it is a challenge to accommodate time-based events such as delays into concrete law; this remains a research topic for the future. Separately, research on marine insurance such as this study could also be significant in corporate insurance, in particular property insurance. 4.6

Alteration of Risk and Warranty

The risks covered by marine insurance are diverse and can involve significant sums of money. Additionally, insurance cover spreads across a geographically wide range of the world and fluctuations in risk may arise beyond the assumptions upon which the insurance is based. While the volatility of risks exists also in consumer contracts, it is particularly problematic in marine insurance because of its characteristics. Insurers pool their risk across their entire portfolios of policyholders, in this context limiting the risk borne by the insurer to a certain extent. Such limitation is not only in the interest of the insurer but also beneficial to the policyholders because it increases the likelihood of acceptance of the insurance and keeps premiums reasonable. If the insurer were to bear the increased risks, the insurer would have to add the premium for such risks.

Subsection (ii) is exclusions for intention or gross negligence and subsection (iii) is exclusions for war or any other social disturbance. These are the same as the exclusions of the Insurance Act. 34 The requirements are stated in Article 739. This exclusion does not apply to the loss or damage if the policyholder or the assured proves that they did not neglect to exercise due care in satisfying the requirements prescribed in that item at the time of departure of the Ship (Article 826(1)). 33

18  Research handbook on marine insurance law In Japan, the Insurance Act provides for only two provisions concerning changes in the risk in insurance contracts: Article 11 gives the policyholder the right to request a reduction in premiums for future periods in the event of a significant decrease in risk, and Article 29 grants the insurer the right to terminate the contract for future periods in the event of an increase in risk beyond certain important facts that were subject to the duty of disclosure.35 However, in the context of marine insurance, significant risk can always increase beyond the scope of facts subject to disclosure. The Commercial Code on marine insurance now addresses changes in risk based on whether the changes occurred before or after the commencement of risk and whether the assured was at fault, and specifies legal remedies such as exempting an insurer from liability for the increased risk or for terminating the contract because of the increase. That is, the Commercial Code on marine insurance widens the legal effects and remedies as follows:36 Article 822(1): If there is a change to a voyage before the commencement of the insurance period, the Marine Insurance Policy ceases to be valid. (2) If there is a change to a voyage during the insurance period, the insurer is not liable to compensate for loss or damage arising from any accident that may occur after the change; provided, however, that this does not apply if the change is due to reasons not attributable to the policyholder or the assured. Article 823: In the following cases, an insurer is not liable to compensate for loss or damage arising from any accident that may occur after the relevant fact arises; provided, however, that this does not apply if the fact has no influence on the occurrence of the accident or the fact is due to reasons not attributable to the policyholder or the assured: (i) the assured has neglected to cause the Ship to depart or continue the voyage, (ii) the assured has changed the route or (iii) beyond what is set forth in the preceding two items, the policyholder or the assured has significantly increased the risk. Article 824: If there is a change to a Ship specified in a Cargo Insurance Policy, the insurer is not liable to compensate for loss or damage arising from any accident that may occur after the change; provided, however, that this does not apply if the change is due to reasons not attributable to the policyholder or the assured.

Warranties are legal techniques frequently used in common law countries. Under the warranty in insurance law, if a breach occurs the insurer is exempted from responsibility, regardless of the causal relationship of the breach with the loss and/or damage, in principle.37 In Japan, the ITC and ICC are widely used, along with their special clauses and endorsements, which include various warranties. This means that warranties are used as contractual terms in the policy. While Japanese law does not have any legal concept corresponding to warranty in English insurance law, various insurance warranties are valid as contractual agreements so long as they are not against the principle of good faith under Civil Code Article 2, which is mandatory. In the UK, because of the strictness of the effect of warranties on the assured, the 2015 Insurance Act revised past case law including the 1906 MIA. In Japan, it is rare that insurers

This means that the items the insurer asked to disclose are only considered. Even where the risk increases significantly, the insurer has no rights if the insurer did not ask a question on the risk for disclosure at the time of making the contract. 36 Translation by the Japanese government. See note 20. 37 MIA 1906 s.33, UK Insurance Act 2015. 35

Marine insurance law as a treasure trove of legal principles  19 seek exemption from responsibility for a breach of warranty if there is no causal link between the breach and the loss. In the author’s view this is because Japan’s Civil Code mandatory principle of good faith may restrict the insurer’s ability to deny payment on the basis of breach of a warranty with no causal link between the loss and the breach.38 Although there are no precedents in Japanese courts regarding the interpretation of contract warranties, insurers are likely to refuse payment only when the breach of warranty caused loss and/or damage or significantly increased risk. Japan’s civil law including the Insurance Act does not recognise warranties, although the Commercial Code marine insurance provisions exempt insurers or terminate contracts under certain circumstances, which produces effects similar to warranties. Limiting risks is important for insurers, but it also has the effect of increasing the acceptability of highly individualised risks, thus benefiting the policyholders. Systems that limit hazards, including warranties, have importance not limited to marine insurance but rather are essential in the study of other insurance contracts as well. 4.7

Constructive Total Loss and Abandonment

In marine insurance, losses are categorised as total or partial, and under some types of coverage the insurer pays only for total losses, which makes these losses important. The concept of a total loss, including constructive total loss, dates to the beginning of the history of marine insurance. However, the total loss in marine insurance was based on the economic interests related to ships and cargo and not limited to the physical complete destruction, which can be seen in Italian marine insurance policies from the fourteenth century.39 Such commercial necessity created the law of abandonment.40 Marine insurance laws in many countries incorporate provisions on abandonment, but there are differences between their applications under civil versus common law.41 In civil law, generally, abandonment is recognised as a unilateral act by which the assured obtains a claim for the full insurance amount in exchange for providing the insurer with the property of the insured ship and/or cargo, and Japan followed this model based on the old German Commercial Code. Japan’s 1899 original Commercial Code stated five circumstances where the assured has a right of abandonment; the sinking of a ship, the disappearance of the ship, the inability to repair the ship, the seizure of the ship or cargo and the confiscation by the government of the ship or cargo and not released for six months.42 An assured making a claim for the full policy

38 Regarding the application of the principle of good faith, the Supreme Court interpreted that in a policy with a clause that the insurer would not assume any liability in the event of a violation of the notification obligation, the insurer would be exempt only for the damages caused by the violation of the notification obligation unless there were any deliberate neglect not allowable in the light of good faith (Supreme Court Decision 20/2/1987.Minshu41/1, p.159). 39 For instance, one marine insurance policy dated 9 September 1388, in Florence, shows that the insurer paid when a vessel was missing for more than ten months (Kimura, note 31, p.43). 40 In the UK, abandonment dates back to the sixteenth century. Rob Merkin, Marine Insurance A Legal History (Edward Elgar Publishing, 2021), para. 2-048. 41 Kyriaki Noussia, The Principle of Indemnity in Marine Insurance Contracts: A Comparative Approach (Springer, 2007), p.114. 42 Article 833.

20  Research handbook on marine insurance law amount based on abandonment must file the claim within three months.43 The insurer acquires the insured property and assumes the responsibility of paying the entire insurance amount.44 In the UK, a total loss is divided into two types – actual or constructive45 – and in cases of an actual total loss, the insurer immediately recognises the claim for the entire insurance amount. Where there is a constructive total loss, the assured may either treat the loss as a partial loss, or abandon the subject-matter insured to the insurer and treat the loss as an actual total loss.46 When the assured elects to abandon the subject-matter insured to the insurer, he must give notice of abandonment.47 If the insurer accepts it, the insurer acquires the insured property and pays the total loss amount.48 The assured cannot unilaterally transfer the insured property to the insurer.49 In comparison, Japan’s Commercial Code previously provided for abandonment based on continental law, but the assured’s unilateral ability to transfer property rights to the insurer caused a number of problems. For instance, shipwrecks and cargo losses usually become negative assets, and insurers that acquire wrecked vessels or ruined cargo could sometimes be responsible for the significant removal costs and other costs in addition to the insurance payments. Originally, abandonment was an excellent remedy for two reasons. First, abandonment expanded the legal rights of the assured by requiring the full policy amount to be paid in the event of a situation amounting not strictly to total loss but nearly the same economically. Second, it eliminates the possibility of gaining profit while valuable residual properties exist. These advantages were based on the assumption that wreckage would have value, but although this might have been true in ancient times, the situation is very different today. In Japan, insurers generally recognise total economic loss regardless of whether it was related to abandonment, which expands the rights of the assured without requiring abandonment. With regard to the second point of eliminating unjust enrichment, Japanese insurance law has a system of subrogation to residual property in the event of a total loss50 that gives the insurer the rights to any valuable residual property on paying the total loss; therefore, it is not necessary to use abandonment.51 Because total loss coverage was available without claims of abandonment, neither party to a policy contract benefited from abandonment claims, and insurers began to stipulate in their policies that they did not accept abandonment;52 therefore, the provisions related to abandonment in the Commercial Code were meaningless and actually

Article 836. Article 833. 45 MIA s.56(2). 46 MIA s.61. 47 MIA s.62. 48 MIA ss.61–3. 49 The insurer is entitled, not bound, to take over the property. See Jonathan Gilman et al., Arnould: Law of Marine Insurance and Average (20th ed. Thomson Reuters, 2021), sections 30–6. 50 Article 24, Insurance Act. 51 As to the transfer of property, it is the right of the assured in abandonment, whereas in the subrogation over the property in total loss, it is the right of the insurer. In the latter, the insurer can choose whether to obtain the property or not. 52 Overseas cargo or global hull insurance often use London’s standard clauses, which are governed by English law. Since English law allows abandonment, Japanese insurers insert a clause not allowing abandonment. 43 44

Marine insurance law as a treasure trove of legal principles  21 had adverse effects. Against this background, in the Commercial Code amendment of 2018, all the provisions related to abandonment (Articles 833 to 841) were abolished. In the modernisation of Japan’s Commercial Code, there were opinions from practitioners that although the provisions related to abandonment should be abolished, it might be necessary to build in contingencies for exceptions that would be contrary to the strict principle of indemnity. The study group also suggested introducing the concept of constructive total loss used in English law and establishing provisions to treat it as a total loss. However, these opinions were all rejected for multiple reasons. First, payment as a total loss that is not strictly totally lost is not unique to marine insurance.53 In addition, as noted above, the Insurance Act contains a provision for subrogation over the property,54 which reflects that the total loss in the Act is not limited to physical complete destruction but also allows for total loss treatment even when there is residual value. Furthermore, the provisions of the Insurance Act are voluntary, and the parties can change their policy provisions if they agree on the changes.55 The main theoretical issue here is how to evaluate loss in insurance. Although the provisions related to abandonment in Japanese law were abolished, research on evaluating loss in marine insurance remains important also for other types of insurance. The Commercial Code on marine insurance contains more articles than are referred to above.56 They fall outside the scope of this chapter, as they are specifically only on marine insurance, and they will therefore not be discussed here.

5.

THE SIGNIFICANCE OF THE MARINE INSURANCE LAW

Legal principles and laws on marine insurance can be broadly classified into three categories: those specific to trade and shipping, those related to trade and shipping but also common to large-scale corporate risk insurance and those that are common to any type of insurance. The Japanese Commercial Code set forth the principles and rules in the first and second categories.57 While all three are important, the below will focus on the second. Although insurance laws differ from country to country, a common theme is consumer protection. Indeed, in recent years, legislation on insurance contracts has been largely amended to incorporate greater consumer protections in many countries. An important example is

53 Such as when car repair costs will exceed the car’s value or when goods are stolen and not recovered for a certain period of time. 54 Article 24, Insurance Act. 55 Article 24 is an optional rule, and insurers are not bound by it. This weakens the need to include in the Commercial Code a provision on marine insurance that differs from the Insurance Act. If we include a provision on constructive total loss in the marine insurance section, it would imply that constructive total loss applies primarily to marine insurance and not to other types of insurance, including consumer insurance. 56 Article 818 (Insured Value of Ship Insurance), Article 819 (Insured Value of Cargo Insurance), Article 821 (Information Contained in Document to Be Delivered upon Conclusion of Policy), Article 827 (Liability to Compensate Where Cargo Is Damaged), Article 828 (Liability to Compensate Where Cargo Is Sold Due to Force Majeure), Article 830 (Application Mutatis Mutandis to Mutual Insurance). 57 The author of this chapter understands that the UK MIA 1906 contains principles of all three categories.

22  Research handbook on marine insurance law the Principles of European Insurance Contract Law (PEICL).58 PEICL is the result of joint research by a group of European legal scholars including those in the UK. It was based on an extensive and detailed investigation of insurance law in European countries and the UK and presented a model of insurance contract law. This was an enormous achievement not only in Europe but around the world as model for comparative insurance law research. Although PEICL is intended to apply to all insurance contracts except reinsurance, its contents in my view correspond to the third category of the above three classifications. Naturally, PEICL does not contain any rule in the first category, that is, the specific rules for marine insurance, and does not appear to incorporate the principles from the second category, which are peculiar to insurance for large risks in the corporate sector; this could be because corporate insurance encompasses a degree of contractual freedom. The national laws in many civil law countries do not contain default rules specific to corporate insurance except marine insurance. However, situations in common law countries are different, since there is abundant case law on corporate large-risk insurance which are binding as case law. 5.1

Insurance Types59

Private insurance, excluding social insurance, is categorised from various perspectives, such as the predictability of risk. Risk-based policies can in turn be broadly divided into mass-risk and non-mass-risk insurance, with the latter also called large-risk insurance. While it is difficult to draw a clear line between the two, it is useful to analyse the law by focusing on the difference between the two types. 5.1.1 Mass-risk insurance Mass-risk insurance refers to insurance that allows for calculating premiums with a certain degree of accuracy based on past data and underwriting a sufficient number of contracts for the law of large numbers to operate.60 Insurance based on life tables (death, survival) and insurance based on statistical traffic accident or fire data are typical examples of mass-risk insurance.61 In the case of mass risk, stable insurance operations can be achieved by pooling risks through accumulating a large number of contracts and utilising the law of large numbers. Thus, it is important, first and foremost, to gather as many contracts as possible and exclude those with extremely high risk. Conducting detailed risk evaluation for each individual risk is not important, but rather becomes a barrier to contract acquisition, increasing costs and becoming a negative factor in market competition. To collect a large number of contracts,

PEICL 2009 by Project Group Restatement of European Insurance Contract Law. This part is partially derived from Satoshi Nakaide, ‘The Types of Insurance Schemes Identified from a Risk Based Perspective: Implications Derived from the Analysis of the Principles of Insurance’, Japan Institute of Life Insurance Journal, no. 221 (2022), p.1. 60 The law of large numbers is originally a theory of mathematics: the larger the sample size, the more closely the actual results will approach the expected results. In insurance, this principle means that the larger the number of policies an insurer underwrites, the more accurately it can predict the frequency and severity of future losses. George E. Rejda et al., Principles of Risk Management and Insurance (14th ed., Pearson Education Limited, 2022), p.44. 61 My risk classification in this chapter does not specifically correspond to the EU’s classification of mass versus large risks. 58 59

Marine insurance law as a treasure trove of legal principles  23 a certain amount of screening that rationalises insurance premiums and minimises costs is effective. Events not assumed in statistical data, such as war or radiation contamination from a nuclear power plant, for instance, need to be excluded from the coverage, which enhances policy stability. Risk can accumulate, but highly biased risks can be externalised through reinsurance. Mass-risk insurance assumes certain accidents will occur, and the core of the business is managing the stability of the risk pool; therefore, deviations from predictions can have significant impacts. 5.1.2 Non-mass-risk insurance Non-mass-risk insurance refers to insurance coverage in areas for which limited data are available, making it difficult to quantitatively predict risks in advance, or areas in which few contracts can be concluded, making stable insurance underwriting difficult. Examples include aviation insurance including for space travel, marine insurance, reinsurance, property and profit insurance on large plants, corporate liability insurance, large-scale cyber insurance, insurance for intellectual property infringement and insurance for large-scale business interruptions such as caused by pandemics. Non-mass-risk insurance makes it possible to estimate risk only roughly because of the lack of available data, and even with inferences from similar risk situations and mathematical models, the small number of contracts limits prediction accuracy. It is therefore necessary to evaluate risk case by case and customise the coverage. Risk diversification through reinsurance is essential. The method and evaluation vary depending on the insurer’s risk appetite. In addition, in non-mass-risk insurance, predicting risk is difficult and the scope of coverage must be limited. The addition of special clauses is necessary for each underwriting, making drafting skills critical for insurers. Notably, although marine insurance is mostly non-mass-risk, yacht and pleasure boat insurance fall under mass-risk insurance. 5.2

Differences in Legal Principles between the Two Types of Insurance

Japan’s Insurance Act did not clearly separate the two types of insurance, assuming insurance for mass risks in the law but allowing for a broad degree of contractual freedom for insurance of non-mass risks; as a result, the principles and rules peculiar to non-mass-risk insurance are not always explicit. However, if there are differences in the nature of the business, there should also be differences in the law that applies to it. Differences in risk and therefore in policies entail differences in applicable legal principles, some of which will be presented below. 5.2.1 Policy provisions In mass-risk insurance, standardisation and uniformity are particularly important, and policies use the insurer’s standard language. In contrast, non-mass policies are often changed through endorsements, and it is not rare that a policy includes language tailored to the client by an insurance intermediary such as a broker. In turn, policy conditions and wordings are arranged individually between insurer and assured. 5.2.2 Principles of interpretation Premiums for mass-risk insurance are computed under the assumption that a substantial number of contracts can be secured. Deviating from this approach to underwrite only specific

24  Research handbook on marine insurance law risks may lead to discrepancies from estimates derived from general statistics, resulting in a detriment to the insurance balance.62 Consequently, the terms of mass-risk insurance policies are standardised to encompass a broad demographic. In the case of these contracts, their interpretation must be consistent and not vary from individual to individual.63 However, in non-mass-risk insurance, the contract is highly individualised and, to that extent, the interpretation of the wording should be allowed to focus on the intention of the parties. 5.2.3 Duty of disclosure In mass-risk insurance, it is crucial to efficiently collect a large number of contracts and eliminate risks that exceed the limits of risk assumption. The significance of the duty of disclosure lies in this kind of mechanism, such as a screening mechanism. Screening involves setting criteria from a uniform perspective. Therefore, the use of pre-contract questionnaires is effective in achieving this objective. In contrast, vendors of non-mass-risk insurance must evaluate risk case by case not only to determine whether to accept an application but also to set appropriate insurance coverage and premiums. As risks vary individually, insurers may not know what is important for each specific case while they know the risk in general. Because of the differences between the mass-risk and non-mass-risk insurance businesses, the function of disclosure is different between them. UK enacted the Insurance Act 2015 to address rules on disclosure for business insurance different from that for consumer insurance and convincingly, introduced the new idea of a duty of fair presentation.64 Japan’s Commercial Code established a rule of duty of disclosure different from that of the Insurance Act as previously explained. The manner in which the duty of disclosure should be carried out theoretically depends on the nature of the risk being assumed. Marine insurance is a typical example of this and it should be, at least theoretically, also applicable to other non-mass-risk insurance. 5.2.4 Exclusions and alteration of risk In non-mass-risk insurance, individual risks are highly specific, and insurers must aim to limit risks as much as possible using various legal techniques. While risk limitation is not unique to non-mass-risk insurance, the methods employed to achieve it are diverse and vary in wording in non-mass-risk insurance. In risk limitation, methods employed include specifying detailed coverage with various exceptions and exclusions, setting various prerequisites or conditions, stipulating effects for changes in risk and excess and deductibles. Because of this, non-mass-risk insurance policy wording tends to be very complicated.

The larger the number of policies an insurer underwrites, the more accurately it can predict the frequency and severity of future losses. This is important for insurance companies to accurately estimate future losses and better manage risks. 63 For the law of large numbers to operate, uniformity of each risk is important. 64 Theoretically, the standard for disclosure duty should be changed according to the nature of the risk assumed, rather than based on whether the contracting party is a merchant or not. For instance, in auto or life insurance contracts where corporations are the contracting party and a large number of contracts are involved, the information to be disclosed can be standardised similarly to insurances for individuals. 62

Marine insurance law as a treasure trove of legal principles  25 5.2.5 Assessment of insurance claim and payment While this area is not governed by the Japanese Insurance Act or Commercial Code, and is not referenced as a legal principle in relation to marine insurance contracts above, I would still like to discuss certain aspects of insurance payments. In the context of insurance payment processing, uniformity and fairness are critical considerations for mass-risk insurance policies operating under the same contract terms. In contrast, non-mass-risk insurance policies exhibit high levels of individuality in terms of the insured items, accident circumstances and situation and condition of loss or damage, making it inherently difficult to clarify and establish evaluation criteria and contractual interpretations. Therefore, it is essential to adopt a flexible approach based on the principles. The above examples are merely illustrative. However, it is evident that there exist differences in the principles of insurance contract law between mass-risk insurance and non-mass-risk insurance in various respects. Furthermore, the characteristics of non-mass-risk insurance share many similarities with the distinct features of marine insurance that were previously discussed in the revision of the Japanese Commercial Code. In other words, the study of marine insurance can be instrumental in elucidating the legal theory of non-mass-risk insurance as well.

6. CONCLUSIONS Many countries have insurance laws, and in maritime nations there are often specific marine insurance laws. Marine insurance law is recognised as a specialised legal field that pertains to a particular type of insurance. In Japan, marine insurance law is a component of the Commercial Code that deals with maritime business and is understood to be the insurance discipline specific to maritime affairs. However, a careful examination of the Japanese Commercial Code reveals that the principles outlined therein encompass principles and rules that are also applicable to other types of non-mass-risk insurance. It appears that many continental law countries codify the rules on mass-risk insurance but leave the rules unique to non-mass risks to contractual freedom. Consequently, the legal landscape on non-mass-risk insurance is ambiguous, and clarifying such rules represents a future challenge. Given the globalisation of corporate insurance, addressing this challenge requires the harmonisation of laws between continental law and common law. Marine insurance law offers a wealth of systems and principles that are not addressed in general insurance law, making it a valuable resource for studying non-mass-risk insurance law as well. In this sense, marine insurance law is a treasure trove of legal principles.

2. Protection and indemnity insurance: is it truly insurance? Rhidian Thomas

1. INTRODUCTION The question raised in this contribution is on occasions directed at protection and indemnity (P&I) insurance of third party risks associated with shipping and related marine activities. Is it truly insurance?1 The insurance is provided principally by P&I Associations (Clubs) that are members of the International Group of Protection and Indemnity Clubs (IGP&I).2 The IGP&I is an unincorporated association which consists of 12 mutual Clubs (usually identified as Group Clubs), and functions through a central administration and a number of specialist committees and other bodies.3 It assumes particular responsibilities for managing the relationship between the individual Group Clubs, procuring and administrating reinsurance and representing Group Clubs at national and international fora.4 It occupies a significant position in contributing to the development of global policy in relation to international shipping and the marine environment.5 The member Group Clubs are the principal source of insurance underpinning the expanding regime of international maritime conventions.6 The Group Clubs collectively insure the liabilities of about 90 per cent of international shipping measured by tonnage.7 They function in a competitive environment, even as between themselves, and provide insurance cover on the basis of their Rules, which range over the entire

Merkin, Marine Insurance Legislation, 5th edn (2014, Informa Law from Routledge), pp.142–3. See generally, Hazelwood and Semark, P&I Clubs: Law and Practice, 4th edn (2010, Lloyd’s List, London); Jonathan Gilman KC et al., Arnould: Law of Marine Insurance and Average, 20th edn (2022, Sweet & Maxwell, London), 4-09–4-15; Hill, Robertson & Hazelwood, Introduction to P&I, 2nd edn (1996, LLP, London). 3 The IGPIC is located at: 3rd Floor, 78/79 Leadenhall Street, London EC3A 3DH. The Group Clubs are the American Steamship Owners Mutual Protection and Indemnity Association Inc.; The Japan Ship Owners’ Mutual Protection and Indemnity Association; Assuranceforeningen Gard (Norway); Assuranceforeningen Skuld (Norway); The Swedish Club (Sweden); The Britannia Steam Ship Insurance Association Ltd; The London Steam-Ship Owners’ Mutual Insurance Association Ltd (The London P&I Club); North Standard Ltd (previously named the North of England Protecting and Indemnity Association Ltd) and Indemnity Association Ltd; The Shipowners’ Mutual Protection & Indemnity Association; The Standard Club Ltd; The Steamship Mutual Underwriting Association (Bermuda) Ltd; United Kingdom Mutual Steam Ship Assurance Association Ltd; and The West of England Ship Owners Mutual Insurance Association (Luxembourg). 4 IGP&I 20/21 Annual Summary. 5 The IGP&I is granted observer status at the IMO and Global Maritime Forum and participates in the deliberations of the CMI. 6 See, generally, De La Rue and Anderson, Shipping and the Environment, 3rd edn, Ch. 20, ‘P&I Clubs and other Liability Insurers’ (2022, Informa Law from Routledge, London). 7 See www​.igpandi​.org. 1 2

26

Protection and indemnity insurance: is it truly insurance?  27 span of potential maritime liabilities.8 Different Classes of cover are available but the Class of relevance to the current discussion is traditionally referred to as P&I. This cover includes risks relating to death and personal injury, passengers, cargo loss and damage, collisions as between ships and fixed and floating objects, salvage and general average, towage, wrecks, pollution from ships, stowaways and quarantine.9 The insurance has responded to the expansion of potential shipping liabilities over the past 60 years, particularly those relating to oil and bunker oil pollution from ships, passenger liabilities, wrecks and the growth of off-shore commercial activities.10 Although the insurance is available primarily to owners, this concept is defined broadly to include charterers, specialist operators, ship managers, international traders and operators in the off-shore industry.11 Beyond their roles as primary insurers, Group Clubs also pool larger claims between themselves on the basis of an agreed formula.12 The management of Group Clubs is governed by the Articles of Association and Rules. The various branches of management are vested in a board of directors, members’ committee (or equivalent body) and managers, with the latter playing the major role in the day-to-day management of the insurance.13 The managers are either directly employed by the Club or a management company engaged under contract.14 All is supplemented by an international network of correspondents.15 The managers play a dominant role in underwriting and handling claims. The members’ committee, which, as its name indicates, is composed of elected members, plays an important role in determining whether in prescribed circumstances a member’s claim should be paid.16 This ensures that in the event that a problematic or controversial matter arises in connection with a claim it is resolved by representatives of the members and not management, with some issues reserved to the directors.17 The Group Clubs domiciled in England are part of the structure of national insurance and subject to regulation under the Financial Services and Markets Act 2000, as amended by the Financial Services Act 2012, and the rules promulgated by the Prudential Regulatory

8 In this contribution the Rules of the UK P&I Club are cited as representative of the rules of Group Clubs generally. In their substance the rules of the 12 Group Clubs are very similar but may be different in structure and drafting. 9 UKP&I Rule 2 – Risks Covered. 10 The international conventions in question are the International Convention on Civil Liability for Oil Pollution Damage 1992 (CLC 1992); the International Convention on Civil Liability for Bunker Oil Pollution Damage 2001 (Bunker Convention 2001); the Athens Convention Relating to the Carriage of Passengers and Their Luggage by Sea 2002 (Athens Convention 2002); and the International Convention on the Removal of Wrecks 2007 (Nairobi Convention 2007). 11 UKP&I, Articles of Association, art 2 (Interpretation): ‘owner’ in relation to an entered Ship means owners, owners in partnership, owners holding separate shares in severalty, part owner, mortgagee, trustee, charterer, operator or builder of such Ship and any other person. 12 This practice is based on a Pooling Agreement which is renewed annually and currently applies to claims in excess of US$10 million. 13 Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn, Chs 2 and 3 (2010 Lloyd’s List, London). 14 The UK Club is managed by the Thomas Miller Group. 15 See IGP&I publication Guidelines for Correspondents (2022). 16 See Thomas, Direct and third-party claims against P&I Clubs, Ch 3 in Modern Law of Marine Insurance, vol 5, ed Prof D Rhidian Thomas (2023, Informa Law from Routledge, London); Tilley, Protection and Indemnity Rules and Direct Actions by Third Parties (1986) 17 JMLC 427. 17 See the ‘Omnibus rule’ below.

28  Research handbook on marine insurance law Authority and Financial Conduct Authority.18 Following Brexit, some have also established a corporate presence in Europe so as to be in position to insure risks throughout the EEC.19 This question raised in this contribution might be considered surprising given the history and contemporary significance of P&I insurance. But neither consideration is conclusive of the question; nor is the customary description of the cover as insurance. To describe a facility as insurance is not necessarily conclusive of its true character but, at the same time, it is not necessarily devoid of any significance.20 Ultimately the issue is one of substance and not form.21 There are many commercial agreements that may be regarded as serving a similar function as insurance, in that they are beneficial to the interests of parties exposed to loss, disadvantage or liability, but which are not categorised as insurance agreements. Contracts of indemnity, maintenance, guarantee and product warranties are identifiable examples,22 and how they are to be construed continues to be the subject of debate.23 But if P&I cover is not insurance, to what legal category does it belong. It continues to be a valid commercial agreement; it is not illegal, unlawful, contrary to public policy or otherwise void. And if not insurance, what would be the consequences? It would, presumably, not be subject to the law and legislation applicable to contracts of insurance.24 There might also be implications for market regulation.25 It is probable that the question about the status of P&I insurance arises as a consequence of its mutual nature and the significant degree of discretion embodied in the Rules for determining if claims should be paid.26 Underpinning the question is the more fundamental issue: what is insurance? This represents the unavoidable starting position, but it is not a hurdle capable of being negotiated with ease. Notwithstanding the universal familiarity of the practice of insurance, when it comes to the question of definition, difficulties abound.27 It is hardly a source of encouragement that an English judge has expressed the opinion that it might be best not to attempt an exhaustive definition.28 With some trepidation, that advice is ignored.

For a concise overview, see Birds & Richards, Birds’ Modern Insurance Law, 12th edn, Ch. 2 (2022, Sweet & Maxwell, London). 19 For example, the UK Club has established a corporate presence in the Netherlands: the UK P&I Club N.V. 20 Fuji Finance Inc v Aetna Life Ins Co Ltd [1997] 173, 189 per Morritt LJ. 21 Department of Trade and Industry v St Christopher Motorists’ Association Ltd [1974] 1 Lloyd’s Rep 17, 21 per Templeton J. 22 Above. See, generally, Clarke, Law of Insurance Contracts, 6th edn, Ch. 1 (2009, Informa, UK); McKendrick (ed.), Goode on Commercial Law, 5th edn, para. 30.13 (2016, Penguin, UK); Birds and Richards, Birds’ Modern Insurance Law, 1-09 et seq. 23 Fuji Finance Inc. v Aetna Life Insurance Co Ltd [1997] Ch. 173(CA); Digital Satellite Warranty Cover Ltd v Financial Services Authority [2012] Lloyd’s IR 112(CA), Lloyd’s Rep IR 236(SC). 24 Of particular significance, insurance contracts are contracts of utmost good faith and dependent on the existence of an insurable interest. Regulatory and private law legislation relating to insurance and insurance contracts might not apply except, possibly, by analogy. 25 Above n.18. 26 Considered below under the title ‘The Nature of P&I Insurance’. 27 The Medical Defence Union Ltd v Dept of Trade [1979] 1 Lloyd’s Rep 499, 505 per Sir Robert Megarry V-C. Also Clarke, The Law of Insurance Contracts (now in looseleaf format, latest issue 11 November 2022). 28 Above p.505 per Sir Robert Megarry V-C. Prof Clarke has declared that ‘Convincing definitions of insurance do not exist’: see Clarke, The Law of Liability Insurance, 2nd edn (2013, Informa Law from Routledge, London), para. 1.2.1. 18

Protection and indemnity insurance: is it truly insurance?  29

2.

INSURANCE AND CONTRACTS OF INSURANCE: QUESTIONS OF DEFINITION?

The discussion in the present context is confined to contracts ‘of’ and not ‘for’ insurance,29 and to indemnity and not contingent insurance.30 The search for a definition of insurance may follow many paths, each promoting a response that reflects the particular perspective or purpose in question.31 In the present context the focus is on matters legal and commercial. But even adopting this approach is not to identify a single route of travel. The approach and outcome may again vary against the backdrop of different classes of risk and the underlying commercial purpose of the insurance. In the realm of insurance the sometime practice of the law is to omit any attempt at a definition or, at best, to adopt a descriptive definition. The singular example of the former approach is the Financial Services and Markets Act 2000, which regulates insurance companies and businesses, without defining insurance.32 The Marine Insurance Act 1906, Section 1, is an example of the latter approach. It defines a contract of marine insurance as ‘a contract whereby the insurer undertakes to indemnify the assured, in manner and to the extent thereby agreed, against marine losses, that is to say, the losses incident to marine adventure’.33 This descriptive approach is useful but fails to address the constituent essence of marine insurance.34 Nonetheless, it is often adopted by express reference in related legislation.35 So far as the common law is concerned, there may on occasion be a disposition to follow the understanding of commercial people. The more an agreement looks like insurance and is acted upon as such in the course of commerce, the greater may be the judicial leaning to accept it to be an insurance agreement.36 This is to define insurance in terms of commercial recognition and identify the commercial community as arbiter. Of course, there are limits to this approach: in the final analysis, substance trumps perceptions. But at the same time, it has to be recognised that there has to be a strong reason to reject the consistent understanding of the commercial community.37 Less difficult in point of principle is such a reference as a makeweight. Having concluded in the face of a dispute that an agreement is to be categorised as insurance, it may be anticipated that it would be reassuring for a judge to be able to observe that the same view is adopted by the commercial community. A framework contract which provides for insurance to be obtained subsequently is construed as a contract ‘for’ insurance and the insurance later obtained as a contract ‘of’ insurance: see also Gould v Curtis [1913] 3 KB 84, 94, 98; The Medical Defence Union Ltd v Dept. of Trade, above p.507 per Sir Robert Megarry V-C. 30 Under contingency insurance the insurer must pay the sum agreed upon the occurrence of the specified event: see Virk v Gan Life Holdings Plc [2000] Lloyd’s Rep IR 159, 162. 31 See, generally, Merkin and Steele, Insurance and the Law of Obligations, Ch. 2 (2013, OUP, Oxford); Lowry, Rawlings & Merkin, Insurance Law: Doctrines and Principles, 4th edn, Ch. 1 (2022, Bloomsbury Publishing plc, London). 32 The traditional reason given for this is that it avoids the risk of accidentally omitting agreements that should be the subject to regulation. 33 By contrast, the Insurance Act 2015 does not provide a definition of ‘contract of insurance’. 34 See, generally, Gurses, Marine Insurance Law, 3rd edn, Ch. 1 (2023, Routledge, London). 35 For example, Third Parties (Rights against Insurers) Act 2010, section 9(7). 36 Department of Trade and Industry v St Christopher Motorists’ Association Ltd, above p.14 per Templeman J; Home Insurance Co v Admin Asigurarilor de Stat [1983] 2 Lloyd’s Rep 674. 37 Fuji Finance Inc. v Aetna Life Ins. Co Ltd., above, p.189, per Morritt LJ. 29

30  Research handbook on marine insurance law Although the judicial and commercial approaches to an understanding of what is a contract of insurance may run in parallel, this is not necessarily the case. The judicial approach may be inclined to be more searching, more inwardly analytical, and less instinctive than commercial perception. In this process, the following indicia or constituent elements might be identified. 1. Insurance is based on a contract between parties, namely insurer and assured,38 which gives the assured protection against the consequence of specified happenings. The insurer is invariably a corporate or mutual entity engaged in the business of providing insurance.39 The contract of insurance is a mechanism for the transfer of risk borne by the assured to the insurer for a monetary consideration.40 2. The agreement is invariably embodied in a written contract which may be accompanied by a policy.41 It follows that the mutual duties, rights and powers of the parties are legally binding. The assured has an enforceable right to payment or other compensation for losses within the insurance, and the insurer is obliged to pay or otherwise compensate the assured as the contract prescribes.42 3. The happening that triggers the insurance involves an element of uncertainty: an event that may or may not occur. There must be present an element of chance, a risk, not certainty. Insurance is antithetical to events that are certain to occur or to occur at a particular time.43 4. It is probable that the happening must also be outside the control of the insurer.44 5. The happening must be prejudicial or adverse to the interests of the assured. It must involve loss or disadvantage or detriment.45 6. The assured agrees to pay a premium or equivalent payment as consideration for the agreement of the insurer to assume the risk.46 Calls paid by members of a P&I Club are an equivalent payment.47 The word ‘assured’ is preferred but has the same meaning as ‘insured’. Hampton v Toxteth Co-operative Provident Society Ltd [1915] 1 Ch. 721(CA). 40 It is for this reason that the market concept of ‘self-insurance’ is not strictly regarded as insurance. 41 A policy is generally not essential and its role appears to be of dwindling significance in the evolving law. The Law Commission has recommended that the requirement of a policy in marine insurance be repealed, see Consultation Paper No 201. 42 Hampton v Toxteth Co-operative Provident Society Ltd. Above. See also section entitled ‘Distinction between “right” and “discretion”’. 43 This point is particularly relevant to life insurance. See Prudential Insurance Co v IRC [1904] 2 KB 658, 663; Department of Trade and Industry v St Christopher Motorists’ Association Ltd, above pp.20–1. 44 This proposition has not been affirmed in the authorities but it is accepted by many commentators: see Birds and Richards, Bird’s Modern Insurance Law, 12th edn, paras 1–10; MacGillivray on Insurance Law, 15th edn, Gen ed. Leigh-Jones, 1-006 (2022, Sweet & Maxwell, London). In Department of Trade and Industry v St Christopher Motorists’ Association Ltd, above, and The Medical Defence Union Ltd v Dept. of Trade, above, the proposition was raised but left open. Even in the absence of supporting case law it does appear to be a logical proposition. 45 Prudential Insurance Co v IRC, above pp.662–3; Department of Trade and Industry v St Christopher Motorists’ Association Ltd, above p.21 per Templeton J; The Medical Defence Union Ltd v Dept. of Trade, above p.505, per Sir Robert Megarry V-C. This point is not necessarily applicable to contingent insurance: see Gould v Curtis [1913] 3 KB 84 (CA). 46 Marine Insurance Act 1906 s.85(2) where it recognised that ‘a guarantee, or such other arrangement as may be agreed upon’ may be substituted for the premium. See also Wooding v Monmouthshire & South Wales Mutual Indemnity Society [1938] 3 All E R 625; [1939] 4 All E R 570. 47 Above. 38 39

Protection and indemnity insurance: is it truly insurance?  31 7. The benefit payable by the insurer is in the form of money or ‘money’s worth’.48 The latter alludes to a benefit which is the equivalent or corresponding to money payment. Examples are rebuilding a house or repairing a ship.49 Beyond this, it is not the case that ‘any benefit’ will suffice.50 8. The assured must have an insurable interest in the subject matter of the insurance. Only then can the assured claim to have suffered an adverse consequence, in the form of loss or detriment or disadvantage. Without an insurable interest the agreement may take on a different characterisation, such as a wager.51 To this extent, insurable interest is a prerequisite for a valid insurance agreement. Channel J adroitly incorporated many of these indicia in the succinct definition he proposed of insurance in the following terms: ‘A contract of insurance, then, must be a contract for the payment of a sum of money, or for some corresponding benefit such as the rebuilding of a house or the repairing of a ship to become due on the happening of an event, which event must have some amount of uncertainty about it, and must be of a character more or less adverse to the interest of the person affecting the insurance.’52 This definition has achieved general approval but with it acknowledged that it has to be qualified in the case of life insurance.53 2.1

Distinction of ‘Right’ and ‘Discretion’

Of the indicia identified, the most significant to the current discussion is the emphasis on the contractual entitlement to benefit. Once loss within the terms of the insurance is suffered, the assured has an enforceable right to compensation in whatever form specified. An implication of this premise is that an agreement under which payment or other benefit is discretionary is not insurance.54 Consequently the division between enforceable right and discretion is critical, with the precise nature of any agreement a question of construction. In Department of Trade & Industry v St Christopher Motorists Assn55 the question before the court was whether the defendant Association carried on business as insurers to which the then applicable Companies Act 1958 applied. Under the rules of the Association, if an event occurred which prevented a member from driving his car, the Association would provide him with a driver, or a car and a driver, for stated periods.

48 The older view that payment under a contract of insurance can be in money only, as enunciated by Brett LJ in Raynor v Preston (1881) 18 Ch. D 1, 9, is now rejected. 49 Prudential Insurance Co v IRC, above p.664. 50 The Medical Defence Union Ltd v Dept. of Trade, above p.505, where the wider view based on ‘some benefit’ was rejected by Sir Robert Megarry V-C. It was also left open if services to be provided and paid for by the insurer could amount to a benefit, at p.506. 51 MIA 1906 s.4. In the modern law the concept of insurable interest is increasingly viewed less restrictively. See Quadra Commodities SA v XL Insurance SE and Others [2022] EWHC 431(Comm.) for recent judicial analysis of insurable interest in personal property. 52 Prudential Insurance Co v IRC, above pp.662–3. This definition was approved in Department of Trade & Industry v St Christopher Motorists Assn, above p.21, with Sir Robert Megarry V-C also cautioning that it might not be exhaustive. 53 Above. Also Fuji Finance Inc. v Aetna Life Insurance Co Ltd; Re Sentinal Securities Plc [1996] 1 WLR 316. 54 Hampton v Toxteth Co-operative Provident Society Ltd [1915] 1 Ch. 721(CA). 55 [1974] 1 Lloyd’s Rep 17.

32  Research handbook on marine insurance law There existed a contractual relationship between each member and the Association under which members paid an annual membership fee. As for claims, the Rules provided that a ‘claim by member to be examined by a committee which shall assess the merits of each claim and decide whether or not the claim is to be accepted’. The question in issue was did this scheme amount to insurance, and the answer depended on whether the benefits were discretionary or contractual entitlements. Templeman J concluded that on the proper construction of the Rules the right to benefits was not discretionary. The Committee had the right to reject a claim if it considered a claim not to be valid. But when faced with a valid claim, namely a claim truthfully made and within the terms of the Rules, the claimant member was entitled to the services provided by the Association. In these circumstances a claimant could not be deprived of the benefits.56 The Rules were construed as giving the member a contractual right to have a driver when the necessary conditions were satisfied, and not as giving the member a mere right to have his application fairly considered by the club committee, with discretion in the committee whether or not to grant the application. This distinction the judge considered touched upon the essence of a contract of insurance.57 In The Medical Defence Union Ltd v The Department of Trade58 the plaintiff union, a company limited by guarantee, under its memorandum and articles of association provided medical practitioners and dentists with protection in matters relating to the conduct of legal proceedings on behalf of members. The union was governed by an elected council and officers. The protection included an indemnity for claims made and related costs, and for the giving of advice. The question before the court was whether the union was an insurance company subject to the control of the Department of Trade under powers set out in the Insurance Companies Act 1974.59 The core issue concerned the meaning of ‘insurance’ and ‘contract of insurance’, which were not expressly defined by the Act, in relation to the contract between each member and the union. The articles provided that the Council/committee ‘may […] grant from the funds of the Union to any member […] an indemnity wholly or in part with regard to any […] proceeding […] and the indemnity may extend to any incidental or consequential losses’ and that ‘it shall rest only in the absolute discretion of the Council or any authorized committee in every case to limit or restrict the grant of an indemnity or altogether to decline to grant the same or to determine any indemnity so granted’. Sir Robert Megarry V-C noted the frequent use of the word ‘may’, which in this context was to be construed as permissive, and concluded that on a proper construction of the articles the right of a member ‘is merely a right to have his request fairly considered by the Council or one of the committees. Only if the request is granted is the member entitled to have the proceedings conducted by the Union and to have an indemnity’.60

Above, p.19. The approach adopted by Templeman J was approved by Mocatta J in CVG Siderurgicia del Orinoco SA v London Steamship Owners Mutual Insurance Association Ltd (The Vainqueur Jose) [1979] 1 Lloyd’s Rep 557, 580. 58 [1979] 1 Lloyd’s Rep 499. 59 Superseded by the Financial Services and Markets Act 2000 and Orders made thereunder. 60 Above, p.504. 56 57

Protection and indemnity insurance: is it truly insurance?  33 This, in the opinion of the Vice Chancellor, was to be compared with the position of an assured under a contract of insurance: ‘When a person insures, I think he is contracting for the certainty of payment in specified events, and not merely for the certainty of proper consideration being given to his claim that a discretion to make a payment in those events should be exercised in his favour.’61 The union was held not to be a company carrying on insurance business within the meaning of the 1974 Act. The Vice Chancellor firmly rejected the contention ‘that a right to have an application properly considered suffices for a contract of insurance’.62 These two cases are addressing the meaning of insurance and insurance contract in the context of the statutory regulation of the insurance market and this must elicit a degree of caution. Nonetheless, there is good reason to believe, derived from the way the two cases have been responded to subsequently by the judiciary and commentators, that they are also of significant relevance to the broader discussion about the core indicia of insurance in the context of commercial and marine risks.63

3.

THE NATURE OF P&I INSURANCE64

In addressing this subject the analysis is confined to P&I mutual cover as a distinct Class of insurance, separate from the other Classes and insurances provided by Group Clubs, such as FD&D (legal expenses), Charterers’ Risks and War Risks.65 The availability of a widening range of cover is in keeping with the policy of Group Clubs to endeavour to embrace all risks incidental to commercial developments in the maritime sphere.66 Although P&I insurance is predominantly mutual in character, it is also the case that risks may be accepted on fixed premium terms.67 3.1

The Contract

A member of a Group Club occupies two positions. First, as a member of the corporate entity, the Club (Association), the member’s position is governed by the Rules and Articles of Association.68 The second position is as assured under a contract of insurance with the Club as insurer.69 This establishes a degree of connection with the historical view of mutual insurance as groups of similarly positioned individuals who insure each other, with each member of

Above, p.506. Above, p.506. 63 See, generally, MacGillivray on Insurance Law, above Ch. 1. 64 See, generally, Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn; Hill, Robertson & Hazelwood, Introduction to P&I, 2nd edn; Jonathan Gilman KC et al. (eds), Arnould: Law of Marine Insurance and Average, 20th ed., 4-09–4-15. 65 UKP&I Rule 2 sets out the full ambit of the basic mutual cover. 66 For example, UKP&I Rules 3 & 4. 67 For ‘Fixed Premium Entries’ see UKP&I Rules 1(7) & 9. 68 The Articles of Association of the UKP&I are published in the 2022 Rules. 69 The Rules may endorse this fact by referring to ‘contracts of insurance effected by the Association’: see, UKP&I Rule 7(A). 61 62

34  Research handbook on marine insurance law the group both an assured and insurer.70 A party insured by a Club automatically becomes a member of the corporate entity.71 The contract of insurance is established when the proposal by a member, presented in accordance with the procedure laid down by the Club, is accepted by the managers on behalf of the Club.72 Following acceptance a Certificate of Entry is issued which provides conclusive evidence of the terms of the insurance.73 Any subsequent amendment is conclusively evidenced by an Endorsement Slip.74 The range of risks covered by a Group Club is published annually in its Rules.75 The particular risks covered in a negotiated contract of insurance are drawn from the Rules and identified in the Certificate of Entry.76 Beyond defining the risks covered, the Rules set out the terms of the cover, including the exclusions, limitations, conditions, warranties, procedure for making claims, dispute resolution provisions and governing law.77 Unlike market insurance, the assured member does not pay a one-off premium but is obliged to pay calls.78 The assessment of an initial call (mutual premium) is made on entry, but if it becomes necessary in order to meet its obligations the Club may make additional call(s) in respect of each policy year that has not been closed. In this regard a three-year accounting period applies. There is a range of different calls (Call Entries) that may be made to meet different circumstances.79 On the face of matters the position of a member is more precarious than that of a market assured whose premium obligation is fixed on placement of the risk and cannot thereafter be adjusted, unless permitted under the terms of the insurance. Nonetheless, Club managerial practices are strongly slanted towards avoiding or minimising the risk of additional calls.80 3.2

The Cover

In broad terms the insurance covers the third party liabilities and expenses of members incurred in relation to an entered ship. An Owner is insured against loss, damage, liability or expense incurred by it which arises: (i) out of events occurring during the period of entry of a ship in the Association, (ii) in respect of the Owner’s interest in the entered ship; and (iii) in connection with the operation of the ship by or on behalf of the Owner.81

Marine Insurance Act 1906, section 85. See also Firma – Trade S.A. v Newcastle Protection and Indemnity Association (The Fanti) and Socony Mobil Oil Co, Inc. and Others v West of England Ship Owners Mutual Insurance Association (London) Ltd (The Padre Island) (No 2) [1990] 2 Lloyd’s Rep 191, 194, per Lord Brandon (HL). 71 UKP&I Rule 14 (Membership). 72 UKP&I Rule 7. 73 UKP&I Rule 12(C). 74 UKP&I Rule 12(B) & (C) (Certificate of Entry and Endorsement Slip). 75 For example, the UKP&I Rules 2022. 76 UKP&I Rules 6 (Owners and Successors Bound by Rules); 7 (Applications for Insurance); Rule 12 (Certificate of Entry and Endorsement Slip). 77 UKP&I Rules 1, 5, 36, 40 and 42. 78 UKP&I Rule 1(6). 79 UKP&I Rules 8, 19–23 (supplementary and overspill calls). 80 The reserve funds of the Club may be applied to the purpose of stabilising supplementary and overspill calls: see UKP&I Rule 24. 81 UKP&I Rule 1(5). 70

Protection and indemnity insurance: is it truly insurance?  35 ‘Owner’ and ‘ship’ are defined very widely.82 Liability in this context means legal liability as established by a judgment, arbitral award or agreement.83 Where the legal liability is limited by law, the cover is limited to the same extent.84 The cover is for a period of 12 months running from noon GMT on any 20 February to noon GMT on the following 20 February.85 Although sometimes described as third party liability insurance, it is more accurately categorised as indemnity insurance.86 There is a significant legal difference between the two. With regard to liability insurance the cause of action exists once the liability of the assured is established in any of the ways previously discussed.87 In the case of indemnity insurance the cause of action arises when the assured discharges the liability by payment.88 The position is made clear by the ‘pay to be paid’ or ‘pay first’ provision in the Rules. The UK Club Rules provide: Rule 2 – Unless otherwise agreed between an Owner and the Managers, the risks covered by the Association are as set out in Sections 0 to 26 below, PROVIDED ALWAYS that: i. Unless and to the extent that the Directors otherwise decide, an Owner is only insured in respect of such sums as it has paid to discharge the liabilities or to pay the losses, costs or expenses referred to in those sections; ii. […] Rule 5 (A) titled ‘Payment first by the Owner’ states – Unless the Directors in their discretion otherwise decide, it is a condition precedent of an Owner’s right to recover from the funds of the Association in respect of any liabilities, costs or expenses that the Owner shall first have discharged or paid the same out of funds belonging to it unconditionally and not by way of loan or otherwise.

The effect of the rule is that a member must first discharge the liability or expense by payment and thereafter recover from the Club. Prior to discharge or payment the member’s right is contingent and only following payment is the right to be indemnified established.89 There can be no doubting the validity of such a provision. The ‘pay first’ rule is jealously protected by Group Clubs because they regard it as an essential condition of mutual insurance that members possess the financial resources to respond to calls. The condition is made even more demanding by the prohibition on borrowing to fund the precondition. There are, however, possible concessions. The directors have discretion not to insist on the performance of the condition precedent. This would likely occur when a Club settles directly with third parties. Also when if it would facilitate an out-of-court settlement. For some time it has been the practice of Clubs not to apply the condition precedent to death and personal injury claims, an exception which is now included in the Rules.90 It continues to be applicable to These words are defined in UKP&I Rule 44 (Definitions). UKP&I Rule 5(B)(i). 84 UKP&I Rule 5(B)(i). 85 UKP&I Rule 44 (Definitions) for the meaning of ‘policy year’. 86 Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn, Ch. 20. 87 See n.83. Post Office v Norwich Union Fire Ins. Society Ltd [1967] 2 Q B 363, 373(CA), per Lord Denning MR; Bradley v Eagle Star Ins Co Ltd [1989] AC 957. 88 The Fanti and Padre Island above, per Lords Brandon and Goff. 89 The Fanti and Padre Island above, per Lord Goff. 90 UKP&I Rule 1(9). 82 83

36  Research handbook on marine insurance law third party direct rights of action against Clubs, but again not in relation to death and personal injury claims.91 It may, nevertheless, be excluded in third party rights of action introduced by particular statutes giving effect to international maritime conventions.92 3.2.1 Discretionary cover Another characteristic is that cover may be dependent on the exercise of discretion in the management of the insurance by the directors, management committee or managers, as the case may be. This is a significant feature of the cover and a substantial point of contrast with market insurance. The discretion may come into play in different ways. It may relate to the availability of cover, the application of conditions, limitations and exclusions, and the claims procedure. The first of these possibilities is of material significance to the theme of this contribution. 3.2.2 Omnibus rule The commonly labelled ‘Omnibus Rule’ is set out in Rule 2, Section 24, of the UK P&I Rules, under the title ‘Expenses incidental to the operation of ships’. It provides cover for: Liabilities, costs and expenses incidental to the business of owning, operating or managing ships which in the opinion of the Members’ Committee fall within the scope of the Association; PROVIDED ALWAYS that: a) Subject to paragraph (b) of this proviso there shall be no recovery under this Section in respect of liabilities, costs and expenses, which are expressly excluded by other provisions of these Rules; b) […] c) Any amount claimed under this Section shall be recoverable to such extent only as the Members’ Committee in its discretion may determine without having to give any reasons for its decision.

The effect is that a claim which is not expressly covered or excluded93 under the Rules, but which is considered by the Members’ Committee to be ‘incidental’ to the business of owning, operating or managing ships, may be indemnified as a claim under the insurance, with the sum payable again determined in the discretion of the Committee. Within the boundaries of the provision, all is left to the discretion of the Members’ Committee, without the requirement to give reasons. The member has no enforceable right to payment, the member is at the mercy of the Members’ Committee. Many questions may be raised about the way this provisions works in practice. There is an overriding duty on the Club and those who act on its behalf to manage and determine issues fairly.94 The broad impression that prevails is that only in exceptional circumstances would a claim be indemnified under this provision.95 This position is again very different from market insurance where predominantly the strict terms of the insurance prevail. An assured

Third Parties (Rights against Insurers) Act 2010, ss.9(5) and (6). See n.10 above for the relevant international maritime conventions. The legislation incorporating or giving effect to them establishes third party rights of action against insurers which are not subject to the ‘pay to be paid’ condition. See also n. 16 above and Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn, Ch. 27. 93 Subject to the exception in para (b); see UKP&I Rules, Rule 2, 24(b) referring to R. 5(G). 94 The CVG Siderurgicia del Orinoco SA v London Steamship Owners Mutual Insurance Association Ltd (The Vainqueur Jose) above n.57. 95 Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn, 10.277 et seq. 91 92

Protection and indemnity insurance: is it truly insurance?  37 with a claim outside the express terms of the insurance stands little chance of a recovery.96 The discretion in P&I insurance is probably to be explained by reference to the mutual character of the insurance and the resulting concern that members are protected from risks that genuinely may be associated with their respective positions in the business of shipping. 3.2.3 Other circumstances Beyond the generality of the Omnibus Rule there are individual claims which may be wholly or partly dependent on the discretion of the Members’ Committee. Liabilities arising under an indemnity or contract entered into by a member with regard to services rendered by an entered ship are recoverable only in specified circumstances, one of which is when ‘the Members’ Committee in its discretion decides that the owners should be reimbursed’.97 And certain specified cargo claims are ‘payable only at the discretion of the Members’ Committee’.98 In the face of an exclusion from cover, the Members’ Committee may have discretion to allow the claim to the extent it considers appropriate. This is narrower than the discretion under the Omnibus Rule because it applies only to particular claims that would otherwise be excluded. An example is provided in the body of the Omnibus Rule,99 which provides that ‘The Members’ Committee may authorise payment of claims which are excluded by Rule 5(G) of these Rules but only if a majority of threequarters of those members of the Members’ Committee present when the claim is considered so decide’. This is an exception to the general rule that discretion under the Omnibus Rule cannot be exercised in relation to an exclusion from cover. Rule 5(G) relates to exclusions from cover, but, nonetheless, the Members’ Committee may in its discretion authorise payment, provided it is supported by 75 per cent of the members present at the time the claim is considered. Another interesting example is provided in The Vainqueur Jose.100 The question in issue was whether a claim for forwarding expenses in relation to cargo was recoverable. The applicable rule in the case provided: Unless the Committee shall in its sole discretion otherwise determine, there shall be no recovery under this rule in respect of a Member’s liability for the cost of forwarding cargo to the port of destination stipulated in the contract of carriage from another port at which the cargo was discharged from an entered ship.

This provision appeared in the rule setting out the risks covered in relation to cargo carried on board an entered ship. Part of the claim against the Club related to forwarding expenses and was rejected. In the litigation that followed in which the refusal was challenged, the judge summarised the submission on behalf of the defendant Club in the following terms.101 The rule ‘made it clear that there was no insurance of a claim in relation to forwarding expenses, but

98 99

Ex gratia and without liability payments are not unknown. UKP&I Rule 2(14). UKP&I Rule 2(17(c). UKP&I Rule 2(24)(b). 100 C.V.G. Siderurgicia Del Orinoco S.A. v London Steamship Owners’ Mutual Insurance Association Ltd (The Vainqueur Jose) above n.57. 101 London Steamship Owners’ Mutual Insurance Association Ltd. 96 97

38  Research handbook on marine insurance law merely the possibility of what one might regard as an ex gratia payment in respect of them’.102 The Committee refused to exercise discretion in favour of the member or the plaintiff standing in the shoes of the member. It regarded the exclusion as a proper disincentive against any over-readiness to abandon a voyage. The judge considered that this may not have been the only reason for the decision. In the result the court refused to interfere, for there was no reason to consider the refusal improper or unjustified or misguided.103 Many of the instances identified could doubtless be analysed by reference to the general principles relating to waiver and estoppel. Although there are limits, it is generally open to a party to waive or to be estopped from asserting a condition, limitation or exclusion in an insurance contract. For example, sue and labour costs are recoverable provided they have been incurred by the member with the prior agreement of the managers. Nonetheless, even in the absence of prior agreement, they are recoverable if the member’s committee in its discretion decides that the owner should recover, and the extent of the recovery.104 Again, for the period that class and statutory conditions are not fulfilled, the member is not entitled to any recovery unless and to the extent the member’s committee otherwise decides.105 In other words, in each instance the pre-condition may be waived.

4. CONCLUSION It is clear that the insurance provided by Group Clubs is in its form, structure and management significantly different from market insurance. Apart from the nature of the risks covered, the administration of the cover is closely allied to the managerial structure of the Club, with the directors, members’ committee106 and managers playing material roles. This model exemplifies the mutual nature of Clubs, with its strong emphasis on serving the interests of members. This objective is best achieved by the establishment of a stable and flexible managerial platform. Some cover is conditional on post-entry consent. This may relate to the terms and conditions on which continuing cover depends or the procedure for making claims. They may, however, be effectively waived by the relevant managerial entity, although the process is not expressly described in the language of waiver in the rules. These decisions involve the exercise of discretion but there is nothing particularly exceptional about the process. It runs in parallel with the general legal concepts of waiver and estoppel.107 However, there are occasions when the process assumes a very different character, when discretion alone is the basis on which a beneficial payment is determined. The Omnibus Rule provides the clearest example.108 It is this provision, probably more than any other, that explains why the status of P&I cover as insurance is sometimes questioned. It has been observed that one essence of insurance is the existence of an ‘enforceable right’ to the agreed benefit. To possess ‘merely a right to have [a] request fairly considered’ is not Above, p.580. Above, p.580 per Mocatta J. 104 UKP&I Rule 2(25). 105 UKP&I Rule 5(k). 106 Or an equivalent entity. 107 See, generally, Treitel, The Law of Contract, 15th ed, Edwin Peel (2020, Sweet & Maxwell, London), 3-066 et seq. 108 See under the title ‘Omnibus Rule’. 102 103

Protection and indemnity insurance: is it truly insurance?  39 sufficient.109 If all that a party has is the contractual right to request that claims be considered and determined on a discretionary basis the agreement does not amount to insurance.110 This principle appears not necessarily to apply to the entirety of the cover. If there exists a mix of ‘enforceable rights’ and ‘rights to make discretionary requests’ the status of the agreement as a contract of insurance survives.111 What is not clear is whether the two groups are to be separately categorised for the purpose of legal analysis.112 This is the position that prevails with regard to the cover provided by Group Clubs. It cannot be said that the cover is in its entirety based on discretion; it is a mix of legal entitlements and requests to have claims considered.113 There is, as yet, no indication in the law that the ratio between the two categories may be of relevance. The same result may follow even when the part relating to ‘discretionary requests’ is dominant.114 There are two makeweight considerations to this conclusion. First, issues relating to P&I cover have been contested in litigation and arbitration for a considerable period of time without the status of the cover as insurance being seriously contested. Of particular note are The Vainquer Jose115 and The Fanti and The Padre Island.116 The first case is a significant authority which analysed inter alia the exercise of discretionary powers under the rules. In a comprehensive and extensive judgment, at no time did Mocatta J cast doubt on the insurance provided by the Club in question.117The second case is a decision of a unanimous House of Lords confirming that a Club may raise a defence based on the ‘pay to be paid’ provision against a statutory third party claimant. Lord Brandon, who delivered the leading speech, observed: ‘When shipowners enter one of their ships in a P&I club there comes into being a policy of marine insurance relating to that ship on the terms of the club’s rules.’118 That dictum is unequivocal and has never been questioned. Also noteworthy is The Allobrogia where, in a case relating to third party rights against a P&I Club, Slade J considered, albeit cautiously and without wishing to lay down a binding precedent, the relation between Club and member to be based on a contract of insurance.119 The second supportive, and more mundane, reason is that the commercial and maritime markets have been of the same clear opinion for some time, and, it is suggested, with good reason. There would appear to be only one possible conclusion. Not only does P&I cover look like insurance, it is insurance.120 The insurance falls within the parameters of the Marine Insurance

See nn.61 and 62. Hampton v Toxteth Co-operative Provident Society Ltd above. 111 Cf C.V.G. Siderurgicia Del Orinoco S.A. v London Steamship Owners’ Mutual Insurance Association Ltd (The Vainqueur Jose); Tilley, above, at p.442. 112 C.V.G. Siderurgicia Del Orinoco S.A. v London Steamship Owners’ Mutual Insurance Association Ltd (The Vainqueur Jose) above. 113 Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn, Ch.5. 114 See n.57. 115 See n.57. 116 Firma – Trade S.A. v Newcastle Protection and Indemnity Association (The Fanti) and Socony Mobil Oil Co, Inc. and Others v West of England Ship Owners Mutual Insurance Association (London) Ltd (The Padre Island)(No 2) [1990] 2 Lloyd’s Rep 191. 117 The London Steamship Owners’ Mutual Insurance Association Ltd. 118 Ibid, p.194. 119 In re Allobrogia Steamship Corporation (The Allobrogia) [1979] 1 Lloyd’s Rep 190. 120 Cf. Fuji Finance Inc v Aetna Life Ins Co Ltd, p.189, per Morritt LJ. See also Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn, Ch. 5, paras 5.1–5.9. 109 110

40  Research handbook on marine insurance law Act 1906 and Insurance Act 2015.121 It relates to third party liability incurred by an owner or other person interested in or responsible for an entered ship (insurable property) by reason of maritime perils.122 The cover is a time policy within the meaning of the 1906 Act.123 The Group Clubs domiciled in England all adopt English law as the governing law of the Rules and insurance contracts entered into.124 The Acts of 1906 and 2015 are expressly identified and incorporated save to the extent they have been excluded, as is permitted under the 2015 Act.125 That power has been used to the maximum extent.126 Finally, there is a reservation about the principle that discretion in its fullness is inconsistent with the essence of insurance. The case law at the root of the principle has survived the test of time,127 but it is also of a time when the practice of insurance was more conservative than is presently the case. The innovative products of contemporary insurance markets often challenge conventional ideas about the role and boundaries of insurance. Within the world of P&I insurance, FD&D (legal expenses) cover may be provided by an independent Defence Club, with the cover wholly discretionary.128 At this advanced stage it would be commercially unrealistic to deny this cover the status of insurance. If discretionary benefits in part do not disturb the traditional categorisation of insurance, why should the same outcome not prevail when discretion underpins the entire cover. There is nothing anomalous about discretionary contractual powers. The difference between the two circumstances may be marginal and the fact that discretionary insurance might be less attractive commercially is beside the point. Let the market decide the kind of insurance it requires.

UKP&I Rule 5(L). Marine Insurance Act 1906, section 3, defines ‘maritime perils’. 123 Section 25(1). See, also, Compania Maritima San Basilio S.A. v The Oceanus Mutual Underwriting Association (Bermuda) Ltd (The Eurysthenes) [1976] 2 Lloyd’s Rep 171, 181, per Roskill LJ; In re Compania Marabello San Nicholas S.A. [1973] 75, 81 per Megarry J. 124 UKP&I Rule 42. 125 UKP&I Rule 5(L). 126 Consequently, for example, the remedy for failure to make a fair presentation of the risk continues to be ‘avoidance’ regardless of whether the breach is innocent, deliberate or reckless. See also UKP&I Rule 7. 127 See section titled ‘Distinction of “right” and “discretion”’. 128 Hazelwood & Semark, P&I Clubs: Law and Practice, 4th edn, Ch. 26. 121 122

PART II  

3. Has the principle of ‘utmost good faith’ in marine insurance law become like the Cheshire Cat – now you see it and now you don’t? Richard Aikens

1.

THE ISSUE

The Insurance Act 2015 (‘the IA’) was the result of years of labour by the English and Scottish Law Commissions. It tackled a number of ‘problems’, one of which was the way in which sections 17 to 20 of the Marine Insurance Act 1906 (‘the MIA’) dealt with the concept of ‘utmost good faith’ in insurance contracts.1 Section 17 of the MIA had set out a general principle that ‘[a] contract of marine insurance is a contract based upon the utmost good faith and, if the utmost good faith be not observed by either party, the contract may be avoided by the other party’. Sections 18 to 20 of the MIA dealt with particular applications of the general principle in relation to the duty of the insured2 and its agent concerning disclosure and representations prior to the formation of the contract of insurance, which were material to the risk to be insured. There was no statutory elaboration of any duty of utmost good faith on the part of the insurer prior to the formation of the contract. Nor did any other section of the MIA deal with the question of whether there was any continuing ‘duty’, ‘principle’ or ‘obligation’ of utmost good faith by either the insured or insurer relating to the performance of contract of insurance after its formation. All those issues were left to the courts to work out.3

It had long been accepted that the terms of sections 17 to 20 of what became the MIA 1906, which had been drafted and redrafted by the great Parliamentary Draftsman Sir Mackenzie Chalmers through its long gestation in Parliament, represented a codification of the common law on the obligation of ‘utmost good faith’ in relation to contracts of both marine insurance and non-marine insurance. See for example the statement of Lord Mustill in Pan Atlantic Insurance Co Ltd v Pine Tope Insurance Co Ltd [1995] 1 AC 501 at 518D. All the other four law lords agreed on this point, which was not contentious. 2 The MIA refers always to ‘the assured’, but the more modern term ‘insured’ will be used here. 3 Lord Mansfield had himself given a possible example of a duty of the insurer prior to the formation of the contract in Carter v Boehm (1766) 3 Burr, 1905 at 1909. The leading House of Lords case on the nature and extent of the insured’s post-formation obligation of utmost good faith before the IA came into force was Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd: The ‘Star Sea’ [2003] 1 AC 469, hereafter ‘The Star Sea’. The vexed issue of whether the duty of the insured not to make a fraudulent claim was based upon an obligation of ‘utmost good faith’ post-contract formation or some other implied term in the contract of insurance was debated, but not resolved in Versloot Dredging VB v HDI Gerling Industrie Versicherung AG: The ‘DC Merwestone’ [2017] AC 1 (hereafter ‘Versloot’); see the views of Lord Sumption at [8] and Lord Hughes at [54]–[55]. Even if, at common law, an insurer might have had a legal duty to pay a claim in a reasonable time, there was no remedy available: Sprung v Royal Insurance (UK) Ltd [1999] Lloyd’s Rep IR 111 (CA). (But it was not argued in that case that any duty rested on utmost good faith). On this last point see now s.13A of the IA and further below. 1

42

Has ‘utmost good faith’ become like the Cheshire Cat?  43 In Part 5 of the IA, entitled ‘Good Faith and Contracting Out’, section 14(1) abolishes ‘any rule of law permitting a party to a contract of insurance to avoid a contract on the ground that the utmost good faith has not been observed by the other party’. Section 14(2) states that ‘any rule of law to the effect that a contract of insurance is a contract based on the utmost good faith’ is modified ‘to the extent required by the provisions of this Act’ and also the provisions of the parallel Act dealing with consumer insurance contracts. It followed, as set out in section 14(3)(a) of the IA, that the last part of section 17 of the MIA had to be repealed, or ‘omitted’ as the wording of the IA coyly puts it. The consequence of these statutory changes, which are, obviously, applicable to non-consumer insurance contracts of both the marine and non-marine variety, is not clear.4 Does there remain some sort of legal ‘principle’ or ‘obligation’ of utmost good faith between the parties to the contract and does it operate both before and after its formation? If there is some remaining legal principle, a failure to observe this does not have the legal consequence it used to; so can it have any other legal consequences? If there is no legal ‘principle’ or ‘obligation’ or no legal consequences for its breach, then what is the point of retaining a statutory ‘principle’ or ‘obligation’ of utmost good faith at all? The Law Commissions apparently thought the principle would have some use, saying in its final report that it was intended to retain the role of ‘utmost good faith’, both as an ‘interpretive principle’5 with regard to the new duty of ‘fair presentation’ and also in relation to whether to imply contractual terms in a policy. The Law Commissions also considered that the principle might assist courts in future and possibly ‘provide a solution in an especially hard case or emergent difficulty’.6 The Law Commissions noted that there had been substantial opposition to the idea that damages should be made available for breach of the duty of utmost good faith by an insurer. Thus it stated that ‘we do not make any recommendation along these lines’.7 The fact remains that section 17 of the MIA, even as amended, states that a contract of marine insurance is a contract ‘based upon the utmost good faith’. Section 17 is, even in its new form, a codification of the common law concept of ‘utmost good faith’ in relation to insurance contracts. The issue that this chapter attempts to tackle is whether that concept has any substance and whether, despite the Law Commissions’ refusal to make any recommendations

4 For a general review see B Soyer and AM Tettenborn: ‘Mapping (utmost) good faith in insurance law-future conditional?’: 132 LQR (2016) 618. 5 The current editors of Arnould on the Law of Marine Insurance and Average (20th ed. 2021) concluded that the ‘interpretive principle’ was unlikely to have any role to play in relation to Part 2 of the IA 2015. Effectively, the editors appear to regard s.17 of the MIA, as amended, as a dead-letter: see paras 15–13 and 18B-51. 6 Paras 30.22–23 of the joint report of the Law Commission and the Scottish Law Commission ‘Insurance Contract Law: Business Disclosure; warranties, insurers’ remedies for fraudulent claims; and late payment’ (Law Com No 353) (Scots Law Com No 238), Cm 8898, published July 2014: hereafter ‘LC 2014’. 7 LC 2014 para. 30.24, referring to responses made to the idea that damages should be made available for breach of the duty of utmost good faith by the insurer. This, however, was in the context of the issue of late payment of claims. See: CP2, Insurance Contract Law: Post Contractual Duties and other Issues, Law Commission Consultation Paper No 201, Scottish Law Commission Discussion Paper No 152, (December 2011), paras 4.31 to 4.41. See also para. 116 of the Explanatory Notes to the IA and below.

44  Research handbook on marine insurance law for remedies for any ‘breach’ of that principle by either party,8 the existing law can provide a possible answer.9 Hence the reference to the Cheshire Cat: will the principle disappear as soon as you try to grasp it, or will it purr – or growl, as the Cheshire Cat would have put it?10

2.

THE CONTROVERSIES RELATING TO THE PRINCIPLE OF ‘UTMOST GOOD FAITH’ BEFORE THE INSURANCE ACT 2015

As noted, there are four different circumstances where, before the reforms in the IA, the courts had to consider the question of whether a duty of utmost good faith arose and the consequences of its breach. The first pair concern the two parties to the insurance contract in the period before its formation; the second pair concern both parties after the contract’s formation. So far as the pre-contractual situations were concerned, the two most controversial issues were, first, the wide-ranging scope of the duty of the insured with respect to disclosure and representations of the risk and the rule that any breach of those duties gave the insurer the automatic right, in all cases, to avoid the contract altogether;11 second, the corollary to the existence of the remedy of avoidance for breach of utmost good faith. The courts concluded that there was no remedy of damages for breach of the duty. This meant that if an insurer was held to be in breach of a duty of utmost good faith before the contract was concluded, an insured could not obtain damages and, of course, the ‘remedy’ of avoidance was, in virtually all cases, of no practical use.12 8 As opposed to the particular duties of the insured in relation to ‘fair presentation’ of the risk, set out in Part 2 of the IA, and the consequences of an insured making a ‘fraudulent claim’: see Part 4 of the IA. 9 For an analysis which suggests that the role of ‘good faith’ in insurance contracts is one that is now diminished, perhaps to vanishing point, see Margaret C Hemsworth: ‘The fate of “good faith” in insurance contracts’ [2018] LMCLQ 143. 10 Alice in Wonderland Ch 6. The Cheshire Cat insisted that it growled when pleased, and wagged its tail when angry. 11 As the CA noted in Banque Kaiyser Ullman SA v Skandia (UK) Insurance Co Ltd [1990] 1 AB 665 at 771, the insured’s obligation in relation to pre-formation disclosure and representation was an ‘absolute one which is not negatived by the absence of fraud or negligence’. Insurance lawyers would habitually divide the types of non-disclosure or misrepresentation into three categories: fraudulent, negligent or innocent. 12 See Banque Keyser Ullman SA v Skandia (UK) Insurance Co Ltd [1990] 1 QB 665. Steyn J at first instance had held that the insurer was in breach of a pre-formation duty of utmost good faith and he was prepared to award damages for the breach. The CA upheld the existence of the duty and its breach but concluded that no damages remedy was available. Although the appeal to the House of Lords was dismissed on different grounds, Lords Templeman and Jauncey agreed (obiter) with the view of the CA that there was no remedy of damages for breach of the duty of utmost good faith: Banque Financière de la Cité SA v Westgate Insurance Co Ltd [1991] 2 AC 249 at 280 and 281. In Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd (the ‘Good Luck’) [1990] 1 QB 818 the CA rejected the conclusion of the trial judge, Hobhouse J, that the juridical nature of the post-formation duty of utmost good faith was contractual in nature; the CA held it was based on the ‘general law’, with the conclusion, following the Skandia decision, that there was therefore no remedy in damages. The CA in The Good Luck accepted that if the post-formation duty of utmost good faith were based on an implied term, then a breach of it would have sounded in damages: see 886. The premises for these conclusions have been doubted by a convincing analysis of the cases: Howard N Bennett: ‘Mapping the doctrine of utmost good faith in insurance contract law’ by [1999] LMCLQ 165, in particular at fns 101–3, and see further below.

Has ‘utmost good faith’ become like the Cheshire Cat?  45 On the position after the formation of the contract, the debate had concentrated on two aspects in particular: first, whether there was any continuing ‘duty’ of utmost good faith based on a principle of ‘general law’ as opposed to being based on an implied term in the contract of insurance; second, whether the duty of the insured not to make a fraudulent claim is one aspect of a duty of utmost good faith or entirely independent of that. As noted,13 these controversies had not been resolved when the IA came into force on 12 August 2016.

3.

WHAT ISSUES ARE RAISED BY THE AMENDMENTS TO SECTION 17 OF THE MIA MADE BY SECTION 14 OF THE IA?

The first issue must be whether the amended version of section 17 of the MIA (now reading ‘a contract of marine insurance is based upon the utmost good faith’)14 creates any separate and enforceable legal duty at all on the part of either or both the insurer and insured, and, if so, what its nature might be. The second issue, assuming the section does create legal duties, is whether any legal duty before the contract exists independently of those created by Part 2 of the IA before the contract is concluded. If some duty does exist, does it relate only to the insurer, or is there another duty of utmost good faith on the insured in addition to the statutory duties of ‘fair presentation’ dealt with by Part 2 of the IA?15 The third issue is whether, given the terms of section 12 of the IA (on the consequences of an insured making a ‘fraudulent claim’) and the introduction of section 13A of the IA16 (creating an implied term that an insurer will pay a valid claim within a reasonable time), there is scope for any further duty of utmost good faith on either party after the formation of the contract and, if so, when it is triggered. The fourth issue which must arise whether considering either a pre or post-formation duty of utmost good faith is: what remedies (if any) are available for a breach of such duty? In particular, can the courts now re-open the question of whether damages for breach of a duty of utmost good faith are a possible remedy in the light of the changed wording of section 17?

4.

DOES THE AMENDED VERSION OF SECTION 17 OF THE MIA CREATE ANY ENFORCEABLE LEGAL DUTY ON THE PART OF EITHER THE INSURED AND INSURER AND IF SO WHAT IS ITS NATURE?

The answer to the first part of this question must depend upon the correct construction of the amended section 17. It has to be assumed that Parliament was aware of the previous common law on ‘utmost good faith’ in the context of insurance contracts, the previous wording of the See fn 3 above. Italics added. 15 In Peter Macdonald Eggers KC and Sir Simon Picken, Good Faith and Insurance Contracts (4th ed. 2018), the authors appear to treat the duty of ‘fair presentation’ created by the IA as an aspect of the obligation of utmost good faith rather than a separate statutory duty: see ch. 7. Contrast Colinvaux’s Law of Insurance (13th ed. 2022) edited by Prof Robert Merkin KC, LLD and others, at para 6-018 (hereafter Colinvaux). 16 Added by s.28 of the Enterprise Act 2016. 13 14

46  Research handbook on marine insurance law unamended section 17 of the MIA and all of the decisions of the courts on the construction and effect of the old version. The Law Commissions’ reports and the Explanatory Notes can also be taken into account in interpreting the amended section 17. Paragraph 116 of the Explanatory Notes to the IA states: ‘The intention is that good faith will remain an interpretive principle, with section 17 of the 1906 Act and the common law continuing to provide that insurance contracts are contracts of good faith.’ Unfortunately, none of these aids to interpretation leads to an unequivocal answer to the correct construction of the revised section 17. Perhaps the key lies in the last phrase of paragraph 116 of the Explanatory Notes set out above. English law now appears to recognise that in certain types of contract, dubbed ‘relational contracts’, a term can be implied (if it is not express) by which the parties will owe a duty to one another to act in ‘good faith’ in the formation and the performance of the contract.17 Insurance contracts are very often not ‘relational contracts’ in the sense developed in the cases since Yam Seng. However, if, as stated in Explanatory Note paragraph 116, insurance contracts are to be regarded as contracts of ‘good faith’, then the phrase ‘a contract based on the utmost good faith’ in section 17 must mean the contracts are based on a legal principle of ‘good faith’. To give that concept substance, it must involve the existence of some legal right and correlative duty on the parties to the contract, which duty is operative in certain circumstances. As Lord Mustill pointed out in his seminal speech in Pan Atlantic Insurance Ltd v Pine Top Ltd,18 the doctrine that a contract of marine insurance is one uberremae fidei had been ‘firmly established for decades’ before the great writers on insurance in the nineteenth century (Duer, Parsons, Phillips, Arnould, Chancellor Kent and, earlier, Park). Section 17 of the MIA put the common law doctrine into statutory form. The existence of a legal obligation and possible correlative rights, unless expressly abolished (as was the right to avoid for breach), must surely remain, given its firm place in legal doctrine and history for more than 200 years. If there is such a legal principle, then the second part of this first issue becomes relevant: is this legal right/duty one that is based on the general law or an implied term of the contract of insurance that is created? The significance of the juridical basis of the obligation lies in the remedies that might follow from the correct characterisation.19 Under the previous law, the Court of Appeal decided in two cases that the obligation of disclosure by an insurer, in relation to a contract of marine insurance, was based on a general law principle, rather than being an implied term of the contract itself. The first case concerned the position before the contract was concluded.20 The second concerned disclosure after the 17 The sequence of cases started with the decision of Leggatt J in Yam Seng Pte Ltd v International Trade Corp Ltd [2013] EWHC 111 (QB). A ‘checklist’ of indicators for a ‘relational contract’ was compiled by Fraser J in Bates v Post Office (No 3) Common Issues [2019] EWHC 606 (QB) at [725–6] which was approved by the CA in Candy v Bosheh [2022] 4 WLR 84 at [31]. In Times Travel (UK) Ltd v Pakistan International Airlines Corpn [2023] AC 101, the Supreme Court appeared to reject a general principle of ‘good faith’ in the formation of contracts, at least in the context of ‘lawful means duress’: see [1] and [95]. 18 [1995] 1 AC 501 at 543. 19 As is pointed out in Colinvaux at 6-019. 20 Banque Keyser Ullmann SA v Skandia (UK) Insurance Co Ltd [1990] 1 QB 665; see the discussion relating to the position concerning a failure to disclose before the contract is made under Issue 5, at 773–80. Having held that there was a duty but it arose from ‘the general law’, the CA then concluded that the duty of disclosure was not a tort and that a breach of the duty to disclose gave rise to no right to damages. As already noted, in the House of Lords, sub nom. Banque Financière de la Cité SA v Westgate

Has ‘utmost good faith’ become like the Cheshire Cat?  47 contract was concluded.21 Both cases also concluded that the duty of utmost good faith was not a tort which, if committed, gave rise to a right to damages.22 Those decisions were the subject of a rigorous analysis by Professor Howard N. Bennett.23 He convincingly challenged both their conclusions, viz that the duty of utmost good faith was non-contractual and that a breach could not give rise to a right to damages. What is the position for the revised wording of section 17 of the MIA? First, if the revised section 17 creates a legal right and a correlative duty, is this contractual in nature? In the absence of any express term, does it result in an implied term in the contract of marine insurance? The foundations for the conclusion of the Court of Appeal in the Banque Keyser Ullmann case rested on three matters: first, the terms of section 17 as it then existed; second, the balance of previous case law, in particular the dissenting judgment of Lord Esher MR in Blackburn, Low & Co v Vigors,24 which dissent was upheld in the House of Lords;25 and, lastly, the alleged origin of the right to avoid a contract (including an insurance contract) for misrepresentation being ‘by reason of the jurisdiction originally exercised by the courts of equity to prevent imposition’.26 As Professor Bennett has demonstrated, these foundations were not rock-solid even with the old section 17. But when the only remedy for breach of the obligation expressly recognised in section 17, viz avoidance of the contract, is removed, then the foundations for the conclusion of the Court of Appeal in the Banque Keyser Ullmann case are swept away. All need to be reconsidered. First, the revised section 17 does not refer to any remedy for breach of ‘utmost good faith’ by either party. Either that means that the statement gives rise to no legal right, or if, as is argued above, a legal right is still created by the revised section, the new wording must leave open the question of possible remedies for its breach. The old (originally Latin) maxim of ‘where there is a right there is a remedy’ should apply.27 Given the history of section 17 of

Insurance Co Ltd [1991] 2 AC 249, the appeal was dismissed on other grounds, but Lords Templeman and Jauncey accepted the Court of Appeal’s analysis on this issue: see 280 and 281. 21 Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd [1991] 1 QB 818, in particular at 885–8. The CA expressly accepted that if the duty of utmost good faith were based on an implied term in the contract then, if that term were broken, it would give rise to a remedy of damages for breach: see 886E. See also the Banque Keyser Ullmann case at 778. After the two CA decisions, in Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co Ltd [1995] 1 AC 501, Lord Mustill, giving the leading speech of the majority, said that the ‘controversy’ as to the legal source of ‘the mechanism whereby the validity of the [insurance] contract was compromised’ where there had been non-disclosure or misrepresentation, had still not been resolved: see 544. 22 See respectively: [1990] 1 QB 665 at 780–1 and [1990] 1 QB 818 at 888. 23 ‘Mapping the doctrine of utmost good faith in insurance contract law’ [1999] LMCLQ 165, hereafter ‘Mapping’. In The Star Sea, Lord Hobhouse described the article as containing ‘a penetrating discussion of the conceptual difficulties’ of good faith in insurance law: see [42]. 24 (1886) 17 QBD 553 at 561–2; hereafter ‘Blackburn, Low’. 25 (1887) 12 App Cas 531 at 539 per Lord Watson and 542 per Lord Fitzgerald. But Lord Watson’s formulation is equivocal, characterising full disclosure of material facts as a ‘condition precedent of every contract of marine insurance’; in short, putting it in contractual terms. Lord Halsbury LC, at 536, doubted a resolution that the debate on whether the obligation of disclosure was contractual or not would help in deciding the case, so he hardly gave a ringing endorsement of the view of Lord Esher. 26 See the Keyser Ullmann case at 779, quoting the judgment of Luxmoore LJ in Merchants and Manufactuirers Insurance Co Ltd v Hunt [1941] 1 KB 295 at 318. The remark, with which Scott LJ agreed at 313, was obiter. No authority for this wide proposition was cited by Luxmoore LJ. 27 William Blackstone, Commentaries on the Laws of England vol 3 at 23.

48  Research handbook on marine insurance law the MIA and the fact that the remedy of avoidance for breach of the duty of utmost good faith has now been abolished by section 14 of the IA, if there is to be a remedy at all, then that of damages for a breach has to be one obvious choice. It must be accepted that damages could be a remedy for either a breach of contract or a tort. Thus it still prompts the question of juristic analysis: is the nature of the duty of utmost good faith contractual or tortious, or, as some of the cases put it, a duty which arises in ‘general law’? This leads on to the second foundation of the reasoning of the Court of Appeal in Keyser Ullmann: its reliance upon the reasoning of Lord Esher in Blackburn, Low. The issue in that case was whether non-disclosure of material facts by a broker who did not actually place the risk could vitiate the policy. The majority of the Court of Appeal said it did; Lord Esher dissented. The passage of Lord Esher’s judgment referred to in Banque Keyser Ullmann occurs in the part of it where he investigates ‘what circumstances have been held to prevent the enforcement of a contract of insurance and what are the principles under which those circumstances have been held to have that effect’.28 Lord Esher quotes from three principal (and influential) eighteenth and nineteenth-century authors of the texts on insurance: Arnould on Marine Insurance; Duer on The Law and Practice of Marine Insurance; and Phillips’ Treatise on the Law of Insurance.29 Lord Esher states that Duer ‘holds that it is a part of the contract that full disclosure shall be made as well as that every representation shall be accurate’. Lord Esher comments that if this is so then ‘the contract should never be set aside or treated as void on the ground of concealment; the contract should stand and be treated as broken by the assured’. This does not follow, particularly as the common law right to avoid an insurance contract for concealment existed. Lord Esher then goes on to quote from Phillips, whom he describes as ‘the more accurate guide’. Phillips rejects the analysis of both Arnould and Duer. He argues that the ‘practical doctrine’ can be stated in ‘direct terms’ which are ‘that it is an implied condition of the contract of insurance that it is free from misrepresentation or concealment, whether fraudulent or through mistake’; and then ‘the forfeiture of the insurance by misrepresentation or concealment is a forfeiture by a breach of condition of the contract. So it seems also to have been considered by Chancellor Kent.’30 As Professor Bennett commented in Mapping, all these references to conditions of the contract strongly suggest that both judges and commentators considered that the duty of utmost good faith was of a contractual nature.31 Given that the remedy of avoidance for ‘concealment’ or misrepresentation in the formation of the insurance contract had long been firmly established, it is unsurprising that the condition should be characterised as a ‘contingent’ condition rather than as a promissory one. In William

At 560. As Lord Mustill pointed out in Pan Atlantic, in the nineteenth century the ‘writers of the texts’ were a far more potent source of general principle of marine insurance than ‘the scattered decisions of the courts’: [1995] 1 AC 501 at 537. 30 The italics are as in the law report. James Kent (1763–1847) was successively a law professor at Columbia University and the equity judge in the New York Supreme Court, then subsequently Chief Justice of that court. He then became ‘Chancellor of New York’ from 1814 to 1823. It is from this appointment that he derived the title ‘Chancellor Kent’ by which he was known throughout the rest of his life and to posterity. His ‘Practical Treatise on Commercial and Maritime Law’ was published in 1837. See: John H Langbein: ‘Chancellor Kent and the history of legal literature’: 93(3) (1993) Columbia Law Review 547. 31 Lindley LJ, one of the majority, also referred to ‘a condition of the contract’: see 562. 28 29

Has ‘utmost good faith’ become like the Cheshire Cat?  49 Pickersgill & Sons v London and Provincial Marine & General Insurance Co Ltd,32 Viscount Sumner, then Hamilton J, held that the judgments of the Court of Appeal and of Lord Watson33 in Blackburn, Low constituted binding authority for the proposition that the duty of utmost good faith was the subject of an implied term of the contract.34 It is strongly arguable that the analysis of the remarks of Lord Esher in Blackburn, Low by the Court of Appeal in Keyser Ullmann was misplaced.35 If, as Hamilton J stated firmly, the obligation of disclosure and the consequence of a failure were all founded on an implied contractual term, then two further questions arise. First, is it still reasonable and necessary in order to give a contract of marine insurance business efficacy that a term creating a duty of utmost good faith must be implied?36 While the scope of the term can be debated, the general answer must be ‘yes’, because of the terms of section 17 of the MIA as amended, the nature of a marine insurance contract and the expectations of the parties, backed by more than 200 years of legal history. There will be debate as to whether the term extends to pre-formation duties on the insured, given the statutory code in Part 2 of the IA. There must also be a debate as to the necessity for a general implied duty post-formation in the light of section 12 (consequence of fraudulent claims) and section 13A (implied duty to pay valid claims within a reasonable time) of the IA. These issues are considered below. A question which would then arise must be: what is the precise nature of the contractual implied term? With the change made to section 17 of the MIA by the IA, which abolishes the right to avoid the contract for breach of the duty of utmost good faith, it is impossible to describe the proposed implied term as a ‘condition precedent to performance’ as stated in Keyser Ullmann. The only alternative would be that the contractual term is a ‘promissory’ one. The implied term would therefore be that both parties promise that they have entered into the contract in the ‘utmost good faith’ and that they will perform it in the same manner. There is nothing strange in this: although the acts or omissions occur before the contract is concluded, why should that prevent parties from promising each other than what they have done was done ‘in the utmost good faith’? After all, if there is an allegation of a breach of this term, it is [1912] 3 KB 614 at 621, where Hamilton J states: ‘the rule imposing an obligation to disclose upon the intending assured does not rest upon a general principle of common law, but arises out of an implied condition, contained in the contract itself, precedent to the liability of the underwriter to pay.’ Mr Mackinnon (later Mackinnon LJ) had specifically argued that the proposition that ‘the avoidance of policies of marine insurance through non -disclosure of material facts is part of the general law’ was wrong; he argued that it arose from a breach of a condition in the contract, which was a condition precedent to performance of the contract by the insurer. 33 (1886) 12 App Cas 531 at 539. 34 As stated, albeit obiter, by the distinguished commercial judge Parke B in Moens v Heyworth (1842) M&W 147 at 157–8, although, as the CA pointed out in Keyser Ullmann, that was not an insurance case. 35 In the 16th edition of Arnould on the Law of Marine Insurance and Average (1981, editors Sir Michael Mustill and Mr Jonathan Gilman – perhaps the two most learned scholars on marine insurance of the day, as Mr Gilman KC is today as lead editor of Arnould) took the view that the basis for the rules on misrepresentation and non-disclosure in section 17-20 of the MIA was to be found in an implied condition in the contract that there was no misrepresentation or concealment. They relied on Blackburn, Low. The current editors of the 20th edition have changed their mind, on the basis that this analysis has been disapproved, saying that the matter had ‘arguably’ been determined in the HL in The Star Sea, particularly in the speech of Lord Hobhouse at [45]–[46]. I remain unconvinced for the reasons here given. 36 The ‘reasonable and necessary’ test still applies: Marks and Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2016] AC 742. 32

50  Research handbook on marine insurance law always made after the contract is concluded, never before; the aim of the allegation is always to provide a defence to performance of the contract. It makes sense to link it to contractual terms rather than something called ‘the general law’. The third foundation for the conclusion of the Court of Appeal in Keyser Ullmann on the nature of the duty of utmost good faith was said to be the equitable origin of the powers of the court to give relief in the case of innocent pre-contractual misrepresentation by a party to a contract of utmost good faith. This is not now important, because the remedy of avoidance for breach of the duty of utmost good faith has been abolished in relation to insurance contracts.37 More significant, perhaps, is the statement, obiter, of Lord Atkin in Bell v Lever Brothers Ltd38 that the duty of utmost good faith in insurance contracts ‘does not arise out of contract; the duty of a person proposing an insurance arises before a contract is made’. It is, of course, correct to say that the duty arises before the contract is concluded; but it does not follow that the duty and the rights that flow from it cannot be embodied in an implied term in the subsequent contract. Thus, while acknowledging that Lord Atkin, like his erstwhile teacher Scrutton LJ, was a matchless commercial lawyer, this statement does not consider what was said in Vigors, Low, nor does it consider Hamilton J’s analysis in Pickersgill. It should not be given undue weight. In Keyser Ullmann, once the Court of Appeal had concluded that the duty of utmost good faith was not contractual, it considered whether the duty was tortious in nature. It held that it was not. Four reasons were advanced:39 First was the equity origins of the remedy for non-disclosure in the case of a contract of utmost good faith. As already discussed, this is not solid. The second, based on the Court of Appeal’s decision in Container Transport International Inc v Oceanus Underwriting Association (Bermuda) Ltd,40 is now baseless. The conclusion in that case that there did not have to be inducement of the actual insurer who entered the contract of insurance was overturned by the House of Lords’ decision in Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co Ltd.41 The third reason, that section 17 gives the remedy of avoidance but no other, is also no longer valid. On the current statutory wording there is no basis for a conclusion that Parliament ‘did not contemplate that a breach of the obligation would give rise to a claim for damages’.42 The fourth reason was directed to whether a tort should be recognised. The court held that it would be wrong to create a tort of absolute liability which could give rise to a claim for damages, even in the absence of blameworthy conduct. That reasoning is no longer valid, at least so far as an insured is concerned, in In any case, as Prof Bennett demonstrated in Mapping (at 190–2) the doctrine of utmost good faith in relation to insurance contracts was not, or not solely, of equitable origin. As he demonstrated, it had common law roots as well and there is no warrant for the bald and unconvincing statement of Luxmoore LJ in Merchants and Manufacturers Insurance Co Ltd v Hunt [1941] 1 KB 295 at 318. See also the comments at fn 24 of Prof Soyer’s chapter ‘The insurer’s duty of good faith: is the path now clear for the introduction of new remedies’ in Malcolm Clarke and Baris Soyer, eds, The Insurance Act 2015 (2017). 38 [1932] AC 161 at 227. That too, of course, was not an insurance case. Lord Hobhouse referred with approval to this statement of Lord Atkin at [46] in The Star Sea. However, the derivation of the duty of utmost good faith was not directly in issue in the latter case. 39 [1988] 1 QB 665 at 780–1. 40 [1984] 1 Lloyd’s Rep 476. 41 [1995] 1 AC 501. 42 [1988] QB 665 at 781. In The Star Sea it was conceded by counsel for both parties that, in the light of the case law, in particular Keyser Ullmann, that there was no remedy in damages for any want of good faith, whether it related to matters occurring before or after the formation of the contract: see [49]. 37

Has ‘utmost good faith’ become like the Cheshire Cat?  51 the light of the changes to the law made by Part 2 of the IA. That imposes duties on an insured which are absolute in the sense that breach can occur without negligence of the insured or its agent. The remedies set out in Schedule 1 of the IA do not include damages in the classic sense, but the insurer is compensated for the insured’s breach of its statutory duties. As for an insurer, it depends on the scope of the implied term as to nature of the duty of utmost good faith and the circumstances in which it can be invoked. All these arguments are as yet untested in the courts. The conclusion on this issue in the present circumstances can only be that it is plausible that the revised section 17 of the MIA gives rise to a legal duty of utmost good faith which is contractual in nature and can give rise, in appropriate circumstances, to a remedy in damages.43 It is now necessary to consider when the duty arises, both on the insured and the insurer.

5.

DOES THE LEGAL DUTY OF UTMOST GOOD FAITH NOW APPLY TO PRE-FORMATION ISSUES IN THE CASE OF THE INSURED OR INSURER?

An implied term cannot exist if it is inconsistent with an express term of the contract. The same principle must apply to prevent there being an implied term which would be inconsistent with any obligation that is imposed on a party by statute. In both cases it would be legal nonsense to permit the existence of an inconsistent implied term. That principle must be at the forefront when examining the question of whether there is any general obligation of utmost good faith on either party to a contract of marine insurance concerning pre-formation matters. First, consider the case of the insured. Part 2 of the IA creates detailed obligations on a potential insured to make a ‘fair presentation’ of the risk to be insured. There is no mention of ‘utmost good faith’ in any of the provisions in Part 2. In the light of this detailed statutory framework on the pre-contract duties of the insured, is there any scope for any additional duty of utmost good faith on the part of an insured in relation to pre-formation matters? Although the wording of section 17 as amended makes the concept of a contract ‘based on utmost good faith’ a free-standing one, it is difficult to formulate circumstances in which the duty would apply to the insured pre-formation outside the statutory duty of ‘fair presentation’. It is significant that there were no decisions under the old law which created duties on the insured prior to contract formation apart from those concerning disclosure and representations. The position with the insurer is different, because even after the IA there is no specific statutory regime concerning its pre-contract formation obligations.44 If there are any ‘utmost good faith’ obligations on the insurer they must therefore arise out of a duty imposed by the amended section 17 of the MIA. In Keyser Ullmann the Court of Appeal held that, by virtue of the old section 17, the insurer was under a duty, before the contract was concluded, which extended at least ‘to disclosing all facts known to him which are material either to the nature of the risk sought to be covered or the recoverability of a claim under the policy which a prudent

This possibility is accepted by Colinvaux at 6-019. On the insurer’s duty of good faith and possible remedies for breach post the IA see Soyer, ‘The insurer’s duty of good faith’. 43 44

52  Research handbook on marine insurance law insured would take into account in deciding whether or not to place the risk for which he seeks cover with that insurer’.45 Logically, the same type of duty on an insurer should exist under the revised section 17 of the MIA. It would surely be a major flaw in English law if it were to be held that the obligation that had existed under the old law had been eradicated by the changes made to section 17 on the issue of the remedy of avoidance, but not otherwise. The formulation of the Court of Appeal in Keyser Ullmann identifies a scope of the duty and the need for ‘materiality’. Whether there is now any wider scope to the duty pre-formation of contract is a more difficult question. There were no decisions that established it under the old law. But circumstances might arise where a court has to consider this issue, for example, in the context of insurance contracts (even non-consumer marine insurance contracts) being concluded online. One issue might be: what is the scope of the duty of an insurer in presenting an online form and the terms of insurance to a potential insured, particularly as processes become more automated?46 Failure to present the terms in a manner that constitutes a ‘fair presentation’ of what the insurer proposes by way of cover and terms could be regarded as a breach of a duty of ‘utmost good faith’. The more vexed question is what remedy, if any, an insured may have if there were a breach of the duty by the insurer. That is considered in the last section of this chapter.

6.

TO WHAT DOES THE DUTY OF UTMOST GOOD FAITH APPLY POST-FORMATION OF THE CONTRACT: GENERAL

The cases have distinguished between a more general duty of utmost good faith post-formation of the contract and a particular duty on the part of the insured not to present a fraudulent claim. The first of these received very little judicial attention before the later years of the twentieth century.47 The Star Sea48 was the leading House of Lords case on the insured’s post-formation duty of utmost good faith before the IA was passed. Although the House of Lords examined the issue generally, the actual decision was on the short point that any post-formation of contract duty of utmost good faith on the part of the insured was replaced by the procedural rules of court once a formal claim had been made. In the subsequent Court of Appeal decision, K/S Mer-Scandia XXXXII v Certain Lloyd’s Underwriters and others (‘The Mercandian Continent’),49 both the post-formation duty of utmost good faith and the ‘fraudulent claim’ rule See 772. The legal analysis of a proposal of a potential insured made online and when and how they are accepted by either human insurers or by artificial intelligence and the legal consequences are matters that merit a study on its own. 47 Some issues arose in Piermay Shipping Co SA v Chester (‘The Michael’) [1979] 2 Lloyd’s Rep 1. The duty was considered in detail in Black King Shipping Corpn v Massie (‘The Litsion Pride’) [1985] 1 Lloyd’s Rep 437, then in Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd [1988] 1 QB 818 (CA). The topic was extensively examined by Rix J in Royal Boskalis Westminster NV v Mountain [1997] LRLR 523 and dealt with by the House of Lords in The Star Sea, which disapproved the reasoning of Hirst J in The Litsion Pride on the creation and remedies available to an insurer for breach of a post-formation duty of utmost good faith by the insured: see per Lord Hobhouse at [71]. 48 [2003] 1 AC 469. 49 [2001] 2 Lloyd’s Rep 563 (CA) Although The Mercandian Continent is more of a case on ‘fraudulent claim’ there is a general discussion of the post-formation duty of utmost good faith on the part of the insured in the judgment of Longmore LJ. 45 46

Has ‘utmost good faith’ become like the Cheshire Cat?  53 were analysed by Longmore LJ. In Brotherton v Aseguradora Colseguros SA, Mance LJ made some general but important comments about the scope of the post-contract duty of utmost good faith.50 The duty was re-examined, in the context of an alleged ‘fraudulent claim’, by the Supreme Court in Versloot.51 The particular duty on an insured not to present a fraudulent claim was well established by the mid-nineteenth century.52 The issue of whether this ‘fraudulent claims rule’ was a particular manifestation of the duty of utmost good faith and a part of the ‘general law’, or (in the absence of an express term) the subject of a term of the insurance contract ‘implied or inferred by law or at any rate an incident of the contract’ remains unresolved.53 Part 4 of the IA created a statutory remedy for the insurer in the case of an insured making a fraudulent claim. This therefore presupposed a duty on the insured not to do so. But the IA neither defined the extent of the duty nor the definition of what constituted a ‘fraudulent claim’. Those issues were deliberately left to the courts to determine.54 Versloot was a case on facts that took place before the IA came into force, although the judgments in the Supreme Court were given after it had done so. The decision therefore does not, strictly speaking, deal with the meaning of ‘fraudulent claim’ for the purposes of Part 4 of the IA. The majority55 of the Supreme Court held that the use by an insured of a ‘collateral lie’, which was not material to the claim or an insurer’s defence to it, but was an attempt to give the claim verisimilitude, did not turn an otherwise genuine claim that was within the terms of the policy into a fraudulent one. Four general conclusions can be drawn from the existing cases and the statutory wording of Part 4 of the IA. First, the Law Commission was well aware of decisions such as The Star Sea and The Mercandian Continent and proposed nothing in its reports that led to the IA, that detracts from them. Thus, insofar as those cases deal with the existence, nature and scope of a duty of utmost good faith post-formation of contract (as opposed to the remedies for breach, now partly altered by section 14 of the IA), they must remain important in determining the law on those issues under the new regime. Second, the post-formation of contract duty of utmost good faith must be contractual in nature, as Lord Hobhouse stated in The Star Sea.56 Third, in discerning the circumstances in which the duty on the insured is triggered there is helpful guidance from Longmore LJ in The Mercandian Continent.57 A particularly important point, accepted also by Mance LJ in Brotherton v Aseguradora Colseguros SA,58 is that the scope of any post-contractual duty of utmost good faith is limited to ‘circumstances of repudiatory 50 [2003] 2 All ER (Comm) 289 at [34]. Buxton LJ added some comments rejecting an argument that the duty of utmost good faith on an insurer, post contract formation, could create a restriction on the right to rescind: [46]. 51 [2017] AC 1. 52 The leading case is Britton v Royal Insurance Co (1866) 4 F&F 905, (which was a property insurance case, not marine) in which Willes J gave a direction to the jury that ‘if the claim is fraudulent, it is defeated altogether’, despite the fact that there was an express clause in the policy in that case that if a fraudulent claim was made the insurer would not be liable to pay. 53 In Versloot, Lord Sumption, who gave the leading judgment of the majority, inclined to the latter view at [8]; Lord Hughes appears to agree, although his conclusion is tentative: see [64] to [68]. 54 Law Commission Report No 353 (Cm 8898 July 2014) at Ch 22.19 and Ch 23.6. 55 Lord Mance dissented. The decision was met with dismay in the insurance industry. 56 [50] and [51]. 57 [2001] 2 Lloyd’s Rep 563. 58 [2003] 2 All ER (Comm) 298 at [34]. Mance LJ refers to The Star Sea, The Mercandian Continent and Agapitos v Agnew [2002] EWCA Civ 886.

54  Research handbook on marine insurance law breach or fraudulent intent’. Fourth, because all the cases were dealing with the old section 17 wording, remarks in them about possible remedies for breach of the duty are now of limited value.

7.

TO WHAT DOES THE DUTY OF UTMOST GOOD FAITH APPLY POST-FORMATION OF THE CONTRACT: THE INSURED

There are four issues that require discussion. First, what are the circumstances in which the insured’s duty of utmost good faith to the insurer might be triggered? Second, what is the nature of the act or omission by the insured that will constitute a breach of the duty? Third, does there have to be a causative link between the breach and either the claim under the policy or the insurer’s defence to a claim? Last, what remedies will the insurer have for a breach by the insured of its duty? The first three points are considered here; the last will be dealt with in the final section of this chapter. The leading case in which the first three aspects were considered is The Mercandian Continent.59 Shipowners had obtained a default judgment against ship repairers for damages as a result of faulty repairs to a ship. The shipowners then made a claim directly60 against the insurers of a liability policy. The insurers resisted the claim and purported to avoid the policy on the ground that a fraudulent letter concocted by the insured ship repairers (although not one directed at the insurers, and made long before any claim could be made on the policy)61 was a breach of the duty of utmost good faith by the insured after the insurance contract had been concluded. This defence was rejected by Aikens J and the Court of Appeal, who heard the case soon after the House of Lords had given its reasons in The Star Sea. Longmore LJ, with whom Carnwath and Robert Walker LJJ agreed, gave the leading judgment. He traced the development of the law on the duty of utmost good faith owed by the parties after the formation of an insurance contract. Leaving fraudulent claims to one side,62 Longmore LJ identified four specific circumstances where the post-formation of contract duty of utmost good faith on the part of the insured was, or might be, triggered. They were: (1) when there is a variation of the risk; (2) when there is a renewal of the risk; (3) when an insured is entitled (on fulfilling certain conditions) to be ‘held covered’; and (4) where an insurer has a right to demand information from the insured during the policy term. Longmore LJ also 59 [2001] 2 Lloyd’s Rep 563. It was discussed and effectively approved by the SC in Versloot at [16], [17] and [44]. At the very least no aspect of its reasoning was criticised by the majority. 60 Under s.1 of the Third Parties (Rights Against Insurers) Act 1930, (‘the 1930 Act’), since replaced by the Third Parties (Rights Against Insurers) Act 2010, which simplifies the process. 61 A third party who wishes to sue an insurer under a liability policy under the 1930 Act has to establish that the insured is liable to the third party: Post Office v Norwich Union Fire Insurance Society Ltd [1967] 1 Lloyd’s Rep 216; Bradley v Eagle Star Insurance Co Ltd [1989] QC 957. If the third party then wishes to sue the insurers directly under the 1930 Act, it must then establish that the insured has become insolvent. An insurer can rely on any defence under the policy that would be available against the insured: Firma C-Trade SA v Newcastle Protection and Indemnity Association (‘The Fanti’) [1991] 2 AC 1. 62 Longmore LJ stated, at [22], that it was likely that the ‘fraudulent claims rule’ had developed independently of the duty of utmost good faith post contract. As noted in fn 3 above, the question of whether there are two rules or one remains unresolved.

Has ‘utmost good faith’ become like the Cheshire Cat?  55 noted that, in the case of liability policies, where an insurer takes over a claim on behalf of an insured, both parties will owe a duty ‘to act in good faith towards one another’.63 As Longmore LJ pointed out at [31] of his judgment, categories (1) to (3) are all really examples of the pre-contract duty because they arise before the variation, the renewal or the application of the ‘held covered’ provision in a policy. The first two of Longmore LJ’s categories are now covered by Part 2 of the IA.64 It is arguable that the third constitutes a form of ‘variation’ so is also covered by Part 2. If it is not, then the duty must be triggered once the ‘held covered’ clause is invoked. Longmore LJ considered that ‘the operation’ of section 17 post-formation was not limited to cases analogous to the pre-contract context and fraudulent claims.65 However, apart from the example of the mutual duty of utmost good faith that arises in the case of a liability policy where an insurer has taken over the conduct of a defence, Longmore LJ did not identify any other ‘triggers’ for the duty post-formation of the contract. The next question is what standard of conduct will result in a breach of the duty? This was considered by Rix J in Royal Boskalis Westminster NV v Mountain.66 The owners of dredgers that had been detained in Kuwait during the first Gulf War claimed on marine policies for the constructive total loss of the vessels and the value of claims that the dredger owners had ‘waived’ in a secret deal with Iraq under which they had also paid the Iraqi government US$20 million, contrary to UN sanctions. The owners then deliberately concealed this deal from the insurers when making their claims, although they subsequently informed the insurers of the facts. One of very many issues in the trial before Rix J was whether the fact of the concealment was a breach of a post-contractual duty of utmost good faith by the insureds, giving the insurers a defence to all the claims. Rix J found that there had been ‘deliberate and culpable’ misrepresentations and non-disclosures by the shipowners, although he did not find that they had been fraudulent; nor did he find they had presented a fraudulent claim. The deliberate and culpable concealment was a breach of the insured’s post-formation of contract duty of utmost good faith. However, Rix J also held that before any culpable and deliberate misrepresentation and non-disclosure could give the insurers any right to avoid the contract it had to be shown

See [24] sub-para (7). Longmore LJ there referred to the Delphic comment of Sir Thomas Bingham MR in Cox v Bankside Agency Ltd [1995] 2 Lloyd’s Rep 437 at 462 that if there were a breach of duty by the insurer in conducting the defence on behalf of the insured, then he (Sir Thomas Bingham MR) could not ‘for one instance accept […] [the] suggestion that a breach of this duty, by an insurer, once a policy is in force, gives the assured no right other than recission’. This seems a particularly odd statement given that Bingham LJ was a party to the CA’s judgment in Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd. [1990] 1 QB 818, in which, at 888, the court held that any breach of the duty of utmost good faith post-contract by the insurer could not give rise to a claim in damages by the insured or its assignee. 64 See s.2(2) on the application of Part 2 to variations. As for a renewal, as this constitutes a new contract of insurance, Part 2 must apply by virtue of the wording of s.3(1): ‘Before a contract of insurance is entered into…’. 65 See [34]. 66 [1997] LRLR 523. This issue was not discussed in the CA, because it was accepted by both sides that the court was bound by the test as set out by the CA in The Star Sea. The HL gave leave to appeal on all aspects of the CA’s decision in Boskalis but the parties settled the case before a hearing was fixed. The case would have been heard together with The Star Sea. The views of Hirst J in Black King Shipping Corpn v Massie (‘The Litsion Pride’) [1985] 1 Lloyd’s Rep 437 on this point are not considered because his reasoning was disapproved by the HL in The Star Sea: see [71] of Lord Hobhouse’s speech. 63

56  Research handbook on marine insurance law that the conduct of the insured had to be causally relevant to the insurers’ ultimate liability or to a defence to the claim.67 The third issue is what causative link is needed between the behaviour that constitutes a breach of the post-formation duty of utmost good faith and the claim being made. Rix J’s conclusion that the concealment, non-disclosure or other act or omission had to be materially relevant to the claim or a defence to the claim was approved and adopted by the Court of Appeal in The Mercandian Continent. In that case Longmore LJ also emphasised that it had to be shown that the material concealment, non-disclosure or other act or omission had actually induced the insurer to take the action it had.68 There is nothing in the IA that forces any change in the analysis of these issues. The courts would be likely to follow that analysis if any case arose concerning a breach of the duty of utmost good faith by an insured post-contract formation. What of fraudulent claims? As already noted, the common law rule that the presentation of a fraudulent claim by an insured entitles an insurer to reject it and, possibly, to terminate the insurance contract on the ground of repudiatory breach had been firmly established since the mid-nineteenth century.69 In The Mercandian Continent Longmore LJ stated that there was ‘no evidence that Sir Mackenzie Chalmers had this line of authority in fire insurance cases in mind when he drafted section 17 of his marine insurance code’.70 This suggests the obligation not to present a fraudulent claim and the consequences if there is a breach are not based on the duty of utmost good faith. The early cases are mixed in their reasoning as to whether the ‘fraudulent claim rule’ rests on the duty of utmost good faith.71 In the leading case on fraudulent insurance claims, Lek v Mathews, neither Branson J nor the Court of Appeal nor the House of Lords, in their elaborate judgments on the facts, mention the juridical basis of the rule that if a person makes a fraudulent claim, then the claim is forfeit. This is probably because the policy, which insured a collection of stamps against theft, contained a clause which stated that ‘if the assured shall make any claim knowing the same to be false or fraudulent as regards amount or otherwise, this policy shall become void and all claim thereunder shall be forfeited’. Branson J concluded that the claim was fraudulently made and so, by the terms of the policy, it was avoided. The majority of the Court of Appeal (Scrutton LJ dissenting) reversed that conclusion on the facts, but the House of Lords restored the judge’s findings and conclusion.72 The case therefore gives no help on the juridical basis of the ‘fraudulent claim’ rule.

See 591–601 of the report. See [26] to [29]. 69 In The Mercandian Continent, Longmore LJ traced the origin of the rule to fire insurance cases, such as Levy v Baillie (1831) 7 Bing 349; Goulstone v Royal Insurance (1858) 1 F&F 276 and Britton v Royal Insurance (1866) 4 F&F 905. 70 [22]. 71 In Levy v Baillie (1831) 7 Bing 349 there was an express clause in the policy that if there was any fraud in the claim or ‘false swearing or affirming in support, the claimant shall forfeit all benefit under the policy’. In Goulstone Pollock CB directed the jury that ‘if the claim was fraudulent the plaintiff cannot recover’. In Britton when Willes J directed the jury he said that an insured who made a fraudulent claim could not recover, but he also said: ‘the contract of insurance is one of perfect good faith on both sides, and it is most important that such good faith should be maintained.’ He pointed out that there was an express term in the fire insurance in that case that it would be ‘void in the event of a fraudulent claim’ and said that ‘such a condition is only in accordance with legal principle and sound policy’. 72 See respectively (1926) 24 L Ll Rep 191; (1926) 25 L Ll Rep 525 and (1927) 29 LLl Rep 141. 67 68

Has ‘utmost good faith’ become like the Cheshire Cat?  57 However, there is help in the judgments of Lords Sumption and Hughes in Versloot.73 Lord Sumption agreed with the view of Lord Hobhouse, expressed in The Star Sea,74 that once the contract is made, ‘the content of the duty of good faith and the consequences of its breach must be accommodated within the general principles of the law of contract’. Thus the ‘fraudulent claim rule’ had to be regarded as a term ‘implied or inferred by law, or at any rate and incident of the contract’. Lord Sumption noted that the difference mattered because if the rule was part of the duty of utmost good faith, then breach would mean, on the wording of section 17 as it then stood, that the contract was voidable; whereas, if the rule was a term of the contract and not a part of the duty of utmost good faith, then the rights would depend on contractual principles. Normally that would mean that a contract could be terminated only prospectively.75 Lord Hughes regarded the ‘fraudulent claim rule’ as a ‘rule of common law, grounded in sound policy, rather than depending on an implied term in the contract’.76 In a further passage, Lord Hughes reviews early cases and concludes: ‘There is no occasion in the present case to pursue the elusive matter of definitive analysis of the content of the post-contract duty of good faith, for it is enough that it plainly includes the fraudulent claim rule.’ However, Lord Hughes then goes on to say that the present case cannot be answered by ‘predicating it on the basis that the fraudulent claims rule is merely a manifestation of the duty of utmost good faith’.77 This is both inconclusive and also puzzling. If the fraudulent claims rule is a part of the post-contractual duty of utmost good faith, then the consequence of a breach was, under the old section 17, clear; the insurer had the right to avoid the policy. As Lord Sumption pointed out, that must apply to all breaches, whether in the form of submitting a fraudulent claim or supporting a valid claim with fraudulent, untruthful or recklessly inaccurate information (as happened in Versloot). But the majority of the Supreme Court came to the opposite conclusion. This debate may now be regarded as academic, in the light of the change to section 17 and the introduction of Part 4 of the IA, which sets out the consequences of the presentation of a fraudulent claim. The statutory remedies set out in section 12 of the IA are clear and robust. The insurer does not have to pay the claim; any sums previously paid can be recovered; and the insurer can, by notice, terminate the contract of insurance with effect from the time of the fraudulent act.78 It seems best to conclude that, from now on, the duty not to make a fraudulent claim is not to be regarded as a manifestation of the continuing duty of utmost good faith. The duty not to present a fraudulent claim will arise either as a result of an express term of the contract, or one that is implied. A breach gives rise to statutory remedies, as set out in section 12 of the IA. Those can be modified by contractual agreement between the parties, provided

[2017] AC 1. At [50] and [61]–[62]. 75 See [7] in Lord Sumption’s judgment. Lord Hobhouse had made the same point about the consequences of a contractual analysis in The Star Sea at [61]. 76 [55]. 77 [67] and [68]. 78 Section 12(1) IA. The prospective nature of the remedy is confirmed in s.12(2), which stipulates that if the insurer does treat the contract as terminated, then it may refuse all liability to the insured under the contract in respect of ‘a relevant event occurring after the time of the fraudulent act’: in other words it can reject all claims made after the ‘fraudulent act’; but not claims beforehand: s.12(3). 73 74

58  Research handbook on marine insurance law that the ‘transparency’ safeguards in the ‘contracting out’ provisions in Part 5 of the IA have been met by the insurer.79 Is there any further scope for the duty of utmost good faith in relation to claims on the part of the insured? What if the claim itself is a good one, as in Versloot, but the insured has attempted to reinforce it with fraudulent statements or documents or ones that are recklessly made, even though immaterial to the claim itself? What if the consequence is that the insurer has to do a great deal of research and investigation as a result, even though, ultimately, the claim has to be paid as it is sound. Does the insurer have any right to some compensation for all the extra expenses and trouble it has had to undertake? Of course, if the claim results in a trial, the insurer may recover costs in relation to that work, but how much may be doubtful. Can the insurer assert that the actions of the insured were a breach of the duty of utmost good faith, for which he can claim damages? There are difficulties with this. First, any such rights would appear to be contrary to the principles that were first considered by Rix J in Royal Boskalis Westminster NV v Mountain80 and confirmed by the majority of the Supreme Court that the only non-disclosures or misrepresentations concerning a claim which matter are those that are legally relevant to the claim or defence. Second (or perhaps it is putting the same point another way), any rights based on the principle that the concealment or misrepresentation of matters that were immaterial to the claim or defence would appear to undermine the clear distinction that the Supreme Court made in Versloot between a ‘fraudulent claim’ and a ‘collateral lie’. Third, if the correct analysis is that there is a breach of the duty of utmost good faith post-formation of the contract only when the breach is so great as to give rise to a right to terminate as being repudiatory, then an insurer could terminate the contract and claim damages. However, on Lord Sumption’s analysis, the insurer would still have to pay the valid claim.81

8.

TO WHAT DOES THE DUTY OF UTMOST GOOD FAITH APPLY POST-FORMATION OF THE CONTRACT: THE INSURER82

Before the change to section 17 brought about by the IA, there was general acceptance that the duty of utmost good faith on insurers exists post-contract. If, as is argued above, the effect of section 17 as amended is to retain a legal duty (and correlative right) of utmost good faith, then it must apply to insurers after the formation of the contract as well as before. Given the

See IA ss.16 and 17. This gives rise to an interesting conundrum: as has been seen, many policies contain an express clause that if a fraudulent claim is made or there is any ‘false swearing or affirming in support’ (Levy v Baillie), then the claim is forfeit. However, if the definition of a ‘fraudulent claim’ excludes ‘collateral lies’, as decided in Versloot, then a clause in a policy that states the claim would be forfeit even if genuine because supported by ‘false swearing in support’ would not be subject to the ‘transparency’ obligations in ss.16 and 17, because a ‘collateral lie’ is not within the definition of a ‘fraudulent claim’, so outside Pt 4. Was that intended? 80 [1997] LRLR 523 at 597. 81 See [23]–[36]: the key is that the right to indemnity arises as soon as the loss occurs. Can anything that is immaterial to the claim or a defence retrospectively remove or bar the insured’s pre-existing cause of action? The majority of the SC said ‘No’. 82 See, generally, ch.3 of The Insurance Act 2015 (Clarke and Soyer eds 2017) for a discussion of this topic by Prof. Soyer. 79

Has ‘utmost good faith’ become like the Cheshire Cat?  59 analysis that has been applied to the duty owed by insureds, it must follow that the duty on the insurers arises out of an implied term of the contract, which is itself founded on the existence of the duty of utmost good faith. As with insureds, however, similar questions arise: when does the duty manifest itself, what is the standard of behaviour which constitutes a breach and what are the consequences if there has been a breach by the insurer? There are very few cases prior to the IA and none since to provide answers. On the first question, on the authority of Keyser Ullmann, there must be a duty on an insurer to disclose all facts known to him which are material to the recoverability of a claim to be made by the insured under the policy.83 In Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co,84 Lord Lloyd of Berwick took up the point made by Lord Mansfield in Carter v Boehm85 that the insurer, in asserting its rights under the policy, has to act with the utmost good faith. The same point has been referred to in subsequent cases, most notably in Drake Insurance plc v Provident Insurance plc,86 where one of many issues considered was whether an insurer could avoid a policy of motor insurance or whether, in the circumstances, to do so would be a breach of a duty of utmost good faith. Rix LJ said that the obligation of utmost good faith could operate to limit an insurer’s right to avoid, although he did not give any examples where this might be done. He held in that case that the insurer was entitled to avoid.87 The insurer must also owe a duty of utmost good faith in acting in the conduct of litigation where the insurer has taken over the defence from the insured under the terms of a liability policy.88 It must be arguable that the insurer is under a general duty of utmost good faith in considering claims by the insured, apart from the obligation now imposed by statute, to pay sums due in respect of the claim within a reasonable time.89 Section 13A(1) of the IA imposes the duty on an insurer to pay sums due in respect of a claim made under an insurance contract ‘within a reasonable time’. The statutory duty is expressed as an implied term of the contract. There is no mention of ‘utmost good faith’ in the new section 13A. The whole statutory regime about timely payment of claims due therefore appears to have been placed outside any post-formation of contract duties of utmost good faith on the insurer.

See fn 42 above. [1995] 1 AC 501 at 555. 85 (1766) 3 Burr 1905 at 1918–19. 86 [2004] QB 601. 87 See in particular [87]–[88]; Clarke LJ was ‘inclined to agree with the views expressed’: [143], while Pill LJ considered that a failure by the insurer to make a relevant inquiry of the insured before avoiding the policy was a breach of the duty of utmost good faith: [177]. The remarks of both Rix and Clarke LJJ were obiter. Pill LJ was in the minority on the main issue on which the appeal turned, but was clear in his view on this point. The CA in Brotherton v Aseguradora Colseguros SA [2003] 2 All ER (Comm) 298, seems to have reached the opposite conclusion, again obiter: see the judgment of Mance LJ at [34], with whom Buxton and Ward LJJ agreed, Buxton LJ commenting specifically on this point at [46]. 88 The point was not squarely dealt with in Groom v Crocker [1939] KB 194, where the emphasis was on the duty of the solicitor to its client, the insured, although the solicitor was appointed because the insurer was entitled under the policy to take over the defence of claims. Sir Wilfred Greene MR considered that the insurer itself had a duty to consider the interests of the insured: see page 203. The matter was adverted to by Sir Thomas Bingham MR in Cox v Bankside Agency Ltd [1995] 2 Lloyd’s Rep 437 at 462. 89 Section 13A(1) IA. 83 84

60  Research handbook on marine insurance law The question of what standard of conduct is sufficient to constitute a breach of an insurer’s duty of utmost good faith post-formation of the contract remains undecided. In Drake Insurance plc v Provident Insurance plc, Rix LJ commented that The Mercandian Continent had limited ‘the operation of the right to void for want of good faith on the part of the insured in the post-contractual context to situations where the law of contract would justify termination on repudiatory grounds’.90 This might suggest that there is only a breach of the post-formation of contract duty if it is sufficiently serious to demonstrate that the party in breach intends no longer to be bound by the terms of the contract or has renounced the contract. That is a high standard. The consequences of breach of the duty by the insurer are dealt with in the next section.

9.

REMEDIES AVAILABLE FOR A BREACH OF THE DUTY OF UTMOST GOOD FAITH BY EITHER INSURED OR INSURER

Before the amendment to section 17 of the MIA the only remedy available for a breach of the duty of utmost good faith was avoidance of the policy. There were suggestions that a breach of the obligation of good faith on the part of an insurer could give rise to a ‘defensive’ remedy for the insured which would prevent an insurer from being able to avoid the contract for breach of the good faith obligation by the insured.91 The equitable remedies of injunction, specific performance or declaration could be available in appropriate circumstances. The remedy for the presentation of a fraudulent claim (in the absence of an express term in the contract to other effect) was that the claim was forfeit entirely and the insurer had the right to terminate the contract.92 The discussion of remedies for breach of the duty of utmost good faith under the amended section 17 must assume that the concept has some legal effect, as argued above. Thereafter, the only certainty would appear to be that a breach of the duty by either insured or insurer cannot give rise to a right to avoid the contract, that is, to treat it as if it had never existed. If the analysis undertaken above is correct, then the circumstances in which the duty of utmost good faith can arise under the new section 17 are limited. First, it could arise in relation to acts or omissions of an insurer before the contract is concluded. There is no scope for additional duties on the insured outside those imposed by Part 2 of the IA. Second, there could be situations post-contract where the duty could arise in relation to acts or omissions of either the insured or the insurer, but not in relation to either fraudulent claims or the timely payment of sums validly due under a claim by an insured. It has been argued above that if there is a legal duty of utmost good faith that derives from the new section 17, it should be regarded as an implied term of the contract of marine insurance, whether the duty is one that arises in relation to pre-contract matters or post-formation of the contract. If it is an implied term, then, at least in relation to the pre-contract duty (on an insurer), there could be arguments about its nature. Is it the equivalent of a condition, the breach of which gives a right to terminate, however slight the breach? Is it a ‘warranty’ 90 [2004] QB 601 at [87]. In Brotherton v Aseguradora Colseguros SA [2003] 2 All ER (Comm) 298 Mance LJ made the same point: [34]. 91 See above and fns 81 to 83. 92 The Star Sea at [50] and [61] of Lord Hobhouse’s speech.

Has ‘utmost good faith’ become like the Cheshire Cat?  61 which could only ever give rise to a right to damages, if a breach. Or is it an ‘innominate’ or ‘intermediate’ term, where breach gives a right to damages but will only give rise to a right to terminate if the breach is so serious as to deprive the innocent party of substantially the whole benefit of the contract. The increasing tendency of the English courts is to treat terms of contracts, especially implied terms, as ‘innominate’ or ‘intermediate’. That appears to have been the approach towards the scope of any post-formation of contract duty of utmost good faith.93 Thus Mance LJ described the scope as limited to ‘circumstances of repudiatory breach or fraudulent intent’.94 Thus circumstances in which a right to claim that there had been a breach of the duty of utmost good faith, whether before or after the formation of the contract, would be limited to those where the innocent party was also entitled to terminate the contract on the grounds that the breach had been repudiatory. In practice, it may not be commercially prudent to terminate the contract. If an insurer had concealed or misrepresented the scope of the cover before entering the contract, then the insured may wish to terminate and claim damages, but, equally, depending on the circumstances, may wish to continue to have cover but claim damages for any loss of cover suffered. The same would be true if the insurer had done the same thing after the contract was concluded in the way in which the contract worked. (Delays in handling a claim would come under section 13A of the IA). As already noted above, the circumstances in which an insurer could claim that an insured had been in breach of the duty of utmost good faith post-formation of the contract appear to be very limited or even non-existent, given the introduction of Part 2 and sections 12 and 13A of the IA.

10. CONCLUSION This chapter is, like marine insurance itself, a speculation. It is an attempt to predict how the courts will deal with the amended version of section 17 of the MIA alongside the changes in the law of insurance brought about in particular by Parts 2, 4 and 4A of the IA and in the light of existing case law. The predictions may turn out to be incorrect, but it is right that there should be consideration of the possibilities. It is hard to accept a proposition that, quite apart from provisions of Parts 2 and 4 of the IA, there is no legally enforceable right or duty of utmost good faith in marine insurance contracts after more than 200 years of the concept being central to marine insurance law. Yet, given the other reforms, in particular in Parts 2 and 4 of the IA, any scope for the free-standing legal duty, assuming it exists at all, is much reduced. Its nature, that is whether it is characterised as an implied term or some principle of ‘general law’, is also for future debate. So are the circumstances in which the duty may arise, both before and after the contract of marine insurance has been made. The conclusions that courts will reach on what remedies there are for breach of the duty, and, in particular, whether it is now possible to consider damages as a possible remedy, are unpredictable.

93 See per Lord Hobhouse in The Star Sea at [50] and [61], Longmore LJ in The Mercandian Continent at [26] and Mance LJ in Brotherton v Aseguradora Colseuros SA [2003] 2 All ER (Comm) 298 at [34]. 94 Brotherton v Aseguradora Colseuros SA [2003] 2 All ER (Comm) 298 at [34].

62  Research handbook on marine insurance law So the only conclusion that can be reached with certainty is that the principle of ‘utmost good faith’ in marine insurance law is like the Cheshire Cat: now you see it and now you don’t. Whether the Cheshire Cat would be likely to growl or wag its tail is anybody’s guess.

4. Co-insurance and rights of subrogation post-Gard Marine And Energy v China National Chartering Company Ltd Kyriaki Noussia and Yavuz Can Aslan

1.

CO-INSURANCE IN HULL POLICIES

1.1 Introduction Co-insurance policies protect an interest that is common to more than one party. However, where this common interest is exact, such as the joint ownership of a property, it is said that the co-insurance policy is joint. On the other hand, if the parties’ common interests are of different legal natures, such as the interests of an owner and a tenant, then the co-insurance policy is qualified as a composite.1 From one aspect, this distinction significantly affects the co-insured’s right to indemnity. In joint insurance, if each and all co-insureds do not have the right to sue for the loss, the obligation of the insurer to indemnify does not arise. In the composite co-insurance policies, it is sufficient that at least one co-insured has a right to sue for claiming the insured amount from the insurer.2 From another aspect, the insurer’s subrogation rights may also depend on the qualification of the co-insurance policy as joint or composite, as this qualification is a repercussion of the legal relationship between the co-insureds on the scale of the main contract, according to which the rights of actions between the co-insureds are defined. This chapter examines co-insurance and the rights of subrogation after the case of Gard Marine and Energy v China National Chartering Company Ltd.3 It discusses the impact of clauses 12 and 13 of BARECON 89 standard form4 on co-insurance and clause 29, which was added to the standard BIMCO form by the parties; the legal nature of subrogation; and the impact of the safe port warranty on the right of subrogation. In doing so it examines the decision of the Supreme Court in Gard Marine5 that the port was safe and its further elaboration that had the port been unsafe, Gard would not have been entitled to recover because of the clause 12 of the BARECON 89.6 Further on, it discusses Haberdashers’ Aske’s Federation Özlem Gürses, Marine Insurance Law (3rd edn, Routledge 2023) 416; Rob Merkin and Jenny Steele, Insurance and the Law of Obligations (1st edn, Oxford University Press 2013) 172–3; Ahmed Tolulope Olubajo, ‘The Law of Co-Insurance Policies’ (University of Southampton 2003) 1; Diane Donnelly and Robert Hogarth, ‘Insurance Law for the Construction Industry’ in Robert Hogarth (ed), Insurance Law for the Construction Industry (1st edn, Oxford University Press 2008) 51–2. 2 Olubajo (n 1) 2; Donnelly and Hogarth (n 1) 52. 3 [2017] UKSC 35. 4 BIMCO, BARECON89, www​.bimco​.org/​Contracts​-and​-clauses/​BIMCO​-Contracts/​BARECON​ -89 5 [2017] UKSC 35. 6 BARECON89 (n 4). 1

63

64  Research handbook on marine insurance law Trust Ltd & others v (1) Lakehouse Contracts Ltd & (2) Cambridge Polymer Roofing Ltd7 and the missed opportunity in further discussing subrogation rights issues, whereby it was taken as granted that the sub-contractor caused the damage and, as such, was liable for it. It also examines the aftermath effect of Gard Marine,8 which prompted BIMCO to consider the case and issue BARECON 20179 clause 17(a), which establishes provisions that apply regardless of who might issue the insurance and as per which the existence of the parties as co-insureds in the hull policy does not exclude them from liability, even if it does not solve the problem of subrogation holistically, with regard to the contractual rights that stem from the demise charter itself. Last but not least, it purports that further case law is needed to elaborate on the missed opportunity to settle and clarify the law of subrogation in this regard. Before classifying an insurance policy as joint or composite in nature, one must first be able to qualify it as a co-insurance policy. There are several methods of construction in this regard. For instance, the type of legal relationship between the co-insureds is an important tool for interpreting the insurance policy.10 In Stone Vickers Ltd v Appledore Ferguson Shipbuilders Ltd,11 it was argued that in order to be able to ascertain the intention of the parties, it is essential to not only take into account the policy documents as such, but to also look at and interpret the contract between the assured and the alleged co-assured. The scope of the contractual provisions agreed upon between the co-insureds is not the only means for such an evaluation.12 As elaborated in National Oilwell (UK Limited) v Davy Offshore Limited,13 such evidence is enough to ascertain whether in fact, in any case, either the main party assured or any other contracting party had such intentions. This can be derived by the terms of the policy itself, or by the terms of any contract between the main assured, or any other contracting party and the party ascertained as co-assured, or by any other evidence which can lead to the conclusion that this was the intention of the main assured party. The breadth of protection provided for the co-insureds in the insurance policy may be another criterion to assess whether the same interest has been insured.14 If this interest is not the same, it is impossible to argue that there is a co-insurance policy. Following National Oilwell,15 Rugby Football Union v Clark Smith Partnership Limited & FM Conway Limited,16 though believed to be controversial, is a straightforward application of National Oilwell,17 now also reaffirmed at appellate level.18

[2018] EWHC 588 (TCC). [2017] UKSC 35. 9 BIMCO, BARECON2017, www​.bimco​.org/​Contracts​-and​-clauses/​BIMCO​-Contracts/​ BARECON​-2017 10 Stone Vickers Ltd v Appledore Ferguson Shipbuilders Ltd [1992] 2 Lloyd’s Rep 578. 11 Ibid. 12 National Oilwell (UK Limited) v Davy Offshore Limited [1993] 2 Lloyd’s Rep 582. 13 Ibid. 14 BP Exploration Operating Co Ltd v Kvaerner Oilfield Products Ltd [2005] 1 Lloyd’s Rep 307 [99]: ‘The contract between the parties contains an invitation to insure the sub-contractor. However the contractor obtains a cover that has a wider scope than what has been agreed upon in this contract. The sub-contractor should have the benefit of cover in respect of the whole of the project and not the much more limited scope which the sub-contract’s construction would provide for’ (Haberdashers’ para. 47). 15 [1993] 2 Lloyd’s Rep 582. 16 [2022] EWHC 956 (TCC). 17 [1993] 2 Lloyd’s Rep 582. 18 [2023] EWCA Civ 418. 7 8

Co-insurance and rights of subrogation post-Gard Marine And Energy  65 Rugby Union19 dealt with the issue of co-insurance under project CAR policies. In particular, it attempted to shed light on the difficulties faced by contractors of all levels when trying to demonstrate the extent of cover in the face of a subrogated claim from project insurers. The case serves as a reminder to parties to construction projects that they need to ensure that the contractual arrangements for any project accurately reflect the intention and authority of the party obtaining insurance cover for others. In Rugby Union20 the Rugby Football Union (‘RFU’) was undertaking significant works at Twickenham in 2012 in view of the 2015 Rugby World Cup and contracted Clark Smith to design buried ductwork including power cables, and Conway to install it under a JCT Standard Building Contract. RFU asserted that the ductwork was defective, caused damage to the cables and resulted in replacement costs of £3,334,405.26, which were indemnified by the project insurers, Royal & Sun Alliance Plc (‘RSA’). An exclusion in the policy meant that the cost of addressing the defective ductwork was excluded, but the remedial cost of the consequential damage to the cables was covered. RSA sought to recover those sums from Conway as subrogated on the basis that damage had been caused by its defective workmanship, and Conway sought a declaration that it was a co-insured under the project policy and had the benefit of cover to the same extent as RFU as principal insured, hence there was no ground for the subrogated claim to be put forward. In effect, Eyre J had to rule on the issue of whether the sums paid by RSA to RFU were irrecoverable because RSA could not exercise subrogation rights. The judgment contains a summary of the law regarding co-insurance to date, including the cases when subrogated claims can be barred by reason of circuity of action, as per Cooperative Retail Services Ltd,21 and the cases when one insured may obtain cover for another, as per Gard Marine,22 but it is notable that the issue in Rugby Union23 was distinct in that the case dealt with the extent of that cover for a co-insured and not with the existence of cover in the first place. That is, applied to the facts of the case, the issue was whether Conway had the benefit of the full cover under the project policy, or if the cover was restricted to damage caused by the third party, that is, the specified perils as provided for by the unamended part of the JCT. Guided by National Oilwell,24 Eyre Justice stated that the contract is key when determining the extent of the provided cover, and went on to claim that ‘when a person becomes a party as a consequence of the actions of another person then the terms of the contract between the insured party and that other govern the extent of the insurance’.25 He went on to rule that the contract documents as a whole did not demonstrate an intention for the project policy to act as the sole remedy for any loss suffered by RFU as a consequence of a breach by Conway; that even if Conway was an insured, it was not a co-insured in relation to the damage for which RSA had indemnified RFU; and that following National Oilwell26 the waiver of subrogation clause in the policy only related to the matters for which Conway had cover under the policy, and so did not prevent a claim by RSA. Rugby Union27 does not alter the previous state of the law and serves to remind the need for [2022] EWHC 956 (TCC). Ibid. 21 [2002] UKHL 17. 22 [2017] UKSC 35. 23  [2022] EWHC 956 (TCC). 24 [1993] 2 Lloyd’s Rep 582. 25  [2022] EWHC 956 (TCC) at 88. 26 [1993] 2 Lloyd’s Rep 582. 27 [2022] EWHC 956 (TCC). 19 20

66  Research handbook on marine insurance law contractual documents to reflect the exact expectations of the parties. Eyre J stated that if the parties wanted a different cover, this has to be depicted in the JCT, so as to reflect the parties’ intentions.28 The appellate decision of 19 April 2023 in Rugby Union29 upheld the decision stating that the judge’s approach – in determining the scope of cover under a joint names policy by reference principally to the terms of the underlying building contract between the parties – was correct, and noted that regarding authority and intention in co-insurance, the investigations will rely on the underlying contractual arrangements agreed between the parties.30 Lord Justice Coulson stated that in relation to the applicable principles in such a notoriously complex area of law as this, the following broad propositions can be derived from the authorities: (a) the mere fact that A and B are insured under the same policy does not by itself mean that A and B are covered for the same loss or cannot make claims against one another;31 (b) in circumstances where it is alleged that A has procured insurance for B, it will usually be necessary to consider issues such as authority, intention and the related issue of scope of cover;32 (c) an underlying contract between A and B is not a necessary pre-requisite for a proper investigation into authority, intention and scope, but a contract may well be implied in any event;33 (d) where there is an underlying contract then, in most cases, it will be much the best place to find evidence of authority, intention and scope.34 Coulson stated, as a first impression, that the judge applied the correct principles of law. He considered the underlying contract between the parties, but he also went on to consider wider matters. His conclusion, both on the law and on the facts, was that the co-insurance defence failed.35 Coulson argued that in all the authorities, not just Gard Marine,36 and regardless of the precise facts of each case, as a matter of principle, when the court considers authority and intention in the co-insurance context it is inevitable that its investigations will start and finish with the underlying contractual arrangements agreed between the parties. Coulson rejected Ground 1 of the appeal and considered that the judge made no error of principle.37 In relation to Ground 2 on the relevance of the building, Coulson considered that, whatever the position in respect of the ‘Letter of Intent’ may be, the judge was entitled to have regard to the subsequent building contract in any event, and that the issue was not a question of construing the policy but a question of authority and intention. Hence, he rejected Ground 2 of the appeal.38 In relation to Ground 3, on whether Conway only has to show authority and not intention, Coulson denied the approach which, in certain circumstances, dispenses with intention altogether and reiterated that the approach of Colman J in National Oilwell39 did not convey that intention was not required in the case of 28 Rob Goodship (2022) ‘Co-insurance, it’s a bit of a scrum’, Fenchurch Law, www​.fenchurchlaw​.co​ .uk/​co​-insurance​-its​-a​-bit​-of​-a​-scrum/​ 29 [2023] EWCA Civ 418. 30 [2023] EWCA Civ 418, at 67; Gateway Chambers, Court of Appeal Upholds Decision in RFU, 23.4.2023, https://​gatehouselaw​.co​.uk/​court​-of​-appeal​-upholds​-decision​-in​-rfu/​ 31 [2023] EWCA Civ 418 at 53.1. 32 Ibid at 53.2. 33 Ibid at 53.3. 34 Ibid at 53.4.5. 35 Ibid at 61. 36 [2017] UKSC 35. 37 [2023] EWCA Civ 418 at 72. 38 Ibid at 82. 39 [1993] 2 Lloyd’s Rep 582.

Co-insurance and rights of subrogation post-Gard Marine And Energy  67 an identified or identifiable insured.40 In relation to Ground 4 and Conway’s alternative case as to an undisclosed principal, he stated that this was another attempt to elevate the Higgs Morris understanding into a complete answer to the first preliminary issue, and to ignore the other findings which the judge made and which comprehensively demonstrated that such an understanding had no legal effect or relevance. Ground 5 was found to have raised a separate point, which was that Conway – being a party insured under the policy – was entitled to the benefit of the policy’s ‘waiver of subrogation’ clause in respect of all losses, and not merely those within the scope of Conway’s cover. The appellate court rejected that argument as being contrary to commercial common sense, as well as contrary to Colman’s decision on the same point in National Oilwell.41 It also accepted that the express terms of the policy, which defined parties as being insured ‘for their respective rights and interests’, supported the conclusion that Conway’s cover was limited to the scope of cover envisaged by the underlying JCT contract.42 Accordingly, the appeal was dismissed.43 Coulson concluded, in summary, that the judge was right for the reasons that he gave, and rejected the appeal.44 In order to understand the insurer’s subrogation rights, the parameters of the background mentioned above must always be considered, and in the case of hull insurance there is no exception to the general rules stemming therefrom. 1.2

The Position of a Demise Charter as Co-Insured in Hull Insurance Policies

1.2.1 The legal relationship between the owner and the demise charterer In a demise charter contract, the ship owner leases his vessel to the charterer, who becomes entitled and responsible for managing the vessel.45 The command of the crew and officers, as well as the possession of the vessel, passes to the charterer.46 The latter aspect of the demise charter is crucial for the insurable interest as the persons entitled to the possession of a property, such as a demise charterer of a ship, are entitled under English law to proprietary rights like owners.47 This is interesting, at first glance, because the interests of an owner and a tenant are regarded as being of different legal nature when interpreting the co-insurance policies, which in turn gives rise to a composite co-insurance policy.48 This might ultimately lead to the question of why the legal nature of the interests of the demise charterer and the owner would not suit this pattern. The ship owner or the demise charter may contract the hull insurance policy. Both such probabilities necessitate separate analysis.

[2023] EWCA Civ 418 at 94. [1993] 2 Lloyd’s Rep 582. 42 [2023] EWCA Civ 418 at 109–11. 43 Gateway Chambers (n 30). 44 [2023] EWCA Civ 418, at 112. 45 [2015] EWCA Civ 1, para 70. 46 Ibid. 47 Aleka Mandaraka-Sheppard, Modern Maritime Law Volume 2: Managing Risks and Liabilities (Inorma Law 2013) 474. It is arguable that, in this regard, there is a strikingly sharp schism between the common law and the continental legal regimes which differentiate the possessory rights from the proprietary rights: Gard Marine & Energy Ltd [2015] EWCA Civ 1, para 79. 48 This approach, which we find absolutely coherent, is also embraced by some legal scholars: see Gürses (n 1), p.420. 40 41

68  Research handbook on marine insurance law 1.2.2 The shipowner as the party to the insurance contract The very first, albeit uncommon, probability is that the ship owner concludes the insurance contract but also covers the risk incurred by the demise charterer against the loss of the ship. The primary rationale behind such a contract would be based on manifold reasons. For example, the ship owner may have better bargaining power against the insurance company and may like to reflect the marginal decrease in the hull insurance policy premium on the monthly fees of the demise charter contract. To achieve this, the contract in favour of third parties, such as a demise charterer, would serve as a legal instrument. One should keep in mind that the ship owner can always have a clause inserted in the policy to cover risks of future demise charterers, as it is not necessary to determine the benefiter of a contract in favour of a third party at the outset. In other words, the benefiter may be described in general and appointed later.49 The reason for a benefiter to be defined at a later stage, as demonstrated in Haberdashers’,50 was because normally a sub-contractor would be an existing and named member of a definable class, whereas if a sub-contractor were to be subsequently appointed and ascertained and the policy of insurance had already been in existence, the latter ascertained sub-contractor would also be covered by the project insurance.51 However, it is common practice in insurance policies to exclude the application of the Contracts (Rights of Third Parties) Act 1999;52 therefore, it is unlikely to encounter any legal conflict where the demise charterer may be regarded as the co-insured being the benefiter of a hull insurance policy. In consequence, other construction methods must be employed to explain the position of the demise charterer as co-insured, and legal precedent in other insurance branches can serve as an asset to illuminate the related general principles regarding this matter. This was also affirmed in Rugby Union.53 1.2.2.1 The standing offer theory The standing offer theory is one method of including persons not a party to the insurance policy into the coverage as co-insureds. According to this approach, the insurers make a standing offer to a defined grouping in the insurance contract, such as sub-contractors or demise charterers. By signing and executing the sub-contract, the sub-contractors acquire the position of co-insured.54 However, if the sub-contract includes a provision that obliges the sub-contractor to make its own and separate insurance policy, this must be understood, as the

[2018] EWHC 588 (TCC). Ibid. 51 Ibid para. 35. 52 John Lowry, Philip Rawlings and Robert Merkin, Insurance Law Doctrines and Principles (4th edn, Hart Publishing 2022) 207. 53 [2022] EWHC 956 (TCC). 54 ‘The sub-contractor would become insured under the Project Insurance by means of a term implied into the sub-contract […] This implication of a term would occur by means of a standing offer to a sub-contractor to be included in the Project Insurance, the offer being accepted by the sub-contractor by execution of the sub-contract’ (Haberdashers’ para. 35; Stone Vickers (n 10); National Oilwell (n 12)). More consistent with the time at which the insurable interest must exist than the agency, John Birds, Ben Lynch and Simon Milnes, McGillivray on Insurance Law (15th ed., Sweet and Maxwell, 2022), at 1-202: ‘The offer is said to be one made by the insurers. The offer is made by the insurer to insure persons who are subsequently ascertained as members of the defined grouping. The offer would be accepted by a sub-contractor joining upon execution of the sub-contract.’ (Haberdashers’ (n 7) para. 59). 49 50

Co-insurance and rights of subrogation post-Gard Marine And Energy  69 standing offer of the insurance policy is not addressed to it,55 and therefore it cannot enjoy the status of the co-insured. The calculation method of the insurance premium may also be another criterion to assess whether sub-contractors of demise charterers can be regarded as co-insured, under the insurance contract.56 As the implied terms of the policy require some consideration, a standing offer directed to the sub-contractor or demise charterer to adhere to the insurance policy must increase the insurance premium. It follows that if the insurance premium is higher than that in standard similar insurance policies, this may incur a piece of evidence in favour of the existence of a co-insurance contract. As elaborated in Rugby Union,57 co-insurance has to be evidenced to reflect the benefits to the parties and their intentions. 1.2.2.2 The agency theory Another approach to qualifying the sub-contractors as co-insured under the main insurance policy is to confer the status of agent on the party who concluded the insurance contract, which is the principal or the ship owner in the case of the demise charter contract, and bind the sub-contractor or charterer as co-insured under the insurance contract.58 Rugby Union59 refers to this too. 1.2.2.3 Waiver of subrogation clauses The insurance policy may also contain an explicit waiver of the subrogation clause for the benefit of any insured.60 In this case, as long as the policy does not contain a counter-effective provision, such as an explicit provision to oblige the demise charterer to obtain its insurance, it may be strongly argued that the insurance contract also covers the demise charterers or sub-contractors, and therefore is a co-insurance contract. For instance, clause 28 of International Hull Clauses 200361 waives the insurer’s subrogation rights against those sub-charterers who are affiliated companies of the assured, unless they did not obtain a separate liability cover. However, the letter of this clause does not compel the sub-charterers

If the sub-contractor is required to obtain, provide or have its own insurance, then this is tantamount to an express term that prevents it from joining the defined grouping (Haberdashers’ (n 7) para. 60). 56 The nature of the implied term requires some consideration (Haberdashers’ (n 7) para. 35). 57 [2022] EWHC 956 (TCC). 58 National Oilwell (n 12); Haberdashers’ (n 7) para. 39; Stone Vickers (n 13); there are doubts that it works in every situation (Haberdashers para. 39). The person for whom the agent (the main insured) professed to act must be a person capable of being ascertained at the time the insurance was effected, and the additional party should be able validly to ratify if he had no insurable interest at the time the insurance was effected (Peter Watts and F. Reynolds, Bowstead and Reynolds on Agency (22nd edn, Sweet and Maxwell, 2020) at 2-067; Haberdashers’ (n 7) para. 43). The agency clause under the sub-contract can either be objectively or subjectively construed (Haberdashers’ (n 7) para. 54). There is as yet no consensus on which avenue should be adopted. Less consistent with the time at which the insurable interest must exist than the agency, Birds, Lynch and Milnes (n 54) at 1-197. 59 [2022] EWHC 956 (TCC). 60 ‘An express waiver of subrogation clause by the insurer for the benefit of any insured under the policy’ (National Oilwell (n 12)). See also Gürses (n 1) 406. 61 In the event of the vessel being chartered by an associated, subsidiary or affiliated company of the Assured, and in the event of loss of or damage to the vessel by perils insured under this insurance, the Underwriters waive their rights of subrogation against such charterers, except to the extent that any such charterer has the benefit of liability cover for such loss or damage. 55

70  Research handbook on marine insurance law to obtain their separate liability cover. That said, in Tyco Fire & Integrated Solutions Ltd v Rolls-Royce Motor Cars Ltd,62 the Court of Appeal has cast doubt, on the assumption that parties, who are jointly insured, can never bring a claim against one another in respect of a risk covered by their joint insurance. In Tyco Fire63 the employer claimed damages and the contractor accepted negligence, but resisted the employer’s claim, arguing its liability was excluded because the contract required the employer to maintain joint names insurance against damage to the existing structure. However, this argument did not succeed in the Court of Appeal, as the contract was thought to not require the contractor to be named as a joint insured, and Lord Justice Rix emphasised the need for clear drafting of joint insurance provisions, explicitly stating the exclusion or not of claims between jointly insured parties to be brought. 1.2.3 The demise charterer as the party to the insurance contract Although it will depend on the definition of ‘assured’ and how such definition is interpreted, nevertheless, where the demise charterer concludes the insurance contract, the overall odds and percentage of it being qualified as co-insured, alongside the ship owner, is substantially higher than when the ship owner concludes the insurance contract. However, this does not preclude the risk of the charterer being held liable for the loss and the insurer’s suing it on the grounds of the right of subrogation.64 The demise charterer’s conclusion of the insurance contract is insufficient to qualify it as a joint co-insurance policy, although the parties may have called it such. In other words, it is still possible to qualify the insurance policy as a composite co-insurance policy, at least with regards to some risks covered in favour of the owners, where the remaining facts would give ground to such a construction. The crux of the legal debate in Gard Marine65 was such an interpretation problem weaved around clauses 12 and 13 of the BARECON 89 and clause 29, which was added to the standard BIMCO form by the parties. 1.2.4

The impact of clauses 12 and 13 of the BARECON 89 standard form on co-insurance Clause 12 and clause 13 of BARECON 89 are two alternative provisions and the parties of the bareboat charter have to opt for only one. They are similar in many aspects with slight differences, which, however, can significantly impact the construction of the insurable interests of the parties under insurance contracts, which are to be subsequently concluded. In Gard Marine,66 this was one of the issues put under the limelight by the Supreme Court. The tender spot of clause 12 is subparagraph (a), which reads as follows: During the Charter period the vessel shall be kept insured by the Charterers at their expense against marine, war and Protection and Indemnity risks in such form as the Owners shall in writing approve, which approval shall not be unreasonably withheld. Such marine, war and P. and I. Insurances shall

[2008] EWCA Civ 286. Ibid. 64 Kodros Shipping Corp of Monrovia v Empresa Cubana de Fletes (The Evia) (No 2) [1983] 1 AC 736; D/S A/S Idaho v Clossus Maritime DA (The Concordia Fjord) [1984] 1 Lloyd’s Reports 385. The mere fact that the charterer pays for the hull insurance is not enough to exempt him from liability for breach of his obligations under the charterparty (para. 193). 65 [2017] UKSC 35. 66 Ibid. 62 63

Co-insurance and rights of subrogation post-Gard Marine And Energy  71 be arranged by the Charterers to protect the interests of both the Owners and the Charterers and mortgages (if any), and the Charterers shall be at liberty to protect under such insurances the interests of any managers they may appoint. All insurance shall be in the joint names of the Owners and the Charterers as their interests may appear.

The last sentence of this provision is of particular importance, as it shows that the parties intend to cover their common interests on the vessel of the exact legal nature. This is not surprising given that, at least under English law, the owner and the charterer are deemed to have the proprietary right of the same legal nature on the vessel.67 Clause 13 subparagraph (a) of BARECON 8968 is the same as clause 12 in that it sets out that ‘All insurance policies shall be in the joint names of the Owners and the Charterers as their interests may appear.’ The main difference of clause 13 is that insuring the vessel against marine and war risks falls upon the owners as an obligation. This provision is also silent on the protection and indemnity insurance which should prudently be concluded by the charterers. There is one phrase in this part of the provision of clause 13 which opens another debate as to both its own and that of clause 12 legal consequences: The Owners and/or insurers shall not have any right of recovery or subrogation against the Charterers on account of loss of or any damage to the Vessel or her machinery or appurtenances covered by such insurance, or on account of payments made to discharge claims against or liabilities of the Vessel or the Owners covered by such insurance.69

Adding this phrase to clause 13 is understandable in that when the charterer will not conclude the insurance contract, it may be an issue of legal discussion whether the insurer would use its right of subrogation against the charterer. To prevent such constructions, clause 13 explicitly bestows upon the owner the right to conclude such an insurance contract, by which the insurer does not have any right of subrogation against the charterer. This is tantamount to protecting the default structure of the co-insurance policies, where both the charterer and the owner are co-insured. The letter of clause 13 stirred legal debates and criticism, as depicted in the case law, about the actual meaning of clause 12, which does not explicitly refer to the subrogation rights of insurers, in that the silence of the latter provision laid the door open for insurers to sue charterers after paying insurance proceeds to owners. It is supported, hereby, that, as clause 12 sets out expressly an obligation to contract out of a joint co-insurance policy, it would be against the letter of this provision to conclude, through an a contrario interpretation of clause 13, that insurers are entitled to use their right of subrogation against charterers.70 1.2.5 Impact of safe port guarantees on joint hull insurance policies Gard Marine71 introduced an additional dimension to an already complicated legal problem by discussing the impact of safe port guarantees added in bareboat charter parties on the joint hull

Mandaraka-Sheppard (n 47) 474. BARECON89 (n 4). 69 Ibid. 70 Gard Marine and Energy Ltd v China National [2017] UKSC 35, para. 117. 71 [2017] UKSC 35. 67 68

72  Research handbook on marine insurance law insurance policies. The issue arose because of an additional clause, numbered 29,72 that obliges the charterer ‘not to trade the vessel to unsafe ports where she may suffer an insured loss’.73 As a result of this provision, the Supreme Court had to deliberate on whether the hull insurance policy, which was to be treated as a joint co-insurance contract under clause 12 of the BARECON 89 form, was still providing coverage for the charterer in case of a breach of the safe port guarantee in addition to clause 29. The Supreme Court has not provided a clear solution as it first ruled that there was no breach of the safe port guarantee, and therefore it was not necessary to find an answer to the above technical problem. However, a minority of the judges, including Lord Sumption in the Supreme Court and Teare J in the first instance court, expressed their opinions as obiter dictum as to the impact of the breach of the safe port guarantee on the coverage provided for the charterer and the insurer’s subrogation rights. In effect, in Gard Marine,74 the Supreme Court clarified the ‘unsafe port’ test, the position under clause 12 of BARECON 89 concerning joint insurance and the charterers’ rights to limit their liability under the 1976 Convention. The facts of the case are summarised as follows: the Ocean Victory, pursuant to clause 12 of an amended BARECON 89, was insured by the demise charterers for the respective rights and interests of themselves and the owners. In August 2006, the demise charterers time-chartered the vessel to the intermediate charterer, who sub-time-chartered the vessel to the charterers. Each charter party contained a safe port warranty, whereby it was undertaken to trade the vessel between safe ports. The vessel was ordered to the port of Kashima, which had a history of strong gale-force weather and, to a lesser extent, a history of long waves. The ship was exposed to long waves in berth, and there was a general idea that the moorings could be compromised for a ship for this size if the vessel remained in port. The vessel then left – the master bowing to local experts, such as the charterer’s agent and local pilots – and, when leaving, gale-force winds forced the vessel into the breakwater and she became a total loss. The insurers paid for the total loss, but one of the insurers, Gard Marine & Energy Ltd, later took assignment of the rights of both the owners and demise charterers. Gard Marine & Energy Ltd commenced proceedings against the intermediate charterers for damages for breach of the safe port warranty, and they then joined the charterers in the proceedings. The charterers denied there was a breach of the safe port warranty because the conditions were an abnormal occurrence. The charterers further argued that clause 12 of the demise charter provided for joint insurance without any right of recovery by the owners against the demise charterers, who, in turn, had no liability to pass down the chartering chain to the charterers as they had not suffered any loss. At first instance, Teare J held that there was a breach of the safe port warranty. He held that the risk stemmed from ‘prevailing characteristics’ of the port, with waves of this type having occurred several times, and that the gale-force wind was not ‘an unusual meteorological [event]’.75 There was thus a ‘real, as opposed to fanciful, risk’76 that the danger could occur, and although the concurrence, as he admitted, was a ‘rare event in the history of the port’, this event ‘flows from the characteristics of the port’.77 However, it was not the weather events

74 75 76 77 72 73

Ibid para. 96. Ibid para. 105. [2017] UKSC 35. QBD (Comm Ct) (Teare J) [2013] EWHC 2199 (Comm) – 30 July 2013, [110]. Ibid [106]. Ibid [128].

Co-insurance and rights of subrogation post-Gard Marine And Energy  73 themselves which rendered the port unsafe; it was the failure in port set-up to ensure that vessels of this size were ‘only left in weather conditions with which they could cope’.78 Teare J’s judgment was overruled by the Court of Appeal.79 Longmore LJ remarked that the necessary test was whether the critical combination of long waves and gale-force winds were normal characteristics of the port or an abnormal occurrence, rather than the meteorological events individually. The key reason for this understanding was to place a limit on the scope of the port’s characteristics.80 It was held that there was no breach of the safe port warranty. Although it was accepted that it was not uncommon for the port to experience the two weather conditions separately, they had never before been encountered simultaneously in the port’s 35-year history, and this, coupled with the fact that they occurred very suddenly, made the conditions unexpected and abnormal events. As such, the Court ruled that the charterers should not assume responsibility for them. The Court of Appeal’s judgment was appealed on three grounds: (a) on the safe port issue, as per which the question posed was whether, as a matter of law in the circumstances, there was a breach of the safe port warranty; (b) on the recoverability issue, as per which the question posed was whether clause 12 of BARECON 89 precluded the rights of subrogation of hull insurers and the rights of owners from recovering damages against the demise charterers for breach of the safe port warranty; and (c) on the issue of whether the charterer was entitled to limit its liability for Gard’s losses or any of them. With regard to the safe port issue, the appeal was unanimously dismissed. It had been established in the case of Leeds Shipping Company Ltd v Societe Francaise Bunge (The Eastern City)81 that the charterers would not be in breach of the safe port warranty if the damage sustained by the vessel was caused by abnormal occurrences. The Supreme Court stated that the term ‘abnormal occurrence’ had its ordinary meaning as per guidance that had already been provided for in numerous cases, for example in Compania Naviera Maropan SA v Bowater’s Lloyd Pulp and Paper Mills, Ltd (The ‘Stork’),82 which allowed determining what constituted an abnormal occurrence. The Supreme Court reaffirmed the Court of Appeal’s view that the integral point was not that each adverse weather condition individually was common enough to make it a feature of the port, but that the combination of these two circumstances had never before been encountered. By this reasoning, the event was abnormal. In relation to the joint insurance, Gard argued that pursuant to clause 12 of BARECON 89 terms, they were able to recover the insured value of the vessel from the time charterers as the demise charterer’s assignee. The appeal on this issue was dismissed by a 3:2 majority and with a dissenting opinion by Lord Clarke and Lord Sumption. Clause 12 precluded such a claim.83 The Court was actually split in two with regard to the issue involved. On one side, Lord Clarke and Lord Sumption opined that if there were a breach of the safe port guarantee, despite clause 12, the owner would be entitled to claim damages from the charterer, and the insurer could use its right of subrogation against the charterer after paying the insurance proceeds to

Ibid [118]. [2015] 1 Lloyd’s Rep 381. 80 Ibid [59]. 81 [1958] 2 Lloyd’s Rep 127, 131. 82 [1955] 2 QB 68, 105. 83 Hill Dickinson, ‘The Ocean Victory’, Hill Dickinson Insights, 18 May 2017, www.hilldickinson​ .com/​insights/​articles/​ocean​-victory 78 79

74  Research handbook on marine insurance law the owner. Lord Mance, Lord Hodge and Lord Toulson, on the other hand, strictly separated the issue of safe port guarantee from the construction of the co-insurance policy and prioritised the explicit letter of clause 12, which stated that the insurance should be contracted and construed as a joint policy. In relation to the classification of an unsafe port issue, there were three criteria set out: first, that the date for judging a breach of the safe port promise is the date of nomination of the port; second, that the promise is a prediction about safety when the ship arrives in the future; and third, that safe port disputes should be reasonably clear and defined by the answer to the culminating question of whether the danger constitutes an abnormal occurrence, which is rare and unexpected, or whether it is something which was typical for the particular port and vessel in this port at the time.84 There may be three possible ways to explain the dynamics behind such reasoning. The first two would seem to be more directly grounded upon principles of insurance law and aim at tactically surmounting technical obstacles that arise from the legal qualification of the hull insurance as a joint insurance contract. In contrast, the third one would strategically attempt to employ the principles of English property law as leverage; more specifically, to split the legal interests of the co-insureds or the owner and the charterers, thus overall reinterpreting the hull insurance contract as a composite insurance policy in order to let an action of subrogation against charterers as co-insured be admissible. 1.2.5.1 The wilful misconduct or intent of the charterer as co-insured Section 55(2)(a) of the Marine Insurance Act 1906 sets out that ‘The insurer is not liable for any loss attributable to the wilful misconduct of the assured’. Although the assured can still enjoy insurance protection where its negligence or misconduct is attributable to the damage, if the degree of its fault reaches wilful misconduct85 or intent, the insurer is no longer liable to pay the insurance coverage. Wilful misconduct denotes and reflects a degree of fault which results from a reckless act whose consequences the actor is aware of, and yet the actor does not care whether loss will result or not. The safe port guarantee can be a parameter to construe the level of fault of the assured in marine insurance contracts. Although not embedded in the insurance contract itself, the safe port guarantee defines the scope of protection sought by the co-assured persons from the insurer. In other words, the insurer can be deemed to have offered the insurance policy under the condition that the charter party included the said safe port guarantee, provided there is an express term for this whereby a violation would also be a breach of the insurance contract. In many instances, a breach of the safe port guarantee by the charterer involves intent. It is hard to imagine a charterer unwillingly or negligently berthing the ship to a port; therefore, if the port is unsafe, there is an intentional breach of the charterparty. Berthing is intentional but the intent must be berthing to an unsafe port, and unsafe nature needs to be known by the assured. This intentional breach of the charterparty automatically triggers an intentional breach of the insurance contract by the charterer, due to a term on safe port, whereby he is legally qualified as the co-assured under the latter because the insurer only undertakes to provide coverage for the specific charterparty with a safe port guarantee. Is it still possible for the charterer to prove that the breach of the safe port guarantee under the charterparty was not intentional, as it would be in the case of wilful misconduct, whereby Ibid. National Semiconductors (UK) Ltd v UPS Ltd, [1996] 2 LL Rep 212.

84 85

Co-insurance and rights of subrogation post-Gard Marine And Energy  75 intentional damage would be assumed? If so, would it result in the insurer being liable to pay the insurance coverage under section 55(2) of the Marine Insurance Act 1906? The answer to these questions should be yes, although this probability would be limited to exceptional cases. For instance, certain acts of God, such as tempests, blizzards or fire, might compel the charterer to berth the ship to the closest harbour. In such cases, the visit to the unsafe port must not be regarded as trading the vessel under the letter of the above-mentioned additional clause 29, as such a visit is the result of a necessity rather than a commercial decision. Therefore, at least, it is feasible to argue that the charterer’s decision cannot be regarded as wilful misconduct and intentional breach from the perspective of the insurance contract. Another possible defence, as to the level of the fault in the breach of the insurance contract via the breach of the safe port guarantee, may be that there is no consensus among the maritime sector as to the unsafe status of the port. If proven, such a defence may also invoke the insurer’s liability, refuting the application of section 55(2) of the Marine Insurance Act. 1.2.5.2 Breach of an insurance warranty by the charterer as co-insured Another technical ground to discharge the insurer from liability due to the breach of the safe port warranty under the charterparty is to exalt this provision to the level of a warranty in the insurance contract.86 Supposing that a safe port guarantee of the charterparty may turn out to be a promissory warranty in the insurance contract, the insurer may be able to repudiate its liability while the co-insured’s obligation to pay the premium would still survive. In case of a breach of warranty, the level of the fault of the insured or the gravity of the breach is also irrelevant; however, if the warranty is repaired, the cover is then reinstated. Prior to the Insurance Act 2015 if a warranty was breached, automatic and permanent termination of cover would be the insurer’s sole remedy for breach of warranty. After the Insurance Act 2015, automatic termination of the risk is no longer available. If a warranty is breached under section 10 of the Insurance Act 2015, cover will simply be suspended from the time of the breach until the breach has been remedied, provided the breach can be remedied. Once the insured has remedied the breach, the insurer will be liable for subsequent losses, unless the loss is attributable to something that happened during the period that cover was suspended. In a co-insurance policy, a breach of warranty affects the co-insured who breached the warranty, in other words, the demise charterer. The owner should still be able to claim the insurance proceeds from the insurer. The insurer, in return, can use its right of subrogation against the charterer for whom the insurance contract has been terminated. However, the contract is not terminated for breach of a warranty, either in the old or new regime, but upon the breach of warranty, and whereby liability has arisen under the demise charter upon the violation of the safe port warranty. Any attempt to raise the safe port clause to an insurance warranty must also survive the contra proferentem rule from the perspective of the insurer.87 First, the insurer must show that the wording of the safe port warranty is so precise as to be construed as a warranty for the insurance contract, which was highly likely signed subsequently. In other words, the insurance contract must clearly express that the previously agreed safe port warranty among

86 For more information on the meaning of the term warranty in general contracts law and insurance law see Lowry, Rawlings and Merkin (n 52) 217; Alastair Owen, The Law of Insurance Warranties (1st edn, Routledge 2021) para. 3.2. 87 Merkin and Steele (n 1) 55.

76  Research handbook on marine insurance law the co-insureds is regarded and accepted as a promissory warranty. Second, the letter of the safe port guarantee clause must also be apt for constructions to exclude the charterer from the insurance protection. In particular, one should not be able to deduce from other provisions of the charterparty that the owner waived its rights to claim damages from the charterers despite the safe port warranty, a probability which would be surprisingly more commonly expected if the contract were not professionally drafted. In this second step, it would be crucially important if another provision of the insurance policy came to provide help as leverage justified in order to reach that conclusion, because it is highly hypothetical to consider that the parties to the charterparty intended to create such a promissory warranty in favour of the undertaker of a prospective insurance policy. 1.2.5.3 Differentiating the legal interests of the owner and the charterer on the vessel In the background of the above observations, the decisive point at issue seems to be whether the breach of a safe port guarantee by the charterer results in differentiating, or more specifically, splitting its legal interests in the vessel from that of the owner. Although the demise charterer of a ship is entitled under English law to proprietary rights like an owner,88 the breach of the safe port guarantee may put the charterer in a position whereby he would not have a proprietary right on the vessel because it is logical and righteous to expect any holder of a proprietary right on a vessel to be cautious so as to not damage the vessel intentionally. The breach of the safe port guarantee shows that the charterer did not act as someone who would have a proprietary right on the vessel. Consequently, it becomes conceivable to assume the role of such charterers to be that of legal persons who cannot technically enjoy any protection under hull insurance as they lack an insurable interest. A similar conclusion cannot be deduced for the owners of the vessel whose property rights are not subject to discussion. However, in Suez Fortune Investments Ltd v Talbot Underwriting Ltd (The Brillante Virtuoso)89 the owners were found to have deliberately destroyed their ships and the High Court ruled that the vessel had been scuttled by its owner and gave guidance on the legal tests for establishing ‘wilful misconduct’ on the part of an insured, that is, it clarified that insurers need be able to prove fact-dependent findings that the owner wilfully and intentionally damaged their own vessel. It is to be noted that, in such cases, the owners lose their right to insurance protection not due to an assumption of the lack of property rights and insurable interest but as a result of the explicit provision of section 55(2)(a) of the Marine Insurance Act. For a co-insured charterer who violates a safe port guarantee, both the lack of property rights and section 55(2)(a) of the Marine Insurance Act can extinguish the right to insurance protection. The case of The Brillante Virtuoso90 has demonstrated the importance of acknowledging that underwriters cannot be expected to uncover the full radius of the fraud committed by a ship owner. By stripping the charterer out of his legal status of co-insured, the above-mentioned theoretical distinction can prevent further discussions as to the non-liability of the insurer in case of the wilful misconduct of one of the co-insureds. Furthermore, any insurance proceeding must only be paid to the owner; however, the owner may still be entitled to sue the charterer as long as its loss would not be totally covered by the insurer, and the insurer shall be entitled to use its right of subrogation for the paid amount of insurance proceeds against the charterer. As an

Mandaraka-Sheppard (n 47) 474. [2019] EWHC 2599 (Comm). 90 Ibid. 88 89

Co-insurance and rights of subrogation post-Gard Marine And Energy  77 additional result of differentiating the legal interests of the charterer from those of the owner – thus alienating the former from the insurance protection – an important theoretical obstacle before the subrogation claim is eliminated.

2.

RIGHT OF SUBROGATION

2.1 Definition Except for contingency insurances,91 the right of subrogation confers upon the insurer a right to direct any action that the insured has against the wrongdoer who caused the loss in connection with the risk defined in the insurance contract,92 To avail itself of the right of subrogation, the insurer must first pay the insurance proceeds to the insured.93 If the third party made any payment to the insured in relation to the loss, after the payment of the insurance proceeds, the insurer is entitled to recover from the insured the amount that exceeds the loss up to the amount paid by the insurer.94 Within this scope, the right of subrogation helps insurers mitigate their loss due to the occurrence of the risk, continue their business and uphold the insurance sector. Thus, it prevents placing the legal liability of the wrongdoers on the insureds who paid premiums to the insurer. Consequently, it favours all the insurance market players and economic actors. Finally, the right of subrogation prevents the insured from enjoying a double recovery. 2.2

Legal Ground and Legal Nature

From what legal source the right of subrogation actually stems is still a debate in English law.95 One view put forward by Lord Diplock states that the right of subrogation is an implied term of the insurance contract.96 Another opinion links it to equity law, as a result of which the courts are bestowed discretionary power to compel to pay damages to those insureds who are reluctant to give their consent to the insurer to use the right of subrogation.97 Whatever its legal ground in English law is, there is consensus on the fact that the right of subrogation is not independently provided for the insurer. In other words, the insurer always needs consent from the insured to bring action against the wrongdoer, as it does so on behalf of the latter. As the right of subrogation is an extension or reflection of the rights of action of the insured, it can prejudice this right by totally or partially waiving its claim against the wrongdoer. In order to counterbalance such disadvantages against the insurer, the legal precedent has

Lowry, Rawlings and Merkin (n 52) 352. Ibid 351. 93 Ibid 353. 94 Ibid 352. 95 In Continental European law jurisdictions the right of subrogation directly emanates from the statutory provisions. For instance, Article 86 entitled ‘the assignment of claims’ (‘Übergang von Ersatzansprüchen’) of the German Insurance Contract Act 2008 (Gesetz über den Versicherungsvertrag). It is also not uncommon to come by legislation defining the right of subrogation in the United States, such as the Riots (Damages) Act 1886, s. 2(2) and Indiana Code 27-7-5-6 (see ibid 350–1, fn 13). 96 Yorkshire Insurance Co. Ltd. v Nisbet Shipping Co. Ltd. [1962] 2 QB 330 at 340; Merkin and Steele (n 1) 99, fn 16. 97 Lowry, Rawlings and Merkin (n 52) 355. 91 92

78  Research handbook on marine insurance law crafted a principle to raise the contractual liability of the insured who abandons its claim to the detriment of the insurer.98 2.3

Insurer’s Right of Subrogation against Co-insureds in Hull Insurance

2.3.1 Legal nature of the right of subrogation The ancillary nature of the right of subrogation to the right of action of the insured against the wrongdoer furnishes further need for scrutiny when the same insurance policy also covers the wrongdoer. Such a situation may arise in the case of joint insurance or co-insurance, where one co-insured causes harm to the other. In joint insurance subrogation is not permitted against the co-assured, simply because the interests are joint, and not separable. On the other hand, when the interests of the insureds in the insured property are not the same – in other words, when there is a composite insurance policy – it is not easy to use the above patterns in order to filter out the right of subrogation of the insurer against the co-insured whose actions resulted in the occurrence of the risk.99 This is because the different legal nature of the interests covered concerning each insured may open doors for damages claims and actions to be directed by one against another. Indeed, in this case, it is more plausible to differentiate the legal liabilities of both co-insured regarding rights of a different kind and hold one of them liable against the other. To accept that payment of the insurance proceeds by the insurer makes good the loss of the prejudiced co-insured and thereby ends the liability of the co-insureds between each other and does not affect subrogation rights is problematic, as it would weaken the intention to be insured together, and the fact that the guilty co-assured is insured under the same policy cannot be overlooked in relation to subrogation. Liability is always determined, as per the underlying contract, between assured and co-assured, and also depends on whether the contracting parties’ liability is excluded or retained by the terms of the contract or not.100 Rix LJ’s judgment in Tyco Fire101 (a) distinguished the case where the underlying contract clearly provides that there is to be no liability of a contractor to his employer in the area of the regime for joint names insurance, being of the opinion that the true basis of the rule is that the contract between the parties works in a straightforward way (as per Cooperative Retail Services Ltd v Taylor Young Partnership Ltd);102 and (b) stated that there is clarity in favour of the contractor’s continued liability to his employer for his negligence. However, in Tyco Fire103 Rix LJ also distinguished Cooperative Retail Services104 and held that the liability provision must not be ignored in assessing a claim by one co-assured, or by the insurer in the assured’s

Commercial Union Assurance Co Ltd v Lister (1874) LR9 Ch App 483. This probability is also worthy of analysis because separating the legal interests of the Charterer and the owner has been accepted as a valid ground for reasoning in favour of the insurers who direct their subrogation rights against Charterers. 100 Cooperative Retail Services Ltd v Taylor Young Partnership Ltd [2002] Lloyd’s Rep IR 555, para. 65; Rathbone Brothers Plc v Novae Corporate Underwriting Ltd [2015] Lloyd’s Rep IR 95, para. 89; Gard (n 3); Özlem Gürses (2017), ‘Subrogation against a contractual beneficiary – a new limitation to insurer’s subrogation?’ Journal of Business Law, 7, 556–74, 560. 101 [2008] 1 CLC 625 CA (Civ Div). 102 [2002] Lloyd’s Rep IR 555. 103 [2008] 1 CLC 625 CA (Civ Div). 104 [2002] Lloyd’s Rep IR 555. 98 99

Co-insurance and rights of subrogation post-Gard Marine And Energy  79 name in subrogation against another. The above was also discussed in obiter in Gard Marine.105 Lord Sumption relied on the general rule that insurance recoveries are ignored in the assessment of damages arising from a breach of duty, and stated that protecting insurers’ subrogation rights is necessary as an important part of the economics of insurance; hence against the third party wrongdoer, the insurance payment is res inter alios acta, in other words, one could argue it as not relevant and not applicable.106 However, it is notable that in Gard Marine107 clause 13 of the charterparty provided an express waiver of subrogation by the owner against the charterer but was optional; that is, if the parties did not expressly agree that clause 13 applied, then clause 12 did, and, as per Lord Sumption, clause 13 was designed for a very different kind of chartered service. However, precluding insurers’ subrogation rights in such cases is not forbidden, since enforcing such rights might have led parties to seek alternative own insurance to cover their liability for the same risks.108 2.3.2 Impact of the safe port warranty on the right of subrogation In light of the above explanations, if the charterer can be deemed as co-insured in an insurance contract, there is substantial ground to argue that the owner of the vessel, as the other co-insured, has waived its right to claim damages against it for losses occurring within the sphere of the insurance protection. The safe port warranty clause inserted in the charterparty provides a pathway to a possible assumption that the owner has only abandoned his right to sue the charterer without prejudice to losses arising from the violation of the safe port warranty. Therefore, the right of action concerning losses due to violating the safe port warranty would still be valid between the owner and the charterer. This, in return, can be assigned to the insurer, either under section 79 of the Marine Insurance Act 1906 or in line with the provisions of the insurance contract once the insurance proceeds have been paid to the owner.109 In Gard Marine110 the court did not base its decision on the implied term, and Lord Mance pointed out that the owners had no claim against the charterers as it was understood that there would be no such claim.111 However, Lord Clarke did not justify the exclusion of the demise charterer’s liability to the owner by virtue of clause 12 or accept that this could be implied.112 Nonetheless, the right of action and its assignment to the insurer is a necessary but insufficient condition to claim damages from the charterer through the right of subrogation. The other necessary condition for successful subrogation action is the substantive existence of the claim that the obligation to provide insurance security of the insurer against the charterer as the co-insured in the insurance contract must not be overridden. In other words, the charterer can still argue that the insurance protection is valid for him as co-insured, which prevents the right of subrogation. We argue that such a construction is at least coherent on a logical basis and deserves further study as a probability that must be refuted. Undeniably, if the charterer has shown misconduct, the very nature of his misconduct removes the ‘accidental’ nature of the loss. Section 55(2)(a) MIA 1906 allows recovery 107 108 109 110 111 112 105 106

[2017] 1 Lloyd’s Rep 521 SC. Gürses (n 100) 561–3. [2017] 1 Lloyd’s Rep 521 SC. Gürses (n 100) 571–2. Ibid, 575. [2017] 1 Lloyd’s Rep. 521 SC. Gard Marine (n 3), at 122. Ibid at [53]; Gürses (n 100) 565.

80  Research handbook on marine insurance law for wilful misconduct of the assured owner’s master and crew, but not for a co-assured. If a co-assured deliberately causes the loss, this will prevent the owner’s claim against the insurer, as demonstrated in Samuel v Dumas113 and The Brillante Virtuoso.114 Both cases held that a co-assured’s wilful misconduct or deliberate loss prevents the loss from being a peril insured against. If the loss was not caused by a peril insured against, such as a peril of the seas, the co-assured – in these cases the owner – had no right to claim. The charter’s misconduct automatically removes the ‘accidental’ element of the nature of the loss. If there is no ‘accidental’ element, the insurer is not under an obligation to pay compensation to any of the assureds. However, if the charterer negligently nominates an unsafe port, then the owner can claim against the insurer, but so can the charterer, as ‘negligent’ nomination would be covered in principle. The charterer, however, will still be liable to the owner under the charterparty given that negligent breach of a charterparty is still a valid breach. Whether the safe port warranty furnishes a warranty clause under the insurance contract stops being a problem for the insurer due to section 55(2) of the Marine Insurance Act, because the breach of such a warranty is sufficient to prove that the charterer acted on wilful misconduct as assured and extinguished the insurer’s liability with respect to himself, but not against the owner. Consequently, the insurer no longer needs to rely on a breach of a warranty clause. If the charterer’s violation of the safe port warranty can be conceived as non-intentional, then its defences to the continuing validity of the insurance protection do not automatically fall. Indeed, such allegations may successfully support the charterer’s case unless the safe port warranty cannot be construed as a warranty under the insurance contract, which in return necessitates an explicit provision in the policy. Here also, there is no sphere to apply the contra proferentem rule. The gist of the use of the contra proferentem rule in insurance cases is that the wording in a contract is to be construed against the party seeking to rely on it and against the party who proposed it, and the courts would leap at the chance to use it against an insurer who has failed to express themselves clearly.115 In Pratt v Aigaion Insurance Company SA (The ‘Resolute’),116 in allowing the appeal, the Court of Appeal held that the typed warranty was ambiguous and should therefore be read contra proferentem, that is, contrary to the interests of the party that seeks to rely upon the warranty, that is, against the insurer. In addition, given the provisions of condition 26 in the incorporated standard Aigaion ‘Trawler Wording’, the typed warranty’s phrase ‘at all times’ had to be given some qualification so as to limit its scope to circumstances when the vessel was navigated, or in other circumstances where the presence of the owner/owner’s experienced skipper and one crew member on board would be appropriate. However, the contra proferentem rule should not be considered to apply as a rule, as so far it has only been applied in motor insurance cases, such as in consumer insurance, and in the marine case of ‘Resolute’.117 In ‘Resolute’118 the appellant’s fishing trawler, The Resolute, was seriously damaged by a fire in December 2006. The contract of insurance under which the vessel was covered contained a warranty provided by the appellant providing for the owner

[1924] UKHL J0225-1. [2019] EWHC 2599 (Comm). 115 Jeremy Hill (2009), ‘Terminal traps in insurance contracts’, In-House Lawyer, Legal Briefing, Dec/Jan 2009, www​.inhouselawyer​.co​.uk/​legal​-briefing/​terminal​-traps​-in​-insurance​-contracts/​ 116 [2008] EWCA Civ 1314. 117 Ibid. 118 Ibid. 113 114

Co-insurance and rights of subrogation post-Gard Marine And Energy  81 and/or owner’s experienced skipper to be on board and in charge at all times, as well as one experienced crew member. At the time of the fire, the vessel was moored in port and empty. The owner claimed on the insurance policy but was denied compensation by the insurer on the grounds that the above warranty had not been complied with. While the owner argued that the clause was clearly intended to refer only to periods when the vessel was navigating or working and, if applied literally, would give rise to non-intended results for the assured, the insurer submitted that the clause should be given its literal meaning. The Admiralty Court found in favour of the insurer and dismissed the claim on the grounds that the appellant had not complied with the warranty. In the Court of Appeal, the main principles of contract interpretation considered in the judgment of Sir Antony Clarke MR were: (a) that a clause in a contract must be construed according to its factual matrix, that is, in relation and with regard to the context, as per Investors Compensation Scheme Ltd v West Bromwich Building Society,119 whereby it was held that the essence and nature of interpretation is to ascertain the meaning which the document would convey to a reasonable person having all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract. In addition, the argument supported in Sirius Insurance Co v FAI Insurance120 – that there has been a shift from literal methods of interpretation towards a more commercial approach – was cited. There is a generally accepted argument that the first relevant rule of construction is that the apparently literal meaning of the words in a warranty must be restricted if they produce a result inconsistent with a reasonable and business-like interpretation of such a warranty, and that the second principle of construction which assists the assured who contends that he has complied with the warranty is that any ambiguity in the terms of a policy must be construed against the insurer. Moreover, the EU regulations relevant at the time limited the number of days for which a vessel like Resolute is permitted to fish to 227 per year; as a result of this such vessels are intermittently tied up for periods or stored, and in any case the skipper is expected to go ashore, whether in connection with the vessel or not. The Court of Appeal held that the warranty should be interpreted in a commercial fashion, and while the natural inference of the words ‘Warranted Owner and/or Owner’s skipper on board and in charge at all times’ was that the owner or skipper should be on board to protect the vessel against any navigational hazards, the second part of that warranty, requiring in addition to the owner and/or skipper ‘one experienced crew member’, strongly supported the assumption that the underlying purpose of the warranty was to protect the vessel during navigation or operation as a fishing vessel, when a minimum of two crew members would be required. Also, the significance of the requirement to have experienced crew ‘on board and in charge at all times’ was considered and found appropriate to give this statement, and therefore – as per the contra proferentem rule – if the insurer had wanted two crew members on board also in non-navigating times, that is, while the boat was docked, this should have been clearly stipulated. The appeal was therefore allowed and the owner was entitled to recover under the contract of insurance.121 However, applying the contra proferentem rule in a commercial

[1998] 1 WLR 897. [2004] UKHL 54; [2004] 1 WLR 3251. 121 Michael Mendelowitz (2009), ‘Insurance Updater’, 5 February 2009, Norton Rose Group, https://s3.amazonaws.com/documents.lexology.com/dd9a984c-3b6e-4f0c-ab71-edccd515b481​ .pdf?AWSAccessKeyId​=​AKIAVYILUYJ754JTDY6T​&​Expires​=​1680179129​&​Signature​=​FwCCD​ %2BHR2EJRXZB1z7xGBOlRKHY​%3D 119 120

82  Research handbook on marine insurance law case like this, where an asset worth millions of pounds is insured and the parties are generally regarded by the courts as being able to defend their interest at the outset of the contract, cannot be accepted. Hence, the rule of contra proferentem should not apply as a rule to lift ambiguities in commercial insurance cases, due to the volume of sums insured involved, even if the reason for using it in insurance is the apparent imbalance of bargaining power between insurer and insured and the rationale is to continue teaching insurers the benefits of clear drafting.122 Rounding up, whether the violation of the safe port warranty enables the insurer to use its right of subrogation against the charterer seems likely to depend on a series of circumstances. On this background, Gard Marine123 has opened the Pandora’s box wide, and detailed studies of various probabilities affecting the insurer’s right of subrogation regarding marine insurance policies are now more necessary than ever.

3. DISCUSSION 3.1

Gard Marine

In Gard Marine124 the Supreme Court decided that the port was safe; it went on to say that if it had been unsafe, Gard would not have been entitled to recover because of clause 12 of the BARECON 89. The liability issue and the limitation of the insurer’s subrogation right apart, it is notable that the Supreme Court was divided in terms of whether liability either had never arisen or was discharged with the insurance payment, and if this could be extended to sub-charters. It is also notable for questioning whether the view of the minority in the Supreme Court (including Lord Sumption), as well as the judge at first instance, would have allowed claims between co-insureds, which is against the established rule. Another question is whether the minority view might have been limited in effect, either depending on the type of the particular contract that BARECON 89 was and its clauses 12 and 13, or simply because this was the view of a minority, which was obiter anyway.125 It upheld the Court of Appeal’s decision that there was no breach of the safe port undertaking and reaffirmed the position established by the Court of Appeal that within the remit of a safe port, the term ‘abnormal occurrence’ should be attributed its ordinary meaning. Essentially, what the Supreme Court herein achieved was to reaffirm the Court of Appeal’s view that such an event means something not normal, unexpected and outside the remit of anything of an ordinary course of events, the occurrence of which would not be anticipated by the notional charterer or owner. The Supreme Court considered whether the port was unsafe within the meaning of the safe port undertaking so that the charterers were in breach. Alternatively, was there an abnormal occurrence so that there was no breach as a consequence? Lord Clarke stated that the date for judging a breach of the safe port undertaking was the date of nomination of the port and that

Hill (n 115). Gard Marine (n 3). 124 [2017] UKSC 35. 125 Jeremy Roberts and Jonathan Spencer, (2017) ‘Storm in a teacup? The Gard Marine case: subrogated claims and co-insurance’, Simmons and Simmons, 17 May 2017, www​.simmons​-simmons​ .com/​en/​publications/​ck09r2​2em62pf0b9​4417y97yf/​170517​-gard​-marine​-case​-subrogated​-claims​-co​ -insurance 122 123

Co-insurance and rights of subrogation post-Gard Marine And Energy  83 it also assumed normality. In arguing this, he cited Robert Goff J in Kodros Shipping Corp of Monrovia v Empresa Cubana de Fletes (The Evia) (No 2),126 whereby it was affirmed that any test of whether a port is unsafe must assume normality and exclude danger caused by some abnormal occurrence. 3.2 Haberdashers’ Haberdashers’127 was settled and hence did not serve the purpose of further discussing subrogation rights issues. It represents a missed opportunity to answer the question left open in Gard Marine128 on whether – in a case of a co-insurance policy where a sub-contractor causes loss, where the co-insurance policy does not cover the sub-contractor – the insurer can bring a subrogated claim against the sub-contractor, or whether it would first need to be proven that the sub-contractor is liable for the loss. Rugby Union129 answered this question. In Haberdashers’,130 the focal point of concentration was whether the sub-contractor was or was not co-insured. The actual question regarding the liability of the sub-contractor as such was not examined. Arguably, a missed opportunity occurred when it was taken as given that the sub-contractor caused the damage and, as such, was liable for it. The majority view had stated that co-insurance excludes liability as between co-insureds. It follows from this that the main contractor itself bears no liability, and as a result there is no justification for a back-to-back claim against the sub-contractor. The minority view in Gard Marine131 expressed an opposite view and stated that there is a liability, but the latter is discharged on the basis of the existence of co-insurance. As the majority in the Supreme Court did not agree, this is where it was left. In addition, regarding liability and back-to-back claims, the issue was left entirely open because, on the facts of Gard Marine,132 it was not tested. It appears that post-Gard Marine,133 the main issue and question to be asked is whether, if we assume that co-insurance results in the exclusion of liability and not in the discharge of liability between the co-insured parties, and what the justification is for a subrogated claim against the third party wrongdoer. More specifically, and in relation to Haberdashers’,134 if the project policy excluded liability between the main contractor and the employer, what liability existed in relation to the sub-contractor to find the basis of the project insurers’ subrogated claim against it? Again, in Haberdashers’,135 the above question was not considered and was left entirely open.136

[1981] Lloyd’s Rep 613. [2018] EWHC 588 (TCC). 128 [2017] UKSC 35. 129 [2022] EWHC 956 (TCC). 130 [2018] EWHC 588 (TCC). 131 [2017] UKSC 35. 132 Ibid. 133 Ibid. 134 [2018] EWHC 588 (TCC). 135 Ibid. 136 Natalie Wardle (2019), ‘A missed opportunity: Haberdashers and subrogation’, Thomson Reuters Practical Law Construction Blog, 11 June 2019, http://​constructionblog​.practicallaw​.com/​a​ -missed​-opportunity​-haberdashers​-and​-subrogation/​ 126 127

84  Research handbook on marine insurance law

4.

FINAL REMARKS

Co-insurance is a complex area, with conflicting legal principles at play and fragmented case law.137 Case law such as Gard Marine138 affirms in the Court of Appeal and the Supreme Court judgments the view that insurance and subrogation can be a complicated matter. In Gard Marine139 there was no specific exclusion of subrogation rights. The test for assessing whether a nominated port was safe for the purposes of a ‘safe port’ warranty was based not on reasonable foreseeability but on whether, absent an ‘abnormal occurrence’, a reasonable shipowner with all relevant knowledge would proceed to it. The phrase ‘abnormal occurrence’ bore its ordinary meaning and referred to a rare, unexpected event, including a rare concurrent occurrence of two or more commonplace events. The Court examined the effect of clause 12 of the BARECON 89 contract regarding composite insurance. Lord Sumption140 dealt with the issue of the safe port and the way this interacted with the BARECON 89 form. He noted that the latter was amended by deleting the trading limits clause (clause 5) and adding to clause 29 a safe port warranty141 and that it should not be disputed that if Kashima was an unsafe port, there was a breach of clause 29. He also noted the provisions which are said to exclude a right to recover the value of the ship as damages for breach of clause 29 and clauses 12(a) and (c), that the argument is that the clause as a whole is a complete code governing financial liability for loss or damage to the ship and that the words quoted provide for the relief, and the only relief, available as between the head owner and the demise charterer for a total loss. His analysis showed that, even if the starting point is the general rule that insurance recoveries are ignored in the assessment of damages arising from a breach of duty,142 the courts have traditionally been concerned with preserving the subrogation rights of insurers against those who are legally responsible for the loss – which is an integral part of the economics of insurance – but that the effect of the collateral payments exception is that, as between the insured and the wrongdoer who has caused the loss, they are not treated as making good the former’s 137 James Clarke (2018), ‘Co-insurance and subrogation rights under the spotlight once more’, Construction Blog, Practical Law, May 2018, http://​constructionblog​.practicallaw​.com/​co​-insurance​ -and​-subrogation​-rights​-under​-the​-spotlight​-once​-more/​ 138 [2017] UKSC 35. 139 Ibid. 140 Ibid [96–106]. 141 Ibid [96], the exact wording being in the following terms: ‘29. Trading Exclusions: “Vessel to be employed in lawful trades for the carriage of lawful merchandises only between good and safe berths, ports or areas where vessel can safely lie always afloat, always accessible within IWL except NAABSA in River Plate where it is customary for similar size or similar dimension vessels to safely lie aground, specially excluding Abkhazia, Albania, Angola, Bosnia-Herzegovina, CLS Pacific ports, Democratic Republic of Congo (formerly Zaire), Eritrea, Israel, North Korea, Lebanon, Liberia, Libya, Sierra Leone, Somalia, Sri Lanka, Federal Republic of Yugoslavia, Zimbabwe, in Arabian Gulf and adjacent waters including the Gulf of Oman North of 24 deg. North, any United Nation embargo countries/ports. Charterers have right to send vessel to the war/warlike zone or other zones for which additional insurance are levied by vessel’s war risk insurers. In such event, Charterers are fully responsible to pay for all additional war risk premium upon demand by vessel’s underwriters and/or P+I club with all risks/consequences to be for Charterers’ account. Charterers shall have right to break IWL in which case Charterers are fully responsible to pay for all additional premium upon demand by vessel’s underwriters and/or P+I Club for breaching IWL with all risks/consequences to be for Charterers’ account. Any ice affected port(s) and/or place(s). No direct sailing between PRC and Taiwan or vice versa’ (BIMCO (ν4)). 142 Bradburn v Great Western Railway Co (1874) LR 10 Ex 1; Parry v Cleaver [1970] AC 1.

Co-insurance and rights of subrogation post-Gard Marine And Energy  85 loss or as discharging the latter’s liability for, as far as the wrongdoer is concerned, insurance is res inter alios acta. Our discussion has allowed us to conclude that the main line of argument followed in the decision of the Court of Appeal was that the prima facie position, where a contract requires a party to that contract to insure, should be that the parties have agreed to look to the insurers for indemnification and not to each other, and that insurance is issued to cover for the consequences of their own negligence and no rights of subrogation will arise even if the insurance was intended to be for the joint benefit of the parties. The Court of Appeal concentrated on whether the scheme of insurance and the implied term that the parties would not subrogate overrode the safe port warranty; it answered this in the positive, and it also found that an express exclusion of subrogation rights did not mean that subrogation would be allowed. This was reiterated by the Supreme Court, which asked whether liability exists between the two co-insured parties or whether the effect of the co-insurance arrangements is to exclude that liability. Lord Sumption stated that it was erroneous, on the basis of the insurance arrangements, to find the sub-charterer with no liability, as liability existed but the insurance rectified it, that is, satisfied it. The ability to claim against a third party wrongdoer sub-charterer survives.143 However, Gard Marine left a number of questions unanswered, such as (a) the strength of the implied term that one co-insured party may not sue another, and when it may be rebutted; and (b) the juridical basis for the implied term and its consequential impact on sub-contractors. On the positive side, it has helped to increase clarity on the issue of subrogation and prompted BIMCO to introduce the BARECON 2017 clause 17(a), to apply regardless of who issues the insurance and establish that the existence of the partiers as co-insureds in the hull policy does not exclude them from liability – even if BARECON 2017 does not solve the problem of subrogation holistically but instead, through clause 17(a)(ii), establishes that payment by the insurer shall be no bar to a claim for indemnity against the charterer by way of subrogation.144 The issues left unanswered in Gard Marine145 have also arisen in later cases, with no full answers. Haberdashers’146 offers yet another reminder to contracting parties of the need to read and fully understand the terms of their contract before signing it and to avoid assuming the existence of coverage as co-assured. As demonstrated in Rugby Union,147 joint name policies for all-risks insurance may not provide the extent of cover that their titles suggest; hence each party needs to consider and fully understand the extent of cover actually provided for in the underlying construction contract, to ensure identification and coverage of any gaps with express language and wording.148 The decision of the Court of Appeal in Rugby Union149 has important implications, as: (a) the decision is seminal also because it is the first at appellate level where the conclusions on co-insurance form part of the ratio; that is, in relation to the Court’s explanation of the prin143 Natalie Wardle (2017), ‘Joint insurance and rights of subrogation revisited’, Thomson Reuters Practical Law Construction Blog, 31 May 2017, http://​constructionblog​.practicallaw​.com/​joint​-insurance​ -and​-rights​-of​-subrogation​-revisited/​ 144 Gilabert Gascón (2021), ‘Insurance related problems in bareboat charter agreements’, Journal of Shipping & Trade, 6, 12–32, 28–32. 145 [2017] UKSC 35. 146 [2018] EWHC 588 (TCC). 147 [2022] EWHC 956 (TCC). 148 Clarke (n 137). 149 [2023] EWCA Civ 418.

86  Research handbook on marine insurance law ciples governing co-insurance and the scope of cover, it is the first time that these principles have formed part of the ratio of any decision at appellate level. As such, while the previous decisions remain important authorities, the Court of Appeal’s decision in Rugby Union is the main authority on these issues; (b) the underlying building contract between parties insured under the same policy remains the key document for the available cover; (c) the possibility of appealing to the broader dealings as an alternative source of authority/intention was left open for an appropriate case; (d) the definition of insureds by reference to their ‘respective rights and interests’ may provide an additional reason why the scope of a contractor’s cover under a joint names insurance policy is limited to that envisaged by their underlying contract.150 Subrogation may be contested as a restitution mechanism as unfair and interfering with the assured’s rights against the party who is liable for the insured loss. However, any such accusations are superseded by the fact that the main rationale, aim and objective behind the principle and mechanism of subrogation is to prevent the wrongdoer from escaping liability after the entitled party has received a payment from the claimant.151Where there is an enforceable claim which has been avoided by the party at fault, subrogation is well placed to intervene. It also serves the goal of avoiding over-indemnification of the assured, whether we are dealing with indemnity insurance or co-insurance.152 On the basis of the existing case law, it stems that the insurers can be prevented from exercising subrogation rights against co-insureds, in construction risk insurance, on the basis of policy considerations. It is also apparent from the examined case law that subrogation rights are secured when the language is certain in prescribing them.153

Gateway Chambers (n 30). Gürses (n 100) 574; Phillip James (1971), ‘The fallacies of Simpson v Thomson’, 34 Mod. L. Rev., 149, 152; Charles Mitchell, ‘The law of subrogation’ [1992] L.M.C.L.Q. 483, 486; Merkin and Steele (n 1) 100. 152 Gürses (n 100) 575. 153 David Edwards (1998), ‘Subrogation against co-insureds – an international perspective’, International Law Review, 6(10), 311–16, 316. 150 151

5. Scuttling, fortuity and marine perils – from the mortgagee’s and the cargo owner’s points of view Özlem Gürses1

1. INTRODUCTION Where a loss is attributable to the assured’s wilful misconduct, the insurer will not be liable for that loss.2 Scuttling or deliberate casting away of a ship with the connivance of the insured shipowner will be assessed under this exception. If the loss was caused deliberately but without the insured shipowner’s involvement, it could either be a barratry (if the crew cast away the ship) or a ‘malicious act’ of a third party aiming to harm the shipowner. ‘Barratry’ and ‘malicious acts of third parties’ are commonly included among the risks insured against under the shipowner’s hull and war risks insurance policy.3 Both occur when there is an act done against the owners of the vessel, hence they apply only when the insured shipowner is not privy to the relevant act that caused the loss.4 However, where the loss is attributable to the wilful misconduct of the shipowner, not only the shipowner but also the mortgagee of the vessel and the cargo owner whose cargo sank together with the ship will be barred from recovery. This brings into question the purpose of s 55(2)(a) of the Marine Insurance Act 1906 (UK) (the MIA 1906), which provides: The insurer is not liable for any loss attributable to the wilful misconduct of the assured, but, unless the policy otherwise provides, he is liable for any loss proximately caused by a peril insured against, even though the loss would not have happened but for the misconduct or negligence of the master or crew.

Namely, does this section aim to prevent any recovery for any loss in which the shipowner’s wilful misconduct is involved, or to impose a personal bar to recovery for the shipowner only? As the courts’ view has been firm on disallowing recovery for an innocent co-assured mortgagee under the shipowner’s hull insurance where s 55(2)(a) of the MIA 1906 applies, the market has developed a mortgagees’ interest insurance (MII) as a way of further securing recovery for the loss of the vessel. An MII aims to protect mortgagees against the possibility of their security, the mortgaged ship, proving insufficient in the circumstances that are likely to 1 This chapter is an expanded version of a lunchtime seminar held at CML during the author’s research visit from 31 January to 25 February 2022. The author gratefully acknowledges the CML funding which supported this research visit. 2 Marine Insurance Act 1906 (UK), s 55(2)(a). 3 Institute Time Clauses (Hulls) 1/10/83, cl 6.2.5; Institute War and Strikes Clauses Hulls – Time (1/10/83), cl 1.2. 4 Shell International Petroleum Co Ltd v Gibbs [1983] 2 AC 375; Atlasnavios Navegacao Lda v Navigators Insurance Co Ltd [2018] Lloyd’s Rep IR 448.

87

88  Research handbook on marine insurance law be listed in the policy.5 Furthermore, in order to protect the cargo owner’s interest, cargo insurance policies include a clause allowing the assured to recover where the shipowner’s wilful misconduct is involved in the occurrence of the loss. The MII, and similarly cargo insurance policies, usually insert a clause which allows recovery for ‘any deliberate or fraudulent casting away of or damage to the Mortgaged Vessel’.6 Recovery under an MII will be closely linked with the insured shipowner’s entitlement to indemnity under the hull policy. The shipowner’s claim against the hull insurers may be rejected for a number of reasons other than deliberate casting away of the vessel. Recent authorities have raised doubts as to whether an express inclusion of deliberate casting away will suffice to protect the mortgagee’s interest, as desired, at the outset of the contract. This chapter proposes that attempting to find a solution through a contractual wording is unlikely to mitigate the unjust outcomes that were created by Samuel v Dumas.7 Construing a marine insurance policy is subject to a number of different principles and rules, both statutory and judicial, that can impede recovery for the mortgagee, despite an express clause as such. A more effective solution for an innocent mortgagee could be found by revisiting Samuel v Dumas,8 and allowing the co-assured to recover where it is not privy to the shipowner’s wilful misconduct.

2.

‘INCLUDED AND EXCLUDED LOSSES’

The title of s 55 of the MIA 1906 is ‘included and excluded losses’. Under s 55(1), the insurer is liable for any loss proximately caused by a peril insured against, and is not liable for losses that are not caused by perils insured against. Section 55(2) states that the insurer is not liable for any loss attributable to the wilful misconduct of the assured. Where, however, a loss is proximately caused by a peril insured against, the insurer is liable, even though the loss would not have happened but for the misconduct of the master or crew.9 The language the subsection adopted for ‘delay’ is different, as ‘the insurer on ship or goods is not liable for any loss proximately caused by delay’.10 Similarly, s 55(2)(c) excludes losses caused by rats or vermin, or for any injury to machinery not proximately caused by maritime perils. Some other exclusions, however, are named as types of loss under s 55(3): ‘the insurer is not liable for ordinary wear and tear, ordinary leakage and breakage, inherent vice or nature of the subject-matter insured.’ In all the cases listed above, the parties are free to agree otherwise. Analysing s 55 as a whole is outside of the scope of this chapter. Understanding the proximate cause of the loss is essential to any evaluation of insured or excluded risks, and highlighting the above list is necessary to point out the variations of the wording of several different exclusions under this section. Fundamentals of the proximate cause will be mentioned below, before moving on to the proof of wilful misconduct of the assured. 5 Schiffshypothekenbank Zu Luebeck AG v Norman Philip Compton (The Alexion Hope) [1988] 1 Lloyd’s Rep 311, 313. 6 Institute Mortgagees Interest Clauses – Hulls 1997, cl 2.1.4. Shell International Petroleum Co Ltd v Gibbs (n 4). 7 Samuel v Dumas [1924] AC 431. 8 Ibid. 9 MIA 1906, s 55(2)(a). 10 Emphasis added.

Scuttling, fortuity and marine perils  89

3.

PROXIMATE CAUSE OF THE LOSS

Section 55 of the MIA 1906 does not define the phrase ‘proximate cause’. Numerous cases11 discussed its meaning, until it was eventually settled that it refers to the ‘efficient’ cause of the loss.12 The word proximate can also be used interchangeably with determining,13 predominant14 or real causes,15 as opposed to remote,16 indirect17 or distant18 causes. The question of ‘what caused the loss’ is essential to any insurance claim, given that the insurer will not be liable for any loss regardless of what brought it about. While the proximate cause criterion sets the legal standard for the relevant burden of proof, it is ultimately a matter of judgment of the facts and evidence to determine what was the efficient cause of the loss. Hence, in the words of Lord Roskill, questions of causation can give rise to problems both of law and of fact, and opinions on them may, and often do, differ.19 The proximate cause of loss is determined by first formulating the problem, then identifying the applicable legal principle(s) and point of construction, and finally determining the cause as a matter of fact.20 As noted by Erle CJ in Ionides v The Universal Marine Insurance Co:21 ‘the relation of cause and effect is a matter which cannot always be actually ascertained: but, if in the ordinary course of events a certain result usually follows from a given cause, the immediate relation of the one to the other may be considered to be established.’ For the insurer to argue the assured’s wilful misconduct under s 55(2)(a), the insurer does not have to prove that the relevant act of the assured was the proximate cause of the loss. This is due to the choice of language in the subsection: instead of ‘caused by’, the words ‘attributable to’ precede the exclusion. While terminology is not always conclusive,22 it has authoritatively been held that the words ‘attributable to’ refer to a different level of causal connection to that of required under the proximate cause from ‘proximate cause’.

11 Thompson v Hopper (1858) El Bl & El 1038; Lawrence v Aberdein (1821) 5 B & Ald 107; Dudgeon v Pembroke (1877) 2 App Cas 284; Pink v Fleming (1890) 25 QBD 396. 12 Reischer v Borwick [1894] 2 QB 548, 550; Leyland Shipping Co Ltd v Norwich Union Fire Insurance Society Ltd [1918] AC 350. 13 Larrinaga Steamship Co Ltd v The King [1945] AC 246, 253 (Viscount Simon LC). 14 Yorkshire Dale Steamship Co Ltd v Minister of War Transport [1942] AC 691, 702 (Lord Macmillan). 15 ENE Kos 1 Ltd v Petroleo Brasileiro SA Petrobras (No 2) (The Kos) [2012] 2 AC 164 [37], [48] (Lord Mance). 16 Thompson v Hopper (n 11). 17 Lawrence v Aberdein (n 11) 110. 18 Ionides v The Universal Marine Insurance Co (1863) 14 CB NS 259, 289 (Willes J). 19 Shell International Petroleum Co Ltd v Gibbs (n 4) 392. 20 The Ann Stathatos (1949–50) 83 Ll L Rep 228, 236. 21 Ionides v The Universal Marine Insurance Co (n 18) 285. 22 The UK Supreme Court, in the context of the meaning of ‘as a result of’, ‘arising from’, and ‘in consequence of’, expressed the view that ‘it is rare for the test of causation to turn on such nuances’: Financial Conduct Authority v Arch Insurance (UK) Ltd [2021] Lloyd’s Rep IR 63 [162]. Previously, however, in Handelsbanken ASA v Dandridge (The Aliza Glacial) [2002] 2 Lloyd’s Rep 421 [60], [62], it was held that the words ‘arising from’ and ‘as a result of’ do no more than import the usual test of causation as between peril and exception, namely that of proximate cause. The words ‘thereby’, ‘proceed from’ and ‘owing to’ were treated similarly: Thompson v Hopper (n 11) 1043.

90  Research handbook on marine insurance law

4.

‘ATTRIBUTABLE TO’

These words also appear in s 10(2) of the Insurance Act 2015 (UK) (the IA 2015), the effect of which is that breach of warranty by an insured suspends the insurer’s liability under the insurance contract from the time of the breach, until such time as the breach is remedied. This assumes that the breach is remediable, and that it has been remedied by the assured. The result of this will be restoration of the insurance coverage by lifting the suspension. The section adds that if a loss occurs after the breach was remedied, the insurer will not be liable for it if the loss is attributable to something which happened when the cover was suspended. In other words, the insurer will have no liability for anything which occurs, or which is attributable to anything occurring, during the period of suspension. The Explanatory Notes to the IA 2015 state:23 ‘A direct causal link between the breach and the ultimate loss is not required. That is, the relevant test is not whether the non-compliance actually caused or contributed to the loss which has been suffered.’ The Notes add that the ‘attributable to something happening’ wording is intended to cater for the situation in which loss arises as a result of an event which occurred during the period of suspension, but is not actually suffered until after the breach has been remedied.24 The Law Commission most probably did not mean ‘caused by’ when they used the phrase ‘as a result of’ in this explanation. In Thompson v Hopper,25 Cockborn CJ’s reference to the words ‘attributable’ supports the above conclusion. In response to the allegation that the assured’s loss was attributable to unseaworthiness of the ship, Cockborn CJ said: ‘it is not necessary that the unseaworthiness should have been the proximate and immediate cause of the loss, provided it can be shewn to have been so connected with the loss as that it must necessarily have led to it.’ The words ‘must have led to it’ express the view that it must be identified clearly as having occurred prior to the loss, that it must have some connections with the loss, but that the connection does not have to be the proximate cause. While the proximate cause test eliminates a remote cause, a loss may be attributable to the latter. Hence, the concept of ‘attributable to’ is broader than ‘caused by’ which includes the proximate cause only. The choice of ‘attributable to’ instead of ‘caused by’ in s 55(2) is in order to oust the proximate cause principle and to allow the insurer to deny recovery for more remote acts which were committed with wilful misconduct of the assured.26 The conduct which caused the loss and the mental state of the person who committed that act are treated separately for the purpose of assessing the insurer’s liability. The relevant mental state could be ‘negligent’, ‘reckless’ or ‘deliberate’, and each of these elements is subject to different evaluations, whether they are the master’s, the crew’s or the assured shipowner’s.

Explanatory Notes [96]: www​.legislation​.gov​.uk/​ukpga/​2015/​4/​notes. Explanatory Notes [89]. 25 Thompson v Hopper (n 11) 1054. 26 Colinvaux’s Law of Insurance, 13th ed, [5-115] fn 289. 23 24

Scuttling, fortuity and marine perils  91

5.

NEGLIGENCE AND FORTUITY

A loss may be attributable to an error or defect in judgement, namely an act of carelessness or negligence in the ordinary navigation of the vessel. For instance, a vessel may have caught fire through the negligence of the master or crew, or may have been stranded in a river because the cargo was loaded carelessly. Such error or defect does not itself suffice to excuse the insurer from liability.27 In Dudgeon v Pembroke28 Lord Penzance said that ‘the assured has hitherto always been held protected from loss through the perils insured against, though that loss was brought about through the negligence of his captain or crew’. This is because negligence is not a cause for this purpose,29 but is a quality of the way the insured peril has occurred. It was held that where it is clear that the loss is immediately occasioned by perils of the sea, the cause of the loss is still perils of the sea, despite being brought about by negligent navigation.30 In The Xantho,31 Lord Herschell said: ‘I am unable to concur in the view that a disaster which happens from the fault of somebody can never be an accident or peril of the sea.’ In the context of grounding, it was held that where there was neither any suggestion nor any evidence that the grounding was deliberate, and the grounding was not the natural and inevitable result of the action of the wind and waves, it must have been fortuitous even if it was brought about by negligence.32 The key issue is the definition of fortuity under rule 7 of Schedule 1 to the MIA 1906 which states: ‘The term “perils of the seas” refers only to fortuitous accidents or casualties of the seas. It does not include the ordinary action of the winds and waves.’ The short definition of fortuity is ‘accidental loss’,33 and whether a loss was accidental or deliberate is ‘a straightforward factual issue’.34 What caused the loss is determined on the evidence in each case. The well-known definition of fortuity was provided by Lord Herschell in the Xantho: it is ‘an accident which might happen, not an event which must happen’. It therefore implies something unexpected ‘which could not be foreseen as one of the necessary incidents of the adventure’.35 Where cargo was damaged because seawater entered into the ship, what led the water into the vessel would be investigated. At a time when ships were made of wood, seawater entry led by a hole opened by rats was accidental, as it was ‘sea damage occurring at sea and nobody’s fault’.36 Where a cargo of rice

Sadler v Dixon (1841) 8 M & W 895, 899. Dudgeon v Pembroke (n 11) 297. 29 Yorkshire Dale Steamship Co Ltd v Minister of War Transport (n 14) 711 (Lord Wright); Venetico Marine SA v International General Insurance Co Ltd [2014] Lloyd’s Rep IR 243 [285]. 30 Trinder Anderson & Co v Thames & Mersey Marine Insurance Co [1898] 2 QB 114, 123 (AL Smith LJ); Sheean v Lloyds Names Munich Re Syndicate Ltd [2017] FCA 1340. 31 Thomas Wilson, Sons & Co v Owners of Cargo per the Xantho (1887) 12 App Cas 503, 511. 32 McKeever v Northernreef Insurance Co SA [2019] 2 Lloyd’s Rep 161 [46]. 33 National Justice Compania Naviera SA v Prudential Assurance Co Ltd (The Ikarian Reefer) (No 1) [1995] 1 Lloyd’s Rep 455, 459. 34 Ibid. 35 NE Neter & Co Ltd v Licenses & General Insurance Co Ltd (1944) 77 Ll L Rep 202; The Catharine Chalmers (1875) 32 LT (NS) 847. 36 Hamilton Fraser & Co v Pandorf & Co (1887) 12 App Cas 518, approving Lopes LJ at first instance: (1885) 16 QBD 629, 635. 27 28

92  Research handbook on marine insurance law was damaged because of sweat and moisture when the crew had to close ventilators to prevent sea water entry, the cause of the loss was held to be accidental.37 While negligence, whether that of the crew members or the shipowner assured, does not prevent the event from being accidental, wilful misconduct on the part of the assured, according to the authorities,38 takes the relevant event outside the scope of this definition. Section 55 of the MIA 1906 also refers to misconduct with regard to the master or crew but omits the word ‘wilful’ in this case. Subsection 2(a) states that the insurer will be liable for ‘any loss proximately caused by a peril insured against, even though the loss would not have happened but for the misconduct or negligence of the master or crew’. The words ‘but for’ here serve a purpose similar to the words ‘attributable to’, so that a link (as described by Cockborn CJ in Thompson v Hopper)39 between the misconduct of the master or crew and the loss suffices to satisfy the ‘but for’ test. The misconduct of the master or crew does not have to be the ‘direct cause’ of the loss. If a collision occurs, and if the master’s or crew’s misconduct contributed to the occurrence of the collision, the insurer will still be liable for the loss if collision is an insured peril. The MIA 1906 therefore looks at whether the loss was fortuitous from the assured’s point of view, and neither the assured shipowner’s nor the master and crew’s negligence prevents the loss from being fortuitous.40 Misconduct of the master and crew will have a similar effect. In a similar vein, the words ‘attributable to wilful misconduct’ is an abstract concept and cannot be a peril in its own right; it will merely be the reason for the occurrence of a peril.41 If the assured deliberately sets fire to his property, his loss is proximately caused by fire, and wilful misconduct operates as an implied exception to the coverage granted by the policy rather than as an uninsured proximate cause in its own right. 42 In exceptional circumstances where the assured brings himself within the test for insanity laid down in Daniel M’naghten’s Case, the assured’s state of mind may excuse deliberate destruction of the subject-matter by him.43 In the MIA 1906 s 55(2)(a) the word ‘wilful’ was added to ensure that pure negligence was not a defence to a claim.44

6.

THE MEANING OF ‘WILFUL’

Both deliberate and reckless acts can be wilful but not all ‘wilful’ acts are deliberate. While the natural meaning of wilful includes deliberate, wilful is capable of having a wider meaning, depending on the context.45 A reckless act could be wilful, but a ‘deliberate’ act does not

Canada Rice Mills Ltd v Union Marine & General Insurance Co Ltd (1940) 67 LI L Rep 549. Samuel v Dumas (n 7); Suez Fortune Investments Ltd, Piraeus Bank AE v Talbot Underwriting Ltd, The Brillante Virtuoso [2020] Lloyd’s Rep IR 1. 39 Thompson v Hopper (n 11) 1038. 40 Although the shipowner’s or their manager’s failure to exercise due diligence may bar recovery under some standard clauses. Institute Time Clauses (Hulls) 1/10/83, cl 6. 41 Colinvaux, 13th ed, [5-115]. 42 Colinvaux, [5-115]. 43 (1843) X Clark & Finnelly 200. 44 R Merkin, Marine Insurance: A Legal History, Vol 2, Edward Elgar Publishing, 2021, p.800. 45 [2021] UKSC 12 [53]. 37 38

Scuttling, fortuity and marine perils  93 include recklessness.46 The word ‘deliberate’ connotes ‘consciously performing an act intending its consequences’.47 It involves a different state of mind from recklessness.48 In Ronson International Ltd v Patrick,49 Tuckey LJ held that in an insurance policy which excludes cover for ‘claims and liabilities arising from any wilful, malicious or criminal acts’, the adjectives characterise the excluded acts and look to the quality of the act and the state of mind of the actor.50 For a wilful act it is not necessary to show that the relevant person intended to cause damage of the kind in question.51 Recklessness as to the consequences of their act is also sufficient in this context.52 If the assured is aware that what they are about to do risks damage of the kind which gives rise to the claim, or does not care whether there is such a risk, they will act recklessly if they go ahead and do it. 53 This approach focuses on the state of the assured’s mind when they do the act, rather than on its intended consequences.54 In other words, it is not necessary to prove the assured intended to cause damage of the kind in question. For instance, if the assured was unaware of the risk that their fire might burn down the ship, and there is nothing to show that they did not care whether it might have done so or not, they are not reckless.55 The UK Supreme Court agreed with the above proposition and held that recklessness – in the form of an intentional act that is carried out with no regard for the consequences – equally does not amount to wilful or deliberate conduct. In Burnett v International Insurance Co of Hanover Ltd,56 while he was in a bar in Aberdeen, under the influence of alcohol and cocaine, G was taken out of the bar by three door stewards. An altercation between G and the stewards caused one of the stewards, M, to apply a neck hold for three minutes on G with the help of the other stewards. The pressure applied by M, however, resulted in G’s death due to mechanical asphyxia. M was convicted of assaulting G. The trial judge accepted that M’s actions were badly executed, but not badly motivated. M’s employer was insured under a public liability insurance policy, s 14 of which excluded liability arising out of ‘deliberate acts of wilful default or neglect’ by the insured’s employee. The Supreme Court held that the use of the word ‘deliberate’ in the exclusion indicated that the employee’s act should be intended to cause the type of harm suffered by the victim. The word ‘accidental’ was to be considered from the perspective of the assured – that is, the employer – rather than that of the doorman.57 It was inherent in a public liability policy such as this that the assured would be covered for damages which it had to pay owing to its vicarious liability for its employees’ torts. It was not the act which gave rise to the injury, but the act of causing the injury, that must be deliberate. It followed that the deliberate act meant carrying out an act intending to cause injury. In Burnett the insurers failed to establish that M had intended to injure G.

Ibid [64]. Ibid [52]. 48 Ibid. 49 [2007] Lloyd’s Rep IR 85. 50 Ibid [13]. 51 Ibid [15]. 52 Ibid [14]. 53 Ibid. 54 Ibid. 55 Ronson v Patrick [2007] Lloyd’s Rep IR 85 [17]. 56 Burnett v International Insurance Co of Hanover Ltd [2021] UKSC 12. 57 Hawley v Luminar Leisure Ltd [2006] Lloyd’s Rep IR 307. 46 47

94  Research handbook on marine insurance law Whether deliberate or reckless, the assured shipowner’s wilful misconduct has been discussed most in the case of the scuttling or deliberate casting away of ships. The difficulties brought about by the scarcity of evidence of the assured’s wilful misconduct, on the facts, were nevertheless overcome in some cases by the insurers who persuaded the courts that the loss was caused deliberately with the connivance of the owner.

7.

SCUTTLING – STANDARD OF PROOF

For the shipowner to succeed, the evidence must establish that the event leading to the loss of the vessel was fortuitous. If the shipowner fails to discharge this burden, their claim will have to fail,58 even if the insurers have alleged but failed to prove that the loss was deliberate. The burden of proof of the assured’s wilful misconduct rests unequivocally on the insurer.59 The insurer must satisfy the court that what resulted in the loss was deliberately done with a view to causing it, and that this was done with the connivance of the shipowner. The burden of proof which rests upon the insurer is derived from the civil rather than the criminal standard, although some authorities have formulated the test as follows:60 ‘Do circumstances exist, individually, perhaps, not of decisive consequence, but in their cumulative effect establishing beyond reasonable doubt that the vessel was dishonestly stranded?’ The standard of proof is on a balance of probabilities,61 but is ‘commensurate with the gravity of the allegation made’.62 The degree or standard of proof required appears to be heavier than that which rests upon the shipowner.63 The mere existence of the possibility that, for instance, the fire or grounding was accidental does not mean that the insurer has not satisfied the burden of proof.64 Although Aikens J said in The Milasan65 that ‘effectively the standard of proof will fall not far short of the criminal standard’, in The Ikarian Reefer66 the Court found it unnecessary to pursue the question whether the burden of proof so described by reference to the balance of probabilities is different in practice from the criminal standard of ‘beyond reasonable doubt’.67 In The Grecia Express,68 Colman J concluded that it must be ‘highly improbable’ that the vessel was lost accidentally, and that there must be derived from the whole of the evidence ‘a high level of confidence that the allegation is true’. The facts proven against the owner must be ‘sufficiently unambiguous’ to establish that the owner was complicit in the casting away of their vessel.69 For the court a distinction must be drawn between remote or fanciful possibil-

The Popi M [1985] 2 Lloyd’s Rep 1; The Ikarian Reefer (n 32) 459. The Alexion Hope (n 5) 317. 60 Anghelatos v Northern Assurance Co Ltd, The Olympia (1924) 19 Ll L Rep 255, 257. 61 The Atlantic Confidence [2016] 2 Lloyd’s Rep 525 [139]. 62 The Ikarian Reefer (n 33) 459. 63 Ibid. 64 Ibid 483, 484. 65 The Milasan [2000] 2 Lloyd’s Rep 458 [28]. 66 The Ikarian Reefer (n 33). 67 For a case comment see G Gauci, ‘The Ikarian Reefer and the law of marine insurance’, J.B.L. 1995, Nov, 613–17. 68 The Grecia Express [2002] 2 Lloyd’s Rep 88. 69 The Milasan (n 65) [28]. 58 59

Scuttling, fortuity and marine perils  95 ities, unsupported by any evidence, and the kind of substantial, or substantiated, possibility.70 The burden of proof is not discharged if the evidence fails to exclude a substantial, as opposed to a fanciful or remote, possibility that the loss was accidental.71 The proof of a motive is not conclusive, but it can be relevant.72 Equally, the absence of motive will assist the shipowner to rebut an accusation of this nature.73 The courts will not disregard the character and antecedents of the shipowner.74 The overall financial position of the owner is also taken into account. 75

8.

WITNESS STATEMENTS

The courts acknowledge that if evidence is given through an interpreter, it is not easy to tell whether a witness is telling the truth. This is mostly because a fact-finding judge can gain little from the demeanour of a witness when the witness is foreign, comes from a different culture and does not give evidence in their first language.76 In cases where scuttling is alleged, the assessment of the reliability of a witness depends on a consideration of the extent to which their evidence is consistent with: (1) what is not in dispute; (2) what the witness has said on other occasions; and (3) the probabilities. If no real or substantial explanation can be put forward to explain an accidental loss of the vessel, that will, or may, have a bearing upon whether factual evidence that the loss was accidental is true. Their evidence must be tested in the light of the probabilities and the evidence as a whole.77 The credibility of the evidence will become doubtful where the shipowner’s explanation requires a series of steps to happen in sequence, each of which is improbable or highly improbable. This is also the case especially if some or all the steps have to take place within a tight timescale and involve one or more remarkable coincidences.78 While the improbable can happen, it is difficult to accept that a number of improbable events may have occurred in rapid succession to each other. As Greer J pointed out in The Ioanna:79 ‘One improbability would not be sufficient to justify one coming to the conclusion that the event did not happen. But when there are two improbabilities the likelihood of it happening is still more remote, and when there are three it is more remote still.’ In The Atlantic Confidence,80 the vessel sank in deep water. The wreck had not been inspected with a view to determining the cause of the fire or the cause of the sinking. The available evidence was limited to surveys of the vessel prior to the final voyage, the crew’s statements, and photographs of the vessel taken after the vessel had been abandoned and

The Ikarian Reefer (n 33) 459. The Popi M (n 58); The Ikarian Reefer (n 33) 459; The Atlantic Confidence (n 56). 72 The Ikarian Reefer (n 33) 483, 498. 73 Ibid. 74 The Olympia (n 55) 257; The Ikarian Reefer (n 32) 498; The Atlantic Confidence (n 61); The Brillante Virtuoso (n 38). 75 The Ikarian Reefer (n 33) 498. 76 The Atlantic Confidence (n 61). 77 The Ikarian Reefer (n 33) 483, 484. 78 Ibid. 79 The Ioanna (1922) 12 Ll L Rep 54. 80 The Atlantic Confidence (n 61). 70 71

96  Research handbook on marine insurance law before it sank. What was argued as accidental was described by the judge as ‘cumulative suspicions’, as the judge was not persuaded that fire, then flooding of the engine room caused by the fire, and then flooding of two double bottom tanks on the portside caused by the fire, were all accidental.81

9.

SCUTTLING – INSURANCE CLAIMS

In Thompson v Hopper,82 the majority of the Court of Queen’s Bench held that it is a maxim of our insurance law, and of the insurance law of all commercial nations, that the assured cannot seek indemnity for a loss produced by their own wrongful act. Lord Ellenborough CJ in Cullen v Butler83 approved a quote that ‘I cannot effectually (valablement) contract with any one that he shall charge himself with the faults which I shall commit’. Section 55(2)(a) of the MIA 1906 is a reflection of these authorities. However, as the post-MIA 1906 cases have demonstrated, it is not only shipowners themselves whose claims are invalidated by the shipowner assured’s wilful misconduct. The innocent cargo owner whose cargo sank together with the ship, and the innocent mortgagee who lost their security for the loan granted to the shipowner, will also be disadvantaged by the way that the loss occurred. The insurance market has been trying to find a market-based solution to this problem through contractual wording, as mentioned in the introduction to this chapter. However, the possibility for a mortgagee making a successful claim under an MII, in the relevant circumstances, still appears to be very slim.

10.

INNOCENT CO-ASSURED

An innocent co-assured is not permitted to recover from the hull insurer, even though their interest is not joint with the shipowner, and they were not privy to the scuttling of the ship. Where a mortgagee claims against the hull underwriter as a co-assured, in principle, their position is different from those who claim as a loss payee or assignee of the insurance recovery.84 The mortgagee may participate in the shipowner’s own policy as a loss payee, as an assignee or as a co-assured. The mortgagee does not have to select one of these options: some or all of them may be available to them to protect their interest as a mortgagee.85 The mortgagee, if described as a loss payee in the shipowner’s policy, will be entitled to enforce the insurer’s promise to pay to them directly.86 Alternatively, the shipowner’s entitlement to claim under the insurance policy may be assigned to the mortgagee under s 50 of the MIA 1906, or under s 136(1) of the Law of Property Act 1925 (UK), or in equity. In all four cases, however, the assignee’s title is derivative, so the insurer will be allowed to argue the Ibid [296]–[300]. Thompson v Hopper (1856) 6 El & Bl 172, 191–2; approved by the Exchequer Chamber: Thompson v Hopper (n 11) 1038. 83 Cullen v Butler (1816) 5 M & S 461. 84 D Smith, ‘Using insurance to protect lender’s interest’, J.I.B.L. 1991, 6(3), 107–15. 85 Peter Macdonald-Eggers QC, ‘Mortgagees’ Interest Insurance’ in Baris Soyer and Andrew Tettenborn (eds), Ship Building, Sale and Finance (Informa Law from Routledge 2015), Ch 11, 172. 86 The Contracts Rights of Third Parties Act 1999 (UK), s 1, permits the mortgagee to do so. 81 82

Scuttling, fortuity and marine perils  97 defences that would be available if the claim was made by the shipowner. It follows that where the shipowner assured scuttled the insured vessel, the insurer’s position is the same towards the shipowner, loss payee or assignee under the insurance contract. The mortgagee of a vessel being purchased or under construction may participate in the insuring of the vessel to protect their interest in it. This is confirmed by ss 14(1) and 14(2) of the MIA 1906, which provide respectively that ‘[w]here the subject-matter insured is mortgaged […] the mortgagee has an insurable interest in respect of any sum due or to become due under the mortgage’; and ‘[a] mortgagee may insure on behalf and for the benefit of other persons interested as well as for his own benefit’. Further, under s 10 of the MIA 1906, ‘[t]he lender of money on bottomry or respondentia has an insurable interest in respect of the loan’. Where the mortgagee is a co-assured together with the shipowner, the interest of each party is composite,87 that is, separate, and if one party’s claim is invalidated, for instance because of a breach of contract, the other party’s entitlement remains intact. This is different to joint insurance, where co-assureds’ interests are inseparably connected, so that a loss or gain necessarily affects them both, and the misconduct of one co-assured is sufficient to contaminate the whole insurance. Despite such a clear distinction being drawn between the interests of the shipowner and the mortgagee, and the latter’s innocence in the way that the loss has occurred, the courts have strictly rejected the proposition that the co-assured mortgagee can successfully claim against the insurer for the loss of the vessel, being the security for the loan which they granted to the shipowner. It is not clear whether it is a matter of principle of general marine insurance law or a matter of policy wording that the co-assured mortgagee’s claim is not allowed judicially even though the mortgagee is not privy to the shipowner’s conduct. Viscount Finlay said in Samuel v Dumas:88 ‘The possibility of scuttling is not a peril of the sea; it is a peril of the wickedness of man, and would have to be mentioned expressly in the policy, like barratry or pirates, in order that the assured should recover from the underwriter in respect of it.’ Where the insured vessel is damaged by the deliberate wicked action of another, that does not prevent recovery from the insurer. The loss in such cases may have been caused by barratry or malicious act of a third party towards the assured shipowner. Where the claimant is the innocent co-assured mortgagee under the hull insurance policy, the position is similar, in that the shipowner’s wicked act is not imputed to the innocent co-assured. Bramwell B stated in Thompson v Hopper:89 There is nothing wrongful in sending an unseaworthy ship to sea; though she is insured, there is nothing wrongful in burning her. The wrong is in making a claim founded on such an act […] The act does not become wrongful where a claim is founded on it and its consequences; but the claim is […] ‘Dolus circuitu non purgatur’ means, You cannot fraudulently do that indirectly which you cannot do directly: and I agree that if a man sent his ship to sea with a false compass, in order that she might be lost, and she was lost in consequence, he could not recover.

87 Jonathan Gilman and others (eds), Arnould’s Law of Marine Insurance and Average (20th edn, Sweet & Maxwell Ltd 2021) para 8-13. 88 Samuel v Dumas (n 7) 459. 89 Thompson v Hopper (n 11) 1038, 1045–1046.

98  Research handbook on marine insurance law While this describes the shipowner’s situation, it is arguable that in a composite policy the position of the mortgagee who is not privy to the owner’s wilful misconduct should be separated from that of the owner.

11.

A DOUBLE-EDGED SWORD

The critical issues in the evaluation of the innocent co-assured’s claim against the vessel’s hull (or war risks) insurers are: first, the co-insurance policy is composite, meaning that, in principle, one co-assured’s loss of the claim by their own failure to meet the contractual requirements does not prevent the other co-assured from claiming from the insurer. Second, the shipowner’s wilful misconduct is not imputed to the mortgagee co-assured who is not privy to the shipowner’s misconduct. Third, the innocent co-assured, in principle, should be allowed to claim against the insurer if the loss falls under the insurance cover. Whether the loss is covered by the policy is determined if the cause of the loss is one of the insured perils. Fourth, and crucially, marine insurance contracts insure against fortuity, meaning accidental losses only. On the other hand, the wilful misconduct of the assured, according to the common law,90 takes away the possibility of the loss having occurred as a result of an accident. In Fenton (Pauper) v J. Thorley & Co Ltd91 it was held that in its ordinary sense, an ‘accident’ is a mishap or untoward event not expected or designed. In reference to Fenton, Colinvaux states that if the consequences were intended by the assured, or if the consequences, while unintended, were inevitable so that the assured can be regarded as having acted with reckless disregard for them, then it is clear from the authorities that there is no accident and the assured is precluded from recovery.92 In such cases the principle is that the assured intends to run the risk involved, and hence any resultant injury is not caused by a mishap which is unexpected, nor is the injury itself unexpected so as to be accidental.93 It follows that in the case of loss of a ship the possibility of scuttling or deliberate casting away inevitably excludes a finding of perils of the seas. In other words, the ability of the co-assured to claim against the insurer evaporates upon proof that the loss was not caused by a peril insured against. As examined above, for the insurer to rely on the assured shipowner’s wilful misconduct under s 55(2)(a) of the MIA 1906, it suffices to prove that the loss was attributable to it.94 In Samuel v Dumas,95 the ship Grigorios foundered in calm weather off the coast of Spain and became a total loss. Bailhache J found that the vessel was scuttled with the connivance of the owner.96 The judge was satisfied that this was done by deliberately letting water into the ship and into the bilge connections, and afterwards causing a sham explosion, which induced the innocent members of the crew to leave the ship together with those who were guilty and to Samuel v Dumas (n 7). [1903] A.C. 443. 92 Colinvaux [5-119]. 93 Colinvaux [5-119]. 94 An analogy can be drawn with the ruling in The Cendor Mopu [2011] 1 Lloyd’s Rep 560, where the definition of inherent vice excluded perils of the seas and vice versa. In other words, once inherent vice is found to have caused a loss, there is no possibility of finding perils of the sea for the loss in question. 95 Samuel v Dumas (n 7). 96 Samuel v Dumas (1922) 12 Ll L Rep 73. 90 91

Scuttling, fortuity and marine perils  99 refrain from using the pumps. The assured’s recovery depended on whether their loss was due to a risk covered by the policy. The policy did not include scuttling among the perils insured against. Bailhache J held that although the incursion of seawater into the ship would undoubtedly be a peril of the sea, the co-assured mortgagee could nonetheless not recover, owing to the fact that scuttling was a risk not covered by the policy. In the House of Lords, Viscount Cave said:97 ‘There appears to me to be something absurd in saying that, when a ship is scuttled by her crew, her loss is not caused by the act of scuttling, but by the incursion of water which results from it.’ While the scuttling, the subsequent entry of the sea water, the slow filling up of the hold and bilges, the failure of the pumps and the break-up of the vessel are as much parts of the effect as is the final disappearance of the ship below the waves, the efficient cause was the scuttling.98 Viscount Cave said that the maxim dolus circuitu non purgatur applied only as against persons who were parties to the dolus. However, the judge found it hardly necessary to have recourse to the maxim. The answer lay in the fact that the loss was not fortuitous; the words ‘perils of the sea’ could not extend to a wilful and deliberate throwing away of a ship by those in charge of it. In Small v United Kingdom Marine Mutual Insurance Association99 the insured ship was wilfully cast away by the captain. The captain owned some shares in the ship. He purchased the shares with a loan from Small, whose interest was insured, as well as the owners. The Court decided that Small could recover from the insurers, who had rejected his claim on the basis that the loss was not caused by perils of the sea. The captain’s wilful casting away of the ship, in the absence of Small’s connivance, amounted to a barratry. Moreover, Small was not involved in the appointment of the captain; the captain was a stranger to Small; and a stranger’s wrongdoing to the ship as a result of which the ship sank was a peril of the sea. The majority of the House of Lords in Samuel v Dumas partially overruled the Small case, which now stands on the ground of the barratry alone.100 The words ‘and of all other perils, etc.’ may be added to the insured risks to broaden the coverage beyond perils of the seas. In Samuel v Dumas, the policy included such wording. However, this was still not sufficient to allow the mortgagee’s insurance claim, as these words were to be construed as applying to perils of the same kind as those which have been previously specified.101 Viscount Finlay also rejected the possibility of a separation between the entrance of the sea water and the act which caused it.102 The judge argued that the view that the proximate cause of the loss when the vessel has been scuttled was the inrush of the sea water, and that this was a peril of the sea, was inconsistent with the well-established rule that it was always open to the underwriter on a time policy to show that the loss arose, not from perils of the seas, but from the unseaworthy condition in which the vessel sailed. With respect, s 39(5) of the MIA 1906 requires proof that the ship was sent to sea in an unseaworthy state, and that the loss is Samuel v Dumas (n 7) 446–7. Samuel v Dumas (n 7) 448, referring to Reischer v Borwick [1894] 2 QB 548, 550; Leyland Shipping Co Ltd v Norwich Union Fire Insurance Society Ltd [1918] AC 350. 99 Small v United Kingdom Marine Mutual Insurance Association [1897] 2 QB 311. 100 Samuel v Dumas (n 7) 449. 101 Thames & Mersey Marine Insurance Co Ltd v Hamilton Fraser & Co, The Xantho (1887) 12 App Cas 484. 102 Samuel v Dumas (n 7) 453–4. 97 98

100  Research handbook on marine insurance law attributable to such unseaworthiness. The test is not the proximate cause test, and it is open to the insurer to prove what s 39(5) looks for if the insurer denies liability under the time policy. It is therefore not easy to see how a separation between the entrance of the sea water and the act which caused it results in the consequence about which Viscount Finlay was concerned. Importantly, the members of the House of Lords in Samuel v Dumas referred to Trinder, Anderson & Co v Thames & Mersey Marine Insurance Co,103 where Collins LJ said: The wilful default of the owner inducing the loss will debar him from suing on the policy in respect of it on two grounds […] first, because no one can take advantage of his own wrong, using the word in its true sense which does not embrace mere negligence […] secondly, because the wilful act takes from the catastrophe the accidental character which is essential to constitute a peril of the sea.

Viscount Finlay acknowledged that the first ground would be applicable to the mortgagee only if the policy was joint, namely, the shipowner’s and mortgagee’s interests were inseparable. However, the judge found it unnecessary to decide whether the mortgagee in Samuel v Dumas could be so identified. For Viscount Finlay, the overwhelming consideration was that there were no perils of the sea on the facts. Lord Sumner dissented. His Lordship read Bailhache J’s judgment as being consistent with the proposition that ‘the guilty owner fails to recover on the policy because he is guilty and not because the loss is not otherwise a loss by perils insured against’. Lord Sumner expressed the view that: Contracts of indemnity are intended to make good losses where they happen in certain events, and except where, as with barratry, culpability is a quality of the cause of loss itself, they are not concerned with the guilt or innocence of the action. Loss attributable to the wilful misconduct of the assured is not excluded by any actual term of the contract. There is a restriction placed by law on the right to recover upon it, in order that a contract of indemnity may not serve as an instrument of fraud.

This accords with what was held by Bramwell B in Thompson v Hopper,104 as discussed above. A fortuitous casualty is a matter of chance. Lord Sumner said that when a ship had been holed below the waterline, if nothing was done to close it, the ship would probably eventually sink. The ship’s ultimate fate was a matter of the intervention of something to stop the inflow before the point of sinking was reached. Whether the hole was made by negligence or by crime, by the impact of heavy cargo slipping from the slings or by contact with floating submerged wreckage would not make any difference. Consequently, his Lordship stated: ‘I do not see how it can be affirmed that the ship did not go to the bottom by getting too full of water, whether the owner let the water in at the beginning or not.’ A further matter on which Lord Sumner relied was the personal disability of the shipowner for recovery under insurance due to their wilful misconduct. The assured not being allowed to take advantage of their own wrong is a different matter from what caused the loss. Importantly, Lord Sumner described that the former maxim refers to something which would, in itself, be a matter of right under the contract being denied to a party, because the law is more moral than the contract. For his Lordship, the exclusion of the assured from recovering on the policy for a loss attributable to their wilful misconduct was a separate matter from the fortuitous

Trinder, Anderson & Co v Thames & Mersey Marine Insurance Co [1898] 2 QB 114, 127–8. Thompson v Hopper (n 11) 1038, 1045–6.

103 104

Scuttling, fortuity and marine perils  101 character of perils of the sea. The judge stated that if proximate cause did not apply, it must be because the policy was subjected to an exceptional rule of law to prevent the assured from profiting from their own misconduct, but no further. If wilful misconduct removes the accidental nature of the case, in principle, the third party’s wilful misconduct should also fall outside of the definition of perils of the sea. In Gordon v Rimmington,105 the ship Reliance was set on fire by its captain and crew while on a voyage from Africa to the West Indian islands. The fire was set in order to prevent the ship being captured by French privateers who were chasing it. The crew left the ship in a long boat. Lord Ellenborough held that if the ship was destroyed by fire, it was of no consequence whether this was occasioned by a common accident, or by lightning, or by an act done in duty to the state. Nor could it make any difference whether the ship was thus destroyed by third persons, by subjects of the King or by the captain and crew acting with loyalty and good faith. Fire was still the proximate cause, and the loss was covered by the policy. Moreover, in The Midland Insurance Co v Smith and Wife,106 loss by perils insured against, although brought about by the felonious act of a third party, were held to be a loss within a fire policy. It was suggested107 that in Gordon: The rule is directed against misconduct only, and does not cover cases where the assured, properly and without any fraud on his part, causes the loss in question, as where he causes the loss in the performance of a duty to the state under whose law the policy was made. Thus, where the captain of a British vessel set fire to it to prevent it falling into the hands of French privateers, it was held that the loss fell within the fire cover provided by the policy and that recovery was not precluded.

The courts subsequently confirmed the majority opinion in Samuel v Dumas, even though it is a double-edged sword for the innocent co-assured, as illustrated in The Brillante Virtuoso.108 In this case, the claim was made by the insurers of the MII against the vessel’s war-risk underwriter. The latter argued that the shipowner, with the assistance of the master and chief engineer, arranged for a fake attack by pirates, and for a fire to be deliberately started on board the vessel. The local salvor who attended the vessel was, allegedly, a party to the conspiracy. The judge found for the war-risk underwriter. There were several improbabilities which, when viewed collectively, cogently suggested that the supposed attack by pirates was a fake attack.109 Having recognised that the mortgagee was a co-assured under the war-risk insurance, the judge stated that the bank could claim against the insurers, despite the shipowner’s misconduct. However, this would be possible only if the mortgagee proved that the loss was caused by a peril insured against. The impossibility of satisfying such a burden of proof has been analysed above. Not surprisingly, on the facts, it was not possible for the mortgagee to prove that the loss was caused by a peril insured against. Given that the judge found that the pirate attack was fake, and that the shipowner was privy to it, it followed that neither a malicious act against the owner nor a pirate attack (given that the owner was not a pirate) had caused the loss.

107 108 109 105 106

Gordon v Rimmington (1807) 1 Camp 123. The Midland Insurance Co v Smith and Wife (1881) 6 QBD 561. Colinvaux, 5-117. The Brillante Virtuoso (n 38). Ibid [414].

102  Research handbook on marine insurance law

12.

CLAIMS UNDER MII

The Court of Appeal’s evaluation of the mortgagee’s claim in The Alexion Hope110 is worth noting,111 although the Court’s reasoning seemingly relies on the separation between perils of the sea and fire. The mortgagee purchased an MII if the vessel were to become a total or partial loss, and the mortgagee were to find themselves unable to recover from the hull underwriter. The MII did not spell out the circumstances in which the mortgagee might not be able to recover from the hull underwriter. Lloyd LJ assumed that they included cases where the hull underwriter declined liability on the ground of misrepresentation or non-disclosure, or because the vessel had been wilfully cast away with the connivance of the owners.112 The Alexion Hope was insured by the shipowner, and the benefit of the policy was assigned to the mortgagee. On 23 October 1982, a serious fire occurred in the engine room which led to the loss of the vessel. Unable to recover from the hull insurer, the mortgagee made a claim under the MII, but the claim was rejected on the ground that if there was a fire at all, the fire was not fortuitous, as it was caused by the wilful misconduct of the shipowner. The Court of Appeal held that ‘fire’ in a marine policy is not confined to an accidental or fortuitous fire.113 It includes a fire started deliberately by a stranger to the contract of insurance. If it included any deliberate fire, the consequence would be that if, in the case of an ordinary householder’s policy, the assured set their house on fire through carelessness, they would recover; but if their neighbour set their house on fire through ill-will, the assured would not recover. Lloyd LJ differentiated perils of the sea and fire, since perils of the sea are defined by rule 7 of Schedule 1 to the MIA 1906 as referring only ‘to fortuitous accidents or casualties of the seas’, whereas there is no such limitation in the case of fire.114 Nourse LJ said:115 the suggestion that a fire deliberately caused by, or with the connivance of, the owners and in respect of which hull underwriters have disclaimed liability is not covered by an insurance whose avowed purpose is to protect a mortgagee against the loss of his security is to my mind preposterous.

Significantly, according to Purchas LJ, it would produce an absurd result if the MII were to be interpreted narrowly. His Lordship said: ‘as between the mortgagee and the mortgagee’s interest insurer, it matters not whether the fire was started by an independent agent, or whether by or with the connivance of the shipowner, the master or the crew, or indeed whether it occurred fortuitously.’ However, as observed in Piraeus Bank AE v Antares Underwriting Ltd,116 the link between the MII policy’s insuring clause and the cover provided under the shipowner’s hull insurance is the most crucial. In the Piraeus Bank case the vessel’s war-risk insurer rejected liability on

The Alexion Hope (n 5). For case comments on the case see DG Powles, ‘Protecting the mortgagee’s interest’, J.B.L. 1988, Mar, 176–81; DG Powles, ‘Protecting the mortgagee’s interest even more (sea)’, J.B.L. 1988, Nov, 507–9. 112 Ibid 313. 113 Ibid 316. 114 Ibid. 115 Ibid 319. 116 Piraeus Bank AE v Antares Underwriting Ltd [2022] EWHC 1169 (Comm). 110 111

Scuttling, fortuity and marine perils  103 the ground that the cause of the ship’s detention, a criminal investigation in which the ship and the crew were involved, was excluded from the cover. Clause 1 of the MMI policy provided an indemnity for: loss of, or damage to, or liability arising in connection with the Vessel (i) Which is prima facie covered by the Owners’ Policies […] but in respect of which, there is subsequent non-payment […] resulting from any act or omission of any one or more of the Owners and/or Operators and/or Charterers and/or Managers of the Vessel concerned and/or their Servants and/or Agents or anyone else held responsible ([…] the ‘Relevant Parties’) including any breach of warranty or condition whether express or implied or any misrepresentation or non-disclosure or alleged non-disclosure of any fact or circumstances of any kind whatsoever […] or (ii) Which occurs by virtue of any alleged deliberate, negligent or accidental act or omission or any knowledge or privity of any of the Relevant Parties including the deliberate or negligent casting away or damaging of the Vessel or the Vessel being unseaworthy or inadequately equipped, manned or certified.

Calver J held that the purpose of the MII was to protect the mortgagee against the risk of non-payment under the shipowner’s policy.117 The coverages in clause 1 of the MII were typical of the types of cover that a mortgagee seeks under the MII in protecting themselves against the shipowner’s insurer denying liability by reason of the shipowner’s misconduct regarding the loss: non-disclosure of material facts; breach of the duty of utmost good faith; breach of warranty; and failure to prove that the loss was caused by an insured peril. Calver J held that the MII policy does not cover losses which would not have given rise to a loss covered by the shipowner’s policies, because, for example, there was no constructive total loss under the shipowner’s policies, or the loss was excluded thereunder. Accordingly, there was no prima facie cover for the loss under the war-risk policy for the purposes of the MII policy, as the detention of the vessel would not have been covered. Calver J held that the purpose of the MII was to provide a secondary insurance, and if the MII was held to cover losses that did not fall under the shipowner’s policy, the MII would be providing primary insurance in respect of loss of or damage to the vessel. Furthermore, Calver J rejected the view that The Alexion Hope laid down a general principle which applies to all MII cover regardless of its wording, to the effect that it would defeat the purpose of MII cover if the rights of recovery under a MII policy were subject to the exclusions in an underlying hull policy. In respect of the MII, the shipowner’s act or omission must be the cause of the non-payment for a loss that was covered by the shipowner’s insurance policy.118

13.

CARGO OWNERS

The ship, when scuttled, might be loaded with cargo. The cargo owner’s claim against their insurer will be met by the same issue: the cause of the loss was not accidental. In Shell International Petroleum Co Ltd v Gibbs,119 Shell lost about 200,000 tons of crude oil as a result of a gigantic fraud perpetrated for the express purpose of sending the crude oil to South Africa Ibid [251]. Ibid [265]. 119 Shell International Petroleum Co Ltd v Gibbs (n 4). 117 118

104  Research handbook on marine insurance law in defiance of the ban on the export of oil from the Gulf to that country. The perpetrators of this fraud chartered a ship to load the cargo from Kuwait, ostensibly to be carried to Italy. They instead took the cargo to South Africa, discharged most of it at Durban, and then scuttled the ship with the remainder of the cargo on board. Shell’s claim for the loss of the cargo at Durban was rejected as the occurrence of the loss did not match with any of the risks insured against. However, Shell was successful in the claim of the loss of the cargo scuttled with the vessel. Similar to Clause 1(ii) of the MII policy wording in the Piraeus Antares case, Clause 8 of the cargo insurance policy provided: ‘In the event of loss the assured’s right of recovery hereunder shall not be prejudiced by the fact that the loss may have been attributable to the wrongful act or misconduct of the shipowners or their servants, committed without the privity of the assured.’ It was common ground that this second sentence was added for the benefit of innocent mortgagees of ships and cargo owners to surmount the decision of the majority of the House of Lords in Samuel v Dumas. The Court had no doubt that clause 8 allowed Shell recovery of the cargo scuttled with the ship. The Court held – and the insurer conceded – that clause 8 rendered this risk a peril of the sea. It is arguable, as referred to above, that a party whose interest is not the same as the shipowner should be regarded as distinct from the shipowner in their claims against the insurers for the loss caused by the sinking of the ship. In the words of Viscount Cave in Graham Joint Stock Shipping Co Ltd v Merchants Marine Insurance Co Ltd (The Ioanna) (No 1),120 the mortgagee is independently insured. The same is arguably true for the innocent cargo owner. In Shell, clause 8 recognised separation of the interests involved, but it should still be possible to draw such a distinction in the absence of a clause such as clause 8. Similarly, in The Atlantic Confidence,121 the ship was loaded with cargo when it grounded. The cargo insurer compensated for the loss which the cargo owner suffered, and then claimed against the shipowner. It would be interesting to see what decision a court would reach in a case similar to Piraeus Bank AE v Antares Underwriting Ltd122 if the insurer persuaded the court that the vessel was scuttled with the connivance of the owner. The MII expressly insures instances where the hull insurer rejects liability for scuttling of the ship. Assume that the wording of the MII policy requires prima facie liability. Where the vessel is scuttled, the loss under the hull insurance would fall outside the insurance cover: it will not be possible to argue fortuity. If there is no fortuity, there will be no prima facie cover, or any cover at all. However, the MII expressly includes scuttling of the vessel. Following the approach adopted in Calver J’s judgment, the MII would be treated as covering primary rather than secondary insurance, whereas Calver J’s view in Piraeus Bank AE v Antares Underwriting Ltd appears to suggest that MII policies, by their nature, should provide a secondary cover. It is submitted that the Court of Appeal’s interpretation of the MII in The Alexion Hope is preferable in order to give effect to the aim of the MII, as desired by the parties at the outset of the contract.

[1924] AC 294. The Atlantic Confidence (n 56). 122 Piraeus Bank AE v Antares Underwriting Ltd (n 116). 120 121

Scuttling, fortuity and marine perils  105

14. SUBROGATION In The Atlantic Confidence, the cargo owner’s insurer indemnified the cargo owner whose interest sank together with the ship and brought a subrogated claim against the shipowner by arguing that the loss was caused by their wilful misconduct. Where the innocent mortgagee is indemnified by the loss of their security (the vessel), whether the mortgagee was a co-assured with the shipowner or the assured of a MII policy, the insurer should subrogate into the mortgagee’s right and claim the amount paid from the shipowner who scuttled the ship. If the loss had not occurred, in theory, the shipowner would have repaid the loan to the mortgagee. The mortgagee would be making a claim for loss of their security which was caused by the shipowner’s wilful misconduct. The mortgagee’s interest is clearly separated from that of the shipowner’s, as the insurance policy is composite. The wrongdoing of the shipowner would be the basis for the insurer’s subrogation claim. It is true that the shipowner is also insured under the insurance contract, and the assured cannot sue themselves. In a composite policy, the insurer’s subrogation against the shipowner for their wilful misconduct would not be regarded as the assured suing themselves. The insurer is not stepping into the shipowner’s shoes and trying to claim against the shipowner. Whether the insurer’s subrogation action is prevented where the person indemnified by the insurer, and the defendant of the insurer’s subrogation action, are both co-assureds has been discussed by the courts to a great extent.123 It is outside the scope of this chapter to discuss the authorities on this issue.124 It suffices here to say that the innocent mortgagee’s position under a composite policy is not different from the cargo owner’s position in respect of the shipowner. So far as the shipowner is concerned, neither the mortgagee nor the cargo owner connived in the shipowner’s wilful misconduct. Each suffers loss because of the shipowner’s wrongdoing. Each is insured for the loss of the ship and the cargo, respectively. Given that the cargo owner’s subrogation action against the shipowner is heard by the courts, it is available so far as the mortgagee’s claim is concerned, unless expressly waived by the insurers.

15. CONCLUSION It is submitted that the shipowner’s wilful misconduct is to be regarded as a personal bar for them to recover under the policy. If scuttling or deliberate casting away of the vessel without the privy of the innocent co-assured excludes any possibility that the perils of the seas might be present on the facts, it is fallacious to argue that it is open to the co-assured mortgagee to prove that the loss was caused by a peril insured against, so that it can make a claim under the policy.

123 Petrofina (UK) Ltd v Magnaload Ltd [1983] 2 Lloyd’s Rep 91; Tyco Fire & Integrated Solutions (UK) Ltd (formerly Wormald Ansul (UK) Ltd) v Rolls Royce Motor Cars Ltd (formerly Hireus Ltd) [2008] Lloyd’s Rep IR 617; Gard Marine & Energy Ltd v China National Chartering Co Ltd [2017] Lloyd’s Rep IR 291; FM Conway Ltd v Rugby Football Union v Clark Smith Partnership Ltd [2023] EWCA Civ 418. 124 The matter was discussed in E Blackburn and A Dinsmore, ‘Joint insurance issues in The Ocean Victory: the roads not taken’ [2018] LMCLQ 50–72; O Gurses, ‘Subrogation against a contractual beneficiary: a new limitation to insurers’ subrogation?’ [2017] JBL 557–75.

106  Research handbook on marine insurance law In the words of Bailhache J125 and Viscount Finlay126 in Samuel v Dumas, for the co-assured mortgagee to recover for this loss, scuttling could, and should, be included expressly in the shipowner’s policy among the perils insured against. This appears to be acceptable practice in MII and cargo policies. However, it is doubtful that hull insurers that insure the mortgagor and the mortgagee as co-assureds would agree to do so. Even if they did, would the public policy permit insurance of the shipowner’s wilful misconduct in a co-insurance policy? The unjust outcome of the majority decision for the innocent mortgagee co-assured in Samuel v Dumas could, and should, be authoritatively revisited.

125 126

Samuel v Dumas (n 88). Samuel v Dumas (n 7) 459.

PART III  

6. To what extent should marine cargo insurance be construed to include cover for financial loss? John Dunt

1. INTRODUCTION 1.1

The Issue and its Importance

It is ‘not unlawful to combine two types of insurance in a single policy’.1 On the other hand, a policy should not be construed so that ‘the cover is transformed and exposures wholly alien to the policy are introduced’.2 When a marine cargo insurance policy is extended to cover financial losses in circumstances where no cargo exists, this transforms the underwriting exposure in two ways. First, the exposure is unlimited because the practice in the cargo market is to limit liability by reference to the value of existing cargo.3 It follows that if there is no cargo there is no limit. Second, exposure to financial risks involves a different rating approach and policy terms as compared with risks of physical loss of or damage to cargo.4 There are two connected issues: first, the existence of cargo; second, even where cargo exists, whether the risk insured is physical loss of or damage to that cargo, or purely financial loss to the assured where the cargo is undamaged, for example a loss arising from a failure or default in payment due to the unsound financial position of a buyer or other party. The issue has become more acute in recent years with the expansion of cargo insurance to include increased exposure to storage risks for static cargoes where misappropriation due to fraud, including by the use of fraudulent documents where no cargo exists, has been of increasing importance for the cargo insurance market.5

1 Sir Geoffrey Vos MR in ABN Amro Bank NV v Royal & Sun Alliance Insurance plc [2021] EWHC 442 (Comm); [2021] Lloyd’s Rep IR 467; [2021] EWCA Civ 1789; [2022] Lloyd’s Rep IR 201 (CA) at [70] in concluding that this was what the parties had in fact done in the case in question. 2 Tomlinson J in Outokumpu Stainless Ltd v Axa Global Risks [2007] EWHC 255 (Comm); [2008] Lloyd’s Rep IR 147, at [26]. 3 Marine cargo insurances limit the insurers’ liability to cargo aboard vessels, or in store at a particular location; for example, a typical wording would read ‘LIMIT: UDS 1,000,000 any one vessel and/ or location’. 4 See ABN Amro Bank NV v Royal & Sun Alliance Insurance plc [2021] EWHC 442 (Comm); [2021] Lloyd’s Rep IR 467 at [217] where the differences in rating approach and policy terms in relation to marine cargo risks and trade credit risks are summarised. 5 See Joint Cargo Committee (‘JCC’) Memorandum JC2017/001 of 1 March 2017 (available at lmalloyds.com on the Clauses and Circulars page) which introduced Misappropriation Clauses designed to limit and control storage losses due to misappropriation of cargoes in store.

108

Should marine cargo insurance include cover for financial loss?  109 1.2

Traditional Cargo Insurance Cover for Financial Losses Where Cargo Exists

In order to examine whether a marine cargo policy extends beyond the normal cover for physical loss of or damage to cargo (which, in line with recent authority,6 will be referred to with the acronym ‘PLOD’), it is necessary to identify the circumstances in which traditional cargo insurance covers financial losses and expenses where cargo exists.7 Marine cargo insurance is underwritten on a voyage rather than a time basis, which gives rise to the first exception to PLOD, which is sometimes expressed by saying that the ‘voyage’ of the cargo is insured8 or that the ‘adventure’ is insured.9 The concept of loss of the adventure entitles the assured to claim a total loss where the cargo is undamaged and in the assured’s possession ‘safe and sound’ but the insured transit has been terminated and the cargo has not reached its insured destination.10 The second exception to PLOD relates to traditional marine expenses and liabilities in respect of general average and salvage,11 where liability can accrue even where the cargo is not itself lost or damaged, as where cargo in a container stowed on deck on the bows of a container ship contributes to general average and salvage of ship and cargo where there was a fire confined to the stern of the vessel that never directly endangered the insured cargo. In addition, sue and labour expenses are recoverable12 when incurred to prevent loss of or damage to cargo where an insured peril has begun to take effect13 and such expenses may even be recoverable in circumstances where the cargo, though in immediate danger from perils insured, is, as yet, undamaged.14 These examples of financial losses contemplate the existence of cargo as property insured. The insurance cover for financial losses attributable to ancillary liabilities and expenses connected with that existing property does not expose underwriters to unlimited liabilities or to risks unconnected with the storage and carriage of cargo.

6 ABN Amro Bank NV v Royal & Sun Alliance Insurance plc [2021] EWHC 442 (Comm); [2021] Lloyd’s Rep IR 467 at [41] and thereafter. 7 The traditional exceptions to PLOD are well recognised: see for example J Dunt, Marine Cargo Insurance, 2nd edn, Informa Law from Routledge, 2016 (hereafter ‘Dunt, Marine Cargo Insurance’) at 13.18 to 13.21. 8 Roux v Salvador (1836) 3 Bing (NC) 266; as explained by Roche J in Vacuum Oil Company v Union Insurance Society of Canton Ltd (1926) 24 Ll L Rep 188, at p.190. 9 British & Foreign Marine Insurance Co v Sanday [1916] 1 AC 650 (HL). 10 Ibid. The Institute Cargo Clauses (‘ICC’) recognise and regulate this principle, ICC(A) Cl. 12 (Forwarding Charges). 11 The ICC cover expenses and liabilities, in particular, liability for salvage and general average and for certain collision liabilities, ICC(A) Cl. 2 (salvage and general average) and ICC(A) Cl. 3 (‘Both to Blame Collision’ Clause). 12 The ICC provide cover for sue and labour expenses incurred to avert or minimise a loss under the Duty of Assured Clause ICC(A) Cl.16; see also the Marine Insurance Act 1906 (‘MIA 1906’) s.78. 13 Noble Resources v Greewood (The ‘Vasso’) [1993] 2 Lloyd’s Rep 309. 14 Integrated Container Service Inc v British Traders Insurance Co Ltd [1984] 1 Lloyd’s Rep 154 (CA). (Voyage terminated by the insolvency of the carrier and the goods exposed to physical loss or damage, either by being exposed to weather conditions, or to legal action as by, for example, sequestration for failure to pay warehouse dues for their storage.)

110  Research handbook on marine insurance law 1.3

Insolvency and Credit Risks

An all risks marine cargo insurance policy in the usual form covers loss of possession of the cargo proximately caused by wrongful conversion or misappropriation even where the underlying reason for the conversion is an insolvency. In the leading case of London & Provincial Leather Processes Ltd v Hudson,15 Goddard LJ held that, where the assured was wrongfully deprived of his leather goods as a result of a wrongful conversion, there was an all risks loss even though the conversion resulted from the insolvency of the third party to whom the assured had entrusted the goods. Cargo underwriters’ concern at this type of financial loss resulted in an exclusion of insolvency, which is now incorporated in the Institute Cargo Clauses, but it only applies where, in broad terms, the assured were aware that the operators of the carrying vessel could become insolvent.16 It has been said that underwriters ‘were curiously reluctant to accept that they were exposed to the financial viability of third parties’17 but cargo underwriters’ assessment at the time of the placing of a cargo risk is primarily concerned with such matters as the packing of the cargo18 and the age and classification status of the vessel which is to carry the cargo.19 It has been held more recently that sue and labour expenses are recoverable under an all risks policy where the expenses are incurred to avert or minimise a physical loss that would fall on underwriters, albeit that the underlying reason that the cargo was in peril was the insolvency of the carrier.20 On the other hand, marine cargo underwriters do not insure credit risks, that is to say, insurances where the risk depends primarily on the creditworthiness of the parties involved. Lloyd’s underwriters, in particular, were party to a voluntary agreement where they undertook not to write direct financial guarantee business, defined as ‘the financial default or insolvency of any party’,21 without the previous sanction of the Committee of Lloyd’s. The current practice is for Lloyd’s to authorise specialist credit risk syndicates to underwrite credit risk business22 and credit risk insurance, including trade credit insurance, is now a well-recognised class of business underwritten at Lloyd’s and in the London company market, but not by cargo underwriters.

[1939] 2 KB 74; (1939) 64 Ll. L. Rep. 352. In Glencore International AG v Alpina Insurance Company Limited [2003] EWHC 2792 (Comm); [2004] 1 Lloyd’s Rep 111, conversion by a bailee/ buyer, who later became insolvent, constituted a loss by misappropriation under an all risks policy: the court further held where the assured had later adopted sales made in ignorance of the misappropriation, this was an all risks loss of physical property. Although this did not amount to conversion, the loss by misappropriation was never made good and there was therefore a valid claim under an all risks storage policy, at [224]. 16 See further ICC(A) Cl. 4.6. 17 D.R. O’May and J. Hill, O’May on Marine Insurance, Sweet & Maxwell, 1993 (hereafter ‘O’May on Marine Insurance’) at p.200. 18 See ICC(A) Cl. 4.3 (insufficiency of packing exclusion). 19 Regulated by the Institute Classification Clause. 20 Integrated Container Service Inc v British Traders Insurance Co Ltd, Lloyd’s Rep 460; [1984] 1 Lloyd’s Rep 154 (CA), insurers held liable for expenses incurred to recover insured shipping containers abandoned worldwide by an insolvent carrier. 21 See the definition of financial guarantee risks approved by the Committee of Lloyd’s, 25 November 1936. 22 The Lloyd’s regulations for specialist credit risk insurers have been relaxed over the years, most recently in 2019. 15

Should marine cargo insurance include cover for financial loss?  111 1.4

The Nature of ‘Loss’ in Marine Insurance: Constructive Total Loss (CTL)

A constructive total loss (CTL) of cargo arises by law even though the cargo remains sound and the assured has not been permanently deprived of it. In this respect, the Marine Insurance Act 1906 section 60(2)(i)(a), as applied in Polurrian Steamship Company Limited v Young,23 provides that there is a CTL where the assured is deprived of the possession of their ship or goods by a peril insured against, and it is ‘unlikely’ that they can recover them within a reasonable time. The question of what constitutes a ‘reasonable time’ varies with the circumstances of each case and may be difficult for the assured to establish. The practice has therefore grown up in hull and machinery insurance of agreeing a period of 12 months for a constructive total loss.24 A similar approach to cargo is not standard practice, possibly because cargo underwriters recognise that a much shorter period is appropriate in cargo cases and few disputes arise on this issue,25 which depends on the characteristics of the cargo and its market,26 and may be quite a short period of only a few weeks or months. 1.5

Specialist Insurances Extending Beyond Risks of Physical Loss or Damage

Subject to appropriate underwriting terms and conditions to limit and control the risk, add-ons that go beyond PLOD are available in the London and international markets.27 These add-ons may extend the cover, or may permit claims to be adjusted in a way that does not require the normal proof of physical loss or damage required by the Marine Insurance Act 1906. The extensions reflect the requirements of particular trades, such as foodstuffs or pharmaceuticals, and the needs of particular manufacturers, such as producers of well-known branded products. An extension of cover is illustrated by the cover for Rejection Risks. Rejection Risks cover, which is taken out by food importers in an ever more regulated world, includes the arbitrary rejection, for alleged breach of import regulations, of sound goods by the government of the country of destination, to protect the domestic market. It does not require loss, damage or deterioration of the goods during the duration of the insurance.28 23 [1915] 1 KB 922 (CA). The case decided, inter alia, that recovery must be ‘unlikely’ and not merely uncertain, and that the assured must establish that it could not recover the goods within a ‘reasonable time’: at p.937. 24 See the Detainment Clause in the Institute War & Strikes Clauses Hulls which provides that where the assured has lost free use and disposal of a vessel for a continuous period of 12 months the assured shall be deemed to have been deprived of possession ‘without any likelihood of recovery’. For a detailed description of the genesis of the Detainment Clause, see O’May on Marine Insurance at pp.276–83. 25 It is to be noted that hull war risk underwriters eventually settled the claims on the ships trapped in the Suez Canal from 1967 to 1975 but the ‘Claims on cargo were paid early on’: O’May on Marine Insurance at p.278. 26 See Clothing Management Technology Ltd v Beazley Solutions Ltd [2012] EWHC 727 (QB); [2012] 1 Lloyd’s Rep. 571 (English clothing manufacturer unlikely to regain possession within a reasonable time of garments from unpaid factory workers in Morocco. In assessing what was a ‘reasonable time’ the court took into account the assured’s commitment to its customers, high street retailers, and that the goods were fashion garments with a short shelf life). 27 ABN Amro Bank NV v Royal & Sun Alliance Insurance plc [2021] EWHC 442 (Comm); [2021] Lloyd’s Rep IR 467; [2021] EWCA Civ 1789: [2022] Lloyd’s Rep IR 201 (CA), where the evidence was that ‘coverages which do not require physical loss or damage were generally available in the London marine cargo market in 2015/16 as an add-on to marine cargo and storage policies’: at [209]. 28 See Dunt, Marine Cargo Insurance at [10.62].

112  Research handbook on marine insurance law So far as claims are concerned, the markets provide Brands Clauses to the manufacturers of well-known branded products to protect loss of reputation entitling the assured to claim a total loss and destruction of the damaged product where there is neither an actual nor a constructive total loss. In particular trades, such as the pharmaceutical industry, the policy may be specifically extended by extension of the Control of Loss clause to cover fear of loss, where damage to part of a shipment of pharmaceuticals leads to a fear that other parts of the consignment may have been contaminated though there is no evidence of PLOD.29

2.

MARINE CARGO INSURANCE AND PROPERTY INSURANCE NORMALLY CONFINED TO PHYSICAL LOSS

2.1

The Earlier English Marine Cargo Cases on Physical Loss

The question of whether a marine cargo policy extends to financial loss, rather than actual physical loss of cargo, is not a new one. For example, in De Monchy v Phoenix Insurance Company of Hartford30 Lord Atkin, in holding that a cargo insurance policy was confined to ‘actual physical loss’ and therefore did not extend to notional leakage of a cargo of turpentine, commented that ‘underwriters were insuring against casualties but not of arithmetic’. Following De Monchy, the next case to consider the requirement for loss of physical cargo was Fuerst Day Lawson Limited v Orion Insurance Co Ltd.31 In that case, the assured in England purchased essential oil in drums from Indonesia, which they insured against all risks. On discharge the drums were found to contain water with slight traces of the essential perfumed oil on the surface to deceive customs officers, or other inspectors or surveyors.32 The policy contained the Institute Cargo Clauses warehouse-to-warehouse clause, under which the insurance attached from the time ‘the goods [left] the warehouse or place of storage at the place named in the policy for the commencement of the transit’.33 The assured were unable to establish on balance of probability that the goods left the warehouse so as to trigger ‘attachment’ of the insurance under the Institute Cargo Clauses.34 In Coven SpA v Hong Kong Chinese Insurance Co35 a cargo of broad beans shipped from China to Italy was insured against all risks under the Institute Commodity Trade Clauses (A). The cover was extended to include ‘shortage in weight but subject to an excess of one percent’. The quantity of beans shipped from China was the same as that which arrived in Italy.

See further K.S. Vishwanath, Insuring Cargoes: A Practical Guide to the Law and Practice, 2nd edn, Witherbys Publishing Group Ltd, 2023 (‘Vishwanath, Insuring Cargoes’) paras 16.2 to 16.4 of Chapter 16, ‘Widening of Cover Beyond the Physical’ (pp.471–9). 30 (1929) 34 Ll L Rep 201 (HL) at p.210. 31 [1980] 1 Lloyd’s Rep. 656. 32 It was a large-scale fraud with the deception based on oil floating on water, which was the technique later employed in the United States in the notorious ‘salad oil swindle’. 33 See Cl. 1 of the Institute Cargo Clauses (All Risks) 1/1/63, which were the usual Institute Cargo Clauses in use at that time (1976). 34 Under the ICC(A) 1/1/09 the insurance attaches under Cl. 8.1 ‘from the time when the subject-matter insured is first moved in the warehouse’, rather than when the goods leave the warehouse, but the point remains equally valid. 35 [1999] Lloyd’s Rep IR 565 (CA). 29

Should marine cargo insurance include cover for financial loss?  113 Although there was no loss of existing goods, there was an apparent shortage of more than 1 per cent due to a measurement error. The issue was whether the policy should be construed to cover a measurement error where there was no actual loss of existing goods. Following Fuerst Day, it was common ground that ‘all risks cover only applies to physical loss and damage’.36 The issue was whether the policy words ‘shortage in weight’ extended the marine cargo policy cover beyond physical loss to financial loss without actual loss of existing goods. Clarke LJ, with whom the Master of the Rolls, Lord Woolf MR and Henry LJ agreed, opened his analysis with a brief summary of the general rules of construction, saying:37 That provision [‘covering all risks of loss and damage to the subject matter insured, including shortage in weight’] must of course be construed in the context of the policy as a whole, which must in turn be considered in its factual matrix or against its surrounding circumstances. The policy is […] a typical marine cargo policy. It is entitled ‘Marine Cargo Policy’ and on its face states that the insurers agree to insure against loss, damage, liability or expense. It is thus essentially concerned with insuring physical loss of and/or damage to goods. It would be somewhat surprising if it insured a measurement error such as occurred here.38

Later, in giving more detailed reasons for this approach, Clarke LJ said:39 The parties cannot have intended to insure goods which never existed, but that is the effect of [the submission of counsel for the assured] if correct. If parties to an insurance contract wish to insure against measurement error in circumstances such as this, where the goods never existed, they would, as I see it, have to do so in clear terms. They did not do so here. On the contrary, for the reasons I have tried to give the policy insured against physical loss, including shortage in weight, of goods which existed. A further point which militates against [the assured’s] claim is that it is difficult, if not impossible, to see how the insured could have an insurable interest in the subject matter of the insurance within the meaning of clause 11.1 of the Institute Clauses in respect of the […] beans which never existed. [Counsel for the assured] submits that the effect of the words added to the schedule is to disapply clause 11 of the Institute Clauses in this case. I see no warrant for that conclusion. However if the views which I have expressed thus far are wrong, as already stated, I would uphold the decision of the judge substantially for the principal reason which he gave, namely that, on his findings of fact, the same quantity or weight of goods left the warehouse, or place of storage, as were delivered to the receivers. By clause 8.1 of the Institute Clauses the insurance attached when the goods left the warehouse or place of storage and terminated on delivery. It follows that no shortage in weight occurred while the insurance attached and that [the assured’s] claim must fail.

Clarke LJ also noted that this conclusion was supported by, or at least consistent with, the cases in which it has been held that cover does not attach to goods which never commenced the insured journey.40

At p.568, per Clarke LJ. At p.568. 38 Special cover is now available for measurement errors, which may include non-physical paper losses; see, for example, the Institute Bulk Oil clauses/All risks + Guaranteed Outturn 01/08/2019 and see further Vishwanath, Insuring Cargoes at paras 9.11 and 9.12 (at pp.265–8) which discuss the Bulk Oil Clauses and para 16.6 (at pp.483–90) which discusses paper shortage adjustments under Guaranteed Outturn cover. 39 At p.569. 40 Simon Israel & Co v Sedgwick [1893] 1 QB 303 (CA) and Kallis (Manufacturers) Limited v Success Insurance Limited [1985] 2 Lloyd’s Rep 8 (PC). 36 37

114  Research handbook on marine insurance law The judgment gives a clear rationale for construing marine cargo policies, particularly those based on the Institute Cargo Clauses, as limited to physical loss or damage. First, any extension of the cover to non-existent goods would have to be expressed in ‘clear terms’. Second, there would be no insurable interest as required by Clause 11 of the Institute Cargo Clauses, which require that in order to recover under the insurance the assured ‘must have an insurable interest at the time of the loss’ (the ‘Insurable Interest’ issue). Third, as we have already seen in Fuerst Day Lawson Limited v Orion, non-existent goods could never be covered as they could never ‘leave’ the warehouse in accordance with Clause 8 of the Institute Cargo Clauses, the Transit Clause, and cover could never attach (the ‘Attachment issue’). In addition, the conclusion is supported by the parallel reasoning in the change of voyage cases, where the insurance does not attach if the cargo never leaves on the insured voyage, even if it leaves the warehouse on some other voyage. It is submitted here that the rationale is not based merely on the requirement for ‘clear terms’, important as that is, but also upon the Insurable Interest issue and the Attachment issue, as well as the parallel change of voyage cases. 2.2

‘Damage’ Requires Physical Damage to Goods

The need for the ‘physical’ existence of goods is also present in the cases on the meaning of ‘damage’ where goods are insured against ‘all risks of loss or damage’. The cases require ‘physical damage’ not economic loss,41 and ‘physical damage’ has been narrowly construed to require, at the very least, sub-molecular damage in circumstances where the damage is not visible.42 2.3

The Approach to ‘Physical Loss’ in Related Fields of Insurance Law

The principle that a policy is limited to physical loss or damage, in the absence of clear words to the contrary, extends beyond marine cargo cases to other types of insurance cover, e.g. product liability and builders’ risks, where the courts are reluctant to find in favour of the assured in the absence of physical damage.43 The same approach has been adopted in material damage and business interruption cases. In particular in Outokumpu Stainless Ltd v Axa Global Risks (UK) Ltd44 the court held that a property damage and business interruption policy ‘in general […] responded only to direct physical loss of or damage to property’.45 With regard to the particular clause in issue, Tomlinson J said:46 41 Cator & Others v The Great Western Insurance Co of New York (1873) LR 8 CP 552; Lysaght Ltd v Coleman [1895] 1 QB 49 and Overseas Commodities Ltd v Style [1958] 1 Lloyd’s Rep. 546 (all cases where loss in value due to suspicion of damage was held not to constitute ‘damage’). 42 Quorum v Schramm [2002] 1 Lloyd’s Rep. 249, at p.264. 43 The product liability cases, Bacardi v THP and Ors [2002] 1 Lloyd’s Rep. 62 and Rodan v Commercial Union [1999] Lloyd’s Rep IR 495, take a narrow view of damage, as do the builders risk insurance cases, Cementation Piling and Foundations v Aegon Insurance Co Limited [1993] 1 Lloyd’s Rep. 526; The Nukila [1997] 2 Lloyd’s Rep 146 and Shell v CLM Engineering [2002] 1 Lloyd’s Rep. 612. 44 [2007] EWHC 2555 (Comm); [2008] Lloyd’s Rep IR 147. 45 At [20], cited in a marine cargo policy case, Engelhart CTP (US) LLC v Lloyd’s Syndicate 1221 [2018] EWHC 900 (Comm); [2018] 2 Lloyd’s Rep 24 at [40] for the proposition that the same approach applied to marine cargo insurance. 46 At [26].

Should marine cargo insurance include cover for financial loss?  115 It would in my judgment be surprising if the parties intended by this wording to introduce cover of a wholly different nature from that given by section 1 of the policy, section 1 being headed ‘Property’. If the word ‘loss’ is in this context to be given an unrestrained general meaning then […] the cover is transformed and exposures wholly alien to the policy are introduced.

This approach is further illustrated most recently by TKC London Ltd v Allianz Insurance plc,47 a case of business interruption during the Covid-19 pandemic, where the court accepted the proposition that ‘the word “loss” in property damage insurance usually (though not invariably) has a physical element’,48 and further held that deterioration of a restaurant’s food stocks during the pandemic was due to inherent vice and not an all risks loss.

3.

THE UNITED STATES CASES: THE FRAUDULENT BILL OF LADING CLAUSE

3.1

Introduction to the United States Cases: The Need for the FBOL Clause

It is appropriate at this point to consider the Unites States cases on the Fraudulent Bill of Lading Clause (‘FBOL clause’) which was imported into English marine cargo policies in the 1980s. The need for this type of cover became apparent in the United States as a result of the notorious ‘salad oil swindle’ in 1963 that caused losses of more than US$180 million to banks who had loaned money on non-existent salad oil.49 Vessels supposedly full of salad oil would dock and inspectors and surveyors would certify the quantities of the cargo on board when in fact the ships’ tanks contained mainly water, with a few feet of salad oil floating on top. When salad oil prices fell the fraud was exposed, leading to a series of cases in New York that, ‘like Banquo’s ghost’, were still continuing in 1979, 16 years after the scandal came to light.50 The practice in the United States is to underwrite cargo insurance based on the American Institute Cargo Clauses, which provide cover against named perils including ‘assailing thieves’.51 The practice is then to extend that cover by tailored clauses, often, but not always, including all risks. Where cover is limited to assailing thieves the cover will not extend to losses caused as a result of fraudulent bills of lading or warehouse receipts as such losses are not carried out by assailing thieves, that is, by robbery with violence. So it appears to have come about that there was a need, in the light of the salad oil swindle and earlier cases,52 for cover for fraudulent bills of lading and thus the FBOL clause came into being. [2020] EWHC 2710 (Comm); [2020] Lloyd’s Rep IR 631. At [118]. 49 As a result of this scandal the New York courts took the view that the risk posed by fraudulent bills of lading was clearly known to the insurance industry: see Chemical Bank v Affiliated FM Insurance Company 1993 AMC 1743; 815 F. Supp 115 (S.D.N.Y. 1993). 50 Scarburgh Co v American Manufacturers Mutual Insurance 107 Misc. 2d 772 (N.Y. Sup. Ct. 1979), 435 N.Y.S. 2d 997 per Richard Wallach J in opening his judgment on a case of non-disclosure relating to the salad oil swindle. 51 For the origins of this term in United States policies see J Gilman et al., Arnould’s Law of Marine Insurance and Average, 20th edn, Sweet & Maxwell, 2023 (hereafter ‘Arnould’) at [23–33]. Under English law the term ‘thieves’ does not cover ‘clandestine theft’: MIA 1906 Sch. 1, Rules for Construction of Policy, r.9. 52 Nieschlag & Co v Atlantic Mutual Insurance Co 43 F.Supp.797 (S.D.N.Y. 1941). 47 48

116  Research handbook on marine insurance law The position in England is different to the extent that the usual practice is to use the Institute Cargo Clauses (A), which provide all risks cover. A loss caused by acceptance of a fraudulent bill of lading or warehouse receipt would not be covered under a named perils policy under English or United States practice but would be covered under an all risks policy which covers all types of theft of physical goods including by fraud, such as by what was formerly called larceny by a trick.53 There is therefore no independent role in a standard English marine cargo insurance on all risks terms for a FBOL clause that covers physical loss of goods by theft: theft in all its forms is already covered. Nevertheless, in the 1980s the FBOL clause was to be found in English all risks policies, for example for American clients, and this prompted the Joint Cargo Committee to issue a memorandum to the market on 14 January 1983 (the ‘JCC Memorandum’) which read, so far as material, as follows:54 Fraudulent Bills of Lading The attention of the Joint Cargo Committee has been drawn to the practice of some underwriters who have granted additional cover […] on the following basis: ‘This Policy covers loss or damage through the acceptance by the Assured and/or Agents and/or Shippers of fraudulent Bills of Lading and/or Shipping Receipts and/or Messenger Receipts and/ or shipping documents. Also loss or damage caused by the utilization of legitimate Bills of Lading without the authorization and/or consent of the Assured or its Agents.’ The practice […] could have the effect of committing Underwriters to financial losses unrelated to any maritime perils and where no cargo of any nature exists.

This prescient memorandum remains true today. 3.2

The United States Cases

The first reported marine cargo case in the United States to address the problem of financial loss due to fraudulent documents appears to be Nieschlag & Co v Atlantic Mutual Insurance Co,55 where the assured advanced money against fraudulent warehouse receipts for bags of non-existent cocoa beans. The cocoa beans were insured under a marine cargo policy covering ‘goods the property of the assured’ against named risks including fire, theft, water damage and non-delivery. It was held at that the insurance ‘clearly contemplated that [the insurers] should be liable only if the assured had an insurable interest in the goods insured’. As the assured had no ownership or rights of possession over any cocoa beans, the claim failed. In response to the argument that the policy extended to all risks, the court pointed out that this would not assist

53 Australia & New Zealand Bank Ltd v Colonial & Eagle Wharves Ltd; Boag (Third Party) [1960] 2 Lloyd’s Rep 241, where goods, insured against all risks, were lost due to the issue of false warehouse orders. The argument that there was no ‘physical’ loss because the goods were, as a result of the fraud, delivered to the intended recipient, was rejected. Held physical loss due to larceny by a trick, McNair J at p.251. Applied in a marine cargo case in Hong Kong where the CIF buyer obtained delivery of the goods without payment by fraudulently obtaining the original bill of lading, Anbest Electronic Ltd v CGU International Insurance plc HCCL 82/200 (unreported) Stone J at [169]; [2009] HKEC 6 (HKCA). 54 See Chemical Bank v Affiliated FM Insurance Company 1993 AMC 1743; 815 F. Supp. 115 (S.D.N.Y 1993) where the JCC Memorandum is set out. 55 43 F.Supp.797 (S.D.N.Y. 1941).

Should marine cargo insurance include cover for financial loss?  117 the assured, even if it were the case, as the policy would still only ‘insure goods [the assured] owned or title to which it held as security’. On 8 September 1986 a Preliminary Notice of Loss was distributed by the brokers concerned to some 40 Lloyd’s syndicates and London Market companies (the ‘London Insurers’) advising them of a claim for more than US$100million by Andina Coffee and their bankers based upon ‘Fraudulent Bills of Lading coverage’. The London Insurers subsequently settled the claim. However, the claim against Affiliated FM, the previous insurers, was the subject of continuing proceedings in New York, Chemical Bank v Affiliated FM Insurance Company.56 In those proceedings the court was asked to construe the FBOL clause, which read as follows: ‘This policy covers loss or damage occasioned through the acceptance by the Assured and/or their agents or shippers of fraudulent Bills of Lading and/or shipping receipts and/or messenger receipts and/or warehouse receipts and/or truckmen’s receipts.’ In a summary judgment application the central issue was whether the FBOL clause covered losses of non-existent goods. In essence the question was whether the word ‘loss’ in the FBOL clause covered financial loss when used in a conventional ‘marine open cargo policy’ in the traditional United States form covering maritime perils. District Judge Broderick, in the New York Federal Court for the Southern District of New York, held that the FBOL clause provided coverage to Andina and the banks even where the goods did not exist. Although the policy was a marine cargo policy it could be extended by clear words such as those of the FBOL clause. He said:57 [Insurers’] primary argument is that the core purpose of the policies involved was to cover physical goods and merchandise (consisting principally of coffee), against losses on the high seas and related hazards to which shipments of merchandise may be exposed […] Certainly the Affiliated policy was basically intended to provide coverage against injuries caused by loss of goods. This does not mean that other clauses included in the policy which by their terms extended beyond this central objective were rendered void.

Judge Broderick acknowledged, but distinguished,58 cases on the Transit Clause saying that ‘cases rejecting claims for insurance coverage on the ground that risk did not attach because the contested goods were never in transit are not relevant because those cases did not have clauses analogous to the FBOL clause here’. He held that under New York law a potential ambiguity must be resolved against insurers and, echoing the words of Sir Geoffrey Voss MR, which begin this article, he said:59 ‘The insurer Affiliated correctly points out that contracts should be construed as a whole. But there is no reason why a provision which extends coverage with respect to a particular matter may not be effectively included in a policy which otherwise has more limited coverage.’ The overall approach of the New York court is similar, in principle, to the approach of the English courts to the extent that this case recognises that a marine cargo policy may extend to

58 59 56 57

1993 AMC 1743; 815 F. Supp. 115 (S.D.N.Y 1993). At para IV. Fn 9 to the judgment. At para V.

118  Research handbook on marine insurance law financial losses where clear words are used. The principle is the same though the English and New York courts may differ in their approach when applying the rule.60 Judge Broderick also considered the JCC Memorandum of 14 January 1983, which warned that a FBOL clause could have the effect of committing underwriters to financial losses where no cargo existed, and this was another reason that he declined to give summary judgment. The decision in Chemical Bank v Affiliated FM was later distinguished in the US District Court for the District of New Jersey in Centennial Ins Co v Lithotech Sales LLC.61 In that case a ‘Marine Open Cargo Policy’ insured ‘printing presses and similar merchandise’ against ‘all risks of physical loss or damage from any external cause’. There was no evidence that a type of printing press insured under the policy ever existed. The FBOL Clause read as follows: ‘This policy also covers loss of or damage to the property insured occasioned through the acceptance by the Insured or Insured’s agent or customers or consignees or others of Fraudulent Bills of Lading or Shipping Receipts.’ The court held that the phrase ‘loss of or damage to the property insured’ limited the coverage to losses sustained by the actual subject-matter of the policy, and distinguished the Chemical Bank decision.

4.

THE RECENT ENGLISH CASES

4.1

Fraudulent Documents Where Cargo Did Not Exist

The construction issue on the FBOL clause was raised for the first time in an English court in Engelhart CTP (US) LLC v Lloyd’s Syndicate 1221.62 In this case a cargo of copper ingots in 102 containers was shipped from New York to China and was discovered, on transhipment in Hong Kong, to contain nothing but slag of nominal commercial value. The cargo did not exist. The bills of lading, packing lists and quantity certificates were fraudulent. The so-called copper ingots were insured under a ‘Marine Cargo and Storage Insurance’ on goods and merchandise, including metals, while in transit and/or in store. The policy incorporated the Institute Cargo Clauses (A) 1/1/09, which provide cover against ‘all risks of loss of damage to the subject-matter insured’ and the American Institute Cargo Clauses (Sept 1, 1965), which, as we have seen, are on limited perils, such as assailing thieves, but which were extended in this case, as is often the practice, to cover ‘Against all risks of physical loss of or damage to the subject-matter insured from any external cause’. The policy included a FBOL clause in the following terms: Fraudulent Documents This insurance contract covers physical loss of or damage to goods and/or merchandise insured hereunder through the acceptance by the Assured and/or Shippers of fraudulent documents of title, including but not limited to Bill(s) of Lading and/or […] and/or other document(s) of title.

For example, the rule under NY law that any ambiguity must be resolved against insurers contrasts with English law, where the contra proferentem rule, if applied in this case, would probably have favoured insurers. 61 187 F. Supp. 2d 214 (D.N.J. 2001). 62 [2018] EWHC 900 (Comm); [2018] 2 Lloyd’s Rep 24. 60

Should marine cargo insurance include cover for financial loss?  119 Sir Ross Cranston stated the legal rule as follows:63 Since an all risks marine cargo policy is generally construed as covering only losses flowing from physical loss or damage to goods, there must be clear words indicating a broader intention. That was the approach of Clarke LJ in Coven SpA v Hong Kong Chinese Insurance Co: with a policy entitled ‘Marine Cargo Policy’, which on its face stated that the insurers agree to insure against loss, damage, liability or expense, one is essentially concerned with insuring physical loss of goods. It would be surprising (his Lordship accepted) if it covered, as in that case, a measurement error. The same approach was adopted by Tomlinson J in Outokumpu Stainless Ltd v Axa Global Risks (UK) Ltd, albeit that the policy there was not for marine cargo, but covered all risks material damage and business interruption: see esp. [26]. The leading textbooks in the area, referred to earlier,64 cite Fuerst Day Lawson v Orion Insurance Co and Coven v Hong Kong Chinese Insurance, as authority for the proposition that, in the absence of specific provision, all risks marine cover generally does not extend to paper losses.

The rule is expressed in two ways in this passage and it is submitted that there are in fact twin rules, albeit related. First, based on the authorities, the judge says that ‘there must be clear words indicating a broader intention’ before a marine cargo policy can extend to non-existent goods: second, based on the cases and leading textbooks and authorities, he says that ‘in the absence of specific provision, all risks marine cover generally does not extend to paper losses’. The rule that ‘clear words’ are needed is now well recognised but it should not be overlooked that, because of the nature of marine cargo insurance as property insurance, there should also be a ‘specific provision’ if the policy is to extend to non-existent goods.65 In the Engelhart case it was held that, so far as the Fraudulent Documents clause was concerned, there was an insuperable difficulty for the assured as that clause expressly provided cover for ‘physical’ loss of goods through the acceptance of fraudulent bills of lading, which contrasted with the wording in Chemical Bank v Affiliated FM Insurance,66 where the policy covered ‘loss or damage’ occasioned through the acceptance of fraudulent shipping documents. The clause in the Engelhart policy was akin to that in the subsequent case, Centennial Insurance v Lithotech Sales67 where the policy insured against ‘physical loss or damage’. The judge concluded that ‘there cannot be a physical loss when nothing existed in the first place’.68 4.2

Fraudulent Documents Where Cargo Existed

The rule that a marine cargo policy does not normally extend to non-existent goods was recognised in Quadra Commodities SA v XL Insurance Company SE,69 a case where the goods

At [40], citations omitted. The earlier references in the judgment by Sir Ross Cranston (at [29]) were to Arnould 18th edn [23-72] fn 490 (currently Arnould 20th edn [23-81] fn 513), and to R Merkin, Colinvaux’s Law of Insurance, 11th edn 1st suppl. at [11-007] (currently Colinvaux’s Law of Insurance, 13th edn [11.012]). 65 The words used by Sir Ross Cranston are ‘paper losses’ rather than ‘non existent goods’ but paper losses are but one example of the principle which is equally applicable to policy provisions regarding measurement errors and also to clauses related to fraudulent documents. See also J Dunt, ‘Can marine cargo insurance provide cover for non-existent goods?’ [2018] LMCLQ 464 at 470. 66 1993 AMC 1743; 815 F. Supp. 115 (S.D.N.Y 1993). 67 187 F. Supp. 2d 214 (D.N.J. 2001). 68 At [44]. 69 [2022] EWHC 431 (Comm); [2022] 2 Lloyd’s Rep 541; [2023] EWCA Civ 432; [2023] Lloyd’s Rep IR 455 (CA). 63 64

120  Research handbook on marine insurance law existed. The judgment of Butcher J at first instance was fully endorsed by Sir Julian Flaux C70 in the Court of Appeal, both as to the facts and the issues, which are therefore taken from the first instance judgment unless otherwise indicated. The facts were as follows. Quadra Commodities SA (‘Quadra’) lost substantial sums (quantified at more than US$5.5 million plus sue and labour expenses) as a result of the ‘Agroinvestgroup Fraud’ which was ‘perpetrated by the Agroinvest Group pledging and/or selling the same parcels of agricultural commodity products to multiple traders, via the issuance of fraudulent warehouse receipts’.71 The same cargo was sold or pledged to as many as six different buyers.72 The fraudsters obtained grain from local farmers, stored it in their own silos or elevators, and sold the same grain to Quadra and other buyers who purchased these goods for export. Quadra entered into contracts with two Agroinvest Group entities, Agri Finance SA and Linepuzzle Ltd, for the purchase of various cargoes of grain in store in silos or elevators in the Ukraine, and subsequently received warehouse receipts detailing quantities of grain as referred to in those contracts. Quadra’s surveyors confirmed that the quantities of grain in the silos or elevators exceeded the amounts referred to in the warehouse receipts but, as the grain was held in co-mingled bulk storage with goods belonging to the other buyers, it was not possible to identify any particular parcel of grain. Quadra duly made payment against the receipts in accordance with the contracts, paying the full amount of the invoices to Linepuzzle and 80 per cent of the invoice prices to Agri Finance. It is to be noted that the warehouse receipts passed no title to Quadra. Apart from the difficulties arising from the possibility that Agri Finance and Linepuzzle may have already sold the cargoes to others,73 it was held that no title passed as the goods were unascertained goods for the purposes of the Sale of Goods Act 1979 (SGA) and no title could pass until the goods were ascertained, which never occurred.74 Further, even if this was wrong, the Agri Finance contracts provided for a ‘transfer’ at the inland elevators but for title itself to pass only at the seaport which, of course, was never reached.75 The fact that the warehouse receipts were not documents of title is an important factor distinguishing this case from the fraudulent bills of lading and other cases discussed above. In January 2019 Agroinvestgroup’s offices became difficult to contact; by 17 January they were closed and by late January there were rumours they were in trouble. At the beginning of February 2019, a criminal investigation was started. Some grain was found in the warehouses, elevators and silos but this was wholly insufficient to satisfy the buyers and none of the grain came to Quadra. Quadra were insured under a Marine Cargo Open policy which provided all risks cover under the Institute Cargo Clauses (A) 1/1/2009. The policy also included a Fraudulent Documents clause which covered ‘physical loss of or damage to goods […] through the acceptance by the Assured […] of fraudulent shipping documents, including […] Warehouse Receipts’.

With whom Popplewell LJ and Snowden LJ agreed. At [48]. Evidence of Mr Alastair Scott of Gray Page Intelligence Services. 72 At [49]. 73 Butcher J at [100]. 74 SGA s.16. It was argued that Quadra became owners in common of the bulk under SGA s.20(A) but this argument was rejected: see Butcher J at [101] to [112]. 75 Butcher J at [113]. 70 71

Should marine cargo insurance include cover for financial loss?  121 In addition there was a Misappropriation Clause which covered ‘all physical damage and/or losses’ caused by (1) the disposal of the ‘insured goods’ in bad faith by a contracting party and (2) physical delivery of the ‘insured goods’ in bad faith to persons not entitled to the delivery of the goods. The underwriters argued that there was no loss of physical property and that the assured’s loss was a purely financial loss, in respect of which it was not insured.76 The judge was satisfied, as a matter of fact, that goods corresponding in quantity and description to the cargoes were physically present at the time the warehouse receipts were issued.77 The existence of the goods was evidenced by the receipts themselves and the inspections carried out by the surveyors. It was also corroborated by two deliveries of existing grain and corn.78 Quadra did not have title to the goods, but they had an insurable interest in the goods for two reasons: (1) they had paid in advance for them and were liable to suffer prejudice under the contracts of purchase and (2) they had an immediate right to possession of the goods under Ukrainian law. On the question of whether the marine cargo policy extended beyond physical loss, Butcher J applied the established rule, saying:79 Consistently with what I have said above as to the subject matter of the insurance, I do not consider that the Policy covers a situation where no property has existed (and thus has not been lost or damaged). An all risks marine cargo insurance does not ordinarily cover such a situation, though it can be extended to do so by clear terms.

Butcher J held that Quadra’s claim was covered under the Misappropriation Clause on the facts as he had found them based on the nature of the fraud perpetrated by the various parties in the Agroinvestgroup.80 However, the loss was not covered under the Fraudulent Documents clause because the physical loss of the goods was not caused by Quadra’s acceptance of fraudulent warehouse receipts.81 The decision illustrates and re-states the rule of construction regarding physical loss. However, it analyses what is a common form of cargo fraud, where multiple documents are issued in respect of the same cargo, as a loss of physical goods. This contrasts with the remarks of Jacobs J in the case of ABN Amro Bank NV v Royal & Sun Alliance Insurance plc,82 where, considering the reasons why a Fraudulent Documents clause was expanded to cover ‘financial loss’, he said:83 At [51]. At [69] and following. The Court of Appeal held that there was ample evidence for this finding, Sir Julian Flaux C at [118]. 78 There was a delivery of 800mt to one of Quadra’s sub-buyers, which suggested the existence of grain in the silos, and a delivery of 8,000mt of corn loaded aboard the vessel ‘ATHERINA’: see judgment of Sir Julian Flaux C at [116] to [117]. 79 At [66], citing Engelhart CTP (US) LLC v Lloyd’s Syndicate 1221 [2018] EWHC 900 (Comm), [2018] Lloyd’s Rep IR 368 and the cases assembled by Sir Ross Cranston at paragraphs 23–33 of that authority, including in particular Fuerst Day Lawson Ltd v Orion Insurance Co Ltd [1980] 1 Lloyd’s Rep 656, Coven SpA v Hong Kong Chinese Insurance Co [1999] Lloyd’s Rep IR 565, and the US case of Centennial Insurance Co v Lithotech Sales LLC 187 F Supp 2d 214 (S.D.N.Y. 2001). 80 At [118]. 81 At [119]. 82 [2021] EWHC 442 (Comm); [2021] Lloyd’s Rep IR 467; [2021] EWCA Civ 1789; [2022] Lloyd’s Rep IR 201 (CA). 83 At [211]. 76 77

122  Research handbook on marine insurance law The expansion of the cover in 2015/2016 reflected a widely known problem at that time, referred to at the trial as Qingdao, where a company deployed fraudulent duplicate warehouse receipts to raise trade finance secured against the same stockpile of cargo. Multiple versions of the warehouse receipts were fraudulently created for the same parcel of goods, leaving numerous insureds suffering loss in the absence of any physical loss or damage or theft of the goods.

It is apparent from this passage that Jacobs J considered that the Fraudulent Documents clause in question was amended by the parties because they looked upon the issue of multiple warehouse receipts for the same cargo as causing ‘financial loss’ rather than ‘physical loss’ and they therefore needed to extend the cover to ‘financial loss’.84 It is probable that the amendment also reflects an insurance market view that losses incurred as a result of the issue of multiple documents for the same cargo, such as in the Qingdao and the Agroinvestgroup frauds, are financial losses rather than physical losses because no right to physical cargo passes to the assured, which is effectively tricked out of its money and not its cargo. However, these types of frauds, where multiple documents are issued, may vary in important respects. While it is normally integral to this type of fraud that at some point there is existing cargo that can be inspected by the buyer,85 in some cases real cargoes are interposed with phantom shipments where no cargo exits.86 In the Quadra case the existence of the cargo was evidenced by the warehouse receipts ‘transferring’ or ‘delivering’ the cargo to Quadra87 as well as by the inspection reports. The existence of the grain was further corroborated by the fact that in two cases existing grain was delivered to Quadra to fulfil sub-sales. This distinguishes the Quadra case from many of the other frauds discussed above. Moreover, as we have seen, in the Quadra case the warehouse receipts were not false documents of title such as the fraudulent bills of lading considered above. As the Court of Appeal made clear, the warehouse receipts were evidence of the existence of the grain ‘but could not be relied upon as evidence that Quadra owned the grain in question’.88 The receipts were therefore some evidence of existing grain but not evidence that the grain was the property of Quadra. However, this was not determinative as long as Quadra could prove a sufficient insurable interest in the existing cargo which they successfully did by virtue of their payments for the grain, in circumstances where they were liable to suffer prejudice under the contracts of purchase, and their immediate right to possession of the goods under Ukrainian law. As the grain was sold up to six times in the Quadra case it is likely, though this could not be proved,89 that the insurance markets (including the same underwriters) may have paid claims several times for ‘physical loss’ of multiple cargoes in circumstances where only one cargo existed. The extract from the ABN Amro case cited above suggests that this scenario had been 84 The quoted passage may also be read as indicating that it was also the judge’s view that such a fraud gave rise to a financial rather than a physical loss, but the point was not considered in any detail by Jacobs J and only mentioned in passing in relation to the history of the clause that he was called upon to consider. 85 As was the case in respect of the Agroinvestgroup fraud: see Butcher J at [71] and Sir Julian Flaux C at [111]. 86 See for example Chemical Bank v Affiliated FM Insurance Company 1993 AMC 1743; 815 F. Supp. 115 (S.D.N.Y. 1993) discussed above in the section on United States cases. 87 No title passed to Quadra. The goods were ‘transferred’ in the case of the Agri Finance contracts and ‘delivered’ in the case of Linepuzzle, see Butcher J at [80]. 88 Sir Julian Flaux C at [112]. 89 See Sir Julian Flaux C at [136].

Should marine cargo insurance include cover for financial loss?  123 treated by the market as a ‘financial loss’ for which Quadra was not insured. If this was the understanding, the insurance market will need to consider revised wording that limits ‘physical loss’ to: (1) existing cargo that (2) belongs solely to the insured named under the policy in question, narrowing the insurable interest to one of ownership. 4.3

Financial Risks Where Cargo Exists but Suffers No Physical Loss or Damage

In ABN Amro Bank NV v Royal & Sun Alliance Insurance plc90 a bank suffered losses due to the default of its customers, traders in cocoa beans and cocoa products, to whom the bank had loaned money on the security of cargo which was not damaged by physical perils but proved worthless as it was ‘old’ and in awful condition when the customer became insolvent. There were no marine perils or all risks perils operating on the cargo. The risk was a trade credit risk. The bank needed marine cargo insurance as it took the cargo as security by purchasing it for a set period, say three months, at the end of which the customer was obliged to repurchase the cargo at a pre-agreed price which allowed the bank a profit on the transaction. The bank, as owners of the cargo, were therefore insured in the London marine cargo insurance market under a ‘Marine Cargo and Storage Insurance’. The General Conditions of the policy and Section 1 provided that the policy insured against risks of ‘physical’ loss or damage to the subject-matter Insured. In his opening summary, Jacobs J described the policy as follows:91 ‘The policy was built on a foundation of conventional marine ‘all risks’ terms including the Institute Cargo Clauses ‘A’. However, it also contained a large number of detailed clauses, including extensions to the cover which went beyond ordinary physical loss and damage to the cargo.’ The three92 extensions to the cover that were held to extend beyond PLOD were as follows: (1) ‘Additional Cover [for] Confiscation and Expropriation’, or CEND, cover. This covered ‘loss of and/or damage to the Subject Matter Insured’ caused by confiscation and associated perils including ‘deprivation’, which was defined as loss of use or possession of the subject-matter insured caused by refusal of a foreign government to permit export of the goods for a period of three months. (2) ‘Business Contingency Cover’ or BCC. This included cover against ‘any premium or profit element that the Insured would otherwise have earned but for the delayed delivery of the subject-matter insured to the Exchange in the related Futures Month’. (3) A ‘Fraudulent Documentation’ clause, which covered ‘direct financial loss’ suffered by the Insured by reason of their relying upon, or being supplied with, a counterfeit document or a fraudulently altered document.

That central construction issue in the case turned on the meaning of the Transaction Premium Clause (‘TPC’), which provided: ‘Underwriters note and agree that, in respect of any [2021] EWHC 442 (Comm); [2021] Lloyd’s Rep IR 467; [2021] EWCA Civ 1789; [2022] Lloyd’s Rep IR 201 (CA). 91 At [11]. This passage could be read to mean that the policy ‘contained a large number of extensions to the cover which went beyond ordinary physical loss or damage to the cargo’: see headnote at [2021] Lloyd’s Rep IR 467 (Jacobs J) at p.467, repeated at [2022] Lloyd’s Rep IR 201 (CA) at p.201. However, the way the sentence is punctuated does not suggest that was the intended meaning; nor do the facts of the case. 92 The issue was whether a fourth clause, the Transaction Premium Clause (TPC), went beyond PLOD. 90

124  Research handbook on marine insurance law Transaction, it is hereby confirmed that the Insured is covered under this contract for the Transaction Premium that the Insured would otherwise have received and/or earned in the absence of a Default on the part of the Insured’s client.’ The words ‘Transaction Premium’, as used in the clause, were defined in a series of elaborate definitions, and were held to mean that the policy covered the whole of the loss suffered by the assured bank if the debtor defaulted on the loan and the insured goods, which secured the loan, proved inadequate and had to be sold at a loss. There was no requirement that the goods be lost or damaged during the currency of the policy. Jacobs J said that the TPC ‘was long and tightly drafted and its full import would not necessarily be grasped by an underwriter on a first reading’.93 This assessment may be contrasted with the evidence of Mrs Joyce Webb, the insurers’ underwriting expert. She said that it took her a number of sessions to actually work out how that clause worked and that she had to read it a number of times before she understood it.94 The TPC was obscure to an experienced underwriter such as Mrs Webb because the heading ‘Transaction Premium’, and the other words of the TPC, are not used in a way that would be familiar to a marine insurance underwriter or practitioner but have a special meaning, common, perhaps, in trade finance. Looking at the TPC through the eyes of a marine underwriter or practitioner, the meaning of the words used in the clause is surprising, if not unusual, and recalls what Humpty Dumpty said to Alice: ‘When I use a word it means just what I choose it to mean – neither more nor less.’ However, though the TPC is drafted with wording unfamiliar to cargo insurers, unlike Humpty Dumpty, who used words with a special meaning known only to himself, the draftsman here has provided the key in the definitions in the TPC. The enigma can be solved by careful application of those definitions – the key to the code. Applying these definitions over a number of sessions, the TPC is clear and, indeed, precisely covers the exact loss earlier anticipated by the bank for which they prudently required insurance cover to be added to their policy. It was argued by underwriters that the TPC, when construed in the context of a marine cargo insurance, was a valuation clause providing a formula for the quantification of all risks losses causing ‘physical’ loss of or damage to goods as expressed elsewhere in the policy. The bank argued that the three clauses mentioned above (the CEND, BCC and Fraudulent Documents clauses) showed that the policy was not limited to physical loss or damage, either by section 1, or the General Conditions (which both referred to ‘physical’ loss); that the underwriters were prepared to insure financial loss; and that the words of the TPC were clear and covered default by the bank’s customers. Jacobs J’s judgment reviews the Supreme Court cases on construction of commercial contracts,95 and then considers Engelhart CTP (US) LLC v Lloyd’s Syndicate 122196 in some detail.97 It adopts and applies Sir Ross Cranston’s observation, based on the authorities, that since an ‘all risks marine cargo policy is generally construed as covering only losses flowing At [921]. At [601]. Jacobs J described Mrs Webb as an ‘impressive witness’ with far greater underwriting experience than the broker’s underwriting expert. 95 Rainy Sky SA v Kookmin Bank [2011] UKSC 50; [2012] 1 Lloyd’s Rep 34, [2011] 1 WLR 2900; Arnold v Britton [2015] UKSC 36; [2015] AC 1619; Wood v Capita Insurance Services Ltd [2017] UKSC 24; [2018] Lloyd’s Rep Plus 13; [2017] AC 1173. 96 [2018] EWHC 900 (Comm); [2018] 2 Lloyd’s Rep 24. 97 At [179] to [183]. 93 94

Should marine cargo insurance include cover for financial loss?  125 from physical loss and damage to goods, there must be clear words indicating a broader intention’. Jacobs J went on to say:98 There was some debate as to whether it is appropriate to speak in terms of a ‘presumption’. I consider that this is simply another way of expressing the idea that if a policy such as the present, or a particular clause therein, is to be construed as extending beyond physical loss and damage, there needs to be contractual language which clearly so provides. [Underwriter’s counsel] in his oral closing said that the relevant proposition to be found in Engelhart is that clear words are needed ‘if you’re going beyond the normal risk contemplated in the marine cargo market’. In my view, however, it is simpler to say that, as Sir Ross Cranston’s judgment indicates, there must be clear words if the policy is to be construed as covering losses other than physical loss or damage to the goods.

Jacobs J concluded that the TPC was applicable to and covered the trade credit risk claims made by the bank because: (1) the underwriting experts agreed that extensions that were not limited to PLOD were available in the market at the time; (2) the policy itself contained three separate add-ons which did not require PLOD and (3) the wording of the TPC was ‘clear’. In particular, the language of the TPC, which was not a valuation clause, ‘was clear’ in accordance with the rule in Engelhart. The clear wording of the TPC prevailed over the evidence of market background. The case went to the Court of Appeal99 where the judgment given by Sir Geoffrey Voss MR agrees with the approach of the trial judge on the construction issue,100 and applies the rule in Engelhart that a marine cargo policy covers physical loss and damage unless there are clear words which provide wider cover.101 It states the central issue as follows:102 ‘The problem for the appellant Underwriters was, as the judge also said, that add-ons to standard physical loss and damage cover were common in the market, and there was no reason why such an add-on could not give protection for financial default.’ It is submitted that this part of the judgment, both at first instance and on appeal, is not convincing. As was argued by underwriters at first instance, the fact that the insurers were prepared to accept some risks going beyond PLOD was ‘very different’ to their agreeing to accept ‘trade credit or financial guarantee insurance’.103 At the outset of his judgment Jacobs J acknowledges that it was ‘plain’ that none of the three clauses which were held to extend beyond PLOD ‘provided coverage akin to coverage for losses arising from the default of the Bank’s counterparties’.104 But cover for ‘default’ is only a minor aspect of the difference. The fundamental difference is not the risk of ‘default’ but the risk of default due to ‘the financial default or insolvency of any party’.105 It is the practice in almost every major open cargo cover issued in London and worldwide,106 to cover FOB or C&F/CFR shipments under a Seller’s At [183]. [2021] EWCA Civ 1789; [2022] Lloyd’s Rep IR 201. 100 At [26] Sir Geoffrey Voss appears to indicate that Jacobs J stated the Engelhart rule in terms of ‘clear words before the cover could be held to extend beyond the normal cover in the marine cargo market’ but preferred to state the rule more narrowly, as Sir Ross Cranston did, in terms of PLOD. 101 At [69] and [70]. 102 At [69]. 103 At [236], Luke Parsons KC. 104 At [46]. 105 As defined in the Lloyd’s Agreement referred to earlier in this article as risks that Lloyd’s underwriters would not undertake. 106 Including the ABN Amro cover, see [264]. 98 99

126  Research handbook on marine insurance law Interest clause which extends cover for loss or damage to the goods where the buyer defaults on its contractual obligations because the cargo is lost or damaged. In such circumstances the buyer’s contractual default is triggered by the fact that the goods are lost or damaged and the buyer seeks to avoid payment though it remains financially sound. Cargo underwriters have no problem with ‘default’ where cargo is lost or damaged as this is a cargo risk: the objection is to financial default and insolvency because the other party has no money to pay. This is not a cargo risk and not a risk which cargo underwriters can assess when the risk is placed. In this case, the fact that the policy had extensions which extended to PLOD has only marginal relevance, if any, to the construction of the TPC. A trade credit risk policy, such as the TPC, is not an extension of PLOD, as is commonly found in the marine cargo markets, but a different category of risk in an entirely different class of business. Sir Geoffrey Voss MR concludes that the words of the TPC provided coverage and were not a valuation clause, and that they were clear and apt to cover the loss in question. He closes his judgment by saying that what the parties had done in this case was ‘to combine two types of insurance in a single policy’.107 As discussed earlier, when considered long and carefully in light of the definitions, the TPC is clear. The judgments arrive at the only possible conclusion. Underwriters must be bound if they agree a coverage clause that is clear, however obscure until unravelled, and however unexpected in the type of policy in question. The clarity of the wording of the carefully drafted TPC overrides the market considerations, albeit that it remains surprising that Lloyd’s underwriters, in particular, agreed to write credit risks business.108 The ABN Amro Bank case is also significant in relation to the construction of the three clauses said to provide cover extending beyond PLOD. The first of these, the CEND Clause, provided cover for ‘loss or damage to the subject matter insured’ caused by confiscation and a range of similar perils including ‘deprivation’ which included loss of use or possession of the cargo for a period of three months. The BCC Clause109 was rather specific to the bank’s particular ‘loan’ arrangement and included cover against any profit element that would have been earned but for the delayed delivery of the cargo to the exchange. The judge accepted that there was some analogy with PLOD regarding these two clauses110 as, in both cases, the assured was deprived of the physical cargo for a period of time, and in that sense, might be said to suffer a physical loss, at least ‘pro tem’. However, Jacobs J considered that the analogy was not exact because the cargo may have been entirely unharmed and might have survived in store for many years.111 As indicated earlier in the introduction to this chapter, the PLOD concept is subject to exceptions including the special rule that cargo insurance covers loss of the adventure where the cargo is ‘entirely unharmed’. Moreover, as we have seen, there is a constructive total loss of cargo where it is unlikely that the assured can recover the cargo within a reasonable time, which, in the case of cargo, may be a short period At [70]. See further J Dunt, ‘When marine cargo insurance became credit risk insurance’ [2021] LMCLQ 431. Cf Cheng Lin, ‘Understanding the legal nature of export credit insurance: lessons from ABN Amro v RSA’, J.B.L. 2022, 3, 225–37, which examines the case from the point of view of a specialist in the law of credit insurance. 109 No declarations were made under this clause. 110 At [234]. 111 The evidence was that cocoa beans (as distinct from cocoa products) could survive in store for many years. 107 108

Should marine cargo insurance include cover for financial loss?  127 such as three months. Seen in this light it is submitted that the judge’s remarks, though sympathetic to the ‘analogy’ with PLOD, perhaps do not give sufficient weight to these factors, in particular in relation to the CEND cover. The Fraudulent Documents clause deserves more detailed consideration. The earlier 2015 policy included a ‘Fraudulent Documentation’ clause (Clause 17.3) which covered ‘direct financial loss suffered by the Insured, in good faith during the Period and in the ordinary course of business acting upon a ‘Counterfeit Document’ (as defined below) […] subject to a limit of USD5,000,000, for any one event or loss during the Period’.112 The cover for financial loss is limited to US$5million compared to the much higher general policy limits which, for cocoa and cocoa beans, were US$50million ‘any one conveyance’ and US$100million ‘any one individual warehouse’. This suggests that the underwriters were aware of the need to provide a separate limit for financial losses as, in the absence of cargo, there would be no limit. Furthermore, they imposed a limit ten times less than the normal policy limit for goods in warehouses, suggesting they were aware of the need to control the risk because it went beyond PLOD. The judge described the background to the development of the Fraudulent Documents clause, saying that it had been amended following the Qingdao fraud, where a company deployed fraudulent duplicate warehouse receipts to raise trade finance secured against the same stockpile of cargo. Omitting Clause 17.1 and the definitions, the amended clause provided as follows: 17. Fraudulent documentation 17.2 If as a result of a Document being stolen, lost and/or misappropriated and such document is fraudulently converted such that the Insured suffers physical loss of the related subject matter insured, or the impairment of its interests [sic] in the subject matter insured, such a loss shall be recoverable hereunder. 17.3 The Insured is indemnified by the Underwriters for the direct financial loss suffered by the Insured including by reason of the impairment of the Insured’s interest in the subject matter insured arising by reason of the Insured, either in good faith during the period and in the ordinary course of business acting or relying upon or being supplied with a ‘counterfeit document’ of [sic] ‘fraudulently altered document’.

Significantly there is no limit in the amended clause 17.3 for financial losses and the special lower limit of US$5million to control financial losses no longer appears in the clause.113 Jacobs J concluded, in relation to the Fraudulent Documents clause, that:114 The analogy [with PLOD] is inapt, or at least less persuasive, in relation to clauses 17.2 and 17.3 in the Fraudulent Documentation clause. These clauses clearly go beyond the wording of the clauses considered in Engelhart. Clause 17.2 covered not only ‘physical loss of the related subject matter insured’, but also ‘impairment of its interests in the subject matter insured’. This wording appears to have been added in order to cover the situation in a case such as Qingdao. Furthermore, clause 17.3 is

At [44]. The position on limits cannot be stated with certainty as the full policy is not set out in the judgment of Jacobs J and the extract attached as Appendix 1 to the judgment of Sir Geoffrey Voss MR is not, unfortunately, attached to the Lloyd’s report or to the Bailee version of the CA judgment. 114 At [235]. 112 113

128  Research handbook on marine insurance law also not confined to physical loss, but covers ‘direct financial loss suffered by [the Insured including by] reason of the impairment of the insured’s interest in the subject matter insured’.115

It appears that Clause 17.2 was originally designed to cover ‘physical loss’ caused by the fraudulent use of stolen documents while Clause 17.3 covered financial loss caused by the use of counterfeit documents. The material words added as a result of the Qingdao fraud refer to ‘the impairment of the Insured’s interest in the subject matter insured’ and, in addition, Clause 17.3 was extended to cover misuse of a ‘fraudulently altered document’. Jacob J’s approach to the phrase ‘interest in the subject matter insured’116 is surprising in light of the cases examined earlier in this article. One of the central themes in those cases is that a clause limited to the ‘subject matter insured’ does not breach the PLOD rule because it is concerned with loss of existing goods. The insurable interest covered by a marine cargo policy in the usual form, is property, that is, goods or merchandise.117 As Clause 17.2 only applies to ‘physical loss’ or where the ‘subject matter insured’ is ‘impaired’, the normal construction, in light of the authorities, is that there must be existing cargo. Clause 17.3 is a contrasting clause which does not refer to ‘physical loss’ but refers to ‘direct financial loss’. It covers financial loss where the assured relies on a counterfeit or fraudulently altered document. This is a serious breach of PLOD and, without the limit to US$5million, contravenes the all-too-prescient warning in the JCC Memorandum.

5.

ANALYSIS AND CONCLUSION

The issue raised in this chapter is to what extent should marine cargo insurance be construed to include cover for financial loss. It is summitted that this is best approached by application of the rules of construction at three levels and then, if necessary, by more detailed analysis based on the central characteristics of a marine cargo policy, in particular (1) insurable interest and (2) attachment of risk. The three levels to the rules of construction are as follows. First, the general rules for the construction of commercial contracts, as recently confirmed by the Supreme Court,118 which provide the starting point for the construction of any insurance policy. Second, the rules for the construction of property-related insurance policies, which hold that ‘loss’ under such policies usually (though not inevitably) has a physical element. These rules provide a strong indication 115 This final quotation from the policy omits the words of 17.3 ‘… the Insured including by…’. Without the word ‘including’ the cover would, arguable, be limited to impairment of the assured’s interest in the subject matter insured, which was cargo, so the policy would not extend to financial loss. 116 Any doubt as to the meaning of the words ‘interest in the subject matter insured’ is resolved by clause 17.1 where the words ‘the Insured’s interest in the subject matter insured’ clearly refer to the insured goods being in the legal ownership of, or at the risk of, the assured, using wording very common in brokers’ covers to attach the risk. The words ‘impairment’ of the insured’s interest do not seem to extend the clause to financial loss, whatever ‘impairment’ may mean, as there must be a property interest in cargo. 117 See Quadra Commodities SA v XL Insurance Company SE [2022] EWHC 431 Comm at [64] and [65]. 118 Rainy Sky SA v Kookmin Bank [2011] UKSC 50; [2012] 1 Lloyd’s Rep 34, [2011] 1 WLR 2900; Arnold v Britton [2015] UKSC 36; [2015] AC 1619; Wood v Capita Insurance Services Ltd [2017] UKSC 24; [2018] Lloyd’s Rep Plus 13; [2017] AC 1173.

Should marine cargo insurance include cover for financial loss?  129 of how parallel marine insurance policies, including cargo policies, should be construed. Third, the rules relating specifically to marine cargo policies, particularly the twin related rules: (1) that there must be ‘clear words’ if a marine cargo policy is to extend to financial losses; (2) in the absence of ‘specific provision’, all risks marine cover generally does not extend to financial losses where no goods existed.119 These more specific rules of marine cargo insurance have been applied to resolve the issue raised by this article but the analysis can be taken to a further level by considering the reasons for the rule that ‘clear words’ are needed, and applying those reasons, or that rationale, to the policy in question. As argued earlier in this chapter,120 it is submitted here that the rationale is not based merely on the requirement for ‘clear terms’, important as that is, but also upon the Insurable Interest issue and the Attachment issue. While it is possible to extend a marine cargo policy to financial losses, there must not only be ‘clear words’ but the policy must also deal with the question of when the risk is to attach for non-existent goods and it must widen the insurable interest requirement.121 It is not suggested here that this needs be done by amendments that refer specifically to the Institute Clauses, or other policy terms defining the interest or the transit, but that the ‘clear words’ must nevertheless answer two questions in relation to a policy that extends to financial loss: (1) what is the extended insurable interest122 and (2) when does the financial risk attach. If each of these questions cannot be answered from a consideration of the policy wording then, under English law, a marine cargo policy does not extend to financial loss.

Engelhart v Lloyd’s Syndicate 1221 at [40], as discussed above. See in particular the discussion of Coven SpA v Hong Kong Chinese Insurance Co [1999] Lloyd’s Rep IR 565 (CA). 121 It may be added that the points made by Clarke LJ in Coven SpA v Hong Kong Chinese Insurance Co relating to insurable interest and attachment of the risk are part of a wider point that the Institute Cargo Clauses as a whole contemplate existing cargo and that this is reflected in many of the clauses, such as the cover for salvage and general average in ICC (A) Cl. 2 and the exclusions, such as the exclusion for insufficiency of packing in ICC (A) Cl. 4.3. 122 Clarke LJ rejected the argument that the words ‘shortage in weight’ added to the schedule in Coven SPA v Hong Kong Chinese Insurance Co of themselves amended the insurable interest requirement in Cl.11 of the ICC saying that there was ‘no warrant for that conclusion’. Cf ABN Amro Bank NV v Royal & Sun Alliance Insurance plc where the unique wording of the TPC of itself provided cover and trumped the need for any special provision as to insurable interest. 119 120

7. Total losses under marine policies Peter MacDonald Eggers KC

1. INTRODUCTION Marine insurance contracts are contracts whereby the marine insurer promises to indemnify the assured against marine losses, which are losses incident to a marine adventure.1 The indemnity represents compensation for the assured, in money terms, in respect of such losses.2 Where the subject-matter of the insurance is property, such as a ship, cargo or an offshore platform, the measure of that indemnity is to be calculated by reference to either the sum required to make good any damage or the loss of value of the property asset. The measure of indemnity may not be quantifiable immediately on the occurrence of the loss, but only afterwards. That does not, however, prevent the cause of action against the marine insurer accruing before the claim for an indemnity is or can be quantified. The cause of action under an indemnity insurance contract normally accrues upon the occurrence of the loss or damage.3 There is a limit of any indemnity in the case of a single loss. In the case of an unvalued policy, the assured is entitled to recover up to the insurable value of the subject-matter of the insurance.4 In the case of a valued policy, the assured is entitled to recover up to the insured or agreed value. For this purpose, the measure of indemnity is regulated by reference to whether the loss is characterised as a total loss or a partial loss. The measure of indemnity for a partial loss is, in general terms, the cost of repairing or restoring the insured property or the diminution in value of the property on the assumption that it has been physically damaged.5 The general assumption appears to be that there is no measure of indemnity in a case where the assured has been deprived of possession of the property on a merely temporary basis (that is, in circumstances where there is no total loss); however, Eveleigh LJ has suggested that there might be a partial loss by deprivation of possession.6 In that event, the measure of indemnity would have to be calculated by analogy to the other provisions of the Marine Insurance Act 1906.7

Section 1 of the Marine Insurance Act 1906. Section 67 of the Marine Insurance Act 1906. 3 The Medina Princess [1965] 1 Lloyd’s Rep 361, 517–18; Callaghan v Dominion Insurance Co Ltd [1997] 2 Lloyd’s Rep 541, 544–5. However, the terms of the insurance contract may be such that the cause of action will arise only once the loss is quantified (see, for example, British Credit Trust Holdings v UK Insurance Ltd [2003] EWHC 2404 (Comm); [2004] 1 All ER (Comm) 444). 4 As defined by section 16 of the Marine Insurance Act 1906. 5 Section 69 and 71 of the Marine Insurance Act 1906. 6 Integrated Container Service Inc v British Traders Insurance Co Ltd [1984] 1 Lloyd’s Rep 154, 161. 7 Cf section 75 of the Marine Insurance Act 1906. 1 2

130

Total losses under marine policies  131 What is a partial loss and what is a total loss? These terms are defined in the Marine Insurance Act 1906 by reference to each other. A partial loss is any loss that is not a total loss.8 It is important that the assured identify the nature of its claim, whether a claim for a total loss or a claim for a partial loss. If, having claimed for a total loss, the evidence is that there has been no total loss, the assured is still entitled to claim for a partial loss.9 What then is a total loss? Such a simply stated question. Yet the answer is not always that simple.

2.

DEFINITION OF A TOTAL LOSS

There are two kinds of total loss in marine insurance law:10 1. An actual total loss (or an ‘ATL’). 2. A constructive total loss (or a ‘CTL’). If either an actual or a constructive total loss is proved, assuming the marine policy covers the loss, the measure of indemnity is the sum fixed by the policy (in the case of a valued policy) or the insurable value of the subject-matter insured (in the case of an unvalued policy).11 If the insurance policy covers ‘total losses’, the insurance will respond to both actual and constructive total losses (assuming of course that the insurance contract does not provide otherwise).12 There are a number of important observations to make about total losses under the Marine Insurance Act 1906. First, the terms are exhaustively defined by the 1906 Act. ATLs are defined under section 57 (and section 58) of the Marine Insurance Act 1906. CTLs are defined under section 60 of the Marine Insurance Act 1906. If the facts do not meet either of these definitions, then there is no total loss for the purposes of the Marine Insurance Act 1906. Of course, these definitions may be modified by the terms of the marine insurance policy. That said, only section 60 explicitly allows for such modification with the introductory words ‘Subject to any express provision in

Section 56(1) of the Marine Insurance Act 1906. A partial loss includes a ‘particular average loss’ (section 64(1)), particular charges (section 64(2)), salvage charges (section 65, 73) and general average losses (section 66, 73). 9 Section 56(4) of the Marine Insurance Act 1906. Similarly, even if a loss might amount to a ‘constructive total loss’, the assured may elect to claim only for a partial loss (section 61; Kastor Navigation Co Ltd v Axa Global Risks (UK) Ltd [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119). 10 Section 56(2) of the Marine Insurance Act 1906. 11 Section 68 of the Marine Insurance Act 1906. 12 Section 56(3) of the Marine Insurance Act 1906. 8

132  Research handbook on marine insurance law the policy’.13 There are no such similar words in section 57 (or section 58). That said, it would be odd if the parties could not alter the definition of an actual total loss if they so agreed.14 Second, the definitions of a total loss in the Marine Insurance Act 1906 appear to be limited to cases where the subject-matter of the insurance is property, such as a vessel or cargo, or where the marine policy insures other losses on condition that there is a total loss of property. Other types of loss which might be insured by a marine insurance policy, such as financial loss policies (for example, loss of hire or freight)15 or third party liability policies are not amenable to the characterisation of the relevant loss as a total loss. Third, the definitions of an ATL and a CTL both depend on there being one of two major categories of circumstances which may give rise to a total loss, namely: 1. Where the subject-matter insured is physically damaged. Physical damage requires a physical alteration of the insured property (which need not be permanent or irreparable) which alteration is commercially harmful and such an alteration will be commercially harmful where, for example, the insured property’s value or functional utility has been impaired or prejudiced by reason of the physical alteration. Such a physical alteration need not be visible and can exist at the molecular level. However, a mere impairment of value or utility, without any physical alteration of the insured property, will not constitute damage.16 However, a mere impairment of value or utility, without any physical alteration of the insured property, will not constitute damage.17 2. Where the assured has been deprived of possession, or the free use and disposal, of the subject-matter insured. This may take place for example in cases of theft, confiscation,

The definition of a CTL is commonly altered by reference to the definition of the value of the subject matter insured. The 1906 Act provides that the value is actual or market value, not the agreed value under a valued policy: section 27(4) of the Marine Insurance Act 1906. However, this value is contractually altered to the agreed or insured value by the Institute Time Clauses – Hulls (1/11/1995), clause 19.2 and by the Institute Voyage Clauses – Hulls (1/11/1995), clause 17.2. The International Hull Clauses 2003, clause 21 requires the cost of repair to exceed 80 per cent of the insured value. There is no equivalent provision in the Institute Cargo Clauses 1982 or 2009. 14 Note section 87 of the Marine Insurance Act 1906. The absence of similar words in section 55(2)(a) has led the Court to hold that no loss caused by wilful misconduct is insurable: State of the Netherlands v Youell [1997] 2 Lloyd’s Rep 440; [1998] 1 Lloyd’s Rep 236. Of course, that provision raises separate issues of public policy. 15 If there is a loss affecting the vessel or cargo such that the assured is prevented from earning the entire freight due on the adventure, there will be a total loss of freight. This cover may depend on the occurrence of a total loss of the vessel or the cargo. See Asfar v Blundell [1896] 1 QB 123; Petros M Nomikos Ltd v Robertson [1939] AC 371; Papadimitriou v Henderson (1939) 64 Ll L Rep 345; Vrondissis v Stevens [1940] 2 KB 90; Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 86. See Institute Voyage Clauses – Freight (1/11/1995), clause 12. 16 Ranicar v Frigmobile Pty Ltd [1983] Tas R 113, 116 (Tas Sup Ct); Promet Engineering (Singapore) Pte Ltd v Sturge (The Nukila) [1997] 2 Lloyd’s Rep 146; Transfield Constructions Pty Limited v GIO Australia Holdings Pty Limited (1997) 9 ANZ Insurance Cases 61-336; Quorum A/S v Schramm [2002] 1 Lloyd’s Rep 249, para. 90; Pilkington United Kingdom Ltd v CGU Insurance plc [2004] EWCA 23; [2004] Lloyd’s Rep IR 891, paras 49–53. See also Blue Circle Industries plc v Ministry of Defence [1999] Ch 289, 300–1; Hunter v Canary Wharf Ltd [1997] AC 655, The Orjula [1995] 2 Lloyd’s Rep 395, 398–9; TKC London Ltd v Allianz Insurance plc [2020] EWHC 2710 (Comm); [2020] Lloyd’s Rep IR 631, paras 117–19. 17 Transfield Constructions Pty Limited v GIO Australia Holdings Pty Limited (1997) 9 ANZ Insurance Cases 61-336. 13

Total losses under marine policies  133 seizure or detainment. In such cases, the insured subject-matter need not be physically damaged and yet may be a total loss. Fourth, the contrasting definitions of an ATL and a CTL are aimed at two different general scenarios. The first, in the case of an ATL, where there has been an absolute or actual loss of the asset insured. The second, in the case of a CTL, where there has been a loss not in absolute terms, but in terms which render the insured property of no commercial significance for the assured. As Chalmers and Owens said in The Marine Insurance Act 1906:18 In the majority of cases the distinction between actual total loss and constructive total loss corresponds with the distinction which had been drawn between physical impossibility and mercantile impossibility. A merchant trades for profit, not for pleasure, and the law will not compel him to carry on business at a loss. A commercial operation is regarded as impracticable, from the mercantile point of view, when the cost of performing it is prohibitive.

The doctrine of CTL has been developed only in the realm of marine insurance and does not exist, as a matter of law, in the context of non-marine insurance.19 That said, the parties to a non-marine insurance may regularly provide in their contract that a loss which is not an actual total loss, will entitle the assured to an indemnity under the insurance contract for a total loss, for example as a constructive, or compromised or arranged, total loss. This is commonly done, at least to an extent, in motor and aviation insurance contracts. Thus, with this distinction between ATLs (a total loss based on impossibility of recovery) and CTLs (a total loss based on impracticality of recovery), these two concepts must be examined in more depth.

3.

ACTUAL TOTAL LOSS

3.1

No Hope of Recovery

Section 57(1) of the Marine Insurance Act 1906 provides that ‘Where the subject-matter insured is destroyed, or so damaged as to cease to be a thing of the kind insured, or where the assured is irretrievably deprived thereof, there is an actual total loss’. The Marine Insurance Act 1906 was intended to codify the common law of marine insurance.20 Prior to the passing of the 1906 Act, the essence of an actual (or absolute) total loss was that there was no hope of recovery (spes recuperandi).21

(2nd edn, 1913), at p.92. See also Moss v Smith (1850) 9 CB 94, 102–3. Moore v Evans [1918] 1 AC 185, 194; Webster v General Accident Fire & Life Assurance Corp Ltd [1953] 1 QB 520, 531–2. 20 MacDonald Eggers, ‘The Marine Insurance Act 1906: Judicial Attitudes and Innovation – Time for Reform?’ in Rhidian Thomas (ed.), Marine Insurance: The Law in Transition (Informa, 2006), paras 10.9–10.10. 21 In Carras v London and Scottish Assurance Association [1936] 1 KB 291, 303–303, Lord Wright, MR equated the absence of a ‘spes recuperandi’ with an actual total loss. See also Holdsworth v Wise (1828) 7 B&C 794, 798–9; Stringer v The English and Scottish Marine Insurance Company (1870) LR 5 QB 599, 603–6. The position would be different if the insured goods had been sold: Mullett v Shedden 18 19

134  Research handbook on marine insurance law In Brotherston v Barber,22 Lord Ellenborough CJ said that ‘In cases of capture, a spes recuperandi [hope of recovery] exists: it is not as if the ship were sunk to the bottom; there must be always a greater or less degree of probability that she may ultimately be recovered’. In the same case, Abbott J made it clear that ‘It is not […] argued that the capture only, without abandonment, would have entitled the plaintiffs to recover a total loss; the claim is founded on the notice of abandonment’. Further, in Rankin v Potter,23 Cleasby B (who dissented and whose decision was not affirmed on appeal) stated that ‘The general rule is, that there is no actual total loss while the thing remains in specie, however much it may be damaged. It must either be destroyed or its recovery irretrievably hopeless. This is the rule laid down in the text-books (see Arnould, s. 364), and is the effect of all the authorities.’ This position at common law continues under the Marine Insurance Act 1906. In Marstrand Fishing Co Ltd v Beer,24 the insured vessel had been taken by barratry. Porter J said that it would be wrong to hold that the mere capture of a vessel constituted an ATL, because in any event the matter is now governed by the Marine Insurance Act 1906. The judge said: with regard to an actual total loss, it is said that barratry is analogous to capture and that capture is an actual total loss though that loss may be redeemed by a recapture. I doubt if this ever was the true question. I think it was always a question of fact whether capture was an actual total loss or merely a possible constructive total loss. Capture followed by condemnation no doubt was actual total loss, but that was because the vessel had in fact been condemned; the war was supposed to last indefinitely, and, therefore, there was no chance within any reasonable time of the ship being restored. The capture alone I do not think was ever necessarily an actual total loss. It is possible that if the vessel had been carrying contraband and that condemnation was certain, she might be held to be an actual total loss, but I do not think it is certain, even then, that that result would follow. Normally, I think capture is a constructive total loss, and the confusion which has arisen with regard to whether it is an actual or a constructive total loss arose merely because in the earlier cases the distinction between those two classes of loss was not kept clear. In the same way, damage may amount to a constructive total loss, but I think will not amount to an actual total loss, though it may amount to an actual total loss if it has been followed by sale so as to make the position one in which the vessel was lost to her owners by the proper sale after sufficient damage to justify it. The class of case I am referring to is Dean and Another v. Hornby, 23 L.J. Q.B. 129, and Stringer and Others v. English and Scottish Marine Insurance Co., Ltd., L.R. 4 Q.B. 676, at p. 686. However that may be, whether under the old law capture was or was not an actual or constructive total loss, the case is now governed by Sects. 56 to 60 of the Marine Insurance Act, 1906.

The hope of recovery therefore puts paid to any conclusion that the subject-matter insured is an ATL. The hope of recovery applies both to cases of physical damage and deprivation of possession. In the case of physical damage, it refers to the possibility of repair; in the case of

(1811) 13 East 304, 310. In Thornely v Hebson (1819) 2 B & Ald 513, the vessel was sold by the Admiralty Court for non-payment of salvage; as this was preventable, it was held that there was no total loss. 22 (1816) 5 M&S 418, 421–2, 425. See also Tunno v Edwards (1810) 12 East 488, 491; Cologan v The Governor and Company of the London Assurance (1816) 5 M&S 447. However, in the latter case, Abbott J said (at 456): ‘Capture operates as a total loss, unless it be redeemed by subsequent events […] I do not consider an abandonment as having the effect of converting a partial into a total loss; but here the loss was total in the first instance, and continued so ever after.’ 23 (1870) LR 5 CP 341, 356. 24 (1936) 56 Ll L Rep 163, 172.

Total losses under marine policies  135 deprivation of possession, it refers to the possibility of possession being restored. Of course, the one casualty may give rise to both types of loss.25 In either case, it is the hope of recovery within a reasonable time, as opposed to any time. The hope or possibility of recovery translates to the possibility that the subject-matter insured may be restored to its pre-casualty state, either in terms of its physical integrity or in terms of its availability to the assured as a functioning asset. It is for this reason why the definition of ATL in section 57 of the Marine Insurance Act 1906 employs absolute terms to describe the condition of the subject-matter insured: destruction, ceasing to be a thing of the kind insured, and irretrievable deprivation. These terms echo those used in connection with the common law notion of an ATL. In Roux v Salvador,26 hides were carried from Rio de Janeiro to Bordeaux; a leak in the vessel resulted in their putrefaction and the hides were sold en route because they would have been destroyed before the end of the voyage. The Court had to decide whether the loss of the hides was an ATL or a CTL; it held that the loss was an ATL. Lord Abinger CB addressed the requirements of an ATL first: The underwriter engages, that the object of the insurance shall arrive in safety at its destined termination. If, in the progress of the voyage, it becomes totally destroyed or annihilated, or, if it is placed, by reason of the perils against which he insures, in such a position, that it is wholly out of the power of the assured or of the underwriter to procure its arrival, he is bound by the very letter of his contract to pay the sum insured.

Lord Abinger CB then proceeded to contrast the case of an ATL with the ‘intermediate case’ of a CTL: But there are intermediate cases – there may be a capture, which though prima facie a loss, may be followed by a recapture, which would revest the property in the assured. There may be a forcible detention which may speedily terminate, or may last so long as to end in the impossibility of bringing the ship or the goods to their destination. There may be some other peril which renders the ship unnavigable, without any reasonable hope of repair, or by which the goods are partly lost, or so damaged, that they are not worth the expense of bringing them, or what remains of them to their destination. In all these or any similar cases, if a prudent man not insured, would decline any further expense in prosecuting an adventure, the termination of which will probably never be successfully accomplished, a party insured may, for his own benefit, as well as that of the underwriter, treat the case as one of total loss, and demand the full sum insured.

With this cardinal requirement that an ATL must involve the loss of the subject-matter insured without any hope of recovery, one can then turn to each of the types of ATL referred to in sections 57 and 58 of the Marine Insurance Act 1906. An ATL (formerly referred to as an ‘absolute total loss’) occurs in one of the following four situations: 1. Where the subject-matter insured is destroyed. 2. Where the subject-matter insured is ‘so damaged as to cease to be a thing of the kind insured’. See, for example, George Cohen, Sons & Co v Standard Marine Insurance Co Ltd (1925) 21 Ll L Rep 30. 26 (1836) 3 Bing NC 266, 286. 25

136  Research handbook on marine insurance law 3. Where the assured is irretrievably deprived of the subject-matter of the insurance. 4. Where the ship concerned in the adventure is missing and after the lapse of a reasonable time no news has been received of the ship. Each of these cases is predicated on the notion that it would be physically or legally impossible for the subject-matter of the insurance to survive the relevant insured peril which struck the vessel, cargo or other property.27 3.2 Destruction/Damage If the vessel or cargo is reduced to ashes or splinters, the subject-matter of the insurance has been ‘totally destroyed or annihilated’ and will be an ATL.28 Or as Lord Watson said in Sailing Ship ‘Blairmore’ Co v Macredie:29 ‘A mere congeries of wooden planks or of pieces of iron could not without reconstruction be restored to the form of a ship.’ Where an insured vessel is reduced to a wreck – that is, scattered planks and the like – the mere fact that the constituent elements of what had been a ship are recoverable does not detract from its status as an actual total loss. Similarly, where the ship is not totally destroyed, but is damaged to such a degree that it ceases to be that which it had been, there will be an ATL of the subject-matter insured. The question is whether it is physically possible to restore the vessel into a serviceable condition.30 This question is not concerned with cost or economics, but with physical possibility. Therefore, in Fraser Shipping Ltd v Colton (The Shakir III),31 when a ship was virtually split in two, it was held that the vessel was not an ATL, because there was undisputed evidence that it was possible to salvage and repair the vessel even if at great cost; the Court considered that the vessel retained its original appearance and character as a single ship.32 There was a common law qualification to this position. If it was not practicable for the master to repair a ship – which was merely a constructive total loss – and the master sold the ship as a matter of urgent necessity, the ship would become an ATL. However, with the growth of modern communications or technologies, it is difficult to conceive that this exception would have any realistic application today.33 This possibility exists with more likelihood in respect of cargo insurance, where the cargo might be perishable. Where the goods have deteriorated during the insured voyage to such an extent that they could never have reached their destination in their original character, but are sold at an intermediate port before they reach that degree of deterioration, the cargo might well be an ATL.34 Fraser Shipping Ltd v Colton (The Shakir III) [1997] 1 Lloyd’s Rep 586, 590–3; Masefield AG v Amlin Corporate Member Ltd [2010] EWHC 280 (Comm); [2010] Lloyd’s Rep IR 345; [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630, paras 17–25, 53–5; Venetico Marine SA v International General Insurance Company Ltd [2013] EWHC 3644 (Comm); [2014] 1 Lloyd’s Rep 349, para. 398–436. 28 Roux v Salvador (1836) 3 Bing NC 266, 286. 29 [1898] AC 593, 603. 30 George Cohen, Sons & Co v Standard Marine Insurance Co Ltd (1925) 21 Ll L Rep 30, 31–3; Captain J A Cates Tug & Wharfage Co Ltd v Franklin Insurance Co [1927] AC 698, 704–5. 31 [1997] 1 Lloyd’s Rep 586, 591–2. 32 See also Masefield AG v Amlin Corporate Member Ltd [2010] EWHC 280 (Comm); [2010] Lloyd’s Rep IR 345; [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630. 33 Arnould: Law of Marine Insurance and Average (20th edn, 2021), para. 28-12. 34 Roux v Salvador (1836) 3 Bing NC 266. 27

Total losses under marine policies  137 In the case of perishable or non-perishable goods, the insured cargo will be an ATL if they ‘cease to be a thing of the kind insured’. Therefore, if the goods remain ‘in specie’ at the destination, notwithstanding that they are extremely damaged, the cargo will not be an ATL.35 If the goods lose their commercial identity or value and are incapable of being used for the purpose for which they are intended, there will have been an ATL of the cargo.36 The mere fact that the cargo insured could be used for a purpose for which they were not originally intended does not mean that there has not been an ATL.37 If a part of the insured cargo is totally lost, it is possible for the assured to recover an indemnity for an actual total loss of that part, unless the insured cargo is ‘warranted free from particular average’.38 3.3

Deprivation of Possession

If the subject-matter of the insurance is not destroyed or severely damaged, that is, if the vessel or cargo is physically intact, the assured may yet suffer a total loss if it is deprived of possession of the subject-matter insured. If the assured is irretrievably deprived of the vessel or cargo, there will be an ATL. The very notion of irretrievability resides in it being impossible to recover that possession; that is, the insured subject-matter can no longer be retrieved.39 It does not matter if the impossibility is a physical impossibility (for example, if the insured vessel has sunk and rests two miles below the surface of the sea)40 or a legal impossibility (for example, if the insured cargo has been lawfully sold or confiscated).41 In either case, one may have to allow a reasonable time to elapse in order to determine whether or not there has been an irretrievable deprivation.42 There will be a presumed actual total loss in the case of ‘missing ships’. Under section 58 of the Marine Insurance Act 1906, if the vessel (whether the ship is insured or the cargo on board) is missing and no news of the vessel is received after a reasonable time, an actual total loss will be presumed. This appears to presuppose that there is no information available at all as to what has happened to the vessel. If there is such information, for example if the Court is able to conclude that the vessel has been cast away by the owner43 or if the vessel has been detained by pirates but nothing has been heard of the vessel since, in these circumstances44 the

Francis v Boulton (1895) 65 LJQB 153; Asfar v Blundell [1896] 1 QB 123. Berger and Light Diffusers Pty Ltd v Pollock [1973] 2 Lloyd’s Rep 442, 456. 37 Asfar v Blundell [1895] 2 QB 196; [1896] 1 QB 123. 38 Marine Insurance Act 1906, section 71 and 76(1). 39 Masefield AG v Amlin Corporate Member Ltd [2010] EWHC 280 (Comm); [2010] Lloyd’s Rep IR 345; [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630. 40 Anderson v Royal Exchange Co (1805) 7 East 38. 41 Cossman v West (1887) 13 App Cas 160; Marstrand Fishing Co Ltd v Beer (1936) 56 Ll L Rep 163, 172–3. 42 Forestal Land, Timber and Railways Co Ltd v Rickards (1941) 70 Ll L Rep 173; Scott v Copenhagen Reinsurance Co (UK) Ltd [2002] EWHC 1348 (Comm); [2002] Lloyd’s Rep IR 775, paras 67–73; [2003] EWCA Civ 688; [2003] Lloyd’s Rep IR 696, paras 34–50, 76–83. 43 La Compania Martiartu v Corp of the Royal Exchange Assurance [1923] 1 KB 650; [1924] AC 850. 44 Bayswater Carriers Pte Ltd v QBE Insurance (International) Pte Ltd [2005] SGHC 185; [2006] 1 SLR 69. 35

36

138  Research handbook on marine insurance law Court need not resort to the presumption in section 58 and may conclude that there has been no insured total loss or there has been an ATL on conventional terms.

4.

CONSTRUCTIVE TOTAL LOSS

4.1

Section 60 of the Marine Insurance Act 1906

A CTL is the second type of total loss, after an ATL, recognised by marine insurance law. It is a legal device for determining the measure of indemnity.45 A CTL is a concept known only to marine insurance law. In Moore v Evans,46 jewellery was insured under a non-marine (jewellers’ block) policy. The issue in the case was whether the trial judge was correct in applying principles equivalent to those applicable to constructive total losses to a non-marine policy. The House of Lords held that the trial judge had erred. Lord Atkinson referred to the trial judge’s decision: […] he then proceeds: ‘It does not to my mind follow from this that actual permanent deprivation of the timber is essential in this case. This is a business policy, and if the facts showed a loss of the timber in a commercial sense I should hold the plaintiffs entitled to succeed. If, for instance, the plaintiffs were unable, owing to a peril insured against, to deal with this timber and were liable to be prevented for so long a time that the cost of warehousing would in all probability exceed the value of the timber before the plaintiffs could dispose of it, I should be prepared to hold that there was a loss of the timber within the meaning of the policy.’ It is not necessary in this case to pronounce agreement or disagreement with this expression of opinion, but I confess to me it looks very like an application of the constructive total loss principle, He then proceeds: ‘Since writing this judgment I have read over again some observations of Blackburn J. in advising the House of Lords in Rankin v. Potter. I had forgotten them and had arrived at the conclusion indicated by me unaided by authority, but I am strengthened in my view by what Blackburn J. there says. The result is that although I think it is wrong to use the expression ‘constructive total loss’ in connection with this policy, and unnecessary to allege notice of abandonment, yet it is right in considering whether there has been a loss under this policy to take into account considerations similar to those which one would take into account in determining a question of constructive total loss under a marine policy.’ If that means that principles of the law as to constructive total loss are to be applied in effect, though not in name, but under an alias as it were, to a loss under a non-marine policy, I respectfully dissent from the learned judge’s opinion […] The strange thing is that the learned judge seems to have treated the Marine Insurance Act of 1906 as applicable to the case, and used the definitions given in it of actual and constructive total loss to help him to decide whether there was a loss under the policy or not, although that statute is expressly confined to marine insurance. In my opinion that statute had no application whatever to the policy in that case.

Connect Shipping Inc v The Swedish Club (MV Renos) [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78, para. 11. However, it is to be noted that sections 57 and 60 – defining an ATL and a CTL respectively – are set out under the heading ‘Loss and Abandonment’ in the Marine Insurance Act 1906, and not under ‘Measure of Indemnity’ (sections 67­–78). 46 [1918] AC 185, 195–7. Kuwait Airways Corp v Kuwait Insurance Co SAK [1996] 1 Lloyd’s Rep 664, 686; cf. Scott v Copenhagen Reinsurance Co (UK) Ltd [2003] EWCA Civ 688; [2003] Lloyd’s Rep IR 696, paras 34–50, 76–83. 45

Total losses under marine policies  139 A CTL is defined exhaustively in section 60 of the Marine Insurance Act 1906. It is exhaustive because section 56(1) provides that any loss other than a total loss ‘as hereinafter defined’ is a partial loss.47 There are two sub-sections in section 60 defining a CTL. Section 60(1) defines a CTL in the following terms: ‘Subject to any express provision in the policy, there is a constructive total loss where the subject-matter insured is reasonably abandoned on account of its actual total loss appearing to be unavoidable, or because it could not be preserved from actual total loss without an expenditure which would exceed its value when the expenditure had been incurred.’ Section 60(1) therefore identifies two types of CTL, namely where the subject-matter insured is reasonably abandoned (a) on account of its ATL appearing to be unavoidable, or (b) because the subject-matter insured could not be preserved from an ATL without an expense which exceeds the value of the subject-matter insured after the expense has been incurred. Common to both of these types of CTL is the concept of abandonment of the subject-matter insured. This concept is discussed below, but suffice it to say that it is not concerned with the assured’s decision to transfer the property insured to the insurer, but with the assured’s decision to give up the insured property as lost. Section 60(2) begins with the words ‘In particular’ and then proceeds to identify two instances of a CTL where the subject-matter insured is either ship or cargo. The two instances are concerned with cases where the assured has been deprived of possession of the insured subject-matter or where the insured subject-matter has been damaged. The relationship between section 60(1) and section 60(2) was the subject of debate. One view was that section 60(2) is merely illustrative of section 60(1). If this were correct, it would follow that any CTL under section 60(2) would require an abandonment of the insured subject-matter. That seems unsustainable, given the terms of section 60(2), which do not even suggest that they require any act of abandonment. The other, prevailing, view is that section 60(2) gives an independent definition of a CTL so that an assured could establish a CTL by relying on either section 60(1) or section 60(2).48 Therefore, although sub-section (2) begins with the words ‘In particular’, the House of Lords has indicated that the tests in each sub-section are independent.49 In Robertson v Petros M Nomikos Limited,50 Lord Porter said: That s. 60 is intended to be a complete and not a partial definition appears to follow from the wording of s. 56 when it says: ‘Any loss other than a total loss, as hereinafter defined, is a partial loss.’ But it Irvin v Hine (1950) 83 Ll L Rep 162, 166–8. However, it has been held by the House of Lords that it is possible to recover a constructive total loss of cargo based on the loss of the voyage or the loss of the adventure, even though such a constructive total loss is not identified in the Marine Insurance Act 1906 (British & Foreign Marine Insurance Company v Sanday [1916] AC 650; Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 90–1; Arnould: Law of Marine Insurance and Average (20th ed., 2021), paras 29-51–29-53). 48 Polurrian Steamship Company v Young [1915] 1 KB 922, 936–7; Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 83–4; Clothing Management Technology Ltd v Beazley Solutions Ltd [2012] EWHC 727 (QB); [2012] 1 Lloyd’s Rep 571, paras 34–6. 49 See also Rickards v Forestal Land, Timber and Railways Company Ltd [1942] 1 AC 50, 83–4; The Bamburi [1982] 1 Lloyd’s Rep 312, 314. Cf. Marstrand Fishing Co Ltd v Beer (1936) 56 Ll L Rep 163, 173. 50 [1939] 1 AC 371, 382, 391–2. 47

140  Research handbook on marine insurance law does not follow that the first sub-section lays down the general rule, whereas the second gives certain particular instances already covered by the general rule. Indeed, whatever may be the case with regard to sub-s. 2 (i.), sub-sub-ss. (ii.) and (iii.) do not appear to be covered in terms by the definition in sub-s. 1. But in any case, unless there is some reason to the contrary, a definition must be held to include the whole of its wording, and if particular instances are given which include matters which are outside the more general definition, that is no reason for supposing that their application is limited by the more general words. They do not merely illustrate – they add to the terms of the definition. Sect. 60 does not confine constructive total loss to cases where the subject-matter of insurance has been abandoned, though in some instances there may be no constructive total loss unless abandonment has taken place.

It is open to the parties to modify the tests applicable to define a CTL. For example, it is now almost uniformly the case that the value of the vessel insured which is relevant to determining whether there has been a CTL (which value the estimated cost of repair or recovery must exceed in order to constitute a CTL) is the agreed value in a valued policy, rather than the actual market value of the vessel (as is the default position under section 27(4) of the Marine Insurance Act 1906).51 It is not uncommon for the cost of repair or recovery being agreed by the parties to exceed a proportion of less than 100 per cent of the agreed value for a CTL to exist.52 A more difficult question concerns the effect of clause 13 of the Institute Cargo Clauses (A) (1/1/2009), which provides that: No claim for Constructive Total Loss shall be recoverable hereunder unless the subject-matter insured is reasonably abandoned either on account of its actual total loss appearing to be unavoidable or because the cost of recovering, reconditioning and forwarding the subject-matter to the destination to which it is insured would exceed its value on arrival.

There is a dispute whether or not clause 13 restricts the available tests for a CTL to the ‘abandonment’ test in section 60(1) or whether a CTL might still be established for the loss of a cargo under section 60(2). Given the House of Lords’ decision from the 1930s/40s that sections 60(1) and 60(2) are independent formulations of a CTL, and given the clarity of the language of clause 13 in restricting a CTL claim to cases of abandonment, it is difficult to see how this cannot be viewed as a contractual limitation of what constitutes a valid CTL claim, that is, limiting a CTL to cases of abandonment in section 60(1). Nevertheless, this has been the matter of some debate.53 In Masefield AG v Amlin Corporate Member Ltd,54 the Court of Appeal said in connection with clause 13: ‘There was a special CTL clause, the effect of which was agreed to exclude that form of CTL which depended on the deprivation of possession where its recovery is unlikely (in the absence of agreement there might have been debate about this).’

51 See clauses 19.1 and 19.2 of the Institute Time Clauses – Hulls (1/11/1995); clause 17 of the Institute Voyage Clauses – Hulls (1/11/1995). 52 See clause 21 of the International Hull Clauses 2003. 53 Bennett, The Law of Marine Insurance (2nd edn, 2006), para. 21.76; contra Arnould: Law of Marine Insurance and Average (20th edn, 2021), para. 29-49, footnote 310. 54 [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630, para. 18.

Total losses under marine policies  141 4.2

The Prudent Assured

Traditionally, whether or not a loss should be treated as a CTL was dependent on what a prudent person in the position of the assured would or would not do in response to an insured casualty. This is consistent with the characterisation of a CTL as a total loss based on it being commercially impractical to recover or repair the subject-matter insured. Thus, in Roux v Salvador,55 Lord Abinger said that: In all these or any similar cases, if a prudent man not insured, would decline any further expenses in prosecuting an adventure, the termination of which will probably never be successfully accomplished, a party insured may, for his own benefit, as well as that of the underwriter, treat the case as one of total loss, and demand the full sum insured.

In Irving v Manning,56 the House of Lords adopted the advice of the judges below in the Exchequer Chamber, given by Patteson J in the following terms: The principle laid down in these latter cases is this: that the question of loss, whether total or not, is to be determined just as if there was no policy at all; and the established mode of putting the question, when it is alleged that there has been, what is perhaps improperly called, a constructive total loss of a ship, is to consider the policy altogether out of the question, and to inquire what a prudent uninsured owner would have done in the state in which the vessel was placed by the perils insured against.

This approach was again adopted by the House of Lords in Sailing Ship ‘Blairmore’ Co v Macredie.57 Lord Watson said: the vessel might, nevertheless, in these circumstances be a constructive total loss; and, in my opinion, the proper test for ascertaining whether she had become so or not is the same in Scottish as in English law, although these laws may differ in regard to the date at which the test ought to be applied. The test, as I understand it, is simply this: that in order to instruct a total constructive loss, at the date to which the inquiry relates, it must be shewn that a shipowner of ordinary prudence and uninsured would not have gone to the expense of raising and repairing the vessel, but would have left her at the bottom of the sea, because her market value when raised and repaired would probably be less than the cost of restoration and repair. That, in my opinion, was the test as explained by the consulted judges and accepted by this House in Irving v. Manning.

Lord Shand echoed this approach: I say there was a constructive total loss because I understand the law to be that the test of whether a constructive total loss has or has not occurred is to be found in the answer to be given to the question what would a prudent owner do if not insured. If such an owner, having regard to all the circumstances, would abandon his vessel and would not attempt to raise and repair her because the cost of doing so would exceed her value when thus restored to her former condition, a constructive total loss has been incurred. Cases in which a prudent owner would certainly proceed to raise and repair his ship, as, for example, where it appears that at a cost, say, of 2000l., a vessel worth 10,000l.

(1836) 3 Bing NC 266, 286. See also Stringer v The English and Scottish Marine Insurance Company (1869) LR 4 QB 676, 686–7; (1870) LR 5 QB 599, 602–3. 56 (1847) 1 HL Cas 287, 306–7. See also Rankin v Potter (1873) LR 6 HL 83, 104, 155; The Wild Rose SS Co v Jupe (1903) 19 TLR 289, 290. 57 [1898] AC 593, 603, 612. 55

142  Research handbook on marine insurance law or 5000l. could be recovered and fitted up so as to be substantially as good as before she sank, would not, according to the test I have stated, be regarded as a total loss actual or constructive.

The costs of repair as adopted by the attitude of a prudent uninsured shipowner has also been adopted by the Court after the passage of the Marine Insurance Act 1906.58 Section 60 of the Marine Insurance Act 1906 has made the attitude or approach of a prudent assured less relevant. For example, a prudent assured might not consider it worthwhile to pay 90 per cent of the vessel’s value to recover or repair the insured vessel, but under section 60 there would be no CTL in those circumstances. Section 60(1) provides that there will be a CTL if there has been a reasonable abandonment of the insured property in one of two identified circumstances: either an ATL is unavoidable; or the cost of preserving the insured subject-matter from becoming an ATL exceeds the relevant value. Under section 60(2), the matter of reasonableness has been supplanted by the following tests: 1. In the case of deprivation of possession, the unlikelihood of recovery if the assured has been deprived of possession of the insured vessel or cargo will render the property into a CTL. 2. In the case of damage, the cost of repairing or recovering the insured property such that if the cost exceeds the repaired or recovered value, or the insured value (if so agreed), the subject-matter insured will be treated as a CTL. That is not to say that the role of the prudent assured is irrelevant, even though it may be less relevant. For example, whether or not the assured is unlikely to recover the insured property may depend on the reasonableness of the assured’s decisions or conduct in seeking the recovery of property of which the assured has been deprived. Similarly, whether the cost of repair or recovery would exceed the relevant value depends on identifying the method and cost of repair and recovery. In Suez Fortune Investments Ltd v Talbot Underwriting Ltd,59 Flaux J said: 87. In assessing the costs of repair, the approach the court should take is to ask what a prudent uninsured shipowner in the position of the claimants would have done in deciding whether or not to repair the vessel and where and how the repair should be carried out. This test was formulated in these terms by Patteson J, giving the advice of the judges of the Exchequer Chamber, adopted by the House of Lords in Irving v Manning (1847) 1 HL Cas 287, pages 306 and 307.

Thus, the reasonableness of the assured’s response to the insured property being put in a position of peril or loss or damage remains a defining feature of the doctrine of CTL. 4.3

Section 60(1): Abandonment

Section 60(1) provides that there will be a CTL where the subject-matter insured is reasonably abandoned (a) on account of its ATL appearing to be unavoidable, or (b) because the

Carras v London & Scottish Assurance Corp Ltd (1935) 52 Ll L Rep 34, 40, 42–3; Venetico Marine SA v International General Insurance Co Ltd [2014] 1 Lloyd’s Rep 349, para. 438, 466. 59 [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, para. 87. 58

Total losses under marine policies  143 subject-matter insured could not be preserved from an ATL without an expense which exceeds the value of the subject-matter insured after the expense has been incurred. A CTL under section 60(1) will exist where the subject-matter insured has been abandoned for one of the two reasons set out in the sub-section. Abandonment in this sense means giving up or surrendering the property as lost. It does not mean abandoning the property to the insurer under sections 61 and 63 of the Marine Insurance Act 1906. This is because the latter sense of abandonment refers to an option granted to the assured as a consequence of the insured subject-matter being a CTL, whereas the abandonment referred to in section 60(1) is different in that it is the reason for the insured subject-matter becoming or being treated as a CTL. In Masefield AG v Amlin Corporate Member Ltd,60 David Steel J said: 55. In my judgment these criteria are not met. In the first place the vessel and its cargo were not abandoned in the relevant sense. What is required is not a notice of abandonment in the sense of sections 61, 62 and 63 of the Marine Insurance Act but the abandonment of any hope of recovery. The distinction is spelt out in Court Line Ltd v R (1944) 78 Ll L Rep 390 in the judgment of Scott LJ at pages 395 and 396: When the ship is spoken of as ‘abandoned on account of its actual total loss appearing to be unavoidable’, the word is used in nearly the same sense as when according to the law of salvage the ship is left by master and crew in such a way as to make it a ‘derelict’, which condition confers on salvors a certain but not complete exclusiveness of possession, and a higher measure of compensation for salvage services. But to constitute the ship a ‘derelict’, it must have been left: (a) with that intention (animo derelinquendi) (The John and Jane, 4 C Rob 216); (b) with no intention of returning to her; and (c) with no hope of recovering her. Obviously that sense of the word is frequently inappropriate to the second case to which the first subsection applies, namely, because it could not be preserved from total loss (that is, an economic test) ‘without an expenditure which would exceed its value when the expenditure had been incurred’. Another distinction between those two alternative grounds in subsection (1) for claiming a constructive total loss is that in the latter case the financial estimate is one which normally would be made by the owner; whereas the forecast of the probability of actual total loss would, at any rate a century ago, nearly always have to be made by the master on the spot; and even in these days of easy and quick wireless communication, the decision would very often devolve on the master. The making of the financial estimate is, of course, merely an exercise of business judgment and discretion. The abandonment which follows after it may be expressed in a letter and not in boats as in the first alternative; or be a mere mental decision by the owner that he will exercise the option which Sect 61 allows him. There is a somewhat similar contrast between the two alternatives of subs (2)(i)(a) and (b). In Sect 61 the word ‘abandonment’ seems to import an act on the part of the assured, but in truth it amounts usually to nothing more than his making up his mind to give notice of abandonment to the insurer under Sect 62(1), at the peril of losing his right of election under Sect 61. The legal consequences of a notice of abandonment if accepted by, or established as valid against, the insurer is to pass the property to the underwriter as an abandonment to him under Sect 61. A valid ‘abandonment’ in Sect 63 necessarily means an abandonment by the assured to the insurer and passes the property to him. It cannot be the same act as is contemplated by Sect 60(1), where the act is done in consequence of an actual total loss appearing unavoidable. That abandonment, for example, by the master and crew leaving the ship with the intention of never returning, etc, may lead up to and justify a subsequent abandonment to the insurer, but the two are wholly different acts, and distinct in kind.

60 [2010] EWHC 280 (Comm); [2010] Lloyd’s Rep IR 345, paras 54–7. On appeal, the claim for a CTL was not pursued: [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630. See also Robertson v Petros M Nomikos Limited [1939] 1 AC 371, 382, 392; Court Line Ltd v The King (1945) 78 Ll L Rep 390, 396, 400 (Stable J).

144  Research handbook on marine insurance law In Court Line Ltd v The King,61 referred to by David Steel J, the Court of Appeal considered whether there had been a CTL within the meaning of a charterparty clause. For this purpose, the Court applied the definition set out in section 60 and considered the meaning of ‘abandoned’ in section 60(1). Each of the members of the Court unanimously rejected the argument that ‘abandoned’ referred to the abandonment of the insured subject-matter by the assured to the underwriters.62 Stable J adopted a similar definition of ‘abandoned’ to that of Scott LJ, but dissented on the factual application of the test. Stable J also said that the test must be applied at the time the master and crew leave the vessel. Stable J said that there must be no hope of recovery (spes recuperandi), meaning something more than a ‘bare hope’ or ‘wishful thinking’. The third member of the Court of Appeal, du Parcq LJ, considered that ‘abandoned’ must be given a more flexible meaning because, on the majority’s interpretation, the word is given two meanings when used only once in the same sub-section. Du Parcq, LJ said he understood ‘abandon’ [as used in section 60(1)] to mean ‘give up for lost’, and when I say give up for lost I mean that the owners are renouncing all their rights in the ship except the right to recover insurance […] it is not enough to show that he and the crew left the ship temporarily to her fate, or that, having left, he had grave doubt whether she would be recovered or ultimately saved. It must, I think, be made clear that he so acted as to show an intention to renounce all the owners’ […] rights in the ship, their right to property as well as to possession.

If there has been abandonment of the insured subject-matter in the sense understood by section 60(1), it must have been for one of the reasons set out in that sub-section. As to the unavoidability of an ATL, in Court Line Ltd v The King,63 Stable J, sitting in the Court of Appeal, stated that unavoidability ‘connotes a very high degree of probability with the additional element that there is no course of action, project or plan, present at the time or place in the mind of the person concerned which offers any reasonable possibility of averting the anticipated event’. Scott LJ considered the question of unavoidability from the perspective of the master and the commander of the Royal Naval sloop which had rescued the master and crew. Du Parcq LJ did so by reference to what might ‘reasonably’ have been assumed. Stable J expressly refrained from deciding whether or not the test should be applied by reference to the factual position or by reference to what appeared to the persons concerned. Stable J added that he interpreted ‘unavoidable’ not to mean ‘inevitable’ but as follows:64 It is sufficient to say that I think the word connotes a very high degree of probability with the additional element that there is no course of action, project or plan, present at the time or place in the mind of the person concerned which offers any reasonable possibility of averting the anticipated event, I venture to think there is some affinity between avoidability and the spes recuperandi referred to in the salvage cases.

As to the second of the reasons for the abandonment of the insured property in section 60(1), namely the expense of preserving the vessel from becoming an ATL exceeding the value of (1945) 78 Ll L Rep 390, 399, 400–1. Pages 394 (Scott LJ); 398–9 (du Parcq LJ, although his decision is less explicit in this respect); 400 (Stable J). See also Bradley v H Newsom, Sons & Company [1919] 1 AC 16. 63 (1945) 78 Ll L Rep 390, 401. 64 Cf. C Czarnikow Ltd v Java Sea and Fire Insurance Co Ltd [1941] 3 All ER 256, 262. 61 62

Total losses under marine policies  145 the subject-matter, this is addressed further below when considered in connection with section 60(2). 4.4

Section 60(2)(i): Deprivation of Possession

Even though the insured subject-matter is physically intact, meaning undamaged, there may yet be a CTL where the assured has been deprived of possession of the insured vessel or cargo by reason of an insured peril, if: (a) it is unlikely that it can recover the vessel or cargo; or (b) the cost of recovering the vessel or cargo would exceed its value when recovered. The deprivation of possession of the vessel or cargo means that the assured, having been in possession of the insured property, is deprived of the right to use the vessel or cargo as it chooses. Therefore, the mere fact that the assured or the vessel’s master and crew remain in physical possession of the vessel or cargo but is not, for example, permitted to remove the vessel or cargo from a particular place or port, does not necessarily mean that the assured has not been deprived of possession, because it has been deprived of the free use and disposal of the vessel or cargo.65 There may be some types of policy, such as mortgagees’ interest insurance policies, where the loss of the insured vessel is determined not by reference to the mortgagee (who is an assured) but by reference to the position of the owner of the vessel (who is not an assured under the mortgagees’ interest insurance policy); this is because the deprivation of possession must be determined from the perspective of the party in possession of the insured vessel.66 4.4.1 Unlikelihood of recovery The Marine Insurance Act 1906 is largely a codifying statute.67 However, there is one particular area where the 1906 Act altered the common law. At common law, there could be a CTL if the assured was deprived of possession of the insured subject-matter and it was uncertain – rather than unlikely – that the assured would recover possession. Section 60(2)(i), however, rendered the test of a CTL more rigorous by requiring that it was improbable or unlikely that the insured property would be recovered. In Polurrian Steamship Company v Young,68 a neutral vessel was captured by a belligerent, detained for six weeks and then released. The Court held that there would have been a CTL prior to the passing of the 1906 Act, but now there could be no such CTL, because it could not be proved that the assured was unlikely to recover possession.69 The Court of Appeal explained this difference, at first by reference to the two tests of a CTL in sections 60(1) and 60(2):

65 Fooks v Smith [1924] 2 KB 508; Polurrian Steamship Co v Young [1915] 1 KB 922; Panamanian Oriental Steamship Corp v Wright [1970] 2 Lloyd’s Rep 365; The Bamburi [1982] 1 Lloyd’s Rep 312, 317–21; Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 533–4. 66 Piraeus Bank NE v Antares Underwriting Ltd [2022] EWHC 1169 (Comm); [2022] Lloyd’s Rep IR 441, para. 231. 67 Raiffeisen Zentralbank Österreich AG v Five Star Trading LLC [2001] EWCA Civ 68; [2001] QB 825, para. 64; The Mercandian Continent [2001] EWCA Civ 1275; [2001] 2 Lloyd’s Rep 563, para. 21. See also Chalmers, ‘An Experiment in Codification’ (1886) 2 LQR 125. 68 [1915] 1 KB 922, 936–7. 69 See also Rickards v Forestal Land, Timber and Railways Company Ltd [1942] 1 AC 50, 86–7.

146  Research handbook on marine insurance law One may, I think, without disrespect, express some regret that the two expressions ‘reasonably abandoned on account of its actual total loss appearing to be unavoidable’ and ‘unlikely that he can recover the ship’ should be used apparently to describe the same position of things; for in my view, at any rate, it is one thing to predicate that a total loss of a thing reasonably appears to be unavoidable and another to predicate that its recovery is unlikely. Taking, however, the latter and, as it seems to me, the less severe test of the right to treat a capture as constituting a constructive total loss, I think that the statute has modified the pre-existing law to the disadvantage of the assured. One is always properly afraid of incompleteness in attempting a definition; but I venture to say that the test of ‘unlikelihood of recovery’ has now been substituted for ‘uncertainty of recovery.’ The assured must now show two things: the first, that he has been deprived of the possession of his ship; the second, that it is unlikely that he can recover it.

Such unlikelihood is a factual issue to be established on the balance of probabilities by reference to the evidence.70 As Rix J said in Royal Boskalis Westminster NV v Mountain:71 ‘The test is external and objective. It depends on the judgment of the reasonable man, not of the assured himself. The judgment is to be exercised on the true facts existing at the relevant time, not merely on the facts as known or appearing to the assured.’ Further, it is permissible to take account of subsequent events because they may ‘assist in showing what the probabilities really were, if they had been reasonably forecasted’.72 The requirement that it is ‘unlikely’ that the insured vessel or cargo will be recovered means that the assured has to prove that there is more than a 50 per cent chance that the vessel or cargo will not be recovered.73 The law before the Marine Insurance Act 1906 was that it merely had to be uncertain, not unlikely, that the vessel or cargo would be recovered; uncertainty meant only that there remained a real possibility that the insured property would not be recovered, even though it was likely that it would.74 The unlikelihood of recovery must be assessed on the basis that it would be unlikely to recover the vessel or cargo ‘within a reasonable time’; there is no express provision to that effect in section 60(2)(i), but that is the meaning which has been ascribed by the Court to this provision.75 Further, the likelihood of recovery must be based on the true facts as opposed to the facts as they appeared to the assured.76 Indeed, the likelihood of recovery must be assessed by reference to the steps which assured may legitimately take to recover the property; it is not a test which applies on the assumption that the assured steps back and waits for events to unfold. Thus, whether or not it is likely or unlikely that an insured vessel or cargo would be Arnould: Law of Marine Insurance and Average (20th edn, 2021), para. 29-18. [1997] LRLR 523, 534. See also Bainbridge v Neilson (1808) 10 East 329; Marstrand Fishing Co Ltd v Beer (1936) 56 Ll L Rep 163, 173. 72 Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 534. See also Bank Line Ltd v Arthur Capel & Company [1919] AC 435, 454. 73 Polurrian Steamship Co v Young [1915] 1 KB 922; Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 87; Kuwait Airways Corp v Kuwait Insurance Co SAK [1996] 1 Lloyd’s Rep 664; Piraeus Bank NE v Antares Underwriting Ltd [2022] EWHC 1169 (Comm); [2022] Lloyd’s Rep IR 441, paras 217–25. 74 Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 86–7; Kuwait Airways Corp v Kuwait Insurance Co SAK [1996] 1 Lloyd’s Rep 664, 686. 75 Polurrian Steamship Co v Young [1915] 1 KB 922, 937; Petros M Nomikos Ltd v Robertson [1939] AC 371, 383; Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 534; Clothing Management Technology Ltd v Beazley Solutions Ltd [2012] EWHC 727 (QB); [2012] 1 Lloyd’s Rep 571, para. 32. 76 Marstrand Fishing Co Ltd v Beer (1936) 56 Ll L Rep 163, 173–4; Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 534. 70 71

Total losses under marine policies  147 recovered if seized by pirates must consider the possibility that the vessel or cargo could be lawfully recovered upon the lawful payment of a ransom.77 In some cases, it may not be possible, for the time being, to determine upon the happening of an insured peril whether the insured property will or will not be recovered. In those circumstances, the Court has adopted a ‘wait and see’ approach. This approach, however, was first developed in the context of non-marine insurance where there is no applicable doctrine of CTL. There is no reason, however, why the ‘wait and see’ approach need not be applied in cases of ATL, as it was in Masefield AG v Amlin Corporate Member Ltd.78 The fact-finding exercise requires a consideration of each case,79 having regard to inter alia the intention of those taking the property, whether the property’s location is known and allowing a time within which a view may be formed on the prospects of recovery.80 In Kuwait Airways v Kuwait Insurance,81 Rix J quoted the arbitral decision of Mr Michael Kerr QC in Dawson’s Field Award (March 1972). In that case, three aircraft were hijacked in 1970 by the PFLP and the hijackers threatened to blow up the aeroplanes if certain guerrilla prisoners were not released.82 Mr Kerr QC held that the aircraft were not lost when they were hijacked, but only when they were destroyed (after the hijackers’ demands were not met). In Kuwait Airways v Kuwait Insurance, Rix J quoted from Mr Kerr QC’s Award as follows: ‘Wait and see’ is therefore to some extent always an essential ingredient of a claim for a total loss in circumstances involving deprivation of possession, unless (perhaps) there is a deprivation within the terms of specifically enumerated perils such as ‘capture’ or one can infer from the circumstances that there was a clear intention at the time of the dispossession permanently to deprive the owner of possession and ownership. This is quite different from a ‘ransom’ situation such as in the present case […] In my view, as was said by Parker J. (as he then was) in Webster v. General Accident […] every case in which there has been a dispossession must depend on its own facts as to whether and at what stage a total loss has occurred. One must consider the facts concerning the dispossession, the apparent intention of the person or persons concerned, whether or not or to what extent the whereabouts of the subject-matter are known, and allow for the lapse of a period of time to form a view about the prospects of recovery; i.e. whether the loss is total or partial […] I therefore reject the contention that these aircraft were total losses before they were blown up.

Relying on Dawson’s Field Award, Rix J distinguished the case before him from ‘ransom’ cases, ‘where “wait and see” is the order of the day’. In Scott v Copenhagen Reinsurance Co (UK) Ltd,83 Rix LJ further said: it is impossible on the judge’s findings to say that BA was irretrievably deprived of its aircraft from the first, whatever the content of that test may be. It was a ‘wait and see’ situation. Care must no doubt be taken with that expression, because it is capable of being used in two senses. In its real sense, it refers to a situation which is subject to a process of development and change. Will a ransom be paid and honoured and the property recovered? Will the property be released? That is the sense in which it was used by Mr Kerr in Dawson’s Field and again by the judge in this case […] In the present case, however, the ‘wait and see’ concept is used in its real sense. 79 80 81 82 83 77 78

Masefield AG v Amlin Corporate Member Ltd [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630. [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630. Webster v General Accident Fire & Life Assurance Corp Ltd [1953] 1 QB 520, 531–2. Dawson’s Field Award, para. 7. [1996] 1 Lloyd’s Rep 664, 688, 689. A fourth aircraft had already been destroyed in Cairo. [2003] EWCA Civ 688; [2003] Lloyd’s Rep IR 696, para. 76.

148  Research handbook on marine insurance law In Masefield AG v Amlin Corporate Member Ltd,84 Rix LJ applied this approach to a marine insurance case involving the seizure of an insured cargo by Somali pirates and the question whether there was an ATL (rather than a CTL): In the light of all this material, I conclude that, subject to Sir Sydney’s second point about the public policy of paying a ransom, piratical seizure in the circumstances of this case, where there was not only a chance, but a strong likelihood, that payment of a ransom of a comparatively small sum, relative to the value of the vessel and her cargo, would secure recovery of both, was not an actual total loss. It was not an irretrievable deprivation of property. It was a typical ‘wait and see’ situation. The facts would not even have supported a claim for a CTL, for the test of that is no longer uncertainty of recovery, but unlikelihood of recovery. That is itself recognised by the insured’s dropping of its CTL claim. There is no rule of law that capture or seizure is an ATL. The subject matter is not amenable to a rule of law at all: it is all ultimately a question of fact. The typical case of capture, by a nation’s warship, subject to condemnation as a prize, is not an ATL, although it may mature into one. Piratical seizure, in the absence of a policy of ransom, may amount to an ATL, where the pirates escape with their prize for their own use and there is no prospect whatever of finding or recovering vessel or cargo: but where a chance of recapture remains even such a seizure will not give rise to an immediate ATL, and in any event that is very far from this case.

The policy may include a provision whereby it will be deemed that it is unlikely that the assured will recover possession of the vessel if it has been deprived of the free use and disposal of the vessel for a continuous period, say, of 12 months.85 4.4.2 The cost of recovery Even if it is likely that the insured property can be recovered with reasonable effort on the part of the assured, there might still be a CTL if the cost of recovery would exceed the ‘recovered value’ of the insured property. Section 27(4) of the Marine Insurance Act 1906 provides that the agreed value is not conclusive in determining whether there has been a constructive total loss (unless the policy provides otherwise). With respect to insurance of ship, the Institute Time Clauses – Hulls (1/11/1995), clause 19.2 requires the ‘cost of recovery’ to exceed the insured value of the vessel. It should be noted, however, that clause 19.1 provides that the insured value shall be taken as the ‘repaired value’ corresponding to section 60(2)(ii) (‘value of the ship when repaired’); there is no reference to the ‘value when recovered’ (the words used in section 60(2)(i)), although the sense is reasonably clear.86 With respect to insurance of cargo, clause 13 of the Institute Cargo Clauses (A), (B) and (C) (1982 and 2009) requires the comparison to be made between the cost of recovery, reconditioning and forwarding the cargo to its destination and the value of the cargo on arrival (corresponding to section 60(2)(iii), which concerns damage to goods and not deprivation of possession). It may be that this is wide enough to deal with deprivation of possession in any

[2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630, para. 56. See, for example, the Institute War and Strikes Clauses – Hull (1/11/1995), clause 3; The Bamburi [1982] 1 Lloyd’s Rep 312; Piraeus Bank NE v Antares Underwriting Ltd [2022] EWHC 1169 (Comm); [2022] Lloyd’s Rep IR 441, paras 209–15. 86 See also the Institute Voyage Clauses – Hulls (1/11/1995), clause 17. The International Hull Clauses 2003, clause 21, provides that there will be a constructive total loss if the cost of recovery or repair exceeds 80 per cent of the insured value. 84 85

Total losses under marine policies  149 event, as clause 13 refers to the ‘cost of recovering’. In any event, as discussed above, this clause appears only to apply in the event of abandonment as understood by section 61.87 4.5

Section 60(2)(ii) and (iii): Damage

Where the subject-matter insured is a ship, and the ship has been damaged, the ship will be a CTL if the cost of repair would exceed the value of the ship when repaired (section 60(2) (ii)),88 and not the insured value (section 27(4)). However, clause 19 of the Institute Time Clauses – Hulls (1/11/1995) requires that the cost of repair exceeds the insured value of the ship.89 4.5.1 The cost of repair The cost of repair must be the reasonable cost of repair that would be incurred by a prudent (uninsured) assured. The following principles may be said to apply in determining the quantum of a reasonable cost of repair, which depends on a variety of circumstances, including the purpose of the repair, the method of repair, the choice of repair contractor, the location of the repair90 and the alternative options available. The repairs must be such as to restore the ship to its pre-damage condition.91 An expense will be included within the costs of repair if it ‘would have to be expended to put the [insured subject-matter] right’.92 Such costs include a wide range of expenditure, including, but not limited to, costs which are preliminary to repairs such as surveys and temporary repairs, the cost of transport to a repair yard, the cost of salvage and the cost of safeguarding the property.93 Although section 60(2)(ii) allows account to be taken of ‘expense of future salvage operations’, but makes no reference to past salvage operations, the costs of salvage incurred – both past and future – should be taken into account. It is permissible to take into account the costs of repair, salvaging and safeguarding the vessel incurred and indeed all of the costs of repair, salvaging and safeguarding the vessel as from the date of the casualty, even if they precede the date of a notice

Note Masefield AG v Amlin Corporate Member Ltd [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630, para. 18. 88 Irving v Manning (1847) 1 HL Cas 287. 89 See also the Institute Voyage Clauses – Hulls (1/11/1995), clause 17.2; the International Hull Clauses 2003, clause 21 (requiring the cost of repair to exceed 80 per cent of the insured value). 90 Carras v London & Scottish Assurance Corp Ltd (1935) 52 Ll L Rep 34, 42. 91 North Atlantic Steamship Co Ltd v Bure (1904) 9 Com Cas 164; Lohre v Aitchison (1878) 3 QBD 558, 563; rev’d on other grounds (1879) 4 App Cas 755; Carras v London & Scottish Assurance Corp Ltd (1935) 52 Ll L Rep 34, 42. 92 The Medina Princess [1965] 1 Lloyd’s Rep 361, 520. 93 The Medina Princess [1965] 1 Lloyd’s Rep. 361, 520; The Brilliante Virtuoso [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, para. 246; The Renos [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78, para. 19 and 24. See also Northern Barge Line v Royal Insurance Co [1974] AMC 136. 87

150  Research handbook on marine insurance law of abandonment (discussed below),94 as well as the costs of towage to the port of permanent repair and any intervening port of refuge.95 The cost of repair must be reasonable, in that the work must be necessary and the charges must not be extravagant.96 Generally, the practical way in which the Courts have calculated the cost of repair is to ask how much the reasonable cost of repair so as to put the insured subject-matter back in the state it was in before it was damaged would be, and, if the sum claimed appears to be excessive, would the Court be justified in investigating whether that sum exceeds the cost that the assured would have incurred in having the repairs carried out by a reputable contractor.97 In Stocovaz v Fung,98 the NSW Supreme Court said in a slightly different context: 37 Accordingly, I am of the opinion that a plaintiff, acting reasonably, does not have the untrammelled right to recover from a tortfeasor the cost of repairs to his property. Nor, however, in my opinion, is a defendant simply entitled to show that the work could have been carried out more cheaply and to have a deduction on that account from the amount claimed. 38 In my view, the limitation, which the authorities and principles impose, is a limitation designed only to counter extravagance or unreasonableness. The former of those words looms quite large in the cases, and Mr Alexis submitted that its use gives rise to difficulties of interpretation. However, although its meaning may have altered in the last 150 years, the first two meanings of the word ‘extravagant’ given in the Third Edition of the Macquarie Dictionary, are, ‘1. Going beyond prudence or necessity in expenditure; wasteful. 2. Excessively high, exorbitant.’ They are the connotations which, in my opinion, meet the remarks of Lord Loreburn LC in Lodge Hole Colliery and the much more recent references to the word ‘extravagant’ by Sheller JA in Hyder Consulting and Hely J in Port Kembla Coal quoted above. I see no reason why the assertion that a claim is extravagant does not create a justiciable issue.

On appeal, the NSW Court of Appeal said: 17 Once it is understood that reasonable costs may lie within a range, which may not be narrow, it seems likely that the liability of a defendant to pay something less than the actual costs of repair will turn on evidence that the repairs could have been done at a lower cost and that the plaintiff acted unreasonably in not obtaining an alternative quotation or further quotations, or in not accepting a lower quotation.

The costs of repair may include items of repair which are excluded from cover for a partial loss or particular average.99 Connect Shipping Inc v The Swedish Club (MV Renos) [2016] EWHC 1580 (Comm); [2016] Lloyd’s Rep IR 601, paras 27–47; [2018] EWCA Civ 230; [2018] 1 Lloyd’s Rep 285, paras 69–85; [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78, paras 4–19. See Arnould: Law of Marine Insurance and Average (20th edn, 2021), paras 29-36–29-37. 95 Carras v London & Scottish Assurance Corp Ltd (1935) 52 Ll L Rep 34; rev’d on other grounds [1936] 1 KB 291; Irvin v Hine [1950] 1 KB 555; The Medina Princess [1965] 1 Lloyd’s Rep 361, 520. 96 The Pactolus (1856) Swabey 173; McGregor on Damages (21st edn, 2020), para. 37-006. 97 Coles v Hetherton [2013] EWCA Civ 1704; [2015] 1 WLR 160, para. 27; Technip Saudi Arabia Ltd v Mediterranean and Gulf Cooperative Insurance and Reinsurance Co [2023] EWHC 1859 (Comm); [2023] 2 Lloyd’s Rep 371, para. 198-207; McGregor on Damages (21st edn, 2020), para. 37-006. 98 [2006] NSWSC 1345, paras 37–8; [2007] NSWCA 199, para. 17. 99 Venetico Marine SA v International General Insurance Company Ltd [2013] EWHC 3644 (Comm); [2014] Lloyd’s Rep IR 243, para. 453; Suez Fortune Investments Ltd v Talbot Underwriting Ltd 94

Total losses under marine policies  151 It is not, however, permissible to take account of SCOPIC remuneration payable to salvors in determining whether there has been a CTL. In Connect Shipping Inc v The Swedish Club (MV Renos),100 Lord Sumption explained this decision as follows: 25. The common feature of all the cases where the cost of steps preliminary to repairs have been included in the comparison is that their objective purpose was to enable the ship to be repaired. That will generally be true of salvage charges. The same goes for the cost of temporary repairs, towage, and other steps which are plainly preliminaries to carrying out permanent repairs. The objective purpose of SCOPIC charges was different. It was not to enable the ship to be repaired, but to protect an entirely distinct interest of the shipowner, namely his potential liability for environmental pollution. That purpose has nothing to do with the subject-matter insured, namely the hull. It was no part of the measure of the damage to the ship, and had nothing to do with the possibility of repairing her. The point may be tested by asking what the position would have been if the shipowner, instead of making a single agreement with salvors to salve the ship and prevent or minimise environmental damage, had contracted with one enterprise to salve the ship and another to put floating booms around her with a view to preventing or minimising environmental damage […] 26. I am prepared to assume that a prudent uninsured owner would have done what these owners did and contracted with the salvors for both the salving of the ship and protecting the environment. But I do not think that that makes any difference […] 27. The result is that it is necessary to identify the purpose of the expenditure which it is proposed to take into account, and to apply the prudent uninsured owner test only to expenditure for the purpose of repairing the ship in the larger sense which I indicated above. The fact that a prudent uninsured owner might have contracted with the same contractors for both the protection of the property and the prevention of environmental pollution does not show that both are part of the cost of repairing the damage. Neither does the fact that the charges under both heads are secured on the ship. The two heads of expenditure have quite different purposes, only one of which is related to the reinstatement of the vessel. If they were truly indivisible, this might not matter. But the whole scheme of the SCOPIC clause depends on their being separately identifiable, and the very fact that one is for the hull underwriter’s account and the other for the P&I insurers shows that they cannot be indivisible. In my opinion, SCOPIC charges are not part of the ‘cost of repairing the damage’ for the purpose of section 60(2)(ii) of the Act or the ‘cost of recovery and/or repair’ for the purpose of clause 19.2 of the Institute Clauses, because their purpose is unconnected with the damage to the hull or its hypothetical reinstatement.

The question arises, however, whether SCOPIC costs should be taken into account in estimating or calculating the cost of repair in circumstances where the owner of the insured vessel has no choice but to contract with the salvor on SCOPIC terms in order to have the vessel repaired or ‘put the ship right’. In those circumstances, there must be a case for allowing such SCOPIC costs to enter into the equation for repair costs. In calculating the cost of repair, the residual value of the damaged vessel (that is, the wreck) is not taken into account.101

[2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, paras 86–94, 221–3; PT Adidaya Energy Mandiri v MS First Capital Insurance Ltd [2022] SGHC(I) 14; [2022] 2 Lloyd’s Rep 381, paras 170–89, 218–22 (Singapore). 100 [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78, paras 25–7. 101 See Hall v Hayman [1912] 2 KB 5; Institute Time Clauses – Hulls (1/11/1995), clause 19.1; Institute Voyage Clauses – Hulls (1/11/1995), clause 17.1; the International Hull Clauses 2003, clause 21. Cf. the common law position: Macbeth & Co v Maritime Insurance Co Ltd [1908] AC 144.

152  Research handbook on marine insurance law Where there is difficulty in formulating an estimate of the cost of repair, the Court may resort to other mechanisms to identify the cost of repair, without resorting to a burden of proof. The practical difficulties faced by a shipowner in a case such as this were also taken into consideration by the Court of Appeal in Angel v Merchants Marine Insurance Co.102 Vaughan Williams LJ said:103 Precise estimates are, of course, impossible, and it seems to me that, unless the insured shipowner is to take upon himself risks which ought not to be borne by him (such as the risk whether the ship will be got afloat at all, or, having been got afloat, will arrive at a port for temporary repairs, and ultimately at home for permanent repairs), a large margin ought to be added to the figures of cost of repair to cover risks of this sort – risks which a ‘prudent uninsured owner’ would certainly take into consideration in determining whether he should repair or sell […] Now in my judgment the ‘prudent uninsured owner’ test was clearly accepted as the proper test at least down to 1873. The recognition of the test in Irving v. Manning and in Rankin v. Potter puts the matter, to my mind, beyond argument. Nor do I think that it is possible to say that Moss v. Smith, which was cited in Rankin v. Potter, had then been recognised as substituting for the ‘prudent uninsured owner’ test an arithmetical test turning on the difference between estimated totals. The prudent uninsured owner test was, I think, adopted for the very purpose of covering considerations which cannot be embodied in the figures of an arithmetical calculation.

In Suez Fortune Investments Ltd v Talbot Underwriting Ltd,104 Flaux J adapted this ‘prudent uninsured owner’ approach to cases where it is difficult to estimate the cost of repair either because it is not possible to carry out a comprehensive inspection of the insured vessel or because there is more than one place where repairs may be undertaken at differing costs. Flaux J said: 90. Some assistance as to the approach the court should adopt when faced with such difficulties is to be found in the judgment of Vaughan Williams LJ in Angel v Merchants Marine Insurance Co [1903] 1 KB 811 at pages 816 and 817 […] 92. It seems to me that the effect of this approach is that, in relation to matters which cannot be determined with precision, such as the extent of damage to items of machinery and equipment which were not opened up and tested, the court has to apply to any repair estimate what Vaughan Williams LJ describes as a ‘large margin’. That is by no means the same thing as giving the assured the benefit of the doubt in a manner which reverses the burden of proof, which is always on the assured to prove that the vessel was a constructive total loss. It is simply recognising that a margin of error has to be applied in relation to the extent of the damage where, as in the present case, it was not possible to investigate fully and the assessment of the cost of repair has to take account of the fact that the items which were not opened up and tested might well have required replacement, so that a prudent uninsured owner would have replaced them.

The place of repair may have a substantial impact on the cost of repair, as was the case in Suez Fortune Investments Ltd v Talbot Underwriting Ltd,105 where Flaux J said:

[1903] 1 KB 811, 816–17. Suez Fortune Investments Ltd v Talbot Underwriting Ltd [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, paras 89–92, 252–4; cf. Connect Shipping Inc v The Swedish Club (MV Renos) [2016] EWHC 1580 (Comm); [2016] Lloyd’s Rep IR 601, paras 85–93; [2018] EWCA Civ 230; [2018] 1 Lloyd’s Rep 285; [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78. 104 [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, paras 90–2. 105 [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, paras 93–4. 102 103

Total losses under marine policies  153 93. In relation to the place of repair, some assistance is to be found in the judgment of Porter J (as he then was) in Carras v London & Scottish Assurance Corporation Ltd (1935) 52 Ll L Rep 34 at page 42 col 1 […] 94. In my judgment when one considers that passage as a whole, contrary to the submission which Mr MacDonald Eggers QC appeared to be advancing, at least in his opening submissions, Porter J was not saying that there was some presumption in favour of the nearest port to the casualty as the port of necessity or refuge where repairs should be carried. It is, as the learned judge said, always a question of fact dependent upon all the circumstances of the case, where the prudent uninsured owner would have carried out the repairs. Whilst cost is always an important factor, it cannot necessarily be determinative, given the presence of other factors, such as are present in this case, including the need for cleaning before any long tow, the costs, time and risks of a long tow, the reputation of the rival yards, the risk of delay in those yards and the difficulties of repositioning the vessel for gainful employment after repairs have been undertaken. I return to consider all these factors in more detail in the section of the judgment dealing with the place of repair.

These considerations underlie the approach which the Court should take in estimating the costs of repair. In carrying out such an exercise, the Court must ask itself what a prudent (uninsured) assured would have done in (a) deciding whether or not to repair the insured property and (b) deciding how the repair is to be carried out.106 The cost of repair is to be calculated with reference to all of the circumstances attending the vessel at the place and time of the casualty.107 The port of necessity or the closest such port is properly to be regarded as ‘the natural place of repair’.108 4.5.2 The relevant value Where the subject-matter insured is cargo, the cargo will be a CTL where the cost of repair and of forwarding the cargo to its destination would exceed its value on arrival (section 60(2)(iii)).109 Where the subject-matter of the insurance is a ship and where the ship is damaged, the vessel is a CTL where ‘where she is so damaged by a peril insured against that the cost of repairing the damage would exceed the value of the ship when repaired’ (section 60(2)(ii)). The right to recover an indemnity for a CTL is qualified by clause 19 of the Institute Time Clauses – Hulls (1983). Clause 19 provides: 19 CONSTRUCTIVE TOTAL LOSS 19.1 In ascertaining whether the Vessel is a constructive total loss, the insured value shall be taken as the repaired value and nothing in respect of the damaged or break-up value of the Vessel or wreck shall be taken into account. 19.2 No claim for constructive total loss based upon the cost of recovery and/or repair of the Vessel shall be recoverable hereunder unless such cost would exceed the insured value.

106 Roux v Salvador (1836) 3 Bing NS 266, 420–1; Young v Turing (1841) 2 M & Gr 593, 603–4; Benson v Chapman (1843) 6 M & Gr 792, 810. 107 Arnould: Law of Marine Insurance and Average (20th edn, 2021), para. 29-42. 108 Carras v London & Scottish Assurance Corp Ltd (1935) 52 Ll L Rep 34, 42; Suez Fortune Investments Ltd v Talbot Underwriting Ltd [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, para. 94. 109 Masefield AG v Amlin Corporate Member Ltd [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep 630, para. 18. As to the common law approach to forwarding costs, see Rosetto v Gurney (1851) 11 CB 176.

154  Research handbook on marine insurance law 4.6

The Assured’s Election

In the event that there has been a CTL, whether in accordance with section 60(1) or section 60(2) or the parties’ contractual agreement, the assured has a choice to make, namely whether to (a) treat the loss as a partial loss, or (b) offer to abandon the subject-matter insured to the insurer and treat the loss as if it were an actual total loss. Section 61 of the Marine Insurance Act 1906 sets this option available to the assured in the following terms: ‘Where there is a constructive total loss the assured may either treat the loss as a partial loss, or abandon the subject-matter insured to the insurer and treat the loss as if it were an actual total loss.’110 If the assured elects to treat the loss as a partial loss, the assured is then committed to claiming an indemnity only for a partial loss and cannot thereafter claim for a CTL,111 unless the circumstances change which would give rise to an entitlement to make a fresh election.112 If, however, the assured elects to claim for a CTL, and if as a matter of fact the loss is not a CTL, the assured may nevertheless thereafter claim for a partial loss (section 56(4) of the Marine Insurance Act 1906).113 Of course, if the assured elects to treat the loss of the insured property as a total loss, and the circumstances of the case support that case, the assured is entitled to an indemnity for a CTL under the relevant marine policy. The election to claim for a CTL depends on the assured being prepared to offer to abandon to the insurer the insured property if the insurer indemnifies the assured in respect of the total loss. The abandonment to the insurer in this sense (as opposed to the sense contemplated by section 60(1)) is the transfer of title and control in respect of the insured property from the assured to the insurer. The abandonment cannot be foisted on the insurer; the insurer must itself elect to accept the abandonment in order that the cession of the assured’s interest takes place.114 However, the assured must offer to abandon the property to the insurer as a consequence of the election to claim for a CTL and must at all times be prepared to honour that offer. If, by contrast, the assured elected to claim only for a partial loss, the assured is not obliged to abandon or to offer to abandon the insured subject-matter to the insurer. It is therefore possible for the assured to revoke any election made to abandon the subject-matter insured, which may occur for example where the assured deals with the property as if the property were the assured’s own and does not deal with it for the account of the insurer or for the joint benefit of the insurer (who may accept the abandonment) and the assured (who remains the owner of the property pending the insurer’s own election to accept the abandonment).115

110 The WD Fairway [2009] EWHC 889 (Admlty); [2009] 2 Lloyd’s Rep 191, para. 31; Involnert Management Inc v Aprilgrange Ltd [2015] EWHC 2225 (Comm); [2015] 2 Lloyd’s Rep 289, para. 259. 111 Fleming v Smith (1848) 1 HL Cas 513. 112 Stringer v The English and Scottish Marine Insurance Company (1869) LR 4 QB 676, 686–7; (1870) LR 5 QB 599, 602–6. 113 Bank of America National Trust and Savings Association v Crismas (The Kyriaki) [1993] 1 Lloyd’s Rep 137, 150–2; Kastor Navigation Co Ltd v AGF MAT [2002] EWHC 2601 (Comm); [2003] 1 Lloyd’s Rep 296, para. 21; [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119. 114 The WD Fairway [2009] EWHC 889 (Admlty); [2009] 2 Lloyd’s Rep 191, paras 25–9. 115 Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 555–9; Suez Fortune Investments Ltd v Talbot Underwriting Ltd [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, paras 255–8; Connect Shipping Inc v The Swedish Club (MV Renos) [2016] EWHC 1580 (Comm); [2016] Lloyd’s

Total losses under marine policies  155 The election to claim for a CTL is generally effected, and must in most cases be effected, by the tender by the assured of a notice of abandonment. 4.7

The Notice of Abandonment

If the assured wishes to claim for a CTL, in order to make that election the assured must give a notice of abandonment (‘NOA’) to the insurer, thus signifying its election to claim for a CTL rather than a partial loss. Section 62(1) of the Marine Insurance Act 1906 provides that ‘where the assured elects to abandon the subject-matter insured to the insurer, he must give notice of abandonment’.116 The mere fact that the assured chooses to claim for a CTL will not prevent the assured claiming alternatively for a partial loss if the evidence does not establish that there was in fact a CTL (section 56(4)). The purpose of the NOA is to enable the insurer to accept the abandonment if it chooses to do so.117 The requirement of an NOA exists wholly for the benefit of the insurer. Its purpose is to enable the insurer to exercise the rights which arise in its favour upon an effective abandonment.118 The NOA (if required) must be given ‘with reasonable diligence after the receipt of reliable information of the loss’. However, ‘where the information is of a doubtful character the assured is entitled to a reasonable time to make inquiry’ (section 62(3) of the Marine Insurance Act 1906). What may constitute a reasonable time is a question of fact (section 88 of the Marine Insurance Act 1906).119 It may be that an assured will be in a reasonable position to conclude that there is a CTL soon after the insured peril operates. In other cases, the assured may have to ‘wait and see’ (as discussed above). For example, if the insured vessel or cargo is taken by pirates, it may be that the insured property can be recovered by the payment of a negotiated ransom less than the value of the insured property. Similarly, if the insured property is damaged, it may not be possible to determine that the estimated cost of repair would exceed the relevant value of the property as repaired until the process of carrying out an inspection of the damage, the drawing up of a repair specification and the obtaining of tenders and quotations has been completed, which may be many months after the casualty.

Rep IR 601, paras 6–7; [2018] EWCA Civ 230; [2018] 1 Lloyd’s Rep 285; [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78. 116 As to whether an NOA is required for a loss of freight, see Trinder, Anderson & Co v Thames and Mersey Marine Insurance Co [1898] 2 QB 114, 119–22; Associated Oil Carriers Ltd v Union Insurance Society of Canton Ltd [1917] 2 KB 184. 117 Involnert Management Inc v Aprilgrange Ltd [2015] EWHC 2225 (Comm); [2015] 2 Lloyd’s Rep 289, para. 265. 118 Connect Shipping Inc v The Swedish Club (MV Renos) [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78, para. 14. 119 Anderson v Royal Exchange Assurance Co (1805) 7 East 38; MR Currie & Co v Bombay Native Insurance Co (1869) LR 3 PC 72, 79; Kaltenbach v Mackenzie (1878) 3 CPD 467, 471–5; Involnert Management Inc v Aprilgrange Ltd [2015] EWHC 2225 (Comm); [2015] 2 Lloyd’s Rep 289, para. 261–4; Connect Shipping Inc v The Swedish Club (MV Renos) [2016] EWHC 1580 (Comm); [2016] Lloyd’s Rep IR 601, paras 8–26; [2018] EWCA Civ 230; [2018] 1 Lloyd’s Rep 285, paras 39–68; [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78; PT Adidaya Energy Mandiri v MS First Capital Insurance Ltd [2022] SGHC(I) 14; [2022] 2 Lloyd’s Rep 381, paras 195–7 (Singapore).

156  Research handbook on marine insurance law If the assured fails to tender an NOA where one is required, by section 62(1), the assured will be treated as claiming only for a partial loss and will not be permitted to claim for a CTL, unless the circumstances later change supporting a fresh claim for a CTL.120 The assured is also likely to lose the right to claim for a CTL, where an NOA is tendered but is not tendered in time in accordance with section 62(3). The NOA is not an essential element of the assured’s cause of action to be indemnified in respect of a CTL under the marine policy, which accrues on the occurrence of the casualty. Rather, it is merely a condition to be satisfied if the assured wishes to claim for a CTL. That is, it is a procedural condition, not a substantive condition.121 Further, the NOA is not a condition precedent to the existence of a CTL, but only to the right to claim for a CTL.122 There is no equivalent requirement for an NOA in respect of a claim for an ATL (section 57(2) of the Marine Insurance Act 1906). Furthermore, as discussed below, there is no requirement of an NOA in respect of a claim for a CTL, either where the insurer waives the requirement or whether there is no possibility of benefit to the insurer if the NOA were tendered (sections 62(7) and (8) of the Marine Insurance Act 1906). The importance attached to the NOA such that its not being tendered deprives the assured of a right to claim for a CTL is odd, given that even in the case of an ATL, there may be value attached to what remains of the subject-matter insured, and given that upon the payment of a total loss indemnity, by section 79(1) of the Marine Insurance Act 1906, the insurer is entitled to take over whatever remains of the insured property. The only possible reason why the insurer might be prejudiced by the assured’s failure to tender an NOA, which is otherwise required, would be if the insurer wished to act quickly to secure its rights or to preserve the property without delay and without having to pay the CTL indemnity in the first instance. Nevertheless, it is almost always the case that the insurer does not accept the NOA and so one wonders why the insurer would be prejudiced by the failure to tender or the late tender of an NOA. There is no particular form required for a valid NOA.123 The only requirement is that by the NOA the assured indicates its intention to abandon its insured interest in the subject-matter insured unconditionally to the insurer (section 62(2) of the Marine Insurance Act 1906). Further, the NOA should provide sufficient information to enable the insurer to make proper enquiries.124 The NOA is not a ‘matter of the mind’, that is a merely subjective, internal intention to abandon is not sufficient for the purposes of tendering an NOA; a physical manifestation of that intention must be communicated to the insurer.125 By section 62(6) of the Marine Insurance Act 1906, if the insurer accepts the abandonment offered by the NOA, the insurer is taken to admit conclusively liability for the CTL under

120 Stringer v The English and Scottish Marine Insurance Company (1869) LR 4 QB 676, 686–7; (1870) LR 5 QB 599, 602–6; Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, para. 86. 121 Petros M Nomikos Ltd v Robertson [1939] AC 371, 377, 381; Bank of America National Trust and Savings Association v Crismas (The Kyriaki) [1993] 1 Lloyd’s Rep 137, 150–2. 122 Connect Shipping Inc v The Swedish Club (MV Renos) [2019] UKSC 29; [2019] 2 Lloyd’s Rep 78, para. 14. 123 George Cohen, Sons & Co v Standard Marine Insurance Co Ltd (1925) 21 Ll L Rep 30; Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 554. 124 Knight v Faith (1850) 15 QB 649, 659. 125 Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, paras 77–9.

Total losses under marine policies  157 the marine policy, and that the NOA was sufficient within the requirements of section 62. Nevertheless, the insurer may still be able to argue that the insurance contract is voidable for non-disclosure or misrepresentation126 or that the loss was caused by an excluded or non-covered peril if the NOA was accepted under a mistake of fact.127 By section 62(5), the insurer may accept the abandonment expressly or impliedly from the conduct of the insurer; however, mere silence will not be taken as an acceptance. Equally, the insurer may refuse to accept the abandonment. In that event, the assured will not be prejudiced by that refusal (section 62(4) of the Marine Insurance Act 1906). While the abandonment remains unaccepted, the assured may revoke the NOA.128 In summary, the position is that the tender of an NOA is an executory offer to abandon the insured property to the underwriters and constitutes a claim for an indemnity for a CTL (as opposed to a claim for a partial loss). If the offer is accepted, the abandonment to the insurer takes place and is irrevocable (section 62(6)). If the offer is declined by the insurer, the offer remains open (unlike a contractual offer which when rejected is no longer open for acceptance). Indeed, the continuing offer is a condition precedent to the making of a claim for a CTL, at least until the insurer admits liability for the claim. Accordingly, if the assured revokes or withdraws the offer, the claim for a CTL also fails, leaving the assured only with its claim for a partial loss (assuming there is one). The law was explained by Rix J in Royal Boskalis Westminster NV v Mountain.129 In that case, a dredging fleet had been working in Iraq pursuant to a contract with the Iraqi authorities at the time of the Iraqi invasion of Kuwait in August 1990. Soon afterwards, the Iraqi government passed a resolution seizing all foreign property in Iraq. In September 1990, the assured tendered an NOA, which was declined, the insurers agreeing to put the assured in the same position as if proceedings were commenced that day. Thereafter, the assured and the Iraqi authorities negotiated for the release of the assured’s personnel and the insured equipment. In December 1990, the assured agreed to waive various contractual claims against, and to pay funds to, the Iraqi authorities in exchange for the release of the personnel and equipment, which duly took place. In January 1991, the UN-sanctioned war against Iraq commenced. The Court held, for various reasons, that there had been no CTL at the time of the NOA. Nevertheless, the Court also considered the insurers’ argument that in any event the NOA had been revoked both expressly (because the assured had pressed its sue and labour claim on the basis that a total loss claim had been saved) and by conduct (because the assured used the vessels in their businesses and had sold two of the more valuable vessels and dismantled a third). Rix J concluded that the assured had revoked its abandonment and could not therefore claim for a total loss. The judge considered the nature and effect of an NOA and then made the following comments as regards the issue of revocation: (a) the assured may revoke the NOA; (b) the revocation may take place expressly; (c) the revocation may take place by conduct, by the assured dealing with the property as if it were the assured’s own; (d) the assured, or other person, may deal with the insured subject-matter for the benefit of both insurers and the

Fraser Shipping Ltd v Colton (The Shakir III) [1997] 1 Lloyd’s Rep 586. Norwich Union Fire Insurance Society v Wm H Price Ltd [1934] AC 455. 128 Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 555–8; Kastor Navigation Co Ltd v AGF MAT [2002] EWHC 2601 (Comm); [2003] 1 Lloyd’s Rep 296, para. 21; [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119. 129 [1997] LRLR 523, 555–9. 126 127

158  Research handbook on marine insurance law assured without revoking the NOA,130 but if the master or agent deals with the property for the assured’s own interests, there will be a revocation. In Royal Boskalis v Mountain, the insured vessels were sold after their return from Iraq and, as the judge found, for the benefit of the assured’s business. The test of revocation appears to depend on whether the relevant conduct was undertaken for the assured’s own account alone and whether the assured has acted consistently with its claim for a total loss. Equally, the assured might waive the right to claim for a CTL in advance of the tender of an NOA in that the assured may act in such a way which is inconsistent with the property being abandoned to the insurer.131 Whether the waiver or withdrawal of the abandonment or intended abandonment of the insured property takes place after or before the tender of an NOA, as Rix J said in Royal Boskalis v Mountain, ‘similar principles must arise’ (p.558). The revocation of an abandonment or intended abandonment should, in principle, apply in respect of both CTLs and ATLs, even though Rix J said that in respect of the possibility of an NOA being revoked, ‘a CTL claim is different from that of an actual total loss’.132 Even though the general rule is that an NOA must be tendered as a condition precedent to the making of a valid CTL claim under a marine policy, by section 62 an NOA is excused where: 1. It has been waived by the insurer (section 62(8)).133 2. At the time the assured receives information of the loss, there would be no possibility of benefit to the insurer if notice were given to it (section 62(7)).134 If the assured is excused from tendering an NOA, the assured may still be deprived of the right to claim a CTL if by its words or conduct it has manifested an intention to claim only for a partial loss.135 As notices of abandonment (which are provided for in section 62) must be distinguished from the concept of abandonment referred to in section 60(1), it must also be distinguished from the abandonment itself. The NOA is the offer made by the assured to abandon the insured property to the insurer in the event that the assured elects to claim for a CTL. It is a pre-condition of the assured’s entitlement to claim for a CTL. An abandonment, on the other 130 See, for example, Brown v Smith (1813) 1 Dow 349, relied on by Rix J where there had been no waiver because what was done was consistent with the duty to sue and labour (p.358). See also Suez Fortune Investments Ltd v Talbot Underwriting Ltd [2015] EWHC 42 (Comm); [2015] 1 Lloyd’s Rep 651, paras 255–8. 131 Fleming v Smith (1848) 1 HLC 513, 530–1. 132 See also Arnould: Law of Marine Insurance and Average (20th edn, 2021), para. 28-04. 133 Involnert Management Inc v Aprilgrange Ltd [2015] EWHC 2225 (Comm); [2015] 2 Lloyd’s Rep 289, paras 265–72; PT Adidaya Energy Mandiri v MS First Capital Insurance Ltd [2022] SGHC(I) 14; [2022] 2 Lloyd’s Rep 381, paras 201–17 (Singapore). 134 See, for example, Trinder, Anderson & Co v Thames and Mersey Marine Insurance Co [1898] 2 QB 114; Kastor Navigation Co Ltd v AGF MAT [2002] EWHC 2601 (Comm); [2003] 1 Lloyd’s Rep 296, paras 15–23; [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, para. 65; Clothing Management Technology Ltd v Beazley Solutions Ltd [2012] EWHC 727 (QB); [2012] 1 Lloyd’s Rep 571, paras 37–40; Involnert Management Inc v Aprilgrange Ltd [2015] EWHC 2225 (Comm); [2015] 2 Lloyd’s Rep 289, paras 265–72; PT Adidaya Energy Mandiri v MS First Capital Insurance Ltd [2022] SGHC(I) 14; [2022] 2 Lloyd’s Rep 381, paras 198–200 (Singapore). 135 Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, paras 87–90.

Total losses under marine policies  159 hand, is ‘a cession or transfer of the ship to the underwriter, and of all his property and interest in it, with all the claims that may arise from its ownership, and all the profits that may arise from it, including the freight then being earned’.136 By section 63 of the Marine Insurance Act 1906, where there is a valid abandonment the insurer is entitled to take over the interest of the assured in whatever may remain of the subject-matter insured, and all proprietary rights incidental thereto.137 An abandonment can take place where there has been either an ATL or a CTL.

5.

THE TIMING OF A TOTAL LOSS

Most marine policies do not explain whether the loss or the insured peril must occur during the period of the policy. In Promet Engineering (Singapore) Pte Ltd v Sturge (The Nukila),138 Hobhouse, LJ said that, in the case of marine (hull) policies, it is the loss, not the insured peril, which must occur during the policy period; however, that was a case where the relevant peril was a latent defect in the hull or machinery of the insured vessel, which in most cases exists prior to the commencement of the policy period. In many cases, the occurrence of an insured peril and the resultant loss occur at the same time or in close proximity to each other. There are occasions, however, when the total loss might not develop until a considerable time has elapsed after the occurrence of the insured peril. In such cases, the Courts have developed a principle that – even if the loss must occur during the policy period – it would be sufficient to establish cover under the marine policy if the subject-matter of the insurance received its ‘death blow’ before the marine policy expired, meaning that the insured peril must have occurred during the policy period and the resultant total loss must have developed as the result of a sequence of events following in the ordinary course upon the occurrence of the insured peril, even if the total loss does not materialise until after the expiry of the marine policy.139 Whether or not there has been an ATL or CTL must be assessed as at the date of the commencement of legal proceedings against the marine insurer.140 In the case of a claim for an indemnity for a CTL, not only must the CTL exist as at the date of the commencement of the legal proceedings against the insurer, it must also exist as at the date of the tender of the NOA.141 However, a practice has developed that when a NOA is tendered in support of a CTL claim Rankin v Potter (1873) LR 6 HL 83, 144; Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, paras 52–3, 76, 88. 137 Bennett, The Law of Marine Insurance (2nd edn, 2006), paras 22.33–22.34. 138 [1997] 2 Lloyd’s Rep 146, 151; cf. Kelly v Norwich Union Fire Insurance Ltd [1990] 1 WLR 139. 139 Knight v Faith (1850) 15 QB 649; Kuwait Airways Corp v Kuwait Insurance Co SAK [1996] 1 Lloyd’s Rep 664, 690; Bayview Motors Ltd v Mitsui Marine and Fire Insurance Co Ltd [2002] EWHC 21 (Comm); [2002] 1 Lloyd’s Rep 652, para. 28; [2002] EWCA Civ 1605; [2003] 1 Lloyd’s Rep 131, para. 22, 26). 140 Tunno v Edwards (1810) 12 East 488; Goldsmid v Gillies (1813) 4 Taunt 803; Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 98; Panamanian Oriental Steamship Corporation v Wright [1970] 2 Lloyd’s Rep 365, 383; Arnould: Law of Marine Insurance and Average (20th edn, 2021), para. 28-04. 141 Pesquerias y Secaderos de Bacalao de Espana SA v Beer (1946) 79 Ll L Rep 417, 433; Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 534; cf. Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 84–5. 136

160  Research handbook on marine insurance law and the NOA is declined by the underwriter, it will be agreed between the parties that they will be in the same position as if proceedings were commenced on the date on which the NOA was tendered or declined so that the assured need not commence legal proceedings to support the CTL claim. In that event, depending on the terms of the agreement, whether or not there is an ATL – as opposed to a CTL – may be tested at the agreed date, instead of the date of the actual commencement of legal proceedings.142 It has also been said that this assessment is to be based on the true facts and not merely those facts which were known.143 That said, the test of a CTL by abandonment in section 60(1) depends on an ATL ‘appearing’ to be unavoidable, which would suggest that the test might reasonably be applied to the facts as were known or could reasonably be known.144 If the insured subject-matter is a CTL as at the date of the commencement of proceedings and if the subject-matter of the insurance is restored or becomes salvageable after that date, there remains an entitlement to claim for an indemnity for a total loss notwithstanding. In that event, upon the recovery of the insured property, the insurer upon indemnifying the assured is entitled to the abandonment of the property pursuant to sections 63 and/or 79(1) of the Marine Insurance Act 1906.145 Where the insured property is a CTL and the insurer gratuitously intervenes to recover or restore the vessel before the date of commencement of proceedings, so as to reduce the cost of repair or recovery of the vessel, the assured may still claim for a CTL.146

6.

SUCCESSIVE TOTAL LOSSES

Just as a partial loss may develop into a CTL, a CTL may develop into an ATL. In that latter event, the question arises whether or not the succession of such total losses means that the assured is prevented from recovering in respect of the CTL or the ATL. There appears to be no doctrine of merger in such cases (as there would be if there were an unrepaired damage partial loss followed by a total loss in accordance with section 77(2) of the Marine Insurance Act 1906). Therefore, in those circumstances, the assured could claim for the subsequent ATL.147 Equally, the assured would be entitled to recover in respect of the earlier CTL and not claim in respect of the later ATL.148 Of course, if the assured does not tender an NOA (and should have done so) and so elects to claim for a partial loss in respect of the prior CTL, the occurrence of

See Masefield AG v Amlin Corporate Member Ltd [2011] EWCA Civ 24; [2011] 1 Lloyd’s Rep

142

630.

Marstrand Fishing Co Ltd v Beer (1936) 56 Ll L Rep 163, 173; C Czarnikow Ltd v Java Sea and Fire Insurance Co Ltd [1941] 3 All ER 256, 262. 144 Cf. The Lavington Court [1944] 2 All ER 249, 253; Court Line Ltd v The King (1945) 78 Ll L Rep 390. 145 Ruys v Royal Exchange Assurance Corp [1897] 2 QB 135; Polurrian Steamship Co v Young [1915] 1 KB 922; Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 555–7. 146 Sailing Ship ‘Blairmore’ Co v Macredie [1898] AC 593. 147 Woodside v Globe Marine Insurance Co Ltd [1896] 1 QB 105; Rickards v Forestal Land, Timber and Railways Co Ltd [1942] AC 50, 88–9. 148 Kastor Navigation Co Ltd v AGF MAT [2002] EWHC 2601 (Comm); [2003] 1 Lloyd’s Rep 296, paras 20–3; [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, paras 91, 111–17. 143

Total losses under marine policies  161 the later ATL will result in the merger of the earlier partial loss assuming it concerned unrepaired damage in accordance with section 77.149

7.

RIGHTS UPON PAYMENT OF TOTAL LOSS

Section 63 of the Marine Insurance Act 1906 provides that where there is a valid abandonment the insurer is entitled to take over the interest of the assured in whatever may remain of the subject-matter insured, and all proprietary rights incidental thereto. That entitlement arises only if the insurer pays an indemnity for a total loss (consistently with section 79(1)) and if the insurer elects to accept the abandonment.150 Upon the abandonment of an insured vessel, the insurer is entitled to the benefit of ownership of the vessel but also must bear the liabilities and costs associated with such ownership. In this latter respect, the abandonment of the insured property might constitute a ‘damnosa hereditas’.151 Abandonment therefore involves the divesting of the assured’s interest in the remains of the subject-matter insured in favour of the insurer from the time of the casualty in the event of the insurer’s liability for a total loss. This can take place only if the insurer so chooses.152 It might be thought that if there has been an actual total loss of the subject-matter insured, there can be no abandonment because there is nothing left to abandon to the insurer. Accordingly, in cases of ATL (as opposed to CTL), there is no requirement that an NOA must be given (section 57(2)).153 Even in cases of ATL, however, the concept of abandonment (as opposed to the NOA) is applicable.154 There may be some value in the remaining property which may be abandoned to the insurer.155 By section 79(1) of the Marine Insurance Act 1906,156 pursuant to its right of subrogation,157 in the event of a total loss, upon payment of the claim, the insurer is entitled to take over the assured’s interest in the subject-matter insured as from the date of the casualty. Once that election is made, the underwriters acquire a proprietary interest in the vessel, which will in

149 Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, para. 69, 86. As to clause 18.2 of the Institute Time Clauses – Hulls 1995, see Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, paras 104–10. See also the International Hull Clauses 2003, clause 20.2. 150 Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, para. 76. 151 Allgemeine Versicherungs-Gesellschaft Helvetia v Administrator of German Property [1931] 1 KB 672, 688. 152 Royal Boskalis Westminster NV v Mountain [1997] LRLR 523, 557. 153 Involnert Management Inc v Aprilgrange Ltd [2015] EWHC 2225 (Comm); [2015] 2 Lloyd’s Rep 289, para. 259. 154 Kastor Navigation Co Ltd v Axa Global Risks (UK) Ltd [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, para. 88. 155 Rankin v Potter (1873) LR 6 HL 83, 156–7; Kastor Navigation Co Ltd v Axa Global Risks (UK) Ltd [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, para. 93. 156 Even though section 63 does not refer to the requirement of payment, it has been held that such rights of abandonment depend on the insurer having indemnified the assured for the total loss. See Kastor Navigation Co Ltd v Axa Global Risks (UK) Ltd [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119. 157 The right to take over the insured property is provided for in section 79(1) under the heading ‘Right of Subrogation’.

162  Research handbook on marine insurance law the first instance be constituted by an equitable lien and will include legal title once all legal formalities, if any, have been complied with.158 This principle is often referred to as the principle of ‘salvage’. It is very similar to the doctrine of abandonment (that is, ‘a cession or transfer of the ship to the underwriter’), which occurs upon the insurer’s agreement to accept a claim for a total loss and to accept the abandonment.159 The purpose of this principle is to ensure that the assured is not over-indemnified. If the assured has been paid for a total loss, it follows that it has received in money terms the value of the asset insured so that any residual interest in that asset might operate to put the assured in a better position than it had been in before the casualty. In the circumstances, such salvage is designed to ensure that the assured is left with nothing more than the indemnity for a total loss.160 In the case of a total loss claim, it may be agreed between the parties that the assured will credit the insurer the residual value of the insured property against the total loss indemnity to be paid by the insurer. In that event, the insurer will often waive its right to take over insured property upon its recovery. There is no provision in the Marine Insurance Act 1906 which entitles the indemnifying the insurer to such a credit;161 instead, the insurer is entitled only to take over the assured’s interest in the property upon its recovery, which would render the insurer entitled to the income earned by the use of the property and responsible for any associated liabilities.162

8. CONCLUSION The law of total losses under marine insurance policies has been developed at common law and been enshrined in the Marine Insurance Act 1906. The law has been developed with a remarkable degree of maturity and stability to develop solutions which take account the respective commercial positions of the insurer and the assured. That said, there remain challenging issues which still require analysis and consideration.

The WD Fairway [2009] EWHC 889 (Admlty); [2009] 2 Lloyd’s Rep 191. Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277; [2004] 2 Lloyd’s Rep 119, para.

158 159

76.

Castellain v Preston (1883) 11 QBD 380, 386–7. The WD Fairway [2009] EWHC 889 (Admlty); [2009] 2 Lloyd’s Rep 191, para. 30 (in this case there was a concession that the insurer was entitled to such credit). 162 Dornoch Limited v Westminster International BV [2009] EWHC 1782; [2010] Lloyd’s Rep IR 1. 160 161

8. Forwarding cargo to destination: a review of Rules F and G, York–Antwerp Rules Richard Sarll1

1. INTRODUCTION ‘We so rarely have to consider the law of general average that it is as well to remind ourselves of it’, wrote Lord Denning MR.2 At root, general average (GA) involves a system of risk transfer which pre-dates marine insurance and exists separately from it, but is nevertheless closely linked with it, since usual marine policies provide cover against general average losses. According to one classic definition, the founding principle is as follows: ‘All loss which arises in consequence of extraordinary sacrifices made or expenses incurred for the preservation of the ship and cargo come within general average, and must be borne proportionately by all who are interested. Natural justice requires this.’3 But while that statement may explain why property interests and their insurers contribute rateably to, for example, ransom payments to Somali pirates, it does not however account for why contributions are routinely paid towards expenses that are incurred when cargo is transhipped onto a substitute vessel and forwarded to destination. As will be explained below, the admissibility of this type of expense is a creature of the York–Antwerp Rules, a law reform project seeking the international harmonisation of general average law and practice that began in the Victorian era, and is still continuing.4 Two provisions, in particular, are in issue: Rules F and G. In the specific context of forwarding cargo to destination, these rules throw up numerous problems of interpretation which it is the purpose of this chapter to examine. In view of those difficulties, and also of the hostility often associated with the marine cargo insurance community towards claims by shipowners in general average, it is perhaps surprising that this important topic has been the subject of relatively little forensic attention.5

This article is dedicated to Charlotte Warr in gratitude for her generous tutoring of students preparing for examinations set by the Association of Average Adjusters. 2 Australian Coastal Shipping v Green [1971] QB 456 at 478. 3 Birkley v Presgrave (1801) 1 East 220, 228–9. 4 The first edition of the rules followed a conference in 1877. The most recent edition is YAR 2016. 5 One reason may be as follows. If the expenses incurred in hiring a substitute vessel are not allowable in general average with the result that others must contribute to them, an alternative analysis might involve their being treated as special charges on cargo with the result that cargo interests must bear the entire cost themselves. This is certainly a point to be cautious of. But the necessary ingredients for special charges would still need to be fulfilled: see further N.G. Hudson, Special Charges on Cargo. Part 1: The Obligations of Ship and Cargo Owners [1981] LMCLQ 315. 1

163

164  Research handbook on marine insurance law

2.

FORWARDING EXPENSES IN GENERAL AVERAGE

Forwarding expenses arise when cargo is transhipped onto a substitute vessel for carriage to destination, rather than being carried there aboard the vessel onto which it was originally loaded. Numerous reasons exist why the original voyage may be broken up in this way. For instance, a containership may suffer a fire due to the ignition of dangerous cargo and deviate to a port of refuge where undamaged cargo is discharged and transhipped onto another containership plying the same line. Or else a bulker may become stranded and require lightening by salvors, with the lightered cargo then being forwarded to destination in a substitute bulk carrier engaged for this purpose. On such occasions, issues arise as to whether the expenses incurred in forwarding cargo to destination are allowable in general average. Further issues arise in connection with port of refuge expenses, such as crew wages and the cost of bunkers, that are incurred by the original vessel after the cargo has been transhipped (that is, Non-Separation Agreement, or ‘NSA’, expenses) and concern the question whether cargo interests must continue to contribute to such costs, even though there is no longer any prospect of the original laden voyage being prosecuted. For reasons which will be explored in this article, the sort of forwarding and detention expenses that have just been considered are potentially admissible in general average under the York–Antwerp Rules, with the result that other parties to the common maritime adventure must contribute to them rateably according to their respective contributory values. The liability to contribute will then attract an indemnity under standard forms of marine cargo insurance and hull and machinery (H&M) insurance. Intuitively, such outcomes may strike some readers as heterodox: why ever should interested parties contribute to forwarding expenses, when these do not serve the preservation either of the property at risk or the adventure? Why ever should cargo interests contribute to ongoing port of refuge expenses, once their cargo is long gone and may even already have arrived at its destination? These and other such conundrums will be explored below. One point to note at the outset is that, even if such expenses are potentially admissible, numerous qualifications exist, none of which is straightforward.

3.

LAW OF CONTRACTS OF CARRIAGE – DOCTRINE OF FRUSTRATION

Before examining relevant principles in the field of general average, it is, for at least two reasons, important to begin by explaining the treatment of transhipment in the law of contracts of carriage. First, this is because the limits to admissibility of expenses in general average are, as will be seen, expressed by reference to concepts applicable to contracts of carriage, such as abandonment of the voyage. Second, it is only once the ordinary principles of the law of charterparties and bills of lading have been considered that it becomes apparent just how remarkable the modern law and practice of general average truly is. In case a vessel should be stricken by some casualty and requires repairs in order for the contractual voyage to be completed, a shipowner will ordinarily owe a duty to its counterparty to carry out those repairs and continue with the voyage, or else it will be held in breach of

A review of Rules F and G, York–Antwerp Rules  165 contract and be liable to compensate its charterer in damages.6 Leaving aside the application of contractual exemption clauses such as perils of the sea or navigational error, this unforgiving rule will hold good, unless the shipowner is able to show that the contract is frustrated. In that case, both parties will automatically be discharged from any further contractual duties. If, exceptionally, the contract is frustrated, the shipowner will therefore be excused from carrying out the necessary repairs. The shipowner may, in those circumstances, require the owners of cargo to retrieve it.7 Alternatively, a second freight may be agreed under a new contract enabling the goods to be carried to destination, typically aboard some other vessel that the shipowner has arranged for this purpose. In order to be excused from repairing the vessel and continuing the voyage, the shipowner must fulfil the onerous test for frustration of contract. (Whereas earlier cases speak of ‘abandonment of the voyage’ as though there were some special doctrine akin to frustration that is peculiar to contracts of carriage, it has been held that those cases are properly to be treated as instances of frustration.8) The classic test for frustration is and remains the one explained by Lord Radcliffe in Davis Contractors Ltd v Fareham UDC.9 The enquiry is whether the circumstances in which performance is called for are ‘radically different’ from those originally envisaged. Where charterparties and bills of lading are involved, the two most common situations in which frustration occurs are where repairs are required that would cost more than the value of the ship when repaired, and where there is frustrating delay. As for the first of those instances, it is clear that the repairs in issue are those which are needed to complete the contractual voyage. As such, they need not necessarily restore the vessel’s condition entirely, but could consist simply of temporary repairs.10 For these reasons, it is incorrect to equate the test of whether a vessel is a constructive total loss under its H&M insurance with the separate and distinct test which applies to frustration. As for frustrating delay, it is a notoriously difficult question whether the prospective delay is of sufficient duration as to produce frustration of the contract. One reported decision which is often referred to on such occasions is The Fjord Wind.11 In that case, a vessel suffered a catastrophic engine failure shortly after departing Rosario, bound for ports in Europe laden with a cargo of soya bean meal. Given the need to effect repairs, it was predicted that the delay would last ‘well over three months’.12 Such a delay was held sufficient to give rise to a repudiatory breach of contract which the charterer could lawfully accept as terminating the contract. On that occasion, the delay was due to the fault of the shipowner. Supposing, however, that the cause had instead been an innocent one, the very same period of delay would have sufficed to

Kulukundis v Norwich Union Fire Insurance Society [1937] 1 K.B. 1 at 16. See for example Helmsville, Ltd v Yorkshire Insurance Company, Ltd (The Medina Princess) [1965] 1 Lloyd’s Rep. 361 at 522. 8 Bunge SA v Kyla Shipping Co Ltd (The Kyla) [2013] 1 Lloyd’s Rep. 565 at [43] et seq. 9 Davis Contractors Ltd v Fareham UDC [1956] A.C. 265. 10 Kulukundis v Norwich Union Fire Insurance Society [1937] 1 K.B. 1. 11 Eridania S.p.A & others v Rudolf A. Oektker & others (The Fjord Wind) [1999] 1 Lloyd’s Rep. 307 (Comm. Ct); [2000] 2 Lloyd’s Rep. 191 (CA). 12 Eridania S.p.A & others v Rudolf A. Oektker & others (The Fjord Wind) [1999] 1 Lloyd’s Rep. 307 at 329. 6 7

166  Research handbook on marine insurance law bring about the frustration of the contract.13 As such, the decision in The Fjord Wind provides some indication of the length of delay necessary to bring about frustration in a typical case.14 In sum, purely as a matter of the ordinary15 law of contracts of carriage and leaving aside any consideration of general average, shipowners must carry cargoes to destination even if this may have become very onerous for them financially. It is only if the circumstances have become radically different that the contract will thenceforth be discharged, in which case a new contract might be agreed in consideration for the payment of a second freight.

4.

LAW OF GENERAL AVERAGE – PORT OF REFUGE EXPENSES

Under English law, as codified by section 66(2), Marine Insurance Act 1906, there ‘is a general average act where any extraordinary sacrifice or expenditure is voluntarily and reasonably made or incurred in time of peril for the purpose of preserving the property imperilled in the common adventure’. (Successive editions of the York–Antwerp Rules have, as from 1924, contained similar language in Rule A.) The classic example is jettison, which serves to save ship and cargo from a common peril, in that both would otherwise risk being lost if this extraordinary act were not taken. In the parlance of average adjusters, sacrifices and expenses which are incurred for such purposes are ‘common safety allowances’. English law and practice have largely recognised only this category of general average.16 In addition to common safety allowances, there exists another category of admissible loss and expense which may be called ‘common benefit allowances’.17 These are typically incurred at a port of refuge and include the cost of discharging, storing and reloading cargo if necessary to carry out repairs, as well as port charges, and wages and so on during the period of detention while repairs are carried out. Given that the vessel will already have put into the port or place of refuge before such losses and expenses are incurred, common safety will already have been attained, hence the reference instead to common benefit. In general terms, English courts were reluctant to admit such losses in general average as a matter of maritime law, unadorned by the York–Antwerp Rules.18 However, the attitude in European countries and in the United States of America was different, in that their courts regarded the attainment of safety as only the first step; the ultimate aim of delivering the cargo at its destination had yet to be fulfilled. Hence, See Hongkong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd (The Hongkong Fir) [1962] 2 QB 26. 14 However, any such determination is necessarily fact-sensitive. As observed by R.C. Clancey in his 1965 address to the AGM of the Association of Average Adjusters on ‘Forwarding Expenses’, ‘the type of cargo, the length of the voyage, the cargo storage and repair facilities are all material factors’. 15 In variation of the position at common law, standard forms of charterparty and bills of lading often contain express liberty clauses designed for circumstances beyond the control of the carrier, and may give the carrier a number of options, including an express option of transhipment. 16 Cornah, Shead & Sarll, Lowndes & Rudolf, General Average and York-Antwerp Rules (15th edn, Sweet & Maxwell 2018) at para. 00.70. 17 Cornah, Shead & Sarll (n 16) at para. 00.70; CMI Guidelines Relating to General Average (2nd edn, www​.comitemaritime​.org 2022), para. 1.2. 18 It is fair to say that English law was nevertheless edging towards admission: see the discussion of Atwood v Sellar (1879) 4 Q.B.D. 342 and Svendsen v Wallace (1884–5) L.R. 10 App. Cas 404 HL in Cornah, Shead & Sarll (n 16) at 10.08 et seq. 13

A review of Rules F and G, York–Antwerp Rules  167 expenses that were necessary to enable the ship and cargo to resume the voyage safely (but not the cost of repairing accidental damage to the ship) could be admitted in general average under these foreign systems of law.19 In furtherance of the aim of the York–Antwerp Rules to harmonise the treatment of general average internationally, provision was made for common benefit allowances, albeit in carefully qualified circumstances. In particular, Rule X(b) has, in successive editions of the Rules, enabled the cost of handling on board or discharging cargo to be allowed in general average ‘when the handling or discharge was necessary […] to enable damage to the ship caused by sacrifice or accident to be repaired, if the repairs were necessary for the safe prosecution of the voyage’. Also admitted, under Rule X(c), are ‘the costs of storage […] reloading and stowing’, which are admissible ‘[w]henever the cost of handling or discharging cargo, fuel or stores is allowable as general average’ – wording which takes one back to Rule X(b). Similarly, under Rule XI(c)(i) of successive editions except York–Antwerp 2004, crew wages and maintenance are allowable in circumstances where a ship has been detained at a port in consequence of accident ‘to enable damage to the ship caused by sacrifice or accident to be repaired, if the repairs were necessary for the safe prosecution of the voyage’. Fuel and stores consumed during the extra period of detention are also allowable under Rule X(c)(i), with port charges being allowed under Rule X(c)(ii). As such, the York–Antwerp Rules contain a series of provisions, applicable at the port of refuge, which work to the advantage of the shipowner if there should be any delay due to the need to carry out repairs there. Whereas, under the law of contracts of carriage as considered above, the shipowner would ordinarily be obliged to shoulder such onerous expenses itself unless and until the contract is discharged by frustration, this position is modified by the York–Antwerp Rules, which oblige the other parties to the maritime adventure to ‘chip in’. Nowadays, this hardly seems surprising, given that the York–Antwerp Rules are of such long standing. However, even if, by force of habit, such allowances have lost any capacity to surprise, this is not to say that the dissonance between the position under the law of contracts of carriage and the position under the law of general average should be overlooked altogether. As will later be seen, the ability to allow forwarding expenses and NSA expenses in general average rests on the admissibility of the very type of port of refuge expenses under consideration. Yet inasmuch as such expenses were ordinarily not allowed in general average by English courts, they are inherently controversial. They have also attracted severe criticism from at least one well-respected commentator, who has castigated this type of allowance, labelling it as ‘artificial’ general average.20 Indeed, it is a reflection of the very controversy surrounding the admission of port of refuge expenses in general average that the provisions of Rule X and XI descend into such detail with a view to ensuring that allowances are kept within carefully defined boundaries. One important proviso is contained in the final provision of Rule X, which finds itself repeated in similar terms in Rule XI(b)(v):

19 Buglass, Marine Insurance and General Average in the United States (3rd edn, Cornell Maritime Press 1991) at 197; Cornah, Shead & Sarll (n 16) at para. 00.70. 20 Tetley, Marine Cargo Claims (4th edn, Editions Yvon Blais 2008) at 1799. Professor Tetley considers it unsatisfactory that, following inclusion of the Interpretation Rule in 1950 which gives precedence to numbered rules over lettered rules, it ‘was not necessary that a ship be in “peril” if claims were made under Rule X(b) and XI(b), for example’.

168  Research handbook on marine insurance law But when the ship is condemned or does not proceed on her original voyage, storage expenses shall be allowed as general average only up to the date of the ship’s condemnation or of the abandonment of the voyage or up to the date of completion of discharge of cargo if the condemnation or abandonment takes place before that date.

The meaning of this provision is somewhat unclear, not least because the terms ‘condemnation’ and ‘abandonment of the voyage’ date back to the 1890 edition of the York–Antwerp Rules, when the modern law of frustration had not yet been enunciated. In Lowndes & Rudolf, The Law of General Average and the York–Antwerp Rules, the important suggestion is made that the phrase ‘abandonment of the voyage’ is not confined to circumstances where the contract of carriage is frustrated, but includes the ordinary case where a decision is taken to tranship, when the transhipment is in continuation of the existing contractual voyage.21 That type of scenario – in which the onerous conditions of frustration are not met yet a decision is taken to tranship cargo so that the contract of carriage may be fulfilled – will be considered further below in connection with Rule F, the provision which governs the allowance of ‘substituted expenses’. The point to note at present is that, on occasions when it is decided to tranship cargo yet the contract of carriage is not frustrated, the ‘date of completion of discharge’ may constitute an important dividing line. Up until that date, port of refuge expenses may be directly admitted in general average pursuant to Rules X and XI. After that date, any further port of refuge expenses that are incurred between the completion of discharge from the original vessel and transhipment aboard the substitute vessel could only be admitted in general average indirectly. Indirect admission might occur through the NSA provisions in Rule G, third paragraph, which, as discussed further below, obliges cargo interests to continue contributing towards shipowners’ expenses such as bunker costs and wages and maintenance as though the transhipment had not occurred. Alternatively, indirect admission could take the form of ‘substituted expenses’ under Rule F, and, as such, include the interim storage expenses and the costs of reloading aboard the substitute vessel that are inherent in the transhipment arrangements. But in that case, the allowance of expenses would have to respect the particular conditions of Rule F, including its final cap. It is to Rule F and its various conditionalities that this chapter now turns.

5.

LAW OF GENERAL AVERAGE – ‘SUBSTITUTED EXPENSES’ UNDER RULE F

Rule F is a complicated rule which is deceptive for its apparent simplicity. It provides in part as follows: ‘Any additional expense incurred in place of another expense, which would have been allowable as general average shall be deemed to be general average and so allowed without regard to the saving, if any, to other interests, but only up to the amount of the general average expense avoided.’ The rule exists as a matter of convenience, and allows into general average the incurrence of expenditure which would not otherwise be admissible, if and to the extent that this achieves

Cornah, Shead & Sarll (n 16) at para. 10.71. One justification for this construction is given as follows: ‘This interpretation reflects the earlier phrase – “when the ship … does not proceed on her original voyage”.’ 21

A review of Rules F and G, York–Antwerp Rules  169 a saving. The following explanation, by N.G. Hudson, of the underlying concept has been cited with approval:22 What the practitioner says is, in effect ‘Here are two (or possibly more) permissible courses of action which the shipowner may take consistently with his obligations under the contracts of carriage. Either (or any) of them involve extra expense in addition to that to which the shipowner was committed prior to the casualty. However, only one of those courses of action involves expenditure for which there is specific provision in the York–Antwerp Rules for a general average contribution. It would be unfair to leave the shipowner without recourse in general average if he should decide to take a course of action for which the Rules do not specifically provide; consequently we shall look at each available course of action as an option in fact, and we shall allow the actual additional expense incurred, but not exceeding the cost which would have been incurred and allowed as general average under the specific provisions in the York–Antwerp Rules.

Various examples can be given of the scenarios to which Rule F applies.23 For instance, a vessel suffers an engine breakdown, as a result of which she requires towage assistance. If, in avoidance of the high cost of admissible port of refuge expenses which would otherwise be incurred if the vessel underwent repairs at a port of refuge, she is instead towed all the way to destination, the resultant towage costs will be allowed under Rule F up to the amount of the port of refuge expenses that are thereby avoided. In calculating the amount of the substituted expenses allowed under Rule F, credit must be given of the voyage expenses saved – in this instance, the bunkers which would have been consumed if the vessel had proceeded to destination under her own steam less the additional crew wages for the longer voyage time involved. Of especial relevance to this chapter is the treatment of forwarding costs as substituted expenses under Rule F. For example, a bulk carrier may run aground shortly after leaving the port of loading, suffering damage to her hull. She refloats on the next high tide but requires extensive repairs at the port of loading before the voyage can be prosecuted. It is anticipated that cargo would need to be discharged and stored in a warehouse while repairs are carried out. One course of action would involve discharging the cargo, storing it and reloading it after the completion of repairs, the original voyage then being prosecuted to destination. In that case, the discharging, storage and reloading costs would be allowed under Rule X, while the cost of crew wages and maintenance would be allowed under Rule XI. Another course of action, however, would involve forwarding the cargo to destination aboard a substitute vessel. In that case, the forwarding expenses would serve to avoid the port of refuge expenses allowable under Rules X and XI, and, as such, can be admitted in general average pursuant to Rule F – but only up to the general average avoided. Again, any voyage expenses saved would fall to be credited when assessing the amount of the substituted expenses to be allowed. It is an implicit requirement of Rule F that the hypothetical alternative course of discharging, storing and reloading the cargo and continuing with the original voyage was indeed a real option available to the shipowner. It might not have been a real option if the delay involved in that sequence of events would have brought about the frustration of the contract, such that cargo interests would have demanded the release of their cargo rather than allow it to languish See Mitsui & Co Ltd & others v Beteiligungsgesellschaft LPG Tankerflotte Mbh & Co KG & another (The Longchamp) [2015] 1 Lloyd’s Rep 76 at [68]. 23 For a list of common examples, see Cornah, Shead & Sarll (n 16) at para. F.43 et seq. A useful worked example of adjusting calculations can be found in Hudson & Harvey, York Antwerp Rules, (4th edn, Informa Law 2018) at para. 11.44. 22

170  Research handbook on marine insurance law in the warehouse during the period of repairs.24 This is an important precondition, but it can be overlooked. As with the allowance of port of refuge expenses, the treatment of forwarding expenses as general average has become so conventional as to obscure quite how remarkable it really is. From the explanation above of relevant principles in the law of contracts of carriage, it was seen how the shipowner had to bear the cost of delivering cargo to destination, unless the contract was frustrated, in which case a second freight could potentially be charged. Through the incorporation of the York–Antwerp Rules (and in particular Rule F), the shipowner’s position has been considerably improved. Having already charged one freight for the original voyage, the shipowner may look to the parties to the common maritime adventure to contribute, in addition, to the cost of the substitute vessel; it need not bear the cost of that substitute vessel all by itself. This is so even though the original contractual duties continue to subsist, including the obligation to carry the cargo to destination. While that might seem heterodox, it is not entirely outlandish in the context of general average, which superimposes a scheme of loss distribution that is capable of applying, even though there may already be a relevant contractual obligation. As observed by Hobhouse J in The Bijela:25 Many types of general average expenditure are incurred by the master in fulfilment of his duties to the interests involved in the common adventure, including to the cargo-owners, and the function of general average is to provide a superimposed scheme for the sharing of the burden of that expenditure in an equitable fashion.

In other words, upon the occurrence of a maritime casualty, various additional voyage expenses may prove necessary in order that the condition of the cargo may be preserved, consistently with the shipowner’s contractual obligations. Some of these expenditures will not be treated as general average (such that the shipowner alone must shoulder them alone), whereas others will be (such that the other parties to the adventure must contribute). The question, which is at the heart of this chapter, is: ‘what belongs in general average?’

6.

RULE F – REASONABLENESS CONUNDRUM

While the general concept underlying Rule F can be stated fairly simply, a number of complex issues exist, some of which do not yet have any clear answers. The first issue concerns the question whether the hypothetical alternative course of action, in place of which the actual course of action is undertaken, has to be a reasonable one. For the purposes of this enquiry, let us posit the example of a stricken oil tanker requiring extensive repairs in a gas-free condition. Since no shoreside storage facilities exist of sufficient size to store the cargo during repairs, let us suppose that the cargo would have to be stored aboard 24 See Buglass (n 19) at p.267: ‘It must be stressed that the allowance of forwarding expenses can be justified only if the carrier would have been entitled, in the particular circumstances prevailing, to retain the cargo at the port of refuge; otherwise no alternative would exist.’ 25 Marida Ltd & Others v Oswal Steel & Others (The Bijela) [1992] 1 Lloyd’s Rep 636 at 643. See also R. Cornah, The York–Antwerp Rules 2016 – A Pragmatic and Commercial Approach and the Principle of Substituted Expenses: The Longchamp (2016) 22(6) J.I.M.L. 456.

A review of Rules F and G, York–Antwerp Rules  171 a flotilla of storage barges, which would themselves have to be towed to the location of the casualty from various ports in the surrounding region at great expense. Could the forwarding expenses incurred in engaging a substitute vessel be allowed in place of the cost of the flotilla of barges? Or would this not be possible, on the basis that the cost of arranging for the flotilla would be unreasonable? It was previously considered by many, if not all, average adjusters that the ‘other expense’ in place of which the actual expense was incurred had to be reasonable, in order for the actual expense to be deemed to be general average under Rule F.26 The argument ran as follows.27 Rule F allows ‘[a]ny additional expense incurred in place of another expense, which would have been allowable as general average [to be] deemed to be general average’. In order to have been allowable as general average, that other expense would have had to have been reasonable. This follows from the reasonableness requirement in Rule A, York–Antwerp Rules, which provides that there ‘is a general average act when, and only when, any extraordinary sacrifice or expenditure is intentionally and reasonably made or incurred for the common safety’. Alternatively, for those occasions when Rule A is not engaged, the need for reasonableness in the alternative course of action follows from the Rule Paramount that appears in York– Antwerp Rules 1994 and subsequent editions, which provides: ‘In no case shall there be any allowance for sacrifice or expenditure unless reasonably made or incurred.’ The suggestion that the ‘other expense’ would have had to have been reasonable posed a conundrum. It would mean that, if a shipowner incurs an expense to avoid paying a reasonable sum, it can in principle recover under Rule F, whereas if it incurs expense to avoid paying an unreasonable sum (that is, a larger sum), it cannot recover. The more obvious its duty to mitigate, and the greater the likely benefits of such mitigation, the less likely it would be able to recover. This conundrum had been known to average adjusters for many years. As long ago as 1946, Mr J.R. Danson took the view that the only solution was an express forwarding agreement:28 [I]f I am right in my interpretation of the legal position and I am assured that I am, the only safe method for an Adjuster to adopt in the future, in cases where alternative courses are possible, will be to obtain a special agreement from all the parties concerned to the lesser sum being allowed as a Substituted Expense and charged to General Average, otherwise the correctness of treating the costs of the course adopted as a Substituted Expense will always depend upon the view which any single individual might take regarding the reasonableness or otherwise of the course for which the expenses were substituted.

However, in the controversial case of The Longchamp,29 the Supreme Court discerned a solution to the conundrum within the wording of Rule F itself.

26 See Association of Average Adjusters Advisory Opinion G24a entitled Substituted expenses, Implications on practice with regard to York-Antwerp Rule F in light of the Supreme Court decision in ‘LONGCHAMP’ at para. 1.3. 27 See the Association of Average Adjusters Advisory Opinion quoted in Mitsui & Co Ltd & others v Beteiligungsgesellschaft LPG Tankerflotte Mbh & Co KG & another (The Longchamp) [2015] 1 Lloyd’s Rep 76 at 85. 28 In his Chairman’s address to the AGM of the Association of Average Adjusters in 1946. 29 Mitsui & Co Ltd & others v Beteiligungsgesellschaft LPG Tankerflotte Mbh & Co KG & another (The Longchamp) [2015] 1 Lloyd’s Rep. 76 (Comm. Ct); [2016] 2 Lloyd’s Rep. 375 (C.A.); [2018] 1 Lloyd’s Rep. 1 (S.C.).

172  Research handbook on marine insurance law The facts of that case were extreme, but by no means unusual at the time. It involved a chemical tanker that was seized on 29 January 2009 by Somali pirates, whereupon a ransom demand of US$6 million was made. After negotiation, a ransom amount of US$1.85 million was agreed on 22 March 2009 and the vessel was released. During the period of negotiation, the shipowner incurred expenditure, in particular crew wages (including high-risk area bonus) and maintenance, as well as the cost of fuel consumed during the detention period, totalling approximately US$160,000. This was claimed and allowed in a published adjustment by way of general average under Rule F in substitution for the greater amount of ransom money that would have had to have been paid if the initial ransom demand had been accepted without negotiation. (Direction admission of such detention expenses was impossible, not least because the vessel was not at a port or place of refuge, but instead at a pirate’s lair.) The cargo interests contested the correctness of the adjustment. Their principal contention was that, if the full ransom had been paid on first demand, the payment would not have been ‘reasonably incurred’ for the purposes of Rule A, given the invariable practice in Somali piracy cases of negotiating the ransom down significantly, as in fact occurred. It was held by a majority of the Supreme Court that the detention expenses were indeed allowable. As for the notion that the ‘other expense’ (here, the greater amount of ransom money that would have had to have been paid) needed to be reasonable in order for the actual expense to be deemed as general average under Rule F, this was dismissed. Instead of meaning that the requirements for admission in general average had to be strictly complied with, the reference in Rule F to an ‘expense which would have been allowable’ was, according to Lord Neuberger, simply to an expense of a nature which would have been allowable. A number of reasons were given by Lord Neuberger in support of his view, as follows:30 First, the word ‘allowable’ in Rule F naturally takes one to Rule C, where the similar word ‘allowed’ is used, rather than Rule A, where there is no reference to anything being ‘allowed’ (the same point applies to the French version – ‘admissible’ in Rule F and ‘admis’ in Rule C). Unlike Rule A, Rule C is concerned purely with the type of expense, and not with quantum. Secondly, the opening part of Rule F is unlikely to be concerned with quantum, as that is dealt with in the closing part, which imposes a cap on a sum recoverable under Rule F, namely ‘only up to the amount of the general average expense avoided’. Thirdly, the interpretation assumed in the courts below imposes an unnecessary fetter on the allowability of an ‘extra expense’, as there is already a reasonable fetter in the concluding part of Rule F. Fourthly, the interpretation I favour produces an entirely rational outcome: whenever an expense is incurred to avoid a sum of a type which would be allowable, that expense would be allowable, but only to the extent that it does not exceed the sum avoided.

Let us now return to the example involving the hypothetical course of action of arranging a flotilla of storage barges. If one were seeking to argue against the allowance under Rule F of actual forwarding expenses on the basis that they could not properly be incurred in place of the cost of hiring the barges and towing them to the location, it would not be a valid argument that that alternative course of action would have been unreasonable. For the reasons explained in The Longchamp, the ‘other expense’ need not be reasonable. It need only be of a nature which would have been allowable.

30 Mitsui & Co Ltd & others v Beteiligungsgesellschaft LPG Tankerflotte Mbh & Co KG & another (The Longchamp) [2018] 1 Lloyd’s Rep. 1 at [19].

A review of Rules F and G, York–Antwerp Rules  173

7.

RULE F – NATURE OF HYPOTHETICAL ALTERNATIVE EXPENSE

The second complex issue involves identifying an expense that is of the requisite ‘nature’, as per Lord Neuberger’s test. Would the costs of arranging for the flotilla of storage barges have been of a nature which would have been allowable, or not? As has already been seen, storage expenses are typically allowable where cargo has to be warehoused while repairs are undertaken that are necessary for the prosecution of the voyage. Are the costs of arranging for the storage barges qualitatively any different?31 In this regard, a potentially apposite observation was made in a dissenting judgment of Lord Hofmann in The Bijela.32 There he suggested that Rule F contemplates that ‘the hypothetical alternative would be a course which the owner was at least legitimately entitled to take’. It is unclear from the judgments in The Longchamp whether or not this qualification applies. But if the hypothetical alternative expense does not involve a course which the owner was at least legitimately entitled to take, it is hard to see why it should be of a nature which would have been allowable. Following this reasoning, it would appear that the alternative course should be a natural and logical alternative, not a matter of artificial invention.33 For these reasons, positing the avoidance of storage costs involving a flotilla of barges – some of which would have had to have been towed to the casualty location at great expense – could well be objectionable on the basis that such expenses are simply not of the required nature: taking such action could fall altogether outside of normal shipping practice and instead be pure artifice. The question where to draw the line, is not, however, a simple one. Finally, while it has been useful to take the example of arranging a flotilla of storage barges in order to identify the problem at hand, in practice average adjusters would typically utilise a different hypothetical course of action altogether in such a situation. As noted in Lowndes & Rudolf, ‘the hypothetical storage of oil aboard a transhipment vessel is often treated as an alternative expense, in place of which the expense is incurred in forwarding the oil to destination’.34 But that analysis throws up its own difficulties. What sort of shipowner would use a transhipment vessel for floating storage, when the possibility exists of her instead steaming directly to destination?! This is a problem which is not solved by the decisions in The Longchamp, in which none of the courts accepted the shipowners’ argument that one must assess the reasonableness of the hypothetical course of action (floating storage) on the assumption that the actual course of action (steaming to destination) was not available or had not been taken.35 Whatever

See further Association of Average Adjusters Advisory Opinion G24a entitled Substituted expenses, Implications on practice with regard to York-Antwerp Rule F in light of the Supreme Court decision in ‘LONGCHAMP’ at para. 1.4, in which it is suggested that the cost of hiring ‘several thousand road tankers’ would meet the requirement of being ‘of a nature’ permitted by the rules, on the basis that ‘such an operation falls within the terms of Rule X(c) regarding storage of cargo’. But this must surely be questionable. 32 Marida Ltd v Oswal Steel (The Bijela) [1993] 1 Lloyd’s Rep. 411 at 421. 33 See further Cornah, Shead & Sarll (n 16) at para. F.38. 34 Cornah, Shead & Sarll (n 16) at para. F.40. 35 Mitsui & Co Ltd & others v Beteiligungsgesellschaft LPG Tankerflotte Mbh & Co KG & another (The Longchamp) [2015] 1 Lloyd’s Rep 76 at [105]; [2016] 2 Lloyd’s Rep. 375 at [71]–[74]; [2018] 1 Lloyd’s Rep. 1 at [58]. (While the last of these references is to the dissenting judgment of Lord Mance, the observation is implicit in the judgments of the majority.) 31

174  Research handbook on marine insurance law criteria apply when evaluating the acceptability of the alternative course of action, there is no room, following The Longchamp, for any recourse to such legal fictions, however convenient they might seem. On the basis that it is incorrect to blinker oneself from the option of using the transhipment vessel to sail directly to destination – such that one must instead acknowledge that obvious course as a distinct possibility – it is a challenge to say why it is acceptable to posit the cost of using the ship as a ‘floating warehouse’ as expenses in place of which the steaming costs may be allowed pursuant to Rule F. Given the option of steaming straight to destination, is not the ‘floating warehouse’ theory a mere fantasy?36

8.

RULE F – OPERATION OF ‘CAP’

The third issue of some complexity is as follows. Even if, as per the ruling in The Longchamp, there is no reasonableness requirement inherent in the formula, ‘in place of another expense, which would have been allowable as general average’, might there nevertheless be a reasonableness requirement inherent in the final cap, that is, ‘but only up to the amount of the general average expense avoided’? While the judgments in The Longchamp do not spell it out clearly, it is submitted that they do suggest that the cap should be quantified by applying ordinary principles, including the usual requirement of reasonableness in Rule A and the Rule Paramount. In particular, as Lord Mance pointed out in his dissenting judgment, by acknowledging the existence of the cap as an adequate protection for the other interests, Lord Neuberger appears to have accepted that the entitlement to treat substituted expenses as general average is limited by the extent to which the hypothetical other expense could have been so treated.37 On the facts of The Longchamp, it is readily apparent why it presented no obstacle that the cap should have entailed the usual requirement of reasonableness. It could hardly be said that only US$1.85 million and no more was the reasonable amount to pay for the liberation of the ship and its cargo. Rather, it would have been reasonable for the purposes of the cap to pay ransom in the amount of at least the amount of US$2.4 million (being the actual ransom in the amount of US$1.85 million plus the operating expenses incurred during the period of negotiation plus the cost of the ransom negotiators). Hence, the detention expenses could clearly be allowed by way of substitution, as they were indisputably encompassed within the greater amount which it would have been reasonable to pay – even if the reasonable ransom amount had not been US$6 million but instead some much lesser sum. While the cap therefore presented no obstacle in The Longchamp, this need not always be the case. In the example under consideration, the cost of hiring a flotilla of storage vessels might prove exorbitantly expensive. By dint of the words ‘but only up to the amount of the general average expense avoided’, one could argue that none of the cost should be admitted under Rule F, since it would be unreasonable to incur it and it could not therefore constitute

See Hudson, Substituted Expenses – Fact or Fantasy [1992] LMCLQ 322, 328, who suggests that the ‘true philosophy’ behind the concept of substituted expenses concept must be ‘based on the facts of each case, and not on any fantasy’. That observation was approved in Mitsui & Co Ltd & others v Beteiligungsgesellschaft LPG Tankerflotte Mbh & Co KG & another (The Longchamp) [2015] 1 Lloyd’s Rep 76 at [68]. 37 Mitsui & Co Ltd & others v Beteiligungsgesellschaft LPG Tankerflotte Mbh & Co KG & another (The Longchamp) [2018] 1 Lloyd’s Rep. 1 at [55]. 36

A review of Rules F and G, York–Antwerp Rules  175 general average properly so called, having regard to the reasonableness requirement in Rule A and the Rule Paramount. If, however, the words, ‘but only up to the amount of the general average expense avoided’ do involve an ordinary reasonableness requirement, can it really be said that the ‘reasonableness conundrum’ has been solved? Even if a fact-specific solution could be found in The Longchamp, situations in which that particular solution will apply are likely to be rare. Ultimately, the suggestion made by Mr J.R. Danson in 1946 of a ‘special agreement’ may still commend itself.

9.

LAW OF GENERAL AVERAGE – NSAS

So far, the discussion has concerned the expense of transhipping cargo and forwarding it to destination aboard a substitute vessel. It is now necessary to consider the other type of expense that is typically incurred on occasions of transhipment, namely NSA expenses. NSAs operate as a type of quid pro quo.38 Their purpose is to permit cargo interests promptly to recover their cargo at some intermediate port of refuge or repair where delay might otherwise ensue (the quid), while at the same time obliging the cargo interests to carry on contributing to general average for as long as it would otherwise be owing (the quo). In essence, the agreement involves an acceptance on the part of cargo interests to contribute in general average as though their property had not been separated through forwarding aboard substitute vessels or delivery at the port of refuge – hence, ‘non-separation’. The legal assumption which underlies the NSA is that the shipowner is not bound to accede to a request for delivery up of cargo at an intermediate port, but is instead contractually entitled to carry it to destination, even if this means that cargo interests must contribute in general average to ongoing port of refuge expenses while the vessel is repaired. Insofar as cargo interests are obliged to carry on contributing in general average during the period of the repairs, this will work to the shipowner’s advantage, as it will help defray its expenses. Since the shipowner is not bound to agree to premature delivery, except on its own terms, it has for many years been conventional for an NSA to be given in exchange for the release of the cargo. In effect, this means that cargo interests will carry on contributing to, for example, wages and maintenance and bunker costs expended while repairs are carried out that are necessary for the safe prosecution of the voyage, even though the original voyage will never be performed. In fact, the cargo may already have reached its destination aboard a substitute vessel long before the allowable detention expenses conclude. Previously, the practice was for an ad hoc NSA to be entered into.39 This often appeared in average bonds and average guarantees given in relinquishment of the shipowner’s possessory lien.40 Since 1994, however, successive editions of the York–Antwerp Rules have expressly included an NSA in Rule G, which provides in its third and fourth paragraphs as follows.

38 Or, as per C.T. Ellis in his Chairman’s address at the AGM of the Association of Average Adjusters in 1964, a ‘tit for tat’. See further Buglass (n 19) at p.271. 39 In 1967 the London marine insurance market adopted a Standard Form of NSA, which is reproduced in Buglass (n 19) at p.271. 40 Certain average adjusters still include extensive NSAs in their house forms of average security documentation, although this is unnecessary when general average is to be adjusted according to YAR

176  Research handbook on marine insurance law (3) When a ship is at any port or place in circumstances which would give rise to an allowance in general average under the provisions of Rules X and XI, and the cargo or part thereof is forwarded to destination by other means, rights and liabilities in general average shall, subject to cargo interests being notified if practicable, remain as nearly as possible the same as they would have been in the absence of such forwarding, as if the adventure had continued in the original ship for so long as justifiable under the contract of affreightment and the applicable law. (4) The proportion attaching to cargo of the allowances made in general average by reason of applying the third paragraph of this Rule shall not exceed the cost which would have been borne by the owners of the cargo if the cargo had been forwarded at their expense.

Also relevant is Rule XVII(c) which explains how contributory values are to be assessed in cases of forwarding: In the circumstances envisaged in the third paragraph of Rule G, the cargo and other property shall contribute on the basis of its value upon delivery at original destination unless sold or otherwise disposed of short of that destination, and the ship shall contribute upon its actual net value at the time of completion of discharge of cargo.

As with Rule F, there are a number of puzzles inherent in these provisions.

10.

RULE G(3) – AVOIDING DOUBLE-COUNTING

The first issue concerns the potential for overlap between Rule F and the NSA provisions in Rule G. As has already been seen, forwarding expenses are allowed under Rule F in place of the admissible detention expenses that would otherwise have been incurred if the vessel had undergone repairs and the original voyage had been continued; such expenses could in principle include the discharging, storing and reloading costs of the cargo as well as the wages and maintenance and bunker costs to be incurred by the shipowner. However, under the third paragraph of Rule G, cargo interests will anyway be obliged to contribute towards wages and maintenance and bunker costs as part of its promise to contribute towards general average as though there had been no separation of interests. Given the risk of double-counting, it is the practice of average adjusters to allow under Rule F the forwarding expenses in place of (only) the avoided storage expenses which would have been admitted under Rule X(c). As explained in Lowndes & Rudolf:41 ‘It is important to note that items such as crew wages cannot be used to rank as savings for allowances under r. F […] if they are allowed under a non-separation agreement, since this would result in a double allowance in general average.’

1994 or subsequent editions. In the CMI-approved General Average Bonds and Guarantees appended to the CMI Guidelines on General Average (n 17) at cl. 5(e) it is expressly agreed that the ‘third and fourth paragraph of Rule G of the York–Antwerp Rules 2016 shall be deemed incorporated where the applicable York–Antwerp Rules do not contain a non-separation agreement’. Hence, whereas a contract contained in a bill of lading may call for general average to be adjusted according to, for example, York–Antwerp Rules 1974, the contract contained in the average bond or guarantee will call for general average to be adjusted according to York–Antwerp Rules 1974 but with the addition of the ‘third and fourth paragraph of Rule G of the York-Antwerp Rules 2016’. 41 Cornah, Shead & Sarll (n 16) para. G.12.

A review of Rules F and G, York–Antwerp Rules  177 This has consequences for the quantification of the cap in Rule F, given that the limit imposed by the words, ‘but only up to the amount of the general average expense avoided’ will typically be assessed against the amount of avoided storage expenses, and no more.

11.

RULE G(3) – INTERACTION WITH FRUSTRATION

The second complex issue relates to the interaction between the NSA provisions and the doctrine of frustration as contemplated by the words ‘as if the adventure had continued in the original ship for so long as justifiable under the contract of affreightment and the applicable law’. One particular situation in which this proviso will be applicable is where a frustrating delay or a cost of repair that exceeds the value of the ship when repaired is not yet apparent at the time of forwarding, but only later becomes apparent. Such a situation is by no means uncommon where, for example, it is first necessary to open a crankcase before the extent of the damage to the crankshaft can be ascertained.42 Supposing the cargo had not been forwarded, the shipowners’ admissible detention expenses would have ended following that discovery in accordance with the end date in Rule XI(b)(v), which is referrable to the ‘date of the ship’s condemnation or of the abandonment of the voyage’ and so on. The formula in Rule G, ‘for so long as justifiable under the contract of affreightment and the applicable law’, therefore ensures that the same cut-off applies where the cargo has already been forwarded. But what if it is already apparent at the time of forwarding that there will be a frustrating delay or that the cost of repair will exceed the value of the ship when repaired? Even though the contract of carriage may not itself be frustrated if the shipowner is making use of a liberty to tranship,43 it would not have been justifiable in those circumstances to detain the goods at the intermediate port and no NSA expenses should therefore be payable.44

12.

RULE G(4) – BIGHAM CLAUSE

A third complex issue relates to the fourth and final paragraph of Rule G, which is known as the ‘Bigham clause’. This clause places a cap on cargo’s contributions to allowances made under the NSA provisions in the third paragraph by stipulating that they ‘shall not exceed the cost which would have been borne by the owners of the cargo if the cargo had been forwarded at their expense’. This clause, which was drafted by an eponymous attorney, owes its existence to the decision of the United States Federal Court for the Southern District of New York in Domingo de Larrinaga, in which Mr Bigham’s firm represented the successful consignees.45 It was held that consignees of damaged cargo, which had been discharged at a port of refuge to permit For an example of this, see Corney v Barellier & Francastel (1923) 16 Ll.L.L.R 39 at 41. See, in this connection, n 15 above. 44 As observed in Buglass (n 19) at p.273, in the context of ad hoc NSAs, ‘the problem as to whether in fact the carrier is entitled to hold the cargo under the original contract of affreightment (and is therefore entitled to ask the concerned in cargo to subscribe to such an agreement) remains to be dealt with in each individual case’. 45 Domingo de Larrinaga [1928] AMC 64. The genesis of the clause is explained in The ABT Rasha [2000] 2 Lloyd’s Rep. 575 at [20]. 42

43

178  Research handbook on marine insurance law inspections of the hull, were entitled, on paying full freight, to demand their goods at the port of refuge, even though the shipowners had arranged for towage of ship and cargo to destination. The clause purports to reflect this entitlement46 by ensuring that cargo interests are not to be placed in a worse position than if they had exercised their right to demand delivery of the cargo and had forwarded to destination at their own expense. According to Buglass in Marine Insurance and General Average in the United States,47 the clause represents an ‘excellent solution from the point of view of cargo interests, guaranteeing them as it does the wisdom of hindsight’. From the point of view of shipowners and their insurers, however, the clause is not nearly so favourable, given its ability to cast upon ship interests a greater part of the admissible detention expenses than would ordinarily be consistent with their rateable proportion.48 Whatever the disadvantages to the shipowner that have historically been inherent in the Bigham clause, there has recently been a development which has the potential to weaken the clause considerably. (Whether this truly is a development depends on one’s interpretation of the clause’s original effect – some adjusters maintain it is always how Rule G(4) has worked.) It consists of the addition to the Bigham clause as it appears in the York–Antwerp Rules 2016 of the cryptic words ‘This limit shall not apply to any allowances made under Rule F’. The effect of this apparently innocuous provision is as follows. As explained in the working papers produced for the CMI meeting in Istanbul on 6–7 June 2015, one issue which was thought to require clarification in the forthcoming 2016 edition of the York–Antwerp Rules was whether the cap in the Bigham clause should apply only in respect of allowance made under the NSA provisions in Rule G, or else whether it should apply in respect of both those allowances and allowances under Rule F, such as the expense of forwarding cargo to destination aboard a substitute vessel. In other words, the issue was whether the amount of cargo’s contribution to allowances made under Rule F and the NSA provisions in Rule G should be aggregated before applying the cap, or else whether the cap should only apply to cargo’s contribution to the allowances made under the NSA, with Rule F allowances being allowed separately. By means of the added words which appear in Rule G, York–Antwerp Rules 2016, it has been confirmed that the cap in the Bigham clause should only apply to allowances made under the NSA provisions; the contribution to forwarding expenses falls to be treated separately. Even if that interpretation may be defensible upon a literal construction of the NSA provisions in the third paragraph of Rule G, it is does not seem to accord with the original rationale for the clause. The result of adjusting general average in this manner may place the cargo

As recognised by the New York Courts; there is no English decision on this question. In this regard, it is the recollection of one senior average adjuster (as relayed to this writer) that, formerly, Bigham clauses were only acceptable to the companies market in London where the contract of carriage was governed by American law. Lloyd’s underwriters, however, took a more commercial view and did not impose such a requirement. Any such distinctions were, however, swept away once the Bigham Clause was incorporated into the York–Antwerp Rules. Given the incorporation of a Bigham clause in Rule G, fourth paragraph, it does not matter whether English law of contracts of carriage recognises such an entitlement or not, provided that Rule G(4) applies. 47 Buglass (n 19) at p.272. 48 In The ABT Rasha [2000] 2 Lloyd’s Rep. 575, it was held that H&M insurers were obliged to indemnify their assured for the balance of general average once the Bigham clause cap was applied, as the reference in Section 66(4), Marine Insurance Act 1906 to ‘proportion’ simply meant ‘share’ and there was no reason to restrict the ordinary meaning of the word proportion to rateable proportion. 46

A review of Rules F and G, York–Antwerp Rules  179 interests in a worse position than if they had demanded delivery at the intermediate port and forwarded the cargo at their own expense, given the need to contribute to the forwarding expenses in addition to the capped NSA expenses.49 Moreover, since the amount of cargo interests’ contribution to NSA expenses alone will rarely be more valuable than the cost of a second freight (the contribution to forwarding expenses being treated separately), the result of this interpretation is, according to some average adjusters with whom this writer has spoken, that the Bigham clause will rarely operate when this adjusting methodology is utilised. While this interpretation is mandated by the express words of York–Antwerp Rules 2016, this new version of the Rules is still only infrequently used. It remains to be seen how a court would interpret the effect of the Bigham clause in earlier versions of the rules, if called upon to do so.50

13.

THE USE OF AD HOC FORWARDING AGREEMENTS

It seems there used to be a widespread practice of average adjusters and admiralty solicitors negotiating ad hoc forwarding agreements. In such cases, the leverage available to the shipowner in gaining cargo interests’ assent to the forwarding agreement, even though there might be no contractual obligation to do so, arises from the cargo interests otherwise being required to pay more: in the absence of an agreement, the shipowner would have to pursue an alternative course of action giving rise to allowable general average expenses for which cargo’s contribution would be in a greater financial amount. This formerly widespread practice has, however, largely fallen away in favour of the ensuing expenses simply being left for adjustment according to the York–Antwerp Rules, which have since 1994 incorporated express NSA provisions.51 Given the complexity of the issues that arise, the demise of ad hoc forwarding agreements is perhaps surprising, even if it may achieve a saving in costs. For instance, it is contrary to the prognostications of the well-known American adjuster Leslie Buglass, cited above, who, writing in 1991, expected the practice to become more common, not less.52 In a similar way, the continuing need for ad hoc forwarding agreements has been remarked upon by the British average adjusters Hudson and Harvey as follows:53 49 The fact that the effect of Rule G, third paragraph, York–Antwerp Rules 2016 may indeed be inconsistent with the American decisions which provide its origins is noted in Hudson & Harvey (n 23) at para. 12.47. 50 In this regard, it is notable that the interpretation of Rule G, third paragraph that is expressly agreed upon in the York–Antwerp Rules 2016 was not shared by all average adjusters who attended the CMI conference. Michael Harvey of Harvey Ashby Limited did not think it correct, as he notes in Hudson & Harvey (n 23) at para. 12.47. 51 One particular occasion when forwarding agreements are still negotiated is where sale of cargo occurs at a port of refuge, see Lowndes & Rudolf, General Average and York–Antwerp Rules, 15th edn, para. F.50. Aside from this, it has become common in the case of containership casualties for interim bridging security to be agreed, but this falls outside the scope of the present article. For an example, see The Maersk Neuchatel [2014] 2 Lloyd’s Rep. 377. 52 Buglass (n 19) p.270: ‘With the demise of Rule X (d) [a provision in the York–Antwerp Rules which expressly dealt with the admission of forwarding expenses but was omitted in 1974] no doubt this practice will become even more prevalent.’ 53 Hudson & Harvey (n. 23) at para. 12.50.

180  Research handbook on marine insurance law Prior to 1994 it was not infrequently the case that the provisions of the so-called standard forms of Non-Separation Agreement were not apt to cover the particular circumstances in which the parties found themselves. When this was the case, it was very often the task of the average adjuster to draft a special agreement, incorporating a Non-Separation provision, in order to suit the needs of the situation. It is envisaged that, notwithstanding the introduction of the Non-Separation Clauses into Rule G, cases will continue to arise which will exercise the ingenuity of the average adjuster to find a solution, in partial derogation of the York–Antwerp Rules, which will satisfy the conflicting interests of the parties.

Whenever it is doubtful whether forwarding expenses and detention expenses incurred following transhipment of cargo should be allowed in general average, it would be advisable for forwarding agreements to be expressly negotiated. Otherwise, it may be held that the relevant expenses are simply not encompassed by Rules F and G. The most prominent situations in which such admissions may be doubtful can be listed as follows. The list is not, however, intended to be exhaustive. ● Where the actual course of action undertaken in place of some hypothetical alternative course of action involves the incurring of a loss rather than an expense. This follows from the fact that Rule F refers only to any ‘additional expense incurred in place of another expense’. The need for this could, for instance, arise where it is intended to sell a cargo at an interim port rather than carry it to destination, in the knowledge that it might attract only a distressed price.54 ● Where there is concern about whether the hypothetical alternative is indeed of a nature that would have been allowable. ● Where there is concern that the amount of the cap in Rule F (‘but only up to the amount of the general average expense avoided’) might potentially be exceeded, simply upon comparison of costings. ● Where there is concern about the amount of the cap in Rule F being exceeded, by comparison with reasonable general average that is avoided. ● Where the hypothetical alternative course of action of repairing the ship and continuing with the voyage might entail the frustration of the contract. As has been seen, this is potentially relevant to both Rules F and G. ● Finally, where there is some doubt about whether the actual expenses which are intended to be substituted for admissible general average expenses might, in some sense, be too remote.55

54 Another example might be where it is expected that there may be cargo loss during transhipment operations, something which can occur where the cargo is fine and liable to seep through grabs. 55 While space has not allowed for the addressing of this issue, further explanation can be found in Association of Average Adjusters Advisory Opinion G24b entitled Substituted expenses, Revised implications on practice with regard to York-Antwerp Rule F in light of the Supreme Court decision in ‘LONGCHAMP’. For example, it is suggested at p.G24b/5 that ‘Additional warehouse rent at destination paid during the period’ might ‘possibly’ be allowed as substituted expenses in cases of deliberate delay as in The Longchamp, but this must be considered on a case-by-case basis as it may be too remote. If there are concerns about the remoteness of likely expenses, it could be stated in an agreement what expenses will fall to be admitted before the actual course is undertaken.

A review of Rules F and G, York–Antwerp Rules  181

14.

FINAL THOUGHTS

Finally, the question inevitably arises, quite irrespective of their correctness, of whether the principles considered in this article make for ‘good law’. It is suggested that this is dubious at best. As per a message received by this writer from one senior claims professional, commenting on modern containership casualties: it does not sit well with GA […] that with containers they can be moved so quickly when they have lodged security, unlike say a bulk cargo discharged so that repairs can be carried out before the voyage can continue, and some cargo underwriters might query why they are paying for lengthy GA detentions, when their cargo arrived at destination, whilst the ship was still discharging damaged containers!!

In this regard, it is a simple truth that containers discharged from a stricken containership are never likely to be stored in a container yard until such time as the ship is repaired. Liner operators would be hard pressed to justify such conduct to angry customers, nor would containership owners expect this of them. Instead, the containers will be forwarded to destination as a matter of routine, often aboard the very next containership scheduled to ply the same route. Against that background, the notion that forwarding expenses should be allowed in substitution for avoided detention expenses seems nothing more than a convenient fiction; so too does the notion that the original voyage would ever be prosecuted to destination as envisaged by an NSA. It must therefore be queried whether these complex and intricate rules properly reflect, and do justice to, the modalities of modern sea transportation, especially of containerised cargoes. If not, the justification for cargo interests’ contribution towards forwarding and NSA expenses risks resembling the description in Kingsley Amis’s Jake’s Thing of the disquisitions of Geoffrey Mabbott, one of whose specialities was the ‘inverted pyramid’, that is, ‘a great parcel of attitudes, rules and catchwords resting on one tiny (if you looked long and hard enough) point’. Take away the notion of the original voyage being prosecuted to destination, and what remains? Whisper it, but the expense of merely hiring another ship to carry cargo to its destination is not, and never has been,56 general average.

56

See Casaregis Discursus, cxxi., Nos 16, 17, 18, 19.

PART IV  

9. Protection & indemnity clubs and arbitration clauses Rob Merkin

1.

THE RULES OF P&I CLUBS1

The website of the International Group of P&I Clubs proudly announces that the 13 members of the Group provide marine liability cover for approximately 90 per cent of the world’s ocean-going tonnage. Protection & Indemnity (P&I) Clubs developed some 200 years ago as non-profit-making associations of shipowners formed primarily to pick up the increasing range of liability claims not covered by the standard Lloyd’s SG wording in all but universal use at the time. Various standard features of P&I Club rules and their relationship with ordinary insurance principles have vexed the courts over the years, not least the concept that payment of losses is discretionary rather than mandatory, but for present purposes it is necessary to focus on only two of those features: first, all disputes are to be submitted to arbitration; second, the provision of indemnity is on a ‘pay to be paid’ basis. As between Club and member, the arbitration clause is unexceptional in its operation and standard in most forms of commercial insurance and reinsurance. Pay to be paid, by contrast, is rarely adopted in any other form of liability insurance.2 Property policies even with apparent pay to be paid wording have been interpreted as meaning that as long as the assured incurs liability to a builder to carry out repairs then the insurers’ obligation to indemnify is triggered,3 and in the context of reinsurance ‘actual payment’ to the policyholder by the reinsured has been held to mean established and quantified liability to pay the policyholder.4 As long as the shipowner is solvent and able to pay claims, neither clause will ultimately bar the right to recover. However, where the losses are substantial, as in the case of pollution, collision or loss of life, and the shipowner does not have the financial resources to meet claims, the third party victims may, assuming that the insolvency proceedings are conducted in England, take advantage of the direct claims procedure set out in the Third Parties (Rights against Insurers) Act 2010. Under this legislation, the third party stands in the shoes of the assured and may bring proceedings against the Club itself. The 2010 Act, which replaced the Third Parties (Rights against Insurers) Act 1930 with effect from 1 August 2016, allows the third party to bring proceedings against the Club as soon as the shipowner has become insolvent under one or more of Hazelwood and Semark, P&I Clubs Law and Practice, 4th ed, 2010. It was said in Lambert Leasing Inc v QBE Insurance Ltd (No.2) [2015] NSWSC 1196 that such provisions were ‘inimical’ to the concept of insurance and under Australian law were potentially unenforceable on good faith grounds. 3 Medical Assurance Society New Zealand Ltd v East [2015] NZCA 250; Manchikalapati v Zurich Insurance Plc [2019] EWCA Civ 2163; Endurance Corporate Capital Ltd v Sartex Quilts and Textiles Ltd [2020] EWCA Civ 308. 4 Charter Reinsurance Co Ltd v Fagan [1996] 3 All ER 46. 1 2

183

184  Research handbook on marine insurance law the statutory insolvency regimes operative in England. Under the 1930 Act, the liability of the shipowner and the quantum of the claim had to be established by the third party in proceedings against the shipowner as a precondition of the direct action, whereas under the 2010 Act those matters fall to be determined in the direct action if there has been an earlier insolvency. Under the 1930 Act, and now the 2010 Act, the obligation to arbitrate cannot be sidestepped. If the third party commences judicial proceedings in the English courts, the Club may apply for a mandatory stay under s 9 of the Arbitration Act 1996 on the basis that the third party as statutory assignee is claiming ‘under or through’5 the shipowner as party to the arbitration clause. A stay will be granted as a matter of course as long as the Club has not taken a step in the judicial proceedings without a reservation of rights and thereby forfeited the benefit of s 9. The operation of a pay to be paid clause in a third party claim is more complex. Under the 1930 Act the majority of the House of Lords in The Fanti and the Padre Island6 reached the nonsensical conclusion – although one that was just about supportable on the actual wording of the legislation – that if the shipowner could not afford to pay then there were no sums due under the policy and the third party’s action against the Club could not succeed. Section 9 of the 2010 Act has left enforceability of pay to be paid clauses intact for marine insurance only, and even there has mitigated its worst effects by removing the right of a liability insurer to rely upon a pay to be paid clause in liability claims for death or personal injury. Pay to be paid thus remains in place for all other liabilities covered by P&I Clubs, including cargo, collision and pollution claims against shipowners. So much then for purely domestic claims. Both arbitration clauses and pay to be paid clauses will be enforceable without any real argument. The issue is whether it is possible for a claimant against a P&I Club to avoid those consequences by commencing proceedings elsewhere. It should here be emphasised that the fate of pay to be paid and arbitration clauses are closely intertwined. If the P&I Club can ensure that the claim against it is heard in arbitration in England, the pay to be paid clause will be enforced without argument. However, if the proceedings are brought against the P&I Club in a foreign court which is prepared to hold that the arbitration clause cannot be enforced, the pay to be paid clause is itself at risk. Although many of the cases concerning foreign proceedings are focused on arbitration, pay to be paid is of equal if not greater concern. In what follows, two separate scenarios are discussed. The first scenario is foreign court claims by the shipowner, raising the question of the enforceability of the arbitration clause in foreign proceedings as a matter of contract. The second scenario is foreign court claims not by the shipowner but by a third party exercising rights akin to those under the Third Parties (Rights against Insurers) Act 2010 as conferred by the local legislation. Such legislation may be more generous to the third party than the English measure, that is, that the right to claim is necessarily contingent upon the shipowner’s insolvency: the issue here is not enforceability as a matter of contract but, in Thomas Rafael KC’s important phrase, enforceability as a matter of ‘quasi-contract’.7

Arbitration Act 1996, s 82(2). [1990] 2 All ER 705. 7 Rafael, The Anti-Suit Injunction, 2nd ed, 2019. 5 6

Protection & indemnity clubs and arbitration clauses  185

2.

FOREIGN JUDICIAL PROCEEDINGS BY THE SHIPOWNER: CONTRACTUAL CLAIMS

2.1

The Anti-suit Injunction8

The history of the grant of anti-suit injunction has been expertly traced by Mr Rafael KC.9 It suffices to say here that the grant of relief originally required proof that the foreign proceedings were vexatious, oppressive or abuse of the process of the English court, but a series of cases beginning with Aggeliki Charis Compania Maritima SA v Pagnan SA, The Angelic Grace10 has led to the current position, capable of being summarised in a series of propositions. First, the jurisdiction of the English courts to grant anti-suit relief is founded upon the general power in s 37 of the Senior Courts Act 1981. Section 44(2)(e) of the Arbitration Act 1996, which confers the power on the English court of ‘granting an interim injunction for the purposes of and in relation to arbitration proceedings’, was at one time thought to confer concurrent jurisdiction, but Lord Mance, giving the leading judgment in the Supreme Court in AES Ust-Kamenogorsk Hydropower Plant LLP v Ust-Kamenogorsk Hydropower Plant JSC,11 rejected that possibility. Lord Mance drew a distinction between the positive obligation in an arbitration clause to submit disputes to arbitration, and the negative obligation not to submit disputes to another forum. The Supreme Court ruled that an injunction restraining breach of the negative obligation could not be said to be in relation to arbitration proceedings within the meaning of s 44(2)(e), given that the party seeking relief may well not itself be seeking to initiate arbitration proceedings. It was also noted that s 44(2)(e) did not apply to permanent injunctions, laid down its own conditions for grant that were not replicated in s 37 of the 1981 Act and imposed restraints on appeal. Second, an anti-suit injunction is not aimed at the foreign court but rather at the claimant who has wrongly started proceedings in that court. Third, the English court must have jurisdiction over the claimant in the foreign proceedings. That does not give rise to any issue in the present context, even if permission for leave to serve outside the jurisdiction is required. It was confirmed by the Supreme Court in AES Ust-Kamenogorsk Hydropower Plant LLP that such permission may be granted on either of two grounds: first, it is possible to rely upon CPR 62.5(1)(c), which applies where the applicant ‘(i) seeks some other remedy or requires a question to be decided by the court affecting an arbitration (whether started or not), an arbitration agreement, or an arbitration award; and (ii) the seat of the arbitration is or will be within the jurisdiction’; second, the general provisions of CPR Practice Direction 6B, para 3.1(6), which applies to breach of a contract governed by English law, will also be open to use by the applicant. P&I Club Rules are inevitably governed by English law, and although the arbitration clause is in principle a separate undertaking capable of having its own applicable law,12 it is inconceivable that it would be decided that

Fentiman, ‘Anti-Suit Injunctions – Comity Redux?’ [2012] CLJ 273. Rafael, n 7, Ch 2 ‘The History of an Unusual Remedy’. 10 [1995] 1 Lloyd’s Rep 87. 11 [2013] UKSC 35. 12 Arbitration Act 1996, s 7. For the nature of severability, see DHL Project & Chartering Ltd v Gemini Ocean Shipping Co Ltd, The Newcastle Express [2022] EWCA Civ 1555. 8 9



186  Research handbook on marine insurance law an arbitration clause specifying England as the seat of arbitration in a contract governed by English law would be governed by anything other than English law.13 Fourth, the test for grant is no longer whether the respondent to the anti-suit proceedings has acted oppressively or vexatiously. Although the jurisdiction is to be exercised with caution,14 the threshold question is whether the applicant can demonstrate to a high degree of probability that there is a valid arbitration clause extending to the dispute in question.15 If that is the case, the question becomes whether it is just and convenient for an injunction to be granted, the touchstone being what the ends of justice require. Fifth, the court may in the exercise of its discretion refuse an injunction if the respondent in the anti-suit proceedings can demonstrate that there are exceptional circumstances which militate against the grant of relief. The most important circumstance here is delay by the applicant in seeking anti-suit relief, thereby allowing the foreign proceedings to reach an advanced stage.16 Sixth, no question of forum non conveniens can arise. That was determined by the Supreme Court in Enka Insaat ve Sanayi AS v OOO Insurance Company Chubb,17 which concerned a construction contract with no governing law clause but written in Russian and specifying ICC arbitration in London. A fire broke out at the works and Chubb, having indemnified the employer, commenced subrogation proceedings in the Russian courts against the contractor. Chubb did not dispute the validity of the arbitration clause, but contended that it did not extend to the tort claim brought in the employer’s name. The contractor sought an anti-suit injunction. It was common ground that Russia was the most convenient forum for the hearing of the claim, given that the contracting parties were local to Russia and the issue was the proper Where the law applicable to the main contract and the seat are different, the common law view is that the law applicable to the arbitration clause follows that of the main contract: Enka Insaat v OOO Insurance Co Chubb [2020] UKSC 38. However, the Law Commission’s draft Insurance Bill published in September 2023, LC 1787 would, if implemented, provide as a default rule that the law applicable to the arbitration clause is that of the seat. 14 AES Ust-Kamenogorsk Hydropower Plant LLP v Ust-Kamenogorsk Hydropower Plant JSC [2013] UKSC 35. 15 Navigation Maritime Bulgar v Rustal Trading Ltd [2002] 1 Lloyd’s Rep 106. This formulation is found in all of the most recent cases: see, for example, Michael Wilson & Partners Ltd v Emmott [2018] EWCA Civ 51; Ulusoy Denizcilik AS v COFCO Global Harvest (Zhangjiagang) Trading Co Ltd [2020] EWHC 3645 (Comm); A v B [2020] EWHC 3657 (Comm); ZHD v SQO [2021] EWHC 1262 (Comm); Specialised Vessel Services Ltd v MOP Marine Nigeria Ltd [2021] EWHC 333 (Comm); UAU v HVB [2021] EWHC 1548 (Comm); VTB Bank PJSC v Mejlumyan [2021] EWHC 1386 (Comm); Aquavita International SA v Indagro SA [2022] EWHC 892 (Comm); Aiteo Eastern E&P Company Ltd v Shell Western Supply and Trading Ltd [2022] EWHC 2912 (Comm); AIG Europe SA v John Wood Group Plc [2022] EWCA Civ 781. The same principle applies to exclusive jurisdiction clauses: Ebury Partners Belgium SA/NV v Technical Touch BV [2022] EWHC 2927 (Comm); E-Star Shipping and Trading Company Ltd v Delta Corp Shipping Ltd [2022] EWHC 3165 (Comm); Al Mana Lifestyle Trading Llc v United Fidelity Insurance Company [2023] EWCA Civ 61. 16 Gorthon Invest AB v Ford Motor Co Ltd, The Maria Gorthon [1976] 2 Lloyd’s Rep 720; Alfred C Toepfer International GmbH v Molino Boschi Srl [1996] 1 Lloyd’s Rep 510; Mediterranean Shipping Co SA v Atlantic Container Line [2008] EWHC 213 (Comm); Transfield Shipping Inc v Chiping Xinfa Huayu Alumina Co Ltd [2009] EWHC 3642 (Comm); Essar Shipping Ltd v Bank of China Ltd, The Kishore [2015] EWHC 3266 (Comm); ADM Asia-Pacific Trading PTE Ltd v PT Budi Semesta Satria [2016] EWHC 1427 (Comm); UAU v HVB [2021] EWHC 1548 (Comm); Africa Finance Corporation v Aiteo Eastern E & P Company Ltd [2022] EWHC 768 (Comm). 17 [2020] UKSC 38. 13

Protection & indemnity clubs and arbitration clauses  187 construction of the arbitration clause. The Supreme Court ruled that once the parties had by their arbitration clause submitted to the curial jurisdiction of the English court, there was no discretion to refuse an anti-suit injunction on forum non conveniens grounds. Even if the arbitration clause was found by the English court to be governed by Russian law, expert evidence could be admitted in order to determine its meaning. Any view expressed by the Russian courts was thus advisory but not determinative. Finally, and following on from the last point, a contention that the arbitration clause is invalid under the law of the country in which the judicial proceedings have been brought is not a relevant consideration in the grant or otherwise of an anti-suit injunction.18 2.2

The Intervention of EU Law

The settled ability of an English court to grant anti-suit relief was severely restricted during the UK’s membership of the European Union, as a result of the operation of, successively, the Brussels Convention on jurisdiction and the enforcement of judgments in civil and commercial matters 1968, as replaced by a Council Regulation in 2001,19 and ultimately by the Brussels Regulation Recast in 2012.20 The Regulation operates a ‘first seised’ principle in art 27, whereby the court first seised of a dispute has exclusive jurisdiction over it. The Court of Justice of the European Union ruled in Turner v Grovit21 that the first seised principle applied to exclusive jurisdiction clauses, so that if a question as to the validity of an exclusive jurisdiction was raised in judicial proceedings in the courts of one Member State other than the courts of a state which were to have exclusive jurisdiction, then that question could only be considered by the former. Indeed, any decision on the point was conclusive and had to be recognised and enforced in all other EU courts. The main effect of this ruling was to preclude any possibility of the English courts granting an anti-suit injunction against a claimant who had commenced proceedings elsewhere in the EU in apparent breach of an English exclusive jurisdiction clause, as the English court would not in such circumstances be first seised. It was initially unclear whether the same principle applied to an anti-suit injunction to restrain a claimant from commencing judicial proceeding elsewhere in the EU in breach of an arbitration clause. There was good reason to assume that Turner v Grovit would not be so extended, given that art 1(2)(d) of the Brussels Regulation excludes ‘arbitration’. The point arose in Allianz SpA v West Tankers Inc, The Front Comor.22 The lengthy litigation arose from a collision by Front Comor with a jetty in Syracuse. The vessel was owned by West Tankers and chartered by Erg, and the jetty itself belonged to Erg. Erg’s insurers paid up to the limits of indemnity, and Erg commenced arbitration in London against West Tankers in accordance with the arbitration clause in the charterparty. However, the insurers sought to recover their payment by way of subrogation in judicial proceedings in Syracuse. West Tankers sought an anti-suit injunction. This was a straightforward two-party contractual claim, in that the insurers were clearly bound by the arbitration clause by relying upon derivative rights.

18 Riverrock Securities Ltd v JSC International Bank of St Petersburg [2020] EWHC 2483 (Comm); QBE Europe SA/NV v Generali Espana De Seguros Y Reaseguros [2022] EWHC 2062 (Comm). 19 Council Regulation 44/2001. 20 European Parliament and Council Regulation 1215/2012. 21 [2004] ECR I–3565. 22 [2007] UKHL 4 (HL); Case C-185/07 (CJEU).

188  Research handbook on marine insurance law The House of Lords was firmly of the view that the Regulation did not preclude the grant of anti-suit relief, in that such an action was exclusively about ‘arbitration’ and it sought to protect the right that disputes were to be resolved by arbitration. However, a reference was made to the Court of Justice for a preliminary ruling. In a brief judgment the Court of Justice ignored the blatant steer by the House of Lords and ruled that if the substantive claim fell within the Brussels Regulation,23 then ‘a preliminary issue concerning the applicability of an arbitration agreement, including in particular its validity, also comes within its scope of application’. That meant that the validity of the arbitration clause was to be determined exclusively by the Syracuse court. This judgment was greeted with dismay by those engaged in UK arbitration, as on the face of things it became possible to disregard an arbitration clause and to commence local proceedings, hoping – often with good reason – that the local court would refuse to give effect to the arbitration clause. The ingenuity of the legal profession nevertheless rapidly found ways to overcome this ruling. Three devices have been employed. First, s 38 of the Arbitration Act 1996 empowers the parties to confer upon the tribunal itself the power to grant an interim anti-suit injunction, although if that power is not conferred then it is not included in the list of default powers open to the tribunal under s 38. In principle EU law could have no objection to the exercise of that power: an arbitral tribunal is not a court and thus is not bound by the Brussels Regulation. That conclusion was confirmed by the Court of Justice in Gazprom OAO v Lietuvos Respublicka,24 a decision under the identical wording in the Brussels Regulation Recast. Second, the right of the tribunal under English law (s 30 of the 1996 Act) to determine its own jurisdiction means that irrespective of the existence or progress of any overseas judicial proceedings the tribunal can simply reject any application by the arbitration respondent to stay the arbitration and carry on with the arbitral process.25 By seeking and obtaining an immediate declaration that the arbitration clause is binding on the parties, the arbitration claimant can apply to the Court under s 66 of the 1996 Act for the enforcement of the declaratory award. It was at one time thought that a declaratory award was incapable of enforcement, but it was held in subsequent proceedings in The Front Comor litigation, West Tankers Inc v Allianz SpA,26 that a declaratory award can be the subject of an enforcement order if such an order would serve some useful purpose. Armed with a judgment under s 66, it is possible for the arbitration claimant to resist the enforcement in England of any subsequent judgment of a foreign court awarding damages to the arbitration defendant. It emerges also from that case that the tribunal may grant a declaration to the effect that the arbitration claimant is entitled to damages for breach of the arbitration clause by the arbitration defendant, and is entitled also to an indemnity against any damages that might be awarded in the foreign judicial proceedings in circumstances where that judgment is enforced against the arbitration claimant in a third jurisdiction. Third, it is possible to add to the claim in the arbitration a further claim for damages for breach of the arbitration clause. The tribunal will doubtless have jurisdiction to entertain such a claim, and a damages award is enforceable under s 66 even if the enforcement order is made

Para 26. Case C-536/13. 25 Noori Holding Ltd v Public Joint-Stock Co Bank Otkritie Financial Corporation [2018] EWHC 1343 (Comm). 26 [2012] EWCA Civ 27. 23 24

Protection & indemnity clubs and arbitration clauses  189 after the foreign court has awarded damages: that is the result of s 32 of the Civil Jurisdiction and Judgments Act 1982, under which a foreign judgment given in proceedings brought in breach of an exclusive jurisdiction or arbitration clause may not be enforced in the English courts.27 The principle extends to an arbitration defendant who, while not a party to an arbitration clause, relies on derived rights from a party, most importantly a subrogated insurer.28 It had been hoped that the introduction of the Brussels Regulation Recast in 2012 would result in the abolition of the bar on anti-suit injunctions. It did so partially for exclusive jurisdiction clauses, art 31(2) reversing the result in Turner v Grovit by providing that any court other than the court identified as having exclusive jurisdiction is to stay its proceedings, thereby removing the need for anti-suit relief. However, the Recast Regulation left arbitration untouched, merely modifying the recitals relating to arbitration. Those modifications were enough to persuade the Advocate General in Gazprom OAO v Lietuvos Respublicka that anti-suit injunctions had been authorised by the Recast Regulation, but the Court of Justice completely ignored the issue and satisfied itself with ruling that the tribunal itself was not precluded from granting such relief. Thus, as far as EU law is concerned, judicial anti-suit relief to protect arbitration clauses is not available, although the withdrawal of the UK from the EU has left the issue entirely academic in the UK and all but academic elsewhere in the EU given that the anti-suit injunction is largely a creation of English procedure. One final issue of EU law should here be addressed briefly. If the Brussels Regulation is applicable, in that ‘arbitration’ is not the issue before the court, the general rule in art 4 of the Regulation is that a person domiciled in an EU member state can be sued only in the member state of his or her domicile. There are also special insurance rules whereby a ‘policyholder, insured or beneficiary’ who is resident in an EU state can under art 14 in respect of a ‘matter relating to insurance’ be proceeded against by the insurers only in the courts of that state.

3.

FOREIGN JUDICIAL PROCEEDINGS BY A THIRD PARTY: QUASI-CONTRACTUAL CLAIMS

3.1

Direct Claims against P&I Clubs under Foreign Laws

A key consideration for P&I Clubs is the enforcement of arbitration and pay to be paid clauses when a direct claim is brought by a third party in a foreign court. That third party will be either the victim of the wrongdoing asserted against the shipowner, or the subrogated insurers of the victim. It is immaterial whether the claim is personal or subrogated, as it is trite law that a subrogated insurer obtains no better rights than its assured. Subrogated insurers are subject to arbitration clauses in exactly the same way as the insured victim,29 as are other parties seeking 27 CMA CGM SA v Hyundai Mipo Dockyard Co Ltd [2008] EWHC 2791. (Comm); AdActive Media Inc v Ingrouille [2021] EWCA Civ 313. 28 Argos Periera Espana SL v Athenian Maritime Ltd [2021] EWHC 554 (Comm). 29 Schiffahrtsgesellschaft Detlev von Appen GmbH v Voest Alpine Intertrading GmbH, The Jay Bola [1997] 2 Lloyd’s Rep 279; West Tankers Inc v Ras Riunione Adriatica di Sicurta SpA, The Front Comor [2005] EWHC 454 (Comm); Through Transport Mutual Insurance Association (Eurasia) Ltd v New India Assurance Co Ltd, The Hari Bhum [2004] EWCA Civ 1598; Royal Bank of Scotland plc v Highland Financial Partners LP [2012] EWHC 1278 (Comm); Shipowners’ Mutual Protection and

190  Research handbook on marine insurance law to exercise rights derived from the victim who are attempting to sidestep the arbitration clause.30 Assignees are equally bound by arbitration clauses.31 The initial question is whether the law of the country in which the claim is brought recognises a direct claim. If so, it becomes necessary to move to the second stage and to ask whether the relevant law requires the claimant to rely upon the terms of the insurance contract or confers a wholly independent right of recovery free of policy obligations imposed upon the assured. It is only where there is a direct claim contingent on policy terms that the third question arises – will the English courts grant an injunction to restrain a quasi-contractual direct claim of this nature? The classification question has come before the English courts on a number of other occasions, and on each of them it has been concluded – after the reception of detailed evidence from local lawyers – that the relevant foreign law operated to confer rights based upon contract and thus was to be classified under English law32 as derivative rather than independent.33 The starting point is the decision of the Court of Appeal in Through Transport Mutual Insurance Association (Eurasia) Ltd v New India Assurance Co Ltd, The Hari Bhum.34 Here, the insurers of cargo owners commenced subrogated proceedings in Finland against the shipowner’s P&I Club, disregarding provisions for arbitration in London and for pay to be paid. The shipowners became insolvent. The claim was brought under Finnish legislation that permitted a direct claim against the insurers of an insolvent policyholder. The Club sought anti-suit relief in England. It was not disputed before Moore-Bick J at first instance35 that there was a direct claim under Finnish law, and thus it became necessary for him to classify that legislation as conferring either a derivative claim or an independent claim. His approach was that it was necessary to examine the nature of the Finnish legislation and then to classify it according to English law and not in terms of the classification that might have operated under Finnish law. Applying that two-step analysis of the Finnish legislation, Moore-Bick J’s conclusion was that the effect of Finnish law was to transfer to the third party claimant exactly the same rights as had been possessed by the assured; such a measure was to be classified as a matter of English law as a derivative claim. It followed that the Finnish proceedings had been brought in contravention of the arbitration clause. The Court of Appeal in The Hari Bhum affirmed Moore-Bick J’s analysis.

Indemnity Association (Luxembourg) v Containerships Denizcilik Nakliyat ve Ticaret AS, The Yusuf Cepnioglu [2016] 1 Lloyd’s Rep 641; Fair Wind Navigation SA v ACE Seguradora SA [2017] EWHC 3352 (Comm); Airbus SAS v Generali Italia [2019] 2 Lloyd’s Rep 59; QBE Europe SA/NV v Generali Espana de Seguros y Reaseguros [2022] EWHC 2062 (Comm). 30 Bannai v Erez [2013] EWHC 3689 (Comm) and the authorities there cited. 31 Montedipe SpA v JTP-RO Jugotanker, The Jordan Nicolov [1990] 2 Lloyd’s Rep 11. 32 This is a far from easy conclusion to draw, despite the fact that the English courts have adopted it without any real dispute on the point. See Raphael, paras 10.31 to 10.65, for detailed discussion of the applicable law in the classification process. See also Dickinson, ‘The Right to Rome? The Law Applicable to Direct Claims against Insurers and Anti-Suit Injunctions’ 132 LQR 536 (2016). 33 Riverrock Securities Ltd v International Bank of St Petersburg [2020] EWHC 2483 (Comm); Aspen Underwriting Ltd v Credit Europe Bank NV, The Atlantik Confidence [2020] UKSC 11; QBE Europe SA/NV v Generali Espana de Seguros y Reaseguros [2022] EWHC 2062 (Comm); UK P&I Club NV v Republica Bolivariana De Venezuela, The RCGS Resolute [2022] EWHC 1655 (Comm) (where the court was not persuaded that there was any cut-through provision under Venezuelan law). 34 [2004] EWCA Civ 1598. 35 [2003] EWHC 3158 (Comm).

Protection & indemnity clubs and arbitration clauses  191 The twin points, that the classification of cut-through legislation is a matter for English law rather than the law governing the legislation itself and that restrictions on the reliance of policy terms by insurers,36 are unlikely to be fatal to a finding by the English court that the legislation is derivative, were confirmed by the Court of Appeal in London Steam Ship Owners’ Mutual Insurance Association Ltd v Kingdom of Spain, The Prestige (No 2).37 The complexities of this case are discussed later in this chapter, and here it suffices to focus on the issues arising from the Spanish direct claims legislation. This imposed various restrictions on reliance by insurers on their ordinary rights, including limitation periods. However, the Court of Appeal concluded that the ability conferred upon a third party under the legislation to claim from liability insurers was by its nature contractual, so that the arbitration clause was valid and binding. Moore-Bick LJ gave some guidance on what a foreign law conferring independent rather than contractual rights might look like, and in his view, what was required was some form of redefinition of liability which did not mirror in substance the scope of the contractual obligations. The next case in the line is Shipowners’ Mutual Protection and Indemnity Association (Luxembourg) v Containerships Denizcilik Nakliyat ve Ticaret AS, The Yusuf Cepnioglu,38 which also had the twist of a restriction on the rights of insurers to rely upon all of their rights in a cut-through case. This concerned a Turkish cut-through statute conferring the right to bring a direct claim but outlawing reliance on pay to be paid clauses. The Court of Appeal, affirming the decision of Teare J, concluded that the Turkish legislation had the effect of requiring the claimant to rely upon the policy terms in order to give rise to a cause of action, which meant that the legislation was to be classified under English law as conferring derivative rather than independent rights. The prohibition on the enforcement of pay to be paid was held not to alter the basic nature of the legislation. It followed that the arbitration clause was valid, and it also followed that the restrictions on the Club’s rights under Turkish law had no relevance in any subsequent London arbitration, where the policy would be enforceable in its entirety. Finally, it may be noted that the position established in the English cases has been followed in Hong Kong. In AIG Insurance Hong Kong Ltd v McCullogh,39 a decision of retired High Court judge Sir William Blair, the claimant, who was resident in Texas, fell from a zip-wire while holidaying in St Lucia. A claim was made against the director of the operators, and his Directors and Officers Liability Policy required disputes to be resolved in arbitration in Hong Kong. The Policy excluded personal injury claims, and unsurprisingly the insurers denied liability. The claimant commenced proceedings in Miami against the operators and the director, and applied to have the insurers added as additional defendants. A separate claim was made against the insurers, namely that they had acted in bad faith in refusing indemnity. Sir William Blair held that the claimant’s action against the insurers was a contractual rather than extra-contractual claim and accordingly it was one subject to the terms of the policy including the arbitration clause.

That is in any event the position under the 2010 Act itself, as regards pay to be paid clauses and also claims co-operation clauses. 37 [2015] EWCA Civ 333. 38 [2016] EWCA Civ 386. 39 [2019] HKCFI 1649. 36

192  Research handbook on marine insurance law 3.2

Anti-suit Relief in Derived Rights Cases40

Once it is established that foreign cut-through laws operate to confer a direct action on a claimant, and also that the direct action is to be classified as a matter of English law as derivative on the assured’s rights and thus contractual rather than independent of the contract, the key question arises as to whether the claimant should be restrained by anti-suit injunction from pursuing a claim other than in arbitration in London. This is the province of the quasi-contractual anti-suit injunction. After an uncertain start, the English courts have reached the settled position that anti-suit relief is to be granted on exactly the same basis as in direct contractual claims, namely, that an applicant is entitled to an injunction if it can demonstrate to a high degree of probability that there is a valid arbitration clause extending to the dispute. It is then to the defendant to show strong reasons why relief should not be granted. The uncertain start was that in The Hari Bhum. The Court of Appeal, having held that the respondent had been in breach of an English arbitration clause by suing in Finland under Finnish cut-through legislation, declined to grant an anti-suit injunction and instead issued a declaration. Its view was that declaratory relief would preclude the enforcement of any Finnish judgment in England, and that sufficed to protect the Club. The Court’s reluctance to grant an injunction was doubtless based on caution in sanctioning a new remedy and also concerns that such relief was not permitted within the EU (the latter fear subsequently being confirmed by Turner v Grovit). Although that cautious approach was initially accepted,41 it was abandoned by Teare J and the Court of Appeal in The Yusuf Cepnioglu, a case with all but identical facts although involving Turkish rather than Finnish legislation. EU considerations were not relevant here, and there was little hesitation in protecting the Club by injunction against the respondent’s pursuit of Turkish proceedings in which the Club would not just lose the chosen dispute resolution mechanism of arbitration but would also be denied the protection of the pay to be paid clause. The approach in The Yusuf Cepnioglu was followed by the Court of Appeal in London Steamship Owners’ Mutual Insurance Association Ltd v Spain and The French State, The Prestige (Nos 3 and 4).42 More recently, in QBE Europe SA/NV v Generali Espana de Seguros y Reaseguros,43 Generali were the insurers of an undersea cable allegedly damaged by a yacht insured against liability by QBE UK covering the year 2016 when the damage occurred. The QBE policy contained a London arbitration clause, but Generali – having indemnified the owners of the cable – exercised subrogation rights to proceed against QBE UK in Spain. Foxton J, having classified the Spanish direct action as one derived on the policy, granted anti-suit relief to QBE UK. This line of authority is far from immune from criticism,44 but is now well established.

Raphael, paras 10.66–18.80. Markel International Co Ltd v Craft, The Norseman [2006] EWHC 3150 (Comm). 42 [2021] EWCA Civ 1589. See also: Argos Periera Espana SL v Athenian Maritime Ltd [2021] EWHC 554 (Comm); UK P&I Club NV v Republica Bolivariana De Venezuela, The RCGS Resolute [2022] EWHC 1655 (Comm). 43 [2022] EWHC 2062 (Comm). 44 Briggs, ‘Direct Actions and Arbitration: All at Sea’ [2016] LMCLQ 238. 40 41

Protection & indemnity clubs and arbitration clauses  193 3.3

Anti-suit Relief in Other Quasi-contractual Cases

There is a second form of quasi-contractual anti-suit injunction apparent in the decided cases. This arises where the applicant for the injunction does not assert that it was a party to an arbitration clause, but nevertheless argues that the claimant cannot bring proceedings against the applicant other than in arbitration. This is one step removed from the more obvious form of quasi-contractual claim. The authorities were analysed by Cockerill J in Times Trading Corporation v National Bank of Fujarah (Dubai Branch),45 where the Bank, the holder of 27 bills of lading containing London arbitration clauses, commenced judicial proceedings in Singapore for damages against Times as bareboat charterer for mis-delivery of the cargo following its release without the production of the bills of lading. Times contended that it was not the bareboat charterer and thus was not a party to the arbitration clauses in the bills of lading, but nevertheless sought anti-suit relief. This was granted by Cockerill J, who accepted that there was no consistent juridical underpinning in the decided cases on this scenario46 in that some focused on the vexatious or oppressive nature of the foreign proceedings and others on the notion that the claimant in the foreign proceedings was bound by the arbitration clause in the contract whether or not the applicant itself was a party to the contract.47 She was nevertheless able to derive from them the proposition that The Angelic Grace should be applied by analogy and that no distinction should be drawn between the ‘derived right’ quasi-contractual scenario and that where the applicant for anti-suit relief was not itself relying on the contract. The test for relief is thus the same as for contractual and derived rights cases, namely that the court must be satisfied by the applicant to a high degree of probability that there is a binding arbitration clause, and the respondent then has the burden of demonstrating strong reasons why relief should not be granted. That approach was applied in the insurance context in QBE Europe SA/NV v Generali Espana de Seguros y Reaseguros,48 the basic facts of which were given above. To those facts should be added that the policy under which the direct action was brought had originally been issued by QBE UK, but all of QBE UK’s rights had been transferred to QBE Europe in 2020. The direct claim had nevertheless been brought against both QBE UK and QBE Europe. Foxton J, having held that QBE UK as a contracting party was entitled to an anti-suit injunction, held that QBE Europe should not be treated in any different way even though it was not a party to the policy under which the claim had been brought. Foxton J commented that Generali was seeking to assert a contractual right without respecting the condition of that right requiring the claim to be asserted in arbitration in England.

[2020] EWHC 1078 (Comm). The scope and indeed validity of this jurisdiction has thus been questioned: Myburgh, ‘Non Parties, Forum Agreements and Expanding Anti-Suit Injunctions’ [2020] LMCLQ 345. For a more detailed analysis, see Justice Belinda Ang, ‘Anti-Suit Injunctions in Maritime Disputes: A Trend that Threatens to Be Out of Control’, National University of Singapore, Centre for Shipping Law Working Paper, March 2021. 47 Sea Premium Shipping Ltd v Sea Consortium Pte Ltd [2001] EWHC 540 (Admlty); Jewel Owner Ltd v Sagaan Developments Trading Ltd, The MD Gemini [2012] EWHC 2850 (Comm); Dell Emerging Markets (EMEA) Ltd v IB Maroc.com SA [2017] EWHC 2397 (Comm); Clearlake Shipping Pte Ltd v Xiang Da Marine Ltd [2019] EWHC 1536 (Comm); Hai Jiang 1401 Pte Ltd v Singapore Technologies Marine Ltd [2020] 4 SLR 1014 (a decision of Belinda Ang J herself). 48 [2022] EWHC 2062 (Comm). 45 46

194  Research handbook on marine insurance law

4.

SOVEREIGN IMMUNITY

The State Immunity Act 1978 becomes relevant where the claimant against the P&I Club is a sovereign state. The effect of s 1, subject to defined exceptions, is to confer immunity from suit upon a head of state and the government and its departments of that state, although there is no immunity for any ‘separate entity’ that is distinct from the state’s executive organs (s 14). It is always open to a person entitled to judicial immunity to submit to the jurisdiction of the English court either expressly in writing or by initiating or participating in judicial proceedings (s 2). Sovereign immunity is a concept that is potentially in conflict with the guarantee of the access of private persons to the courts under art 6 of the European Convention on Human Rights, and is thus given a narrow ambit,49 but it may still impose important limitations on the enforcement of contractual rights against that state. There are two important limitations on state immunity relevant to the present discussion. First, there is no immunity as respects proceedings relating to a commercial transaction entered into by a state (s 3(1)(a)), defined as meaning any contract of sale or supply (s 3(3)(a)), any loan or related transaction (s 3(3)(b)) and ‘any other transaction or activity (whether of a commercial, industrial, financial, professional or other similar character) into which a State enters or in which it engages otherwise than in the exercise of sovereign authority’ (s 3(3)(c)). It is settled law that these concepts are very broad and are to be construed generously against the state pleading that its activity is not commercial.50 Second, by s 9(1), ‘Where a State has agreed in writing to submit a dispute which has arisen, or may arise, to arbitration, the State is not immune as respects proceedings in the courts of the United Kingdom which relate to the arbitration’. It was held in Through Transport Mutual Insurance Association (Eurasia) Ltd v New India Assurance Co Ltd (No 2)51 that if a third party made a claim against a Club in overseas cut-through proceedings then it had agreed in writing to submit the dispute to arbitration. This provision is nevertheless subject to s 13, which states in s 13(2)(a) that an injunction may not be granted against a state, and in s 13(2) (b) and s 13(4) that the property of a state shall not be subject to any process for enforcing an arbitration award unless it is property used or intended to be used for commercial purposes. Any issue between a state and a P&I Club will inevitably occur only in the context of ‘quasi-contractual’ claims, where a state has brought proceedings in its own courts against a P&I Club to recover compensation for liabilities incurred to the state by an insured shipowner. The operation of the sovereign immunity legislation in that situation was considered by Sir Ross Cranston in UK P&I Club NV v Republica Bolivariana De Venezuela, The RCGS Resolute.52 The case concerned a collision between the cruise liner Resolute and the patrol vessel Naiguatá belonging to the Republic of Venezuela. The Republic commenced proceedings in the courts of Venezuela and Curaçao local courts against the owners of Resolute and against the P&I Club in which she had been entered. The claim against the Club was brought under domestic legislation in Venezuela which, it was contended, recognised such an action. The issue was whether an anti-suit injunction should be granted. Benkharbouche v Secretary of State for Foreign and Commonwealth Affairs [2017] UKSC 52. Kuwait Airways Corporation v Iraqi Airways [1995] 1 WLR 1147; NML Capital Ltd v Republic of Argentina [2011] UKSC 31; Benkharbouche v Embassy of The Republic of Sudan [2017] UKSC 6. 51 [2005] EWHC 455 (Comm). 52 [2022] EWHC 1655 (Comm). 49 50

Protection & indemnity clubs and arbitration clauses  195 Sir Ross Cranston concluded that even if it was the case that if the law of Venezuela did allow a direct action – and that proposition had not been fully established – then at best it was one to be classified as based upon the terms of the insurance. Accordingly, the claim was subject to the contractual arbitration and pay to be paid provisions. However, the outstanding question was whether anti-suit relief could be granted to give effect to those provisions. The Court ruled initially that adjudicative immunity was not available under either of s 3 or s 9, but that there was no remedial jurisdiction to grant anti-suit relief by reason of s 13. Turning first to s 3, the meaning of the term ‘any other transaction’ in s 3(3)(c) turned on the distinction between a claim arising out of an exercise of sovereign authority and therefore possible only by a state, and a claim that could be made by a private individual whether or not it was actually brought by a state.53 As long as the claim related to a matter with a ‘commercial […] character’ as required by s 3(3)(c) then it was removed from sovereign immunity even though a state was a party. That test was easily satisfied in the present case. The claims brought by Venezuela arose out of a collision, and damages were sought for physical harm done. That was a straightforward civil claim of a type that could be brought by any private individual suffering similar loss, and the claim was brought under the terms of a commercial contract of insurance. As far as the arbitration exception in s 9 was concerned, Sir Ross Cranston’s ruling – made without the citation of authority54 – was that once Venezuela had brought its direct action it had adopted the terms of the Club’s rules setting out insurance coverage, including the arbitration and pay to be paid clauses. There was accordingly an implicit agreement in writing to submit the dispute to arbitration. The case thus turned upon s 13. Sir Ross Cranston drew a distinction between the adjudicative jurisdiction set out in ss 3 and 9 of the 1978 Act and the enforcement jurisdiction in s 13, and held55 that s 13(2)(a) conferred an immunity which was not affected by the limits on state immunity in the adjudicative sections of the legislation. The only real debate was as to the compatibility of s 13(2)(a) with art 6 of the European Convention on Human Rights. The Court had no doubt that the terms of art 6 had been ‘engaged’ by a provision that sought to prevent ordinary private judicial relief, but found that the bar on injunctive relief against a state nevertheless constituted a legitimate and proportional response. The test as laid down by Lord Sumption56 was that ‘in the absence of a recognised rule of customary international law, article 6 is satisfied if the rule applied by a Convention state lies within the range of possible rules consistent with “current international standards”’. Applying that test, Sir Ross Cranston found that ‘There is no clear and settled view in customary international law regarding orders for injunctions and specific performance against states in proceedings relating to their non-sovereign activity or otherwise’. That made it necessary to turn to the second limb of the Sumption test, and on this point the Court was satisfied that s 13(2)(a) lay ‘within the range of possible rules consistent with current international standards’. A number of other considerations persuaded Sir Ross Cranston that the immunity should stand: the sanction for failing to observe an anti-suit injunction was imprisonment for contempt of court, but that was wholly inappropriate against a state; there was considerable The Prestige (Nos 3 and 4) [2021] EWCA Civ 1589. Unnecessary, in the light of The Prestige (Nos 3 and 4) [2021] EWCA Civ 1589. 55 Applying Alcom Ltd v Republic of Colombia [1984] AC 750. 56 In Benkharbouche v Secretary of State for Foreign and Commonwealth Affairs [2017] UKSC 62.

53 54

196  Research handbook on marine insurance law international sensitivity on the subject; and the Court had to be aware of issues of comity and procedural propriety. The outcome, therefore, is that where a state brings a ‘quasi-contractual’ claim in its own courts (or indeed those of another country) against a P&I Club, the English courts may not restrain the state’s conduct. It is thus necessary to revert to the remedies developed in the period during which the UK was governed by the Brussels Convention and Brussels Regulations. Those remedies are, as already noted, damages for breach of the arbitration clause and refusal by the English courts to recognise and enforce judgments obtained by states against P&I Clubs.

5. THE PRESTIGE: A CASE STUDY 5.1

The Factual Background

The vessel Prestige was entered into the London Steamship Club. The insurance contained the usual London arbitration clause and pay to be paid provisions, and the limit of indemnity was US$1 billion. In 2002 Prestige broke in two off Cape Finisterre, discharging much of its cargo of 70,000 tonnes of fuel oil. The result was severe pollution to the shorelines of Spain and France. Prosecutions were brought in Spain against the Master, Chief Officer and Chief Engineer, and the Owners were also prosecuted on the basis of their legal responsibility for the primary wrongdoers. In addition, civil claims were brought in Spain by Spanish and French victims. The Owners’ Club was joined to the proceedings. The Spanish authorities indemnified a number of the victims, and so the proceedings against the Owners and the Club were in part for losses suffered by Spain and France themselves, and in part by way of subrogation proceedings by Spain. The Spanish action proceeded to the Supreme Court, and on 26 January 2016 the Master was convicted of negligence. Judgment on liability was given against both the Master and the Owners, and the Club was found to be directly liable to Spain under Spanish cut-through legislation. A decision on quantum was referred to the Provincial Court, which on 1 March 2019 ruled that the liability was capped at €855,493,575.65 net of sums paid into the limitation fund established under the Civil Liability Convention 1992. In the meantime, the Club had taken steps to protect itself against any judgment in Spain. In January 2012 the Club commenced arbitration proceedings in London against both Spain and France. They refused to participate. A sole arbitrator was appointed, and on 13 February 2013 the arbitrator issued a declaratory award to the effect that the claimants in the Spanish proceedings were bound by the arbitration clause, that the pay to be paid clause meant that the Club was liable to provide indemnity to the Owners only once the Owners had themselves made payment, and that liability was capped at US$1 billion. Steps were then taken to enforce the arbitration award. An application was made to Hamblen J under s 66 of the Arbitration Act 1996 for enforcement on 4 March 2013. This was contested by Spain and France on jurisdictional grounds. In the course of the proceedings they additionally sought to have the award set aside for want of jurisdiction under s 67 of the 1996 Act and relied upon the right of a non-participating party to challenge an award under s 72. Hamblen J rejected the sovereign immunity claim and further held that the award should be enforced: an enforcement order could be relied upon as a defence to proceedings brought

Protection & indemnity clubs and arbitration clauses  197 against the Club elsewhere in the EU and an enforcement order would not subvert any Spanish proceedings. The matter went to the Court of Appeal in London Steam Ship Owners’ Mutual Insurance Association Ltd v The Kingdom of Spain, The Prestige (No 2).57 Spain’s argument was that it was entitled to sovereign immunity and so could not be bound by the enforcement order. The Court of Appeal disagreed on two grounds. First, Spain and France had, for the purposes of s 2 of the State Immunity Act 1978 submitted to the jurisdiction of the English court by contesting enforcement without any reservation of rights and then by seeking to have the award overturned. Second, although it was not necessary to decide the point, the decision in Through Transport Mutual Insurance Association (Eurasia) Ltd v New India Assurance Co. Ltd (No 2), that immunity was lost under s 9 of the 1978 Act by pursuing derivative proceedings in reliance on a policy containing an arbitration clause, was correct. 5.2

The English Proceedings

The position up to the Spanish judgment on quantum in March 2019 was thus that the Club had obtained an enforcement order on a declaratory award holding that the Club did not face liability to Spain and France until the Owner had made payment, whereas Spain and France had a judgment in Spain holding that the Club was liable. Armed with the quantum judgment, Spain and France sought to register the judgment in England under the terms of the Brussels Regulation 2001. In anticipation, in February 2019 the Club commenced an English action against Spain and France, seeking a declaration and damages for failure to honour the arbitration award. On 28 May 2019 Spain and France successfully obtained an order registering the Spanish judgment, although the Club appealed. That was followed in June 2019 by a second action by the Club seeking similar relief, this time for failure to honour the s 66 enforcement order relating to the mix. There were thus three separate sets of proceedings in the English courts. Added to the mix, the Club sought to commence second arbitrations against Spain and France, and applied to the court for the appointment of arbitrators under s 18 of the 1996 Act. In January 2019 the Club served notice of arbitration on France seeking damages and indemnity for any sums that might be awarded against the Club in Spain and seeking also injunctive relief preventing France from enforcing any Spanish judgment. France refused to appoint its own arbitrator, and Foxton J appointed retired Court of Appeal judge Dame Elizabeth Gloster as the sole arbitrator. The application for the appointment of an arbitrator in the Spanish arbitration came before Henshaw J in The Prestige (No 3).58 It was necessary for the Club to persuade the court that there was a good arguable case that the arbitrator would have jurisdiction to hear the claims made by the Club. Those claims were for: (i) a declaration that Spain would, by seeking to enforce the Spanish judgment, be in breach of its obligation under the arbitration clause to proceed only by way of arbitration, and a declaration that the tribunal had jurisdiction to grant an anti-suit injunction and to award compensation in the form of equitable compensation, damages for breach of contract and damages in lieu of an injunction; (ii) equitable compensation or damages for breach of the equitable obligation to arbitrate the claims brought in the Spanish proceedings, encompassing costs arising from Spain’s pursuit of the proceedings in Spain and the enforcement proceedings in England; (iii) damages for breach of contract for the [2015] EWCA Civ 333. [2020] EWHC 1582 (Comm).

57 58

198  Research handbook on marine insurance law sums identified in (ii), arising from Spain’s participation in the section 66 proceedings; (iv) an anti-suit injunction to restrain breach of the equitable obligation to arbitrate, or damages in lieu under (ii) above; and (v) an order that Spain withdraw the claims brought in the Spanish proceedings and cease recognition and enforcement attempts worldwide. Spain pleaded sovereign immunity under ss 2, 3 and 9 of the 1978 Act, but its defence was rejected. Henshaw J proceeded to appoint an arbitrator, but excluded claim (iii) on the basis that it did not fall within the scope of the arbitration clause in the Club’s rules. The effect of Henshaw J’s judgment was that an arbitrator – retired High Court judge Sir Peter Gross – was appointed to hear claims for damages for failure to honour the award. Henshaw J’s ruling that an arbitrator would have no jurisdiction to hear any claim in respect of the enforcement judgment was the subject of a separate action in the High Court, The Prestige (No 4).59 Here, the Club sought damages against Spain and France on two separate grounds: failure to honour the arbitration award of 13 February 2013 (the Award Claims); and failure to honour the Court of Appeal’s judgment enforcing the award (the Judgment Claims). This action came before Butcher J, who ruled that: Spain and France were not entitled to sovereign immunity; the Award Claims gave rise to a serious issue to be tried and should be heard; and the Judgment Claims were not justiciable in England in that they were within the Brussels Regulation – they were not about ‘arbitration’ and so were not excluded by art 1(2)(d), but they were ‘matters relating to insurance’ and thus fell within the special jurisdiction rules for insurance claims under the Brussels Regulation whereby an action by an insurer against a policyholder or injured party could be brought only in the courts of the domicile of the defendant, so that the Judgment Claims had to be brought against France and Spain in their respective courts. 5.3

The Court of Appeal’s Second Judgment

The judgments of Henshaw J and Butcher J were considered by the Court of Appeal in a combined appeal, The Prestige (Nos 3 and 4).60 The initial issue was sovereign immunity. The Court of Appeal, agreeing with each of the first instance judges, held that immunity had been lost under s 3(1)(a) of the 1978 Act in that the actions of the French and Spanish states had, in seeking to enforce the Spanish judgment in England, engaged in ‘commercial activity’. In the Court of Appeal’s view ‘the activity in question comprised the pursuit of civil claims which constituted an attempt to enforce the terms of the insurance contract under a direct action right conferred by Spanish law’ and thus was contractual in nature. The Court of Appeal rejected the states’ argument that a distinction was to be drawn between ‘judicial’ and ‘commercial’ activity. Both the Judgment Claims and the Award Claims were the result of an attempt to enforce the insurance policy, and the application for the appointment of an arbitrator was – as held by Henshaw J – an issue that arose purely from the commercial activity of seeking to enforce the Spanish judgment. That ruling made it unnecessary to consider whether immunity had been lost under s 9 of the 1978 Act in that Spain and France had agreed in writing to submit the dispute to arbitration by relying upon the Club rules. The Court of Appeal was here happy to follow its own previous ruling that bringing proceedings on the policy satisfied the requirements of s 9.

[2020] EWHC 1920 (Comm). [2021] EWCA Civ 1589.

59 60

Protection & indemnity clubs and arbitration clauses  199 The final issue was whether the English court had jurisdiction to hear the various actions brought by the Club. At the time of the proceedings, Brexit had yet to take effect, and so it remained for the Court of Appeal to consider the effect of the Brussels Regulation. The appointment of an arbitrator was straightforward. The Court of Appeal agreed with Henshaw J that Spain and France were parties to an agreement to arbitrate. The dispute was thus one about ‘arbitration’ and fell outside the Brussels Regulation. An arbitrator could therefore be appointed. As regards the Award Claims brought in the High Court, there was no jurisdictional issue. These were plainly about ‘arbitration’ within the exclusion in art 1(2)(d) and thus the Brussels Regulation had no application to them. However, the nature of the Award Claims was somewhat problematic. Butcher J had ruled at first instance that they gave rise to a serious issue to be tried in three respects: an action lay at common law against a person who failed to honour an award, on the basis of breach of an implied term, and a fresh cause of action arose after the award had been made; remedies existed against a third party who sought to rely on the insurance contract by reason, in equity, under contract and in tort; and the obligation to honour the award did not merge into a judgment enforcing an award under s 66 of the 1996 Act. The Court of Appeal disagreed. The award in the present case was purely declaratory and stated merely that if Spain and France wished to claim against the Club, they had to do so in English arbitration proceedings. There was an implied promise in the arbitration agreement that it would be honoured, and no additional obligations arose from the award itself. Accordingly, the effect of the award was that it was capable of being enforced under s 66 of the 1996 Act, thereby precluding enforcement proceedings in England on any foreign judgment in conflict with the award, but failure to honour the award itself did not give rise to any additional substantive rights. The Judgment Claims were, in the view of the Court of Appeal – agreeing with Butcher J – too far removed from ‘arbitration’ to fall within that exemption. They were accordingly governed by jurisdiction rules in the Brussels Regulation. In so deciding, the Court of Appeal doubted the correctness of Butcher J’s preliminary view that a cause of action arose either because there was an implied contractual obligation to adhere to a Court judgment or because contempt of court in failing to adhere to a court order could be regarded as unlawful means for the purpose of establishing the tort of conspiracy to injure. However, there was no need for the Court of Appeal to deal with that question: even if such claims were possible, they were nothing to do with ‘arbitration’ and were thus governed by the Brussels Regulation. Given that the claims arose under an insurance policy they were matters relating to insurance within art 14 of the Brussels Regulation, with the result that the Club could proceed against France and Spain only in their respective courts. 5.4

The EU Sequel

There have been crucial sequels to all of this, involving the potential impact of EU law. The most important aspect for the Club of the prolonged litigation was the ruling in the first Court of Appeal decision that the declaratory awards were to be enforced, thereby precluding any enforcement by Spain and France of the Spanish judgment in England. The later litigation opened the door for the Club to claim damages in arbitration for the losses that it had suffered by having to face judicial proceedings in Spain in breach of the arbitration clause. The arbitra-

200  Research handbook on marine insurance law tions before Sir Peter Gross (Spain) and Dame Elizabeth Gloster (France) were designed for that purpose. However, there remained the problem that Spain and France and been successful in securing the registration of the Spanish judgment in England, so that it could be enforced in this jurisdiction. Butcher J decided that it was necessary to refer the registrability of the judgment for a preliminary ruling by the Court of Justice of the European Union. By way of background, art 33 of the Brussels Regulation requires a judgment given in a Member State to be recognised in all other Member States, the process in England being registration. Art 34 provides limited exceptions where ‘it is irreconcilable with a judgment given in a dispute between the same parties in the Member State in which recognition is sought’ (art 34(3)) or it is ‘manifestly contrary to the public policy’ of the enforcing state (art 34(1)). This provision is reflected in s 32 of the Civil Jurisdiction and Judgments Act 1982, although s 32 was – before Brexit – subject to the operation of the Brussels Regulation. The Club relied upon art 34(3), the argument being that the s 66 enforcement order was a ‘judgment’, and that it was inconsistent with the Spanish ruling so that the latter could not be registered. Three questions were referred to the CJEU for preliminary ruling on the effect of the Court of Appeal’s s 66 order: (1) Was the s 66 order a ‘judgment’ within art 34(3)? (2) Was the s 66 order taken outside the Brussels Regulation by the exclusion of ‘arbitration’ in art 1(2) (d)? (3) If it was the case that art 34(3) did not apply, could recognition and enforcement be refused on public policy grounds under art 34(1)? The CJEU ruled61 in London Steam-Ship Owners’ Mutual Insurance Association Ltd v Kingdom of Spain62 that the enforcement order was about ‘arbitration’ and thus excluded from the Brussels Regulation, which meant that it could not be recognised or enforced in Spain under art 33 of the Regulation. The CJEU nevertheless accepted that the Court of Appeal’s enforcement order was a ‘judgment’ for the purposes of art 34(3). On the face of things, therefore, the earlier English enforcement judgment took priority over the Spanish judgment and art 34(3) prevented any recognition of the Spanish judgment. However, there was a sting in the tail. In the CJEU’s view, the Regulation did not preclude the UK’s obligation to recognise and enforce the Spanish judgment. The CJEU ruled that where an award enforced in a domestic court ‘was made in circumstances which would not have permitted the adoption, in compliance with the provisions and fundamental objectives of that Regulation, of a judicial decision falling within the scope of that Regulation’, then the recognition judgment fell outside art 34(3) and was to be disregarded. In the present case the enforcement order offended the fundamental objectives of the Regulation in two respects. First, the subject matter was related to insurance, and under art 14 of the Brussels Regulation a claim by insurers against a ‘policyholder, assured or beneficiary’ could be brought only in the courts of the domicile of that person. The CJEU was of the view that the protection conferred by art 14 would be undermined if an arbitral award on a matter relating to insurance could be enforced by a judgment on the award. Second, the Spanish action had been commenced before the issue of the arbitration proceedings, which meant that

61 The decision of Butcher J to make a reference to the CJEU was ordered by the Court of Appeal to be reviewed, but before Butcher J had had a chance to do so, the CJEU proceeded to judgment. It was pure fortuity, therefore, that the case ever reached the CJEU. 62 [2022] EUECJ C-700/20.

Protection & indemnity clubs and arbitration clauses  201 the Spanish courts were first seized of the action and thus had exclusive jurisdiction under art 27 of the Regulation. 5.5

The Binding Effect of the CJEU Judgment

The CJEU judgment was handed down while the arbitrations before Sir Peter Gross and Dame Elizabeth Gloster were ongoing. The CJEU judgment taken at face value meant that EU law precluded any reliance on the Court of Appeal’s enforcement order and that the Spanish judgment holding that the Club had to make payment was to be given effect in England. However, the arbitrators took a different view. Sir Peter Gross, in a First Partial Award dated 6 January 2023, held that the CJEU judgment was not binding on the English courts and therefore was not binding on him as arbitrator. By a Second Partial Award dated 22 March 2023 Sir Peter held that: Spain had, by maintaining its civil claims in Spain and in seeking to enforce the Spanish judgments against the Club, acted in breach of its obligations not to pursue claims other than by London arbitration; any further steps taken by Spain in Spain or England to enforce the Spanish Orders would be in breach of its obligations not to pursue its claims other than by way of London arbitration; and if and when Spain obtained a final monetary judgment or any enforcement order in any jurisdiction, Spain would in that jurisdiction indemnify for that amount. Spain obtained permission to appeal against these awards for error of law, under s 69 of the Arbitration Act 1996. Spain’s argument was that Sir Peter had wrongly refused to apply the CJEU judgment. The Club for its part appealed against the registration of the Spanish judgment in England. In London Steam-Ship Owners’ Mutual Insurance Association Ltd v Kingdom of Spain, The Prestige,63 Butcher J dismissed Spain’s appeal and upheld the Club’s application to set aside the registration of the Spanish judgment. Butcher J first considered the effect of the Court of Appeal’s enforcement order, and held that it was irreconcilable with the later Spanish judgment in that the Court of Appeal had given effect to a pay to be paid clause which had been disregarded in the Spanish judgment. What then of the CJEU judgment? Butcher J held that it was not binding in England, for three reasons. First, Hamblen J had determined in 2013 that an enforcement order did not subvert any later Spanish proceedings judgment: there had been no appeal against that ruling, and it thus remained binding and formed the basis for an issue estoppel. Second, and perhaps more dramatically, the CJEU judgment had been reached without jurisdiction. The CJEU had power to give preliminary rulings only on the questions referred to it: here the CJEU had not been asked to determine the effect of the insurance rules or of the fact that the Spanish proceedings had been commenced before the arbitrations, and therefore its answers on those points were not binding. Further, the CJEU had purported to apply the law to the facts by holding that the Spanish proceedings had been commenced earlier, again something beyond its powers.64 Third, issue estoppel was a common law principle which took effect as part of public policy, and accordingly art 34(1) of the Regulation permitted the English court to refuse to give effect to the CJEU judgment insofar as it required the English court to refuse to give effect to its own earlier binding decision that the award should be enforced. [2023] EWHC 2473 (Comm). It was not necessary for Butcher J to determine the substantive correctness of the CJEU’s interpretation, but he expressed the view that he would not have followed it. 63 64

202  Research handbook on marine insurance law Butcher J also dismissed Spain’s appeals on substantive points. Sir Peter had not erred in holding that Spain had an equitable obligation to arbitrate and that breach of the duty gave rise to equitable compensation. This was in his view a logical development of the law. The outstanding dispute was the Club’s arbitration with France. This had not concerned the effect of registration and raised the question of remedies for breach of the obligation to arbitrate. On 8 February 2023 Dame Elizabeth Gloster published her First Partial Award holding that the Club was entitled to equitable compensation for the breach by France of its equitable obligation not to pursue claims other than in arbitration. By a Second Partial Award dated 2 May 2023, Dame Elizabeth confirmed that France would be in breach of its obligation if it sought to enforce the Spanish judgment and that it was liable to indemnify the Club for any sums awarded against it should there be foreign enforcement proceedings. France appealed against this award for error of law, and in French State v London Steam-Ship Owners’ Mutual Insurance Association Ltd65 Butcher J repeated his conclusion that the law now recognised equitable compensation for breach of an equitable obligation. One final point arises from the appeals to Butcher J in these cases. Sir Peter and Dame Elizabeth had both held that they had jurisdiction to grant injunctive relief against Spain and France respectively. Section 13(2)(a) of the State Immunity Act 1978 states that ‘relief shall not be given against a State by way of injunction or order for specific performance or for the recovery of land or other property’ in the absence of agreement. Section 48(5(a)) of the Arbitration Act 1996 confers upon a tribunal the same powers as a court ‘to order a party to do or refrain from doing anything’. Both arbitrators regarded the 1978 Act prohibition as merely imposing a bar on the grant of injunctive relief by a court and not by an arbitrator: Sir Peter had nevertheless refused relief on discretionary grounds, although Dame Elizabeth had granted an injunction. Butcher J held in each of the appeals that the arbitrators had erred in law. The effect of s 48(5) of the 1996 Act was to confer upon an arbitral tribunal the same powers as the court. However, if the court had no power, then neither has a tribunal. Section 13 went to jurisdiction and was not simply a limit on the exercise of a power. Those points aside, an injunction granted by a tribunal could not be enforced by a court by reason of state immunity.

6. CONCLUSIONS The complexities discussed in this chapter spring primarily from two sources: the Brussels Regulation and sovereign immunity. The former is now largely of historical interest, and the latter will be relevant in very few cases. The position of P&I Clubs should, therefore, in future cases, be more or less secure. If judicial proceedings are brought by the assured in breach of the arbitration clause in the insurance, an anti-suit injunction will follow as a matter of course. By contrast, if the proceedings are brought by a third party victim (including its subrogated insurers or assignees) against the Club, an anti-suit injunction will issue as long as the laws under which the proceedings are brought require some reliance by the third party on the terms of the insurance and the English court classifies the foreign law as derivative rather than 65 [2023] EWHC 2474 (Comm). The application for permission to appeal had been made after the 28 days permitted by s 70(3) of the 1996 Act, so much of the judgment is concerned with the questions whether the First Partial Award was sufficiently determinative to constitute an ‘award’ so as to trigger s 70(3) (yes) and whether time should be extended (yes, in respect of the issues relating to remedies).

Protection & indemnity clubs and arbitration clauses  203 independent. In practice, that is almost inevitably going to be the case. The absence of a clear juridical underpinning for extending anti-suit injunctions in this fashion has proved not to be a barrier to the creativity of English judges.

10. Taxonomizing third-party rights of direct action against marine liability insurers Paul Myburgh

1. INTRODUCTION The issue of third-party1 victims’ rights of direct action (TRDAs) against marine liability insurers (typically P&I Clubs)2 has gained significant prominence in recent years, thanks to high-profile transnational proceedings arising from the Prestige,3 Sea Endeavour I4 and Stolt Commitment5 incidents. These and other cases have highlighted some of the fundamental tensions and confusions arising from the judicial taxonomization of such TRDAs, their governing laws and jurisdictional issues. This chapter seeks to analyse through a comparative conflict-of-laws lens how courts have taxonomized the ‘peculiar world’6 of TRDAs brought against P&I Clubs. TRDAs are usually sourced in either international maritime Conventions or national statutes. Most of the international Conventions which provide for a TRDA follow the standard template of art 7.8 of the International Convention on Civil Liability for Oil Pollution Damage 1992 (the CLC):7 Any claim for compensation for pollution damage may be brought directly against the insurer or other person providing financial security for the owner’s liability for pollution damage. In such case the defendant may, even if the owner is not entitled to limit his liability according to Article V, paragraph 2, avail himself of the limits of liability prescribed in Article V, paragraph 1. He may further avail himself of the defences (other than the bankruptcy or winding up of the owner) which the owner 1 I will use the usual terminology here, although of course the victim is a ‘non-party’ to the policy: see Adrian Briggs, ‘Direct Actions and Arbitration: All at Sea’ [2016] LMCLQ 327, 329 n 11. 2 On which, see generally Steven J Hazelwood and David Semark, P&I Clubs: Law and Practice (4th edn, Informa 2010). 3 See, among others, The Prestige, Spanish Supreme Court ECLI: ES:TS:2018:4136; The Prestige, French Cour de Cassation ECLI:FR:CCASS:2019:C100370; The London Steam Ship Owners Mutual Insurance Association Ltd v The Kingdom of Spain and Another (The Prestige (Nos 3 and 4)) [2021] EWCA Civ 1589, [2022] 1 Lloyd’s Rep 539; London Steam-Ship Owners’ Mutual Insurance Association Ltd v The Kingdom of Spain M/T ‘Prestige’ (No 5) [2022] EWCA Civ 238, [2022] 4 WLR 39; London Steam-Ship Owners’ Mutual Insurance Association [2022] EUECJ C-700/20, ECLI:EU:C:2022:488. Also see Merkin, Chapter 9, section 4 for a detailed discussion of the Prestige litigation. 4 Assens Havn v Navigators Management (UK) Ltd [2017] EUECJ C-368/16, ECLI:EU:C:2017:546. 5 The Stolt Commitment I, Norwegian Supreme Court, HR-2018-869-A. 6 Briggs (n 1) 331. 7 On the Conventions, see D Rhidian Thomas, ‘Third Party Direct Rights of Action against Insurers under UK Law and International Maritime Liability Conventions’ in Abhinayan Basu Bal et al (eds) Regulation of Risk: Transport, Trade and Environment in Perspective (Brill 2023) 685, 709 ff; Johanna Hjalmarsson, ‘Direct Claims against Marine Insurers in the English Legal System’ (2010) 18 Asia Pacific LR 259, 262 f; Vibe Ulfbeck, ‘Direct Actions against the Insurer in a Maritime Setting: The European Perspective’ [2011] LMCLQ 293, 293 n 2.

204

Taxonomizing third-party rights of direct action against marine liability insurers  205 himself would have been entitled to invoke. Furthermore, the defendant may avail himself of the defence that the pollution damage resulted from the wilful misconduct of the owner himself, but the defendant shall not avail himself of any other defence which he might have been entitled to invoke in proceedings brought by the owner against him. The defendant shall in any event have the right to require the owner to be joined in the proceedings.

The obvious advantage of this type of (more or less) standardized Convention wording is that it provides universal and clear parameters for the quantum of the insurer’s liability, defences against the third-party victim, and procedural joinder. The wording does not, however, solve issues of jurisdiction or of which law will govern issues not covered by the uniform Convention liability regime,8 which will necessarily have to be referred to national rules of private international law. It also does not elaborate on the conceptual underpinnings of the TRDA created by the Conventions, although the wording does seem to suggest ‘an independent right to proceed against and claim compensation from the insurer who, in these circumstances, appears to be independently and primarily liable’9 rather than a transferred right, or one that is conditional on the insured’s default. Private international law issues become much more pronounced in the context of TRDAs sourced in national statutes. There is little uniformity of statutory approach to be found between different jurisdictions,10 or indeed sometimes between territorial units within federal systems.11 These national statutes run the gamut from relatively strictly circumscribed TRDAs, such as those contained in the Third Parties (Rights against Insurers) Act 2010 (UK)12 – which operate as a statutory assignment to a third-party claimant of an insolvent debtor’s rights to claim against a liability insurer, but only in certain circumstances and typically following a judgment against the tortfeasor insured13 – to more liberal and independent TRDAs, mainly found in civil law countries.14 These differences in national statutes may, and have, led to forum shopping, the issuing of anti-suit injunctions and even parallel conflicting judgments. P&I Club policies (via their Club rules) invariably include either exclusive arbitration or jurisdiction clauses15 which are designed to channel all disputes arising from P&I policies into the liability insurer’s preferred jurisdiction. This tends to encourage strategic forum

So, for example, in the Prestige litigation, the CLC liability was relatively straightforward – all the litigation centred on the non-CLC claims and the different conceptions of third-party RDAs at English and Spanish law: see for example The London Steam Ship Owners Mutual Insurance Association Ltd v The Kingdom of Spain and Another (The Prestige (No 2)) [2013] EWHC 3188 (Comm), [2014] 1 Lloyd’s Rep 309 [5]–[9]. See also https://​iopcfunds​.org/​incidents/​incident​-map/​#1916​-13​-November​-2002 for more detail on the Prestige incident and its aftermath. 9 Thomas (n 7) 711. 10 Ibid 686: ‘There is a broad division of opinion as to the question whether a legal link should be established between third party and insurer and, if so, to what extent and subject to what terms and conditions. This division is readily visible when global legal traditions and national laws are compared.’ 11 See Matthew J Pallay, ‘The Right of Direct Action: Issues Proceeding Directly against Marine Insurers’ (2016) 41 Tulane Maritime LJ 57 for an overview of the different approaches adopted in various US States’ direct action statutes. 12 On which, see Thomas (n 7) 690 ff. 13 Firma C-Trade SA v Newcastle Protection and Indemnity Association (The Fanti) and Socony Mobil Oil Co Inc v West of England Ship Owners Mutual Insurance Association (London) Ltd (The Padre Island (No 2)) [1990] 2 Lloyd’s Rep 191 (HL). 14 See Ulfbeck (n 7) 294 f; Thomas (n 7) 686. 15 See generally Hazelwood and Semark (n 2) Ch 21; Hjalmarsson (n 7) 260. 8

206  Research handbook on marine insurance law skirmishes, where the liability insurer seeks to enforce its choice-of-venue and governing law clauses against the third-party victim, while the victim seeks to bring its TRDA in a venue of its choosing, which may be more convenient and familiar or less costly, and may accord more generous rights and/or remedies. A survey of the choice-of-venue and governing law clauses included in P&I liability policies by the International Group of P&I Clubs (IGP&I),16 which currently insures around 90 per cent of the global shipping fleet,17 confirms a universal pattern of ‘homeward trend’, with English law and dispute resolution being stipulated in the Rules of those Clubs based in the UK;18 Norwegian law19 and arbitration in Oslo for Assuranceforeningen Gard and Assuranceforeningen Skuld; Swedish law and arbitration in Göteborg for The Swedish Club; Japanese law and arbitration20 for the Japan Ship Owners’ Mutual Protection and Indemnity Association; and New York law and the jurisdiction of the United States District Court for the Southern District of New York for the American Steamship Owners Mutual Protection and Indemnity Association Inc.21 The effect of these ‘homeward trending’ choice-of-venue and law clauses in the Rules is that, while the majority of P&I Club disputes involving TRDAs will be settled in London and on an application of English law, a sizeable minority will be dealt with under the EU legal framework, or in accordance with Japanese or New York law and jurisdiction. This chapter will analyse how the courts taxonomize TRDAs brought against P&I Clubs in each of these four legal regimes, evaluate the comparative lack of uniformity of approach, and make some concluding remarks about possible ways forward.

See www​.igpandi​.org/​. D Rhidian Thomas, ‘Direct and Third-Party Claims against P&I Clubs’ in D Rhidian Thomas (ed), The Modern Law of Marine Insurance: Volume 5 (Routledge 2023) Ch 3, 41. 18 The Britannia Steam Ship Insurance Association Ltd; The London Steam-Ship Owners’ Mutual Insurance Association Ltd (The London P&I Club); The North of England Protecting and Indemnity Association Ltd; The Shipowners’ Mutual Protection & Indemnity Association; The Standard Club Ltd; The Steamship Mutual Underwriting Association (Bermuda) Ltd; United Kingdom Mutual Steam Ship Assurance Association Ltd; and The West of England Ship Owners Mutual Insurance Association (Luxembourg). 19 Both Gard and Skuld expressly exclude the provisions of the Norwegian Insurance Contracts Act 1989 in their governing law clauses, presumably as an attempt to avoid the liberal TRDA conferred by s 7-6 of that Act. 20 Japan P&I Club Rules 2022, r 47: ‘Should any dispute arise between the Association and a Member in respect of the insurance contract between the Association and the Member, the dispute shall be referred to the arbitration by the Japan Shipping Exchange, Inc. (Ippan Shadan Hojin Nihon Kaiun Shukaijo), and any award of the arbitration shall be final and binding on the parties involved, provided, however, that, subject to agreement between all the parties involved, the dispute may be referred to the arbitration by an arbitrator registered with the London Maritime Arbitrators’ Association.’ 21 See Thomas (n 17) 41. 16 17

Taxonomizing third-party rights of direct action against marine liability insurers  207

2.

UNITED KINGDOM APPROACH

In the United Kingdom, TRDAs against insurers are governed by the Third Parties (Rights Against Insurers) Act 1930 (the 1930 UK Act),22 now updated and amended as the Third Parties (Rights Against Insurers) Act 2010 (the 2010 UK Act).23 The 1930 UK Act allowed for a TRDA, but only where the insured tortfeasor was insolvent, and the victim had an unsatisfied judgment against the tortfeasor. Aside from these procedural hoops, the major difficulty with the 1930 UK Act was that its already narrowly drafted TRDA was essentially rendered nugatory by the ‘pay to be paid’ or ‘pay first’ clauses which are invariably included in UK P&I policies.24 As the label suggests, such pay to be paid clauses require the insured to have actually paid the sums due to third parties in respect of the insured’s liability before the insured is entitled to an indemnity from the insurer.25 The House of Lords held in The Fanti and The Padre Island26 that where such a pay to be paid clause applied, members’ rights of indemnity against P&I Clubs in respect of any third-party liability were contingent on the members first having discharged that liability themselves. Where a member became insolvent prior to paying the third-party victim (which, rather surreally, was the very prerequisite to the statutory TRDA under the 1930 UK Act), this contingent right to indemnity was extinguished, and there was therefore nothing left to transfer to the third-party victim. This cruel Catch-22 in which third-party victims found themselves was justified on the basis that the 1930 UK Act was never intended ‘to put a third party in any better position as against an insurer than that of the insured himself’.27 The 2010 UK Act ameliorates the harshness of this outcome to some extent. Section 9(5) of the 2010 UK Act provides that the third-party victim’s transferred rights ‘are not subject to a condition requiring the prior discharge by the insured of the insured’s liability to the third party’. However, in the case of a contract of marine insurance,28 s 9(6) provides that s 9(5) applies ‘only to the extent that the liability of the insured is a liability in respect of death or personal injury’.29 The effect of these two provisions is to reverse the rule in The Fanti and The Padre Island in cases involving death or personal injury, but pay to be paid clauses continue to bar third-party property damage and economic loss claims at English law where the insured does not first pay the third-party victim (in which case, of course, a statutory TRDA is otiose).30 Although the 1930 and 2010 UK Acts are perhaps best regarded as creating a rule of insolvency law, they conceptually frame the TRDA in English law as a statutory assignment of the On which, see Hjalmarsson (n 7) 264 f. Ibid 265 ff; Thomas (n 7) 690 ff. 24 For example, see the UK P&I Club Rulebook 2022, r 5(A): ‘Unless the Directors in their discretion otherwise decide, it is a condition precedent of an Owner’s right to recover from the funds of the Association in respect of any liabilities, costs or expenses that the Owner shall first have discharged or paid the same out of funds belonging to it unconditionally and not by way of loan or otherwise.’ 25 See the Law Commission and Scottish Law Commission, Third Parties – Rights against Insurers (Law Com No 272, Scot Law Com No 184, 2001) paras 5.28 ff. 26 [1991] 2 AC 1, [1990] 2 Lloyd’s Rep 191. 27 Ibid at 198, per Lord Brandon of Oakbrook. 28 Section 9(7) of the 2010 UK Act defines ‘contract of marine insurance’ by reference to the meaning given by s 1 of the Marine Insurance Act 1906. 29 Defined in s 9(7) of the 2010 UK Act as including ‘any disease and any impairment of a person’s physical or mental condition’. 30 Hjalmarsson (n 7) 265 f. 22 23

208  Research handbook on marine insurance law contractual rights under the insurance policy. This has undoubtedly also coloured the English courts’ ‘characterization’ of foreign statutory TRDAs brought against UK marine liability insurers. At English law, choice-of-venue and governing law clauses are obviously binding in disputes between P&I Clubs and their members; whether they will also bind third-party victims seeking to enforce TRDAs against the Clubs is a rather more vexed question. On a traditional analysis, such victims are strangers to the insurance policy, and have not agreed to be bound by any choice-of-venue or governing law clauses incorporated into it. However, English law has taken the view in recent years that a third-party victim relying indirectly on the insurance policy between the Club and its insured to sue the Club directly cannot eat its cake and still have it31 – a stance given teeth by the granting of ‘quasi-contractual’ anti-suit injunctions32 by the English courts to dissuade third-party victims from attempting to enforce their TRDAs against Clubs in other than the selected venue in the insurance policy.33 In Youell v Kara Mara Shipping Co Ltd,34 Aikens J in the English Commercial Court was called upon to decide whether to issue a permanent anti-suit injunction to prevent enforcement proceedings brought by a third-party victim, World Tanker, in Louisiana under its direct action statute, contrary to the English exclusive jurisdiction clause in the relevant hull and machinery policies issued by the insurer. The insurer argued that the Louisiana proceedings were vexatious, oppressive and unconscionable, and that it had an equitable right not to be subjected to them. Aikens J acknowledged that World Tanker had ‘not become a party to the policies by a mechanism of statutory novation or of statutory assignment’,35 but nonetheless concluded:36 The nature of the rights that the Direct Action Statute confers on World Tanker is contractual; it confers a statutory right to make a claim on a contract to which World Tanker was not originally a party. And (subject to par. C of the Statute) the rights are confined to the ‘terms and limits of the policy’.

This raises the obvious question as to whether the reference to ‘terms and limits of the policy’ in the Louisiana statute relates only to the substantive terms and conditions of the policy cover, or also includes the English jurisdiction and governing law clauses. Aikens J preferred the latter, broader view, concluding that ‘it must at least be highly arguable’ that World Tanker’s TRDA was subject to ‘all the bundle of rights and obligations contained in that contract,

See Andrew Dickinson, ‘The Right to Rome? The Law Applicable to Direct Claims against Insurers, and Anti-Suit Injunctions’ (2016) 132 LQR 536, 539. 32 On which, see Thomas Raphael, The Anti-Suit Injunction (2nd edn, OUP 2019), Ch 10 ‘Quasi-Contractual Anti-Suit Injunctions’; Myron Phua and Serena Seo Yeon Lee, ‘Taxonomising “Quasi-Contractual Anti-Suit Injunctions”’ [2021] LMCLQ 58; Paul Myburgh, ‘Non-Parties, Forum Agreements and Expanding Anti-Suit Injunctions’ [2020] LMCLQ 346. See also Merkin, Chapter 9, Part 3(a). 33 Unless the Brussels I Regulation (Recast) framework applies: see section 3 below. In Aspen Underwriting Ltd & Ors v Credit Europe Bank NV [2020] UKSC 11, [2020] 1 Lloyd’s Rep 520, the UK Supreme Court endorsed the CJEU approach in Assens Havn (n 4). That is, of course, no longer the position after Brexit (see remarks in section 7 below). 34 [2000] 2 Lloyd’s Rep 102. 35 Ibid [58], comparing the Louisiana direct action statute with the 1930 UK Act. 36 Ibid. This, despite the fact that most US commentators regard the Louisiana direct action statute as creating an independent TRDA: see for example Pallay (n 11) 67 ff. 31

Taxonomizing third-party rights of direct action against marine liability insurers  209 including the EJC’.37 On this reasoning, there was a sufficient contractual nexus to issue an anti-suit injunction constraining the Louisiana proceedings. In Through Transport Mutual Insurance Association (Eurasia) Ltd v New India Assurance Co Ltd (The Hari Bhum (No 1))38 the cargo interests sought to sue the relevant P&I Club directly in Finland under s 67 of the Finnish Insurance Contracts Act 1994, which conferred a statutory TRDA where the insured was insolvent. The Club sought an anti-suit injunction from the English Commercial Court to restrain the cargo interests from pursuing their claim in Finland, on the basis that its policy contained an English arbitration and choice of law clause.39 The cargo interests argued that these clauses did not bind them, as they were not parties to the insurance policy, and their TRDA against the Club was not derived from the policy. Rather, they were pursuing an independent foreign statutory remedy sourced in Finnish law.40 The Club’s contrary position was that the TRDA was, in substance, a direct enforcement of the rights of indemnity available to the insured under the P&I policy. In effect, the cargo interests were seeking to obtain the benefits of a statutory assignment, and were therefore (equitably) ‘bound’ by the English arbitration clause.41 Moore-Bick J saw the issue as being one of characterization: ‘in proceedings before an English Court a dispute about the nature of [the cargo interests’] claim can only be resolved by applying the principles of English law relating to characterization.’42 Applying his version of characterization, the judge considered that the issue in this case was one of legal obligations, and more importantly that the nature of the TRDA under the Finnish Act was essentially a right to enforce the P&I policy in accordance with its terms.43 Moore-Bick J conceded that the Finnish statute rendered void any terms in the P&I policy which restricted, or were inconsistent with, the Finnish TRDA (which should have squarely raised the question of the mandatory nature of the foreign statutory right), but nonetheless concluded (on, with respect, rather unconvincing grounds) that because the Club’s obligations under the insurance policy were governed by English law, Finnish legislation could not, and would not, override those terms.44 The English arbitration clause was thus binding on the cargo interests, and the Club was entitled to an anti-suit injunction to prevent the continuation of the Finnish proceedings.45 On appeal, the Court of Appeal agreed with Moore-Bick J’s approach to the characterization of the Finnish TRDA, holding that the Finnish Act gave the injured third party the right to claim compensation ‘according to the insurance policy’.46 Nonetheless, the Court acknowledged the relevance of foreign public policy considerations, and as a result exercised its discretion against granting an anti-suit injunction, on the basis that it could not fairly be said that

Ibid [59]. [2003] EWHC 3158 (Comm), [2004] 1 Lloyd’s Rep 206; 2004 EWCA Civ 1598, [2005] 1 Lloyd’s Rep 67. 39 EWHC judgment [5]–[8]. 40 Ibid [10]. 41 Ibid [9]. 42 Ibid [11]. 43 Ibid [19]. 44 Ibid [20]. 45 Ibid. The fact that the Finnish Court was first seised of the matter was irrelevant, as the Brussels I framework did not apply to arbitration by virtue of its art 2(d): ibid at [24]. See also the EWCA judgment [22]–[51] agreeing with this assessment. 46 EWCA judgment [58]. 37 38

210  Research handbook on marine insurance law the proceedings in Finland were vexatious or oppressive. Rather, comity required the Finnish proceedings to be allowed to continue:47 This claim is brought in Finland under a Finnish statute conferring rights on third parties against liability insurers in circumstances in which the insured is insolvent. The statute was no doubt passed because, as a matter of public policy in Finland, it was thought that liability insurers should be directly liable to third parties who had suffered loss in respect of which the insured was liable. The public policy behind the Finnish Act was the same as or very similar to the public policy behind the Third Parties (Rights Against Insurers) Act 1930. It appears that the only difference of importance between them is that in England the anti-avoidance provision does not defeat the pay to be paid clause, whereas it may be that s. 3 of the Finnish Act will do so, although it is right to say that that is a matter yet to be determined by the Finnish courts.

Similar arguments played out in the notorious and long-running Prestige litigation, with the French and Spanish governments arguing that they were entitled to bring a TRDA under Spanish law, despite the English law and arbitration clauses in the P&I policy of the Prestige’s insolvent shipowner, Mare Shipping Inc. The relevant P&I Club, the London Steam-Ship Owners Mutual Insurance Association Ltd, acknowledged its limited liability under the CLC regime, but argued that all claims brought by France and Spain outside the CLC framework were subject to the English law and arbitration clauses in its P&I policy, as well as all contractual defences available to the Club, including the pay to be paid clause in its policy, the ultimate effect of which was that the Club had no liability to France or Spain for any non-CLC claims. In The Prestige (No 2)48 Hamblen J adopted the reasoning in The Hari Bhum (No 1) that the issue was whether the foreign TRDA should be characterized according to English conflict-of-laws principles as being either:49 (a) A contractual obligation derived from the contract of insurance; or (b) An independent right of recovery under the relevant statute.

The judge concluded that Spain and France’s TRDAs were to be characterized in English law as claims to enforce English law obligations rather than independent Spanish statutory rights, and that those obligations could be enforced only in accordance with the terms of the relevant P&I Club policy, that is, in English arbitration, subject to English law and subject to the pay to be paid clause. He further held that Spain and France had become parties to the arbitration agreement in the Club rules and were therefore not entitled to State immunity under s 9(1) of the State Immunity Act 1978 (UK); that their claims were arbitrable; and that it was appropriate to give permission to enforce the English awards in favour of the Club as judgments. The Court of Appeal agreed with his approach, holding that50 the obligation which the appellants wish to enforce against the Club is governed by English law. It cannot be enforced otherwise than by arbitration in accordance with the Club rules and [France and

Ibid [94]–[97]. The Prestige (No 2) (n 8). 49 Ibid [49]. 50 The London Steam Ship Owners Mutual Insurance Association Ltd v The Kingdom of Spain and Another (The Prestige (No 2)) [2015] EWCA Civ 333, [2015] 2 Lloyd’s Rep 33 [84] per Moore-Bick LJ. 47 48

Taxonomizing third-party rights of direct action against marine liability insurers  211 Spain] have submitted to the jurisdiction of the English courts in relation to the determination of the arbitrator’s jurisdiction and the Club’s application to enforce the award as a judgment.

In The Prestige (No 3),51 the Club sought to appoint an arbitrator to deal with Spain’s disregard of the English courts’ previous rulings. Henshaw J held that Spain was bound by the full scope of the arbitration clause in the Club rules even though it was a ‘third party’, that is, a stranger to the P&I policy,52 on the basis of the ‘conditional benefit’ principle articulated in The Jay Bola:53 By asserting and continuing to pursue (in Spain or elsewhere) a claim founded on a contract of insurance containing an arbitration clause, Spain has […] assumed the burden represented by the arbitration clause insofar as it appertains to the claim being pursued. The arbitration clause applies to Spain’s claim by obliging those bound by it to arbitrate not only the substantive claim itself but also any dispute about a party’s compliance with the obligation to arbitrate.

Similarly, in The Yusuf Çepnioğlu,54 where the English Commercial Court and Court of Appeal had to decide whether a TRDA under Turkish law was subject to an English arbitration and law clause and pay to be paid clause in the relevant Club rules, the Court of Appeal applied the Jay Bola ‘conditional benefit’ principle to justify issuing an anti-suit injunction against Turkish proceedings, deprecating the earlier softer, comity-deferential approach adopted in The Hari Bhum (No 1). This harder line was articulated by Moore-Bick LJ as follows:55 In my view questions of comity in the established sense do not arise in a case such as this. Parties to a contract of this kind are free, if they so wish, to agree that any disputes arising under the contract are to be referred to arbitration. That necessarily involves giving up the right to pursue a claim by proceedings in the courts of any country. If legislation confers on an injured party the right to recover directly against the wrongdoer’s liability insurer by giving him in substance the right to enforce the contract, he must accept what the legislation gives him, including the obligation to pursue any claim in arbitration. To hold him to that agreement is to give effect to the legislation while preserving the substance of the obligation which he seeks to enforce.

It therefore now seems firmly established in English law that claims of third-party victims relying on foreign TRDAs, if ‘characterized’ through an English conflicts lens as rights in some manner derived from, or conditional on, the P&I policy, will be subject to any exclusive

51 The London Steam Ship Owners Mutual Insurance Association Ltd v The Kingdom of Spain and Another (The Prestige (No 3)) [2020] EWHC 1582 (Comm), [2020] 1 Lloyd’s Rep IR 413. 52 Ibid [82]. 53 Schiffahrtsgesellschaft Detlev von Appen GmbH v Voest Alpine Intertrading GmbH (The Jay Bola) [1997] 2 Lloyd’s Rep 279. See also Charterers’ Mutual Assurance Association Ltd v British & Foreign Marine Insurance Co Ltd [1998] IL Pr 838; National Navigation Co v Endesa Generacion SA (The Wadi Sudr) [2010] 1 Lloyd’s Rep 193; but cf The Atlantik Confidence [2020] 1 Lloyd’s Rep 520; [2020] Lloyd’s Rep IR 274. 54 Shipowners’ Mutual Protection and Indemnity Association (Luxembourg) v Containerships Denizcilik Nakliyat VE Ticaret AS (The Yusuf Çepnioğlu) [2015] EWHC 258 (Comm), [2015] 1 Lloyd’s Rep 567; [2016] EWCA Civ 386, [2016] 1 Lloyd’s Rep 641. 55 EWCA [58]. Nonetheless, comity may play a greater role where the third party seeking to enforce a foreign TRDA is a State: see UK P&I Club NV v Republica Bolivariana De Venezuela (The RCGS Resolute) [2022] EWHC 1655 (Comm) [123], citing General Dynamics United Kingdom Ltd v State of Libya [2021] UKSC 22, [2021] Lloyd’s Rep Plus 109. See also Merkin, Chapter 9 section 4 on the implications of sovereign immunity in this context.

212  Research handbook on marine insurance law English choice of venue and law, and pay to be paid clauses contained in the relevant policy. The most likely outcome is that all third-party victims’ TRDAs (apart from those arising from death or personal injury) will be defeated because the insured shipowner is insolvent.56 The corollary of this position, which was acknowledged in The RCGS Resolute,57 is that if ‘legislation or the law in the foreign jurisdiction creates an independent right which does not mirror the insurer’s liability under the contract of insurance, the arbitration clause is unlikely to bind the third party since the contract is not a necessary condition of the claim’. However, it is telling that in practice, apart from TRDAs derived from international maritime Conventions, English courts have invariably been both able and willing to find that any passing reference to the insurance policy in a foreign statute means that the foreign TRDA should indeed be ‘characterized’ as derived from, or contingent on, some aspect of the UK P&I policy, which acts as a tractor beam to draw the foreign TRDA back into the ambit of English jurisdiction and law, and consequently the problematic limits imposed by the 2010 UK Act. The English approach is susceptible to a number of criticisms. First, despite the insistence by the English courts that they are engaging in conflict-of-laws characterization, this is clearly not the case, or at least not as characterization in private international law is generally understood. As Dickinson points out in his critique of The Yusuf Çepnioğlu,58 in the context of an application for an anti-suit injunction, the starting point should be to identify the rules of English law relied upon to support the grant of an injunction (assuming, as seems likely, that this issue is regarded as a matter of English procedure and jurisdiction, and therefore for the lex fori), before turning to consider (and if necessary to classify) the subject matter of the foreign TRDA insofar as it is necessary for the application of the rules of English law. However, on a traditional English conflicts approach, this would simply be an exercise in classification of the foreign obligation as a factual element in applying the lex fori, that is, English law. It would not be an exercise in full-blown characterization for the purposes of determining the law applicable to the TRDA in English proceedings.59 The second chief difficulty with the conflicts approach adopted by the English courts is that it lacks certainty, predictability and clarity. Instead of reaching a decision on the legal status of foreign TRDAs as a matter of principle, the English casuistic approach of trying to taxonomize the foreign TRDA right as independent from, or indirectly dependent on, or derived from, the UK P&I policy, necessarily involves a complex and lengthy analysis and interpretation of specific foreign statutory provisions, which of course have to be proven by duelling expert witness reports.60 As Briggs aptly points out, this ‘is a recipe for expensive unpredictability’.61 There is also a real risk that English courts will interpret the foreign statutory TRDA out of context, or without due regard for foreign public policy, and that the judicial analysis of the relevant foreign law will, as a consequence, be distorted. Any passing reference to an insurance policy in the relevant foreign statute seems to be rather eagerly magnified and taken up by the English courts as an invitation to invoke a contractual model of transferred rights analogous

See QBE Europe SA/NV & Anor v Generali Espana de Seguros y Reaseguros [2022] EWHC 2062 (Comm), [2022] 2 Lloyd’s Rep 481. 57 Ibid [33]. 58 n 54. 59 Dickinson (n 31) 537 f. 60 See for example The Yusuf Çepnioğlu (n 54) EWHC judgment [7]–[32]. 61 Briggs (n 1) 329. 56

Taxonomizing third-party rights of direct action against marine liability insurers  213 to the 2010 UK Act. Finally, there has been little meaningful discussion by the English courts as to whether the foreign statutory TRDA might create a mandatory or overriding obligation at foreign law – the assumption simply seems to be that an English choice of law clause will always trump a foreign statutory TRDA.

3.

EUROPEAN UNION APPROACH

In sharp contrast to the stance adopted by the English courts, the Court of Justice of the European Union (CJEU) has refused to approach jurisdictional issues arising from TRDAs by reference to either a contractual or tortious/delictual characterization, which would orthodoxly trigger an application of the jurisdictional rules in the Rome I or Rome II Regulations62 respectively. Instead, the Court has characterized the issue as one of jurisdiction in ‘matters relating to insurance’, which is governed by Section 3 (arts 10–16) of the Brussels I Regulation (Recast).63 Importantly, Section 3 of the Brussels I Regulation (Recast) provides that an insurer domiciled in a Member State may be sued in the courts of the Member State in which it is domiciled, but also ‘in another Member State, in the case of actions brought by the policyholder, the insured or a beneficiary, in the courts for the place where the claimant is domiciled’,64 or additionally, in respect of liability insurance or insurance of immovable property, in the courts for the place where the harmful event occurred.65 These special jurisdictional rules, which are clearly informed by a policy of protecting insureds and beneficiaries as the presumptively weaker parties in the insurance relationship,66 apply equally to third parties’ TRDAs.67 Odenbreit,68 a motor vehicle insurance case involving a German victim of an accident that occurred in the Netherlands, provides the locus classicus of this approach. The Dutch tortfeasor’s insurer resisted German jurisdiction and argued for an application of Rome II on a tortious/delictual characterization, which would force the litigation into the Dutch courts on the basis that that was the locus delicti.69 On a reference from the German Supreme Court,

Regulation (EC) No 593/2008 of the European Parliament and of the Council of 17 June 2008 on the law applicable to contractual obligations (Rome I); Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II). 63 Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (recast). 64 Art 11(1). Art 11(2) extends the domicile rule to insurers who have a branch, agency, or other establishment in one of the Member States in respect of disputes arising out of the operations of that branch etc. 65 Art 12. 66 Ulfbeck (n 7) 296. 67 Art 13. 68 FBTO Schadeverzekeringen NV v Jack Odenbreit (C-463/06) [2007] ECR I-11321. 69 Ibid at [13]: ‘Thus, according to the prevailing view […] the right of action of the injured party in German private international law is regarded as a right in tort and not as a right under an insurance contract.’ 62

214  Research handbook on marine insurance law the CJEU held in favour of Odenbreit’s right to bring a TRDA where he was domiciled in Germany, based on the special insurance jurisdiction rules contained in Brussels I.70 The approach adopted in Odenbreit was confirmed in the landmark CJEU decision in Assens Havn.71 In this case, the tug Sea Endeavour I damaged quay installations in Assens Havn, Denmark. As the tug’s charterer, Skåne Entreprenad Service AB, had in the meantime gone into liquidation, Assens Havn brought a TRDA against the charterer’s P&I Club, Navigators Management (UK) Ltd, in the Danish Maritime and Commercial Court. That Court held that it lacked jurisdiction to hear the dispute, as the exclusive English jurisdiction clause in the relevant P&I policy bound the third-party victim, Assens Havn.72 The CJEU disagreed, finding as follows: 30. As in cases involving workers and consumers, actions involving insurance are characterised by an imbalance between the parties […] , an imbalance which the rules laid down […] seek to correct by giving the weaker party the benefit of rules of jurisdiction more favourable to his interests than the general rules provide for […] 31. In particular, those provisions ease the situation of a victim of insured damage by enabling him […] to sue the insurer in question before the courts for the place where the harmful event occurred, provided that the national law permits such a direct action. […] 40. The view must therefore be taken that an agreement on jurisdiction made between an insurer and an insured party cannot be invoked against a victim of insured damage who wishes to bring an action directly against the insurer before the courts for the place where the harmful event occurred, as recalled in para 31 of this judgment, or before the courts for the place where the victim is domiciled, a possibility accepted by the court in its judgment of 13 December 2007, FBTO Schadeverzekeringen NV v Odenbreit Case C-463/06 [2008] Lloyd’s Rep IR 354, para 31).

The CJEU reiterated the above stance in its preliminary ruling on a reference from the English High Court in London Steam-Ship Owners’ Mutual Insurance Association,73 with the Court concluding as follows: 60. As regards, first, the relative effect of an arbitration clause included in an insurance contract, it is apparent from the case law of the court that a jurisdiction clause agreed between an insurer and an insured party cannot be invoked against a victim of insured damage who, where permitted by national law, wishes to bring an action directly against the insurer, in tort, delict or quasi-delict, before the courts for the place where the harmful event occurred or before the courts for the place where the victim is domiciled (see, to that effect, judgment of 13 July 2017, Assens Havn v Navigators Management (UK) Ltd Case C-368/16; EU:C:2017:546; [2018] Lloyd’s Rep IR 10; [2018] QB 463, paras 31 and 40 and the case law cited). 61. It follows that, to avoid that right of the victim being undermined, a court other than that already seised of that direct action should not declare itself to have jurisdiction on the basis of such an arbitration clause, the aim being to guarantee the objective pursued by Regulation (EC) No 44/2001, namely the protection of injured parties vis-à-vis the insurer concerned (see, to that effect, Assens Havn, paras 36 and 41). 62. That objective of protecting injured parties would be compromised if a judgment entered in the terms of an arbitral award by which an arbitral tribunal declared itself to have jurisdiction on the basis

Discussed by Hjalmarsson (n 7) 273 ff. Assens Havn v Navigators Management (UK) Ltd (n 4). 72 Such an exclusive jurisdiction clause would, by virtue of art 15 of the Brussels I Regulation (Recast), bind the insurer and the insured, as they are both parties to the insurance agreement. The third-party victim, however, is not. 73 n 3. 70 71

Taxonomizing third-party rights of direct action against marine liability insurers  215 of such an arbitration clause, included in the insurance contract concerned, could be regarded as a ‘judgment given in a dispute between the same parties in the Member State in which recognition is sought’, within the meaning of article 34(3) of Regulation (EC) No 44/2001.

This stark split between the UK and European approaches to the jurisdiction issue was again underlined in The Prestige (No 4),74 where the P&I Club sought to enforce its arbitral awards and award judgments75 against France and Spain. Butcher J held that, although the English court had jurisdiction over the award claims, because they fell within the arbitration exception to the Brussels I Regulation (Recast), the judgment claims were too far removed from the arbitrations to fall within the arbitration exception.76 Butcher J concluded that the judgment claims were ‘matters relating to insurance’77 for the purposes of the Brussels I Regulation (Recast). As ‘injured parties’, France and Spain were thus entitled to the jurisdictional protections of Section 3 of the Regulation, without having to show that they were in fact economically weaker parties.78 As a result, by virtue of art 14(1) of the Brussels I Regulation (Recast), the Club was required to bring proceedings only in the courts of the Member State where the defendant was domiciled. This finding was affirmed by the English Court of Appeal.79 In conflict-of-laws terms, the EU approach is thus the very antithesis of the UK approach, in the sense that the CJEU has deliberately eschewed any attempt to characterize or interpret the relevant TRDA in terms of the nature of the underlying right/obligation. Instead, the CJEU has deliberately framed the legal issue as being ‘title-based’,80 namely whether the TRDA is a ‘matter relating to insurance’, which now seems settled, and whether the third party may indeed be presumed to be the weaker party in the insurance relationship.81 The most significant criticism of the EU approach is that it bypasses or fudges the more obviously applicable and specific underlying characterization rules which would classify the TRDA as being sourced in contract (Rome I) or arising from the insured’s tort/delict (Rome II), without any clear explanation as to why that should be the case. It may also be argued that the EU approach is too blunt and victim-centric in that it negates the legitimate expectations of

The London Steam Ship Owners Mutual Insurance Association Ltd v The Kingdom of Spain and Another (The Prestige (No 4)) [2020] EWHC 1920 (Comm), [2020] 2 Lloyd’s Rep 356. 75 See The London Steam Ship Owners Mutual Insurance Association Ltd v The Kingdom of Spain and Another (The Prestige (No 2)) [2013] EWHC 3188 (Comm), [2014] 1 Lloyd’s Rep 309. 76 Ibid [104] ff. 77 Ibid [123]. 78 Ibid [132], relying on The Atlantik Confidence (n 53), especially [50], [51] and [56]. 79 The Prestige (Nos 3 and 4) (n 3) [133] ff. 80 See the Opinion of Advocate-General Bobek in Landeskrankenanstalten-Betriebgesellschaft – KABEG v Mutuelles du Mans Assurance – MMA IARD SA (Case C-340/16), [2017] IL Pr 31 [26]: ‘I do not think that it would be either necessary or wise to attempt to provide a general and exhaustive definition of what is a “matter relating to insurance” and, hence, what is “insurance”. That can be left in the hands of legal scholarship. There is, however, one element that emerges from the reviewed case law, naturally tied to the logic of the Brussels Convention/Regulation system: for the purpose of international jurisdiction, the basis for ascertaining what is a “matter relating to insurance” is essentially “title-based”. Is the title for which action is launched against a specific defendant (in other words, the cause of that action) the ascertaining of rights and duties arising out of the insurance relationship? If yes, then the case can be deemed as a matter relating to insurance.’ 81 See for example Group Josi Reinsurance Company SA v Universal General Insurance Company (UGIC), Case C-412/98, [2000] ECR I-05925, ECLI:EU:C:2000:399: the special insurance jurisdiction rules in Section 3 of Brussels I do not apply to claims between insurers and reinsurers. 74

216  Research handbook on marine insurance law P&I Clubs to be able to channel all litigation brought against them and deal with it efficiently in a single selected jurisdiction and under a single selected law.82 The CJEU reasoning also involves something of a bootstrapping exercise or logical sleight of hand, in the sense that art 13(2) of the Brussels I Regulation (Recast) provides that the special insurance jurisdiction rules, which are primarily for the benefit of insureds and beneficiaries, also extend ‘to actions brought by the injured party directly against the insurer, where such direct actions are permitted’. This obviously begs the question: which legal system determines whether TRDAs are permitted? This presumably has to be an issue for determination by the lex causae rather than the lex fori, as it involves an assessment of substantive third-party rights.83 That should in turn arguably mean that jurisdiction and applicable law are determined by the underlying characterization of the nature and legal status of the third-party right adopted by the relevant lex causae, rather than by lumping all TRDAs with foreign elements under the general rubric of ‘matters relating to insurance’.

4.

NEW YORK LAW

The US Supreme Court in Wilburn Boat Co v Fireman’s Fund Insurance Co84 held that marine insurance contracts are governed by state law, provided that no well-entrenched federal admiralty precedent conflicts with state law. The effect of the Wilburn Boat ruling is that TRDAs brought against the American Club will be determined by New York law in the District Court for the Southern District of New York, rather than by federal admiralty law. In general terms, the New York direct action statute is less generous than its Louisiana counterpart.85 Rather like the UK statute, the New York statute is designed primarily to protect a third-party victim from an insured’s bankruptcy, by requiring all liability insurance policies to contain a clause which meets the requirements of the statute.86 Third-party victims may only sue the insurer directly where a judgment obtained against the insured remains unfulfilled. The insurer will then be liable for any judgment against the bankrupt insured within the terms and conditions of the relevant insurance policy. Like the UK model, the New York statute does not create an independent TRDA in favour of a third party against the insurer – rather, the third-party victim stands in the shoes of the insured party as a judgment creditor. However, a major difference between New York and UK law is that § 3420(i) of the New York Insurance Law provides that the TRDA section does not apply in respect of ‘insurance in connection with ocean going vessels against any of the risks specified in paragraph twenty-one of

See Hjalmarsson (n 7) 270: ‘To put this in perspective, the very nature of shipping liabilities is that they may occur anywhere in the world. Consequently, would it really be reasonable to expect the P&I insurer to appear before any court in the world?’ See also Yvonne Baatz, ‘Matters relating to Insurance and Protecting the Weaker Party’ [2018] LMCLQ 2, 9, describing the Assens Havn decision as a ‘very worrying development for liability insurers, who may find themselves sued in courts which they had not anticipated or agreed to’. 83 See Ulfbeck (n 7) 298–9, 303 ff. 84 348 US 310, 320–1, 1955 AMC 467, 476 (1955). For a detailed discussion of the Wilburn Boat decision and its implications for choice-of-law issues in the US, see Sturley, Chapter 11, section 3. 85 Pallay (n 11) 67 ff. 86 Ibid 70; NY Insurance Law § 3420. 82

Taxonomizing third-party rights of direct action against marine liability insurers  217 subsection (a) of section one thousand one hundred thirteen of this chapter’. Section 1113(a)(21) in turn provides: ‘Marine protection and indemnity insurance,’ means insurance against, or against legal liability of the insured for, loss, damage or expense arising out of, or incident to, the ownership, operation, chartering, maintenance, use, repair or construction of any vessel, craft or instrumentality in use in ocean or inland waterways,87 including liability of the insured for personal injury, illness or death or for loss of or damage to the property of another person.

In Dunn v Am Home Assurance Co,88 the Appellate Division of the New York Supreme Court held that a ‘yacht hull and protection and indemnity’ policy fell within the aforementioned statutory exemption. The third-party personal injury victims therefore lacked standing to bring a TRDA against the insurer, because they were strangers to the insurance policy.89 Miller v Am SS Owners Mut Prot & Indem Co90 concerned an injured seafarer who obtained a default judgment against the shipowner for his injuries, but the owner was insolvent. At the time of the seafarer’s injury, the owner was the insured under a liability policy covering injuries to employees. The seafarer filed an action directly against the insurer. The Court granted summary judgment in favour of the insurer, confirming that the New York statutory carve-out covered both liability and indemnity policies – the only issue was thus whether the policy was a marine insurance policy. The Court noted that the ‘exception was consciously made by the New York legislature to eliminate a perceived competitive disadvantage to which New York’s marine insurers were placed by the direct action statute’.91 In this regard, the Court referred to the legislative history of the New York statute, including a memorandum to the then Governor from his counsel, Nathan R Sobel, which stated: This bill seeks to exempt marine insurance companies from the necessity of including all of the […] standard provisions in their policies. The principal reason for such exclusion is that such companies cannot compete with authorized companies such as Lloyd’s whose policies do not have to contain these standard provisions.

Direct action statutes were originally introduced in the US ‘as a response by states to address the inequities of common law which, in the past, left injured parties without recourse when judgment could be had against a tortfeasor, despite the tortfeasor being covered for such claims’.92 However, that third-party benefit was deliberately reversed in New York law, seemingly with an envious eye on the UK P&I Clubs’ ability to enforce pay to be paid clauses in their policies. As a consequence, the effect of the New York statute is even harsher than the

87 There is an obvious inconsistency of scope between the two provisions: Hartford Fire Ins Co v Mitlof 123 F Supp 2d 762 (SD NY 2000), 2001 AMC 2135 held that the statutory carve-out only applies to ocean-going vessels. 88 158 AD 2d 505, 551 NYS 2d 268 (App Div 2nd Dept 1990). 89 Cowan v Cont’l Ins Co 86 AD 2d 646, 446 NYS 2d 412 (App Div 2nd Dept 1982). 90 509 F Supp 1047 (SD NY 1981). 91 Ibid 1049 n 2. See also Pallay (n 11) 73 f; WABCO Trade Co v SS Inger Skou 663 F 2d 369 (2d Cir 1981); Dicola v American SS Owners Mut Protection & Indem Assoc (In re Prudential Lines) 158 F 3d 65 (2d Cir 1988), 1999 AMC 609. 92 Pallay (n 11) 85.

218  Research handbook on marine insurance law 2010 UK Act in terms of negating TRDAs in respect of all marine insurance policies, and with regard to third-party personal injury and death claims, as well as property claims.

5.

JAPANESE LAW

The broad starting point in Japanese law, which is historically heavily influenced by German law, is that if ‘one of the parties promises in a contract that he/she will tender a certain performance to any third party, the third party shall have the right to claim that performance directly from the obligor’.93 The corollary of this rule is that a third party which is not a nominated beneficiary, such as a third-party victim seeking to exercise a TRDA against an insurer, had no right of action under general Japanese civil law. However, a special provision, art 22 of the Insurance Act 2008, specifically regulates the position of TRDAs at Japanese law. Article 22 provides that:94 (1) A person who has a claim for compensation for damages against an insured under a liability insurance policy arising from an insured event under said policy shall have a statutory lien [sakidori tokken95] over the right to claim the insurance payment. (2) An insured may exercise the right to claim an insurance payment from an insurer only up to the amount that the insured paid in connection with the claim for compensation for damages set forth in the preceding paragraph, or the amount to which the person that holds the claim has consented. (3) The right to claim an insurance payment under a liability insurance policy may not be assigned, pledged or attached; provided, however, that this shall not apply in the following cases: (i) where said right is assigned to a person having a claim for compensation for damages as set forth in paragraph (1) or attached based on the claim for compensation for damages; and (ii) where the insured may exercise the right to claim an insurance proceeds payment pursuant to the provisions of the preceding paragraph.

The legal effect of art 22 of the Insurance Act 2008 is to afford the third-party victim priority as a preferred creditor when bringing a TRDA against the insurer in the event of the insured’s insolvency.96 This provision is considered mandatory. As a consequence, the legal status of pay to be paid clauses under the Insurance Act 2008 is unclear. At Japanese law, the event that triggers the liability insurance payment is regarded as the occurrence of the insured event, rather than the insured’s payment of the third-party victim. Pay to be paid clauses may thus be considered to contravene the mandatory rule in art 22.97 It is also unclear whether P&I Clubs

Japanese Civil Code, art 537(1), discussed by Masami Okino, ‘Contracts for the Benefit of Third Parties in Japan’ in Mindy Chen-Wishart et al (eds) Formation and Third Party Beneficiaries (OUP 2018) Ch 13, 258. 94 English translation provided at www​.ja​paneselawt​ranslation​.go​.jp/​en/​laws/​view/​2775/​en. 95 On the conceptual nature of which, see Shusei Ono, ‘A Comparative Study of the Transfer of Property Rights in Japanese Civil Law’ (2003) 31 Hitotsubashi Journal of Law and Politics 1, 2, 18. 96 See Satoshi Nakaide, ‘Marine Insurance Law in Japan – A Structure Based on a Combination of Civil Law and English Marine Policy Wordings’ [2014] 440 早稲田商学 1, 16 ff. 97 Ibid 17 n 30. 93

Taxonomizing third-party rights of direct action against marine liability insurers  219 may contract out of the parameters of art 22 by, for example, specifying English law as the governing law.98 This legal position is reflected in cl 42(4) of the Japan P&I Club’s 2022 Rules, which provides that: Members shall not assign or pledge a right to claim insurance money against the Association which was obtained under the Rules where liability incurred by the Members is based on damages, save as provided in the following instances: (1) Where the right is assigned to the person who is entitled to be compensated for damages by a Member; or (2) Where a Member is entitled to exercise the right to present a claim for insurance money against the Association in compliance with Rule 18.99

The 2018 Guidance on the Japan P&I Club Rules100 explains the reason for this clause: Under Article 22 of the Insurance Act of Japan, as the assureds of the liability insurances such as P&I insurance (the Members) have the right to claim insurance moneys from the Association, a statutory lien is provided for the parties who compensate damages to the assureds. In addition in order to arrange these liens, except in certain cases it is not permitted to assign or pledge, etc. the right to claims for insurance moneys. Rule 42.4 reflects the provisions of the Insurance Law. Members as the assureds cannot assign to third parties their future rights to claims for insurance moneys, or cannot commit to assign. However, according to Rule 42.1 and 42.2, an exception to this is that Members can assign the right to claim from the Association to third parties who have the right to claim compensation for damages from Members.

It therefore appears that Japanese law, like EU law, is more favourable to the rights of third-party victims than is the position in common law jurisdictions, although victims’ interests are protected by means of a different mechanism, namely that of a preferred statutory right of priority.

6. EVALUATION From the foregoing comparative analysis, it is obvious that there are fundamental differences of approach to TRDAs in the four major legal frameworks in which the IGP&I Clubs operate, with the EU and Japanese approaches being far more third-party victim friendly, while the UK and New York approaches favour the insurer at the expense of the third-party victim. This will inevitably lead to inconsistent and arbitrary outcomes. Ibid 18. As discussed above, the Japanese Club’s P&I policy is governed by Japanese law. However, it is apparently common for hull and cargo policies to contain an English choice of law clause: ibid 2–3. 99 Rule 18 stipulates: ‘Unless otherwise agreed between a Member and the Association, the Association shall indemnify a Member according to the stipulations under this Chapter against damages and loss (hereinafter called “damages”) as well as costs and expenses, which arise in respect of his interest in an Entered Ship, out of events occurring during the period of entry of the ship in the Association and in connection with the operation of the ship by or on behalf of the Member, as set out in Rules 19 to 33 below and for which the Member has become liable to pay and has paid (except for the case where the Board of Directors otherwise decides).’ Emphasis added. 100 The 2022 Guidance is only available to members of the Club. 98

220  Research handbook on marine insurance law For example, if the owner of the Prestige had happened to take out a P&I policy with Gard, Skuld, or the Swedish Club, instead of a UK Club, there is no doubt that the CJEU would have ruled that the French and Spanish victims were entitled to sue in Spain and/or France, and would therefore have had access to their TRDAs under their laws of domicile because of the special insurance jurisdiction rules in the Brussels I Regulation (Recast), regardless of any Norwegian or Swedish jurisdiction and choice of law clauses in the relevant P&I policy. On the European approach, most of the time, effort, and no doubt eye-watering expense of 20 years of Prestige litigation and arbitration in multiple jurisdictions could have been channelled to where it properly belonged – compensating the third-party victims. By contrast, the current UK approach, while filling the wallets of conflicts lawyers and expert witnesses, and keeping the Commercial Court occupied with interesting cases on quasi-contractual anti-suit injunctions, arguably does little to achieve system-transcendent transnational justice,101 and still less to enhance the international corporate responsibility credentials of UK P&I Clubs. That said, there is an argument for allowing P&I Clubs the benefit of their choice-of-venue clauses, in that they serve a legitimate economic function in terms of channelling claims efficiently and keeping dispute resolution costs reasonable. Given the nature of the international shipping industry, an incident triggering TRDAs may occur in any jurisdiction – it is arguably unreasonable to foist the extra burden of overseas litigation onto P&I Clubs when this was not within their commercial expectations in dealing with disputes with their members.102 In that sense, the European approach allowing an automatic right to third-party victims to bring claims wherever they are domiciled may be seen as tilting the balance too far in the opposite direction. The counter-argument to that position, of course, is that upholding even an exclusive foreign choice-of-venue clause may prove an insurmountable barrier to access to justice for poorer third-party victims.103 Although there is a logical symmetry in insisting that the insurer’s obligations towards the third-party victim coincide precisely with those owed to the insured, public policy might be seen as favouring the primacy of third-party victim’s rights:104 If the primary object of obligatory insurance and third party rights of action is the protection of identified third parties, it is arguable that third parties should be secure in their expectation that the insurance will pay. This may also mean that third parties should be in an even more secure position than the insured and protected against the risk of defensive counter-measures by insurers, particularly when based on conduct of the insured to which the third party was not privy.

101 See Thomas Raphael, ‘Do as You Would be Done By? System-Transcendent Justification and Anti-Suit Injunctions’ [2016] LMCLQ 256: ‘That different countries will have different approaches to the conflict of laws is inevitable, but national rules of private international law should be capable of being rationalised on a universalisable basis, and be susceptible to “system-transcendent” justification: so that another legal system could, in principle, be rationally persuaded to accept them as legitimate from an international perspective, even if its own rules differ.’ 102 See n 82 above. 103 Recent examples of the devastating impacts of major maritime incidents on less developed countries include the X-Press Pearl incident off the coast of Sri Lanka (www​.unep​.org/​resources/​ report/​x​-press​-pearl​-maritime​-disaster​-sri​-lanka​-report​-un​-environmental​-advisory​-mission) and the MV Wakashio grounding in Mauritius: see Richard Clayton, ‘Wakashio was casualty of poor operations and management’, Lloyd’s List, 5 July 2022 (https://​lloydslist​.​maritimein​telligence​.informa​.com/​LL1141478/​ Wakashio​-was​-casualty​-of​-poor​-operations​-and​-management). 104 Thomas (n 7) 688.

Taxonomizing third-party rights of direct action against marine liability insurers  221 Ultimately, although the issues are dressed up in theoretical and technical conflict-of-laws terms, the real problem lies with the enforcement of pay to be paid clauses. Where such clauses are allowed to be enforced against third-party victims because the insured shipowner is insolvent (which must, after all, be a likely occurrence after a major incident), the relevant P&I insurer is arguably granted an unwarranted windfall in the insolvency lottery. UK P&I Clubs will, of course, be quick to point to the I in P&I and argue the Fanti and Padre Island point that this is simply in the nature of an indemnity policy – as the insured only had a contingent right of recompense against the insurer, that right has now fallen away on the insured’s insolvency. However, although this reasoning may be attractive to private law theorists, it will provide cold comfort to third-party victims who may have suffered considerable loss, and certainly have no control over the solvency of the insured shipowner. The UK legislator recognised public policy concerns surrounding pay to be paid clauses by invalidating them with regard to personal injury and death claims brought against P&I Clubs under the 2010 UK Act. Arguably, the legislator did not go far enough.105 It would be very interesting to see what the reaction of the British public would be to a Prestige-type incident polluting the UK coast. It seems unlikely that affected British third-party victims would be happy for their property and financial livelihoods to be sacrificed in order to preserve party autonomy, uphold the sanctity of pay to be paid clauses at English law, and assist the profitability of UK P&I Clubs. Rather, one suspects that there would be an overwhelming and strident clamour for urgent reform of the 2010 UK Act. A cynic might even go so far as to suggest that English courts have, to some extent, been very comfortable with protecting the interests of UK P&I Clubs and issuing anti-suit injunctions in their favour, simply because the oil has washed up on foreign, rather than British, beaches.

7. CONCLUSIONS Given the current lack of international uniformity of legal approaches to TRDAs brought against P&I Clubs, is there a way forward to unify or at least better harmonize outcomes? The Comité Maritime International (CMI) began some work on the issue, with its Standing Committee on Marine Insurance distributing a standard form questionnaire on TDRAs against insurers to the national maritime law associations (MLAs) in 2019.106 The response was not exactly overwhelming,107 and the project does not seem to have progressed beyond this initial information-gathering stage. The prospects of the CMI producing a successful harmonizing or unifying instrument on the issue seem somewhat slim. In addition, Brexit is likely to have a further disharmonizing effect, in the sense that the UK P&I Clubs will in future be able to enforce their exclusive homeward choice-of-venue and governing law clauses, as well as pay to be paid clauses, with impunity and without further perceived ‘interference’ from European courts and the CJEU. This has already been seized

See Thomas (n 17) 70: ‘In the context of the 2010 Act, which applies when the assured is insolvent, the survival of the “pay to be paid” condition amounts to an absurdity.’ 106 See https://​comitemaritime​.org/​work/​right​-of​-direct​-action​-against​-insurers/​. 107 Only 14 MLAs responded: Italy, the PRC, Norway, Spain, Croatia, Sweden, Belgium, Argentina, Colombia, Turkey (Türkiye), Mexico, South Korea, Singapore and Venezuela. 105

222  Research handbook on marine insurance law upon as a cause for celebration by some pro-Brexit commentators.108 Brexit will, in theory at least, provide UK P&I Clubs with something of an economic and competitive edge over their European and Japanese counterparts. However, given the potential catastrophic nature of some of the international incidents insured by P&I policies, this is arguably a short-term, narrow and ethically unsatisfactory perspective to adopt. In a world in which ever greater emphasis needs to be placed on environmental protection, the UK and New York insurer-protective approaches begin to look increasingly archaic, irresponsible, selfish and contrary to international public policy. Ultimately, it seems unlikely that the P&I Clubs themselves will be able to agree on reform of policy wordings and a unified international approach to the TRDA issue. The only organization that could possibly make a difference in this area is the International Maritime Organization (IMO). The IMO maritime pollution and related Conventions which create TRDAs at international law have thus far been relatively effective, because they are framed in independent non-national terms and operate mandatorily. Given that P&I cover is a crucial supplement and adjunct to the limited liability regimes of the major pollution and related international Conventions, it seems reasonable to suggest that the IMO has a legitimate interest in, and a remit to, ensure international minimum mandatory standards when it comes to TRDAs brought against P&I insurers – after all, if States and national laws and private insurance contracts fail, the international community will inevitably be left cleaning up the mess. Whether IMO Member States have the stomach to attempt to regulate private commercial relationships between P&I insurers and their members or extend the Convention regimes of obligatory insurance to encompass P&I policies, however, remains to be seen. There is no doubt that any attempt, for example, to invalidate pay to be paid clauses, or to guarantee minimum levels of compulsory compensation to third-party victims, will be regarded by the UK P&I Clubs and their cheerleaders as significant international regulatory over-reach.

See for example Andrew Tettenborn, ‘Shipping Law, Brexit and the City of London’ [2022] LMCLQ 124, 133: ‘In short, it is hard to deny that freedom from the shackles of the Brussels system of jurisdiction is a serious and substantial gain that London shipping lawyers can take away from Brexit.’ 108

11. Choice-of-law issues in marine insurance cases in the United States Michael Sturley

1. INTRODUCTION Choice-of-law questions can be challenging regardless of the context in which they arise. They can be particularly difficult for marine insurance cases in the United States. Some of the difficulty arises from the potential need to address choice-of-law issues on three different levels: (1) the “horizontal” choice between US law and the law of some other country; (2) if US law governs, the “vertical” choice between maritime law, which is federal,1 and the law of a US state2 with an interest in the resolution of the dispute; and (3) if US state law governs, the “horizontal” choice among the possibly relevant states. An even greater share of the difficulty arises from the lack of clear judicial guidance on the second and third categories. This chapter, focusing on the second and third categories, addresses which body of US law governs when US law applies. The US Supreme Court established the principles for the first category (the “horizontal” choice between US and foreign law) in the middle of the twentieth century,3 and those principles apply broadly to all maritime cases, not only marine insurance cases. The choice-of-law principles in the second and third categories, by contrast, have been unique to marine insurance law since the Supreme Court’s infamous 1955 decision in Wilburn Boat Co. v Fireman’s Fund Insurance Co.4

Maritime law, consisting of the “general maritime law” (that is, judge-made law, sometimes described as a rare example of federal common law) as supplemented by Acts of Congress, is the law of the nation as a whole (and is theoretically uniform throughout the country). 2 In addition to the 50 states, the United States includes a federal district (the District of Columbia) in which the national capital is located and a few other territories (for example, Puerto Rico), some of which are treated in the same way as states for some purposes. Each state possesses substantial independence in legal matters. The most basic legal subjects—such as property, contracts, and torts—are largely matters of state law. Although state laws are generally similar on those basic subjects, each state has the power to alter its law and develop unique legal doctrines. The extent to which states are competent to create rules of maritime law and the extent to which generally applicable state law applies in maritime cases are highly complicated and controversial issues. The issues addressed here are part of that larger controversy. 3 See Lauritzen v Larsen, 345 U.S. 571 (1953) (establishing a seven-factor test to resolve choice-of-law questions in maritime cases); see also Romero v International Terminal Operating Co., 358 U.S. 354 (1959) (applying Lauritzen); Hellenic Lines Ltd. v Rhoditis, 398 U.S. 306 (1970) (adding an eighth factor to the analysis). 4 348 U.S. 310 (1955). 1

223

224  Research handbook on marine insurance law

2.

THE CONTEXT IN WHICH CHOICE-OF-LAW QUESTIONS ARISE

Some basic background in US constitutional law and the law of federal courts helps put Wilburn Boat and the choice-of-law issues in context. The US Constitution’s Admiralty Clause extends the federal judicial power “to all cases of admiralty and maritime jurisdiction.”5 That provision, as implemented by Congress,6 authorizes federal courts to decide maritime cases and gives them the “power and responsibility […] for fashioning the controlling rules of admiralty law.”7 In conjunction with the Constitution’s “Necessary and Proper Clause,”8 it also empowers Congress to “alter, qualify, or supplement [admiralty and maritime law] as experience or changing conditions might require.”9 When the Constitution was adopted, it was not self-evident that marine insurance cases fell within the “admiralty and maritime jurisdiction.” Contemporary English admiralty jurisdiction did not extend to marine insurance contracts because they were neither made at sea nor performed at sea.10 In De Lovio v Boit,11 however, Justice Story, sitting as a circuit justice, wrote a long and detailed opinion examining admiralty jurisdiction and holding that actions on marine insurance contracts are indeed “admiralty and maritime” cases.12 More than half a century later, the full Court confirmed that conclusion,13 and it has not been seriously questioned since.14 As a result, federal courts have the power to decide marine insurance cases, and both Congress and the federal courts have the power to fashion governing rules of substantive marine insurance law.15 5 Article III, section 2, clause 3 provides that “[t]he judicial Power shall extend […] to all Cases of admiralty and maritime Jurisdiction.” The precise definition of “admiralty and maritime jurisdiction” remains controversial. The Supreme Court has addressed the subject frequently but has not always spoken clearly. Fortunately, the controversial aspects are not relevant here. However unclear or irrational the law may be on admiralty jurisdiction generally, the courts have long held that marine insurance contracts are maritime contracts and marine insurance disputes accordingly fall within admiralty jurisdiction. See notes 11–14 and accompanying text. 6 See 46 U.S.C. § 1333. 7 Fitzgerald v United States Lines, 374 U.S. 16, 20 (1963). 8 Article I, section 8 of the Constitution provides in pertinent part that “[t]he Congress shall have power […] to make all laws which shall be necessary and proper for carrying into execution the foregoing powers, and all other powers vested by this constitution in the government of the United States, or in any department or officer thereof.” 9 Panama Railroad Co. v Johnson, 264 U.S. 375, 393 (1924). 10 See, for example, 4 Coke, Institutes of the Laws of England 134–9. A marine insurance contract would typically have been concluded on land (perhaps even in a coffee house) and it would have been performed on land (where the premium was paid and where any claim was paid). 11 7 Fed. Cas. 418, 444, 2 Gall. 398, 1997 AMC 550 (C.C. D. Mass. 1815) (No. 3776) (Story J). 12 Many of the landmark maritime cases in the nineteenth century involved the US rejection of artificial English constraints on admiralty jurisdiction. See, for example, The Genesee Chief v Fitzhugh, 53 U.S. (12 How.) 443 (1851) (holding that admiralty jurisdiction is not limited to tidal waters). 13 See Insurance Co. v Dunham, 78 U.S. (11 Wall.) 1, 29–36 (1871). Interestingly, the successful counsel’s argument rested in part on the need for admiralty jurisdiction in order to apply uniform federal maritime law to marine insurance cases. See 78 U.S. (11 Wall.) at 13–14. 14 Even the Wilburn Boat Court reaffirmed the well-established jurisdictional point. See 348 U.S. at 313. 15 The Wilburn Boat Court acknowledged the power of Congress to legislate in the field. See 348 U.S. at 319. It also acknowledged that courts have the power to formulate new rules of marine insurance,

Choice-of-law issues in marine insurance cases in the United States  225 Despite the undoubted federal power over “admiralty and maritime” cases, the First Congress also preserved a role for the states. The First Judiciary Act, which created the federal district courts and gave them admiralty jurisdiction, expressly recognized a plaintiff’s right to bring most16 maritime causes of action in non-admiralty courts—including state courts.17 In pertinent part, the 1789 formulation provided: “The district courts […] shall […] have exclusive original cognizance of all civil causes of admiralty and maritime jurisdiction […] saving to suitors, in all cases, the right of a common law remedy, where the common law is competent to give it.”18 The “saving-to-suitors” clause has been treated as a significant qualification on the “exclusiv[ity]” of the federal grant. If a state court has the power (under its own jurisdictional rules) to decide a marine insurance case—as it typically will—then Congress has preserved that power, notwithstanding the federal district court’s otherwise “exclusive original cognizance” of the matter. The existence of state court jurisdiction, however, does not necessarily mean that state substantive law applies. When a maritime case19 is brought in state court under the saving-to-suitors clause, the state court is obligated to apply the same substantive law that the federal admiralty court would have applied.20 The ordinary presumption in US maritime law is even though it declined to exercise that power in the context of that case. The Supreme Court has exercised its power to create new rules for maritime contracts in other contexts. See, for example, Norfolk Southern Railway Co. v James N. Kirby, Pty Ltd., 543 U.S. 14 (2004) (fashioning a “limited” agency rule to bind a shipper to a Himalaya clause in a contract to which it was not a party). And lower federal courts have sometimes exercised their power to create new rules of substantive marine insurance law. See, for example, Aasma v American Steamship Owners Mutual Protection & Indemnity Association, 95 F.3d 400 (6th Cir. 1996) (fashioning a federal admiralty rule upholding a P&I Club’s pay-to-be-paid clause in the face of a direct action by a plaintiff with a judgment against a bankrupt member). 16 Federal admiralty jurisdiction is truly exclusive over actions in rem and over several statutory maritime actions that are typically not relevant in the marine insurance context. 17 The “saving to suitors” clause also protects a plaintiff’s ability (if applicable) to bring a maritime action in federal court on some jurisdictional basis other than admiralty. The most common bases for federal jurisdiction are “federal question” and “diversity.” Under federal-question jurisdiction, a federal court can decide cases arising under federal law. See 28 U.S.C. § 1331. Under diversity jurisdiction, a federal court can decide disputes between “diverse” litigants (for example, litigants from different states). See 28 U.S.C. § 1332. Federal courts have admiralty jurisdiction under 28 U.S.C. § 1333, which is quoted in note 18. 18 First Judiciary Act, c. 20, § 9, 1 Stat. 76–7 (1789). The present formulation, 28 U.S.C. § 1333(1), provides that “[t]he district courts shall have original jurisdiction, exclusive of the courts of the states, of […] any civil case of admiralty or maritime jurisdiction, saving to suitors in all cases all other remedies to which they are otherwise entitled.” The Supreme Court has held that the current version and the original have the same meaning. See Madruga v Superior Court, 346 U.S. 556 (1954). Prof. Black described the saving-to-suitors clause as a “sort of quasi-constitutional statutory law.” Charles L. Black, Jr., Admiralty Jurisdiction: Critique and Suggestions, 50 Colum. L. Rev. 259, 260 (1950). 19 A “maritime” case is a case—wherever filed—that would have fallen within the admiralty jurisdiction if the plaintiff had chosen to file it in admiralty. 20 This rule emanates from the Constitution’s grant of federal admiralty power in article III, section 2 (see note 5) and the Supremacy Clause in article VI, which provides that federal law “shall be the supreme Law of the Land.” The Supreme Court has recognized this rule, known as the “reverse-Erie” doctrine, at least since Chelentis v Luckenbach S.S. Co., 247 U.S. 372 (1918). See also, for example, Offshore Logistics, Inc. v Tallentire, 477 U.S. 207, 222–3 (1986). (The “reverse-Erie” doctrine is so named because it mirrors the rule of Erie Railroad Co. v Tompkins, 304 U.S. 64 (1938), under which a federal court in a “diversity” case must apply the same law that the local state court would have

226  Research handbook on marine insurance law that “[w]ith admiralty jurisdiction comes the application of substantive admiralty law.”21 Prior to Wilburn Boat, therefore, it was widely presumed that federal maritime law would apply in preference to inconsistent state law in any action involving marine insurance.22 The presumption in favor of the application of substantive federal maritime law in maritime cases is not absolute. Justice Frankfurter, writing for the Court in Romero v International Terminal Operating Co.,23 cautioned against applying the presumption too broadly: “[T]o claim that all enforced rights pertaining to matters maritime are rooted in federal law is a destructive oversimplification of the highly intricate interplay of the States and the National Government in their regulation of maritime commerce.”24 The Supreme Court has upheld the application of state law in such diverse contexts as maritime pollution,25 transnational forum non conveniens,26 and the death of a recreational boater in territorial waters.27 But to this day, Wilburn Boat remains the most striking example of the Supreme Court’s recognition of the applicability of state law in an admittedly maritime case.

3.

THE SUPREME COURT’S WILBURN BOAT DECISION

For more than two-thirds of a century, vertical choice-of-law principles in US marine insurance cases have been based on the Supreme Court’s 1955 decision in Wilburn Boat Co. v Fireman’s Fund Insurance Co.28 Because the analysis today still turns on that case, it is valuable to examine it in some detail. The decision originated in some very mundane facts.29 In May 1947, the Fireman’s Fund Insurance Company, a California corporation, issued a marine hull policy on the Wanderer, a small houseboat then located in Mississippi waters. The insurer issued the policy in Illinois through an Illinois broker to assureds who resided in Iowa and Illinois. It provided, among other things, that the vessel could neither be sold nor pledged without the

applied.) Similarly, when a maritime case is brought in federal court under the “saving to suitors clause,” that is, without invoking admiralty jurisdiction, the non-admiralty federal court still applies the same law that it would have applied if it had been sitting in admiralty. See, for example, Pope and Talbot, Inc. v Hawn, 346 U.S. 406, 410–11 (1953). 21 East River S.S. Corp. v Transamerica Delaval Inc., 476 U.S. 858, 864 (1986). See also, for example, Executive Jet Aviation, Inc. v City of Cleveland, 409 U.S. 249, 255 (1972) (satisfying jurisdictional requirements “invokes […] the full panoply of the substantive admiralty law”). 22 See, for example, Edwin W. Patterson, Essentials of Insurance Law § 10, at 58 (2nd ed. 1957). 23 358 U.S. 354 (1959). 24 358 U.S. at 373. Justice Frankfurter cited the “saving to suitors” clause as authority for the role that Congress had preserved for the states in the development of substantive maritime law. See 358 U.S. at 371–2. 25 See, for example, Askew v American Waterways Operators, Inc., 411 U.S. 325 (1973). 26 American Dredging Co. v Miller, 510 U.S. 443 (1994). 27 Yamaha Motor Corp., U.S.A. v Calhoun, 516 U.S. 199 (1996). 28 Note 4. 29 As is typical, the Supreme Court’s opinion gives only a bare outline of the facts. More details can be found in some of the lower courts’ opinions (particularly on remand from the Supreme Court), the Transcript of Record, and in the secondary literature. For the most detailed account of the case that is readily available to most readers, see Joel K. Goldstein, The Life and Times of Wilburn Boat: A Critical Guide, 28 J. Mar. L. & Com. 395 (1997).

Choice-of-law issues in marine insurance cases in the United States  227 insurer’s consent.30 Furthermore, the assureds could use the vessel “solely for private pleasure purposes,” and it could not be “hired or chartered” without the insurer’s permission.31 In June 1948, three brothers from Denison, Texas—Glenn, Frank, and Henry Wilburn— purchased the Wanderer, and the insurer indorsed the policy in favor of the new owners doing business as a partnership known as “Wilburn’s Boat Company.” They proceeded to move the vessel to Lake Texoma, an artificial lake on the border between Texas and Oklahoma that had been created in 1944 by the Denison Dam. A policy indorsement authorized the trip and provided that the Wanderer would thereafter be confined to Lake Texoma. In September 1948, the brothers sold the Wanderer to the “Wilburn Boat Company,” an Oklahoma corporation that they owned. The insurer did not consent to that sale. On three occasions, the Wilburn brothers or their corporation pledged the vessel to secure promissory notes. The insurer did not consent to those transactions. Finally, the owners leased the vessel on several occasions and carried passengers for hire. Although a survey sent to the insurer in February 1949 partially disclosed the planned commercial use of the vessel, the insurer did not give its required permission. It was undisputed that each of those actions breached the policy. On February 25, 1949, a fire destroyed the Wanderer while it was moored approximately 300 feet off the Oklahoma shore of the lake. The origin of the fire remains unknown, but it was undisputed that the policy breaches did not cause the loss. When the Wilburn brothers made a claim under the policy, which by its terms covered loss due to fire, Fireman’s Fund declined to pay the claim and instead returned the premiums. It argued that when an assured breaches a warranty in a marine insurance policy the general maritime law permits the insurer to avoid paying a claim for a subsequent loss—even if the breach of warranty was unrelated to the loss. The Wilburns argued that Texas law, rather than the general maritime law, governed the policy. Under Texas law, policy breaches relating to the sale and use of the vessel would not defeat coverage unless they had contributed to the loss,32 and the anti-encumbrance provision in the policy would be ineffective.33 In June 1949, the litigation odyssey began when the three brothers and their company sued the insurer in a Texas state court to recover more than $40,000 under the insurance contract. Fireman’s Fund, asserting diversity jurisdiction,34 removed35 the case the following month to federal district court. In December 1951, the district court ruled that federal maritime law governed and that—because of the “literal compliance” rule for marine insurance warranties—the Wilburns were not entitled to any recovery. On appeal, the US Court of Appeals for the Fifth

Wilburn Boat, note 4, 348 U.S. at 311 n.1. Ibid. 32 See 348 U.S. at 312 n.3 (quoting applicable Texas statute). 33 See 348 U.S. at 312 n.2 (quoting applicable Texas statute). 34 Diversity jurisdiction gives a federal court the power to hear certain cases when the parties are “diverse”—typically from different states. 28 U.S.C. § 1332. When a federal court hears a case under its diversity jurisdiction, it sits as a non-admiralty court. Cf. note 17. Different procedures apply depending on whether the court sits in admiralty or hears the case under its diversity jurisdiction, but the substantive law is the same regardless of the court’s jurisdictional basis for hearing the case. See note 20. 35 “Removal” permits a defendant, under certain circumstances, to defeat a plaintiff’s choice of forum and transfer a case from a state court to federal court. See generally, for example, Michael F. Sturley, Removal into Admiralty: The Removal of State-Court Maritime Cases to Federal Court, 46 J. Mar. L. & Com. 105 (2015). 30 31

228  Research handbook on marine insurance law Circuit affirmed.36 Because of the “[i]mportance of the questions involved,”37 the Supreme Court “granted certiorari,” meaning that it agreed to decide the case on the merits. On February 28, 1955, just over six years after the fire, the Supreme Court reversed the Fifth Circuit’s decision and remanded the case to the district court “for a trial under appropriate state law.”38 Justice Black wrote the Court’s opinion. Justice Frankfurter concurred in the Court’s judgment but rejected much of Justice Black’s reasoning. Two justices dissented. Justice Black, having noted that no relevant federal legislation applied, began his analysis with two questions: “(1) Is there a judicially established federal admiralty rule governing these warranties? (2) If not, should we fashion one?”39 In answering the first question, Justice Black distinguished or ignored several cases that appeared on their face to establish the literal compliance rule.40 He thus concluded that the rule “has not been judicially established as part of the body of federal admiralty law in this country.”41 He did not offer any guidance on what is required for a rule to become “judicially established”42 or any justification for why the question was worth asking in the first place. Turning to the second question, the Court declined to fashion a “new” admiralty rule for two principal reasons. First, the regulation of insurance has historically been a matter for the states (although the federal government has the power to regulate insurance if it chooses), and Congress has recognized and acted upon that division of responsibility.43 Second, even if the Court wished to fashion a new rule, doing so would be a complex and difficult task that courts

36 Wilburn Boat Co. v Fireman’s Fund Ins. Co., 201 F.2d 833 (5th Cir. 1953), rev’d, 348 U.S. 310 (1955). 37 Wilburn Boat, note 4, 348 U.S. at 313. After Wilburn Boat, the Court’s view of the importance of marine insurance cases changed quickly. As of this writing, the Court has not decided a marine insurance case since Wilburn Boat. But see notes 97, 145–9 and accompanying text (discussing a marine insurance case that is pending at the Supreme Court as this publication goes to press). 38 348 U.S. at 321. The Supreme Court did not discuss which state’s law was “appropriate.” See 348 U.S. at 313 n.6. On remand, the litigation odyssey continued for over seven more years. Ultimately, the lower courts held that Texas law (rather than Illinois law) governed the warranty question. But the insurer could still avoid the policy under the uberrimae fidei doctrine because the Wilburns were guilty of eight material misrepresentations or nondisclosures. See Fireman’s Fund Insurance Co. v Wilburn Boat Co., 300 F.2d 631 (5th Cir. 1962). The court of appeals concluded that federal maritime law should govern that question because, under the Supreme Court’s decision, state law is relevant “only where ‘entrenched federal precedent is lacking’ with respect to a specific issue.” 300 F.2d at 647 n.12. But the “rule of concealment in marine insurance is solidly entrenched in our body of federal maritime law.” Ibid. Because the result was the same under Texas or federal maritime law, however, the court found the point to be “of minimal significance to a decision here.” 39 348 U.S. at 314. 40 Scholars have found the rejection of those prior cases to be highly questionable. See, for example, Grant Gilmore & Charles L. Black, Jr., The Law of Admiralty, § 2–8, at 68 & n.71a (2nd ed. 1975); Goldstein, note 29, at 419–25. 41 348 U.S. at 316. 42 See notes 54–7 and accompanying text. 43 348 U.S. at 316–19. This argument appears to overlook the distinction between the substantive law of marine insurance and the regulatory rules governing those in the marine insurance industry. The substantive law addresses the private, commercial law aspects of the field, covering issues such as the formation and interpretation of marine insurance contracts, subjects of marine insurance and remedies available under marine insurance contracts. The regulatory rules, in contrast, address the public, administrative law aspects of the subject, covering issues such as the requirements that must be satisfied before a company or a broker is entitled to conduct business, the regulation of insurance companies, and the like.

Choice-of-law issues in marine insurance cases in the United States  229 are poorly equipped to undertake.44 In Wilburn Boat, for example, Justice Black was clearly uncomfortable with the “harsh” literal compliance rule, but apparently felt more uncomfortable at the prospect of choosing a new rule to replace it.45 Deferring the problem to Congress or the states, with their greater expertise and experience, was much easier. Justice Frankfurter wrote an opinion “concurring in the result,”46 which meant that he accepted the Court’s judgment (ordering a remand to decide the case under state law) but that he did not accept the reasoning in the majority opinion. In essence, he argued for a middle ground under which cases requiring a uniform rule throughout the country would be governed by federal maritime law while cases of essentially local interest could be governed by state law. Because he thought this case arising on an inland lake was of merely local interest, he had no objection to the application of state law.47 But because he thought the reasoning in the majority opinion was unnecessarily broad and could be “directed with equal force to oceangoing vessels in international maritime trade,”48 he refused to join—and, indeed, harshly criticized—the majority opinion.49 Justice Reed, joined by Justice Burton, dissented.50 He hinted that he would be prepared, as a matter of federal maritime law, to modify the literal compliance rule “insofar as the breached warranty does not contribute to the loss.”51 Until Congress or the Court modified the rule, however, he argued that it should apply in all maritime cases to preserve uniformity.

4.

VERTICAL CHOICE-OF-LAW IN LIGHT OF WILBURN BOAT

Immediately after Wilburn Boat, it was unclear what impact the case would have. Soon after the decision, Professors Gilmore and Black speculated in the first edition of their highly respected treatise:

Ibid at 319–20. Ibid at 320. Professor Goldstein found considerable evidence in several of the Justices’ private papers, which have since become available to scholars, that the result in Wilburn Boat was largely driven by the equities of the case. See Goldstein, note 29, at 410–17. It is clear that Justice Black, in particular, wished to avoid the “harsh” literal compliance rule. Requiring the application of state law was an easy way to accomplish this result (assuming that the lower courts applied Texas law on remand—an assumption that was not only obvious in the Court’s opinion but also justified by the ultimate events; see note 38). Formulating a new rule to replace the literal compliance rule would probably have been no more difficult than many of the other tasks that common-law courts regularly undertake, but the Wilburn Boat Court may have felt that it still was not worth the effort. 46 348 U.S. at 321 (Frankfurter J, concurring in the result). 47 Ibid at 322 (Frankfurter J, concurring in the result). 48 Ibid at 323 (Frankfurter J, concurring in the result). 49 Justice Frankfurter—in language foreshadowing his subsequent opinion in Romero; see notes 23–4 and accompanying text—was also critical of Justice Reed’s dissent: “[T]he demand for uniformity is not inflexible and does not preclude the balancing of the competing claims of state, national and international interests […] In rejecting abdication of all responsibility by this Court for uniformities in marine insurance and its complete surrender to the States, one is not required to embrace another absolute, complete absorption by this Court of the field of marine insurance and entire exclusion of the States.” 348 U.S. at 323–4 (Frankfurter J, concurring in the result). 50 348 U.S. at 324 (Reed J, dissenting). 51 Ibid at 326 (Reed J, dissenting). 44 45

230  Research handbook on marine insurance law Wilburn may mean merely that the States are to have a limited competency to regulate certain terms of marine policies. It could as a matter of cold logic be read to mean that there is no federal maritime law at all. It may very well turn out to mean anything between these extremes.52

In practice, the subsequent cases occupy a broad range between those extremes. A principal reason for the wide range of views in the lower courts is the Supreme Court’s failure to provide any meaningful guidance. The Wilburn Boat opinion did not explain how the new rule should be applied. And the Supreme Court has provided no guidance in the intervening decades; it has not decided a marine insurance case since Wilburn Boat.53 The lack of guidance starts with the most fundamental issues. The Court declared that the first question to consider was whether “a judicially established federal admiralty rule” governed the relevant issue,54 but it did not explain what was required for a rule to become “judicially established.” Presumably two court of appeals decisions would not be enough because the Wilburn Boat Court concluded that the literal compliance rule was not sufficiently established when “only two circuits appear to have thought of the rule as a part of the general admiralty law.”55 Would one Supreme Court decision have been enough? Professors Gilmore and Black argued that Insurance Co. v Thwing56 “seems squarely to have decided this very point, or at least inevitably to have rested on the assumption of the correctness of the strict rule.”57 But the Wilburn Boat Court did not cite Thwing, so its potential impact remains a mystery. Kossick v. United Fruit Co.,58 which was not a marine insurance action, clarified at least that Wilburn Boat did not require state law to govern in every admiralty case. Applying federal maritime law rather than the New York statute of frauds to a contract between a seaman and his employer, the Kossick Court distinguished Wilburn Boat with the observation that “the situation presented here [in Kossick] has a more genuinely salty flavor than that [in Wilburn Boat].”59 Some lower courts picked up on that cue and attempted to limit Wilburn Boat to the maritime-but-local context60 as Justice Frankfurter suggested in his concurring opinion.61 Most lower courts have applied Wilburn Boat more broadly,62 at least in the marine insurance context. Grant Gilmore & Charles L. Black, Jr., The Law of Admiralty, § 2–8, at 63 (1st ed. 1957). As this publication goes to press, the Supreme Court is poised to decide its first marine insurance case since Wilburn Boat. See notes 97, 145–9 and accompanying text. But the application of the Wilburn Boat rule is not an issue before the Court. 54 Wilburn Boat, note 4, 348 U.S. at 314. 55 Ibid at 315. 56 80 U.S. (13 Wall.) 672 (1872). 57 Gilmore & Black, note 40, § 2–8, at 68 and n.71a. 58 365 U.S. 731 (1961). 59 Ibid at 742. Kossick is particularly relevant to the interpretation of Wilburn Boat because four of the five members of the Wilburn Boat majority (including Justice Black, the author of the Wilburn Boat opinion) joined the Kossick opinion. 60 See, for example, Aasma v American Steamship Owners Mutual Protection & Indemnity Association, 95 F.3d 400, 404 (6th Cir. 1996). 61 See 348 U.S. at 322–3 (Frankfurter J, concurring in the result). See also notes 47–9 and accompanying text. 62 As Professor Goldstein has noted, “if the Court hoped its reinterpretation of Wilburn [in Kossick] would cabin the decision’s mischievous potential its efforts met with limited success. Some failed to get the message; others concentrated on the discussion in Wilburn rather than on the dicta in Kossick.” Goldstein, note 29, at 571. 52 53

Choice-of-law issues in marine insurance cases in the United States  231 Commentators and lower courts have suggested several other ways to limit Wilburn Boat.63 At one end of the spectrum, some have suggested that it requires the application of state law only to the effect of warranties (the precise issue before the Supreme Court),64 although this is a difficult distinction to defend.65 One district court read Wilburn Boat to say “that federal admiralty law should apply to issues that are maritime in nature and that state law should apply to issues that are common to all sorts of insurance contracts.”66 Within the marine insurance field, the choice-of-law principles remain unclear.67 Despite widespread criticism of Wilburn Boat,68 the lower courts have not uniformly or predictably limited or distinguished it.69 For every case that cuts back on the broad application of Justice Black’s reasoning, another case extends the reach of the decision. If anything, the sporadic efforts to distinguish or limit the case have probably made the situation worse, as each distinction becomes just one more variable for the parties to consider when predicting how a marine insurance dispute will be resolved. Even on questions of methodology, lower courts are spread out along a spectrum. Some courts continue to apply federal law in marine insurance cases, usually because they find an established federal admiralty rule70 but sometimes because they conclude that they should

Professors Gilmore and Black propose some potential distinctions that they then reject as unjustifiable. See Gilmore and Black, note 40, § 2–8, at 69–70; cf. Goldstein, note 29, at 580–1. 64 See, for example, Goldstein, note 29, at 435–7; ibid at 579 and nn.424–5. Professor Goldstein finds some support for this reading in both Kossick, 365 U.S. at 742, and Romero v International Terminal Operating Co., 358 U.S. 354, 373 (1959). See Goldstein, note 29, at 569 & n.348. 65 See, for example, Gilmore & Black, note 40, § 2–8, at 69–70. 66 Home Insurance Co. v Vernon Holdings, 1995 AMC 369, 372 (S.D. Fla. 1994). 67 Wilburn Boat ultimately had little influence outside the marine insurance context. 68 See, for example, George Waddell, Current Issues and Developments in Marine Insurance, 6 U.S.F. Mar. L.J. 185, 187 (1993) (“The Wilburn Boat decision has been universally criticized. Indeed, there appears to have been little if any favorable comment in the subsequent literature—at least none that is widely known”); Goldstein, note 29. 69 A good example of the lower courts’ unwillingness forthrightly to limit Wilburn Boat can be found in the Second Circuit’s two opinions in Youell v Exxon Corp. See 48 F.3d 105 (2d Cir.), vacated 516 U.S. 801 (1995); and 74 F.3d 373 (2d Cir. 1996) (per curiam). Although the court was willing essentially to disregard Wilburn Boat (without principled explanation), it did not take the simple step of declaring that the Exxon Valdez disaster “has a more genuinely salty flavor than” the Wilburn Boat fire on Lake Texoma. 70 Most of the courts of appeals to address the issue have ruled that uberrimae fidei doctrine is an established federal admiralty rule. See, for example, Quintero v Geico Marine Insurance Co., 983 F.3d 1264, 1270–1 (11th Cir. 2020); Catlin (Syndicate 2003) at Lloyd’s v San Juan Towing & Marine Services, Inc., 778 F.3d 69, 81 (1st Cir. 2015); AGF Marine Aviation & Transport v Cassin, 544 F.3d 255, 262 (3d Cir. 2008); Certain Underwriters at Lloyd’s v Inlet Fisheries Inc., 518 F.3d 645, 654 (9th Cir. 2008); Ingersoll Milling Mach. Co. v M/V Bodena, 829 F.2d 293, 308 (2d Cir. 1987); but see Albany Insurance Co. v Anh Thi Kieu, 927 F.2d 882, 890 (5th Cir. 1991). The courts of appeals have also found some other established federal admiralty rules governing marine insurance. See, for example, GEICO Marine Insurance Co. v Shackleford, 945 F.3d 1135, 1142 (11th Cir. 2019) (express navigational-limit warranty); Galilea, LLC v AGCS Marine Insurance Co., 879 F.3d 1052, 1058 (9th Cir. 2018) (enforcement of arbitration clauses); Hilton Oil Transport v Jonas, 75 F.3d 627, 630 (11th Cir. 1996) (express trading-limit warranty in the absence of a “held covered” clause); American National Fire Insurance Co. v Kenealy, 72 F.3d 264, 270-271 (2d Cir. 1995) (award of attorney’s fees); Thanh Long Partnership v Highlands Ins. Co., 32 F.3d 189, 193–4 (5th Cir. 1994) (implied warranty of seaworthiness and the interpretation of Inchmaree clauses). 63

232  Research handbook on marine insurance law fashion one.71 Not all courts dutifully proceed through Justice Black’s two-question analysis. Indeed, the second question—should the court fashion a federal admiralty rule—is regularly ignored.72 Some courts ignore even the first question, apparently applying state law simply because the case before it involves marine insurance.73 To further complicate the issue, it is surprisingly unclear—at least in one circuit—whether a court will treat a rule as sufficiently “established” under Wilburn Boat even when it has announced in a previous case that it is. In Albany Insurance Co. v Anh Thi Kieu,74 the principal issue was whether the doctrine of uberrimae fidei was an established rule of federal admiralty law.75 In 1962, on remand from the Supreme Court,76 the Fifth Circuit declared that the doctrine was “solidly entrenched in our body of federal maritime law.”77 Five years later, that court described the doctrine as “established”78 in the federal law of marine insurance. But in 1991, the Anh Thi Kieu court distinguished those cases, dismissed the relevant statements as mere dicta, and held that the “doctrine is entrenched no more.”79 The court reasoned that the “spotty application” of the uberrimae fidei doctrine “in recent years” (“even in other circuits”) “suggests” that the doctrine is no longer sufficiently established.80 One impact of asking whether a rule is “entrenched” or “judicially established” is that the choice of law depends to some extent on the frequency with which issues are litigated. Some of the most basic legal principles are never litigated (or at least have not been litigated for

See, for example, Aasma v American Steamship Owners Mutual Protection & Indemnity Association, 95 F.3d 400, 404 (6th Cir. 1996). In Aasma, personal injury plaintiffs with default judgments against a bankrupt shipowner brought direct actions against the owner’s former P&I Clubs, which had provided coverage during the period in which the injuries arose. The Clubs asserted defenses under their “pay-to-be-paid” clauses. The Sixth Circuit concluded that no existing federal admiralty rule addressed the validity of such clauses, but that the need for a single, uniform rule in this “uniquely maritime” context justified fashioning one. The court thus recognized the validity of a “pay-to-be-paid” clause as a matter of federal maritime law. 72 See, for example, Travelers Property Casualty Co. of America v Ocean Reef Charters, LLC, 996 F.3d 1161, 1163, 1171 (11th Cir. 2021); Royal Insurance Co. of America v KSI Trading Corp., 563 F.3d 68, 73 (3d Cir. 2009); Carrier v RLI Insurance Co., 854 F. Supp. 2d 1324, 1326 (S.D. Ga. 2010). A First Circuit panel at least explained why it did not give careful consideration to Justice Black’s second question: “Under a well-established principle of federal law applicable to cases of this genre [that is, marine insurance cases], if federal and state law collide, then the federal rule prevails […] But in the absence of such a conflict, Wilburn Boat has generally been interpreted, ‘in deference to state hegemony over insurance, to discourage the fashioning of new federal law and to favor the application of state law.” Acadia Insurance Co. v McNeil, 116 F.3d 599, 603 (1st Cir. 1997) (quoting Windsor Mount Joy Mutual Insurance Co. v Giragosian, 57 F.3d 50, 54 (1st Cir. 1995), quoting Albany Insurance Co. v Wisniewski, 579 F. Supp. 1004, 1013–14 (D. R.I. 1984)). 73 See, for example, Cal-Dive International, Inc. v Seabright Ins. Co., 627 F.3d 110, 113 (5th Cir. 2010) (citing Wilburn Boat for the proposition that “[t]he interpretation of a marine policy of insurance is governed by relevant state law”). 74 927 F.2d 882 (5th Cir. 1991). 75 Most circuits treat the uberrimae fidei doctrine as an established rule of federal admiralty law. See note 70. 76 See note 38. 77 Fireman’s Fund Insurance Co. v Wilburn Boat Co., 300 F.2d 631, 647 n.12 (5th Cir. 1962). See also note 38. 78 Gulfstream Cargo, Ltd. v Reliance Insurance Co., 409 F.2d 974, 980 (5th Cir. 1969). 79 Anh Thi Kieu, note 74, 927 F.2d at 890. 80 Ibid at 889–90. 71

Choice-of-law issues in marine insurance cases in the United States  233 decades) because they are so basic that no one would challenge them. As a result, courts do not rule on those principles and litigants can argue that they are not “judicially established.” That alone may not be a problem; the most basic principles will likely be the same under state or federal law, so it does not matter which applies. But Anh Thi Kieu raises the possibility that a rule may be established in federal law and go unchallenged for decades precisely because it is established. If state law evolves in the meantime, a litigant seeking the application of that new state law may then argue that the previously settled federal rule is “entrenched no more,” in other words, is no longer “judicially established.”

5.

HORIZONTAL CHOICE-OF-LAW AFTER WILBURN BOAT

Although the Wilburn Boat Court gave virtually no guidance on how to decide whether federal or state law should apply in any given situation, it at least offered the illustration of its own analysis concerning the literal compliance rule.81 The Court gave absolutely no guidance on how to decide which state’s law should apply when federal law does not. It simply noted that the horizontal choice-of-law problem was not before it82 and remanded the case to permit the lower courts to resolve the issue.83 Justice Frankfurter’s concurring opinion implicitly suggested a choice-of-law rule when he asked, “Is it to be assumed that were the Queen Mary, on a world pleasure cruise, to touch at New York City, New Orleans and Galveston, a Lloyd’s policy covering the voyage would be subjected to the varying insurance laws of New York, Louisiana and Texas?”84 It is doubtful that Justice Frankfurter himself would have adopted a location-of-the-loss rule or a law-of-the-forum rule for determining the law governing a marine insurance contract if he had actually been called upon to make that decision. It is at least clear that he did not think the law governing the Queen Mary’s insurance policy should vary as the vessel called at different ports. In any event, the lower courts have not been applying either rule in marine insurance cases.85 The complications and difficulties facing the lower courts in deciding horizontal choice-of-law issues are well illustrated by a quartet of marine insurance decisions within a single circuit over just six years. In 1985, the Fifth Circuit held that “the law of the state where the marine insurance contract was issued and delivered is the governing law.”86 Two years later, the same court instead declared that “the law of the state in which the [marine insurance] contract was formed” governs.87 Another two years later the court announced yet a different rule: “In identifying the appropriate state law to apply, we look to the state having the greatest interest in the resolution of the issues.”88 Two years after that, the court reviewed See notes 54–7 and accompanying text. See Wilburn Boat, note 4, 348 U.S. at 313 n.6. 83 See ibid at 321. 84 348 U.S. at 323 (Frankfurter J, concurring in the result). 85 See notes 86–91 and accompanying text. 86 Elevating Boats, Inc. v Gulf Coast Marine, Inc., 766 F.2d 195, 198 (5th Cir. 1985). See also Gulf Fleet Marine Operations, Inc. v Wartsila Power, Inc., 797 F.2d 257, 261 (5th Cir. La. 1986). 87 Graham v Milky Way Barge, Inc., 811 F.2d 881, 885 (5th Cir. 1987). 88 Truehart v Blandon, 884 F.2d 223,226 (5th Cir. 1989). See also Transco Exploration Co. v Pacific Employers Insurance Co., 869 F.2d 862, 863 (5th Cir. 1989). 81 82

234  Research handbook on marine insurance law the field and, in an attempt to reconcile the cases, declared that it would follow a two-step process. In step one, the court must identify the states in which the policy was formed, issued, and delivered. In step two, it must decide which of those states has the greatest interest in the application of its law.89 Ironically, on remand in Wilburn Boat itself the courts in the Fifth Circuit did not follow any of those four choice-of-law approaches. The Wilburn Boat policy was originally formed, issued, and delivered in Illinois, but the district court on remand held that the parties “in effect” concluded a new policy when the insurer indorsed the policy in favor of the new owners.90 Even that new policy was apparently formed, issued, and delivered in Illinois, where the broker was located. The district court nevertheless held (and the Fifth Circuit agreed) that Texas law applied because a Texas statute required the application of Texas law when a company doing business in Texas (such as Fireman’s Fund) issues an insurance policy under which the proceeds would be payable to any citizen or inhabitant of Texas (such as the Wilburn brothers).91 One problem with the horizontal choice-of-law analysis is that many states do not appear to have much interest in the resolution of marine insurance disputes. Indeed, many states explicitly exclude marine insurance from significant portions of their insurance legislation.92 To further complicate the analysis, when a court has chosen a particular state’s law it is often difficult or impossible to find a relevant judicial decision or statute in the maritime context.93 The court must therefore resolve a marine insurance dispute with reasoning designed for automobile or homeowners’ insurance. In 5801 Associates, Ltd. v Continental Insurance Co.,94 for example, a decision involving the sinking of a barge in the open seas off the coast of South Carolina, the federal court felt compelled to look to the law of the inland state of Missouri. Finding no marine insurance decision on point, it followed an automobile insurance decision.95 An even more striking example of the problem arose in the litigation between Exxon and its insurers to determine coverage under a global corporate excess policy for hundreds of millions of dollars of clean-up expenses following the Exxon Valdez oil spill. One issue in the coverage dispute was whether the loss was fortuitous. The insurers argued it was not fortuitous because it was caused by Exxon’s reckless conduct in permitting a vessel-owning subsidiary to employ

Anh Thi Kieu, note 74, 927 F.2d at 890–1. See Wilburn Boat Co. v Fireman’s Fund Insurance Co., 199 F. Supp. 784, 791 (E.D. Tex. 1960), rev’d on other grounds, 300 F.2d 631 (5th Cir. 1962). 91 See ibid. 92 See, for example, Ala. Code § 27-14-2(3) (excepting “[w]et marine and transportation insurance” from chapter governing the insurance contract); La. Rev. Stat. Ann. § 22:851(A) (excepting “ocean marine and foreign trade insurances” from chapter 4 of the Insurance Code, which governs insurance and insurance contract requirements); Va. Code Ann. § 38.2-300(1) (excepting “[o]cean marine insurance other than private pleasure vessels” from chapter governing insurance policies and contracts); Wash. Rev. Code Ann. § 48.18.010 (excepting “ocean marine and foreign trade insurances” from chapter governing the insurance contract). In St. Paul Insurance Co. v Great Lakes Turnings, Ltd., 829 F. Supp. 982, 984–5 (N.D. Ill. 1993), the court relied in part on a statutory exclusion for marine insurance to decide that Illinois had no interest in the application of its law in general to the pending dispute. 93 See, for example, Acadia Insurance Co. v McNeil, 116 F.3d 599, 600 (1st Cir. 1997) (“This case involves an issue of New Hampshire law as to which we have found no decisive New Hampshire precedent”). 94 983 F.2d 662 (5th Cir. 1993) (per curiam). 95 See ibid at 666 and n.10 (following Shelter Mutual Insurance Co. v Brooks, 693 S.W.2d 810 (Mo. 1985) (en banc) (automobile insurance decision)). 89 90

Choice-of-law issues in marine insurance cases in the United States  235 a known alcoholic as the captain of the Exxon Valdez. Exxon not only denied that it had been reckless but also argued that the loss in question would have been fortuitous even if it had been reckless. Exxon contended that the fortuity rule was the same under any law (state or federal) that might be relevant, but it had filed suit in a Texas state court and taken the position that Texas law generally governed. Exxon accordingly needed authority to support its contention that Texas law permits insurance coverage for the unforeseen consequences of reckless or even intentional acts. In the absence of any statute or decision in the marine insurance context, its principal authority on that central issue was a decision of the Texas Supreme Court addressing whether a homeowners’ policy covered liability for transmitting genital herpes to a sexual partner.96

6.

POSSIBLE SOLUTIONS TO THE CHOICE-OF-LAW PROBLEMS UNDER WILBURN BOAT

Multiple efforts have been made on a global level to correct the choice-of-law problems arising from the Wilburn Boat decision. Not only have those efforts been unsuccessful thus far, but none of them seem likely to succeed in the foreseeable future. Individual solutions address the problems only one case at a time, but at least they enable the affected parties to better predict what law will govern their transactions. 6.1

Supreme Court Reconsideration of Wilburn Boat

Supreme Court reconsideration of the issue is the most obvious solution on a global level, and many have urged reconsideration, but the Court has shown no interest in addressing the problems that it created. Not only has the Court rejected every request to reconsider Wilburn Boat, but until 202397 it had declined to hear every marine insurance case since Wilburn Boat.98 Although Exxon was able to persuade the Court to review the punitive damages award in the Exxon Valdez case99 (despite the rarity with which punitive damages are awarded in maritime cases), the Court showed no interest in hearing Exxon’s coverage dispute with one of its insurers in the same case100 (despite the frequency with which insurance is a central aspect in maritime cases). Perhaps the Supreme Court will someday reconsider Wilburn Boat. A plausible opportunity is in the pipeline. In Travelers Property Casualty Co. of America v Ocean Reef Charters,

See, for example, Petition for a Writ of Certiorari at 13 and n.9, Exxon Corp. v Youell, 516 U.S. 801 (1995) (No. 94-1871) (citing State Farm Fire & Casualty Co. v S.S., 858 S.W.2d 374 (Tex. 1993)). 97 In Great Lakes Insurance SE v Raiders Retreat Realty Co., 143 S. Ct. 999 (2023), the Supreme Court agreed to hear its first marine insurance case in 68 years—but the Court denied certiorari on the question that might have allowed it to reconsider Wilburn Boat in order to focus on a narrower choice-of-law issue. See notes 145–9 and accompanying text. 98 See, for example, Morales-Vázquez v Optima Seguros, 142 S. Ct. 424 (2021) (denying cert. to QBE Seguros v Morales-Vázquez, 986 F.3d 1 (1st Cir. 2021)). 99 See Exxon Shipping Co. v Baker, 554 U.S. 471 (2008). 100 See Exxon Corp. v Youell, 517 U.S. 1251 (1995) (denying cert. to 74 F.3d 373 (2d Cir. 1996) (per curiam)). Question 1 in the petition raised the Wilburn Boat issue. 96

236  Research handbook on marine insurance law LLC,101 the Eleventh Circuit was bound to apply Wilburn Boat (as best it could), but the court of appeals’ opinion was highly critical of the Supreme Court’s decision.102 The lower court concluded by practically begging the Supreme Court to grant certiorari and “‘resolve the[] perplexities and contradictions’ created by Wilburn Boat.”103 The justices generally find requests for guidance from lower courts more persuasive than arguments by losing parties, so Ocean Reef Charters may provide a vehicle for the Supreme Court’s return to the subject. The Court has not (yet) acted on the Eleventh Circuit’s plea, however, because the losing insurer instead chose to proceed to trial in the district court. The insurer lost on remand, and the Eleventh Circuit recently affirmed.104 Perhaps the insurer will now petition for certiorari, and that will be the petition that finally persuades the Supreme Court to reconsider Wilburn Boat—but the history is not encouraging. Even if the Supreme Court were to revisit the Wilburn Boat issue, the result might well be the same.105 The Court has often recognized a role for state law in maritime cases,106 conservative justices have long shown considerable deference to “state sovereignty,”107 and the Court has shown no desire to reduce the states’ role in insurance.108 In any event, at this late date, overruling Wilburn Boat might not solve the problems. For more than 65 years, the general maritime law of marine insurance has failed to develop because in most cases the courts have either applied principles that were already settled in 1955 or looked to state law (without the authority to change it or the desire to make it part of the general maritime law).109 As one prominent practitioner explained 30 years ago, “to a great extent the federal law of marine insurance is permanently frozen where it was in 1955.”110 If the Supreme Court were to announce that the lower courts should start applying federal law again, they would have to make up for more than 65 years of inaction and develop the modern rules that are needed today. The task would take time, and meanwhile there would still be a lack of predictability and uniformity—particularly if the Supreme Court left the work entirely to the lower courts.

996 F.3d 1161 (11th Cir. 2021). Ibid at 1162, 1164–7, 1170–1. 103 Ibid at 1171 (quoting Gilmore & Black, note 40, § 2–8, at 70) (alteration by Eleventh Circuit). 104 Travelers Property Casualty Co. of America v Ocean Reef Charters, LLC, 71 F.4th 894 (11th Cir. 2023). 105 See generally, for example, Goldstein, note 29, at 587–91. Cf. US MLA doc. no. 686, at 9726 (Nov. 2, 1990) (statement of George L. Waddell) (expressing the fear that “the Supreme Court might, with all due respect, make the situation even worse” if it revisited the issue). 106 See, for example, Yamaha Motor Corp., U.S.A. v Calhoun, 516 U.S. 199 (1996) (holding that state law governed the wrongful death claim of a “non-seafarer” killed in territorial waters); American Dredging Co. v Miller, 510 U.S. 443 (1994) (permitting state statute governing forum non conveniens to apply in maritime case in state court). 107 See, for example, Seminole Tribe v Florida, 517 U.S. 44 (1996) (holding that Congress lacks the power to abrogate the states’ immunity from suit in federal court). 108 See Northeast Bancorp, Inc. v Board of Governors of the Federal Reserve, 472 U.S. 159, 179 (1985) (O’Connor J, concurring) (“the local nature of insurance is firmly ensconced in federal law”). 109 See, for example, George L. Waddell, Current Issues and Developments in Marine Insurance, 6 U.S.F. Mar. L.J. 185, 189 (1993). 110 Ibid. 101 102

Choice-of-law issues in marine insurance cases in the United States  237 6.2

A Single-Sentence Federal Statute Overruling Wilburn Boat

In 1991, the Maritime Law Association of the United States (“US MLA”) adopted a resolution calling for the enactment of a single-sentence federal statute to overrule Wilburn Boat.111 The proposed statute simply provided: “The interpretation and effect of policies of marine insurance and other insurances of marine risks shall be governed by the general maritime law and statutes of the United States.”112 Congress undoubtedly has the power to federalize marine insurance, as Wilburn Boat itself reaffirmed.113 A single-sentence statute would also have the virtue of simplicity; it would overrule Wilburn Boat and do nothing more. Whatever its virtues, the proposal was described as “dead in the water” in 1992 because no congressman or senator could be found to sponsor it.114 In 1994, the chair of the sponsoring committee reported that the proposal, “while moribund for quite some time, isn’t dead yet.”115 The proposal is undoubtedly dead now; it has not been seriously discussed in almost 30 years. Even if Congress had enacted the proposed statute, it might have been inadequate to solve the problem—for the same reason that a simple judicial overruling of Wilburn Boat might have been inadequate.116 The general maritime law of marine insurance has, for the most part, failed to develop during the last 65 years, so there is no longer an adequate body of federal law to apply.117 6.3

A Comprehensive Federal Statute to Govern Marine Insurance

In 1993, when it was clear that the proposed single-sentence statute was not progressing,118 the US MLA created an Ad Hoc Committee to study the British Marine Insurance Act of 1906119 and consider how the British Act might serve as a model for comparable US legislation. The “goal [was] a federal act which will return marine insurance to maritime law.”120 In 1995, the group published a detailed study examining the extent to which US marine insurance cases are generally consistent with the British Act121 and prepared a second study focusing on the more controversial cases.122

See US MLA doc. no. 688, at 9853 (May 3, 1991) (unanimous voice vote). Ibid at 9892. 113 See 348 U.S. at 319. 114 US MLA doc. no. 699, at 10156 (Nov. 6, 1992) (statement of George L. Waddell). 115 US MLA doc. no. 710, at 10464 (May 6, 1994) (statement of George L. Waddell). 116 See notes 109–10 and accompanying text. 117 See note 110 and accompanying text. 118 See notes 114–15 and accompanying text. 119 6 Edw. VII, c. 41 (UK). At the time, the British 1906 Act had not yet been amended. If the proposal were to return today, it would presumably consider more recent developments such as the Insurance Act 2015, c. 6 (UK), and the Consumer Insurance (Disclosure and Representations) Act 2012, c. 6 (UK). 120 US MLA doc. no. 716, at 10663 (May 5, 1995) (statement of Edward V. Cattell, Jr.); see also, for example, Edward V. Cattell, Jr., An American Marine Insurance Act: An Idea Whose Time Has Come, 20 Tul. Mar. L.J. 1, 4 (1995). 121 ‘Marine Insurance Survey: A Comparison of United States Law to the Marine Insurance Act of 1906’, 20 Tul. Mar. L.J. 5 (1995). 122 See U.S. MLA doc. no. 728, at 10548–9 (Oct. 18, 1996) (statement of Edward V. Cattell, Jr.). 111 112

238  Research handbook on marine insurance law In many ways, a comprehensive statute would be an ideal solution. The Supreme Court invited such a result in Wilburn Boat when it declared that “Congress could replace the presently functioning state regulations of marine insurance with one comprehensive Act.”123 It would avoid a fundamental problem with a simple overruling of Wilburn Boat, for there would be uniform, substantive law to fill the gap created by years of relative inactivity in the field. And lower courts would be obligated to follow a governing federal statute. The major problem is the lack of any evidence that Congress might enact such a statute. Even in the absence of opposition, inertia is a powerful force. If significant opposition arose, which seems likely,124 the prospects for Congressional action would be very slim indeed. 6.4

Restating the Law of Marine Insurance

The US MLA’s final suggestion to resolve the Wilburn Boat problem was a joint project with the American Law Institute (“ALI”) to produce a Restatement of the Law of Marine Insurance, which would follow the model of the well-known and highly respected Restatements in other fields. A Restatement could accomplish indirectly what a comprehensive federal statute might have achieved more directly. Although they are not formally binding in most jurisdictions, Restatements have tremendous persuasive authority and thus the potential to bring uniformity and predictability to a field.125 Choice-of-law issues would become irrelevant if a Restatement of the Law of Marine Insurance supplied the substantive law under any choice. And a Restatement could address the lack of new development in marine insurance law since 1955 more directly. It would not be limited to describing the law as it existed in 1955 or translating the British Act126 to correspond to the constitutional position in the United States. Rather, the ALI would take the opportunity to update the law to govern modern circumstances, looking to state,127 federal128 and international129 sources to determine the best rules available. 123 348 U.S. at 319. Cf. 348 U.S. at 334 (Reed, J., dissenting) (“broad legislative approach might be desirable”). 124 In meetings of the US MLA’s Ad Hoc Committee, representatives of the American Institute of Marine Underwriters expressed opposition to efforts to seek comprehensive federal legislation on marine insurance. 125 Restatements have long been influential in the development of maritime law. See, for example, Air & Liquid Systems Corp. v DeVries, 139 S. Ct. 986, 993–4, 2019 AMC 631, 635–7 (2019); Saratoga Fishing Co. v J. M. Martinac & Co., 520 U.S. 875, 879, 1997 AMC 2113, 2116 (1997); McDermott, Inc. v AmClyde, 511 U.S. 202, 208–9, 1994 AMC 1521, 1525 (1994); Cosmopolitan Shipping Co. v McAllister, 337 U.S. 783, 799 & n.25, 1949 AMC 1031, 1043 and n.25 (1949). See also note 154. 126 See notes 119–22 and accompanying text. 127 Looking to state law for inspiration is very different from relying on state law as authority. The Supreme Court has long looked to state law for inspiration when creating new federal maritime law. See, for example, East River S.S. Corp. v Transamerica Delaval Inc., 476 U.S. 858, 864 and n.2 (1986) (creating uniform general maritime principles drawn from state and federal sources, but not looking specifically at the law of any particular state); Moragne v States Marine Lines, Inc., 398 U.S. 375, 390 (1970) (relying on state wrongful-death statutes to justify creation of maritime wrongful death remedy). 128 Relying on federal sources would not require a return to the law of 1955. Even under Wilburn Boat, some federal courts have developed some new marine insurance principles. See note 71 and accompanying text. 129 Prior to Wilburn Boat, the Supreme Court regularly recognized the importance of maintaining international uniformity (to the extent possible) in the law of marine insurance. See, for example, Queen

Choice-of-law issues in marine insurance cases in the United States  239 When the US MLA’s Ad Hoc Committee130 decided that a federal statute was not worth pursuing, it focused its efforts on the Restatement alternative. A study group prepared a Design Proposal for submission to the ALI in May 1999. After reviewing the proposal, however, the ALI decided that the project was not a priority that the Institute would pursue. The US MLA could have pursued a similar project on its own, but the result would not have been as persuasive without the ALI’s participation and the prestige of the Restatement process. The US MLA discontinued its efforts in the field and has not returned to them in the past 20 years. 6.5

Private Solutions on an Individual Level

In the absence of a global solution, parties to insurance transactions have worked out their own solutions on an individual level. In some cases, this has meant litigation to determine the governing law. In many cases, the parties have acquiesced in the application of state law, and one state’s law has been the clear choice. But some parties have taken advantage of the obvious individual solution to the Wilburn Boat problem and included a choice-of-law clause in the insurance contract. As a general rule, US courts will enforce a contractual choice-of-law clause, at least when the chosen law has a sufficient connection with the underlying transaction.131 One insurer has been particularly aggressive in its reliance on choice-of-law clauses, and a remarkable number of reported decisions have recently addressed the issue. It appears that Great Lakes Insurance SE has routinely included the following clause in its insurance contracts for more than 15 years: It is hereby agreed that any dispute arising hereunder shall be adjudicated according to well established, entrenched principles and precedents of substantive United States Federal Admiralty law and practice[,] but where no such well established, entrenched precedent exists, this insuring agreement is subject to the substantive laws of the State of New York.132

Insurance Co. of America v Globe & Rutgers Fire Insurance Co., 263 U.S. 487, 493 (1924) (noting “special reasons for keeping in harmony with the marine insurance laws of England”); Calmar S.S. Corp. v Scott, 345 U.S. 427, 442–3 (1953) (declaring a House of Lords decision to be “persuasive authority” in a marine insurance case, citing Queen Insurance Co.). 130 See notes 118–22 and accompanying text. 131 See, for example, Great Lakes Insurance SE v Lassiter, 2022 U.S. Dist. LEXIS 78515, *15, 2022 WL 1288741, *6 (S.D. Fla. Apr. 29, 2022) (upholding the choice of New York law because New York had a substantial connection with the transaction, even if Florida had a stronger connection). 132 Great Lakes Insurance SE v Raiders Retreat Realty Co., LLC, 47 F.4th 225, 228 (3d Cir. 2022) (quoting policy) (alteration by court), cert. granted, 143 S. Ct. 999 (2023). See also, for example, Great Lakes Insurance SE v Andersson, 66 F.4th 20, 23 (1st Cir. 2023) (quoting the same clause); Great Lakes Insurance SE v Wave Cruiser LLC, 36 F.4th 1346, 1350, 1353 (11th Cir. 2022) (quoting and paraphrasing excerpts from the same clause); Great Lakes Insurance SE v Lassiter, 2022 U.S. Dist. LEXIS 78515, *12–13, 2022 WL 1288741, *5 (S.D. Fla. Apr. 29, 2022) (quoting the same clause); Great Lakes Insurance SE v Gray Group Investments, LLC, 550 F. Supp. 3d 364, 369 (E.D. La. 2021) (same). Some other insurers have also incorporated substantially the same clause into their policies. See, for example, Clear Spring Property & Casualty Co. v Viking Power LLC, 2022 U.S. Dist. LEXIS 91192, *4–5, 2022 WL 17987099, *2 (S.D. Fla. May 20, 2022) (quoting choice-of-law clause).

240  Research handbook on marine insurance law Sometimes that choice of law has not been controversial, and a court deciding a controversy arising out of such a contract has simply applied New York law without much discussion.133 But in some contexts, the choice of law is dispositive, and the enforcement of the clause is the primary issue in the case. Clear Spring Property & Casualty Co. v Viking Power LLC134 illustrates how a choice-of-law clause can protect an insurer not only from Wilburn Boat problems but also from substantive claims that might otherwise have been effective. The policy at issue in the case included a New York choice-of-law clause and a warranty that the fire-extinguishing equipment would be “maintained in good working order. This includes the weighing of tanks once a year, certification/tagging and recharging as necessary.”135 After the assured’s covered vessel was destroyed in a fire, the insurer preemptively filed an action for a declaratory judgment that it was not liable on the policy due to a breach of the warranty—failing to weigh the tanks when they were serviced each year—even though “the fire-suppression system functioned correctly on the day of the fire.”136 The assured counter-claimed for breach of contract. Because “New York law permits marine insurers to deny coverage for breaches of promissory warranties regardless of whether the breach is causally connected to a later loss,”137 the court granted summary judgment for the insurer. Under Florida law, in contrast, the insurer at least arguably138 would not have been allowed to avoid the policy unless it could prove that the breach of warranty had some causal connection to the specific loss.139 New York choice-of-law clauses have presumably allowed insurers to avoid many claims that might have succeeded if a different state’s law had applied, but two very recent cases show that a choice-of-law clause may not always achieve its intended goal. In Great Lakes Insurance SE v Andersson,140 the assured’s covered vessel was a constructive total loss after it stranded on a breakwater. The insurer preemptively filed an action for a declaratory judgment that it was not liable on the policy due to a breach of two warranties. The assured in his counterclaim alleged a breach of contract and various unfair claims-settlement practices in violation of Massachusetts insurance law. The district court dismissed the state law counterclaims on the ground that New York—not Massachusetts—law applied, and New York law recognizes no comparable causes of action.141 On appeal, the First Circuit reversed. It concluded that the choice-of-law clause was ambiguous with respect to extra-contractual claims, resolved those ambiguities against the insurer that drafted the document, and held that New York law did not apply to the assured’s statutory See, for example, Wave Cruiser, note 132, at 1353–4 and n.5; Gray Group Investments, note 132, at 371. 134 Note 132. 135 2022 U.S. Dist. LEXIS 91192, at *3, 2022 WL 17987099, at *1. 136 2022 U.S. Dist. LEXIS 91192, at *4, 2022 WL 17987099, at *2. 137 2022 U.S. Dist. LEXIS 91192, at *10, 2022 WL 17987099, at *4. 138 Because the court ruled that New York law applied, it did not discuss whether the assured would have prevailed under Florida law. Perhaps the facts of the case would ultimately have justified the same result. At the very least, however, it seems unlikely that the assured would have prevailed on summary judgment if Florida law had governed. 139 See, for example, Fla. Stat. § 627.409(2) (“A breach or violation by the insured of a warranty […] of a wet marine […] insurance policy […] does not void the policy […] unless such breach or violation increased the hazard by any means within the control of the insured.”). 140 Note 132. 141 Great Lakes Insurance SE v Andersson, 544 F. Supp. 3d 196, 200 (D. Mass. 2021). 133

Choice-of-law issues in marine insurance cases in the United States  241 claims. The court of appeals found it significant that the first branch of the clause (calling for the application of established federal law) applied to “any dispute arising hereunder” while the second branch of the clause (calling for the application of New York law) applied to “this insuring agreement.”142 Although the assured’s allegations of unfair claims-settlement practices admittedly arose under the insurance contract, resolving those allegations did not require interpretation of the insuring agreement. Because only contractual claims were explicitly subject to New York law, it was reasonable to construe the clause to permit Massachusetts law to apply to extra-contractual claims.143 Even if the insurer’s construction of the clause was also reasonable, the ambiguity had to be resolved in the insured’s favor.144 An insurer could easily address the ambiguity identified by the Andersson court. The boilerplate language in the standard form could be revised to specify that, in the absence of established federal law, New York law governs “any dispute arising” under it. The second recent case may—depending on what happens in the next phase of the litigation—present more fundamental problems. In Great Lakes Insurance SE v Raiders Retreat Realty Co.,145 the assured’s covered yacht ran aground. The insurer preemptively filed an action for a declaratory judgment that it was not liable on the policy because “the yacht’s fire-extinguishing equipment had not been timely recertified or inspected” even though “the vessel’s damage was not caused by fire.”146 The assured in its counterclaim alleged a breach of contract and various violations of Pennsylvania law. The district court dismissed the state law counterclaims on the ground that New York— not Pennsylvania—law applied. On appeal, the Third Circuit vacated and remanded the case so that the district court could consider whether Pennsylvania has a strong public policy in favor of its relevant law that would preclude the enforcement of the New York choice-of-law clause. The insurer petitioned for certiorari, asking the Supreme Court to address two issues. The first question presented—broadly asking the Court to clarify the standard under maritime law for enforcing a choice-of-law clause—would have given the Court the opportunity to reconsider Wilburn Boat if it had wished to do so.147 The Court declined to hear that question. The second asked more narrowly whether “a choice of law clause in a maritime contract [is] unenforceable if enforcement is contrary to the ‘strong public policy’ of the state whose law is displaced.”148 The Court granted certiorari on that question,149 and at the time of writing it looked likely that the case would be argued in the autumn of 2023 (with a decision likely by June 2024). It is difficult to predict with any confidence what the Supreme Court will decide in Raiders Retreat. Like most US courts, the Supreme Court has rarely hesitated to enforce a choice-of-law or forum-selection clause. But the Court has also declared that if those clauses operate “as See text at note 132. Andersson, 66 F.4th at 26–7. 144 Ibid at 27–8. 145 Note 132. 146 Raiders Retreat, note 132, at 227. 147 See Petition for Certiorari at i, Raiders Retreat, No. 22-500 (filed Nov. 23, 2022). The insurer did its best to frame the case as one of general maritime law rather than marine insurance. 148 Ibid. 149 Great Lakes Insurance SE v Raiders Retreat Realty Co., LLC, 143 S. Ct. 999 (2023) (granting cert to 47 F.4th 225, 228 (3d Cir. 2022)). 142 143

242  Research handbook on marine insurance law a prospective waiver of a party’s right to pursue statutory remedies,” it “would have little hesitation in condemning the agreement as against public policy.”150 The insurer argues that in a maritime case, that principle applies only to federal statutory remedies.151 It is easy to think that a deeply conservative Court is likely to be more sympathetic to the insurer, but some of the most conservative members of the Court are also highly protective of “States’ rights.”152 And the Court’s apparent unwillingness to reconsider Wilburn Boat suggests that it will accept that Pennsylvania law—not federal law—would apply in the absence of the choice-of-law clause.153 It is also plausible that the Court might look to the Restatement (Second) of Conflict of Laws § 187(2), which would protect the interests of “the state whose law would otherwise apply” when “that state has a materially greater interest […] in the determination of the particular issue” than does the state named in the choice-of-law clause.154 If the Supreme Court rules for the insurer in Raiders Retreat, it will confirm the power of a choice-of-law clause in a marine insurance contract—and likely encourage more insurers to include such clauses in their contracts. But even if the Supreme Court affirms the Third Circuit, it will be a relatively narrow decision. Choice-of-law clauses will almost certainly be enforceable in the majority of cases that do not implicate public policy concerns. Even in Raiders Retreat, New York will govern the interpretation of the contract (and thus the underlying coverage dispute).

7. CONCLUSION “Hard cases,” the maxim tells us, “make bad law.”155 Wilburn Boat was apparently a hard case for several of the justices who decided it.156 Almost everyone who has since discussed the case agrees that it made bad law. It has created such confusion for insurers, assureds, and lower courts that repeated efforts have been made over the decades to solve the resulting problems.157 Despite the failure of the many efforts to achieve a global solution to the Wilburn Boat problems, recent litigation suggests that the parties to an individual marine insurance contract have the power to solve the problem for themselves by the simple expedient of including

150 Mitsubishi Motors Corp. v Soler Chrysler-Plymouth, Inc., 473 U.S. 614, 637 n.19 (1985). See also, for example, Vimar Seguros y Reaseguros, S.A. v M/V Sky Reefer, 515 U.S. 528, 540, 1995 AMC 1817 (1995). 151 See Brief for Petitioner at 18–26, Raiders Retreat, No. 22-500 (filed May 26, 2023). 152 Cf. notes 106–8 and accompanying text. 153 Indeed, it is implicit in the insurer’s phrasing of the question presented that Pennsylvania law would apply in the absence of the choice-of-law clause. The question explicitly refers to “the ‘strong public policy’ of the state whose law is displaced.” 154 The Supreme Court has frequently relied on the Restatement of Conflict of Laws; see, for example, Sosa v Alvarez-Machain, 542 U.S. 692, 705–6, 709–10 (2004); Baker v General Motors Corp., 522 U.S. 222, 235, 238–9 (1998); Mississippi Band of Choctaw Indians v Holyfield, 490 U.S. 30, 46 n.21, 48 (1989), including in maritime cases; see American Dredging Co. v Miller, 510 U.S. 443, 454, 1994 AMC 913, 921 (1994); M/S Bremen v Zapata Off-Shore Co., 407 U.S. 1, 11 & n.13, 1972 AMC 1407, 1415 and n.13 (1972); Lauritzen v Larsen, 345 U.S. 571, 586 n.19, 1953 AMC 1210, 1221 n.19 (1953). Cf. note 125. 155 Cf. Winterbottom v Wright, 10 M. & W. 109, 116, 152 Eng. Rep. 402, 406 (Ex. 1842) (Rolfe B). 156 See generally, for example, Goldstein, note 29, at 410–17. 157 See notes 97–130 and accompanying text.

Choice-of-law issues in marine insurance cases in the United States  243 a choice-of-law clause in their agreement. That solution might not work in every case. If a clause offends important public policies, the courts will be less likely to defer to the parties’ choice.158 But a choice-of-law clause is likely to succeed most of the time. The big mystery now may be why US courts still see so many marine insurance contracts that do not include a choice-of-law clause.

See notes 145–54 and accompanying text.

158

PART V  

12. Insurance requirements in Incoterms® 2020 Margarida Lima Rego

1. INTRODUCTION Under Incoterms® 2010 Rules, as in prior editions of these Rules, cost, insurance and freight (CIF) and CIP clauses set forth the seller’s duty to insure the goods under Institute Cargo Clauses (C) of the Institute of London Underwriters. Under Incoterms® 2020 Rules, the International Chamber of Commerce increased the default level of insurance coverage required on the goods under CIP, while maintaining the previous default level of insurance coverage required on the goods under CIF. The former, but not the latter, now points towards the use of Institute Cargo Clauses (A). This chapter aims to identify and assess the legal and commercial implications of this change. After this brief Introduction, the first main section will explain the concept of the passing of risk in a sale of goods. The following section will set out the legal framework of the insurance requirements in these trade standards, with a special focus on the default rules on the passing of risk set forth in the United Nations Convention on Contracts for the International Sale of Goods, also known as the Vienna Convention. After this, the chapter will provide an overview of the history and legal nature of the Incoterms® Rules of the International Chamber of Commerce. It will then move on to compare the insurance requirements in Incoterms® 2010 Rules with those in Incoterms® 2020 Rules. This is where the change in insurance requirements is set out in detail. The legal and commercial implications of this change are then discussed, before the final section expresses the chapter’s conclusions.

2.

THE LEGAL CONCEPT OF THE PASSING OF RISK

In an international sales contract, as, more widely, in other contracts that involve the transfer of a right in rem over things, it is important to determine the exact moment of occurrence of a phenomenon called the ‘passing of risk’ from one party to the other party (hereinafter referred to simply as seller and buyer). Identifying the exact moment at which the risk passes from the seller to the buyer lies at the heart of contracts of sale, as with many other contracts where rights in rem over things are transferred. ‘Risk’, which is a polysemic term, in this context assumes a broad yet rigorous meaning. Its core elements include the notions of (i) possibility and (ii) negativity.1 In other words, its use conveys the ideas (i) that the mere possibility of an occurrence is at stake – this occurrence not being perceived either as an impossibility or as a certainty, but comprising all degrees of probability between 0 and 1; and (ii) that this occurrence is negatively valued. The risk at stake, in this context, is that of loss of or damage to the goods.



1

See ML Rego, Contrato de Seguro e Terceiros. Estudo de Direito Civil (Coimbra 2010) 67–79.

245

246  Research handbook on marine insurance law In a contract of sale, typically the seller is bound by an obligation to deliver the goods to the buyer. If the goods suffer a misfortune, the corresponding loss or damage must be borne by someone. At a crucial moment in time, the risk of occurrence of such loss or damage to the goods shifts from the seller to the buyer. That is the moment of the passing of risk. It is also the moment at which the seller is deemed to have legally completed its contractual obligation to deliver the goods to the buyer.2 It is a matter of interpretation of the sale contract which determines if the risk has passed from the seller to the buyer, namely, if the seller has performed its obligation to deliver under the contract. The risk of loss of or damage to the goods is relevant in any sales contract, but it is particularly important in the vast number of contracts involving the transfer of rights over goods that have to be carried from point A to point B. Should something go wrong before the passing of risk, this is the seller’s problem. Should something go wrong after the passing of risk, the risk is borne by the buyer. From that moment onwards, any loss of or damage to the goods, except for wrongdoing on the part of the seller or their agents or subcontractors, shall entail the seller’s exoneration. The seller is thus exonerated from its contractual obligations, particularly of the obligation to deliver the goods. This is called performance risk: if the risk has passed, such obligations are treated, to all intents and purposes, as having been performed. In other words, the seller has no duty to deliver new goods, to replace those that have disappeared or deteriorated, or otherwise to remedy the consequences of their loss or damage, while retaining in full its right to consideration: namely, the sale price. The latter is called the price risk: if the risk has passed, the seller preserves its right to be paid the sale price. The passing of risk should be distinguished from the transfer of ownership or other rights in rem over the goods. The rules governing them are distinct, so the two moments do not have to coincide in time – although historically they did, in accordance with the maxim res perit domino suo (the destruction of a thing is the loss of its owner).3

3.

THE PASSING OF RISK ACCORDING TO THE VIENNA CONVENTION

The passing of risk in contracts for the international sale of goods, that is to say, in contracts of sale where the parties have their places of business in different countries – is regulated in Chapter 4 of Part III, Articles 66 to 70 of the United Nations Convention on Contracts for the International Sale of Goods, also known as the Vienna Convention (commonly abbreviated ‘CISG’). These rules have as their main antecedent Articles 96 and following of the Uniform Law on the International Sale of Goods, approved by the Hague Convention of 1964. The Vienna Convention is the closest humanity has got to enacting a truly global trade law. It is often pinpointed as ‘one of the success stories in the field of the international unification of private law’.4 According to the United Nations, a total of 95 States have adopted the

See M Davies and DV Snyder, International Transactions in Goods: Global Sales in Comparative Context (OUP 2014) 245. 3 Ibid 246–8. 4 P Huber, ‘Comparative Sales Law’ in M Reimann and R Zimmermann (eds), The Oxford Handbook of Comparative Law (2nd edn, OUP 2019) 933–60, 935. 2

Insurance requirements in Incoterms® 2020  247 Convention.5 Its rules are applicable to international sales of goods, when the parties have their places of business in adopting states or when the rules of private international law lead to the application of the law of an adopting state.6, 7 According to Article 66: ‘Loss of or damage to the goods after the risk has passed to the buyer does not discharge him from his obligation to pay the price, unless the loss or damage is due to an act or omission of the seller.’ This provision only deals explicitly with the passing of the price risk,8 or the risk of counter-performance (in German, Gegenleistungsgefahr), as opposed to the performance risk (Leistungsgefahr).9 The former is the seller’s risk of losing the purchase price, or the buyer’s risk of having to pay it without receiving the goods. Not explicitly regulated in this provision is the latter, which concerns a party’s obligation ‘to perform in spite of unexpected impediments’ (including the occurrence of loss of or damage to the goods).10 Although this wording is not too clear on the exact nature and types of occurrences that are included in the concepts of loss or damage to goods, the terms ‘loss’ and ‘damage’ were preferred over alternative terms, such as ‘deterioration’, so as not to convey the idea that the rules on the passing of risk dealt only with phenomena of natural, not human, origin.11 Case law, legal commentaries and other scholarly writings have since made it clear that the types of events covered by this language correspond to all kinds of accidental physical damage to the goods. This includes disappearance, fortuitous destruction, spoilage, evaporation, improper stowage or careless handling, theft, vandalism, discharge of cargo, the goods being sent off to the wrong destination and the goods being mixed up or blended in with other goods, as well as any kind of degeneration of the physical properties of the goods, without any relevant fault of the parties. The common feature in all these cases is the accidental nature of the occurrence causing the loss or damage.12 It does not include devaluation of the goods, that is, loss of market value not connected with a degeneration of the physical properties of the goods.13

See https://​iicl​.law​.pace​.edu/​cisg/​page/​cisg​-table​-contracting​-states (accessed 31 March 2023). See Article 1 of CISG. See P Huber, ‘Comparative Sales Law’ in M Reimann and R Zimmermann (eds), The Oxford Handbook of Comparative Law (2nd edn, OUP 2019) 933–60, 936. 7 ‘In analyzing the Convention, one must bear in mind that its provisions play a supporting role, supplying only those terms to a contract that the parties have failed to include’: SS Grewal, ‘Risk of Loss in Goods Sold During Transit: A Comparative Study of the U.N. Convention on Contracts for the International Sale of Goods, the U.C.C., and the British Sale of Goods Act’ (1991) 14 Loy LA Int’l & Comp LJ 93, 94. 8 See PM Roth, ‘The Passing of Risk’ (1979) 27 Am J Comp L 291, 291–2. 9 See N Jansen and R Zimmermann, ‘Sale of Goods’ in Commentaries on European Contract Law (OUP 2018, online edn, https://​doi​.org/​10​.1093/​oso/​9780198790693​.003​.0019, accessed 31 March 2023), 2022–4. 10 Ibid 2023. 11 See B Nicholas’s commentary to Article 66 in Bianc and Bonell (eds), Commentary on the International Sales Law: The 1980 Vienna Sales Convention (Giuffrè 1987) 484. 12 Z Valioti, ‘Passing of Risk in International Sale Contracts: A Comparative Examination of the Rules on Risk under the United Nations Convention on Contracts for the International Sale of Goods (Vienna 1980) and Incoterms 2000’ (2004) Nord J Com L 1, 7–8. 13 See F Mohs, ‘Passing of Risk, Introduction to Articles 66–70’ in I Schwenzer (ed), Schlechtriem & Schwenzer: Commentary on the UN Convention on the International Sale of Goods (4th edn, OUP 2016) 950–7, para 3. See also A Romein, The Passing of Risk: A Comparison between the Passing of Risk under the CISG and German Law (Pace Law School Institute of International Commercial Law, 5 6

248  Research handbook on marine insurance law The issue of transfer of ownership or other rights in rem over things is not regulated in the Vienna Convention.14 Its treatment of the passing of risk is based on the assumption that the two matters are independent from one another, it being possible and even desirable to deal with each of these issues separately, independently from the rules applicable to the other. In any event the rules governing the transfer of ownership and other rights in rem over things vary greatly from jurisdiction to jurisdiction, so it would be very unwise to have the passing of risk depend on such a geographically volatile legal institution. A more stable solution was needed, the drafters being acutely aware of the fact that many of the people who would be called upon to apply these rules in practice on a daily basis would not be legally trained. There are thus two basic issues which confront the draftsmen of uniform rules on the passing of risk. The first is one of substance: What practical policies should the provisions concerning risk seek to achieve; which party, in a given case, should bear the risk? Secondly, there is a question of form. How can the articles be drafted so as to ensure that these policies are realised effectively, coping with the immense range of factual situations which may arise, without resorting to undue technicality?15

As a starting point, the physical acts (or ‘overt commercial events’16) of handing over the goods, or of placing the goods at the disposal of the buyer took the stage. The abstract, legal concept of delivery was relegated to a supporting or subordinate role, being defined by reference to the former.17 The abstract, legal concept of transfer of property was not given any significant role to play. A conscious choice was made to select a practical issue as the decisive criterion for determining the default rules on the passing of risk, in lieu of an abstraction – a particularly geographically volatile legal abstraction.18 That issue was that of physical control over the goods: at any given time, the party who is in control of the goods bears the risk of loss of or damage to the goods. This rule is at its simplest in sales that do not involve a carriage of goods from point A to point B. In this case, risk passes when the buyer takes over the goods or, should the buyer fail to do so by breaching its duty to take delivery, when the buyer should have taken over the goods.19 In the great majority of sales involving carriage of goods, there is a period of time when often only a third party – the carrier – has physical control over the goods. In these cases, the passing of risk will depend on whether the seller has undertaken to hand over the goods to

July 1999); J Erauw, ‘CISG Articles 66–70: The Risk of Loss and Passing It’ (2005–6) 25 J L & Com 203, 204–8. 14 See Article 4(b) of CISG. 15 PM Roth, ‘The Passing of Risk’ (1979) 27 Am J Comp L 291, 291–3. 16 Ibid 296. 17 See Articles 31 ff. of CISG. See B Nicholas’s commentary to Article 67 in Bianca and Bonell (eds), Commentary on the International Sales Law: The 1980 Vienna Sales Convention (Giuffrè 1987) 487–9. See also the discussion in PM Roth, ‘The Passing of Risk’ (1979) 27 Am J Comp L 291. 18 See J Erauw, ‘CISG Articles 66–70: The Risk of Loss and Passing It’ (2005–6) 25 J L & Com 203, 214. 19 Article 69(1) and (2) of CISG. CISG attaches importance to the individualization of the goods, determining that, when the contract concerns goods that have not yet been individualised, that is, when the seller’s obligations are generic, the passing of risk cannot take place without their clear identification for the purposes of the contract (art. 69(3) of CISG).

Insurance requirements in Incoterms® 2020  249 a carrier at a particular place – handing over here referring to the physical act, rather than the legal concept of delivery. If so, the risk passes when the goods are handed over to a carrier at that particular place. If not, risk passes when the goods are handed over to the first carrier.20 The rules on the passing of risk concern the risk allocation as between both contracting parties in the sale of goods. They do not touch upon the buyer’s or the seller’s contractual risk allocation vis-à-vis the third party carrier. Consequently, the general default rule, when neither party has physical control over the goods, is that the buyer bears the transit risk. Again, this rule is based on pragmatic considerations: it is usually the buyer who checks the goods upon receiving them from the carrier, being in a better position to assess the loss or damage and to make a claim against the carrier and/or the insurer.21 Lastly, the passing of risk occurs at the time of conclusion of the contract when the goods are sold in transit.22 A similar solution to that set forth in the Vienna Convention can also be found, more recently, in Article 20 of the Consumer Rights Directive, according to which ‘[i]n contracts where the trader dispatches the goods to the consumer, the risk of loss of or damage to the goods shall pass to the consumer when he or a third party indicated by the consumer and other than the carrier has acquired the physical possession of the goods’.23 The Vienna Convention’s default rules on the passing of risk seldom apply in practice, the same happening in most of the world’s national default rules on the passing of risk, mostly due to the popularity of the International Chamber of Commerce’s Incoterms® Rules.24 This is so because default rules are meant to apply only if and to the extent that contracting parties abstain from setting them aside by stipulating their own rules. As the following section will highlight, Incoterms® Rules are precisely aimed at making it easy for contracting parties to stipulate on a number of matters of relevance to their contracts, including the passing of risk. Hence, the greater their popularity, the scarcer the instances where default rules are actually called upon to determine the outcome of a case.

4.

INCOTERMS® RULES OF THE INTERNATIONAL CHAMBER OF COMMERCE

Incoterms®, an acronym which is short for ‘international commercial terms’, are sets of standard commercial terms, or pre-defined contractual clauses, that have been codified, developed

Article 67(1) of CISG. Once again, the transfer presupposes the individualisation of the goods by means of their clear identification for the purposes of the contract (art. 67(2) of CISG). 21 See SS Grewal, ‘Risk of Loss in Goods Sold During Transit: A Comparative Study of the U.N. Convention on Contracts for the International Sale of Goods, the U.C.C., and the British Sale of Goods Act’ (1991) 14 Loy LA Int’l & Comp LJ 93, 97. 22 Article 68 of CISG. 23 Directive 2011/83/EU of the European Parliament and of the Council of 25 October 2011 on consumer rights. 24 See F Mohs, ‘Passing of Risk, Introduction to Articles 66–70’ in I Schwenzer (ed), Schlechtriem & Schwenzer: Commentary on the UN Convention on the International Sale of Goods (4th edn, OUP 2016) 950–7, para 23. 20

250  Research handbook on marine insurance law and published by experts and practitioners under the guise of the International Chamber of Commerce since 1936.25 Incoterms® Rules do not technically have the force of law anywhere in the world, nor can they be chosen by contracting parties as the law governing their contract. They are purely contractual, gaining their binding force from their incorporation by the parties into their contract.26 Incoterms® Rules are not contract templates to be used either in contracts of sale of more broadly for transfer of goods. They do not restrict the party autonomy on the negotiation of the contract terms either.27 They are terms that can be incorporated by reference amid the other terms included in the contract. They will displace the default rules of the law governing the contract, but their incorporation will be subject to the principles of freedom of contract and the mandatory provisions of the law governing the contract.28 Incoterms® Rules are aimed at simplifying their individual negotiation by allowing certain core matters to be regulated per relationem, that is, by reference to these pre-defined, ready-to-use contract terms. For ease of reference, each Incoterms® rule is known by a series of three letters, so the parties need only insert those letters in their contract in order for the entire set of clauses it refers to, to be deemed as an integral part of that contract. The International Chamber of Commerce’s successive editions of the Incoterms® Rules include each rule’s wording, or what the authors put forward as the wording of the rule, strictly speaking. The rules are preceded by an authored introduction – the only part of the text that is presented as having been written by an individual author – and are supplemented by what may be described as instructions, guidelines or explanatory notes for users.29 Although these supplements are aimed at facilitating the handling of the rules by users that may not be legally trained, the whole volume is of relevance to legal professionals as a tool for the interpretation of those sets of three letters, whenever they are inserted in a contract that is to be interpreted and applied, for instance, by a court of any given jurisdiction. The core matters that are regulated in the Incoterms® Rules are those arising from contracting parties’ need to make arrangements for the carriage of goods from point A to point B. The Incoterms® Rules deal with the very common circumstance, in international trade, of there being a geographical distance between buyer and seller. Such core matters are the tasks, risks and costs associated with the transportation and physical handing over of the goods. Incoterms® Rules describe who undertakes to do what (the tasks), where and when the seller is deemed to have legally delivered the goods to the buyer (the risks), that is to say, where and when risk of loss of or damage to the goods passes from

See https://​iccwbo​.org/​resources​-for​-business/​incoterms​-rules/​ (accessed 31 March 2023). See also Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), foreword by JWH Denton. 26 See O Toth, The Lex Mercatoria in Theory and Practice (OUP 2017) 34. 27 See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 6 of the Introduction. 28 Ibid 34–5; and F. Mohs, ‘Passing of Risk, Introduction to Articles 66–70’ in I Schwenzer (ed), Schlechtriem & Schwenzer: Commentary on the UN Convention on the International Sale of Goods (4th edn, OUP 2016) 950–7, para 24. 29 These notes were entitled ‘Guidance Notes’ in Incoterms® 2020 and are now called ‘Explanatory Notes For Users’ in Incoterms® 2020. See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 77 of the Introduction. 25

Insurance requirements in Incoterms® 2020  251 the seller to the buyer, and finally who pays for what (the costs).30 These matters are regulated in two symmetrical series of ten articles, respectively identified as A1 to A10 and B1 to B10. The seller’s undertakings are regulated in A1 to A10, the buyer’s undertakings being set forth under B1 to B10.31 Because they were written with the purpose of identifying each party’s undertakings, what each of them commits to do or to endure, little attention is directly paid to each party’s contractual rights, which must hence be inferred from the former.32 Incoterms® do not regulate the contract for the carriage of goods, nor do they lay down the third party carrier’s undertakings vis-à-vis the seller and/or the buyer. Absent from the Incoterms® Rules are other at least equally important matters, such as what goods are being sold for which contract price. Other aspects which are entirely left out are the determination of the law governing the contract and the regulation of the transfer of title to the goods.33 Although the first known edition of the International Chamber of Commerce Incoterms® is dated to 1936, Incoterms® were originally not a product of the International Chamber of Commerce, some of them having been in use since much earlier times.34 Competing, albeit less influential definitions of some such terms can be found, for instance, in the American Revised Foreign Trade Definitions (1941) – which could be said to have been largely pushed into obsolescence by the success of Incoterms® Rules35 – or in the US Uniform Commercial Code.36 However, the acronym ‘Incoterms’ appears to have been coined by the International Chamber of Commerce (ICC), the latter being the owner of the registered trademark Incoterms®.37 Since their publication in 1936, the Incoterms® Rules have been amended in 1953, 1967, 1976 and 1980, and since then once every decade, in 1990, 2000, 2010 and 2020.38 Some authors have warned that Incoterms® Rules as such should rigorously not be characterised as codified trade usage within the meaning of Article 9(2) CISG.39 Incoterms® Rules,

Ibid, para 5 of the Introduction, by C Debattista. Ibid, para 5 of the Introduction, by C Debattista. 32 CF Almeida, ‘Incoterms: estrutura e qualificações’ (2018) Special Ed. 6 Themis 57, 58. 33 See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 7 of the Introduction, by C Debattista. 34 See, for instance, EA Craighill, Jr, ‘Sales of Goods on C.I.F. Terms’ (1920) 6 Va L Rev 229. The author compares contracts of sale CIF and FOB, referring to the former as ‘long and widely used in international commerce’ (229). See also G Saumier, ‘Trade Usages in the Convention on Contracts for the International Sale of Goods’ in F Gélinas (ed), Trade Usages and Implied Terms in the Age of Arbitration (OUP 2016) 125–44, 140. 35 See, for instance, www​.globalnegotiator​.com/​international​-trade/​dictionary/​american​-foreign​-trade​ -definitions/​(accessed 31 March 2023). 36 FOB and FAS are defined in § 2-319 of the Uniform Commercial Code. CIF and C&F are defined in § 2-320 of the same Code. See JA Spanogle, ‘Incoterms and UCC Article 2: Conflicts and Confusions’ (1997) 31 Int’l Law 111. 37 Incoterms® and the Incoterms® 2020 logo are trade marks of ICC. Use of these trade marks does not imply association with, approval of or sponsorship by ICC unless specifically stated above. The Incoterms® Rules are protected by copyright owned by ICC. Further information on the Incoterms® Rules may be obtained from the ICC website https://​iccwbo​.org/​resources​-for​-business/​incoterms​-rules/​ incoterms​-rules​-copyright/​(accessed 31 March 2023). 38 See https://​iccwbo​.org/​resources​-for​-business/​incoterms​-rules/​incoterms​-rules​-history/​ (accessed 31 March 2023). 39 Incoterms® Rules ‘are not in and of themselves a trade usage within the meaning of Article 9(2), as otherwise one would have to clarify, which clause is a usage for which type of contract’. F Mohs, 30 31

252  Research handbook on marine insurance law and the experts who contributed to each edition thereof, aim to keep up with the developments of international trade usage when carrying out their revisions, which purport to be a reflection of ‘business-to-business practice’.40 Be it as it may, the practical significance of Incoterms® Rules is such that, for instance, US courts have repeatedly held that trade terms used in international sales contracts can be interpreted in accordance with Incoterms® Rules because the latter are incorporated into the Vienna Convention through its Article 9(2).41 This may also be the case even when contracts do not explicitly refer to Incoterms® Rules, by application of Article 9(1) CISG.42 In sum, courts in different countries appear to treat the Incoterms® Rules in fact like uniform private law.43 Currently, 11 rules are in use, according to Incoterms® 2020. Four of these rules are intended for use only in non-containerised sea or inland waterway transport, the other seven being intended for use in any mode of transport, including multimodal transport.44 In some cases, this includes using one of the parties’ mode of transport, in which case no third party carrier will be involved in the transaction.45 Whenever one of the maritime rules is used, delivery occurs; that is, risk of loss or damage to the goods passes when the goods are placed on board or alongside a vessel.46 This is why they are not suitable for the transfer of containerised goods, as those goods are handed over to a carrier at a container terminal, rather than being placed on board or alongside a vessel. In any case, risk allocation varies greatly, from EXW, which conveys the least amount of tasks, risks and costs on the seller and the most on the buyer, to DDP, which places the most tasks, risks and costs on the seller and the least on the buyer. Table 12.1 provides a list of the 11 Incoterms® 2020 rules, from the most onerous to the least onerous rules from the perspective of the buyer.

‘Passing of Risk, Introduction to Articles 66–70’ in I Schwenzer (ed), Schlechtriem & Schwenzer: Commentary on the UN Convention on the International Sale of Goods (4th edn, OUP 2016) 950–7, para 23. 40 See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 4 of the Introduction, by C Debattista. 41 See, for instance, US District Court, Southern District of Texas, United States, 7 February 2006, available at www​.cisg​.law​.pace​.edu (accessed 31 March 2023): ‘Because Incoterms is the dominant source of definitions for the commercial delivery terms used by parties to international sales contracts, it is incorporated into the CISG through article 9(2).’ 42 See G Saumier, ‘Trade Usages in the Convention on Contracts for the International Sale of Goods’ in F Gélinas (ed), Trade Usages and Implied Terms in the Age of Arbitration (OUP 2016) 125–44, 141. 43 See G-P Calliess and IS Jarass, ‘Private Uniform Law and Global Legal Pluralism’ in PS Berman (ed), The Oxford Handbook of Global Legal Pluralism (OUP 2020) 746–67, 764–7. The second author carried out an in-depth analysis of English and German case law involving Incoterms® Rules from 1990 to 2015. ‘[H]er analysis revealed a high degree of effectiveness of the Incoterms despite their legal status as contract language. First, the Incoterms were not declared invalid, and second, the Incoterms were even applied by the courts without a contractual reference to the Incoterms by the parties and moreover, without providing any legal theoretical justification for their application, for example, by referring to trade usages or customary law’ (764). 44 See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 42 of the Introduction, by C Debattista. This distinction was introduced in Incoterms® 2010. 45 Ibid paras 30 and 71–3 of the Introduction, by C Debattista. This clarification was absent in prior editions of the Incoterms® Rules. 46 Ibid para 43 of the Introduction, by C Debattista.

Insurance requirements in Incoterms® 2020  253 Table 12.1

Incoterms® 2020. ICC rules for the use of domestic and international trade terms

EXW

Ex Works (named place of delivery)

any mode of transport

FCA

Free Carrier (named place of delivery)

any mode of transport

FAS

Free Alongside Ship (named port of shipment)

sea/ inland waterway

FOB

Free On Board (named port of shipment)

sea/ inland waterway

CFR

Cost and Freight (named port of destination)

sea/ inland waterway

CIF

Cost, Insurance & Freight (named port of destination)

sea/ inland waterway

CPT

Carriage Paid To (named place of destination)

any mode of transport

CIP

Carriage and Insurance Paid (named place of destination)

any mode of transport

DPU

Delivered at Place Unloaded (named place of destination)

any mode of transport

DAP

Delivered At Place (named place of destination)

any mode of transport

DDP

Delivered Duty Paid (named place of destination)

any mode of transport

Source:

© 2019 International Chamber of Commerce (ICC).

The rule to be used when one wishes to bind the seller to a mere obligation to make the goods available for collection by or on behalf of the buyer at the seller’s premises (in German, Holschuld) is Ex Works, or EXW (followed by the location of the seller’s factory). Because this rule allocates all the tasks, risks and costs involved in the carriage of the goods to the buyer, the International Chamber of Commerce does not recommend its use except in purely domestic sales. Exactly the same can be said of DDP, which, at the other extreme, allocates all the tasks, risks and costs involved in the carriage of the goods to the seller. When import/ export of goods is at stake, some degree of cooperation between the parties is recommended for the transaction to be successful.47 F rules are those where the tasks, risks and costs associated with the carriage of goods are to be borne by the buyer (F meaning free). The most basic of these rules is FCA (Free CArrier), which has a broad spectrum and is relatively balanced and suitable for any mode of transport, including multimodal transport, which makes it the International Chamber of Commerce’s go-to rule when in doubt.48 In this rule, if the place designated for delivery is the commercial establishment of the seller, the loading of the goods onto the means of transport provided is still carried out at the seller’s risk, the risk passing to the buyer only when this operation is concluded. In FOB (Free On Board), which is one of the rules intended for use only in non-containerised sea or inland waterway transport of goods, risk passes when the goods are loaded on the ship. If there is a problem in stowage and the goods fall into the water or crash ashore, or even into the ship’s hold, before being safely deposited on the ship, such risk will fall on the seller. If the goods are lost or damaged after the loading operation has been completed, the buyer will bear the loss. Should the ship sink after only part of the loading has been completed, the risk remains with the seller. If the goods are transported in containers, this rule is not the most appropriate, which is one of the reasons why FOB, once a very popular rule, has been dethroned by the more flexible FCA. Ibid para 22 of the Introduction, by C Debattista. The ICC Model International Sale Contract (Manufactured Goods) in its 2020 edition, as in the prior 2013 edition, include a list of recommended Incoterms® Rules for the parties to choose from. In the absence of choice, the default rule shall apply, that being FCA (seller’s premises). See para 5 of the Introduction. 47 48

254  Research handbook on marine insurance law In this context, the most relevant evolution was observed in the transition from Incoterms® 2000 to Incoterms® 2010. Until then, the risk would have passed with the transposition of the cargo of an imaginary vertical line drawn from the ship’s rail. As a result, if, during the stowage, the cargo fell into the water, the risk would remain on the seller. However, if it fell into the hold of the ship, the risk would be borne by the buyer. This rule was amended in 2010. D rules are those where the seller commits to bringing the goods to the buyer’s premises (in German, Bringschuld) (D for delivery). In this case, the most suitable rule may be, for instance, DAP or DPU. The latter was formerly DAT, in editions prior to 2020. The letter ‘T’ literally referred to delivery at a given terminal, but, although the word was broadly defined and broadly interpreted by experts, its literal reliance on the word ‘terminal’ was thought to be potentially misleading to the rules’ users; hence the new wording, DPU, by replacing T with P, now makes it clear to everyone that delivery may occur at any given place.49 As regards the passing of risk, the difference is that in the former the risk passes to the buyer when the goods are placed at its disposal in the mode of transport of arrival, ready to be unloaded, whereas in the latter the risk only passes to the buyer on completion of the unloading operation. Risk does not always pass at the same time as the costs associated with transport are transferred from the seller to the buyer. C rules disconnect the passing of the risk from the bearing of the costs (C for costs). As a result, they are arguably somewhat counterintuitive and therefore more difficult for non-expert users to handle and may give rise to unexpected practical consequences at times. These rules determine that the seller undertakes to arrange and also pay for the transport of the goods sold. In some cases the seller is also expected to insure the goods sold, although it does not assume the corresponding risk of loss of or damage to the goods during transport. C rules are hence the only clauses that name only the place of destination, rather than the place where delivery is to take place. This is because such rules bind the seller to an obligation to send (in German, Schickschuld) – that is to organise, contract and pay for the carriage of goods, in what are often called ‘shipment sales’ – rather than to an obligation to bring the goods to the buyer.50 This is the case, for example, of CIF. CIF is one of the Incoterms® Rules intended for use only in non-containerised sea or inland waterway transport of goods. In this wording, the risk of loss of or damage to the goods passes to the buyer as soon as the goods are loaded on board the vessel, this being legally the moment of delivery, that is, of the seller fulfilling its obligation. The seller, however, still bears the costs of the carriage of goods to the port of destination, including transport insurance and customs clearance costs. If something happens to the goods after the passing of the risk from the seller to the buyer, the latter’s obligation to pay the contract price is not affected. The buyer will still have to pay the contract price, but, by virtue of the insurance contract, it is expected that the buyer will acquire a right to claim compensation as against the insurer. This disconnection between the passing of the risk and the bearing of the costs will be further explored in the following sections, as it has an impact on the insurance requirements under Incoterms® Rules.

49 50

Ibid para 74–5 of the Introduction, by C Debattista. Ibid para 12, 24 and 26 of the Introduction, by C Debattista.

Insurance requirements in Incoterms® 2020  255

5.

INSURANCE REQUIREMENTS: FROM INCOTERMS® 2010 TO INCOTERMS® 2020

While insurance is always available both to the seller and the buyer, regardless of their chosen Incoterms® Rules, only two out of the 11 Incoterms® Rules set forth a contractual duty to procure insurance coverage: CIP (carriage and insurance paid to destination) and CIF (cost, insurance and freight). This is due simply to the fact that in the other nine Incoterms® Rules, the party who decides whether to seek insurance coverage is the same party that has been contractually burdened with the risk of loss of or damage to the goods. Thus, in EXW it is for the buyer to decide whether to insure its own risk of loss of or damage to the goods, because the risk passes at the seller’s premises, before any transport takes place. In F rules, the buyer also has to decide whether to insure its own risk of loss of or damage to the goods, for the same reason. In D rules, conversely, it is for the seller to decide whether to insure its own risk of loss of or damage to the goods, as the seller bears the risk of any problem arising before delivery at the place of destination of the goods. As mentioned above, C rules are the most complex to understand and handle, because they bring forth a disconnection between the passing of the risk and the bearing of the costs. In these so-called shipment rules, the seller undertakes to send the goods to the buyer, which entails organising, contracting and paying for the carriage of goods. Under CPT and CFR, the risk of loss of or damage to the goods is transferred from the seller to the buyer when the goods are handed over to the carrier (CPT) or when they are on board the vessel (CFR). This means that, while delivery takes place early on in the transaction, the seller will have had to organise, contract and pay for the carriage of the goods to its destination. The seller is, however, not bound to procure insurance coverage. In this case, it is therefore for the buyer to decide whether to insure its own risk of loss of or damage to the goods. Finally, under CIP and CIF, the risk of loss of or damage to the goods is transferred from the seller to the buyer when the goods are handed over to the carrier (CIP) or when they are on board the vessel (CIF). Importantly, however, under these rules, in addition to the seller having to organise, contract and pay for the carriage of the goods to its destination, the seller is also bound to obtain insurance cover against the buyer’s risk of loss of or damage to the goods. As per CIP article A5: Unless otherwise agreed or customary in the particular trade, the seller must obtain at its own cost cargo insurance complying with the cover provided by Clauses (A) of the Institute Cargo Clauses (LMA/IUA) or any similar clauses as appropriate to the means of transport used. […] The insurance shall cover, at a minimum, the price provided in the contract plus 10% (i.e. 110%) and shall be in the currency of the contract. The insurance shall cover the goods from the point of delivery set out in A2 to at least the name place of destination […].

CIF’s article A5 is identical, except it refers to ‘Clauses (C) of the Institute Cargo Clauses (LMA/IUA)’. These are the only two rules where the party who undertakes to obtain insurance differs from the party that bears the insured risks. This aspect is crucial to the understanding of why these are also the only two rules that impose a duty to insure, as it would be contrary to principle for a clause to impose on a contracting party a duty to protect itself from any given risk. A duty to insure is only justified, on principle as well as in practice, when the person or

256  Research handbook on marine insurance law entity that is to be protected is different from the person or entity which is to be in charge of protecting the former. A legal duty imposed on a contracting party to obtain insurance cover so as to protect itself against its own risk of loss of or damage to its property would have been regarded by most as too paternalistic and therefore would be deemed to amount to an unjustified restriction of one’s private autonomy. A contractual duty to insure against one’s own risks would be senseless, as the other party would have no interest in enforcing such a duty as against the other. A contractual duty to insure is therefore only relevant when enforceable by the party who bears the insured risks as against the party who does not but is nonetheless contractually bound to protect the former against such risk. Under Incoterms® 2010, as in prior editions of these Rules, the duty to insure required the seller to insure the goods for 110 per cent of the contract value under Institute Cargo Clauses (C) of the Institute of London Underwriters.51 Under Incoterms® 2020, CIP, but not CIF, was changed to require the seller to insure the goods under Institute Cargo Clauses (A) of the Institute of London Underwriters.52 Therefore, since Incoterms® 2020, we now have two quite different sets of insurance clauses to choose from. The following paragraphs will set out the differences between Institute Cargo Clauses (A), (B) and (C). For the most part, the wordings of all three sets of clauses are identical. The main difference between them is to be found in clause 1, which defines the risks covered under each set. (A) is the only one which qualifies as all risks insurance,53 hence providing the broadest scope of cover of the three types of Institute Cargo Clauses: ‘This insurance covers all risks of loss of or damage to the subject-matter insured except as excluded by the provisions of Clauses 4, 5, 6 and 7 below.’ (B) and (C), on the other hand, both qualify as named perils insurance: This insurance covers, except as excluded by the provisions of Clauses 4, 5, 6 and 7 below, 1.1 loss of or damage to the subject-matter insured reasonably attributable to 1.1.1 fire or explosion 1.1.2 vessel or craft being stranded grounded sunk or capsized 1.1.3 overturning or derailment of land conveyance 1.1.4 collision or contact of vessel craft or conveyance with any external object other than water 1.1.5 discharge of cargo at a port of distress 1.1.6 earthquake volcanic eruption or lightning, 1.2 loss of or damage to the subject-matter insured caused by 1.2.1 general average sacrifice 1.2.2 jettison or washing overboard 1.2.3 entry of sea lake or river water into vessel craft hold conveyance container

The current version of such clauses, dated 1 January 2009, is protected by copyright of the Lloyd’s Market Association (LMA) and the International Underwriting Association of London (IUA), who have established a Joint Cargo Committee in charge of periodically updating the Institute Cargo Clauses. See www​.lmalloyds​.com/​LMA/​Underwriting/​Marine/​JCC/​JCC​_Clauses​_Project/​Cargo​_Clauses​.aspx (accessed 31 March 2023). The first set of Institute of London Underwriters clauses dates back to 1888. See H Bennett, ‘Reading Marine Insurance Contracts: Determining the Scope of Cover’ (2019) 27 Asia Pac L Rev 239, 240. A direct predecessor can, however, be found on the ‘SG Form’, which stands for Ship and Goods Form, formally adopted by Lloyd’s as early as 1779. See J Dunt, ‘History and Harmonization’ in J Dunt (ed), International Cargo Insurance (Routledge 2012) 1–12, 4 (paras 1.5–1.6). For a more recent by-product of the Institute Cargo Clauses, see Institute Cargo Clauses (Air), also by the Institute of London Underwriters. See MA Clarke and G Leloudas, Air Cargo Insurance (Routledge 2016) 25. 52 See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 70 of the Introduction, by C Debattista. 53 See J Dunt, ‘English Law and Practice’ in J Dunt (ed), International Cargo Insurance (Routledge 2012) 49–109, 67–77 (paras 3.34–3.55); O Gürses, Marine Insurance Law (2nd edn, Routledge 2017) 189–90. 51

Insurance requirements in Incoterms® 2020  257 or place of storage, 1.3 total loss of any package lost overboard or dropped whilst loading on to, or unloading from, vessel or craft.54

The scope of Institute Cargo Clauses (A) is broader than the others because loss or damage resulting from all perils deemed to constitute a ‘risk’, or any combination thereof, are covered unless explicitly excluded. (B) and (C) policies include only loss or damage resulting from any one of those named perils are covered, (C) providing the narrowest scope of cover because its list of perils is shorter than that contained in (B). The burden of proof favours the insured in (A), when compared to that in (B) or (C), because under (A), as Bennett explains: [a]ll the insured has to prove in order to recover on a claim is that, on a balance of probabilities, the cause of the loss or damage was such a ‘risk’ or a combination of such risks, but there is no need to prove precisely which risk or risks impacted on the insured goods. As a matter of contractual construction, the precise cause of the loss or damage is irrelevant since the insurer accepts liability for loss or damage caused by any and all risks […] Under a named perils policy, in contrast, there is no such general acceptance of liability. The insurer’s acceptance of liability is confined to loss or damage caused by one or more perils listed in the insuring clause. It follows that the insured must prove, on a balance of probabilities, that the loss or damage sustained was caused by one or a combination of the listed perils, and to do that the insured must necessarily identify the relevant peril or perils.55

The second and last difference between Institute Cargo Clauses (A), (B) and (C) can be found in exclusion clauses 4 and 6. (B) and (C) remove from the scope of cover ‘deliberate damage to or deliberate destruction of the subject-matter insured or any part thereof by the wrongful act of any person or persons’ (4.7). (B) or (C) also exclude s number of perils characterised as different forms of deprivation of possession: ‘capture seizure arrest restraint or detainment, and the consequences thereof or any attempt thereat’ (6.2). The same exclusion can be found in (A), but, in line with its cover of deliberate damage or destruction, the latter places loss or damage resulting from acts of piracy within the scope of cover of (A).56

6. DISCUSSION We have seen that, under Incoterms® 2010, as in prior editions of these Rules, the duty to insure required the seller to insure the goods under Institute Cargo Clauses (C) of the Institute of London Underwriters.57 Explicitly, the Rules gave (C) clauses their seal of approval. The Rules required the seller to ‘obtain at its own expense cargo insurance complying at least with the minimum cover as provided by Clauses (C) of the Institute Cargo Clauses’,58 but implic54 Italics added to mark the perils which are present in type B wording, but which are absent from type C. 55 H Bennett, ‘Reading Marine Insurance Contracts: Determining the Scope of Cover’ (2019) 27 Asia Pac L Rev 239, 241. References to the balance of probabilities should be understood to refer to how the matter would be decided in the courts of England and Wales. See also J Dunt, ‘English Law and Practice’ in J Dunt (ed), International Cargo Insurance (Routledge 2012) 49–109, 68 (para 3.36). 56 On piracy, or forcible robbery at sea, see Ö Gürses, Marine Insurance Law (2nd edn, Routledge 2017) 186–7. 57 See above n 45. 58 Clause A3 of both CIF and CIP according to Incoterms® 2010 Rules. Stress added by the author.

258  Research handbook on marine insurance law itly, this requirement also conveyed the message that (C) clauses were the benchmark against which any other insurance clauses should be measured, hinting that those who procured this level of insurance protection would conform with best trade practices. The change from Incoterms® 2010 to Incoterms® 2020 increased the level of insurance coverage required on the goods under CIP while maintaining the same level of insurance coverage required on the goods under CIF. The former, but not the latter, now points towards the use of Institute Cargo Clauses (A). In sales which involve carriage by sea or inland waterway, this change gives the parties a choice, where before there was none, on the default levels of insurance coverage to be contractually demanded. Contracting parties now have two different default levels of insurance coverage to choose from, one considerably more robust than the other. Although both sets can still be said to have been granted the Rules’ seal of approval, there is now a clear subliminal message that those who seek the highest standard of protection should go for (A). Both before and after the change, contracting parties were and still are, of course, entirely free to deviate from the Rules. One could argue that an additional unwritten implication of this change is that, in the sale of luxury or high-value goods, Institute Cargo Clauses (A) are to be preferred, so CIP should be preferred to CIF, whereas in shipments of commodities or other lower-value goods the preference might still lie with (C) clauses and CIF.59 This is in keeping with market practice.60 The indirect impact of this change goes well beyond the parties’ choice whether to use CIP or CIF in sales which involve carriage by sea or inland waterway. Even when contemplating to use one of the Incoterms® 2020 Rules that set forth no duty to insure, and regardless of the mode of transport that is to be used, there is a new implicit message being conveyed in Incoterms® 2020. This new implicit message is that the insurance protection which was previously seen as the norm is now often too narrow and therefore insufficient to meet the parties’ insurance needs. In other words, before 2020, only one type of insurance protection was mentioned and therefore this was implicitly the benchmark to look for when seeking insurance coverage. As from 2020, two different insurance standards are used, one broader than the other. This prompts an invitation to think which of the two is more appropriate to each set of circumstances. Insofar as the change came as a result of market pressures to increase the required minimum level of insurance protection, as hinted in the commentaries,61 the change may also have the effect of reducing the number of cases of double insurance. Double insurance occurs where two or more insurance contracts cover, wholly or partly, the same risks and the aggregate sums insured exceed the value of the cargo.62 If indeed it was recently often the case that a seller

59 See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 70 of the Introduction, by C Debattista. According to this paragraph, during the consultations leading to the change, there were advocates for and against the change. Among the latter were those involved in the maritime trade of commodities. 60 According to J Dunt, ‘English Law and Practice’ in J Dunt (ed), International Cargo Insurance (Routledge 2012) 49–109, 67 (para 3.34), ‘[t]he usual practice is to insure cargo on all risks terms under the Institute Cargo Clauses (A)’, while (B) and (C) clauses ‘are only used for special commodities and trades and for cargo reinsurance’ (n 172). 61 See Incoterms® 2020. ICC Rules for the Use of Domestic and International Trade Terms, © 2019 International Chamber of Commerce (ICC), para 70 of the Introduction, by C Debattista. 62 See ‘Marine Assureds Overinsured by Double Insurance’ [2004] Shipping and Trade Law, Westlaw UK, 1–4 June; see also J Dunt, ‘English Law and Practice’ in J Dunt (ed), International Cargo

Insurance requirements in Incoterms® 2020  259 would obtain insurance for the buyer as per Institute Cargo Clauses (C), to meet the contractual standard, but then the buyer would find it lacking and would purchase additional insurance as per the wider coverage of Institute Cargo Clauses (A), there would be waste in these instances of double insurance. In other words, financial resources were being spent inefficiently in the purchase of superfluous insurance coverage. It is now hoped that, as a result of this change in Incoterms® 2020, at least in some of those situations the parties would choose to contract under CIP, thereby demanding that the seller obtains the highest level of insurance protection as per Institute Cargo Clauses (A). If this happens, the buyer might no longer feel the need to supplement the seller’s contractually required insurance coverage with a second, more robust insurance contract of its own.

7. CONCLUSIONS Under Incoterms® 2020 Rules, the International Chamber of Commerce upgraded the default level of insurance coverage required on the goods under CIP, while maintaining the previous default level of insurance coverage required on goods sold under CIF. The change from Incoterms® 2010 to Incoterms® 2020 increased the amount of insurance coverage required on the goods under CIP. This development gives the parties a choice, where before there was none, on the default level of insurance coverage to be contractually required. Now both Institute Cargo Clauses (A) and (C) get the International Chamber of Commerce’s seal of approval. Nonetheless, because the former, an all risks type of insurance, provides for substantially more robust insurance coverage than the latter, which corresponds to a named perils insurance, there is now a clear subliminal message that those who seek the highest standard of protection should choose Institute Cargo Clauses (A). Consequently, it seems that in the sale of luxury or high-value goods, Institute Cargo Clauses (A) are to be preferred, so CIP should be preferred to CIF, whereas in shipments of commodities or other lower-value goods the preference might still lie with CIF. Regardless of the actual rule that comes to be chosen by the parties, Incoterms® 2020 now encapsulates an additional unwritten caveat: that the level of insurance protection which was previously seen as the norm might nowadays often prove too narrow and therefore insufficient to meet the insurance needs of present-day parties in an international sale of goods. It is also submitted that the recent developments are likely to reduce the number of cases of double insurance.

Insurance (Routledge 2012) 49–109, 108 (para 3.123). The author states, however, that, ‘in practice, double insurance is comparatively rare in marine cargo insurance’.

13. The use of insurance documents in international trade: enabling digitalisation Miriam Goldby1

1. INTRODUCTION Trading in goods across borders is a high-risk activity that would be significantly impeded if not for insurance products, whereby these risks (or a portion of them) are transferred to a third-party insurer. While in recent years there has been a drive to harness technologies promoting innovation within the insurance industry and across other sectors supporting international trade (including transport and finance), international trade activities continue to rely heavily on paper documents for the information required to support transacting. Relying on paper documents is inefficient and not without its risks, as has been demonstrated by disputes that have played out in the courts in recent years,2 and there has been a drive to harness emerging technologies to digitalise the information supply chain underpinning international trade in goods.3 It has been proposed that certain of these technologies, including distributed ledger technology (DLT) or blockchain, the Internet of Things (IoT), smart contracts technology and machine learning and Artificial Intelligence (AI) can also be helpful in the cargo insurance context, in particular when it comes to claims notification, handling and settlement.4

Much of the work for what became this chapter was done while the author was a visiting professor at the Centre for Maritime Law in January 2023. The author wishes to acknowledge the valuable support and insights received from CML’s permanent academic and research staff. All errors and omissions remain the author’s own. 2 See, for example, ABN Amro Bank NV v Royal and Sun Alliance Insurance plc [2021] EWCA Civ 1789, [2022] 1 WLR 1773; Quadra Commodities SA v XL InsuranceCo SE [2022] EWHC 431 (Comm), [2022] 2 Lloyd’s Rep 541. 3 For a good general overview see Ziyang Fan et al, The Promise of TradeTech: Policy Approaches to Harness Trade Digitalization (World Economic Forum and World Trade Organization 2022) www​ .wto​.org/​english/​res​_e/​publications​_e/​tradtec​hpolicyhar​ddigit0422​_e​.htm accessed 30 June 2023. 4 See, for example, N Mavrias and M Lin, ‘Blockchain: Some Potential Implications for Marine Insurance’ Standard Bulletin (March 2018) 6–7, www​.standard​-club​.com/​fileadmin/​uploads/​ standardclub/Documents/Import/publications/bulletins/split-articles/2018/2678970-blockchain-some​ -potential​-implications​-for​-marine​-insurance.pdf accessed 28 June 2023; M Goldby et al, ‘Triggering Innovation: How Smart Contracts Bring Policies to Life’, Lloyd’s Emerging Risk Report 2019: Understanding Risk (Lloyd’s of London 2019) www.lloyds.com/news-and-risk-insight/risk-reports/​ library/technology/triggering-innovation accessed 28 June 2023, esp. 24–5; W Yong, ‘Is Artificial Intelligence Relevant to Insurance’, IBM Blog, 1 May 2023, www​.ibm​.com/​blog/​is​-artificial​-intelligence​ -relevant​-to​-insurance/​accessed 28 June 2023; A Clere, ‘Using AI to Improve the Insurance Experience for Good’, InsurTech, 11 May 2023, https://insurtechdigital.com/articles/using-ai-to-improve-the​ -insurance-experience-for​good accessed 28 June 2023; Insurance Europe, ‘AI in the Insurance Sector’ (November 2021), www​.insuranceeurope​.eu/​publications/​2608/​artificial​-intelligence​-ai​-in​-the​ -insurance​-sector/​ accessed 28 June 2023, esp 4; R Maull et al, ‘Taking Control: Artificial Intelligence 1

260

The use of insurance documents in international trade: enabling digitalisation  261 One of the barriers to adopting and using digital alternatives is often the legal framework governing this documentation and its use in transactions. With a focus on UK law and the marine insurance market in London, this chapter discusses to what extent legal reform may be required to enable effective technological innovation when digitalising transactions in the cargo insurance space. The chapter also analyses how the UK Electronic Trade Documents Act 2023 may achieve the necessary enabling effects.

2.

MARINE INSURANCE DOCUMENTS

Marine insurance documents come in two primary forms: policies and certificates. Marine policies are not explicitly defined in the Marine Insurance Act 1906. However, reference is made in the Act to the form of policy found in the Schedule to the Act, namely the SG form, which after centuries of use, became defunct in the 1990s, making this part of the Schedule something of an anachronism.5 Other provisions of the Act provide that a policy must fulfil the following formal requirements: it must identify the assured, or ‘some person who effects the insurance on his behalf’,6 and be signed by the insurer.7 Present-day marine policies in the London Market are issued mainly in the MAR 91 form (formally issued policies)8 or simply using the Market Reform Contract (MRC) form, which legally may be able to function as both a slip and a policy under the Act.9 Both formal policy documents and MRCs are often in electronic form.10 Centralised facilities for electronic issue are available in the London Insurance Market, a subscription market.11 MIA 1906, s 22, a leftover from now-defunct stamp duty legislation, provides that the policy is the only admissible evidence of the existence of an insurance contract. This is another anachronistic provision in the Act which had the purpose of disincentivising tax evasion.12 While it remains on the statute book, it is essential that documentation issued by the industry be recognised as constitutive of a policy by the courts.13 The documentation used for Hull and Machinery (H&M) policies differs from that used in cargo cover. Most cargo cover is placed on an open cover basis, whereby the contract establishes uniform terms that will cover declared shipments over an agreed period (usually one year). Assureds that require cover of individual shipments to be documented to be able to and Insurance’, Lloyd’s Emerging Risk Report 2019: Technology (Lloyd’s of London 2019) https://​ assets​.lloyds​.com/​assets/​pdf​-taking​-control​-aireport​-2019​-final/​1/​pdf​-taking​-control​-aireport​-2019​-final​ .PDF accessed 28 June 2023, 42. 5 Law Commission of England and Wales and Scottish Law Commission (hereinafter, ‘Law Commissions’), The Requirement for a Formal Marine Policy: Should Section 22 be Repealed? Reforming Insurance Contract Law Issues Paper 9 (October 2010) www​.lawcom​.gov​.uk/​project/​insurance​-contract​ -law/​accessed 30 June 2023, para 5.33. 6 MIA 1906, s 23(1). 7 Ibid, s 24(1). 8 See discussion in M Goldby, Electronic Documents in Maritime Trade: Law and Practice (2nd edn, OUP 2019) para 7.12. 9 Ibid, paras 7.17–7.18. 10 Ibid, paras 12.09 and 12.18. 11 Ibid, para 8.29. 12 Law Commissions (n 5), paras 2.1–2.20. 13 Ibid, paras 2.21–2.22.

262  Research handbook on marine insurance law tender an insurance document to their buyer or present it to a bank14 are issued a cargo insurance certificate covering the specific shipment.15 Insurance certificates can be of two distinct types. They can either be documents issued by an insurance broker, certifying that the broker has taken out insurance on behalf of the insured,16 or they can be issued directly by the insurer (sometimes referred to as ‘American certificates’).17 Cargo insurance certificates tend to be of the latter type and to be issued by declaration under (often facultative-obligatory)18 open cover agreements. As such, they import a direct contractual nexus between the assured and the insurer and document the latter’s rights against the former. Electronic cargo insurance certificates are widely used in international trade transactions, typically administered over insurers’ portals or using platforms such as QuickAssure,19 which in the London Market is used ubiquitously. In the London Market, Lloyd’s Agency20 used to administer the issuance of cargo insurance certificates under the Insurance Certificate Byelaw.21 However, this service has been discontinued, and the Lloyd’s Certificate Office closed. Guidelines22 are available for issuing Lloyd’s branded certificates. The byelaw itself has not yet been repealed at the time of writing. While there is no longer any need for much of its content, which pertains to the role of the Society and Lloyd’s Agency, there are still certain provisions which may remain relevant to the issue and use of Lloyd’s branded cargo insurance certificates. These are Rule 4, which provides safeguards against unauthorised alterations of issued certificates; Rule 5, which confers on the certificate holder a right to have the certificate replaced by a formal policy document and Rule 6, on brokers’ obligations towards underwriters. The power to make regulations and rules conferred on Lloyd’s Council under Rule 8 might also need to be partially preserved to maintain control over branding. To be clear, there are no instances of which the author is aware where resort to these rules was had or needed, so these issues may not be ones which, in practice, require to be formally provided for in a byelaw. However, the Rules provide a framework to resolve such issues should they arise. Preserving the right in Rule 5 might be particularly important because, as indicated above, under MIA 1906, s 22, the certificate holder might require a policy.23

As discussed below, section 5. Goldby (n 8), paras 7.13–7.14. 16 Ibid, para 7.20. 17 Ibid, para 7.21. 18 Ibid, para 7.14. 19 World Assurance, QuickAssure, www​.worldassurance​.com/​quickassure/​accessed 28 June 2023. 20 Lloyd’s Agency Network, www​.lloyds​.com/​resources​-and​-services/​lloyds​-agency​-network accessed 28 June 2023. 21 Insurance Certificates Byelaw No 1 of 2006 (15 February 2006) https://​assets​.lloyds​.com/​assets/​ pdf​-insurance​-certificates​-byelaw/​1/​pdf​-insurance​-certificates​-byelaw​.pdf accessed 28 June 2023. 22 Lloyd’s Marine insurance certificates – guidelines for use of the Lloyd’s brand, v3, September 2021, https://​assets​.lloyds​.com/​media/​3752e11e​-bbb9​-4f05​-9f6a​-d67311ea786b/​Lloyd’s​-Marine​-Insurance​ -Certificates​_guidelines​.pdf accessed 28 June 2023. 23 See text to n 12 and n 13. 14 15

The use of insurance documents in international trade: enabling digitalisation  263

3.

THE NOTION OF THE ASSIGNABLE DOCUMENT: ASSIGNMENT BY TRANSFER OF THE DOCUMENT ITSELF

3.1

Assignment of Policy

A feature of marine policies is that the document itself (as distinct from the rights under the marine insurance contract it evidences) is assignable. Assignment of the policy document itself is provided for by MIA 1906, s 50.24 A marine policy is assignable unless its terms expressly prohibit its assignment25 and may be assigned before or after the loss occurs.26 Where the policy is assigned, the assignee may sue on the policy in its own name and obtains better rights than it would have obtained from the legal or equitable assignment of contractual rights because when the policy itself is assigned, the insurer may raise only defences that it would have been entitled to raise if the action had been brought in the name of the person by, or on behalf of whom, the policy was effected if those defences arise out of the contract itself27 (including any defences relating to an alleged breach of the duty of fair presentation at the time of formation of the contract). For the assignment to be effective under s 50, the entire interest must be assigned because the expression ‘beneficial interest’ in s 50(2) is taken to mean the whole beneficial interest.28 Under s 51, assignment must take place before or concurrently with the transfer of the assignor’s interest in the subject matter insured, otherwise the assignment is not operative.29 MIA 1906, s 50(3) provides that assignment may be by indorsement on the policy or other customary manner. Indorsement requires the transferee’s signature or that of someone authorised to act on its behalf. The cases on the effectiveness of an assignment of the policy in the absence of such an indorsement are conflicting,30 and the result is likely to depend on proving that the method used in the relevant case is a ‘customary manner’ of assignment (although if this is not proven, an assignment in equity may still be found to have occurred). The question MIA 1906, s 50(1). It is unclear whether this has effect only as between the assignee and the insurer, or also as between the assignee and assignor. See H Bennett, The Law of Marine Insurance (2nd edn, OUP 2006) paras 20.27 and 20.28. 26 MIA 1906, s 50(1). The buyer must pay the seller against tender of compliant documents, even if the goods are lost or damaged prior to arrival at destination: see Goldby (n 8) para 3.05. For this reason, it is essential that the policy is capable of being assigned also after loss. 27 MIA 1906, s 50(2). See also Barker v Adam (1910) 15 Com Cas 227. An example of a defence arising out of the contract is non-disclosure of material facts by the assured at the time the contract is entered into (William Pickersgill & Sons Ltd v London & Provincial Marine & General Insurance Co Ltd [1912] 3 KB 614) or any breach of the duty of utmost good faith, or fraud on the part of the original assured (Black King Shipping Corp v Massie (The Litsion Pride) [1985] 1 Lloyd’s Rep 437, 519). See also Graham Joint Stock Shipping Co Ltd v Merchants Marine Insurance Co Ltd (The Ioanna) (No 1) [1924] AC 294 (HL), 297; cf P Samuel & Co Ltd v Dumas [1924] AC 431 (HL), in which the mortgagee was a party to the policy in his own right, rather than an assignee. 28 Williams v Atlantic Assurance Co [1933] 1 KB 81 (CA), 100; Raiffeisen Zentralbank Österreich AG v Five Star Trading LLC (The Mount I) [2001] QB 825, [72]. Note however, that a partial interest may be assigned in equity. See The Mount I (ibid), [83]. 29 See Powles v Innes (1843) 11 M&W 10; North of England Oil Cake Co v Archangel Maritime Ins Co (1875) LR 10 QB 249 (Div Ct). 30 See Barker (n 27); Safadi v Western Assurance Co (1933) 46 Ll L Rep 140, 144; Iraqi Ministry of Defence v Arcepey Shipping Co SA (The Angel Bell) [1979] 2 Lloyd’s Rep 491, 497. 24 25

264  Research handbook on marine insurance law here is whether, to prove this, the court must be satisfied that the relevant custom is ‘reasonable, certain and notorious’,31 rather than simply that a prevailing trade practice be proved.32 The British Insurance Law Association (BILA) has expressed an opinion that the latter should at least arguably be sufficient33 and, given this, expressed concern at the Law Commissions’ proposal to retain the word ‘customary’ within s 50(3) in the intended reforms to MIA 1906,34 although these reforms have not to date taken place. The Association added that its concern was that the retention of the word would prevent the achievement of the intended result of the reform, which ‘is to allow assignment in modes which the market practice now regards as acceptable’.35 The assignability of the policy itself was ‘designed to respond to the demands of international sale of goods contracts, especially on CIF terms’,36 but in practice, it is the cargo insurance certificate that now performs this particular function. 3.2

Assignment of a Certificate

Cargo insurance certificates have been used for more than 100 years and were designed by the insurance industry specifically for use in international sales of goods.37 As such, they were conceived as documents granting rights to the holder (that is, the person in possession of them to whom they had been validly transferred).38 In Diamond Alkali, McCardie J held that MIA 1906, s 50, does not apply to certificates, and he raised doubts as to their assignability;39

See Devonald v Rosser & Sons [1906] 2 KB 728 (CA), 743. See General Reinsurance Corp v Forsakringsaktiebolaget Fennia Patria [1983] QB 856 (CA), 874 (Slade LJ): ‘There is, however, the world of difference between a course of conduct that is frequently, or even habitually, followed in a particular commercial community as a matter of grace and a course which is habitually followed, because it is considered that the parties concerned have a legally binding right to demand it. As Ungoed-Thomas J pointed out in Cunliffe-Owen v Teather & Greenwood [1967] 1 WLR 1421, 1438: “What is necessary is that for a practice to be a recognised usage it should be established as a practice having binding effect.”’ 33 Law Commission of England and Wales and Scottish Law Commission, Insurance Contract Law: Summary of Responses to Second Consultation Paper Post Contract Duties and Other Issues – Chapter 4: Policies and Premiums in Marine Insurance (February 2013) www​.lawcom​.gov​.uk/​document/​insurance​ -contract​-law​-post​-contract​-duties​-and​-other​-issues​-consultation/​ accessed 28 June 2023 (hereinafter Law Commissions, Summary of Responses), para 2.49. 34 Law Commission of England and Wales and Scottish Law Commission, Insurance Contract Law: Post Contract Duties and Other Issues: Consultation Paper (CP No 201, 20 December 2011), www​ .lawcom​.gov​.uk/​document/​insurance​-contract​-law​-post​-contract​-duties​-and​-other​-issues​-consultation/​ accessed 28 June 2023, para 17.38: ‘We propose to amend section 50(3) to say, quite simply, that a marine insurance contract may be assigned in any customary manner or as agreed between the parties to the transfer.’ This amendment is no longer required in light of the imminent coming into force of the Electronic Trade Documents Act, discussed below, text to n 75. The issue it purported to resolve has been addressed otherwise. 35 Law Commissions, Summary of Responses (n 33), para 2.49. 36 Bennett (n 25), para 20.15. 37 PW Thayer, ‘Marine Insurance Certificates’ (1935) 49 Harvard LR 239, 239–44. 38 See Goldby (n 8), para 7.21. 39 Diamond Alkali Export Corp v Fl Bourgeois [1921] 3 KB 443, 457: ‘[Section 50(3)] only applies, so far as I can see, to that which is an actual marine policy […] If, as is admitted, this document be a certificate only and not a policy, it therefore seems not even to be admissible in evidence before me. If the certificate does not fall within the Marine Insurance Act it appears to be only assignable if at all by 31 32

The use of insurance documents in international trade: enabling digitalisation  265 nonetheless, they continued to be used widely in Cost, Insurance and Freight (CIF) transactions, and their assignability appears to have been accepted in several subsequent cases. For example, in Koskas,40 it was held that: The next point taken was that the certificate required that a person to sue on it should have it indorsed and that though the plaintiff was the holder of and in possession of the certificate, and though he appeared to have bought the goods which have been lost from the person who was the original assured, he could not recover because the original assured had not indorsed on the back of the certificate his name so as to amount to an indorsement. Now the language of the certificate is extremely puzzling. It conveys all the rights of the original policy holder as if the property were covered by a special policy direct to the holder of this certificate. The loss is payable to the order of assured or order. It looks rather tautological in itself; and in my view the English underwriters have very properly decided not to press that point. They have issued a very ambiguous document.

This suggests that documentary assignment of certificates occurred as a matter of course and that, unless clearly required, indorsement was not necessary to effect an assignment. An essential feature of the certificate is that it is viewed as a contract between the insurer and the certificate holder separate from the underlying open cover under which it was issued.41 In De Monchy v Phoenix Insurance Co of Hartford,42 it was held that: [A]ll clauses of the [underlying] policy which are essential to the contract of marine insurance must be read into the certificate, but beyond that there is no necessity to go. The condition in question43 is a collateral stipulation imposing a condition precedent. It has nothing particular to do with insurance, but might be applied to any contract. Common sense and fairness revolts against the idea of this being enforced against the holder or indorsee of the certificate. Neither the holder as here nor a possible indorsee could ever have seen the policy. There is not even expressed in the certificate a right to ask for exhibition of the policy. Against them it may be fair to assume ordinary insurance clauses, but not to assume a collateral agreement of this sort […] I think that the contract of insurance to which the assignee becomes a party is expressed in the certificate of insurance which becomes in his hands a policy.

A consequence of this separateness is that assignment of the certificate will not allow the insurer to raise against the certificate holder defences available to the former against the policy assured, even if they arise out of the underlying contract.44 Thus, the certificate holder should receive cover free from any liabilities – an important advantage in the case of CIF contracts, under which the cost of insurance is included in the price of the goods. Indeed, it would appear that certificates were developed specially to address the particular feature of policy assignment

writing in accordance with the provisions of the Judicature Act, 1873, s 25, sub-s. 6.’ The latter provision has since been re-enacted as s 136(1) of the Law of Property Act 1925. 40 Koskas v Standard Marine Insurance Co Ltd (1927) 27 Ll L Rep 59 (CA), 61. 41 See Goldby (n 8), para 7.23. 42 (1929) 34 Ll L Rep 201 (HL), 205 and 209. Regarding the application of open cover terms to the certificate contract see MacLeod Ross & Co Ltd v Cie d’Assurances Generales l’Helvetia [1952] 1 Lloyd’s Rep 12 (CA). On the treatment of discrepancies, see Evialis SA v SIAT [2003] EWHC 863 (Comm), [2003] 2 Lloyd’s Rep 377, esp [17] and [26]–[42]; Craft Enterprises (International) Ltd v AXA Insurance Co [2004] EWCA Civ 171, [2005] Lloyd’s Rep IR 14. 43 This refers to a one-year limitation clause contained in the underlying open cover policy. 44 See De Monchy (n 42) and Glocom Ltd v Eagle Star Insurance Co Ltd (CA, 4 October 1996).

266  Research handbook on marine insurance law that allowed the insurer to raise against the assignee defences arising out of the contract that he would have been able to raise against the assured under the underlying policy or open cover.45

4.

TRANSFER OF TITLE TO GOODS, INSURABLE INTEREST AND INDEMNITY: THE LAW IN BRIEF

Following a valid assignment, the assignee becomes an assured under the insurance contract. Having been established as an assured, and therefore a person with standing to make a claim, it is incumbent upon the claimant to show several things: not only that it has suffered a financial loss caused by a peril insured against, but also that the subject matter of the loss was covered and that the assured had an insurable interest in it at the time of loss,46 as required by MIA 1906, s 6. ‘Insurable interest’ is defined by MIA 1906, s 5(1), as an interest in a marine adventure. By way of illustration, s 5(2) provides: In particular a person is interested in a marine adventure where he stands in any legal or equitable relation to the adventure or to any insurable property at risk therein, in consequence of which he may benefit by the safety or due arrival of insurable property, or may be prejudiced by its loss, or by damage thereto, or by the detention thereof, or may incur liability in respect thereof.

The courts have elaborated upon the concept of insurable interest by establishing that only a person with a direct concern in the subject matter insured has an insurable interest.47 The definition of the subject matter insured is essential to determining whether or not the claimant had an insurable interest, and, in cargo insurance, may require wording able to capture the safe arrival of the goods rather than (just) the goods themselves.48 Being able to show insurable interest is particularly relevant in cargo insurance, especially where the goods in question are sold in transit and are subject to possible rejection by the buyer. This is covered by MIA 1906, s 7, which provides expressly that even a defeasible interest (such as that of the buyer who might reject the goods or treat them as at the seller’s risk by reason of delay in delivery or otherwise) is insurable. Further, in an international sale of cargo, others besides the seller and buyer may be interested in the safe arrival of the goods, such as a financing bank. Provision for these cases is made under MIA 1906, s 8, which provides that

45 Thayer (n 37), 241. It must be noted, however, that, under an open cover that is facultative for the underwriters, precontractual duties of utmost good faith attach to each declaration (Berger & Light Diffusers Pty Ltd v Pollock [1973] 2 Lloyd’s Rep 442), and any breach would therefore relate to the contract in the certificate rather than the underlying contract. 46 Regarding the importance of the assignee’s insurable interest in the subject matter of the insurance see Comlex Ltd v Allianz Insurance plc [2016] CSOH 87, [2016] Lloyd’s Rep IR 631. 47 See Macaura v Northern Assurance Co Ltd [1925] AC 619 (HL), 630. The direct concern must be in the form of a legal or equitable interest. 48 See Wilson v Jones (1867) LR 2 Ex 139 and its treatment in Macaura, ibid, 627–8. See also Quadra Commodities v XL Insurance and others [2022] EWHC 431 (Comm), [65]: ‘when read with the other terms of the Policy, I consider that what the clause provides is that the insurance is on all types of property (“goods and/or merchandise and/or cargo and/or interest of every description”), whether it is being used incidentally to the business of the Assured, or otherwise, in which the Assured has a relevant insurable interest.’

The use of insurance documents in international trade: enabling digitalisation  267 a partial interest of any nature is insurable. Similar rules to these apply in jurisdictions that have modelled their laws on MIA 1906,49 as well as in the US.50 These provisions must be read in conjunction with MIA 1906 s 15, which provides that where the assured transfers his interest in the subject matter insured, he does not thereby transfer to the transferee his rights under the contract of insurance unless there is an express or implied agreement with the transferee to that effect. The timing of the assignment is thus crucial. Accordingly, the insurance document is normally delivered and (where necessary) indorsed to the transferee alongside other documents, including a bill of lading, which is a document of title to the goods covered by the insurance.51

5.

THE USE OF INSURANCE DOCUMENTS IN INTERNATIONAL TRADE

The performance of international trade and trade finance transactions often involves the transfer of documents. In CIF transactions, typically, these documents include an invoice which sets out a description and the cost of the goods, an insurance document which evidences that cargo insurance has been arranged to cover the goods while in transit,52 and a transport document that evidences that a contract of carriage whereby the goods will be delivered at the agreed destination has been entered into, and relevant freight paid.53 Under the ICC INCOTERMS 2020,54 insurance documents are a required part of performance by the seller not only in the performance of contracts on CIF terms but also in transactions on the basis of Carriage and Insurance Paid To (CIP) terms. Trade finance provided by a documentary credit is almost invariably provided subject to the Uniform Customs and Practices on Documentary Credits (UCP) version 600 (2007). These contain rules specific to insurance documents presented in fulfilment of the documentary credit conditions. One important feature that the insurance document has to comply with, under INCOTERMS, is that it must give the buyer the right to claim directly against the insurer.55 The insurance

See, for example, Federal Marine Insurance Act SC 1993, c 22 (Canada), s 10, and Marine Insurance Act 1909, No 11 of 1909 (Australia), s 11. 50 For an overview of US law in this regard, see DT Rave and S Tranchina, ‘Marine Cargo Insurance: An Overview’ (1991–2) 66 Tulane LR 371, 382–4. 51 See below, text to n 52. 52 For example, under art 28(f)(iii) of the UCP 600 the insurance document must evidence that risks are covered during transport of the goods between the places specified in the credit. 53 Ireland v Livingston (1872) LR 5 HL 395, 406–7. 54 International Chamber of Commerce (ICC) Incoterms® 2020: ICC Rules for the Use of Domestic and International Trade Terms (2019). 55 CIF INCOTERMS 2020, para A5. See commentary on this provision in Trade Finance Global, Cost, Insurance & Freight (CIF) Incoterms® 2020 Rules – A 2023 TFG Walkthrough. www​ .tradefinanceglobal​.com/​freight​-forwarding/​incoterms/​cif​-price​-cost​-insurance​-and​-freight/​ accessed 30 June 2023: ‘The seller must provide the buyer a separate contract or a certificate under an existing policy giving the details of the shipment to enable the buyer, or anyone else having an insurable interest in the goods, to claim from the insurer. This document usually shows the seller as the insured and is then endorsed by the seller on the back of the original/s in blank or with a specific endorsement.’ Under paras A1 and B1 of the INCOTERMS 2020 CIF rule, ‘an electronic form’ will satisfy the requirement to provide a policy or other evidence, ‘as agreed, or where there is no agreement, as is customary’. 49

268  Research handbook on marine insurance law document presented under a documentary credit arrangement must comply with the requirements found in Article 28 of the UCP 600. Article 28(a) provides that the parties who can sign the insurance document include a proxy, and the signature of an agent or proxy must indicate whether it is for or on behalf of the insurance company or underwriter. Article 28(c) disqualifies cover notes which are not issued by or on behalf of the insurer from constituting a compliant presentation. Article 28(d) provides that ‘[a]n insurance policy is acceptable in lieu of an insurance certificate or a declaration under an open cover’. Taken together, these provisions indicate that the document presented should be one issued by the insurer and legally capable of forming the basis of a claim against the insurer. Article 28(b) also requires that the full set of originals be presented to the banks, where the insurance document is issued in a set of more than one original. This requirement is in line with the bank’s need to receive the document not only for the purpose of checking it for compliance but also, where needed, to facilitate the transfer to it of rights pertinent to its position as a secured creditor. The transfer of such rights occurs by operation of laws determining the legal effects of being the ‘holder’ of the insurance document in question.56 Electronic presentation of documents in fulfilment of documentary credit requirements is provided for in the electronic supplement to the UCP600, the eUCP v 2.1 (2023).57 While the eUCP covers presentation rules, and includes a definition of ‘Electronic Transferable Record’ (Article e3(b)(5)), it does not specify how the transfer of rights will be effected where electronic data are being presented in place of paper documents. Article e9, on ‘Originals and Copies’, simply provides ‘Any requirement for presentation of one or more originals or copies of an electronic record is satisfied by the presentation of one electronic record’. It is true that there is likewise no provision in the UCP600 regarding the transfer of rights through (indorsement and) delivery of paper documents. However, such a provision is not needed in the paper context as these methods of transferring rights are recognised as effective by the laws of most jurisdictions, deriving as they do from the lex mercatoria.58 But in the absence of law reform this does not apply to the transfer of rights using documents in electronic form, so that the question of how important rights, normally received by the bank pursuant to the assignment of a cargo insurance certificate, are to be acquired when data rather than documents are being presented remains unaddressed. Therefore, the method for transferring those rights needs to be carefully considered and built in when entering into the letter of credit arrangement. A more recent set of standard rules prepared by the International Chamber of Commerce, the Uniform Rules on Digital Trade Transactions v 1.0 (2021) (URDTT)59 do on the other As discussed above, text to n 24. Supplement to the Uniform Customs and Practice for Documentary Credits for Electronic Presentation (Version 2.0) (2019) with additional text inserted on 30 June 2023 intended to align it with the UNCITRAL Model Law on Electronic Transferable Records (MLETR) in respect of issues pertaining to electronic transferable records (version 2.1) (2023) https://​iccwbo​.org/​news​-publications/​ policies​-reports/​eucp​-version​-2​-1​-icc​-uniform​-customs​-and​-practice​-for​-documentary​-credits/​#single​ -hero​-document accessed 1 July 2023. It was emphasised on the publication of v 2.1 that ‘[i]t is very important to note that this was not a revision nor an update of the eUCP. It was solely an alignment with MLETR with respect to electronic transferable records.’ Ibid, 1. 58 Law Commission of England and Wales, Digital Assets: Electronic Trade Documents, Consultation Paper 254, 30 April 2021 www​.lawcom​.gov​.uk/​project/​electronic​-trade​-documents/​ accessed 28 June 2023 (hereinafter LCCP254), para 3.6. 59 International Chamber of Commerce, Uniform Rules on Digital Trade Transactions, Version 1.0, 2021, https://​2go​.iccwbo​.org/​uniform​-rules​-for​-digital​-trade​-transactions​-urdtt​-version​-1​.html accessed 28 June 2023. 56 57

The use of insurance documents in international trade: enabling digitalisation  269 hand, contain provisions envisaging a transfer of electronic records having the same effects as a transfer of paper documents when it comes to the transfer of rights. Article 7(f) provides: ‘Where the applicable law requires or permits delivery, transfer or possession of an Electronic Record, that requirement or permission is met by the transfer of that Electronic Record to the exclusive control of the Addressee.’ It, therefore, adopts a similar approach to the United Nations Commission on International Trade Law (UNCITRAL) Model Law on Electronic Transferable Records 2017. Article 11 of this Model Law provides that legal requirements for documentary possession are met where a reliable method is used to enable an electronic record to be exclusively controlled.60 The URDTT are a very recent addition to the landscape regarding trade and trade financing transactions, and it is not clear how widespread their use is at the time of writing. This author has not found any evidence indicating where and how they are being used. Pending the wider adoption of URDTT by the trade and trade financing communities, the assignment of rights under the insurance contract may be, and is in fact, often achieved by naming additional assureds when declaring the shipment61 or by identifying the assured in general terms so that all relevant parties can claim under the insurance provided they satisfy the insurable interest requirement. Assignment of the certificate becomes unnecessary, as a sufficiently equivalent effect may be achieved otherwise from the perspective of transferees of the goods in an international trade chain. A consequence of this approach, however, is that a person’s right to claim has to be ascertained when the claim is made by reference to supporting documentation that demonstrates that the claimant is indeed the person with insurable interest to claim.62 This hampers the industry’s ability to increase the automation of claims processes, increasing the speed and efficiency and reducing the cost of claims settlement.63 As is discussed below,64 enabling electronic assignment of the document alongside the transfer of the bill of lading as a document of title to the cargo would generate real-time information as to who has an interest in the goods at the time they are lost or damaged, and who therefore has the standing to claim under the insurance.

6.

NEW TECHNOLOGIES AND CARGO INSURANCE

Blockchain and other new technologies, such as smart contracts technology, have been viewed as having great potential in the insurance space to improve record keeping, to process the data necessary for placing the risk, to settle claims more efficiently and in an environment of increased trust and auditability, and to increase automation.65 For the past few years, work has been under way to harness these technologies within the insurance industry by establishing

See in particular, art 11(1)(a). As was done, for example, in Craft Enterprises (above, n 42). 62 See the discussion above, text to n 46. 63 As noted in Goldby (n 4), 25, ‘In order to assist in enhancing the efficiency of the cargo claims process, and ensure that the payout is made to the correct person, it might be worthwhile for the market to consider building electronic assignment into the functionality of electronic cargo insurance certificates platforms’. 64 See below, section 6. 65 See above, n 4. 60 61

270  Research handbook on marine insurance law collaboratives,66 an insurance-specific blockchain architecture called Canopy in partnership with Corda67 and the launch of pilot schemes. In 2019 Lloyd’s of London published a report exploring potential applications for smart contracts technology in insurance, with one of the case studies being cargo insurance.68 They also launched Parsyl, a cargo insurance application focusing on perishable cargoes.69 In 2021 the Law Commission of England and Wales published a Consultation Paper on Electronic Trade Documents, which was followed by a Report in 2022. These publications explained how the (indorsement and) physical transfer of a paper document could be replicated in the electronic space, for example, using blockchain technology.70 New technologies that have emerged in the past decade and are becoming established can be particularly useful in the cargo insurance space. Blockchain can be used to provide proof of insurance and to effect assignment by transfer of the document, simultaneously with the transfer of an insurable interest in the cargo from seller to buyer. As a source of verifiable data, blockchain can also be used to trigger smart contracts, facilitating crucial steps in the process such as First Notice of Loss.71 IoT, when applied, for example, to sensors fitted in containers, can be used to generate and record real-time data about the goods, for example, the temperatures, humidity and vibrations to which they were exposed in the course of the voyage. Importantly for our purposes, smart contracts technology, in combination with IoT data, can be used to streamline the claims process in case of loss or damage. For example, Lloyd’s cargo insurance application, Parsyl, applies IoT together with smart contracts technology: ‘[w]e recently demonstrated this with an air cargo client – our smart sensors identified a temperature issue and we paid a claim in a record breaking eight-hours.’72 All these technologies, taken together, can enable the development of innovative cargo insurance products, for example, parametric products that reduce or eliminate the pain points for the insurer in settling claims and for the assured in obtaining a payout. However, crucially, in order to harness these technologies in a ‘joined up’ way, the insurer must have easy visibility into where the insurable interest lies as soon as a loss is notified, without having to Most notably the Institutes Risk Stream Collaborative: see ‘The Institutes Risk Stream Collaborative’, www​.riskstream​.org/​accessed 28 June 2023. 67 N Morris, ‘RiskBlock Launches Canopy 2 Blockchain Insurance Platform’, Ledger Insights, 17 September 2018, www​.ledgerinsights​.com/​riskblock​-launches​-canopy​-2​-blockchain​-insurance/​ accessed 28 June 2023. See also Risk Stream Collaborative (n 66): ‘RiskStream leverages a world-class, SOC2 and ISO certified (security) blockchain infrastructure with off-chain data sharing capabilities which ensure personally identifiable information data is exchanged in compliance with rules/regulations.’ 68 Goldby (n 4). 69 Lloyd’s, ‘Parsyl’, www​.lloyds​.com/​news​-and​-insights/​futureset/​join​-the​-community/​product​ -simplification/​parsyl accessed 28 June 2023. See also ‘Parsyl’, www​.parsyl​.com/​ accessed 28 June 2023. 70 LCCP254 (n 58), paras 2.48–2.53. 71 P Schmid on behalf of The Institutes Risk Block Alliance, ‘Blockchain Technology and Insurance’, Presentation given at the American Institute of Marine Insurers (AIMU) Marine Insurance Day Seminar, 5 October 2018, https://​aimuedu​.org/​aimupapers/​3​_Blockchain​_and​_Insurance​_Patrick​ _Schmid​.pdf accessed 28 June 2023, slide 23. The Institutes Risk Block Alliance is the former name of the Risk Stream Collaborative. See ‘The Institutes Changes Name of Its Global Blockchain Consortium to The Institutes RiskStream Collaborative’, 6 May 2019 https://​web​.theinstitutes​.org/​institutes​-changes​ -name​-its​-global​-blockchain​-consortium​-institutes​-riskstream​-collaborative accessed 28 June 2023. 72 Lloyd’s (n 69): ‘Parsyl uses a trusted risk trigger via a smart sensor, leading to faster and simplified claims process.’ 66

The use of insurance documents in international trade: enabling digitalisation  271 rely on time- and resource-intensive manual document verification processes. This would be facilitated by electronic documentary assignment. The next and final section of this chapter is therefore devoted to discussing the extent to which this can be done under current and forthcoming laws.

7.

CAN ASSIGNMENT OF DOCUMENTS BE DONE ELECTRONICALLY? RECENT LEGAL REFORMS

Under English law, the understanding of the concept of document is broad. Section 13 of the Civil Evidence Act 1995 and Rule 31.4 of the Civil Procedure Rules 1998 both provide that ‘“document” means anything in which information of any description is recorded’. Section 7C of the Electronic Communications Act 2000 defines a document as ‘anything stored in electronic form, including text or sound, and visual or audiovisual recording’. At first sight, therefore, it would appear that marine insurance documents would equally be considered documents if issued electronically. However, under UK law as it stood prior to the coming into force of the Electronic Trade Documents Act 2023 (ETDA), assignment by indorsement and delivery of a document could only occur if the document was in tangible form because ‘delivery’73 required a transfer of possession. The law recognised only tangible things as possessable so that if property was not tangible, it was automatically disqualified from being a thing in possession.74 The Law Commission of England and Wales’s Electronic Trade Documents project precisely addressed this ‘possession problem’, a major obstacle to paper trade documents that function on the basis of possession being used in electronic form.75 The project resulted in the ETDA, which received Royal Assent on 20 July 2023 and came into force in September 2023.76 Certain basic notions underpinning the ETDA capture features inherent to paper that are not equally inherent to information in electronic form, and that therefore need to be built in to enable an electronic document to do the same things as a paper one that functions based on possession. A paper trade document generally retains its integrity because it is composed of permanent marks on a physical medium. The paper document also has its own identity: it exists as a thing, distinct from all other things. Therefore, a person who has the document can easily 73 Delivery is defined in the Sale of Goods Act 1979, s 61(1), as the ‘voluntary transfer of possession from one person or another’ while under the Factors Act 1889, s 1(2) states, ‘A person shall be deemed to be in possession of goods or of the documents of title to goods, where the goods or documents are in his actual custody or are held by any other person subject to his control or for him or on his behalf.’ In Forsythe International (UK) Ltd v Silver Shipping Ltd (The Saetta) [1993] 2 Lloyd’s Rep 268, it was held that this definition applies for the purposes of the Sale of Goods Act 1979. See, however, Michael Gerson (Leasing) Ltd v Wilkinson [2001] QB 514 (CA), [34]–[35] and LCCP 254 (n 58), para 5.82 and fn 71. 74 See in particular, OBG Ltd v Allan [2007] UKHL 21; [2008] 1 AC 1; Your Response Ltd v Datateam Business Media Ltd [2014] EWCA Civ 281, [2015] QB 41. See also fuller discussion in M Goldby and W Yang, ‘Solving the Possession Problem: An Examination of the Law Commission’s Proposal on Electronic Trade Documents’ [2021] LMCLQ 605, 607–10. 75 LCCP254 (n 58) and Law Commission of England and Wales, Electronic Trade Documents: Report and Bill, Law Com No 405, 15 March 2022 www​.lawcom​.gov​.uk/​project/​electronic​-trade​ -documents/​accessed 28 June 2023 (hereinafter LCR405). 76 The final version is available at https://​bills​.parliament​.uk/​bills/​3344 accessed 21 July 2023. See s 8(2).

272  Research handbook on marine insurance law show that they have the document, exclude others from it and transfer it to another person (after which they will no longer have it). Because these features are not inherent to information in electronic form, the information has to be stored and structured to give it these features if it is to perform the same functions and achieve the same effects as its paper counterpart. In order to be used in the same way as a paper document, therefore technology must be used to ensure that an electronic document that contains information that makes it a document of a certain type (that is, a cargo insurance certificate or a marine policy) meets the following requirements: (i) the document is identified (that is, given its own identity); (ii) the document’s integrity is ensured; (iii) the document is capable of being controlled exclusively by one person; (iv) the person who has the document is divested of it upon transferring it to another. These requirements appear in the gateway criteria for a document to be an electronic trade document under s 2(2) of the ETDA: 2 Definition of ‘electronic trade document’ (1) This section applies where information in electronic form is information that, if contained in a document in paper form, would lead to the document being a paper trade document. (2) The information, together with any other information with which it is logically associated that is also in electronic form, constitutes an ‘electronic trade document’ for the purposes of this Act if a reliable system is used to— (a) identify the document so that it can be distinguished from any copies, (b) protect the document against unauthorised alteration, (c) secure that it is not possible for more than one person to exercise control of the document at any one time, (d) allow any person who is able to exercise control of the document to demonstrate that the person is able to do so, and (e) secure that a transfer of the document has effect to deprive any person who was able to exercise control of the document immediately before the transfer of the ability to do so (unless the person is able to exercise control by virtue of being a transferee).

In defining an electronic trade document, s 2(2) of the ETDA adds to the definition of ‘document’ for the purposes of English law, set out above, ‘any other information with which it is logically associated’ to capture the composite nature of an electronic trade document (including a human-readable component and a data structure). This is an acknowledgement that a method must be found of giving a structure of some permanence to information in electronic form, as paper does, giving it a specific identity, to fulfil the requirement in s 2(2)(a), namely ‘identify[ing] the document so that it can be distinguished from any copies’. The Act does not prescribe using any particular technology and merely specifies the outcomes to be achieved. However, based on currently available techniques, cryptographic hashing techniques used in blockchain technology provide one way of achieving these outcomes. Cryptographic hashing forms a ‘data structure’ consisting of functional code, which is logically associated with (and specifically identifies) human-readable information.77 As seen from s 2(2) above, a crucial element in the gateway criteria is the notion of exclusive control. The combined effect of s 2(2)(c) and s 2(2)(e) is to establish that the document must be capable of exclusive control, as would be the case with a paper document. Thus, the notion of control is used to require that, for an electronic document to be considered an ETD at law, it be given features that echo features inherent in a paper document that make it possess77

LCR405 (n 75), paras 6.17–6.22.

The use of insurance documents in international trade: enabling digitalisation  273 able. Technologically, the criteria amounting to exclusive controllability of the document can also be fulfilled in practice using blockchain technology. In order to address the possession problem mentioned above, s 3 of the Act provides that if the gateway criteria in s 2(2) are met, making the document an electronic trade document, the document is deemed to be possessable at law: 3 Possession, indorsement and effect of electronic trade documents (1) A person may possess, indorse and part with possession of an electronic trade document. (2) An electronic trade document has the same effect as an equivalent paper trade document. (3) Anything done in relation to an electronic trade document has the same effect (if any) in relation to the document as it would have in relation to an equivalent paper trade document.

It should be noted that s 3 does not refer to control. Indeed, the Act does not seek to answer the question ‘who is in possession of an ETD?’ Instead, it answers the question, ‘what is required to make an electronic document possessable?’ It answers this question by setting out, in s 2, the gateway criteria that must be fulfilled for a document to fall within the definition of an ETD. The former question, that is, ‘who is in possession?’ – a question of fact – is left to be answered according to the circumstances of the case.

8. CONCLUSION This chapter sought to demonstrate, first, the potential benefits of digitalising marine cargo documentation and, second, the extant legal obstacles to using these electronic documents in the same way and to achieve the same effects as their paper counterparts. While these legal obstacles may be overcome with appropriate contractual drafting, these contractual solutions are not such as to facilitate the harnessing of all the benefits that digitalisation may achieve. In particular, the achievement of fuller automation of the claims process based on emerging digital technologies such as DLT, IoT, smart contracts technology and AI is hindered if the insurer does not have a means of ascertaining who the assured is and whether that person has an insurable interest without a resource-intensive process of manual submission and checking of documentation. Establishing who the assured with insurable interest is can be challenging in an international trade scenario, where the cargo can change hands several times before the completion of its transit (and therefore within the duration of cover). In the paper environment, the person with the right to claim is identified by submitting paper documentation in its possession by the assured, including the (indorsed) insurance certificate and the bill of lading (proof of title to the covered goods). In the digital environment, where the electronic system over which the certificate is issued does not provide for electronic assignment, reliance will need to continue to be placed on the claimant to submit the required evidence, stalling further automation of the claims process. In the UK, the recently adopted ETDA will provide a legal framework for electronic assignment which may enable the building of necessary functionality into electronic cargo insurance certificate systems. However, whether these developments take place will ultimately depend on commercial considerations, including perceived demand, and will need to occur alongside digitalisation of other trade documents, particularly the bill of lading.

PART VI  

14. Marine insurance fraud and emerging technology Gary Meggitt

1. INTRODUCTION ‘Fraud and falsehood only dread examination. Truth invites it.’ This observation, which is often erroneously attributed to Samuel Johnson,1 captures one of the most essential aspects of fraud. Its need for ignorance. In the absence of knowledge, fraud can thrive. And fraud is thriving.2 In 2020, UK insurers detected £1.1bn in general insurance fraud, although it is estimated that a further £2bn remains undetected each year.3 In the US, the Coalition Against Insurance Fraud suggests that the annual cost of insurance fraud is at least $80bn.4 A common view among insurers is that approximately 15 per cent of all claims are fraudulent.5 While there is a dearth of public information on the true extent of marine insurance fraud, it seems likely that the losses to insurers and – indirectly – policyholders will be substantial, if only because of the huge sums involved in cases such as Suez Fortune Investments Ltd & Anor v Talbot Underwriting Ltd & Ors (Brillante Virtuoso).6 It therefore comes as little surprise that the UK insurance industry spends £200m per annum on combating fraud and that insurers across the world are investing in emerging technology in the struggle against such criminality. That emerging technology includes blockchain, smart contracts, ‘Big Data’ and AI-enabled underwriting and claims handling (collectively referred to as ‘insurtech’ when discussed in respect of insurance).7 In addition, the development of autonomous vessels and ports gives rise to numerous challenges and opportunities for both would-be fraudsters and marine insurers (and honest policyholders). Yet, while the use of technology to defeat fraud is to be welcomed, it raises an important issue. Is the turn towards

Thomas Cooper was its originator. Some commentators have suggested that insurance itself can ‘lead to loss or claim creation not only by insureds but also by uninsured third parties’. See G. Parchomovsky and P. Siegelman, ‘Third-Party Moral Hazard and the Problem of Insurance Externalities’, 51 J. Legal Stud. 93 (2022). 3 See Association of British Insurers (ABI) Report ‘No Time to Lie: The Average Fraudulent Insurance Claim Rises to £12,000 as Insurers Continue Protecting Honest Customers’ (6/10/2021) available at www​.abi​.org​.uk/​news/​news​-articles/​2021/​10/​detected​-fraud​-2020/​ and IFR, ‘Frequently Asked Questions’ available at https://​www​.theifr​.org​.uk/​en/​faqs/​ 4 See CAIF, ‘Impact’ available at https://​insurancefraud​.org/​fraud​-stats/​ 5 See HKFI Insurance Fraud Prevention Claims Database available at www​.hkfi​.org​.hk/​ifpcd/​en/​ index​.html 6 [2019] EWHC 2599 (Comm), [2019] 2 C.L.C. 403. Discussed further below. 7 Emerging or emergent technology includes AI, gene therapy and nanotechnology. See D. Rotolo, D. Hicks and B. Martin, ‘What Is an Emerging Technology?’ 44(10) Research Policy 1827–43 (2015). 1 2

275

276  Research handbook on marine insurance law a technological solution an admission that the law and the courts have failed when it comes to the fight against marine insurance fraud?8 This chapter begins with a relatively brief examination of marine insurance fraud and how it is addressed in current UK law and regulation. It then discusses the nature and development of ‘insurtech’. The characteristics of autonomous vessels and ports are also touched upon. The chapter concludes with an analysis of the issues arising from the encounter between marine insurance fraud and emerging technology. Given the scope of, and the speed of development, in the subjects involved, a chapter of this length can only serve as an introduction to the use of emerging technology by insurers, by ship owners generally and in the fight against fraud. It is, however, hoped that it will prove useful to readers despite its relative brevity.

2.

MARINE INSURANCE FRAUD9

While there is no UK statutory definition of ‘insurance fraud’, the Fraud Act 2006 provides that someone may be guilty of the offence of ‘fraud’ by way of false representation, failing to disclose information or abuse of position.10 Insofar as civil proceedings are concerned, one must rely on case law. The classic definition of ‘fraud’ was set down by the House of Lords in Derry v Peek11 as follows: ‘Fraud is proved when it is shewn that a false representation has been made knowingly, or without belief in its truth, or recklessly, without caring whether it be true or false.’12 The common law position is that a claim (and a policy itself) may be forfeit in the event of fraud, even in the absence of an express clause to that end in the policy.13 That position, which is discussed further below, has been adjusted by the Insurance Act 2015 (the 2015 Act).14 The Insurance Fraud Register (IFR) has its own – somewhat prolix – definition of insurance fraud15 but, to summarise, it is dishonest (or reckless) conduct directed against an insurance provider for the purpose of financial gain.16 As for maritime fraud, this was defined by International Maritime Bureau, as occurring when a party ‘succeeds, unjustly or illegally, in 8 Zalewski and Majewski discuss the difficulty of assessing what they refer to as the ‘dark number’ of insurance frauds, and the need for technology in doing so, in W. Zalewski and P. Majewski, ‘The Dark Number of Insurance Crimes’, 26 Bialostockie Studia Prawnicze 85 (2021). 9 B. Soyer, Marine Insurance Fraud (Informa Law from Routledge, 2014) remains, arguably, the leading detailed text on marine insurance fraud despite its publication having preceded the passage of the 2015 Act. Other useful guides include Ö. Gürses, Marine Insurance Law (Routledge, 2016), Chapter 12, Fraudulent Claims; and J. Birds, B. Lynch and S. Paul, MacGillivray on Insurance Law (15th ed, Sweet & Maxwell, 2022), Chapters 16–19. 10 The text of the 2006 Act is available at www​.legislation​.gov​.uk/​ukpga/​2006/​35/​contents 11 (1889) 14 App Cas 337. 12 See also Roche J in Wisenthal v World Auxiliary Insurance Corpn Ltd (1930) 38 Ll Rep 54, 62 as cited in Versloot Dredging BV and Another v HDI Gerling Industrie Versicherung AG and Others [2016] 4 All ER 907. The Insurance Fraud Register (IFR) has its own – more prolix – definition, which is available at www​.theifr​.org​.uk/​en/​faqs/​ 13 See Orakpo v Barclays Insurance Services [1995] Lloyd’s Rep IR 443. 14 The text of the 2015 Act is available at www​.legislation​.gov​.uk/​ukpga/​2015/​4/​contents/​enacted 15 Which is available at www​.theifr​.org​.uk/​en/​faqs/​ 16 As to the meaning of ‘dishonest’, see Ivey v Genting Casinos (UK) Ltd t/a Crockfords [2017] UKSC 67. As to the meaning of ‘reckless’ in this context, see Aviva Insurance Ltd v Brown [2012] Lloyd’s Rep IR 42 and Versloot.

Marine insurance fraud and emerging technology  277 obtaining money or goods from another party to whom, on the face of it, he has undertaken specific trade, transport and financial obligations’.17 Fraud against an insurance provider may be carried out by a variety of actors, on a variety of occasions in a variety of ways. In terms of actors, most frauds are perpetrated by policyholders,18 but it is not uncommon for third parties to be involved, either on their own behalf or as confederates of policyholders, and dishonest insurance professionals have also perpetrated frauds. In terms of timing, the fraud may be conducted prior to the inception of the policy, when cover is sought, and afterwards, when a claim is made. In terms of the ways in which insurance fraud may be conducted, these are myriad and limited only by the imagination of those malefactors who seek to obtain that to which they are not entitled. 2.1

Pre-contractual Fraud

Policyholders may commit fraud prior to inception by misrepresenting the magnitude of the risk in order to reduce the cost of the premium, a practice described as fraudulent under-insurance. It is, of course, quite possible for under-insurance to be the result of a honest mistake and such instances are usually addressed by section 81 Marine Insurance Act 1906 (MIA) in the case of marine or valued policies.19 Non-marine policies are not generally ‘subject to average’ unless there is an express condition to that effect,20 which can cause difficulties for insurers irrespective of the presence of fraud. Another type of fraud committed prior to or at the inception is for the prospective policyholder to seek cover for property (usually cargo, rather than vessels) which it does not own or which does not exist at the time the policy is incepted (that is, property in which it had no insurable interest). This is one of the ‘pernicious practices’ condemned in the preamble to the Marine Insurance Act 1745 (the 1745 Act).21 The mis-selling of inappropriate cover by an insurer or ‘ghost broking’ (that is, the marketing and sale of non-existent policies by individuals who are not brokers or insurers) are other categories of pre-inception fraud, albeit perpetrated against policyholders rather than by them.22

See Guide to Prevention of Maritime Fraud, ICC Publication No. 370, 3, 1980. This chapter uses the term ‘policyholder’ rather than insured or assured for the sake of clarity. 19 The text of the MIA 1906 is available at www​.legislation​.gov​.uk/​ukpga/​Edw7/​6/​41/​contents. S81 reads: ‘Effect of under insurance. Where the assured is insured for an amount less than the insurable value or, in the case of a valued policy, for an amount less than the policy valuation, he is deemed to be his own insurer in respect of the uninsured balance.’ 20 Joyce v Kennard (1871) LR 7 QB 78. 21 The preamble reads: ‘It has been found by experience, that the making of insurances, interest or no interest, or without further proof of interest than the policy, hath been productive of many pernicious practices, whereby great numbers of ships, with their cargoes, have […] been fraudulently lost or destroyed.’ 22 ‘Ghost broking’ is more prevalent in consumer insurance, such as home or motor cover, but it is a growing threat. See Z. Alexander, ‘Ghost Broking – Buyer Beware!’ (28 October 2020) available at https://​kennedyslaw​.com/​thought​-leadership/​blogs/​fraud​-blog​-fundamentally​-honest/ghost​-broking​ -buyer​-beware/​ 17 18

278  Research handbook on marine insurance law 2.2

Post-contractual Fraud: Fraudulent Claims

Whereas fraudulent under-insurance may be successfully undertaken without a claim ever being made on the policy, the ‘reward’ for obtaining cover of a risk in which the policyholder has no interest is contingent upon a claim. Consequently, it can be regarded as a form of fraudulent claim, of which Mance LJ (as he then was) provided the following categorisations in Agapitos v Agnew (‘The Aegeon’):23 1) Where there is no actual loss or the loss is the result of a deliberate act by the policyholder;24 2) Where the policyholder has suffered a loss but has exaggerated it when seeking recovery from the insurer; 3) Where policyholder, having apparently sustained a loss, subsequently finds that it has suffered no loss, or a smaller loss, but maintains its original claim nonetheless; 4) Where policyholder makes a claim against the insurer knowing that there is a defence to the claim under the policy; and 5) Where a genuine claim is made with the use of fraudulent means or devices.

The first of these categories includes those situations where the policyholder (or someone colluding with the policyholder) brings about the loss which is the subject-matter of the claim. This includes attempts (successful or otherwise) to scuttle a vessel as in Kairos Shipping Ltd and Another v Enka and Co Llc and Other, The Atlantik Confidence,25 Brillante Virtuoso and National Justice Compania Naviera SA v Prudential Assurance Co Ltd (‘The Ikarian Reefer’) (No.1).26 The second category involves those situations where a policyholder suffers a genuine loss but deliberately or recklessly exaggerates the claim. While this is a common phenomenon in motor vehicle and household insurance, the size and complexity of marine insurance claims – and the attendant resources employed in their investigation by insurers – is such that exaggerated claims are not as prevalent in relation to ships as they are in relation to, say, cars. It is also important to note that the fact that a claim may have been exaggerated does not automatically imply fraudulent intent. The courts are prepared to accept that policyholders may bolster their claims to some extent in the expectation that the insurers will try to undercut them. This reality was acknowledged by Hoffmann LJ (as he then was) in Orakpo v Barclays Insurance Services, who observed that ‘One should naturally not readily infer fraud from the fact that the insured has made a doubtful or even exaggerated claim’, albeit adding that where fraud has been proved ‘it should discharge the insurer from all liability’. That said, as held in Galloway v Guardian Royal Exchange (UK) Limited and elsewhere, if a claim is partly genuine and partly fraudulent, the latter’s blemish will taint the former.27 Earlier, unrelated claims will not, however, be so tainted.28

[2002] Lloyd’s Rep IR 573. Such a deliberate act would also fall foul of s 55 MIA 1906. 25 [2016] 2 Lloyd’s Rep 525. 26 [1995] 1 Lloyd’s Rep. 455. The first instance decision in The Ikarian Reefer is also notable for Cresswell J’s comments on the duties of expert witnesses: see National Justice Compania Naviera S A v Prudential Assurance Company Ltd (‘The Ikarian Reefer’) [1993] 2 Lloyd’s Rep 68. 27 [1999] Lloyd’s Rep IR 209. Also see Danepoint Ltd v Allied Underwriting Insurance Ltd [2005] EWHC 2318; [2006] Lloyd’s Rep IR 429. 28 Axa General Insurance Ltd v Gottlieb [2005] Lloyd’s Rep IR 369. 23 24

Marine insurance fraud and emerging technology  279 The third category is generally considered to be a variation on or subset of the second category, whereas the fourth category was identified by Mance LJ in The Aegeon thus: ‘[A] claim cannot be regarded as valid, if there is a known defence to it which the insured deliberately suppresses. To that extent, at least, fraud in relation to a defence would seem to me to fall within the fraudulent claim rule.’ The fifth category comprises those situations in which the policyholder has a genuine claim but employs fraudulent means or devices (such as forged documents) to advance it. Mance LJ maintained in The Aegon that the public policy objective of deterring insurance fraud justified the forfeiture of such claims. This clarity of reasoning was displaced by the Supreme Court in Versloot Dredging Bv and Another v Hdi Gerling Industrie Versicherung Ag and Others. Instead, Lord Sumption stated that ‘collateral lies’ which did not materially effect the essence of the claim should not lead to the failure of the claim and, accordingly, ‘[t]he extension of the fraudulent claims rule to lies which are found to be irrelevant to the recoverability of the claim is a step too far. It is disproportionately harsh to the insured and goes further than any legitimate commercial interest of the insurer can justify.’ Unsurprisingly, Lord Mance (as he by then was) dissented, remarking: ‘Abolishing the fraudulent devices rule means that claimants pursuing a bad, exaggerated or questionable claim can tell lies with virtual impunity.’ The need for such deterrence had been expressed previously by, among others, Millett LJ in Galloway and Lord Hobhouse in Manifest Shipping Co Ltd v Uni-Polaris Shipping Co Ltd (The Star Sea).29 Finally, there are the questions of materiality and inducement. There has been much judicial and academic debate over the years as to the nature of both in respect of insurance fraud and their relationship to the doctrine of utmost good faith. In Versloot, Lord Sumption remarked that the ‘test of materiality [within the doctrine of utmost good faith] cannot apply to lies told in the course of making a claim’. When it came to claims (fraudulent or not), the only question is whether or not the insurer has a legal liability to pay the policyholder, which is not determined by the insurer’s state of mind at the time. His Lordship concluded: ‘For this reason, although a lie uttered in support of a claim need not have any adverse impact on the insurer, I consider that it must at least go to the recoverability of the claim on the true facts.’ Lord Hughes added that an insurer is legally obligated to pay a claim if it is ‘good’: A lie told in the making of the claim may well affect his handling of the claim, or the speed at which he pays it, or the inquiries which he calls for, but it can make no difference to his liability to pay. It may well be material (relevant) to his behaviour, but it is immaterial (irrelevant) to his liability. So “materiality” means something different at the two stages.

The ‘different’ nature of materiality to which Lord Hughes referred would seem, on the face of it, to be clear, but that has not ended the debate over this or other aspects of the treatment of insurance fraud by the courts.30 Many insurers have sought to avoid this – to them – fruitless

[2001] Lloyd’s Rep IR 49. His Lordship remarked: ‘the logic is simple. The fraudulent insured must not be allowed to think: if the fraud is successful, then I will gain; if it is unsuccessful, I will lose nothing.’ 30 Among the many expert commentaries on Versloot are B. Soyer, ‘Lies, Collateral Lies and Insurance Claims: The Changing Landscape in Insurance Law’, 22(2) Edinburgh Law Review 237–65 (2018) and P.J. Rawlings and J.P. Lowry, ‘Insurance Fraud and the Role of the Civil Law’, 80(3) The Modern Law Review 525–39 (2017). 29

280  Research handbook on marine insurance law debate by including clauses within policies that give them the right to decline claims supported by ‘collateral lies’. Other clauses, tying cover to the use of telematics and disclosure of ‘real-time’ information, have also been employed by insurers in face of the courts’ apparent irresoluteness in the face of fraud. 2.3

Proving Fraud

The burden of proving that insurance has been improperly obtained or that a claim on a policy has been brought fraudulently lies upon the insurer (or whoever else makes the allegation). Indeed, as Eder J remarked in Otkritie International v Urumo, ‘It is, of course, trite law that the burden of proof lies on the claimants’.31 Snowden J elaborated in Bilta (UK) Limited (In Liquidation) v (1) NatWest Markets PLC and (2) Mercuria Energy Europe Trading Limited32 thus: The first [principle] is that the burden of proving dishonesty lies upon the claimant. The second is that the standard of proof is the usual civil standard on the balance of probabilities - i.e. it must be established that the fact or conduct in dispute more probably happened than not. Thirdly, there is no rule of law that if the allegation is more serious (e.g. of dishonesty) more cogent evidence is required to prove it. Fourth, the court can take into account, as a matter of common sense, any relevant “inherent improbabilities” as to the defendant’s behaviour.

In doing so Snowden J was drawing upon the decisions of the (then) House of Lords in Re H (Minors) (Sexual Abuse; Standard of Proof)33 and Re B (Children) (Sexual Abuse: Standard of Proof).34 In particular, there is only one civil standard of proof, which is the straightforward balance of probabilities test. Indeed, as Lady Hale pointed out, the ‘inherent probabilities’ of something having taken place should be taken into account, where relevant, but ‘there is no logical or necessary connection between seriousness and probability’. This approach was reiterated by the Court of Appeal in Bank St Petersburg PJSC v Arkhangelsky35 in which Sir Geoffrey Vos C echoed the words of the House of Lords in Re H and Re B in that the standard of proof in civil proceedings was the balance of probabilities and that this was not altered by the seriousness of the allegations nor the seriousness of the consequences in a particular case. The learned judge further observed that ‘in commercial cases, there will be a wide spectrum of probabilities as to the occurrence of reprehensible conduct’. The trial judge had had failed to look at the matter holistically and considered how his findings on one point effected how he should address other, related points or the matter as a whole. Consequently, the Court of Appeal remitted the matter for retrial. The trial judge in Bank St Petersburg PJSC had, however, stated correctly that ‘cogent evidence is required to justify a finding of fraud or other discreditable conduct’.36 Unfortunately, [2014] EWHC 191. [2020] EWHC 546 (Ch). Snowden J’s decision in this case was overturned by the Court of Appeal and the matter remitted for retrial in Bilta v Natwest Markets Plc [2021] EWCA Civ 680 but his comments on the burden and standard of proof in dishonesty claims remain valid. 33 [1996] AC 563. 34 [2009] 1 AC 11. 35 [2020] 4 WLR 55. 36 The learned judge cited Andrew Smith J in Fiona Trust v Privalov [2010] EWHC 3199 at 1438, who in turn had cited per Moore-Bick LJ in Jafari-Fini v Skillglass Ltd [2007] EWCA Civ 261 at para 73. 31 32

Marine insurance fraud and emerging technology  281 it can be very difficult to obtain cogent evidence in marine fraud cases, especially direct documentary evidence. The importance of obtaining cogent evidence was touched upon in Brillante Virtuoso, where Teare J remarked that scuttling was a serious allegation and that the facts proved against an owner must be ‘sufficiently unambiguous’37 to establish that an owner of a vessel was complicit in its loss. As will be seen below, the learned judge held that facts of that case were ‘sufficiently unambiguous’ to justify his conclusion that the owner was indeed responsible for the Brillante Virtuoso’s loss. Ultimately, the court may find itself having to assess the merits of a claim of fraud against the policyholder on the basis of witness testimony. In such circumstances, the observation by Robert Goff LJ (as he then was) in The Ocean Frost38 is often applied by judges: It is frequently very difficult to tell whether a witness is telling the truth or not; and where there is a conflict of evidence such as there was in the present case, reference to the objective facts and documents, to the witnesses’ motives, and to the overall probabilities, can be of very great assistance to a Judge in ascertaining the truth.

One of the motivations behind the development of insurtech is the elimination of the need to rely on humans, not only in terms of assessing applications for cover and claims but also in terms of obtaining and collating information on policyholders and risks, some of which may be submitted as evidence in court. Moreover, the need for judges to rely on their experience to determine who is lying to them (and, by extension, the insurers) may be reduced. Such enhanced information gathering and analysis is also useful to insurers when bringing claims of fraud in the first place, rather than having to rely on other causes of action. In this respect, it should be appreciated that fraud litigation is fraught with practical difficulties, from the procedural requirements39 to the obligations upon the parties and their advisers.40 Insofar as those other causes of action are concerned, the 1745 Act and MIA legislated for the need for policyholders to have an insurable interest in the subject-matter of a policy in order, among other things, to deter fraud. There have been numerous instances where insurers have declined claims because of an absence of insurable interest, albeit the true reason may have been a suspicion of fraud – not least Macaura v Northern Assurance Co Ltd.41 Indeed, the ABI submitted to the Law Commission’s Insurance Contract Law review that insurable interest should remain part of the law because it guards against moral hazard and invalid claims.42

37 As per Aikens J in Brownsville Holdings Ltd v Adamjee Insurance Co. (The Milasan) [2000] 2 Lloyd’s Reports 458. 38 Armagas Ltd v Mundogas S.A. (The Ocean Frost) [1985] 1 Lloyd’s Rep 1. 39 Pursuant to CPR Practice Direction 16 para 8.2, allegation of fraud must be ‘specifically set out […] in the particulars of claim’. Available at www​.justice​.gov​.uk/​courts/​procedure​-rules/​civil/​rules/​ part16/​pd​_part16​#8. 40 In England and Wales, Rule C9.2 of the Bar Standards Board Code of Conduct states that a barrister ‘must not draft any statement of case […] or other document containing: […] any allegation of fraud, unless you have clear instructions to allege fraud and you have reasonably credible material which establishes an arguable case of fraud’. Available at www​.barstandardsboard​.org​.uk/​the​-bsb​-handbook​ .html. There are similar professional requirements placed upon lawyers in other jurisdictions. 41 [1925] AC 619. 42 See Law Commission ‘Insurance Contract Law: Insurable Interest’, available at www​.lawcom​.gov​ .uk/​project/​insurance​-contract​-law​-insurable​-interest/​

282  Research handbook on marine insurance law The difficulty of bringing complex, risky and expensive claims for marine insurance fraud is therefore avoided – but not the existence of the fraud itself.43 2.4

The Consequences of Marine Insurance Fraud

There are effectively three potential remedies available to insurers for marine insurance fraud: (1) forfeiture (at common law or pursuant to the 2015 Act); (2) action pursuant to a fraudulent claims clause; and (3) a claim for tortious deceit.44

As noted above, the common law position is that a claim that is fraudulent in whole or in part is forfeited.45 The insurer may also terminate the policy but the policyholder does not forfeit any moneys already paid out in satisfaction of legitimate claims. Unfortunately, the jurisprudential basis of this position has been a matter of some dispute. In Orakpo v Barclays Insurance Services the Court of Appeal held that it was an implied term of the policy which was incorporated to give effect to the doctrine of utmost good faith. This approach was endorsed in Black King Shipping Corp v Massie (The Litsion Pride),46 but was overturned by the House of Lords in The Star Sea, where it was emphasised that there was no relationship between the fraudulent claim rule and the doctrine of utmost good faith. In Agapitos v Agnew Mance LJ stated that the fraudulent claim rule did not arise out of the doctrine of utmost good faith and reiterated this point in Axa General Insurance Ltd v Gottlieb. The 2015 Act addresses the remedies for fraudulent claims and clarifies the relationship – or rather the absence of a relationship – between the fraudulent claims rule and the doctrine of utmost good faith. Section 12 of the 2015 Act provides that, in the event of a fraudulent claim, the insurer is not liable to pay the claim; may recover from the policyholder any sums it may have paid in respect of the claim; and treat the policy as having been terminated from the time of the fraudulent act.47 In the event that the insurer elects to treat the policy as terminated, it may refuse to pay claims relating to ‘relevant events’48 occurring after the fraudulent act and may

Soyer also notes this problem in Marine Insurance Fraud paras 1-15 to 1-20. By contrast, Hjalmarsson argues that ‘the law of England and Wales on insurance fraud, from a mostly unregulated state in the 19th century, has, with few exceptions, followed the trajectory posited by Galanter in incrementally nudging the law in the direction favoured by insurers’: J. Hjalmarsson, ‘Insurers and the Law of Fraud: A Success Story and the Case for Regulatory Intervention’ in D. Rhidian Thomas, The Modern Law of Marine Insurance vol. 5 (Routledge, 2023), Chapter 6. 44 There is also evidence to suggest that insurers have also been more inclined to bring private prosecutions against fraudsters in recent years, albeit the number of such actions is small. See A. Wilkinson and D. Cole, ‘Tackling Fraudsters: Private Prosecutions and Contempt’, 2 J.P.I. Law 116–22 (2021). 45 Levy v Baillie (1831) 7 Bing 349; Britton v Royal Insurance Co (1866) 4 F&F 905. 46 [1985] 1 Lloyds Rep 437. 47 A ‘fraudulent claim’ is not the same as a ‘fraudulent act’: the latter is the conduct that makes a claim fraudulent, which may follow the submission of the claim (that is, where a policyholder finds that it has suffered no loss but maintains its original claim). 48 A ‘relevant event’ is defined as whatever gives rise to the insurer’s liability under the contract (and includes, for example, the occurrence of a loss, the making of a claim or the notification of a potential claim, depending on how the contract is written). 43

Marine insurance fraud and emerging technology  283 retain any premiums received from the policyholder.49 The insurer remains liable, however, for any relevant events prior to the policyholder’s (or an appropriate third party’s) fraudulent act. Section 13 of the 2015 Act applies the provisions in section 12 to group insurance. Finally, section 14(1) adds ‘Any rule of law permitting a party to a contract of insurance to avoid the contract on the ground that the utmost good faith has not been observed by the other party is abolished’.50 This provision goes beyond severing any possible link between the fraudulent claims rule and the doctrine of utmost good faith that was suggested in Orakpo v Barclays Insurance Services. It effectively strips the doctrine of any sanction, the related provisions in sections 18 to 20 of the MIA having been substituted by the provisions on disclosure and fair representation in the Consumer Insurance (Disclosure and Representations) Act 2012 and the 2015 Act.51 In light of the confusion created by the conflicting judicial decisions on the remedies available to them, not least as to whether a policy should be treated void ab initio and the status of legitimate claims, it is hardly surprising that many insurers include fraudulent claims clauses in their policy wordings. Such clauses typically set out what is meant (for the purposes of the policy) by the term ‘fraudulent claim’ and the remedies of which the insurer may avail itself in the event of such a claim.52 Unfortunately, many such clauses fail to address even these basic requirements, leading to litigation such as that in Insurance Corporation of the Channel Islands v McHugh.53 Finally, the tort of deceit may provide an alternative remedy, in the form of damages. Insurers may be able to recover monies expended in their efforts to deal with and investigate the claim, including any expenditure on third party consultants, and any sum paid to (incorrectly) indemnify the policyholder for its (non-existent) losses. They may also be awarded aggravated or exemplary damages in accordance with the principles laid down by Lord Devlin in Rookes v Barnard.54 In Axa Insurance UK Plc v Financial Claims Solutions Ltd55 the Court of Appeal awarded an insurer exemplary damages against parties which had staged fictitious motor accidents. Flaux LJ observed that exemplary damages ‘remain anomalous and the exception to the general rule’ of damages as compensation but added that this case was a ‘paradigm for such an award given that the fraudsters’ had been ‘calculated to make a profit for [themselves] which may well exceed the compensation payable to the [insurer]’. It is arguable, and indeed it has been argued, that these remedies are inadequate.56 In particular, common law or statutory forfeiture causes little distress for the policyholder that In the case of ‘claims made’ policies, such ‘relevant events’ may be the notification of a claim against the policyholder, even where no loss has occurred at that point. 50 The remainder of s 14 makes consequential amendments to s 17 of the MIA and s 2 of the Consumer Insurance (Disclosure and Representations) Act 2012. 51 The text of the 2012 Act is available at www​.legislation​.gov​.uk/​ukpga/​2012/​6/​contents/​enacted/​ data​.htm. There is a comparative discussion of the state of the doctrine in the US and UK in A.M. Costabel, ‘Utmost Good Faith in Marine Insurance: The Reports of My Death Have Been Greatly Exaggerated’, 46 Tul. Mar. L.J. 115 (2022). 52 See Lloyd’s Market Association’s model clause LMA5256, ‘Insurance Act 2015 – Fraudulent Claims Clause’ available at www​.lmalloyds​.com/​actclauses 53 [1997] LRLR 94. 54 [1964] AC 1129. 55 [2018] EWCA Civ 1330. 56 See K. Richards, ‘Time’s Up for Wholly Fraudulent Insurance Claims: The Case for New Statutory Remedies’, 7 JBL 580 (2020). 49

284  Research handbook on marine insurance law has brought a wholly fraudulent claim. The loss of the premium may be treated by it as an ‘occupational hazard’ or ‘opportunity cost’ of its criminal enterprise, which can be weighed against the potentially lucrative gains to be made for a successful fraudulent claim. Insurers may include various ‘anti-fraud’ clauses in their policies but these must be carefully written to avoid dissuading (honest) policyholders and being struck down by the courts on any number of contractual grounds, not least the rule against penalty clauses.57 While the decision in Axa v FCS in respect of exemplary damages is welcome (from the point of view of insurers), the fact that Flaux LJ referred to them as ‘anomalous’ and declined to expand upon the principles laid down by Lord Devlin in Rookes v Barnard should not be overlooked.58 Ultimately, the desirability of a technological solution to marine insurance fraud becomes self-evident when one reflects upon the jurisprudential problems, procedural difficulties, evidential burdens and pyrrhic victories (in terms of financial redress) offered to insurers seeking a civil law solution to the problem. The move to ‘self-help’ can also be explained by the widespread perception in the insurance industry and beyond that those responsible for enforcing the criminal law vis-à-vis fraud have neglected to do so.59 In May 2023, the UK Government published ‘Fraud Strategy: Stopping Scams and Protecting the Public’, in which it proclaimed that ‘Our ambition is to cut fraud by 10% from 2019 levels, down to 3.33 million frauds by the end of this Parliament’.60 The scope of this ‘ambition’ should be evaluated in light of the fact that the number of fraud offences reported to the UK authorities increased by 17 per cent between 2014 and 2019 before growing by a further 50 per cent by 2022.61

3.

EMERGING TECHNOLOGY

Emerging technology is being applied to various fields of human endeavour including insurance and marine transport. ‘Insurtech’, a portmanteau of ‘insurance’ and ‘technology’, describes the development and application of emerging technology to the creation, distribution and administration of insurance products and business.62 It is also applied to technology-led companies that have entered the insurance market.63 For the avoidance of confusion, however, references to ‘insurtech’ in this chapter are references to the technology rather than those using it. While there is no comparable term for their use in marine transport (such as ‘marinetech’), Cavendish Square Holding BV v Talal El Makdessi and ParkingEye Ltd v Beavis [2015] UKSC

57

67.

58 See E. Katsampouka, ‘Exemplary Damages and Insurance Fraud’, 135(Jul) L.Q.R. 380–5 (2019), and F. Alexandrakis, ‘The Remedy of Exemplary Damages In Case of Fraudulent Insurance Claims’, 1 J.B.L. 1–21 (2023). 59 C. Hymas, ‘Police have “lost interest in catching fraudsters”’ (The Daily Telegraph, 4 January 2023). 60 The strategy document is available at https://​assets​.publishing​.service​.gov​.uk/​government/​uploads/​ system/​uploads/​attachment​_data/​file/​1154660/​Fraud​_Strategy​_2023​.pdf 61 See the ONS report ‘Nature of Fraud and Computer Misuse in England and Wales: Year Ending March 2022 available at www​.ons​.gov​.uk/​peoplepop​ulationand​community/​crimeandjustice/​articles/​ natureof​fraudandco​mputermisu​seinenglan​dandwales/​yearendingmarch2022 62 ‘Insurtech’ is, depending on one’s point of view, part of or a counterpart to ‘Fintech’. See J. Madir, FinTech: Law and Regulation (Edward Elgar Publishing, 2021). 63 There are a number of media websites dedicated to coverage of insurtech technology and companies, such as The Digital Insurer, available at www​.the​-digital​-insurer​.com/​

Marine insurance fraud and emerging technology  285 many of the same emerging technologies are also being utilised by ship owners, agents, port authorities and others. There follows a brief review of some of the more significant emerging technologies and their application to both insurance and marine transport. 3.1 Blockchain IBM, the proverbial ‘Big Blue’, defined blockchain as ‘a shared, immutable ledger that facilitates the process of recording transactions and tracking assets in a business network’.64 Iansiti and Lakhani defined blockchain as ‘an open, distributed ledger that can record transactions between two parties efficiently and in a verifiable and permanent way. The ledger itself can also be programmed to trigger transactions automatically.’65 Mainelli and Manson gave a lengthier description: A mutual distributed ledger – more simply referred to as a blockchain – is a computer data structure with the following capabilities: • Mutual – blockchains are shared across organisations, owned equally by all and dominated by no-one; • Distributed – blockchains are inherently multi-locational data structures and any user can keep his or her own copy, thus providing resilience and robustness; • Ledger – blockchains are immutable, once a transaction is written it cannot be erased and, along with multiple copies, this means that the ledger’s integrity can easily be proven.66

The crucial aspect of blockchain is that it is a form of distributed ledger technology (DLT), in which ‘blocks’ of information are used to record transactions in a distributed network of computers (also referred to as ‘nodes’). For example, a participant in the network may seek to inform another of the arrival of a cargo at its destination. This is notified to the network (via its node or some other conduit), which in turn validates the action in accordance with a pre-agreed ‘consensus’ mechanism. When each participating ‘node’ has validated the action, a new ‘block’ will be added to the ‘chain’. Each such transaction ‘block’ is also publicly time-stamped and linked to the previous block in the ‘blockchain’. The block cannot be retroactively altered or deleted (that is, it is immutable) but the chain may be updated (that is, new blocks added) and synchronised across the network. A blockchain may be ‘open’ or ‘public’, in which case anyone with a computer can examine it, or it may be ‘private’, with only authorised individuals having access. Blockchains may also be ‘permissionless’, by which anyone can transmit and validate transactions. On the other hand, they may operate as a ‘permissioned ledger’, with only certain participants being able to

M. Gupta, Blockchain for Dummies®, 2nd IBM Limited Edition (John Wiley & Sons, Inc., 2018). Despite its title, this is a useful introduction to the subject for the non-technologist before one moves onto more detailed texts addressing, for example, the legal aspects of blockchain. 65 See M. Iansiti and K.R. Lakhani, ‘The Truth about Blockchain’ 95 Harvard Business Review 118 (2017) available at https://​hbr​.org/​2017/​01/​the​-truth​-about​-blockchain 66 See M. Mainelli and B. Manson, ‘Chain Reaction: How Blockchain Technology Might Transform Wholesale Insurance’ (PwC and Z/Yen Group, Long Finance, 2016) available at www​.longfinance​.net/​ publications/​longfinance​-reports/​chain​-reaction​-how​-blockchain​-technology​-might​-transform​-wholesale​ -insurance/​ 64

286  Research handbook on marine insurance law add to or manage the blockchain and others merely being able to view its contents. 67 Clearly, there will be circumstances in which only certain participants should be allowed to add to the blockchain but this does not detract from the general principle that the integrity of the blockchain is dependent upon its participants rather than external ‘arbiters’ or ‘authorities’.68 A growing number of shipping companies and associated businesses are employing blockchain in their operations.69 The main advantages of the technology are the digitalisation of paperwork (it is estimated that upwards of 15 per cent of transportation costs in the container shipping industry are due to ‘paperwork’),70 the tracking and tracing of vessels and cargo, customs and combating fraud.71 In 2018, Maersk and IBM created ‘Tradelens’, by which shipping and freight operators could validate the transaction of goods on a digital ledger.72 Unfortunately, its creators announced late in 2022 that they intended to wind down the platform as it apparently had failed to attract a sufficient number of users.73 Nevertheless, the use of emerging technologies continues, with the ‘Global Shipping Business Network’ (GSBN) and ‘Container xChange’ being among those developing and utilising DLT in marine transportation.74

Many people think of blockchain solely in terms of cryptocurrencies. Intriguingly, the original ‘white paper’ on the latter, entitled ‘Bitcoin: A Peer-to-Peer Electronic Cash System’, by S. Nakamoto (who may or may not actually exist), does not actually use the term blockchain: available at https://​ bitcoin​.modeapp​.com/​bitcoin​-white​-paper​.pdf 68 There is a very large and growing volume of academic and professional literature on blockchain technology, its use in business and commerce (and elsewhere) and the law relating to the same. In addition to those texts already cited, by way of introduction, see J. Golden et al. (ed) Blockchain Technology: Fundamentals, Applications, and Case Studies (CRC Press, 2020); D.A. Tran et al. (ed) Handbook on Blockchain (Springer, 2022); P. De Filippi and A. Wright, Blockchain and the Law: The Rule of Code Copyright Date (Harvard University Press, 2018); and A. Grinhaus, A Practical Guide to Smart Contracts and Blockchain Law (2nd ed, LexisNexis Canada, 2022). The Stanford Journal of Blockchain Law and Policy is available at https://​stanford​-jblp​.pubpub​.org/​ and the UCL Centre for Blockchain Technologies website is available at http://​blockchain​.cs​.ucl​.ac​.uk/​. The Global Blockchain Business Council publishes ‘The International Journal of Blockchain Law’, available at https://​gbbcouncil​ .org/​. The Legal 500 also publishes a guide to blockchain law and practice across a number of major jurisdictions, authored by leading practitioners, which is available at www​.legal500​.com/​guides/​guide/​ blockchain/​ 69 Texts relating to emerging technologies, including blockchain, and the marine transport industry include B. Soyer et al., New Technologies, Artificial Intelligence and Shipping Law in the 21st Century (Routledge, 2020); M.K. Proshanto et al., Maritime Law in Motion (Springer International Publishing, 2020); and J. Kraska and Y-K. Park, Emerging Technology and the Law of the Sea (CUP, 2022). 70 See www​.forbes​.com/​sites/​tomgroenfeldt/​2017/​03/​05/​ibm​-and​-maersk​-apply​-blockchain​-to​ -container​-shipping/​?sh​=​335f0d833f05 71 Chung-Shan Yang, ‘Maritime Shipping Digitalization: Blockchain-based Technology Applications, Future Improvements, and Intention to Use’, 131 Transportation Research Part E: Logistics and Transportation Review 108–17 (2019); M. Jović, M. Filipović, E. Tijan and M. Jardas, ‘Review of Blockchain Technology Implementation in Shipping Industry’, 33 Scientific Journal of Maritime Research 140–8 (2019). 72 See Maersk, ‘A game changer for global trade’ (20 September 2019) available at www​.maersk​ .com/​news/​articles/​2019/​09/​20/​a​-game​-changer​-for​-global​-trade 73 See Maersk, ‘A.P. Moller-Maersk and IBM to discontinue TradeLens, a blockchain-enabled global trade platform’ (29 November 2022) available at www​.maersk​.com/​news/​articles/​2022/​11/​29/​ maersk​-and​-ibm​-to​-discontinue​-tradelens 74 See GSBN’s website available at www​.gsbn​.trade/​ and XChange, ‘Blockchain shipping: what is it and why is it important?’ [2022] (24 March 2022) available at www​.container​-xchange​.com/​blog/​ 67

Marine insurance fraud and emerging technology  287 Similarly, insurers operate their own blockchain ledgers, in conjunction with brokers and others, to streamline both the underwriting and claims handling processes in addition to having access to the ledgers of their shipping company policyholders. The first blockchain-enabled marine insurance system or platform was ‘Insurwave’, a joint venture between EY and Guardtime which was established in 2018.75 Such blockchain ledgers improve the efficiency of the respective processes (not least in the saving of time), reduce the number of intermediaries involved and diminish the potential for human error.76 3.2

Smart Contracts

Nick Szabo first used the term ‘smart contract’ in the mid-1990s: A smart contract is a computerized transaction protocol that executes the terms of a contract. The general objectives of smart-contract design are to satisfy common contractual conditions (such as payment terms, liens, confidentiality, and even enforcement), minimise exceptions both malicious and accidental, and minimise the need for trusted intermediaries. Related economic goals include lowering fraud loss, arbitration and enforcement costs, and other transaction costs.77

He went on to expound that it was ‘A set of promises specified in digital form, including protocols within which the parties perform on these promises’.78 The, now classic, example Szabo gave of a ‘primitive ancestor’ of a smart contract was the ‘humble vending machine’, which accepts people’s coins and, in return, dispenses the product which they have selected in an entirely automated process.79 Szabo explained that smart contracts went ‘beyond the vending machine’ to encompass agreements relating to ‘all sorts of property that is valuable and controlled by digital means’. It is important to appreciate that Szabo’s is not the only definition of a smart contract. Moreover, while his premise, and the vending machines, predated the emergence of blockchain, it has now become almost synonymous with it.80 It is also recognised that not all aspects

blockchain​-shipping/​#:​~:​text​=​Blockchain​%20shipping​%20improves​%20efficiency​,payments​%20faster​ %20and​%20more​%20secure 75 See T. Roughton ‘“World first” marine insurance blockchain platform welcomed’ (OUT-LAW NEWS, 15 Jun 2018) available at www​.pinsentmasons​.com/​out​-law/​news/​world​-first​-marine​-insurance​ -blockchain​-platform; and ‘Insurwave: blockchain-enabled marine insurance’, a description of the system on EY’s own website, available at www​.ey​.com/​en​_gl/​insurance/​blockchain​-marine​-insurance 76 See A.D. Shetty et al., ‘Block Chain Application in Insurance Services: A Systematic Review of the Evidence’, 12(1) SAGE Open, January 2022; A.A. Amponsah and B.A. Weyori, ‘Block Chain in Insurance: Exploratory Analysis of Prospects and Threats’, 12(1) IJACSA (2021); and D. Popovic et al., ‘Understanding Blockchain for Insurance Use Cases’, 25(e12) British Actuarial Journal 1–23 (2020). 77 N. Szabo, ‘Smart Contracts’ (1994) available at www.fon.hum.uva.nl/rob/Courses/​Information​ InSpeech/​CDROM/Literature/​LOTwinterschool2006/​szabo​.best​.vwh​.net/​smart​.contracts​.html 78 N. Szabo, ‘Smart Contracts: Building Blocks for Digital Markets’ (1996) available at www​.alamut​ .com/​subj/​economics/​nick​_szabo/​smartContracts​.html 79 Some commentators question whether it is the contractual or sales process that is automated. They suggest instead that only the ‘vending’ process is automated. 80 Smart contracts can be, and commonly are, used in a distributed ledger, with the parties’ contractual obligations expressed in the relevant software and performed automatically by computers. The execution or fulfilment of the parties’ legal obligations takes place without the need for further human intervention.

288  Research handbook on marine insurance law of a smart contract will be free of human participation and that there is a distinction between ‘smart contract code’ and ‘smart legal contracts’. The former comprises the actual software, which may be a form of DLT, with the latter being legal contracts expressed in the form of that software.81 Although many involved in the development of smart contracts anticipated that they would encompass the legal relationship between the contracting parties, it is now accepted that this may not always be the case. Consequently, Clack, Bakshi and Braine produced the following definition to cover these disparate aspects: ‘A smart contract is an automatable and enforceable agreement. Automatable by computer, although some parts may require human input and control. Enforceable either by legal enforcement of rights and obligations or via tamper-proof execution of computer code.’82 In November 2021, the Law Commission published its advice to the UK Government on the nature of smart legal contracts.83 It concluded that the legal system in England and Wales ‘is clearly able to facilitate and support the use of smart legal contracts, without the need for statutory law reform’. In the process of doing so, it drew a distinction – touched upon by Clack and his colleagues – between smart contracts and smart legal contracts. The former are ‘Computer code that, upon the occurrence of a specified condition or conditions, is capable of running automatically according to pre-specified functions’, whereas the latter are A legally binding contract in which some or all of the contractual terms are defined in and/or performed automatically by a computer program. There are essentially three forms a smart legal contract can take, depending on the role played by the code. These are: • natural language contract with automated performance; • hybrid contract; or • solely code contract.

The first ‘form’ was described as ‘a natural language contract in which some or all of the contractual obligations are performed automatically by the code of a computer program’. The code or software does not ‘define any contractual obligations’ and is used only to carry them out. The Law Commission observed that these appeared to be most commonly used form of smart legal contract and ‘[did] not raise any novel legal issues’. The second ‘form’ expressed the parties’ rights and obligations in either or both natural language and computer code. The third form expresses and performs the contractual terms ‘automatically by […] the code of a computer program’. Such forms have no natural language version and, consequently, ‘[present] the most challenges from a contract law perspective, in terms of determining whether and when a legal contract is formed, and how that contract can be interpreted’. In terms of marine transportation, a smart contract (within a blockchain) might, as the Law Commission suggested, provide:

It is important to appreciate that blockchains and DLTs are not the only facilitators of smart contracts. For example, there has been a debate over whether Guardtime’s ‘Keyless Signature Infrastructure’ (KIS) differs from DLT or blockchain technology. 82 C.D. Clack, A.V. Bakshi and L. Braine, ‘Smart Contract Templates: Foundations, Design Landscape, and Research Directions’ (15 March 2017), arXiv:1608.00771v3, 2 available at www​ .researchgate​.net/​publication/​305779577​_Smart​_Contract​_Templates​_foundations​_design​_landscape​ _and​_research​_directions​_CDClack​_VABakshi​_and​_LBraine​_arxiv160800771​_2016 83 See Law Commission, ‘Smart Legal Contracts: Advice to Government’ (November 2021) available at www​.lawcom​.gov​.uk/​project/​smart​-contracts/​ 81

Marine insurance fraud and emerging technology  289 Alice (as seller of the goods) is bound to load the goods onto a ship nominated by Bob (as buyer of the goods). It is a term of the contract that Bob must nominate the ship suitable for loading by a certain date. Once Alice receives notification of the relevant ship for loading, Alice will load the goods onto the ship. Upon submission of proof that the goods have been loaded, the smart legal contract will automatically transfer the purchase price to Alice.

These terms are expressed in natural language here but they could be expressed in the form of a code (alone or in conjunction with natural language) in a smart legal contract. Such terms and others, such as a provision for demurrage, are relatively easy to translate into code. The difficulty arises, as acknowledged by the Law Commission, when one party decides – for commercial reasons – not to insist on the performance of a particular term or if the term itself provides for flexibility in its performance or if performance has been frustrated. A code may not be able to cover such behaviour. Nevertheless, despite such concerns, the use of smart contracts in marine transportation (based on DLT or otherwise) is growing. Although Tradelens may be defunct, other platforms such as Blockfreight continue in operation.84 Similarly, Guardtime’s Insurwave is only one of a number of smart contract operations in the field of marine insurance.85 The advantages to both those involved in the marine transport and insurance industries include transparency, the automation of tasks (that is, eliminating incompetent or dishonest humans), claim verification (for both insurers and policyholders), the digitisation of policy documents and the use of sophisticated risk assessment models.86 3.3

The Internet of Things

The term ‘Internet of Things’ (IoT) was first used by British technologist Kevin Ashton in a presentation he gave in 1999. In 2009, he added: If we had computers that knew everything there was to know about things – using data they gathered without any help from us – we would be able to track and count every thing, and greatly reduce waste, loss, and cost. We would know when things needed replacing, repairing, or recalling, and whether they were fresh or past their best.

Ashton continued: ‘We need to empower computers with their own means of gathering information, so they can see, hear and smell the world for themselves, in all its random glory. RFID and sensor technology enable computers to observe, identify and understand the world – without the limitations of human-entered data.’87

The Blockfreight website is available at https://​blockfreight​.com/​ For example, Tokio Marine is actively engaged in developing various insurtech initiatives, see BeInsure, ‘Japanese insurer Tokio Marine and Arbor Ventures building future of insurtech & digital transformation’ (12 August 2022) available at https://​beinsure​.com/​news/​tokio​-marine​-insurance​-and​ -arbor​-ventures​-building​-future​-of​-insurtech/​ 86 See Lloyd’s Risk Report ‘Triggering Innovation: How Smart Contracts Bring Policies to Life’ (2019) available at www​.lloyds​.com/​news​-and​-insights/​risk​-reports/​library/​triggering​-innovation; and A. Borselli, ‘Smart Contracts in Insurance: A Law and Futurology Perspective’ in P. Marano and K. Noussia (eds), InsurTech: A Legal and Regulatory View (Springer, 2020), 101–25. 87 See K. Ashton, ‘That “Internet of Things” Thing’ (RFID Journal, 22 June 2009) available at www​ .itrco​.jp/​libraries/​RFIDjournal​-That​%20Internet​%20of​%20Things​%20Thing​.pdf 84 85

290  Research handbook on marine insurance law As with ‘blockchain’, others have sought to define IoT. The OECD (citing a report by McKinsey) stated: ‘The Internet of Things (IoT) is when sensors and actuators embedded in physical objects – from roadways to pacemakers – are linked through wired and wireless networks, often using the same Internet Protocol (IP) that connects the Internet. The connection permits large volumes of data to flow to computers for analysis.’88 Insofar as what the IoT actually does, McKinsey identified six applications, which could be divided into two broad categories: 1. Information and analysis – a. tracking behaviour (of people, data or things); b. enhanced situational awareness (of the physical environment); and c. sensor-driven decision analytics 2. Automation and control – a. process optimisation (automated controls); b. optimised resource consumption (by smart meters etc); and c. complex autonomous systems (such as autonomous vehicles).89 In terms of their relevance to marine transport, the tracking of cargo by the use of sensors on a ‘smart container’ or upon the individual items within is a clear example of 1(a) above. Similarly, damage to cargo during the voyage caused by, say, environmental condition, as in Noten BV v Harding,90 may be detected and possibly prevented by the presence of sensors within the vicinity of the cargo as per 1(b) above. Such is the potential for the IoT to marine transport that some commentators have spoken of the ‘Internet of Ships’.91 At the highest level, 2(c), there are Maritime Autonomous Surface Ships (MASSs), which are able, to a varying degree, to operate independently of human intervention. The extent of their independence ranges from partially automated vessels – whereby the crew operate some shipboard systems with others being automated – to fully autonomous vessels which can make decisions and determine actions by themselves. In 2019, the 3200-ton Yara Birkeland (reputedly the first all-electric autonomous cargo vessel) travelled 43 miles from Horten to Oslo in Norway.92 In September of the same year, the 70,000-ton Iris Leader made a three-day voyage from Xinsha, China, to Nagoya, Japan.93 In 2022, an autonomous 750-gross-ton commercial cargo ship completed a 500-mile voyage in the heavily trafficked waters of Tokyo Bay, oper-

OECD (2017), Technology and Innovation in the Insurance Sector available at www.oecd​ .org/​pensions/​Technology​-and​-innovation​-in​-the​-insurance​-sector​.pdf. The report the OECD cited is McKinsey, The Internet of Things (March 2010) available at www​.mckinsey​.com/​industries/​high​-tech/​ our​-insights/​the​-internet​-of​-things 89 While others have produced differing analyses, this one serves the purposes of this chapter. 90 [1989] 2 Lloyd’s Rep 527. 91 See S. Aslam et al., ‘Internet of Ships: A Survey on Architectures, Emerging Applications, and Challenges’, IEEE Internet of Things Journal (May 2020) DOI:10.1109/JIOT.2020.2993411. 92 See N. Lavars, ‘World’s first electric autonomous cargo ship takes to the water’ (21 November 2021) available at https://​newatlas​.com/​marine/​worlds​-first​-electric​-autonomous​-cargo​-ship​-yara​ -birkeland​-debut/​ 93 See ‘NYK conducts world’s first maritime autonomous surface ships trial’ (30 September 2019) available at www​.nyk​.com/​english/​news/​2019/​20190930​_01​.html 88

Marine insurance fraud and emerging technology  291 ating without human intervention for 99 per cent of its journey.94 Such is the speed of progress in this field that the International Maritime Organisation is developing a MASS Code for implementation in 2028.95 In terms of insurance, the true value of IoT is in terms of the information which it can produce, manifesting the ‘information and analysis’ side of McKinsey’s categorisations of IoT. This information can facilitate loss prediction and prevention, risk monitoring and claims handling. Concirrus and Parsyl are among those insurance providers which have made use of IoT in this regard.96 In the view of the Geneva Association,97 the growth of IoT not only ‘improves risk assessment and transfer, it creates the potential to predict and prevent risks, as well as offer wider insurance coverage’.98 3.4

Big Data and AI

As with ‘Internet of Things’, the expression ‘Big Data’ has been in use since the late 1990s, although some claim that it was common parlance long before then.99 In 2001, Laney stated that Big Data possessed three characteristics (which he termed the ‘3Vs’): 1. volume (it is vast); 2. velocity (it is generated in real-time); and 3. variety (it may be structured, semi-structured and/or unstructured).100 As with IoT, others have added their own definitions, with Uprichard having identified ‘versatility, volatility, virtuosity, vitality, visionary, vigour, viability, vibrancy […] virility […] valueless, vampire-like, venomous, vulgar, violating and very violent’ as among the ‘v’s alone to be added to the original 3Vs.101 Yet, despite the plethora of definitions that have been applied to Big Data, it is generally accepted that it refers to volumes or collections of data that are too large or complex to be managed by traditional data-processing application software. It

See S. Doll, ‘Autonomous cargo ship completes 500 mile voyage, avoiding hundreds of collisions’ (13 May 2022) available at https://​electrek​.co/​2022/​05/​13/​autonomous​-cargo​-ship​-completes​-500​-mile​ -voyage​-avoiding​-hundreds​-of​-collisions/​ 95 See IMO, ‘Autonomous shipping’ available at www​.imo​.org/​en/​MediaCentre/​HotTopics/​Pages/​ Autonomous​-shipping​.aspx 96 See the Concirrus website, available at https://​www​.concirrus​.ai/​and the Parsyl website, available at www​.parsyl​.com/​ 97 The Geneva Association was founded in 1973 as the International Association for the Study of Insurance Economics. Its website is available at www​.genevaassociation​.org/​about​-us/​history 98 See Geneva Association Report ‘From Risk Transfer to Risk Prevention’ (2021) available at www​.genevaassociation​.org/​sites/​default/​files/​research​-topics​-document​-type/​pdf​_public/​iot​_insurance​ _research​_report​.pdf 99 See S. Lohr, ‘The Origins of “Big Data”: An Etymological Detective Story’ (New York Times, 1 February 2013) available at https://​archive​.nytimes​.com/​bits​.blogs​.nytimes​.com/​2013/​02/​01/​the​-origins​ -of​-big​-data​-an​-etymological​-detective​-story/​ 100 D. Laney, ‘3D Data Management: Controlling Data Volume, Velocity and Variety’, META Group Research Note (2001). 101 See E. Uprichard, ‘Big Data, Little Questions’ (Discover Society, 1 October 2013), available at https://​archive​.discoversociety​.org/​2013/​10/​01/​focus​-big​-data​-little​-questions/​ 94

292  Research handbook on marine insurance law should also be appreciated that Big Data is not limited to the type of data, but extends to the processing and analysis of the same.102 As with IoT and Big Data, the definitions of AI are expansive. The term was coined by McCarthy, for a conference at Dartmouth College in 1956, as ‘the science and engineering of making intelligent machines’.103 In the six decades since then, the meaning of the term and its relationship with the rival term ‘machine intelligence’ has been debated at length.104 For example, Kaplan and Haenlein define AI as ‘a system’s ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation’.105 Russell and Norvig explain: We define AI as the study of agents that receive percepts from the environment and perform actions. Each such agent implements a function that maps percept sequences to actions, and we cover different ways to represent these functions, such as reactive agents, real-time planners, decision-theoretic systems, and deep learning systems.106

In essence, and at the risk of being criticised by those who work in the field, the term covers all ‘intelligent agents’ (that is, computer systems) that are able to learn, adapt and function in a variety of situations and environments.107 There is also a distinction between ‘Narrow AI’ or ‘Weak AI’ in which the ‘agent’ carries out a specific task (such as speech or facial recognition) and ‘Artificial General Intelligence’ (AGI), in which the agent is able to carry out a wide variety of tasks including those which it has not carried out before, much like a human being.108 There has been a flurry of interest and commentary, much of it woefully misinformed, on AGI in recent months, arising from the release of ChatGPT. A number of governments are in the process of developing ‘strategies’ to address the possible emergence of AGI (often after months or years of indolence).109 This chapter does not join this veritable ‘scrum’ and instead,

See FCA (2016), ‘FS16/5: Call for Inputs on Big Data in Retail General Insurance | FCA’, No. FS16/5, available at www​.fca​.org​.uk/​publications/​feedback​-statements/​fs16​-5​-call​-inputs​-big​-data​-retail​ -general​-insurance 103 J. Roberts, ‘Thinking Machines: The Search for Artificial Intelligence’, 2(2) Distillations 14–23 (2016) available at www​.sciencehistory​.org/​distillations/​magazine/​thinking​-machines​-the​-search​-for​ -artificial​-intelligence 104 The term ‘Machine Intelligence’ was employed by Donald Michie, whose long and distinguished career in computer sciences began when he worked at Bletchley Park with Alan Turing and others on decrypting German military communications during the Second World War. 105 A. Kaplan and M. Haenlein, ‘Siri, Siri, in My Hand: Who’s the Fairest in the Land? On the Interpretations, Illustrations, and Implications of Artificial Intelligence’, 62(1) Business Horizons 15–25 (2019). 106 S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach (4th ed, Prentice Hall, 2020). 107 It should be appreciated that not every emerging technology involves or utilises AI. 108 Other important concepts relevant to AI include ‘machine learning’, by which algorithms ‘learn’ from examples and improve their performance with more data over time. Machine learning involves decision trees and Bayesian networks. ‘Deep machine learning’ involves complex statistical models and algorithms in ‘neural networks’. See www​.ibm​.com/​cloud/​blog/​ai​-vs​-machine​-learning​-vs​-deep​ -learning​-vs​-neural​-networks 109 The UK Government released a white paper on its AI strategy in March 2023, available at www​ .gov​.uk/​government/​publications/​ai​-regulation​-a​-pro​-innovation​-approach 102

Marine insurance fraud and emerging technology  293 readers are advised to consult informed sources on the nature of AGI and the opportunities and risks which it presents.110 In respect of the maritime transport industry, Big Data and AI are being employed in a variety of ways, including: advanced scheduling; route forecasting (including the most efficient route in terms of fuel optimisation); predictive maintenance systems; maximising vessels’ cargo capacity; and surveillance systems on vessels.111 MASS, as already noted, are at the high point of this process but by no means the only point or end point. Indeed, Lloyd’s Register estimated that the marine industry spent US$931 million on AI in 2022, with this figure set to increase to US$2.7 billion USD by 2027.112 In terms of insurance, Big Data is being employed in product development, marketing, distribution and sales, underwriting and claims processing.113 For example, in respect of underwriting, telematic devices in vehicles permit insurers to collect real-time behaviour and usage data to determine appropriate premiums. In respect of claims, text analytics can be employed to identify potential fraud in terms of recognising particular trends in respect of policyholders’ claims or loss adjusters’ reports. There are also a number of ways in which AI has been employed to improve the efficiency of insurance operations. In terms of consumer insurance business, the use of robo-advice has become prevalent in some markets (but not all). That said, EIOPA’s thematic review in 2019 identified fraud detection to be the most common use of Big Data analytics in respect of motor and health cover.114

4.

TECHNOLOGY V FRAUD

As indicated above, marine insurance fraud may be carried out by a variety of actors, on a variety of occasions and in a variety of ways. While policyholders are the most common perpetrators, it is not unknown for insurance professionals to be behind fraudulent activities. As mentioned above, frauds may occur prior to inception, as in dishonest under-insurance, and after the policy has been put in place, when a claim is made. This variety is limited only by the imagination and ingenuity of their perpetrators. Perhaps the best way in which to examine how emerging technology may address the myriad forms of marine insurance fraud is to consider how it might have affected the outcome of some fraudulent enterprises.

See, for example, S. McLean et al., ‘The Risks Associated with Artificial General Intelligence: A Systematic Review’, Journal of Experimental & Theoretical Artificial Intelligence (2021) available at www​.tandfonline​.com/​doi/​full/​10​.1080/​0952813X​.2021​.1964003. See also Journal of Artificial General Intelligence available at https://​journals​.scholarsportal​.info/​browse/​19460163. 111 See D. Owczarek, ‘AI in Maritime Industry: How Artificial Intelligence Solutions Benefit the Shipping Sector’ (13 March 2022) available at https://​nexocode​.com/​blog/​posts/​ai​-in​-maritime​-artificial​ -intelligence​-solutions​-in​-the​-shipping​-sector/​ 112 See E. Palmejar and N. Chubb, ‘The Learning Curve’ available at www​.maritime​.bg/​wp​-content/​ uploads/​2022/​08/​2203​_Thetius​_LR​_AI​_F3​.pdf 113 See Y-L. Grize et al., ‘Machine Learning Applications in Nonlife Insurance’, 36 Appl Stochastic Models Bus Ind. 523–37 (2020). 114 Eiopa (2019), ‘Big Data Analytics in Motor and Health Insurance: A Thematic Review’, http://​dx​ .doi​.org/​10​.2854/​54208 110

294  Research handbook on marine insurance law 4.1 Scuttling Scuttling, where a ship owner deliberately wrecks or sinks its own vessel, is potentially the most lucrative form of marine insurance fraud, given the value of the vessels and cargo involved. An additional incentive for the dishonest ship owner is that the physical evidence of its fraud may be destroyed or rendered inaccessible (that is, miles beneath the surface of the ocean) to those investigating the vessel’s loss. There have been many instances of such practices over the years, some of which were eventually the subject of (often protracted) litigation. In Shell International Petroleum Co Ltd v Gibbs, The Salem,115 the court had to deal with a particularly audacious fraud. In November 1979, the South African Strategic Fuel Fund Association (SASFFA) agreed to purchase from American Polamax International Inc. 1.5m barrels of Saudi Arabian light crude oil to be delivered in Durban. American Polamax was, unbeknownst to SASFFA, a fraudulent enterprise. The fraudsters bought a vessel for the task and reregistered her in the name of ‘Salem’. She was then chartered to an innocent party, Pontoil SA, which contracted her to carry a cargo of Kuwaiti (not Saudi) oil to Italy (not South Africa) and declared the voyage under an open cover with the insurers. The Salem actually headed for South Africa and Pontoil (being unaware of this fact) on-sold its cargo to Shell for US$56m. The vessel discharged in Durban in December 1979, under a different name, and her tanks were filled with seawater. After collecting the payment from SASFFA, the fraudsters instructed her master and crew to scuttle the Salem off the coast of Senegal. Shell subsequently sought to recover its losses from the insurers. In his judgment in April 1981, Mustill J held that there was no loss by barratry, which is an act directed against a ship owner. Nor was Shell’s loss the result of persons acting maliciously, as it had been caused by a criminal enterprise for the purposes of financial gain. The loss was, however, covered by virtue of the scuttling of the ship being regarded as a ‘peril of the sea’ vis-à-vis the cargo on board and under ‘takings at sea’, this being a situation where the goods had been lost through the deliberate act of the crew. The matter proceeded to the Court of Appeal and ultimately the House of Lords, where it was held (in February 1983) that ‘takings at sea’ did not cover wrongful misappropriation of a cargo by a ship owner; that barratry had not been committed; and that the scuttling was a ‘peril of the sea’, thus enabling Shell to recover under the policy.116 In Kairos Shipping Ltd, Teare J held that a ship owner was not entitled to a declaration that its liability for losses following the sinking of one of its vessels was limited under the Convention on Limitation of Liability for Maritime Claims 1976. On 30 March 2013, the Atlantik Confidence was in the Gulf of Aden in the course of its voyage from Ukraine to Oman with a cargo of steel products. At 0530 hours, a fire alarm sounded indicating a fire in the unmanned engine room. Despite firefighting efforts by the crew, the vessel began to list to port and was abandoned. She continued to take on water until she sank on 3 April. The wreck could not be inspected because it lay in very deep water. The ship owner’s liability would have been limited if the cargo insurer had not proved that it resulted from the owner’s own act or omission, committed with the intent to cause such loss, or recklessly and with knowledge that such loss would probably result. In the event, the insurer successfully argued at trial in 2016 that the ship owner had indeed scuttled the vessel by delib [1982] 1 All ER 225. [1983] 2 AC 375.

115 116

Marine insurance fraud and emerging technology  295 erately starting the fire. Teare J held that it was more likely than not that the fire originated in the storeroom and, as an accidental cause of the fire could not be identified, there was a real and substantial possibility that it was started deliberately. Moreover, although the insurers could not produce direct evidence that the chief engineer had flooded the engine room deliberately or that the double bottom tanks had also been flooded deliberately, the ship owner’s explanations for these events were described by the learned judge as ‘unlikely’ and ‘remote’. The learned judge observed that this succession of ‘improbable events’ had been preceded by a change of route into deep water five days before the fire (which the master and ship owners had tried to conceal and then sought to explain as an anti-piracy precaution) and an unscheduled ‘abandon ship’ drill four days before the fire. When these matters were considered together with the master’s and chief engineer’s conduct and the behaviour of senior personnel of the ship owner, Teare J concluded that the vessel had been deliberately sunk to ameliorate the ship owner’s financial difficulties. Finally, the Brillante Virtuoso was a motor tanker owned by Suez Fortune Investments Limited, the first claimant in the Brillante Virtuoso litigation.117 In June 2011, she was carrying 140,000 metric tonnes of fuel oil from Ukraine to China.118 Crucially, she carried a Voyage Data Recorder (VDR)119 but, as Teare J noted in his judgment, ‘[u]nfortunately, the VDR [installed on the Brillante Virtuoso] must have been an early one of its type because it did not record engine movements. There was on board an engine logger which would have recorded engine movements but that has not survived.’ An unarmed security team was to board the vessel before she entered the Internationally Recommended Transit Corridor in the Gulf of Aden. At 2200 on 5 July 2011 her engines were stopped and she began drifting close to the Yemeni 12-mile limit. A small boat carrying seven armed and masked men, who claimed to be ‘security’, approached and the master allowed them on-board (despite the fact he was expecting an unarmed three-man team to arrive several hours later).120 These men detained most of the crew in the accommodation section of the vessel, demanded that it be sailed to Somalia and – after it travelled in the opposite direction for more than a hour before stopping – detonated an improvised explosive incendiary device (IEID) in part of the engine room complex. They left and the crew subsequently abandoned ship. A salvor vessel, the Poseidon, arrived from Aden within a very short time and, by 1030 on 6 July, the fire appeared to be dying out. It then resurged and effectively destroyed the accommodation section, wheelhouse and engine room before it ceased on 8 July. The vessel was towed to Khor Fakkan (where the cargo was offloaded) and was eventually scrapped.

The principal decisions in the litigation comprise Flaux J’s determination in 2015 that the vessel was, as argued by the claimants but denied by the defendant insurers, a constructive total loss: see [2015] EWHC 42; the striking out of the first claimant’s claim by Flaux J in 2016, see [2016] EWHC 1085 (Comm); and Teare J’s judgment following the trial in 2019. 118 It is noteworthy that the first claimant had informed the relevant authorities that the cargo was bitumen (in order to avoid customs duties). 119 The IMO SOLAS (Safety of Life at Sea) Convention Chapter V, Regulation 20 required all non-passenger vessels of 3000 gross tonnage or more to be fitted with a VDR not later than July 2010. A VDR must record a variety of details about the vessel, including its position, speed and heading; bridge audio; communication audio; radio; radar data; main alarms; and rudder order and response. 120 Teare J held that the decision to allow the vessel to drift and the absence of any ‘vigilance’ were in breach of the Piracy – Best Management Practices, Third Edition (BMP3) issued by the shipping industry. 117

296  Research handbook on marine insurance law The claimants sought the vessel’s insured value of US$55m (plus US$22m for disbursements and increased value) on its war risks cover as a constructive total loss. The defendant insurers declined to pay and the claimants issued proceedings in February 2012. The insurers contended the first claimant had acted in breach of warranty by, among other things, failing to apply BMP3. In January 2015, the court held that the vessel was a constructive total loss but, within two months, the insurers amended their defence to plead that the fire was started deliberately with the first claimant’s consent, contrary to s 55(2)(a) MIA. In May 2016, the first claimant’s claim was struck out for its (and its beneficial owner’s) failure to comply with an unless order for disclosure. The second claimant bank (and its insurers) continued the claim. At trial in late 2018, seven years after the events in question, Teare J accepted the defendants’ contentions that the Brillante Virtuoso had been scuttled because the first claimant had arranged the ‘attack’ by the ‘pirates’ and the fire which followed the detonation of the IEID. The learned judge also concluded that the master, the chief engineer and the salvor had been a party to this conspiracy. The second claimant had, therefore, failed to establish that the vessel’s loss was caused by an insured peril (that is, piracy). Again, as with The Kairos, a series of ‘improbable events’ – including the armed men’s use of an incomplete IEID, the master’s lack of vigilance and the chief engineer’s tampering with the engines – and the revelation of the first claimant’s precarious finances led to the learned judge’s conclusion. As noted above, the facts were ‘sufficiently unambiguous’ to support such a finding against the first claimant.121 One of the more striking characteristics of each of these three cases is the length of time from the scuttling to the resolution in the courts. In The Salem, Mustill J’s judgment was issued approximately 18 months after American Polamax’s scheme reached its fulfilment off the coast of Senegal. This, admittedly, is a relatively short time in the world of civil and commercial litigation, but there was a further two-year delay until the House of Lords confirmed the substance of Mustill J’s determination. This constituted a significant delay for Shell in the recovery of its US$56 million outlay. In The Kairos, there was a three-year interval between the scuttling of the Atlantik Confidence and Teare J’s judgment – a significant period of commercial uncertainty for the innocent parties concerned. In Brillante Virtuoso, the time which expired between the scuttling of the vessel and Teare J’s (final) judgement was more than eight years in length. The learned judge himself referred to the significant length of the litigation, including the 52-day trial, in his judgment.122 Such lengthy delays between the commencement of a claim and its resolution at trial or appeal are a common feature of complex commercial disputes and the almost inevitable battles over significant volumes of conflicting witness, expert and documentary evidence (or battles over the absence of the same) are a major cause of this phenomenon. As noted in the discussion above on the difficulties of bringing such claims, these battles can be exacerbated by animosity between the parties, which is clearly a factor when fraud is alleged. As noted, the first claimant’s case in Brillante Virtuoso fell – after almost five years – because of its failure to comply with its disclosure obligations rather than because of the weakness of its claim. The

A book entitled Dead in the Water: A True Story of Hijacking, Murder, and a Global Maritime Conspiracy (Portfolio, 2022) by journalists Matthew Campbell and Kit Chellel details the extent and ramifications of the conspiracy. 122 Teare J observed that the ten scuttling cases tried between 1985 and 2018 had lasted between 14 and 87 days. 121

Marine insurance fraud and emerging technology  297 digitisation of contractual documentation relating to marine transport and marine insurance, including the use of smart contracts, and the making of it available to all the relevant parties prior to inception (and certainly before litigation has commenced) would help curtail protracted litigation and, arguably, fraudulent claims at the outset. After all, as Shell would attest, ‘Justice delayed is justice denied’. This leads onto a second characteristic of these three cases: the absence or unreliability of the documentation involved. In The Salem, the fraudsters were able to concoct documentation to avoid the consequences of the oil embargo against South Africa; to ensnare the innocent charterers, Pontoil; and to deliver the oil (from Kuwait, not Saudi Arabia) in South Africa rather than Italy. In The Kairos, ‘the brief case holding the deck log, GPS log, movement book and crew passports and the vessel’s working chart was left on board’ when the vessel sank and other documents were not forthcoming. A major aspect of the Brillante Virtuoso case was the dispute over concocted or missing documentation, not least the archives which the first claimant refused to disclose. Once again, the digitisation of relevant contractual documentation, whether in the form of a smart contract, or otherwise, can assist in containing the use of such fraudulent documentation to advance the schemes of would-be criminals. Perhaps the most significant impact of the development of emerging technology in relation to marine insurance fraud is that of the ability to plan, optimise and track the movements of vessels and cargo across the globe in ‘real time’. The tracking of ships was not unknown in 1979, but it is arguable that the fraudsters’ ability in The Salem to route a vessel from its ostensibly expected destination in Italy to South Africa; to rename her not once but twice (upon purchase and then upon her arrival in Durban); and to empty her holds of her cargo before her arrival would be severely curtailed by the existence of the technology which exists today. Similarly, in The Kairos, the ship owner’s attempt to conceal the vessel’s change in course in the Gulf of Aden (to ensure that it sank in inaccessible waters) would have been apparent long before its discovery in the course of the subsequent litigation. Again, in Brillante Virtuoso, the fact that the vessel had been inexplicably drifting close to the Yemeni coast would have raised proverbial ‘alarm bells’ long before the scuttling. Moreover, the advent of MASS vessels removes much of the ‘human element’ from the proverbial equation. In the Brillante Virtuoso, modern ‘real-time’ on-board sensors may well have detected and recorded (with digital still or moving footage, temperature recordings, and the like) the fact that the main engine did not break down but was stopped by the chief engineer; the exact location and cause of the initial fire; the cause of the second or reignited fire; and the movements of the armed men and crew. Some of these events were recorded on the VDR, but it provided only partial, after-the-event information which was debated at length in the subsequent litigation.123 Beyond this, a truly autonomous vessel, controlled either by on-board systems or by on-board systems in conjunction with remote operators, would be even less vulnerable to the actions of real or fake pirates and collaborators within the crew. While the ship owners, operators and insurers would have been anxious not to lose the vessel or its cargo, the emotional or psychological pressure that would have resulted from potential harm to a human crew would have been absent (the collaborators may have been absent too). Although it may be sometime before autonomous ships can make journeys as complex as those from the

123 The presence of oil rather than bitumen in the hold would also have been apparent from on-board sensors.

298  Research handbook on marine insurance law Middle East to Europe or South East Asia it is only a matter of time before much more marine transportation is conducted by such vessels. Finally, there is the impact of Big Data and AI. It became apparent during the course of the litigation in each of these three cases that the ship owners (and their associates) were in grave financial difficulty. These difficulties were often uncovered only after extensive investigations during the course of litigation by the defendant insurers, other parties and their respective legal teams. In the Brillante Virtuoso, for instance, there are references in Teare J’s judgment to the fact that, when it was scuttled, the vessel was running at an operating loss of US$12m; that the first claimant had defaulted on payments to its lenders; that various trade debts were also overdue; and that notices of default had been served on the first claimant by the second claimant. There are similar findings in The Kairos. While much financial information will remain confidential or otherwise unavailable (until uncovered in the disclosure process) to insurers and their legal teams, the volume of data available to them prior to the inception of policies has increased dramatically. The Coalition Against Insurance Fraud has reported that ‘[t]he most frequent data source relied upon continues to be internal data (100%). Other sources include industry fraud-watch lists (88%), unstructured data (81%), public records (79%), third-party data aggregators (55%), social-media data (48%), and data from connected devices (15%)’.124 The data collected by such methods is increasingly being subjected to sophisticated AI tools which incorporate ‘automated red flags’, predictive modelling, exception reporting and data visualisation/link analysis to detect potential fraud both before inception and in the event of claims. The growing use of both predictive modelling and ‘image-based fraud prevention’ techniques may prove to be of particular significance in the detection and prevention of marine insurance fraud.125 4.2 Overvaluation The facts of Eagle Star Insurance Co Ltd v Games Video Co, The Game Boy126 will be familiar to many but they bear a brief recapitulation here. In the early hours of 13 January 1999 the Game Boy was moored at the Avlis shipyard in Greece when an explosive device attached to its hull was detonated. The resulting damage caused the vessel to list and partially sink. The Game Boy’s owners and operators sought a declaration that its insurer was liable to indemnify them for US$1.8m. The insurer contended that cover had been validly avoided and that it was under no liability to them. The heart of the insurer’s case was that the Game Boy’s true value was only US$100,000 as scrap. Prior to its demise in Avlis, the Game Boy led an undistinguished 34-year life as the Admiral Lunin,127 the Kara Dag and the Seniorita, having been constructed in 1965 as a Soviet naval 124 www​.sas​.com/​content/​dam/​SAS/​documents/​marketing​-whitepapers​-ebooks/​third​-party​-whitepapers/​ en/​coalition​-against​-insurance​-fraud​-the​-state​-of​-insurance​-fraud​-technology​-105976​.pdf 125 See S. Pandhare, ‘Big data analytics: new whistleblower on insurance fraud’ (Digital Insurer, 2015) available at www​.the​-digital​-insurer​.com/​library/​big​-data​-analytics​-new​-whistleblower​-on​-insurance​ -frauds​-an​-infosys​-research​-brief​-by​-sachin​-pandhare/​; and ‘How AI is Transforming Insurance Claims Fraud Detection’ (16 March 2022) available at https://​accern​.com/​blog/​how​-ai​-is​-transforming​-insurance​ -claims​-fraud​-detection/​ 126 [2004] Lloyd’s Rep IR 867. 127 Presumably named in honour of the Second World War Soviet submarine commander Nikolai Lunin.

Marine insurance fraud and emerging technology  299 training ship and, from 1989, spending its time in Greece either laid up or trading as a bar/ cafeteria. In 1998, the vessel was purchased by Games Video Co (GVC) with the intention that it would be managed by a Mr Ghiolman (who took a minority holding in GVC) and Casinomar SA as a restaurant and casino.128 Among a faxed request from local insurance consultants to London brokers in March 1998 to obtain cover was a description of the vessel in which it was described as ‘For Sale’, together with ‘various “value ideas” for the navigating (US$2,500,00) and port risks (US$1,800,000)’. The Hull and Machinery of the vessel (as well as Casinomar SA’s interest in her) was subsequently insured for US$1.8m. In the course of his 2004 judgment, Simon J made the following observation: First, in many cases which rely heavily on the evidence of witnesses the Court can test the veracity of those witnesses by reference to the contemporary documents. This is not such a case. On any view of the case, many of the crucial documents have no provenance: there are no drafts of these documents showing a process of negotiation, and there are no copies of documents with fax headers, although it is clear that some of the documents were sent by fax.

The learned judge refused to grant the declaration sought by the owners and operators and instead held that, GVC did not genuinely believe that the vessel was worth US$1.8 million. In particular, GVC knew that the documents produced in support of its claim were fraudulent and were ‘created because the Assureds knew very well that the valuation of $1,800,000 could not be justified without them’. Simon J held that true value of the Game Boy at the material time was ‘approximately $100–150,000; and I further find that the Assureds knew this’. The learned judge added: ‘I specifically reject Mr Ghiolman’s evidence that he believed that the vessel’s value was $1,800,000 and that he had good grounds for that belief.’ The insurer was entitled to avoid the policy and was discharged from liability. There are three traditional methods for valuing a marine vessel. The first is the ‘market approach’, by which the vessel is valued by comparison to the most recent transaction of a comparable vessel. This figure may then be adjusted (upwards or downwards) by reference to a number of factors, including any change in the freight market, the respective vessels’ trading routes and the idiosyncrasies of either vessel (for example, on-board equipment such as cranes). The second method is the ‘replacement value’, by which a vessel is valued on the basis of how much it would cost to have a similar vessel built (in the same yard) to its original condition and that figure is then depreciated to reflect her present condition. The third is the ‘income approach’, which is intended to reflect the value of the vessel in terms of the monies that she is expected to earn for her prospective owner. The valuation of ships is customarily carried out by sale and purchase (‘S&P’) shipbrokers, usually in accordance with the above methods. As one would expect, prior to the advent of modern technology this was a paper-heavy exercise which was very much dependent upon the professional abilities of the relevant broker. Valuations and other data relating to the shipping market such as daily freight market prices and maritime shipping cost indices are also provided by, among others, the Baltic Exchange and Clarkson’s World Fleet Register.129 Despite the fact that there are a wide variety of data sources and qualifications, such as the Lloyd’s Maritime

The aforementioned owners and operators. Clarkson’s website is available at www​.clarksons​.net/​wfr/​

128 129

300  Research handbook on marine insurance law Academy Certificate in Vessel Valuation, available to members of the shipping industry, there is no guarantee that the valuation produced by a broker will necessarily be ‘the correct one’.130 In The Game Boy, the valuation of the vessel appears to have been compiled from a number of fraudulent documents.131 Indeed, it does not appear that any of the traditional valuation methods (presumably the ‘income approach’ would have been appropriate here) were employed by a S&P shipbroker or anyone else. In any event, had they been pressed at the time, it is possible that the fraudsters could have sought the assistance of a ‘friendly’ shipbroker or other third party to buttress their fraud (NB: shipbroking, as with other professions, is sadly not free of those with criminal inclinations). Twenty years after the events of The Game Boy, the attempts to conceal or mislead the insurers as to the true value of the vessel would have been much more difficult to achieve, and would have been exposed much sooner than at trial five years later, given the advances of AI and Big Data-based ship valuation. For example, MarineTraffic, a company which provides ship tracking information and maritime analytics, is able to offer vessel valuations on its online platform, which it created in conjunction with the maritime data specialist Signal Ocean.132 According to MarineTraffic, this information is ‘live and historical’, is updated on a weekly basis and is ‘derived from recent transactions and a sophisticated algorithm which takes in a range of data points including the vessel’s age, shipyard built, size, equipment or refits’. Clarkson Valuations also provides ship valuations with the use of ‘market expertise with leading research and technology’.133 Given the growth in volume and accessibility of data to companies involved in ship valuations, and the ability of insurers and others to make use of that information, the ability of would-be fraudsters to overvalue their vessels has been markedly curtailed. Their difficulties are exacerbated by the fact that the use of such Big Data can be combined with the employment of blockchain, whether in smart contracts or otherwise, to verify the documentation purporting to give the value of vessels. As a result, fraudulent valuations can – ideally – be identified prior to inception, or robust evidence of the same will be available upon the making of a fraudulent claim. 4.3

Non-existent Cargo

Although it is not a case involving fraud, Coven SpA v Hong Kong Chinese Insurance Co134 demonstrates one of the essential aspects of the coverage of cargo. The claimant was the assignee of a marine cargo policy issued by the defendant insurer for a consignment of broad beans that was shipped from China to Italy in 1990. The bill of lading and the policy (along with other documentation) described the cargo as weighing 2787 metric tonnes, whereas only 2401 metric tonnes of broad beans were actually delivered. The claimant sought the difference in value between those two figures under the policy on the basis that there was a shortage. The defendant insurer argued that consignment of beans loaded at the start of the voyage was Details of the certification are available at https://​informaconnect​.com/certificate​-in​-vessel​ -valuation/​ 131 It is appreciated that, following Versloot, the use of ‘fraudulent devices’ is no longer an absolute bar to a claim, but the magnitude of the fraud in The Game Boy would place it beyond Mance LJ’s late, lamented fifth category. 132 Its website is available at www​.marinetraffic​.com/​blog/​live​-vessel​-values​-now​-available​-on​ -marinetraffic/​ 133 Its website is available at www​.clarksons​.net/​ShipValue/​#!/​ 134 [1999] Lloyd’s Rep IR 565. 130

Marine insurance fraud and emerging technology  301 the same as that delivered in Italy and, moreover, that the difference was the result of a measurement error which was not covered by the policy. The trial judge agreed with the defendant insurer and the claimant appealed. In 1998, the Court of Appeal dismissed the appeal, reiterating that the policy insured against loss of and damage to cargo and did not cover a paper loss which was the result of a measurement error. It was possible to obtain cover against a measurement error but there were no such terms in the policy before the court. As Clarke LJ observed: The parties cannot have intended to insure goods which never existed […] A further point which militates against Coven’s claim is that it is difficult, if not impossible, to see how the insured could have an insurable interest in the subject matter of the insurance […] in respect of the 385.6 mt of beans which never existed.

The trial judge had been right in deciding that there was no shortage between the quantity of beans actually embarked and the quantity delivered. The insurers were therefore not liable for the claimant’s alleged ‘loss’. The decision in Coven SpA was applied in the more recent case of Engelhart CTP (US) LLC v Lloyd’s Syndicate 1221 for the 2014 Year of Account.135 In mid-2015, the claimant policyholder bought 9000 mt of copper ingots and sold the same to Chinese receivers. The first shipment of 7000 mt of ingots was successfully delivered. The second shipment, however, did not contain copper ingots, but slag of nominal value instead. The related bills of lading, packing lists and quality certificates were fraudulent. The claimant sought recompense for the loss of the cargo and insured expenses from the defendant insurers but the claim was refused. The claimant consequently issued proceedings, submitting that the policy covered physical loss claims where an insured had been defrauded into taking up documents of title for non-existent goods. Giving judgment in April 2018, Sir Ross Cranston observed that the purpose of all risks marine cargo insurance was to cover loss of or damage to property. In this case, however, it was agreed by the parties that there was no physical loss of goods. The claimant policyholder had suffered an economic loss as a consequence of accepting fraudulent documents. The bespoke conditions of the policy expanded the all risks coverage very broadly, but the presumption that it covered physical losses had not been displaced. The claimant policyholder’s reliance on the concealed damage clause, the container clause and the fraudulent documents clause did not alter this fact. Had the parties intended to cover losses resulting from the acceptance of fraudulent documents of title in relation to the ingots, they should have agreed clear words to that end in the policy wording.136

[2019] 1 All ER (Comm) 583. In Quadra Commodities SA v XL Insurance Co SE [2022] 2 All ER (Comm) 334, the defendant insurers contended that goods which the claimant policyholder had purchased had never existed rather than, as the policyholder argued, been misappropriated. Butcher J accepted that warehouse receipts, independent inspectors’ reports and the physical delivery of some of the goods to the policyholder demonstrated, on a balance of probabilities, that such goods had existed: a ‘victory’, then, for analogue documentation, but one which took three years from the alleged misappropriation to Butcher J’s decision to achieve. The decision is also noteworthy for the learned judge’s comments on insurable interest and s13A Insurance Act 2015. The Court of Appeal upheld Butcher J’s decision in Quadra Commodities SA v XL Insurance Company SE & Ors [2023] EWCA Civ 432. 135 136

302  Research handbook on marine insurance law The emergence of IoT, and its application to maritime transport (whether one refers to this as IoT or as the Internet of Ships), enables the owners of cargo to track their cargo in real time during its voyage from, say, China to its arrival in Italy. Technologies which enable the rapid identification and verification of items, such as barcodes and intelligent sensors on shipping containers and the items themselves, are now widely used in the maritime industry to this end. Not only do such technologies reduce the transit time for the delivery of goods to their buyers (such as in speeding up the check-in time for the reception of goods through customs, and so on); they also enable the cargo owners and others to determine whether or not those goods actually exist or have been stolen, damaged or interfered with during the course of their journey. Other IoT devices, such as motion tilt sensors and temperature and humidity sensors in containers or vessels’ cargo holds can also be used to monitor the condition and existence of cargoes. 4.4

From Law to Technology?

While it is clear from the above examples that the application of emerging technology to marine transport and marine insurance can reduce the opportunity for the commission of fraud, it would be remiss to treat its introduction as a panacea. The ingenuity of criminals should not be underestimated. The same technologies that may be employed to combat their activities may be employed to advance them. Moreover, there are a significant number of specific problems which arise in relation to the application of emerging technology to marine transport and marine insurance. First, there are many private commercial providers of maritime tracking services and there is no all-encompassing or universal ‘public’ system which combines, replicates or complements their data. Reference has already been made to, among others, Tradelens and MarineTraffic. Consequently, the information gathered by these private tracking systems may – despite the expertise of the providers – be incomplete, inaccurate or unreliable in relation to specific vessels, at specific times, in specific locations. Further, some vessels (especially older ones that may be more vulnerable to the predations of fraudsters, pirates and other criminals) may not have IoT sensors installed so as to provide real-time information to ship owners and insurers about their condition and the condition of their cargoes. Even now, not all vessels are required to carry VDR, despite that relatively ‘mature’ technology having been available prior to the emergence of the IoT or Internet of Ships. It may be several years before data collection technology is advanced and integrated enough to enable a sufficient volume of accurate information to be available to those who require it at an affordable price. The heterogeneous nature of the data collected, when it is collected, by the shipping industry and insurers also presents problems in terms of its analysis. The development and integration not only of data collection but data management services between ship owners and insurers is far from complete. Predictive modelling and other anti-fraud technologies are dependent upon accurate information but the software also operates in different ways from provider to provider (naturally so, given that the providers are competing with each other to offer these products to their customers), some of which may not be appropriate in some circumstances. As with the absence of uniform data, the absence of accurate analysis provides opportunities for fraudsters to continue their nefarious activities. There is also the fact that reliance on wireless and satellite communication technologies and on computer networks, including DLTs, present privacy and security issues for those involved

Marine insurance fraud and emerging technology  303 in their use. The need to develop, maintain and update safe and secure connections for ship owners, agents, brokers and insurers and their maintenance in the face of ongoing cybersecurity threats such as route control attacks and eavesdropping is essential for the long-term development of marine transport-related insurtech. Even in the absence of cyber attacks on wireless, satellite and computer systems, the availability of information relating to ships and their cargo (even in ostensibly private platforms) presents a risk of their abuse by would-be fraudsters and other criminals. Finally, there are the challenges presented by the rise of MASS. Questions about matters such as the liability of ship owners and insurers for autonomous vessels in the event that they are involved in some misadventure (be it a collision or piracy) have not yet been resolved.137 As previously noted, the IMO is developing a code of conduct for the operation of MASS, but this has yet to be implemented and there may be issues which emerge and require addressing after it has been put into effect in or about 2028.138 It is the nature of such codes that they often require constant amendment. Moreover, the code may not necessarily be relied upon by the courts in all relevant jurisdictions when it comes to addressing such issues. Further, as with satellite, wireless and computer systems, the IoT devices on vessels and MASS vessels themselves will be targets of would-be hackers.139 That said, these far from trivial problems do not detract from the nature and the magnitude of the potential paradigm shift represented by the application of emerging technology, whether in the form of the ‘Internet of Ships’ or of insurtech, to the fight against marine insurance fraud. As Brownsword has written at length, the increasing use of emerging technology will result in a move from what he describes as Law 1.0 and Law 2.0 to Law 3.0.140 He explains that Law 1.0 is the traditional ‘application of rules, standards and general principles to particular fact situations’ by the courts in civil or criminal litigation. In Law 2.0 ‘the centre of gravity of law shifts from the courts and historic codes to the political arena’, with governments and regulators seeking to address problems which cannot be resolved effectively (for example, with sufficient haste) by the courts. The rise of the regulatory state and the increasing role and power of what were once referred to as quangos encapsulates this shift from Law 1.0 to Law 3.0. Brownsword explains further that Law 3.0 is ‘characterised by a sustained focus on the potential use of a range of technological instruments’ to resolve various problems. Brownsword offers several practical examples of this shift towards technological solutions to problems that would in the past have been addressed by the courts or regulators. Perhaps the most germane insofar as the subject-matter of this chapter is concerned is that of the disruption

See B. Soyer, ‘Autonomous Vessels and Third-party Liabilities: The Elephant in the Room’ in B. Soyer, New Technologies, Artificial Intelligence and Shipping Law in the 21st Century (Routledge, 2019) 105–16 and F. Stevens, ‘Carrier Liability for Unmanned Ships: Goodbye Crew, Hello Liability?’ in the same at 148–61. See also B. Soyer, ‘Insuring Remote-controlled and Autonomous Shipping: A Paradigm Shift in Law and Insurance Markets Required?’ in D. Rhidian Thomas, The Modern Law of Marine Insurance vol. 5 (Routledge, 2023), 105–16. 138 See P.M. Eggers, ‘Maritime Autonomous Surface Ships: Marine Insurance Response to Risks’ in B. Soyer and A. Tettenborn (eds), Artificial Intelligence and Autonomous Shipping (Bloomsbury, 2021), 139–59. 139 See G. Leloudas, ‘Cyber Risks, Autonomous Operations and Risk Perceptions’ in B. Soyer and A. Tettenborn (eds) Artificial Intelligence and Autonomous Shipping (Bloomsbury, 2021) 101–17. 140 See R. Brownsword, Law, Technology and Society: Re-imagining the Regulatory Environment (Routledge, 2019) and Law 3.0 Rules, Regulation, and Technology (Routledge, 2020). 137

304  Research handbook on marine insurance law at London Gatwick Airport in December 2018 caused by an unauthorised drone aircraft operating in close vicinity to the terminals and runways. In the interests of safety, all flights were suspended and the airport was closed for two days. The resultant disruption to thousands of passengers’ Christmas holidays was headline news across the world. Various individuals and bodies, including BALPA, demanded that action be taken to prevent the recurrence of such incidents.141 In the event, the UK Government announced that it would provide the authorities with new powers to tackle illegal drone use, and that drone no-fly zones would be extended to three miles beyond UK airports. This was the archetypical Law 1.0 or Law 2.0 response to such a problem; by contrast, others suggested that the drone problem could be resolved by way of technology, namely, by methods which would render it impossible for a drone to be flown near an airport. In doing so, they demonstrated a key difference between the mentality of Law 1.0 and 2.0 and that of Law 3.0. The technological approach to problem-solving is focused on preventing the wrongdoing, such as flying a rogue drone near an airport, in advance, whereas the ‘traditional’ approaches focus on punishment and compensation for the wrongdoing after the event.142 Such is the aim of many ship owners and insurers in their adoption of emerging technology in responding to marine insurance fraud. The digitisation of contractual documents, use of smart contracts, employment of IoT sensors on ships and cargo containers and use of Big Data and AI in underwriting and claims management are all intended to ‘nip’ the problem of marine insurance fraud in the proverbial ‘bud’. Brownsword does not, of course, suggest that the adoption of emerging technologies or a move to Law 3.0 will solve all problems and eliminate the need for laws, regulations, regulators and courts. Indeed, he states: ‘My conclusion is not that, with law so reinvented, all will go well.’ What he argues is that emerging technology can complement and enhance the rule of law rather than supplant it. In practical terms, this view that emerging technology is not a magic anti-wrongdoing elixir must be correct, given that although it may well curtail the opportunities for fraud (for example, fraudsters may no longer be able to present falsified physical bills of lading to take delivery of cargo), it will not eliminate all possible wrongdoing (for example, criminals may just use physical violence to seize the goods instead, or try to hack the DLT network). What is clear, however, is that insurers and others are right to reduce their reliance on the courts and regulators in the fight against marine insurance fraud. This, as Brownsword notes, reflects an observation made almost 30 years ago by Barlow: Law adapts by continuous increments and at a pace second only to geology in its stateliness. Technology advances in […] lunging jerks, like the punctuation of biological evolution grotesquely accelerated. Real world conditions will continue to change at a blinding pace, and the law will get further behind, more profoundly confused. This mismatch is permanent.143

See G. Topham, ‘Runway reopens after days of drone disruption at Gatwick’ (The Guardian, 21 December 2018) available at www​.theguardian​.com/​uk​-news/​2018/​dec/​20/​tens​-of​-thousands​-of​ -passengers​-stranded​-by​-gatwick​-airport​-drones 142 See L. Bygrave, who makes this point in ‘Hardwiring Privacy’ in R. Brownsword, E. Scotford and K. Yeung (eds) The Oxford Handbook of Law, Regulation and Technology (OUP, 2017). 143 See J.P. Barlow, ‘The economy of ideas: selling wine without bottles on the global net’ (Wired, 1 March 1994) available at www​.wired​.com/​1994/​03/​economy​-ideas/​ 141

Marine insurance fraud and emerging technology  305

5. CONCLUSION Insurance is as old as human civilisation. The Babylonian Code of Hammurabi, written almost 4000 years ago, describes a form of cover known today as bottomry. Gaius Marius, seven-time Consul of Rome, organised ‘burial clubs’ to pay for the funerals of his fallen legionaries. The first academic text on marine insurance was compiled in 1488 by Pedro de Santarém. Over the centuries, the basic tenets of insurance have developed at an almost glacial pace. Premiums, claims and the pooling of risk are as familiar to modern consumers as they were to the patrons of Edward Lloyd’s Coffee House. Underwriters, brokers and agents continue to thrive and to fall. Emerging technology, in the form of insurtech, may bring this ancien regime to an end.144 Yet, even if it does, insurtech will be addressing the same problems addressed by Richard Sayer at BILA’s seventh annual conference many years ago when he spoke of the need for insurers to collect and share information in order to defeat marine insurance fraud.145 Those problems and those needs have not gone away. The former will be ameliorated and the latter met only by the use of emerging technology in a way that counteracts the shortcomings of the courts and regulators in the fight against marine insurance fraud.

Although, hopefully, without the need for a Robespierre or Committee of Public Safety this time. See R.J. Sayer, ‘Marine Insurance’ available at https://​bila.org​.uk/wp​-content/uploads/old/​ 4ff4163f8c8cf4​.38380052​.pdf 144 145

15. The role of insurance in regulating cyber risks in the shipping industry Feng Wang

1. INTRODUCTION Cyber attacks have become pervasive, and the shipping industry is not immune. The cost of cyber attacks has climbed from $112 billion in 2012 to about $2 trillion in 2019.1 The total possible maximum loss for cyber globally by 2025 will be around $10.5 trillion.2 According to a survey involving the Global Maritime Forum and the International Union of Marine Insurance (IUMI), cyber risks and data theft have been ranked second for lack of preparedness by the maritime industry.3 Cyber attacks, whether detectable or undetectable, are a fairly common experience in daily business, as evidenced in another survey in 2020 which found that one in five participating companies had experienced a cyber attack in the preceding three years and that phishing and ransomware were viewed as the most common forms of attack.4 In July 2021, South Africa’s port of Durban was shut down due to a criminal cyber attack.5 Carriers have also become a common target of cyber attacks: most big names in the industry, including Maersk and COSCO, have suffered losses resulting from cyber incidents in recent years.6 Even worse, the IMO itself – an international public organisation that issues cyber safety rules – has been a victim of pervasive cyber attacks.7 In recent years, the shipping industry has become less and less based on physical platforms and is increasingly built on digital ones. Therefore, with the increased use of digital technologies such as maritime autonomous surface ships (MASS) and the internet of things (IoT), more cyber attacks and losses are expected.8

‘Cybersecurity: a global priority and career opportunity’ https://​ung​.edu/​continuing​-education/​ news​-and​-media/​cybersecurity​.php accessed 9 March 2023. 2 ‘Cybercrime to cost the world $10.5 trillion annually by 2025’ https://​c​ybersecuri​tyventures​.com/​ cybercrime​-damages​-6​-trillion​-by​-2021/​ accessed 9 March 2023. 3 ‘Global Maritime Issues Monitor 2021’, www​.globalmaritimeforum​.org/​content/​2021/​10/​Global​ -Maritime​-Issues​-Monitor​-2021​.pdf accessed 24 December 2022. 4 Department for Digital, Culture, Media & Sport (UK), ‘Cyber Security Breaches Survey 2022’ www​.gov​.uk/​government/​statistics/​cyber​-security​-breaches​-survey​-2022/​cyber​-security​-breaches​ -survey​-2022 accessed 24 December 2022. 5 Zandi Shabalala and Tanisha Heiberg, ‘Cyber attack disrupts major South African port operations’ www​.reuters​.com/​world/​africa/​exclusive​-south​-africas​-transnet​-hit​-by​-cyberattack​-sources​-2021​-07​ -22/​accessed 24 December 2022. 6 For some recent cyberattacks in the shipping sphere, see Victor Chacon, ‘Web Portals and Data Sharing by Ocean Carriers’ in Stephen Girvin and Vibe Ulfbeck (eds) Maritime Obligation, Management and Liability: A Legal Analysis of New Challenges in the Maritime Industry (Hart Publishing 2020). 7 ‘Global maritime regulator hit by cyberattack’ www​.wsj​.com/​articles/​global​-maritime​-regulator​ -hit​-by​-cyberattack​-11601560294 accessed 24 December 2022. 8 The Guidance on Cyber Security Onboard Ships version 4, www​.ics​-shipping​.org/​wp​-content/​ uploads/​2021/​02/​2021​-Cyber​-Security​-Guidelines​.pdf accessed 9 March 2023. 1

306

The role of insurance in regulating cyber risks in the shipping industry  307 The cyber insurance market is also growing. In 20159 coverage of the global market was just $2.5 billion,10 but by 2020 it had reached $7 billion in gross written premium (GWP), and it is expected to rise to $20.56 billion by 2025.11 The insurance industry is trying to catch up with this new emerging risk not only in the shipping industry, but also in other spectrums. Due to the reasons which will be introduced later, it is overly simplistic to treat cyber risk as just an IT issue. On the contrary, it is a systemic risk that should be tackled systematically and should include interdisciplinary collaboration. Therefore, as will be unfolded in this chapter, although some public-made instruments (such as the Insurance Act 2015) might arguably be useful for providing guidance to counter this issue, their insufficiencies make it highly inadvisable to rely solely upon instruments of this kind. In addition, cyber attacks also require a cross-border regulatory regime because both the shipping and insurance industries are transnational, and a fragmented regulatory landscape would do more harm than good to the smooth running of the shipping industry. This chapter will focus on the building of a mechanism for the reasonable mitigation and allocation of risk and explore the role of cyber insurance as an effective risk management measure against cyber risk in the shipping industry. The main part of this chapter is divided into four sections. Following this Introduction, section 2 will offer a brief introduction to cyber risk, its nature and characteristics and the cover provided by insurers to the maritime industry. The third will examine the role of cyber insurance as a regulatory tool in the shipping industry. However, although cyber insurance can contribute to tackling the cyber risk issue, it is not without flaws. This section will also cover these issues. Section 4 will examine the hurdles that impede cyber insurance from being fully functional. In particular, this section will consider how cyber insurers’ regulatory role in the UK might be hurdled by the Insurance Act 2015. In section 5, the potential measures which can promote cyber insurance’s regulatory function will be discussed.

2.

THE RISK

Cyber incidents are becoming more common and more frequent. According to the Cyber Security Breaches Survey 2021 published by the UK government, 39 per cent of businesses had reported cyber security breaches or attacks in the preceding 12 months.12 In the maritime industry, the number of cyber attacks rose by 900 per cent within a three-year timeframe.13 Cyber insurance coverage differs significantly between industries. For instance, the education and health industries tend to have the highest percentage of cyber insurance coverage while other sectors, such as financial institutions, might only have 30 per cent coverage. For detail, see International Association of Insurance Supervisors (IAIS), ‘Cyber Risk Underwriting Identified Challenges and Supervisory Considerations for Sustainable Market Development’ www​.iaisweb​.org/​uploads/​2022/​01/​201229​-Cyber​ -Risk​-Underwriting​_​-Identified​-Challenges​-and​-Supervisory​-Considerations​-for​-Sustainable​-Market​ -Development​.pdf accessed 9 March 2023. 10 ‘Digital disruption in insurance: cutting through the noise’ www​.mckinsey​.com/​~/​media/​mckinsey/​ industries/​financial​%20services/​our​%20insights/​time​%20for​%20insurance​%20companies​%20to​%20face​ %20digital​%20reality/​digital​-disruption​-in​-insurance​.ashx accessed 9 March 2023. 11 ‘Cyber insurance, 2021 update – thematic research’ www​.globaldata​.com/​media/​insurance/​cyber​ -insurance​-industry​-exceed​-20bn​-2025​-says​-globaldata/​ accessed 24 December 2022. 12 See n 4. 13 ‘Cyber attacks on maritime OT systems increased 900% in last three years’ https://​safety4sea​.com/​ cyberattacks​-on​-maritime​-ot​-systems​-increased​-900​-in​-last​-three​-years/​ accessed 24 December 2022. 9

308  Research handbook on marine insurance law Vessels have many kinds of electronic equipment onboard and much of it is constantly connected to the internet, including automatic identification systems (AIS) and GPS. Attackers can easily find gateways into the vessel’s onboard systems and take control of the vessel or install ransomware or change the reported location of the vessel.14 Cyber attackers can also attack shipping companies’ management systems, thus preventing the loading or discharge of cargo, as was seen in the recent Maersk incident.15 By crippling a business and disrupting the operation of maritime information technology (IT) and operational technology (OT),16 cyber attacks can interfere with the smooth running of global shipping. The shipping industry is thus facing significant cyber risk and must acknowledge the threat and develop means to mitigate the risks.17 While the issue is pervasive, the definition of cyber risk is a murky one.18 Scholars have advanced several definitions from the perspectives of technology,19 sociology20 and anthropology.21 Since the distinction of cyber risks is not the focus of this chapter, it is sufficient to understand that cyber risks come from cyber threats or breaches. There are two kinds of cyber threat. The first involves malicious acts, which may include data theft and cyber extortion,22 and the second non-malicious acts, which may include unintentional acts such as human error and accidental loss of data.23 Not all malicious acts come from the external cyber attacks, such as malware or phishing attacks, that are common in the shipping industry. They may also include internal threats such as employees’ theft of valuable or confidential information. There are various forms of cyber risk, but the four most common in the shipping industry are malware, ransomware, denial-of-service (DoS) attacks and phishing.24 With the development of cyber attack techniques, the emergence of new threats is inevitable. The cyber threat is constantly evolving and, especially in the malicious attack scenario, new techniques can always be deployed to exploit vulnerabilities in IT systems and thus undermine cyber defences.

14 S. Khandker, H. Turtiainen, A. Costin and T. Hämäläinen, ‘Cybersecurity Attacks on Software Logic and Error Handling within AIS Implementations: A Systematic Testing of Resilience’ (2022) 10 IEEE Access 2949–505. 15 ‘Cyber attack hits Maersk Group again’ www​.cybersecurity​-insiders​.com/​cyberattack​-hits​-maersk​ -group​-again/​accessed 24 December 2022. 16 See ‘IET Standard, Code of Practice Cyber Security for Ships, Department for Transport, UK’ https://​assets​.publishing​.service​.gov​.uk/​government/​uploads/​system/​uploads/​attachment​_data/​file/​ 642598/​cyber​-security​-code​-of​-practice​-for​-ships​.pdf at p.5 accessed 9 March 2023. 17 ‘Shipping is falling short in cyber preparedness’ https://​lloydslist​.​maritimein​telligence​.informa​ .com/​LL1139994/​Shipping​-is​-falling​-short​-in​-cyber​-preparedness accessed 24 December 2022. 18 See ‘The role of insurance in managing and mitigating the risk’, HM Government, March 2015 MARSH, www​.gov​.uk/​government/​publications/​uk​-cyber​-security​-the​-role​-of​-insurance at p.8 accessed 9 March 2023. 19 M. Bob Kao, ‘Cybersecurity in the Shipping Industry and English Marine Insurance Law’ (2021) 45 Tul Mar LJ 467, 473. 20 Leo P. Martinez, ‘Cyber Risks: Three Basic Structural Issues to Resolve’ in Pierpaolo Marano and Kyriaki Noussia (eds), InsurTech: A Legal and Regulatory View (Springer 2019) 211–12. 21 Ibid. 22 Dean Armstrong, Thomas Steward and Shyam Thakerar, ‘Cyber Extortion’ in Cyber Risks and Insurance: The Legal Principles (Bloomsbury Professional 2021) 83. 23 Ibid 132–3. 24 See n 8 at p.13.

The role of insurance in regulating cyber risks in the shipping industry  309 This work will not illustrate all forms of cyber risk nor distinguish these risks from the perspective of the terminology. However, it will summarise the common features shared by all forms of cyber risks. The first feature is the transnational nature of cyber risks, meaning cyber risk is not confined to any one geographical area.25 Cyber space is transnational. Attacks can originate from any country and be aimed at any computer anywhere. The risks span every industry,26 including the maritime industry and even healthcare. Some of the industries that have been affected by cyber risks are also transnational in nature, such as finance and maritime. These transnational industries in turn enhance the transnational nature of the cyber risks. Cyber attacks can also be initiated by anyone around the world, and therefore the issue of jurisdiction to deal with the cyber risks and regulate the potential cyber attacks may arise, namely, which country should have the power or responsibility to deal with cyber-related issues. Fragmentation of the regulation of cyber risk is another potential issue, given the risk’s transnational nature, as there are currently no effective international instruments to deal with cyber risk. Different nations may have different ways of dealing with cyber risk, and the detailed rules adopted by nations can also differ; even the definitions of cyber risk differ from one nation to another, and this fragmentation can significantly reduce the effectiveness of regulatory rules. The second feature is that cyber attackers, although traceable, are hard for national authorities to identify and deter;27 thus it is difficult to hold them accountable for the losses they cause. This is because the internet is anonymous and anyone with the ability to access the internet is a potential attacker. Identification of the machines used to mount cyber attacks is complex because of the decentralised nature of the internet.28 The third feature is that, like other ever-evolving technologies, the techniques underlying cyber attacks also evolve, and thus risk management and assessment can be complex.29 The adoption of emerging technologies such as IoT and MASS will expose new vulnerable points to cyber attackers and thus significantly expand the cyber risk landscape. Meanwhile, conventional cyber attack measures are also evolving, with new types of malware and ransomware emerging. These new attack methods may undermine existing cyber security measures and increase the chance of breach, which will further affect the construction of a regulatory framework. The fourth feature is that, unlike traditional marine risks, which are purely risks for the client, cyber risks are for both the client and the insurers. The insurance companies which provide cyber insurance policies will not be immune from cyber attacks and due to the large volume of information held by the insurers and their nature as financial entities, they have increasingly become the targets of cyber attacks.30

25 Alexandra Perloff-Giles, ‘Transnational Cyber Offenses: Overcoming Jurisdictional Challenges’ (2018) 43 The Yale Journal of International Law 192. 26 See general IMF Working Paper: Cyber Risk Market Failures and Financial Stability WP/17/185 by Emanuel Kopp, Lincoln Kaffenberger and Christopher Wilson. 27 Alexandra Perloff-Giles (n 25) at 195. 28 IMF Working Paper (n 26) 6–7. 29 Ibid 6. 30 Vikki Davies, ‘Five cybersecurity threats hitting insurance companies in 2022’ https://​ insurtechdigital​.com/​insurtech/​5​-cybersecurity​-threats​-hitting​-insurance​-companies​-in​-2022 accessed 24 December 2022.

310  Research handbook on marine insurance law

3.

CYBER INSURANCE AS A CYBER RISK GOVERNANCE INSTRUMENT

Given the features of cyber risk, a more dynamic and diversified regulatory framework should be constructed, and regulatory instruments from public sectors should not be the only choice to govern the cyber issue in the maritime industry due to their deficiencies. A common way to deal with a potential risk is to obtain insurance. Compared to other more traditional types of insurance, cyber insurance is still at a primitive stage and it is quite common to find exclusions and exceptions in cyber insurance policies.31 The purpose of cyber insurance is not to prevent cyber attacks or promise immunity from cyber incidents as insurance alone cannot provide the necessary tools such as anti-virus protection or software design. The function of cyber insurance is, as with conventional insurance categories, risk management.32 This section discusses these issues. 3.1

Private Governance and the Insurance Industry

Cyber insurance and its providers will play a critical governance role in the future regulation of cyber risks. This kind of governance is by no means powered by public authorities and neither will its authority derive from statutory instruments. Governance through cyber insurance constitutes self-regulation or private governance. Private governance is a phenomenon where private actors try to pursue public good via rulemaking, implementation of non-mandatory instruments such as guidelines or standard contracts and other means of dispute resolution.33 In the commercial sphere, Trakman argues it can be traced back to medieval Europe, when law merchants and merchants’ courts dominated the commercial disputes between merchants from different regions.34 The medieval merchant law showed the ability of the merchants to regulate their business affairs within the broad

31 ‘Cyber-risk insurance: not as easy as you would think’ https://​lloydslist​.​maritimein​telligence​ .informa​.com/​LL1139955/​Cyber​-risk​-insurance​-not​-as​-easy​-as​-you​-would​-think accessed 24 December 2022. 32 There is a debate as to whether cyber insurance is a risk management system or risk transfer system. In general, please see Livashnee Naidoo, ‘Conceptualising the Cyber Insurance Product: How Insurance can Enhance Risk Management Frameworks’ British Insurance Law Association Journal Platinum Special Issue (2022) https://​bila​.org​.uk/​wp​-content/​uploads/​2022/​11/​9​-Conceptualising​-the​ -Cyber​-Insurance​-Product​-Dr​-Livashnee​-Naidoo​.pdf accessed 4 November 2023. 33 Layna Mosley, ‘Private Governance for the Public Good? Exploring Private Sector Participation in Global Financial Regulation’ in Helen V. Milner and Andrew Moravcsik (eds), Power, Interdependence and Nonstate Actors in World Politics (Princeton University Press 2009) 126–46. 34 William Tetley, ‘The General Maritime Law: The Lex Maritima’ (1994) 20 Syracuse Journal of International Law and Commerce 105–45; G.W. Paulsen, ‘An Historical Overview of the Development of Uniformity in International Maritime Law’ (1983) 57 Tulane Law Review 1065; A.N. Yiannopoulos, ‘The Unification of Private Maritime Law by International Conventions’ (1965) 30 Law and Contemporary Problems 70.

The role of insurance in regulating cyber risks in the shipping industry  311 framework of a legal order.35 Although this assertion is not without its critics,36 it is undeniable that merchant groups do possess a significant degree of self-regulation. The self-regulatory regime in the insurance industry also has a historical background. Rooted in lex mercatoria, the private character of insurance is prominent. Through providing evidence regarding merchant customs and interpreting contracts,37 some early merchant rules and widely accepted customs were first recognised by English courts and later codified into the Marine Insurance Act 1906.38 The insurance merchants’ self-regulation was not only limited to the adoption of customs by the court or the formation of modern substantive insurance principles but a dispute resolution mechanism was also constructed. In medieval Europe, merchants could have their own courts, with ‘judges’ selected by them who applied their ‘law’.39 The insurance industry was not an exception and an independent arbitration centre was even founded by merchants in London, which has been suggested as a symbol of the existence of private ordering40 – although such ordering was still subject to influence and intervention from public authorities.41 These independent merchant courts ceased to exist with the enactment of the Marine Insurance Act 1906, but insurance markets can still have a significant influence on the legislative process.42 Regulation or governance beyond the state has been widely discussed and analysed,43 and several common features identified. First, regulation is about changing the regulatees’ behaviour rather than just allowing or disallowing the regulatees’ acts through normative instruments such as laws.44 Second, private governance means the shift of regulatory responsibility from public authorities to private actors. Third, private regulatory rules are voluntary rather than mandatory. The enforcement of a market-based regulatory scheme relies in general on private autonomy and freedom of association.45 Its legitimacy largely depends upon the regulatees’

Leon E. Trackman, ‘The Twenty-First-Century Law Merchant’ (2011) 48 American Business Law Journal 775–834. 36 Emily Kadens, ‘Order within Law, Variety within Custom: The Character of the Medieval Merchant Law’ (2004) 5 Chicago Journal of International Law 39, 44 and 47. 37 Jeffrey Thomson, ‘A History of English Marine Insurance Law: Merchants, Their Practices, the Courts and the Law’ in Andrew Hutchison and Franziska Myburgh (eds), Research Handbook on International Commercial Contracts (Edward Elgar Publishing 2020) 199–200. 38 Ibid. 39 W.S. Holdsworth, ‘The Development of the Law Merchant and Its Courts’ in Association of American Law Schools (ed.), Select Essays in Anglo-American Legal History (Little, Brown 1907) 293 which states that ‘There are three periods of the law merchant courts, first, the period when the law merchant maritime and commercial is administered in local courts’. 40 Jeffrey Thomson (n 37) at 207–8. 41 Ibid. 42 Ibid. 43 For the general theory of self-regulation, see J. Black, ‘Constitutionalising Self-Regulation’ (1996) 59 Modern Law Review 24; please also see F. Cafaggi, ‘The Many Features of Transnational Private Rule-Making: Unexplored Relationships between Custom, Jura Mercatorum and Global Private Regulation’ (2015) 36 U of Pennsylvania J of International Law 105. 44 Julia Black, ‘Learning from Regulatory Disaster’, LSE Law, Society and Economy Working Papers 24/2014 London School of Economics and Political Science Law Department https://​eprints​.lse​ .ac​.uk/​60569/​1/​WPS2014​-24​_Black​.pdf accessed 9 March 2023. 45 Fabrizio Cafaggi, ‘Enforcing Transnational Private Regulation: Models and Patterns’ in Fabrizio Cafaggi (ed), Enforcement of Transnational Regulation: Ensuring Compliance in a Global World (Edward Elgar Publishing 2012) 79. 35

312  Research handbook on marine insurance law willingness to obey it.46 Finally, private governance regimes are normally not self-contained47 and depend on the existence of international or domestic public institutions to function. The most fundamental rationale for the private sector to take a role in the regulation of social affairs is that it and state authorities have the common goal of preventing loss from happening and maintaining a secure society, but will achieve that same goal through different routes. The state can achieve it through providing public services such as healthcare, transportation or legislation; the insurance industry will exercise its influence by protecting against future losses. By providing insurance coverage such as life, medical, health, liability and property, insurers can help to combat moral hazard and change policyholders’ behaviour. The essence of insurance is the allocation of risk. Through the allocation of risk and the arrangement of insurance contracts, the insurance industry can transfer the responsibility to minimise and reduce risk to every policyholder. This mechanism requires every private actor to take all the necessary preventive security measures to reduce risk, and they can be stimulated by insurance policy wording stating they must do what is required by the insurer or face loss of coverage. The insurance industry will also use statistics or information regarding the clients’ previous behaviour, their security record and the resilience of their security systems to measure the risk exposure.48 Obtaining this information will also enable insurers to monitor the client’s activities. 3.2

The Shortcomings of Public-made Instruments in the Regulation of Cyber Risk

The role of cyber insurance as a regulatory instrument also depends on the deficiencies of the current legal framework regarding cyber attacks. The focus here is on the introduction of their common shortcomings, rather than on the specific rules within each instrument. The effect of cyber risks on the shipping industry and how to regulate these risks have been examined from the perspective of public-made rules such as international maritime conventions,49 with questions raised such as whether a personal injury sustained on a cruise ship caused by a cyber incident should still be covered by the Athens Convention and whether cyber-worthiness should be treated as a kind of seaworthiness under the Hague–Visby Rules.50 The change of liability regime might be an answer to the management of cyber risk, at least in the transportation sector.51 Naturally, given the transnational nature of cyber risk, it is desirable to have international public instruments to deal with this issue. Consequentially, in the maritime sphere, inter-

Ibid 77. Fabrizio Cafaggi, ‘New Foundations of Transnational Private Regulation’ (2011) 38 Journal of Law and Society 29–49. 48 Richard V. Ericson, Aaron Doyle and Dean Barry, Insurance as Governance (University of Toronto Press 2003) 52–3. 49 George Leloudas, ‘Cyber Risks, Autonomous Operations and Risk Perceptions: Is a New Liability Paradigm Required?’ in Baris Soyer and Andrew Tettenborn (eds), Artificial Intelligence and Autonomous Shipping: Developing the International Legal Framework (Hart Publishing 2021) 101–18. 50 For a brief discussion on this point please see ‘Shipping needs to be vigilant over cyber risks’ https://​ lloydslist​.​maritimein​telligence​.informa​.com/​LL1138256/​Shipping​-needs​-to​-be​-vigilant​-over​-cyber​-risks accessed 9 March 2023. 51 Jay P. Kesan, Ruperto P. Majuca and William J. Yurcik, ‘Cyber Insurance as a Market-based Solution to the Problem of Cybersecurity: A Case Study’ [2005] WEIS 23–6. 46 47

The role of insurance in regulating cyber risks in the shipping industry  313 national instruments to reduce cyber risks, such as the Guidelines on Maritime Cyber Risk Management (MSC-FAL.1/Circ.3) and Resolution MSC.428(98),52 have been implemented by the IMO. Doubtless, such instruments will provide basic and relatively unified guidelines regarding how to prevent and deal with cyber risks. However, an international effort of this kind may be undermined, for the following reasons. First, the function of international instruments relies upon cooperation between state authorities; however, state authorities themselves may be the source of the cyber attacks,53 and therefore a business or other private organisation could be the victim of a state-sponsored cyber attack. Such a government would be unlikely to follow the rules and regulate itself. Since the construction of an international positive regulatory framework relies on cooperation, countries may be unable to achieve consensus easily even if they are under the same regulatory framework. An example is UNGGE, which was established by the UN to make states act responsibly in cyber space.54 However, only 25 states have joined the programme and the group has failed to reach a consensus on norms of behaviour.55 The existence of an international harmonised hard-law regime dealing with cyber security issues worldwide would bring great benefits.56 However, the cyber security rules, or even the definition and boundaries of key concepts in this field such as cyber security, cyber attack or cyber risk, can be significantly different from one jurisdiction to another – let alone the various measures or other regulations that have been adopted and implemented by different countries.57 In addition, the world is facing deglobalisation and competition between countries and jurisdictions is now greater than cooperation. What makes it worse in the cyber security arena is that countries including the US, China and Russia have been repeatedly accused of initiating cyber attacks against other nations and organisations.58 Cyber security is a threat not only to international commercial interests, but also to national security.59 Therefore, those countries that have technological power and the ability to initiate cyber attacks will only adopt cyber security laws and instruments that reflect their self-interest. The international cooperation needed to deal with this issue does not exist. An example is the application of one of the most notable international agreements in this field, the Council of Europe’s Convention on Cybercrime, or the Budapest Convention.60 Article 22(1)(a) of the Convention requires signatories to recognise computer crimes that are committed in their territory and Article 23 requires them to cooperate to the widest extent possible,

52 For complete versions of both instruments see www​.imo​.org/​en/​OurWork/​Security/​Pages/​Cyber​ -security​.aspx accessed 24 December 2022. 53 Department of Transport, ‘Code of Practice Cyber Security for Ships, Institution of Engineering and Technology’ A.2.6. 54 See www​.un​.org/​disarmament/​group​-of​-governmental​-experts/​ accessed 24 December 2022. 55 Benjamin Ang, ‘International Law And Interaction Between States’ in Helen Wong (ed.), Cyber Security: Law and Guidance (Bloomsbury Professional 2018) 353, 363. 56 Anton N. Didenko, ‘Cybersecurity Regulation in the Financial Sector: Prospects of Legal Harmonisation in the European Union and Beyond’ (2020) 25 Uniform Law Review 127. 57 Naturally, the construction of different countries’ positive law words can also be different. 58 ‘E.U., U.K. and U.S. accuse Russia of cyberattack on internet provider’ www​.nbcnews​.com/​tech/​ security/​eu​-uk​-accuse​-russia​-cyberattack​-internet​-provider​-rcna28086 accessed 24 December 2022. 59 ‘Cyberconflicts and National Security’ www​.un​.org/​en/​chronicle/​article/​cyberconflicts​-and​-national​ -security accessed 24 December 2022. 60 For the whole context of this Convention please see https://​rm​.coe​.int/1680081561 accessed 24 December 2022.

314  Research handbook on marine insurance law including in the collection of evidence.61 Therefore, it would seem that a mutual legal assistance regime has been constructed. However, there are still some limitations and a failure to provide such assistance in some circumstances.62 The use of soft law might be a solution since its enactment is easier than the adoption of an international positive law instrument. However, the challenge with soft law instruments is that they lack teeth and their effect or application relies on each national government hence differences may arise and the effectiveness of these instruments will be weakened.63 The second limitation of public instruments, international or domestic, is that they are mostly principle-based rules without clear guidance for the regulatees to follow. For instance, the most important positive law instrument on the EU level which touches on cyber risk is the General Data Protection Regulation (GDPR).64 Applicable in all the EU member states,65 it consists of principles rather than operable rules66 and Sections 32, 33 and 34 are concerned with the prevention of cyber attacks.67 Rules of this kind usually mandate ‘implement[ation of] appropriate technical and organisational measures to ensure a level of security appropriate to the risk’68 or place organisations under an obligation to take all necessary measures against potential cyber attacks.69 However, they fail to explicitly state what kinds of measures are required.70 These general and vague words are normally the product of technological neutrality, which means that, given the fast-changing nature of technology evolutions, statutory wordings are always trying to keep up, and that the use of vague and general wordings and principle-based rules makes the constant revision of legal instruments unnecessary. This logic is not without its merits, but the subjects of these instruments will lack clear-cut rules to follow71 and interpretation of these rules would be subject to the discretion of both the regulators and regulates, resulting in differences of opinion and thus non-compliance by the regulatees, from the regulators’ perspective.72 Another drawback is that regulators normally lack expertise in cyber security issues. Therefore, except for some very fundamental measures such as the use of firewalls, secure routers and the establishment of cyber breach protocols, measures mentioned in positive law instruments such as GDPR are advisory rather than mandatory.73 Given the unpredictable nature and scale of the potential cyber challenges ahead,

Ibid. Ang (n 55) 365. 63 Kenneth W. Abbott and Duncan Snidal, ‘Hard and Soft Law in International Governance’ (2000) 54 International Organization 421–56 at 422. Please also see Vera Korzun, ‘Enforcing Soft Law in International Investment Arbitration’ (2023) 56 Vanderbilt Journal of Transnational Law 1–65 at 3. 64 For the official edition of GDPR please see https://​gdpr​-info​.eu/​. GDPR is not an instrument aiming to regulate cyber issues only, but some sections do address the subject. 65 After Brexit the equivalent version of GDPR that is applicable in the UK is the Data Protection Act 2018: please see www​.legislation​.gov​.uk/​ukpga/​2018/​12/​contents/​enacted. 66 Damien Geradin, Dimitrios Katsifis and Theano Karanikioti, ‘Google as a De Facto Privacy Regulator: Analysing the Privacy Sandbox from an Antitrust Perspective’ [2021] European Competition Journal 24. 67 Didenko (n 56) 133. 68 GDPR Article 32(1). 69 Didenko (n 56) 138. 70 Ibid 139. 71 Ibid 139. 72 Ibid. 73 Ibid 142. 61 62

The role of insurance in regulating cyber risks in the shipping industry  315 it is hard for authorities to anticipate what kind of specific measures should be taken by the regulatees to tackle each cyber incident.74 Finally, even though some positive law instruments may refer to standards developed by private actors for the regulatees to follow, the wording of these instruments might be vague. They might use the words such as ‘the most internationally recognised’ or ‘well-accepted’ to describe the regulations that should be followed, but this is ambiguous. For instance, if a regulation passed by an international organisation is only adopted by a handful of countries, is it internationally adopted?75 Questions of this kind can significantly diminish the effectiveness of public regulatory bodies. In practice, the application of the GDPR constantly needs guidelines and explanations from the EU.76 There is currently no positive law instrument similar to GDPR in the maritime sector dealing with cyber issues. However, as has been mentioned, the IMO has taken action to deal with cyber risk by adopting Resolution MSC.428(98) and Guideline MSC-FAL.1/Circ.3. But these two instruments also share the shortcomings of GDPR. Resolution MSC.428(98) states that ‘an approved safety management system should take into account cyber risk management under the objectives and functional requirements of the ISM Code’77 and encourages administrations to properly address cyber risks in their safety management systems. Guideline MSC-FAL.1/ Circ.3 also recommends shipping management companies take action to safeguard shipping from current and emerging cyber threats. However, what is meant by ‘properly address’ and what specific actions should be taken to ‘respond to the risk’ are not specified. The instruments are also advisory in nature and more specific guidelines regarding how to deal with a particular category of cyber risk are absent. Therefore, although these two maritime-related instruments can raise awareness of cyber risk in the shipping industry, they are of limited use given that the IMO itself has been the victim of cyber attacks recently.78 The third limitation of the application of existing international law is that the boundary of cyber attacks defined in these instruments can be too narrow and most cyber attacks will not be covered or regulated, leaving private companies unprotected. This is a prominent feature in instruments relating to acts of war. For instance, under the Law of Armed Conflict (LOAC), a cyber attack is defined by Rule 92 of the Tallinn Manual 2.0 as ‘a cyber-operation that is reasonably expected to cause injury or death to persons or damage or destruction to objects’.79 Thus cyber incidents which may have resulted in significant losses will not be considered as cyber attacks if they do not cause injury, death or destruction.80 The circumstance is the same in the maritime industry. Although Resolution MSC.428 (98) and Guideline MSC-FAL.1/Circ.3 pose some recommendations regarding how to deal with cyber issues, they are drafted in broad terms and they all fail to draw a clear ambit of cyber

Ibid 146. Ibid 150–1. 76 Halme, Ria. ‘The Law’ in Helen Wong (ed.), Cyber Security: Law and Guidance (Bloomsbury Professional 2018) 49–102 at 63. 77 ‘Maritime Cyber Risk Management in Safety Management Systems’ www.cdn.imo.org/​ localresources/en/OurWork/Security/Documents/Resolution%20MSC.428(98).pdf accessed 24 December 2022. 78 See www​.seatrade​-maritime​.com/​technology/​imo​-hit​-cyber attack accessed 24 December 2022. 79 Michael N. Schmitt and Liis Vihul, Tallinn Manual 2.0 on the International Law Applicable to Cyber Operations (CUP 2017) 415. 80 Ang (n 55) 363. 74 75

316  Research handbook on marine insurance law risks. Naturally, since the target is unclear, these two instruments fail to define adequate and practical measures that can be taken to achieve the protection against cyber attacks. The enforcement and consequences of breaching public-made rules are another issue for compliance by regulatees. Some scholars argue that the harmonisation of positive law could be achieved through the harmonisation of key principles or substantive rules.81 They even argue that transnational instruments should have enforceable mandatory provisions to secure compliance.82 Tempting as it is, the current fact is that those abovementioned public-made rules lack teeth. They are either recommendations by international organisations or they lack penalties if a violation occurs. The economic penalties are not strong enough to deter cyber attacks or encourage organisations to take adequate measures.83 In some rare scenarios where penalties and fines do exist, they might be imposed by public regulatory authorities because of breaches of obligations by regulated entities. However, not only are the fines too low, but under some cyber insurance policies they can be recovered.84 The penalties can also differ from one country to another. This fragmentation of the penalty regime would undermine the function of these instruments since cyber issues are transnational in nature and therefore should be dealt with transnationally, otherwise a potential cyber weak point that has been considered carefully in one country might be ignored in another and undermine the whole system. Finally, because of the evolving nature of cyber risk, the legislative and regulatory frameworks that underlie the regulatees’ cyber security obligations and responsibilities are also evolving, and it is difficult for positive law instruments to keep up. National legislative authorities lack knowledge of this technological challenge and the formation of a public-made instrument is time-consuming. The legislative process is slow and struggles to deal with the fast-changing issues posed by technology, and by the time an instrument has been made, the cyber risk may have moved on.85 As well as public-made rules, there are target-led initiatives issued on a transnational level, such as ‘TIBER-EU’, a testing framework that aims to simulate cyber attacks and test the resilience of the entity’s systems.86 However, this is still a voluntary project and the level of adoption can be different from one country to another, which may significantly undermine it.87 Given the significant deficiencies of public-made rules, they are unsuitable as the only method to regulate cyber issues. Although instruments from public institutions may not be able

Didenko (n 56) 151. For the recent development of this solution in EU please see ‘State of the Union: new EU cybersecurity rules ensure more secure hardware and software products’ https://​ec​.europa​.eu/​commission/​ presscorner/​detail/​en/​IP​_22​_5374 accessed 17 March 2023. 83 Didenko (n 56) 145. 84 OECD, ‘Encouraging clarity in cyber insurance coverage: the role of public policy and regulation’ www​.oecd​.org/​finance/​insurance/​Encouraging​-Clarity​-in​-Cyber​-Insurance​-Coverage​.pdf 16 accessed 24 December 2022. However, it is also noticeable that in some other countries the regulatory fines, whether criminal or administrative in nature, are not recoverable under insurance. Ibid 18. 85 These drawbacks are shared by most public-made instruments. For critiques of the most recent EU effort to regulate cyber issues please see Vagelis Papakonstantinou, ‘Cybersecurity as Praxis and as a State: The EU Law Path towards Acknowledgement of a New Right to Cybersecurity?’ (2022) 44 Computer Law & Security Review 105653. 86 Didenko (n 56) 132. 87 Ibid. 81 82

The role of insurance in regulating cyber risks in the shipping industry  317 to tackle cyber security issues directly, they can still stimulate the demand for cyber insurance to fulfil statutory obligations.88 3.3

Cyber Insurance as a Regulatory Tool

Regulation through insurance is a topic with abundant research89. In brief, insurers can excise their regulatory power by regulating or changing the policyholders’ behaviour. One prominent example to demonstrate the effectiveness of insurers’ regulatory role is the significant enhancement of driving security by way of motor insurance.90 For competition, premium and business development reasons, insurers have incentives to encourage assureds to mitigate risks.91 Taking the cyber insurance scenario, there are three reasons for cyber insurance to perform its potential regulatory role. The first is that, because of the pervasive internet vulnerabilities and increased reliance on digital technologies by the maritime industry and wider society, cyber insurance has become more common. Therefore cyber insurers can exercise pre- and post-contractual regulatory power through cyber insurance policies and their terms and conditions. These requirements can be viewed as a method of regulation and can stimulate the client to adopt more cyber-resilient systems, thereby enabling a reduction in the number of cyber attacks. This argument is also based on the theory that standard contracts are a self-regulatory tool.92 In the insurance industry, the use of standard contracts is common and cyber insurance should not be an exception.93 The second reason is the inability of traditional public-made rules or traditional insurance to regulate or cover these new risks. There is a lack of comprehensive, effective, consistent and transnational legal regulations or other public institutional instruments to deal with cyber risks, and therefore the cyber security issue has not been dealt with adequately. Review of public instruments in the maritime sphere (such as IMO regulations) shows that they lack practical and detailed rules for the regulatee to follow. However, we still see insurers stepping in and providing cyber insurance. This kind of insurance will normally provide not only first-party loss coverage but also third party liability coverage for cyber attacks.94 Cyber insurance will encourage best practice in information security management by providing a detailed set of rules relating to the prevention and detection of cyber risks. The last reason is that the cyber insurance market is getting strong enough and large enough to gain significant governance power. As the cyber insurance market expands, insurers can manage the conduct of the client through the arrangement and wording of insurance terms. Ibid 135. Omri Ben-Shahar and Kyle D. Logue, ‘Outsourcing Regulation: How Insurance Reduces Moral Hazard’ (2012) 111 Mich. L. Rev. 197. 90 Ibid 220–3. 91 Ibid 204–5. 92 C.M. Schmitthoff, ‘The Unification or Harmonisation of Law by Means of Standard Contract and General Conditions’ (1968) 17 International Comparative Law Quarterly 551. 93 As will be highlighted later, currently cyber insurance policies still lack standardisation; however, it is noticeable that this issue has been identified and efforts have been made to address it. Please see Saket Modi, ‘The success of cyber insurance lies in risk standardization’ www​.securitymagazine​.com/​ articles/​97391​-the​-success​-of​-cyber​-insurance​-lies​-in​-risk​-standardization accessed 24 December 2022. 94 Shauhin A. Talesh and Bryan Cunningham, ‘The Technologization of Insurance: An Empirical Analysis of Big Data and Artificial Intelligence’s Impact on Cybersecurity and Privacy’ (2021) 5 Utah Law Review 968, 973. 88 89

318  Research handbook on marine insurance law Because of the expansion of the market, more potential organisations will be included in this regulatory framework and the management capabilities of the insurers will become more embedded in all aspects of society. Ultimately, as cyber insurers gain strength, their bargaining power will increase, leading to enhanced management of the client’s behaviour. 3.3.1 The origin of the normative force of cyber insurers The normative force of cyber insurers’ regulatory power derives from insurance companies’ role as corporate regulators and their power to change policyholders’ behaviour.95 They can act as regulators over individuals and organisations because through their underwriting processes, contractual terms and risk mitigation services, they have the ability to shape policyholders’ behaviour and establish industry-wide norms and best practices in cybersecurity.96 Effective regulation requires significant amount of information of the regulatees.97 Cyber insurers’ regulatory power is also embedded in information asymmetry. This information asymmetry is not only between insured and insurer but also between insurers and governments. Insurers have access to more information disclosed by policyholders but not obtained by governments. Benefiting from this informational advantage, insurers can reduce clients’ moral hazard behaviour and provide more detailed and precise cover which may supersede some regulatory frameworks offered by governments.98 This unique informational advantage enables cyber insurers to identify emerging trends, vulnerabilities and threats, and to adjust their underwriting criteria, pricing and coverage offerings accordingly. Therefore, the clients’ concerns and cyber vulnerabilities can be detected by the insurers and more effective regulatory measures taken. The cyber insurers’ ability to gather risk-related data also enable them to price insurance product efficiently and therefore achieve incentive-based regulation.99 Another factor underlying cyber insurers’ regulatory power is insurance law instruments. Insurance law is not designed to deal with cyber issues directly. It is designed to protect both the interests of insurers and insureds while maintaining the financial stability and integrity of the insurance market. As evidenced in the next section, one way in which insurance law enhances insurers’ regulation of insureds’ behaviour is through the implementation of certain legal principles, regulations and requirements that govern the relationship between the two parties in insurance contracts. Pre-contractual and post-contractual regulation through cyber insurance policies The cyber risk regulatory system, as suggested by the IMF, can be divided into three sections: risk reduction, risk transfer and risk avoidance.100 Cyber insurance has been viewed by the IMF as a powerful way to enable risk transfer and can also be used to reduce risk or largely avoid it101 3.3.2

Shauhin A. Talesh, ‘Insurance Companies as Corporate Regulators: The Good, the Bad and the Ugly’ (2017) 66 Depaul Law Review 463. 96 Kyle D. Logue, ‘Encouraging Insurers to Regulate: The Role (If Any) for Tort Law’ (2015) 5 UC Irvine L Rev 1355, 1359–66. 97 Ibid 1359. 98 Talesh (n 95) at 471. 99 Logue (n 96) at 1360. 100 See Guidance on Cyber Security Onboard Ships (n 8) p.13. 101 Ibid p.15. 95

The role of insurance in regulating cyber risks in the shipping industry  319 because the terms of cyber insurance policies can exert significant regulatory effects in both the pre- and post-contractual stages. One very powerful regulatory tool in insurers’ hands at the pre-contractual stage is premium pricing. Premiums can be different based upon the claim or security history of each insured or the risk presented by different insured. Setting premiums allows insurers to price policies in a way that accurately reflects the level of risk associated with each policyholder. By adjusting premiums based on risk factors, insurers can motivate policyholders to adopt risk-mitigating behaviours or practices and such measures will result in a virtuous circle of cyber security. For example, by offering discounts to policyholders who install security systems or maintain cyber hygiene, insurers can significantly reduce the likelihood of claims and losses, which in turn will further reduce policyholders’ future premiums because of the diminished risk.102 Effective pre-contractual regulation can also be achieved through the policyholders’ performance of their obligations at the stage of formation of insurance contracts. Although the wording of cyber insurance varies from one policy to another and there are no universally accepted forms of cyber insurance contracts,103 some fundamental characteristics are shared. For instance, normally, cyber insurance will cover losses resulting from cyber incidents such as cyber extortion caused by ransomware or losses caused by business interruption.104 Cyber insurance will also provide third party coverage regarding a data breach and any liability payments made by the client to a third party whose data has been contaminated by the client due to a data breach.105 This wide range of coverage means that policy providers have a great interest in monitoring policyholders’ behaviour at both the pre- and post-contractual stages. Therefore, as a precondition to providing coverage, risk assessments at the stage of formation are important for all types of insurance policies including cyber insurance. Under English insurance law, before the formation of an insurance contract, the client is under a duty to make a fair representation of risk, to disclose all material facts which are known or ought to be known to them and to make no misrepresentation of these facts.106 In the cyber insurance scenario, before the formation of a cyber insurance contract, insurers will normally evaluate the client’s cyber resilience and vulnerabilities by asking questions and gaining additional information to overcome the information asymmetry. The insurers may require information on the security measures or arrangements that have been adopted, such as technical measures and procedural arrangements taken by the client to reduce its cyber vulnerability. Other incentives at the pre-contractual stage can also contribute to the control of cyber risk. For instance, insurers can encourage the client to take measures or adopt security defences by offering a discount.107 At the post-contractual stage, various kinds of surveillance tools can be embedded in cyber insurance arrangements to facilitate governance at a distance. This is not unique to cyber

A significant example about how adjustment of premium will influence or change assureds’ behaviour can be found in motor insurance. Please see Ben-Shahar and Logue (n 89) 220–3. 103 Celso de Azevedo, Cyber Risks Insurance: Law and Practice (2nd ed., Sweet & Maxwell, 2022), 7-013. 104 National Cyber Security Centre, ‘Cyber Insurance Guidance’ www​.ncsc​.gov​.uk/​guidance/​cyber​ -insurance​-guidance accessed 24 December 2022. 105 Trey Herr, ‘Cyber Insurance and Private Governance: The Enforcement Power of Markets’ (2021) 15 Regulation & Governance 98 at 101–2. 106 Insurance Act 2015 (UK) www​.legislation​.gov​.uk/​ukpga/​2015/​4/​contents/​enacted accessed 24 December 2022. 107 Talesh (n 95) at 471. 102

320  Research handbook on marine insurance law insurance. A typical example is when the insurer deals with marine risks which are at a significant distance from the insurer’s effective management. As can be seen from English marine insurance law and practice, complicated arrangements and rigid rules have been constructed to reduce the vulnerability of the insurer due to the information asymmetry and to allocate the insurer’s liability. Cyber insurance policies will also contain warranties, conditions and exclusion clauses. These contractual provisions, like their functions in other categories of insurance, will require the client to take all reasonable steps to perform their contractual obligations, such as avoiding and mitigating losses, and will require that this be monitored by the insurer. Therefore, both of the insureds’ pre-contractual and post-contractual obligations can be clearly stated in insurance policies, and if breach occurs consequences will follow.108 In addition, in cyber insurance the insurer’s surveillance also relies on self-regulation by the client, meaning that the client has the responsibility for controlling risk. This can be carried out through standards, guidance and other incentives incorporated into cyber insurance policies to make sure that the client’s business activities are carried out according to high and secured recommendations and standards.109 This is to embed surveillance of the client’s moral risk into the organisational arrangements with the client. Many provisions in insurance policies are designed to address potential moral hazard problems. For instance, the client may not be able to claim if it fails to take reasonable action to prevent the losses from happening. The insured also should mitigate losses once they occur. The guidance or rules provided by the insurer can also ensure that the client will participate in the governance of their own moral risk. Insurers can also carry out governance activities through the underwriting rating system.110 Therefore, unless the client can meet a certain standard and have a resilient cyber security system in place, they may not be able to get coverage. Many cyber insurers can also help with risk management through other services such as risk mitigation and accident response services and by running penetration tests to locate vulnerable points in the system.111 Standard is worthy of mention here, since, as introduced, one way to enhance cyber insurers’ regulatory power is to incorporate technical standards into policies. The insurance industry is data-driven. Given that cyber risk is relatively new and unpredictable, the industry is left with very limited data to understand its nature and impact. Therefore, in both the pre- and post-contractual stages, insurers seek to ensure that state-of-the-art practice is followed by the client to try to prevent cyber attacks. Insurance policies can act effectively by transforming vague standards into bright-line rules and therefore provide regulatees with clear instructions regarding how to prevent and mitigate cyber loss.112 Currently there is no cyber security standard for the marine insurance industry. Although very general guidelines on cyber issues in the maritime industry have been issued by the IMO, they are not technical standards. Therefore, to control cyber risks more effectively from the 108 For example, in Columbia Casualty Co v Cottage Health System 2:15-cv-03432 (C.D. Cal. May. 7, 2015), a US case, there was a warranty about the insured’s obligation to implement risk management measures. The warranty specified ‘the minimum required practices that are listed in the minimum required practices endorsement as a condition of coverage under this policy, and also to maintain all risk controls identified in the insured’s application form and any supplemental information provided by the insured in conjunction with insured’s application form for this policy’. 109 Ericson, Doyle and Barry, Insurance as Governance (n 48) 88. 110 Ibid 89. 111 Ben-Shahar and Logue (n 89) 213–14. 112 Ben-Shahar and Logue (n 89) 231.

The role of insurance in regulating cyber risks in the shipping industry  321 insurers’ perspective, some internationally adopted non-marine standards and cyber risk management programmes should be adopted along with the guidelines from the shipping societies such as the ISM Code. For example, ISO/IEC 27001:2013 is the international standard that sets out the specification for an information security management system.113 Companies that have adopted this standard can manage cyber security risks more effectively.114 Therefore, by transforming this standard into insurance policies, insurers can provide clear rules and detailed instructions to the assureds and therefore, proper care and cyber security levels can be achieved. Through this method, the client can be shown how to avoid and reduce cyber attacks and demonstrate that cyber hygiene and pre-loss standards have been met, and then procure the indemnification. Another useful tool is the exclusion clause. Exclusion clauses are normally used to define the boundary of the insurer’s liability. Insurers providing cyber insurance will understandably not wish to risk their money on companies with poor or insufficient security practices. Therefore, they may include an exclusion clause relating to the minimum required practices to be in place to allow a claim under the cyber insurance policy in the event of an attack.115 In addition to standard exclusion clauses, there are also commonly agreed exclusions in cyber insurance policies. In fact, there are various categories of exclusion clauses and they are widely used by cyber insurers. For instance, tangible property damage and personal injury caused by cyber incidents is normally excluded.116 Therefore, any hardware damage caused by a cyber attack will not be covered. Naturally, cyber insurance will also not cover those losses caused by the client’s use of unproven or illegal software or losses resulting from fraud, dishonesty or reckless conduct of a director or senior officer of the client or losses occurring before the start date of the policy or sometimes before the retroactive date of the insurance policy, even if not discovered until the policy is in force.117 In addition, certain exclusions may also refuse coverage for any claim or loss resulting from a breach event or situation of which the insured ‘should have been aware’ prior to the policy’s commencement date.118 Therefore, exclusion clauses play a critical role in helping insurers manage their risks by ensuring that they do not cover organisations with inadequate security practices, or losses arising from certain predefined circumstances. By understanding and adhering to these exclusions, policyholders can better navigate their cyber insurance coverage and align their cybersecurity practices with insurer requirements.

ISO/IEC 27001:2013 Information technology – Security techniques – Information security management systems – Requirements www​.iso​.org/​standard/​54534​.html accessed 24 December 2022. 114 Mark Camillo, ‘Cyber Risk and the Changing Role of Insurance’ (2017) 2(1) Journal of Cyber Policy 53–63, at 60. 115 For a more comprehensive introduction of more exclusion clauses in cyber insurance policies please see Dean, Steward and Thakerar (n 22) 157–68. 116 This is because of the use of the Tangible Property and Bodily Injury Exclusion clause: see Azevedo (n 103) 9-016. 117 Ibid. 118 This is normally known as Prior Knowledge of Circumstances Exclusion please see Azevedo (n 103) 9-005. 113

322  Research handbook on marine insurance law

4.

THE LIMITS OF CYBER INSURANCE POLICY AS A REGULATORY TOOL

Cyber insurance’s function as a regulatory tool is rooted in the wide adoption of cyber risk-related insurance policies or insurance clauses. In essence, it is a market force, and the wider the adoption of cyber insurance, the stronger the market force it generates. However, there are hurdles in front of regulation through cyber insurance, and such insurance is not a panacea for managing cyber risk. 4.1

The Lack of Information

The first difficulty faced by cyber insurers is the lack of reliable cyber-related information. The insurer might not be able to acquire enough data regarding the client’s previous exposure to cyber risk to calculate premiums or implement new requirements in the policy. This is partly because cyber insurance was not common before and therefore no data could be collected. Another contributory factor is that the client may be afraid that disclosure will result in an increased premium. Therefore, if the cyber attacks experienced by the client were not severe and did not result in substantial damage, they may choose not to disclose them to the insurer. Lack of familiarity with cyber risks might also result in entities not being aware of cyber attacks, and therefore there would be no report to the insurer. Cyber issues are concerns not only for commercial entities but also for nations, and so, for strategic reasons, a nation might be reluctant to share information with another state.119 From the client’s perspective, they may not be familiar with issues regarding cyber insurance policies such as coverage, contractual terms, technological requirements and exclusions. This lack of information may result in suspicion regarding the effectiveness and necessity of cyber insurance policies. This lack of information will also result in a lack of mutual trust between the parties, without which the dynamic regulatory regime described above could not be constructed. Because of the lack of information, cyber insurers will be driven by their own interests and may not act in the client’s best interests. Therefore, there might be a risk that insurers will motivate the client to take measures to prevent cyber incidents and construct a more cyber-resilient world, but take advantage of their possession of all the data and act only for their benefit. With this risk in mind, the assured might be reluctant to take out cyber insurance, and therefore the insurers’ regulatory power would be diminished. The cultivation of trust in the insurance industry is not an easy matter, since the whole regime operates on the assumption of distrust. 4.2

The Wide Existence of Exclusion Clauses

As has been covered, exclusion clauses can play a significant regulatory role. However, the wide adoption of various categories of exclusion clauses will have side-effects on insurers’ regulatory function. One point that may obstruct the regulatory nature of cyber insurance is the ambiguity regarding cyber risk coverage. The losses suffered by cyber attack victims are huge; however,

IMF Working Paper (n 26), 13–14.

119

The role of insurance in regulating cyber risks in the shipping industry  323 the compensation they get from their insurers is small.120 Previously, the most significant exclusion clause was the CL 380 clause.121 Drafted in 2003, long before the arrival of many now commonplace information technologies, it was incorporated into many marine insurance contracts and accepted as a standard market practice until 2019, when LMA 5403 began to be accepted.122 The purpose of CL 380 is to exclude losses caused by cyber attacks unless in very limited circumstances. Its application is viewed by both insured and insurers as outdated given the continuing growth of cyber attacks.123 The other reason for the discontinuity of this clause is that it may give rise to silent cyber issues, which were banned by Lloyd’s in 2019.124 ‘Silent cyber’ means that whether cyber incidents are covered or not is unclear according to policy wording. A policy may afford silent cyber cover if CL 380 is not used, or for non-malicious cyber even if it is. After Lloyd’s Market Bulletin Y5258, all policies are required to be clear on whether coverage is provided for losses caused by a cyber event.125 Thus, following this arrangement, the application of CL 380 can be significantly limited. Other than CL 380, there are two standard exclusion clauses in use. The first is LMA 5402, which is similar to CL 380 and aims to function as a blanket exclusion clause.126 The other is LMA 5403, which only covers non-malicious cyber risk and excludes malicious cyber attacks.127 There is also a Lloyd’s market exclusion for losses resulting from malicious cyber incidents. However, in the circumstances of non-add-on cyber insurance policies, whether terms in traditional insurance policies would cover cyber loss or not is still untested and largely depends on the specific policy wording. This is because cyber losses or cyber attacks are always unforeseeable. Finally, cyber attacks that involve the participation of states are excluded from most marine cargo and hull policies. It is also unclear whether these politically motivated cyber attacks can be covered by war and terrorism insurance policies. After the 2017 NotPetya ransomware attacks, the affected shipping companies were unsuccessful in claiming compensation under the war risk clause.128 Therefore, for now, whether cyber risk can be covered by a war risk

IAIS report (n 9) 1.2.3. Institute Cyber Attack Exclusion Clause (CL 380, 10/11/2003). For the defects and limitations of CL380, especially in the autonomous ships scenario, please refer to Luci Carey, ‘Autonomous Ships and Hull and Machinery Marine Insurance in Maritime Obligation’ in Stephen Girvin and Vibe Ulfbeck (eds), Maritime Organisation, Management and Liability: A Legal Analysis of New Challenges in the Maritime Industry (Hart Publishing 2021) 267–70. 122 Cyber Risk for Insurers – Challenges and Opportunities https://​register​.eiopa​.europa​.eu/​ Publications/​Reports/​EIOPA​_Cyber​%20risk​%20for​%20insurers​_Sept2019​.pdf at pp.18–19, accessed 17 March 2023. 123 See https://​lloydslist​.com/​LL1139955/​Cyber​-risk​-insurance​-not​-as​-easy​-as​-you​-would​-think​?vid​ =​Maritime​&​processId​=​8e7fa3b8​-82d2​-48b8​-a124​-c2bfc0dc503a accessed 24 December 2022 124 ‘Lloyd’s moves to rein in “silent” cyber exposures’ https://​insuranceday​.​maritimein​telligence​ .informa​.com/​ID1127905/​Lloyds​-moves​-to​-rein​-in​-silent​-cyber​-exposures, accessed 24 December 2022. 125 Market Bulletin Ref Y5258 www​.ciab​.com/​download/​19427/​accessed 24 December 2022. 126 See www​.lmalloyds​.com/​LMA/​News/​LMA​_bulletins/​LMA​_Bulletins/​LMA19​-031​-PD​.aspx accessed 24 December 2022. 127 Ibid. 128 ‘Russian military “almost certainly” responsible for destructive 2017 cyber attack’ www.ncsc​ .gov​.uk/​news/​russian​-military​-almost​-certainly​-responsible​-destructive​-2017​-cyber​-attack accessed 24 December 2022. 120 121

324  Research handbook on marine insurance law clause might still be an issue of construction and is yet to be tested in court.129 Not covering state-initiated cyber attacks is the choice of the insurance market, not a requirement of law. A recent example of such may be found in a memo issued by Lloyd’s stating that from 31 March 2023, any losses arising from state-backed cyber attacks and acts of war will no longer be covered if the policy does not have a war clause.130 This is due to the war in Ukraine and the massive scale of cyber attacks initiated by both Ukraine and Russia since then. According to Lloyd’s, to continue providing coverage for such risks would expose the whole insurance market to systemic risks which could be unmanageable.131 Although many add-on cyber insurance policies and clauses are available for insureds now, the sheer volume of exclusions still means a great deal of uncertainty.132 The insurance industry’s reluctance to provide insurance cover for this kind of cyber attack can significantly impair clients’ confidence and trust in providers, given the scale of attacks that can be initiated by nation states, and thereby weaken cyber insurance’s role as a regulatory tool.133 4.3

The Vulnerabilities of the Insurance Industry

Some fundamental changes have been brought about by the rapid development of technologies, and the shipping industry is not the only one facing digital challenges. In the insurance sector, the use of technology to innovate or disrupt is known as ‘insurtech’. Some technologies, such as the IoT, sensors in the smart home and medical devices, will enable the insurer to access abundant amounts of data, such as the status of cargo, home burglary-deterrence systems and clients’ health conditions, while other algorithmic technologies such as artificial intelligence and machine learning will enable the insurer to process, analyse and gain insights from these data sources.134 The emerging distributed ledger technologies such as blockchain and blockchain-based smart contracts will make the execution of insurance contracts more cost-effective and reliable.135 The insurance industry could enjoy enormous benefits from this digital or algorithm transformation. For instance, the execution of insurance contracts could be more efficient, the

129 The most recent disputes relating to this issue was Mondelez International Inc v Zurich American Insurance Co was pleaded before the Circuit Court of Illinois. For the trial proceeding see www​.databreachninja​.com/​wp​-content/​uploads/​sites/​63/​2019/​01/​MONDELEZ​-INTERNATIONAL​ -INC​-Plaintiff​-v​-ZURICH​-AMERICAN​-INSURANCE​-COMPANY​-Defenda​.pdf. However, a settlement was reached before the court ruled: www​.law360​.com/​articles/​1544141/​mondelez​-zurich​-settle​ -notpetya​-dispute​-before​-trial​-close both accessed 24 December 2022. 130 Market Bulletin, Ref: Y5381, https://​assets​.lloyds​.com/​media/​35926dc8​-c885​-497b​-aed8​-6d2f87​ c1415d/​Y5381​%20Market​%20Bulletin​%20​-​%20Cyberattack%20exclusions.pdf accessed 24 December 2022. 131 Ibid. 132 For the meaning and extends of all those available cyber insurance exclusions please refer to Azevedo (n 103) Chapter 9. 133 National Cyber Security Centre, ‘NCSC Annual Review 2022’, www​.ncsc​.gov​.uk/​collection/​ annual​-review​-2022/​threats​-risks​-and​-vulnerabilities/​state​-threats accessed 24 December 2022. 134 ‘A practitioners’ guide to insurtech’, www​.lexisnexis​.co​.uk/​legal/​guidance/​a​-practitioners​-guide​ -to​-insurtech accessed 17 March 2023. 135 Angelo Borselli, ‘Smart Contracts in Insurance: A Law and Futurology Perspective’ in Pierpaolo Marano and Kyriaki Noussia (eds), InsurTech: A Legal and Regulatory View (Springer 2019) 102.

The role of insurance in regulating cyber risks in the shipping industry  325 gathering of data could be easier and new innovative products and services could be designed.136 However, with the application of those digital technologies in both the insurance industry and other industries, both the insurer and insured are facing traditional cyber security challenges such as malware, phishing and DoS attacks. One of the most significant challenges for the insurer is that the possession of these important data has made the insurance industry itself a valuable target of cyber attacks. One of the biggest questions facing cyber insurers is how to handle the sheer volume of data that can be commercially sensitive or affect the privacy of the client and so must be properly stored and guarded. Failure to do so would result not only in economic losses and sanctions but also in reputational damage. In addition to intentional cyber attacks, there are design loopholes in the algorithm or software that are used to gather data and decision-making and may also expose the insurer to cyber risks. Therefore, because of the adoption of digital technologies, the insurance industry’s exposure to cyber risks is also increasing.137 As insurance companies provide coverage for their clients, the insurers’ own security must be assured, particularly given the limited number of cyber insurance providers in this field and the real risk that they might become vulnerable points in this cyber security regime. Consequently, it would be hard to persuade entities to take action to fulfil their obligations under cyber insurance policies if the providers of those policies were not immune from cyber attacks and could potentially themselves become the cause of losses suffered by the client.138 4.4

The Abuse of Power by Cyber Insurers

Insurance companies may not always have positive outcomes as private regulators. The function of the industry’s private governance mechanism does not mean that these regulatory activities should be carried out without oversight by the state because cyber security gaps filled by the insurers, private guidelines issued by them and best practice may not always represent the best interest of the client or the public. For instance, the insurer may require the client only to contact service providers approved by the insurer and may blur the criteria for who is eligible for cyber insurance coverage in order to protect the insurers’ own interests.139 Normally, cyber insurers will provide a set of both pre- and post-cyber breach services, including training sessions, risk-prevention tools and investigation services.140 However, these services may not be used by policyholders, for various reasons, such as pricing or lack of trust, and may therefore diminish the function of cyber insurance policies.141

See generally ‘Cyber Risk for Insurers’ (n 122). For insurtech’s challenges to insurers and insurance law see Paola Manes, ‘Legal Challenges in the Realm of InsurTech’ 31 European Business Law Review (2020) 129–68. 138 For the possible catastrophic consequences resulting from cyber attacks on insurance companies please see ‘Cyber risks: what is the impact on the insurance industry?’ www​.eiopa​.europa​.eu/​cyber​-risks​ -what​-impact​-insurance​-industry​-2021​-10​-15​_en accessed 17 March 2023. 139 Talesh and Cunningham (n 94) 970. 140 Ibid 1003–4. 141 Ibid 1015. 136 137

326  Research handbook on marine insurance law 4.5

Ambiguous Legal Instruments

As argued in this chapter, cyber insurers’ regulatory power largely originates from cyber insurance policies and related contractual terms. Therefore, statutory instruments governing insurance disputes, such as the Marine Insurance Act 1906 and the Insurance Act 2015, act as guarantees of the effectiveness and efficiency of such regulatory power as they are the origin of the ‘enforcement power’ of this private regulatory mechanism. However, one of the factors influencing the regulatory role of insurers is the ambiguity in the current insurance legal framework concerning how the assured should address cyber risks and fulfil their obligations under cyber insurance policies. This section will concentrate on the two most essential legal principles related to the insured’s obligations, namely, the duty of fair presentation and warranties. These principles are selected not only due to their fundamental nature but also because they have been fundamentally changed with the introduction of the Insurance Act 2015. Although the Insurance Act 2015 is not mandatory for commercial insurance policies and can be contracted out through mutual agreement between parties,142 it represents the most substantial development in English insurance law since 1906. Similar to other insurance policies, if English law is chosen as the governing law of the policy, cyber insurance policyholders must comply with their obligation to accurately represent the risk according to the Insurance Act 2015. However, this requirement may pose challenges for policyholders. For example, section 3(4)(a) mandates that the insured disclose every material circumstance they know of or should know of.143 Section 4 of the Act further stipulates that the insured must disclose information they are aware of or should be aware of.144 Additionally, section 6(1) extends the insured’s actual knowledge to include blind-eye knowledge.145 For organisations, section 4(3) specifically defines relevant knowledge as information known by senior management and the individual responsible for insurance.146 In practice, these requirements can be difficult for policyholders to fulfil, as crucial information related to cyber insurance may be complex and unknown to senior management or the insurance manager unless previously identified by the IT department.147 In addition, cyber risks are rapidly evolving, and in some cases cyber attacks can be undetectable and unknown to the victims.148 This creates another challenge for policyholders, as the reasonable search required by section 4(6) may not reveal the presence of advanced viruses or past unauthorised data access attempts.149 However, such attempts undoubtedly constitute material circumstances that should be disclosed by the insured, as they are relevant to the risk.

For contracting out sections of the Insurance Act 2015 please refer to sections 16 and 17 of the Act. Insurance Act 2015, s 3(4)(a). 144 Ibid s 4. 145 Ibid s 6(1). 146 Ibid s 4(3). 147 European Economic and Social Committee, ‘Cybersecurity: ensuring awareness and resilience of the private sector across Europe in face of mounting cyber risks’ www​.eesc​.europa​.eu/​sites/​default/​files/​ files/​qe​-01​-18​-515​-en​-n​.pdf accessed 17 March 2023, at p.72. 148 Ibid. It is revealed in this report that many cyber attacks may not be detected at all. 149 Insurance Act 2015 s 4(6). 142 143

The role of insurance in regulating cyber risks in the shipping industry  327 Therefore, even with rules that favour policyholders in place (compared to the previous utmost good faith regime under the Marine Insurance Act 1906),150 policyholders still face the risk of losing coverage since the broad wordings in the abovementioned sections provide little guidance regarding how policyholders can fulfil their duty of fair presentation in cyber insurance scenarios. This situation, in turn, may diminish insurers’ regulatory power, because such a murky position would not encourage stakeholders to take out cyber insurance policies. In addition to pre-contractual obligations such as the duty of fair presentation, as has been introduced, insurers’ post-contractual regulation can also be achieved through policy terms. Among various categories of policy terms, insurance warranties are of great importance in cyber insurance scenarios; however, the new sections introduced by the Insurance Act 2015 can potentially be another Achilles’ heel for the application of the Act in cyber insurance scenarios. Section 10 of the Act effectively transforms warranties into suspensory conditions and abolishes any provisions that fully exempt the insurer from liability in case of a warranty breach.151 Under the new regime, insurers remain responsible for any losses suffered by the insured after the warranty breach, as long as the breach is remedied. Additionally, if a loss happens before the breach and remains unaffected by it, it would generally be recoverable under the new law section 10(4).152 Consequently, determining the timing of the breach and the timing of the remedy is crucial. In the context of cyber insurance, if a warranty requires insureds to comply with a particular industrial standard, then identifying the exact timing of a failure of compliance can be challenging. Due to the complexity of modern systems, the evolving nature of cyber threats and the limitations of current security tools, the detection of such breach of warranty can be technical in nature and is normally found after a cyber breach has happened.153 Thus, since breach of such warranty is normally not detected, and therefore cannot be remedied, before loss, policyholders might encounter obstacles in recovering these losses. In addition to section 10, another new section which is allegedly in favour of the assured is section 11. According to the section wording, if a contractual term (including but not limited to warranty) is not a term that defines the risk as a whole and if the assured can demonstrate that the breach of such term could not have increased the risk of the actual loss, then the assured would still be entitled to recover under the policy.154 However, the current pitfall is that the application of section 11 of the Insurance Act 2015 to this situation has yet to be tested in English courts. In addition, in a cyber risk scenario, since the risk itself is inevitable, what should be the burden of proof upon the insured for it to show that the breach of these terms will not increase risk? All of these questions are waiting to be clarified by English courts and by industry action. In conclusion, the current insurance legal framework, particularly the Insurance Act 2015, presents several challenges in regulating and managing cyber insurance policies. In light of Rob Merkin and Özlem Gürses, ‘The Insurance Act 2015: Rebalancing the Interests of Insurer and Assured’ (2015) 78 The Modern Law Review 1004–27. 151 Ibid at 1018. 152 Insurance Act 2015 s 10(4). 153 For the general information of this please see Adel Alshamrani, Sowmya Myneni and Ankur Chowdhary, ‘A Survey on Advanced Persistent Threats: Techniques, Solutions, Challenges, and Research Opportunities’ (2019) 21 IEEE Communications Surveys and Tutorials 1851–77. 154 Insurance Act 2015 s 11. For the analysis of this section please see Merkin and Gürses (n 150) 1019–22. 150

328  Research handbook on marine insurance law these issues, it is crucial for the insurance industry and regulatory bodies to address the ambiguities and complexities inherent in the current legal framework governing cyber insurance. This can be achieved through a combination of clearer legislative guidance, judicial interpretation and industry-led initiatives. Such efforts will enhance the effectiveness and efficiency of cyber insurance as a private regulatory mechanism and encourage greater adoption of cyber insurance policies, ultimately promoting a more secure and resilient digital environment.

5.

THE WAY FORWARD

Given the limits of cyber insurance and the obscurity of cyber risks, the most significant obstacle to the adoption and development of cyber insurance is that it is hard to reach a consensus on whether cyber insurance coverage is needed in the maritime industry. The market has yet to reach a conclusion on which strategy should prevail and this uncertainty will undermine the cyber insurance regulatory function because the maritime industry still has very little experience with this type of insurance. The numerous exclusions and extra obligations may cause the client to doubt that a claim for cyber damage will be paid. The expense of requests by the insurer to have outdated IT infrastructure replaced is another concern for small companies and the replacement of IT infrastructure will not guarantee the prevention of cyber attacks or indemnification by the insurer. Therefore, the client may lack the incentive to obtain cyber insurance. If no consensus on the adoption of cyber insurance can be reached, this will weaken the regulatory function and credibility of cyber insurance. To better exploit the regulatory function of cyber insurance in the maritime industry, some fundamental challenges should be addressed. The first is how to obtain trustworthy data from the client so that cyber insurers can better assess the scale of cyber risk. The second is the fragmentation of cyber insurance policies and the third is how to cultivate cyber awareness in the maritime industry and enhance its cyber insurance experience. Regarding the assessment of risk due to the lack of expertise and reliable data, it is hard for the insurer to comprehensively evaluate the client’s cyber risk vulnerabilities155 because of the lack of previous reliable data regarding the losses caused by this new phenomenon and applicants will withhold information for reputational or premium concerns.156 Indirect losses such as those due to disruption of business or the negative effects on reputation are also hard to quantify.157 To obtain better data, a more comprehensive assessment regime should be developed. For now, most assessments of the client’s vulnerabilities to cyber risk have been done using questionnaires. However, data collected this way commonly lacks accuracy and comprehensiveness.158 To overcome this, a collation of insurers, state authorities and the client should be established so that more reliable data can be collected and more targeted policies devised. The information will allow insurers to understand the driving factors of cyber attacks and employ experts or collaborate with organisations to identify the technical weaknesses in the clients’ systems and optimise the underwriting process by asking more specific questions. The insurer 157 158 155 156

Talesh and Cunningham (n 94) 994. IMF Working Paper (n 26) 11–12. Ibid 9–10. IAIS report (n 9) 30.

The role of insurance in regulating cyber risks in the shipping industry  329 should also establish a reliable database to comprehensively understand the history of the client’s cyber security breaches and create compliance frameworks to measure this kind of risk. The most effective way to deal with the fragmentation of the cyber insurance policy issue is by standardisation of policies. The cyber insurance industry is relatively new and has been criticised for lacking standardisation and a standard contract.159 However, it is also a growing industry and the market is getting bigger since people’s awareness of cyber security issues has been raised.160 Various categories of cyber insurance policies with different terms have been devised which results in clients not receiving the same level of protection and naturally the obligations on them are also different. The advantages of cyber insurance policies are prominent. Insurance companies which provide cyber insurance coverage will normally have experts who are familiar with the technology and risk and can calculate the potential exposure and losses. Therefore, unlike positive law instruments or regulations implemented by public authorities, the requirements in cyber insurance contracts could be altered promptly in response to developments in cyber attacks and related technologies. Compared to positive law instruments, the flexibility of cyber insurance policy is remarkable. Changing insurance policy wording to align with the development of market reality is not a new thing in the insurance industry. For instance, in maritime insurance, insurers are always issuing new requirements or updating policies to tackle emerging risks.161 Legislators can also have a better understanding of the status quo from the industry’s practice. Standardised cyber insurance policies can ensure the implementation of standards consistently and coherently. There are many cyber risk-related standards, codes of practice and guidance documents in place, but they need to be tailored for application to the maritime industry.162 In the shipping industry, there will be no single set of standards that fits all, and different standards will be specified for particular products. Policies should also require the implementation of a cyber security plan and the appointment of a cyber security officer, both of which are recommended by the UK Department of Transport.163 Another way is to set minimum cyber security standards for SMEs that insurers can refer to and some scholars suggest that such minimum standards should be defined in legislation.164 Clearer wording of insurance policies could also promote their standardisation. Large and influential insurers in this field can recommend standardised wording and promote the unification of cyber insurance. A positive movement in this sphere is that the silent cyber issue has been noticed and regulated by the industry and in July 2019, Lloyd’s mandated that all policies provide clarity regarding cyber coverage by either excluding or providing it. It is important to provide clarity on coverage as this will stimulate the development of the cyber insurance

‘Cyber Risk for Insurers – Challenges and Opportunities’ (n 122) 17. Lucy Fraser, ‘Cyber insurance – growing the market to meet the global threat’, www​.abi​.org​.uk/​ news/​blog​-articles/​2022/​02/​cyber​-insurance​-growing​-the​-market​-to​-meet​-the​-global​-threat/​ accessed 24 December 2022. 161 Such as the Market Bulletin, Ref: Y5381 published by Lloyd’s. 162 Department of Transport (n 53) at 4.2. 163 Ibid. 164 European Union Agency for Network and Information Security (ENISA), ‘Information security and privacy standards for SMEs: recommendations to improve the adoption of information security and privacy standards in small and medium enterprises’ 11–12. 159 160

330  Research handbook on marine insurance law market. For now, cyber insurance is poorly understood not only by the client165 but also by the broker.166 Therefore, the standardisation of cyber insurance policies can start with the unification of terminologies and coverage risks and conditions and exclusions. Standardised cyber insurance policies should include issues such as the boundaries of cyber attacks, as this is the most important issue that must be defined. If all minor cyber breaches are included then perhaps all participants in the shipping industry have experienced cyber attacks.167 Different insurers may also have different opinions regarding how cyber risks should be covered; namely, whether cyber insurance should be treated as an independent risk or as a category of marine risk or, since some cyber incidents are initiated by states, as war and political risk. A standardised cyber insurance policy could also provide a clear position on coverage and a widely adopted standard form contract could contain standard requirements fundamental to cyber risks. Some consistency has already been achieved. In the US and Germany, a standard cyber insurance policy form has been developed by insurance associations.168 The insurance industry should continue with the standardisation of the language of key concepts and encourage such standardised language to be adopted by the shipping industry gradually. The world will benefit from a widely adopted and standardised cyber insurance contract or policy because a unified insurance contract would allow the current existing multinational and multi-industrial cyber security regulations to be simplified and the core requirements in these regulations can be added to the standard insurance contracts. Thus, the inconsistency of terminology in different national and industrial regulatory rules would be overcome; although interpretation would still lie with the national courts scrutinising future litigation, these policies would provide a unified code for the industry to follow. A universally accepted cyber insurance policy can be used as an international instrument and response to the transnational nature of cyber risks. However, on the path towards standardisation, there are some obstacles to consider and overcome. The first is how to assess the value of cyber risks and so set the premium. This is again complicated by the lack of historical data. The second is how to construct and interpret cyber insurance policies to avoid the silent cyber risk issue which originates from the poor construction of the add-on clauses on traditional insurance policies,169 although to a certain extent this has been resolved by Lloyd’s.

165 In a survey of North American (and some international) underwriters and brokers published in 2018, 56 per cent of respondents cited a lack of understanding of the coverage offered as a top obstacle to cyber-insurance take-up. For the full report please visit www​.partnerre​.com/​wp​-content/​uploads/​2018/​ 10/​2018​-Survey​-of​-Cyber​-Insurance​-Market​-Trends​.pdf accessed 9 March 2023. 166 In the UK, 31 per cent of brokers surveyed in 2018 indicated that they had a poor or very poor understanding of cyber insurance and cyber risks. For the full report please see www​.munichre​.com/​content/​ dam/​munichre/​contentlounge/​website​-pieces/​documents/​DAS​_Market​_Barometer​_Cyber​_May2018​.pdf/​ _jcr​_content/​renditions/​original​./​DAS​_Market​_Barometer​_Cyber​_May2018​.pdf accessed 9 March 2023. 167 Lloyd’s List, ‘Cyber security: a special report’ https://​lloydslist​.​maritimein​telligence​.informa​ .com/​special​-reports/​2022/​Cyber​-security/​Report​?vid​=​Maritime​&​processId​=​caa7b01e​-7af9​-4c65​ -a85e​-e0c461414c4a, at p10 accessed 24 December 2022. 168 Ibid 21–2. 169 Jan Martin Lemnitzer, ‘Why Cybersecurity Insurance Should Be Regulated and Compulsory’ (2021) 6(2) Journal of Cyber Policy 118–36.

The role of insurance in regulating cyber risks in the shipping industry  331 Finally, although cyber insurance is vital to cyber resilience, most practitioners in the maritime industry are still unaware that cyber risks can be insured at all.170 Governments and private institutions such as Lloyd’s and the Association of British Insurers should collaborate on issues such as increasing awareness of cyber risk and collecting data from the market.171 The insurance industry should make its clients aware of the existence of cyber risks and cyber insurance policies and regulation instruments. Shipping companies can be the victims of cyber attacks, but the ships can also be vulnerable. Information sharing is one of the best ways to understand the scale of the threat and raise the client’s awareness of cyber risk and tailor the response appropriately. There are already some alliance mechanisms in place including the CSO Alliance which encourages sharing of information and intelligence among its members.172 Clients should also be aware that, although existing marine insurance law and standard contracts may provide some guidance on how to deal with the threats posed by cyber risks, they may be of limited use given the changing nature of those risks and the advancement of new technologies. There are many exclusion clauses and excluded risks in current policies, and clients should be made aware of them and have a clear understanding of the coverage provided. All these proposed measures are designed to build mutual trust between the clients and the insurers and the clients’ recognition of cyber insurance’s regulatory role and regulatory authority. Like the evolution of maritime law, the recognition of cyber insurance’s regulatory function is the ultimate source of the regulatory authority of cyber insurance.

6. CONCLUSION Cyber attacks are inevitable and have become a transnational security issue. It is not a question of whether, but of when. The purpose of the construction of a resilient cyber risk regulatory system is to transfer or mitigate the risk from cyber incidents rather than to prevent them from happening in the first place. Instruments issued by public institutions in the maritime and insurance spheres play a very limited role in transferring and mitigating cyber-related losses due to their potential drawbacks. They are normally principle-based, lack detailed guidance, are incomprehensive and quickly become outdated. Although these instruments may provide some basic guidance regarding cyber attack prevention, they are far from adequate. Additionally, legal instruments such as the Insurance Act 2015 also fail to provide clear and exercisable rules regarding what insureds can do to address cyber risks, even though the law can be used to enhance the insurer’s regulatory function. This chapter has explored the potential regulatory functions of cyber insurance policies and the insurance industry. It has been argued that, by the combination of underwriting practice, contractual terms and consequences of breach by the client, cyber insurance providers can

The Guidance on Cyber Security Onboard Ships (n 8) 34–47. Ibid. 172 For more information, please see https://​csoalliance​.com/​about​-us accessed 24 December 2022. Members are: Norwegian Hull Club, Norwegian Shipowners Mutual War Risks Insurance Association and North of England P&I Club. 170 171

332  Research handbook on marine insurance law become de facto regulators in this field,173 which can significantly encourage greater compliance, decrease moral hazard and deter corporate misconduct. Thus, through reliance upon general contract law and unique insurance law principles, cyber insurance providers can set and enforce cyber security standards on clients, which arguably constitutes a unique instance of private regulation. Therefore, in the absence of any targeted instruments regarding cyber risk, the emergence of the insurers’ regulatory role stands for private advance and public retreat. However, this private governance feature also has drawbacks, such as the wide adoption of exclusion clauses, and cyber insurers may still choose to provide insurance coverage regardless of the client’s condition. The construction of the future regulatory framework also relies on collaboration between authorities and private entities. Without public oversight, it is the insurers who get to control who can or cannot obtain insurance.174 To provide a cyber resilience system and a successful governance regime, interaction between different participants is crucial, and in marine insurance a successful cyber risk regulatory framework needs cooperation between the client, insurer and public authorities. It is also clear from the analysis in this chapter that to become a more effective regulatory tool, the cyber insurance market needs more extensive regulation so that the existing obstacles can be overcome.

Talesh and Cunningham (n 94) 1003. Talesh (n 95) 466.

173 174

16. Legal and regulatory issues for sustainable marine insurance: the Poseidon Principles for Marine Insurance Livashnee Naidoo

1. INTRODUCTION Climate change is one of the most pressing issues of our time and its effects can be seen through the increase in natural disasters across the globe.1 At its core, climate change is a cross-cutting legal and regulatory issue. It is an issue that does not – and cannot – sit neatly within a ‘silo’; rather, it cuts across finance, insurance, shipping, corporate governance, contract law, tort law, to name but a few. Climate change is a global concern that impacts all facets of society and business and therefore requires a coordinated response. The Poseidon Principles for Marine Insurance (‘the ‘PPMI’) were launched in December 2021 and are the first sector-specific initiative by the marine insurance industry to address climate change. The PPMI supplement other initiatives on sustainability and climate change in shipping, insurance and financial markets, and provide an important starting point for any discussion about environmental, social and governance risks (‘ESG risks’) in marine insurance.2 ESG risks engage the three core areas of marine insurance: hull and machinery (‘H&M’), cargo, and protection and indemnity (‘P&I’) insurance.3 Climate-related risks relate to physical risks, transition risks and liability risk – and in marine insurance, climate change as a thematic issue primarily engages the risk criteria of greenhouse gas emissions (‘GHG’), transition risks and physical risks.4 In H&M and P&I markets, climate change is a potential elevated risk due EIOPA, ‘Climate Change: A Challenge for Our Time’ www​.eiopa​.europa​.eu/​publications/​climate​ -change​-challenge​-our​-time​_en accessed 20 July 2023. 2 Sustainability and climate change are often used interchangeably but sustainability is a broader umbrella term, and climate change is an environmental factor of sustainability. ESG risks include, inter alia, climate change, environmental degradation, human rights and unsustainable practices and these criteria are used to assess a company’s environmental, social and governance impact. There are 17 Sustainable Development Goals (‘SDGs’), which provide a global blueprint to improve health, prosperity, education and economic growth. See United Nations, ‘Sustainable Development’ https://​sdgs​.un​.org/​ goals​# accessed 2 April 2023. 3 H&M insurance provides cover for the ‘hull of the vessel’ along with all the articles and pieces of furniture on the ship. This type of insurance is meant to guard against the risks shipowners face arising from the loss of their vessel. This cover is provided by both H&M insurance and by P&I Clubs. P&I Clubs are mutual associations for shipowners that cover insurance not covered by H&M cover, as well as liabilities to third parties not covered by other policies. 4 ‘Physical risk’ would be an example of H&M cover which covers physical damage to the insured vessel itself. ‘Transition risk’ in the context of marine insurance would encompass the transition to new types of technology and fuel, and new insurance products to cover these new types of risk. ‘Liability risk’ relates to liability cover and relates to actions brought against insurers for compensation arising from physical or transition risks from climate change. See also PSI-IUMI Global Webinar, ‘UNEP Principles 1

333

334  Research handbook on marine insurance law to those markets’ nature.5 The PPMI are concerned with climate alignment of the signatories’ H&M portfolios which is the only environmental factor covered by these principles.6 Climate alignment refers to the degree to which a vessel, policy or portfolio’s carbon intensity is aligned with two decarbonisation trajectories. The first is the International Maritime Organisation’s (‘IMO’) decarbonisation trajectory which is concerned with a 50 per cent reduction in GHG emissions by 2050 (‘the 50 Per Cent Reduction Trajectory’) compared to the 2008 target, and the second is the Paris Agreement on Climate Change, which has a goal of net-zero by 2050 (‘the 100 Per Cent Reduction Trajectory’)7 – collectively referred to as ‘the decarbonisation trajectories’.8 The decarbonisation trajectories are regulatory efforts to address climate change and complement other initiatives on emissions trading and decarbonisation.9 To set the scope of the chapter, it is important to understand the taxonomies of climate change. This chapter is concerned with climate change alignment and not with other taxonomies relating to physical and liability risk. The IMO’s decarbonisation agenda has many moving parts, including the potential of alternative fuels and more efficient ship design, but insurance remains part of the IMO’s vision of a Sustainable Maritime Transport System.10 The PPMI are a principles-based, private regulatory tool based on the Poseidon Principles for Financial Institutions (‘the PPFI’) launched in June 2019, the latter of which are aimed at ship finance agreements. Both these initiatives engage a quantitative methodology to assess and

for Sustainable Insurance’s Guide on Environmental, Social and Governance (ESG) risks for the Insurance Industry: An Overview with a Special Focus on Marine Insurance’ (6 October 2020) https://​ iumi​.com/​uploads/​Webinar/​PSI​-IUMI​_webinar​_​-​_Bacani​_Oct​_2020​_final​.pdf accessed 20 March 2023, 12 (‘PSI-IUMI Global Webinar’). 5 ‘PSI-IUMI Global Webinar’ (n 4), 8. This is because P&I and H&M provide cover for a broader array of environmental risks. In relation to ESG risks, ‘environmental’ includes but is not limited to greenhouse gas emissions, waste and illegal fishing; ‘social’ risks encapsulate labour, diversity, safety management, weapons and human trafficking; and governance encapsulates rights and responsibilities between different stakeholders in the governance of the companies. 6 The Poseidon Principles for Marine Insurance, ‘About’ www​.poseidonprinciples​.org/​insurance/​ about/​accessed 6 February 2022 (‘Poseidon Principles’). 7 See the International Maritime Organisation, ‘Addressing Climate Change: A Decade of Action to Cut Greenhouse Gas (GHG) Emissions from Shipping’ https://​www​.imo​.org/​en/​MediaCentre/​PressBriefings/​ pages/​DecadeOfGHGAction​.aspx​#:​~:​text​=​In​%20June​%202021​%2C​%20IMO​%20adopted​,the​%20energy​ %20efficiency​%20of​%20ships accessed 8 November 2023. The Paris Agreement is a legally binding international treaty on climate change, adopted on 12 December 2015 at the UN Climate Change Conference (COP21) in Paris. It entered into force on 4 November 2016. 8 Poseidon Principles for Marine Insurance: Technical Guidance (version 11, 2023), www​ .poseidonprinciples​.org/​insurance/​wp​-content/​uploads/​2021/​12/​Poseidon​-Principles​-for​-Marine​-Insurance​ -Technical​-Guidance​.pdf accessed 30 July 2023 at 18, 24 (‘PPMI Technical Guidance’): ‘A decarbonisation trajectory is a representation of how many grams of CO2 a single ship can emit to move one tonne of goods one nautical mile (gCO2/tnm) over a time horizon.’ The Technical Guidance is a supporting document which states the requirements and expectations for each Principle. The guidance is framed in the context of the existing IMO environmental regulations and climate agreements, as well as the Paris Agreement and its commitments. 9 For example, the UK’s Climate Change Act 2008 which is the UK’s approach to combat and respond to climate change. The target will require the UK to bring all greenhouse gas emissions to net zero by 2050, compared with the previous target of at least 80 per cent reduction from 1990 levels. 10 International Maritime Organisation, ‘A Concept of Sustainable Maritime Transportation System’ https://sustainabledevelopment.un.org/content/documents/1163CONCEPT%20OF%20%20​ SUSTAINABLE​%20MARITIME​%20TRANSPORT​%20SYSTEM​.pdf accessed 11 December 2022.

The Poseidon Principles for Marine Insurance  335 report the climate alignment of H&M insurers and banks’ shipping portfolios, respectively. In doing so, the PPMI and PPFI are intended to engage environmentally responsible behaviour and to embed ESG considerations into decision-making. However, there are differences between the PPFI and PPMI, as, although the banking and insurance sectors are closely aligned financial sectors, the contractual nature of the relationship between signatories (banks versus marine insurers) and their shipowner customers (borrowers versus marine insureds) differs. This chapter examines the status of the PPMI in marine insurance law and practice. The discussion is organised by developing legal and regulatory themes centred on the PPMI and, more broadly, on climate change initiatives within the insurance sector. Section 2 provides an overview of the PPMI. Section 3 discusses the nature of marine insurance contracts within the broader conception of insurance. This is necessary to understand the objectives of the PPMI and how they relate to the specificities – and the vision – of the marine insurance market. Section 4 situates this discussion within the broader regulatory framework of sustainability in insurance and identifies the core legal and regulatory concerns relating to the PPMI. Section 5 examines the contractual architecture of the PPMI. Section 6 analyses the regulatory dimension of the PPMI and offers conclusions as to whether the PPMI (and the type of regulation that they represent) offer an effective blueprint for the marine insurance industry to mitigate climate change. Climate change is a global issue but the nature of insurance contract law is not homogenous between jurisdictions; therefore, while the context for this chapter is English law, other jurisdictions will be referred to as necessary. The PPMI are an important initiative as the signatories represent a significant volume of the world’s ocean-going fleet and is a significant step by the marine insurance sector towards climate alignment with the IMO’s decarbonisation trajectory. This chapter offers novel insights to understand the status, operation and effect of the PPMI in marine insurance markets. In doing so, it clarifies the value of the PPMI as a legal and/or regulatory tool to meet ‘green’ trajectories and contributes to broader scholarship on the role of insurers and the insurance industry in supporting the transition to a net-zero economy.

2.

THE POSEIDON PRINCIPLES FOR MARINE INSURANCE (‘THE PPMI’)

The PPMI are a set of principles created by a group of leading H&M insurers and affiliate members to assess and align their H&M portfolios with the decarbonisation trajectories.11 The Paris Agreement is a significant starting point in that case as it creates a global, binding climate agreement to address climate change by reducing global warming to well below 2°C and pursuing efforts to limit it to 1.5°C compared to pre-industrial levels. The shipping sector was not included in the Paris Agreement and therefore the IMO adopted its own strategy to reduce GHG emissions from shipping activities. The objective of the PPMI is for marine insurers, specifically H&M insurers, to play a role in the transition to a net-zero economy by supporting their shipowner clients to improve decision-making on more environmentally conscious shipping fleets, thereby contributing to the Net-Zero Insurance Alliance (‘NZIA’).12 There are ten signatories and nine affiliate members. ‘The NZIA was launched in 2021 and is an initiative driven by a group of leading insurance companies to commit to decarbonising their insurance and reinsurance underwriting portfolios to contribute 11 12

336  Research handbook on marine insurance law This is achieved through disclosing, assessing and benchmarking the carbon efficiency and climate alignment of signatories’ respective insurance portfolios by relying on a quantitative methodology. Decarbonisation of shipping has driven three interrelated initiatives by the Global Maritime Forum. The PPMI are the most recent of these initiatives and build on the PPFI launched in June 2019 and the Sea Cargo Charter launched in October 2020. The PPFI are concerned with green financing and the Sea Charter is concerned with chartering activities as part of the transition to the IMO’s Sustainable Maritime Transport Agenda.13 These initiatives create a global cross-cutting framework for maritime decarbonisation by aligning responsible environmental behaviours in ship financing, marine insurance, and chartering activities through disclosing and benchmarking their respective climate alignment portfolios against the decarbonisation trajectories. The significance of these initiatives is the weight of signatories, where, in relation to the PPMI, H&M constitutes the second largest marine insurance coverage after cargo insurance, representing a significant portion of the world’s insured fleet. Likewise, the PPFI signatories represent about USD200 billion in shipping finance.14 The PPMI are an initiative driven by private actors, including shipowners, marine insurers, classification societies and the Global Maritime Forum.15 The PPMI are voluntary principles that are intended to evolve in line with developments on the decarbonisation trajectories and on ESG risks. It applies to H&M insurance coverage where the signatory is the lead insurer or where the signatory is a follower, but the lead is also a fellow signatory and applies to all vessels which fall within the purview of the IMO whose carbon intensity can be measured with the IMO Data Collection System (‘IMO DCS’) data.16 Importantly, the objectives of the PPMI are achieved through four principles, which follow the PPFI and to which signatories commit:17 (i) Assessment of climate alignment: Signatories will measure the carbon intensity and assess the climate alignment of their H&M portfolios on an annual basis. (ii) Accountability: To ensure the fairness and accuracy of data that is relied upon for the assessment, signatories will rely exclusively on the data types, data sources and service providers identified in the Technical Guidance.18 (iii) Enforcement: Compliance with the PPMI is intended to be contractual in the signatory’s business activities to ensure that shipowners share accurate data with signatories. It is recommended that the signatories use the standardised covenant clause (‘SCC’) to be included in policy agreements to obtain consent for data use.

to the implementation of the Paris Agreement on Climate Change.’ UNEP FI, ‘Net-Zero Insurance Alliance’ www​.unepfi​.org/​net​-zero​-insurance/​ accessed 10 January 2023. 13 UN Environment Programme, ‘Green Financing’ www​.unep​.org/​regions/​asia​-and​-pacific/​regional​ -initiatives/​supporting​-resource​-efficiency/​green​-financing accessed 20 January 2023: ‘Green financing refers to increasing the financial flows (e.g. banking, insurance, investment) from the public, private and not-for-profit sectors to sustainable development priorities.’ 14 ‘Poseidon Principles for Financial Institutions’ www​.poseidonprinciples​.org/​finance/​signatories/​ accessed May 2023. 15 Others include Swiss Re Corporate Solutions, Swiss Re Institute and University College London Energy Institute and UMAS. 16 The PPMI is supported by affiliate members insurance brokers and collective groups (such as insurance associations, unions, and P&I Clubs). ‘PPMI Technical Guidance’ (n 8) 6. 17 ‘Poseidon Principles’ (n 6) ‘Principles’. 18 See n 8.

The Poseidon Principles for Marine Insurance  337 (iv) Transparency: The signatories will publicly declare that they are signatories to the PPMI and their climate alignment scores will be published annually by the Secretariat of the Poseidon Principles.

The PPMI rely on the carbon intensity metric of the annual efficiency ratio (‘AER’), which uses the parameters of deadweight tonnage at summer draught, distance travelled and fuel consumption.19 This metric was preferred over the Energy Efficiency Operational Indicator (‘EEOI’) used in the Sea Cargo Charter.20 The AER is calculated using data from the IMO DCS, making it a sensible approach for consistency and to reduce the administrative burden on shipowners.21 The IMO DCS is the IMO’s MARPOL Annex VI Data Collection System for Fuel Consumption, which is mandatory for all ships of 5000 gross tonnage and above and engaged in international trade, and shipowners are required to collect and report data per year for ships which fall into this category.22 The data is checked against the regulations of the relevant flag state or recognised organisation (‘RO’) and if it complies, a Statement of Compliance (‘SoC’) is issued by the flag state or RO and reported to the IMO Ship Fuel Oil Database. The AER has been subject to some criticism within the IMO as it is influenced by external factors which are beyond the control of shipowners, such as long ballast voyage or port congestion. The result is that AER is impacted by how the vessel is operated rather than its technical fuel efficiency.23 The assessment of climate alignment of a vessel requires that the vessel’s annual carbon intensity is compared with the decarbonisation trajectory for that ship type, size and class.24 The climate alignment of the insurer’s portfolio is a weighted average of the vessel carbon intensities in each product of portfolio, ‘calculated using the share of deadweight insured (being the product of the insurer’s percentage share insured and the deadweight tonnage of the vessel)’.25 The focus of this chapter is not on the assessment of climate alignment, including the technicalities of the AER and its calculation; rather, it is on the three principles of accountability, enforcement and transparency. The accountability and enforcement provisions are closely related concepts as they are the premise for ensuring that the other two principles (that is, assessment and disclosure of climate alignment) are conducted in a ‘practical, fair

19 ‘PPMI Technical Guidance’ (n 8) 22: ‘In shipping, carbon intensity represents the total operational emissions generated to satisfy a supply of transport work (grams of CO2 per tonne-nautical mile [gCO2 / tnm].’ 20 The EEOI draws on the CO2 emissions, the distances sailed while doing transport work and the amount of cargo carried, and so on, and procures the closest measure of the vessel’s true carbon intensity. The AER is less accurate than the EEOI in estimating a vessel’s carbon intensity because the actual cargo carried by a ship is often less than its maximum capacity and also taking into account ballast voyages. ‘PPMI Technical Guidance’ (n 8) 23, 64. 21 ‘PPMI Technical Guidance’ (n 8) 23. 22 Ibid 22. The IMO DCS requirements are separate from, and pre-date, the PPMI. The IMO DCS is an amendment to MARPOL Annex VI which entered into force in March 2018. The data includes: the amount of fuel consumption for each type of fuel in metric tonnes; distance travelled; hours under way; technical characteristics of the ship, including DWT at maximum summer draught. 23 The pros and cons of the different methodologies will not be considered but see ‘PPMI Technical Guidance’ (n 8) 23. 24 Ibid 21. See also Step 2 regarding calculating vessel carbon intensity and climate alignment and Step 3 regarding calculating portfolio climate alignment in the information flow process below. 25 Ibid 28–33.

338  Research handbook on marine insurance law and accurate manner’.26 Central to the functioning and success of the PPMI is the information flow process, detailed through four steps in the Technical Guidance, for the accountability and enforcement principles. While the PPMI are based on a quantitative methodology, the qualitative aspect remains relevant by relying on data that shipowners are required to report in order to be compliant with the IMO DCS and SoC for the calculation of AER, which can be sourced from the shipowner or the RO with consent from the shipowner (step 1).27 This speaks to the accountability principle. Step 2 is the calculation of the vessels’ carbon intensity using the IMO DCS data and the calculation of the vessels’ alignment with the decarbonisation trajectories. Step 3 is the calculation of the signatories’ portfolio climate alignment. As Steps 2 and 3 deal with assessment, these will not be looked at further. Step 4 deals with the disclosure requirements, which operate as a ‘quality control mechanism’.28 This speaks to the ‘transparency principle’, which is two-fold: it requires first that the signatory (or affiliate member) insurer make known that it is a signatory (or member), and second that the signatory should report to the Secretariat its climate alignment in accordance with the Technical Guidance. The collection and reporting of data in the information flow process occurs through two documents: (i) the Standard Covenant Clause, to be included in policy agreements (SCC); and (ii) a form of letter to be sent by signatories to shipowners to request the data. These documents speak to the enforcement principle and will be discussed below. The PPMI are the first sector-specific principles which support the IMO decarbonisation agenda, but not the first initiative of the insurance sector. The PPMI – as principles-based regulation – represent a normative yardstick for responsible environmental behaviour and are built on accountability and transparency to create a disclosure and reporting framework. Climate alignment is met through a compliance framework, not an enforcement one, raising questions about the impact of the PPMI in marine insurance contracts and about the type of regulation that this signifies in marine insurance markets. Before progressing to that stage, it is necessary to first examine the conception of marine insurance contract law and to situate the PPMI within broader sustainability initiatives in shipping and insurance.

3.

THE MARINE INSURANCE CONTRACT LAW PARADIGM29

Insurance law is not homogenous; it consists of different categorisations, and there are different visions of insurance law.30 How insurance is perceived influences the legal and regulatory response, and this is relevant to understand the value of the PPMI in H&M markets.31 Although Ibid 32. ‘PPMI Technical Guidance’ (n 8) 37–8 The sourcing of data is step 1. 28 Ibid 43. 29 This part is partially derived from Livashnee Naidoo, ‘Applied Contract Theory and the Legal Regulation of Marine Insurance Contracts: The Case of Risk Control Terms and Contracting Out under the Insurance Act, 2015’ unpublished PhD Thesis (2020). 30 For example, (i) consumer versus commercial contracts; (ii) the heterogeneity of commercial parties within commercial contracts; (iii) marine insurance encompasses both consumer and commercial insurance; and (iv) the heterogeneity of commercial parties within commercial marine insurance. 31 It can, for instance, be accepted that the regulatory threshold (whether through statute or the judiciary) for intervention in consumer contracts is generally lower than that of commercial contracts, given that there is a need for the law to protect parties in a weaker position. 26 27

The Poseidon Principles for Marine Insurance  339 a focus on the ‘vision’ of marine insurance law might appear too nuanced, it matters because marine insurance at its core can be viewed as the quintessential ‘pure’ commercial contract which attracts the norms, values and perceptions that are associated with commercial contracting (and regulation).32 This is a contentious statement, and therefore a brief overview of the operation of the marine insurance market will help to explain this point further and provide the context for a discussion on the impact of the PPMI in marine insurance markets. The London market, which is the main marine insurance market, includes three broad classes of marine insurers: Lloyd’s underwriters, insurance companies and P&I Clubs. Marine insurance is indemnity insurance, whereby the insurer undertakes to indemnify the insured against marine losses.33 As the signatories to the PPMI are H&M insurers, this type of marine insurance is the primary focus for this chapter. A significant part of H&M cover is underwritten in the London market and this market operates on a subscription basis with insurers accepting a percentage of the risk and the process occurring through brokers. Brokers play a central role in marine insurance both in terms of placing risks and managing any claims, but also ‘in bridging the knowledge gap’ between insured and insurers and in ensuring that the market remains competitive for the placing of risks.34 Marine insurance underwritten in the London market is largely based on standard wording for marine policies, incorporated into policies by attachment.35 The slip has been abolished and brokers are required to prepare a standard policy in the form of the new Market Reform Contract for risks placed in the London market; subscriptions are confirmed by signing the Market Reform Contract.36 Marine insurance contracts are created by the completion of the policy form and by the incorporation of the relevant Institute Clauses, which form the basis of marine insurance contracts internationally yet apply English law and jurisdiction, and which are amended to suit the individual needs of the parties. Despite the use of standard forms in marine insurance, it would not be entirely correct to refer to a marine insurance policy as a contract of adhesion.37 The wording of contracts in marine insurance markets is also a product of market practice and experience, which results in the evolution and development of wordings

Referring to the position in the UK. See Jonathan Morgan, Contract Law Minimalism (Cambridge University Press, 2013) for an insightful neo-classical account of contract law. 33 The Marine Insurance Act 1906, s 1: ‘A contract of marine insurance is a contract whereby the insurer undertakes to indemnify the assured, in manner and to the extent thereby agreed, against marine losses, that is to say, the losses incident to marine adventure.’ There are other forms of insurance not captured by this definition, such as contingency insurance, with the best example being life insurance. 34 Baris Soyer, Warranties in Marine Insurance (3rd edn, Routledge 2017) 187. 35 The most commonly used standard clauses are the Institute Hull Clauses: the Institute Hull Clauses (Voyage and Time) 1983 and 1995, and the International Hull Clauses 2003. 36 A ‘slip’ summarised the insurance cover and was presented by the broker to the leading underwriter, who has a reputation in the market as an expert and whose lead in subscribing to a percentage of the risk was likely to be followed. The slip was then presented by the broker to successive underwriters, who also then subscribed to a proportion of the risk which they were willing to accept – a process known as ‘scratching’. American Airlines Inc v Hope [1973] 1 Lloyd’s Rep 233. 37 Jay Feinman, ‘Contract and Claim in Insurance Law’ (2018) 25 Connecticut Insurance Law Journal 159, 162. One of the central questions of modern contract law is how to regard such contracts. See also Friedrich Kessler, ‘Contracts of Adhesion: Some Thoughts on Freedom of Contract’ (1943) 43 Columbia Law Review 629; Todd Rakoff, ‘Contracts of Adhesion: An Essay in Reconstruction’ (1983) 96(6) Harvard Law Review 1173. 32

340  Research handbook on marine insurance law through trial and error. If the terms are too onerous on one side, the standard terms may be abandoned by the market.38 To return to the point about the nature of the marine insurance market, it is generally accepted to be a specialised and sophisticated market due to the nature of the contracting parties, the type of large marine risks that are underwritten in the subscription market and the ‘broker’ nature of that market, which acts to balance the inequality of bargaining power that exists in other insurance markets.39 Marine insurance can arguably be viewed as the epitome of commercial contracting and of the neo-formalist model. The author is careful to qualify this statement with the phrase ‘arguably’ for two reasons. It is well accepted in scholarship that there is more than one conception of insurance and Kenneth Abraham’s seminal article has categorised these conceptions as contract, public utility, product and governance.40 Additionally, within (insurance) contract law there is continuing discussion about whether (insurance) contract law encompasses relational elements and is therefore more than the commercial contract between parties.41 It is the author’s contention that, when viewed through the lens of the PPMI, the perception that marine insurance is more than the commercial contract is true – but only as a starting premise. The PPMI are not part of the H&M policy but is part of the conceptually broader insurance contractual relationship. In that context, H&M insurance is something less than contract but simultaneously more (soft) regulation.42 Consequently, the author’s contention is that (marine) insurance, in relation to climate alignment, is not contract-centric, and through the lens of the PPMI, H&M insurance is adopting the conception of ‘insurance as agreement’ and not ‘insurance as contract’ due to an absence of enforceable contractual obligations.43 This may very well change in future as climate considerations solidify into legal obligations.44

38 Broker pressure has ensured that the marine insurance market adheres to the more benign earlier versions of the Hull Clauses as promulgated in 1983 and 1995. 39 Party sophistication in contract law is a largely unstudied area in the UK, whereas the US has seen more of an engagement with this concept – although that too has been very limited. See H. Glenn Beh, ‘Reassessing the Sophisticated Insured Exception’ (2003) 39 Tort Trial & Ins Prac LJ 85; Alan Schwartz and Robert Scott, ‘Contract Theory and the Limits of Contract Law’ (2003) 113 Yale LJ 541; Allen Blair, ‘A Matter of Trust: Should No-Reliance Clauses Bar Claims for Fraudulent Inducement of Contract?’ (2009) 92 Marq L Rev 423; Benjamin Hermalin and Michael Katz, ‘Judicial Modification of Contracts between Sophisticated Parties: A More Complete View of Incomplete Contracts and Their Breach’ (1993) 9 JL Econ & Con 230. 40 Kenneth Abraham, ‘Four Conceptions of Insurance’ (2013) 161 U Pa L Rev 653. 41 See James Davey, ‘The Shape of Insurance Contract Law’ in Julian Burling and Kevin Lazarus (eds) Research Handbook on International Insurance Law and Regulation (2nd edn, Edward Elgar Publishing 2023) (‘Davey’). The more specific issue of whether marine insurance is relational in nature and the related question of whether marine insurance contract law is relational is a question for another paper. 42 Borrowing the formulation from Davey ibid 27 but drawing a different conclusion in this context. 43 How the ‘concept of insurance as agreement’ maps onto Abraham’s four conceptions of insurance will be discussed in section 6. 44 See the very recent Excess Emissions Insurance launched by AXA XL which operates as an extension to the AXA XL UK & Lloyd’s Marine Hull policy. As this initiative was launched in July 2023, when this volume was already in production, it cannot be discussed in detail. AXA, ‘Excess Emissions Insurance’ https://​axaxl​.com/​en​-gb/​insurance/​products/​excess​-emissions​-insurance accessed 10 August 2023.

The Poseidon Principles for Marine Insurance  341 Furthermore, there is an emerging view of the role of insurers as ‘climate enablers’,45 and the PPMI is an example of marine insurers’ role extending beyond the traditional contract of indemnification to support climate alignment. Climate change (at least insofar as it concerns climate alignment in this chapter) is shifting the vision of marine insurance (or more specifically H&M insurance), which is likely to have normative implications. To substantiate this view, the next section will adopt a broader view of sustainable initiatives within the insurance and shipping sectors, and these will serve to highlight the normative issues underpinning the PPMI.

4.

SUSTAINABLE MARINE INSURANCE: LEGAL AND REGULATORY ISSUES?

At this juncture, it is important to ask: why have H&M insurers committed themselves to a set of voluntary principles for ‘green’ marine insurance? The PPMI recognise the role of marine insurers in the shipping industry ‘to promote responsible environmental stewardship throughout the maritime value chain’.46 The Principles of Sustainable Insurance (‘PSI’), developed in June 2020 by United Nations Environment Programme Finance Initiative (‘UNEP FI’), were the first global framework to address ESG risks in the context of insurance business.47 The PPMI build on and supports other initiatives, including the Carbon Disclosure Project,48 the Task Force on Climate-related Financial Disclosure (TCFD),49 the Science Based Targets Initiative,50 the Net-Zero Insurance Alliance (‘NZIA’) and the UNEP FI Sustainable Blue Economy Finance Initiative.51 The signatories of the PPMI are also members of the NZIA and therefore these initiatives can be viewed as complementary. The IMO’s vision for a Sustainable Maritime Transport System requires collaboration between the shipping and marine insurance sectors. The IMO has adopted the 2030 Agenda for Sustainable Development but as a global specialised body of the United Nations its broader vision needs to be granularised into digestible sectoral-specific initiatives. This does 45 Franziska Arnold-Dwyer, ‘A Legal Framework for Net Zero Aligned Insurance Products’ (2023) 29(2) Conn Ins LJ 9. 46 ‘PPMI Technical Guidance’ (n 8) 4. 47 UNEP FI, ‘Principles of Sustainable Insurance’ www​.unepfi​.org/​psi/​the​-principles/​ accessed 10 October 2022. The PSI provides a global framework to address ESG risks but are non-binding on signatories. 48 CDP Disclosure Insight Action www​.cdp​.net/​en accessed 2 April 2023. 49 Financial Stability Board, ‘Task Force on Climate-Related Financial Disclosure’ www​.fsb​-tcfd​ .org/​ accessed 12 March 2023. The Task Force on Climate-Related Financial Disclosures (TCFD) was established in December 2015 to develop a set of voluntary climate-related financial risk disclosures that enables investors and the wider public to be aware about the risks a company may face related to climate change. The TCFD recommends 11 disclosure items structured around governance, strategy, risk management and targets related to climate risks. The TCFD has gained considerable momentum and is now the international reference point for both the financial community and governments. 50 Science-Based Targets https://sciencebasedtargets.org/​ accessed 12 April 2023. The SBTi is developing guidance to support companies to go beyond their science-based targets by channelling additional climate finance towards mitigation activities outside of their value chains. 51 UNEP FI, ‘Sustainable Blue Economy’ www​.unepfi​.org/​blue​-finance/​the​-principles/​ accessed 12 November 2022 provides a global framework for banks, insurers and investors to finance a sustainable blue economy through promoting the implementation of SDG 14 (Life Below Water).

342  Research handbook on marine insurance law not equate to a ‘silo-specific’ approach; rather, the cross-cutting nature of the decarbonisation agenda is being given effect by reference to the idiosyncrasies of specific sectors, such as the PPMI for H&M markets. The importance of the maritime decarbonisation agenda to address climate change is seen through industry initiatives,52 regulatory and policy changes53 and legal and contractual developments.54 Climate change is a prominent topic under the umbrella of ‘ESG’ risks and the EU Taxonomy Regulation (which includes shipping) establishes six environmental objectives as part of its classification for environmentally sustainable economic activities.55 The taxonomy is intended to increase investment in sustainable activities across all economic sectors. In relation to insurance, the taxonomy can be of wider assistance to insurers by integrating ‘sustainability considerations in their investments by providing common definitions and metrics’.56 There is a commonality to the various insurance and broader sustainable initiatives which can be gleaned from a definition of sustainable insurance: ‘A strategic approach where all activities in the insurance value chain, including interactions with stakeholders, are done in a responsible and forward-looking way by identifying, assessing, managing and monitoring risks and opportunities associated with environmental, social and governance issues.’57 The principles of the PPMI (transparency, disclosure, reporting, accountability) echo the core values of initiatives such as NZIA, the SDGs, the PSI, the Paris Agreement and the TCFD. In the context of climate change, ‘public regulatory norms’ are increasingly finding expression in private insurance contractual relationships.58 To return to the initial question as to why H&M insurers have committed to the PPMI, the answer lies in the broader role of insurers which encompasses not only paying claims, but driving sector-specific solutions to foster sustainable economic and social development. The integrating of climate change considerations in their investment strategies and processes is a strategic priority of insurers as part of the sustainability agenda.59 Marine insurance is following suit by developing sector-specific solutions to address climate change in both insurers’ portfolios and their shipowner clients’ 52 Examples include: the PPFI, PPMI, Sea Cargo Charter, the Getting to Zero Coalition, the Green Shipping Corridors. 53 Such as the Emission Trading Scheme (ETS) https://​climate​.ec​.europa​.eu/​eu​-action/​eu​-emissions​ -trading​-system​-eu​-ets​_en accessed 1 April 2023. This initiative will apply to shipping within the EU and for voyages to or from the EU from 2024. See also initiatives by the IMO towards decarbonisation including the Energy Efficient Design Index (EEDI), the Energy Efficiency Existing Ship Index (EEXI). 54 BIMCO issued a Carbon Intensity Indicator (CII) clause on 17 November 2022. This clause intends to help owners and charterers contractually navigate and collaborate on compliance with CII as an operational measure. 55 The EU Taxonomy Regulation (Regulation (EU) 2020/852 of the European Parliament, art 9. 56 Marie Scholer and Lazaro Cuesta Barbera, ‘The EU Sustainable Finance Taxonomy from the Perspective of the Insurance and Reinsurance Sector’ (EIPOA Financial Stability Report, July 2020) www​.eiopa​.europa​.eu/​system/​files/​2020​-07/​20203310​_thematic​-article​-eu​-sustainable​-finance​ -taxonomy​.pdf accessed 2 April 2023, 5. 57 UNEP FI, ‘Principles for Sustainable Insurance’ www​.unepfi​.org/​insurance/​insurance/​#:​~:​text​ =​Sustainable​%20insurance​%20is​%20a​%20strategic​,with​%20environmental​%2C​%20social​%20and​ %20governance accessed 5 April 2023. 58 This will be discussed in section 6. 59 Marayam Golnaraghi, ‘Climate Change and the Insurance Industry: Taking Action as Risk Managers and Investors’ (Geneva Association 2018) www​.genevaassociation​.org/​sites/​default/​files/​ research​-topics​-document​-type/​pdf​_public/​climate​_change​_and​_the​_insurance​_industry​_​_taking​ _action​_as​_risk​_managers​_and​_investors​.pdf accessed 12 April 2023.

The Poseidon Principles for Marine Insurance  343 operations. Insurers who are not seen as being part of the solution to reduce GHG emissions are likely to face reputational harm and this ‘could negatively reflect in the value of company and thus affect the shareholders’ return on equity’.60 Moving away from this broader framework to the PPMI, there are several legal and regulatory issues that can be identified, but the focus of this chapter is on the contractual and regulatory dimensions of the PPMI. The former enquiry relates to whether and, if so, how the PPMI are likely to change the contractual architecture of marine insurance contracts. This question has not yet been considered; nevertheless, this chapter takes on the task of examining whether a principles-based approach is sufficient to meet the decarbonisation agenda or if a contractual framework is also necessary to support and enforce the objectives. The author’s previous contention that the PPMI are not contract-centric and are therefore something less than contract is supported by commentators who argue that the ‘contract’ conception of insurance is not entirely accurate if the terms are not binding on the insured.61 The regulatory enquiry claims that the PPMI, as a private regulatory tool, have opted for regulatory techniques of disclosure and transparency, thereby adopting a focus on public regulatory norms rather than (contractual) enforcement – in line with other insurance-sector initiatives.

5.

THE CONTRACTUAL ARCHITECTURE OF THE PPMI

5.1 Incorporation Becoming a member of the PPMI requires that signatories take the Standard Declaration, a formal commitment that the signatory will follow all legally binding requirements of the PPMI, which extends to the four key principles. This, along with the related reporting, is a public commitment. Signatories to the PPMI undertake a commitment to implement the PPMI in ‘their policies, procedures, and standards’62 to further the objective of supporting their shipowner clients to better align their business portfolios with the trajectories and to promote more responsible environmental behaviour. Although the objectives are aligned with the PPFI, the PPFI have greater potential as an incentivising framework for shipowner clients to ‘green’ their portfolios where finance may be limited for vessels that are less compliant with the decarbonisation agenda. The context for marine insurance differs under the PPMI as the availability of H&M cover is not dependent on or subject to shipowners’ climate alignment. The objective of the PPMI is not indemnification but collaboration, whereby H&M insurers support their shipowner clients’ climate alignment to transition to a greener portfolio. At this early stage of the PPMI, questions arise about the practicalities of how to incorporate the PPMI into the insurance contract and the legal effect (or lack thereof) of that incorporation. The PPMI are to be incorporated into H&M policies through the standardised covenant clauses

Arnold-Dwyer (n 45) 40. Abraham (n 39) 674, referring to Jeffrey W. Stempel, ‘The Insurance Policy as Thing’ (2009) 44 Tort Trial & Ins Prac LJ 813, 818; Daniel Schwarcz, ‘A Products Liability Theory for the Judicial Regulation of Insurance Policies’ (2007) 48 Wm & Mary L Rev 1389, 1405–6; Jeffrey W. Stempel, ‘The Insurance Policy as Social Instrument and Social Institution’ (2010) 51 Wm & Mary L Rev 1489, 1495. 62 ‘PPMI Technical Guidance’ (n 8) 5. 60 61

344  Research handbook on marine insurance law (‘SCC’) and the other relevant document is a letter from signatories to shipowners to request the data.63 The SCC in the PPMI reads: Any/all Assured(s) hereon shall supply or procure the supply to the Claims Leader (whether directly, to such physical and/or electronic address as provided to the Assured(s) for the purposes of this clause, or via the broker(s) for provision to the Claims Leader) the PPMI Data in respect of any/all Ship(s) insured by this policy on or by [10 October] each year. The Claims Leader is authorised to share the PPMI Data with other PPMI Signatories subscribing to this policy, and all PPMI Signatories in receipt of the PPMI Data are authorised to share it with Calculation Provider(s). [The Assured(s) hereon shall also, upon written request by any PPMI Signatory for the PPMI Data, provide it or procure its provision to that Signatory within 10 days of the request. Such request may be made directly to the Assured(s) or the broker].64

The SCC is recommended and not mandated, which lessens its legal value if it is not widely adopted in the industry. The PPMI Technical Guidance states that ‘the Signatory will use its best efforts to include the Definitions and Covenant wording set out in the covenant clause in the relevant documentation, amended, where necessary, to reflect the Signatory’s proposed method of data collection’.65 The information flow process, based on evidence from the IMO DCS and with the provisions of the SoC from an RO, is detailed in the Technical Guidance, as discussed above, but the SCC does not specifically mention the steps in the information flow process. Key to the success of the PPMI are standard clauses drafted by the marine insurance industry and raising awareness of climate alignment to create an industry norm for the reporting of emissions data.66 Given the nature of the subscription market, brokers will have an important role to play by using their best efforts to inform and by including definitions and standard clauses to support signatories and clients in providing the data.67 5.2

(Insurance) Contract Terms and Legal Obligations

The PPMI create not a contractual framework for climate alignment but rather a disclosure and reporting framework through ‘soft’ law obligations. The PPMI are principles-based; membership is voluntary, and a signatory’s reliance on the SCC is also voluntary. The PPMI do not create absolute legal obligations on the shipowner to disclose the climate alignment of its vessel to the H&M insurer, or a legal obligation on the latter to publicly disclose the climate alignment of its H&M portfolios. What, then, is the status of a contractual term(s) that requires but does not mandate disclosure of emissions data? Given the early stages of the PPMI, at best, this section provides insights on the contractual terms that are typically used in insurance contract law and how

‘Poseidon Principles’ (n 6) Principle 3. Poseidon Principles for Marine Insurance, ‘Standard Covenant Clause’ (June 2022) www​ .poseidonprinciples​.org/​insurance/​wp​-content/​uploads/​2022/​06/​Poseidon​-Principles​-for​-Marine​-Insurance​ -Standard​-Covenant​-Clause​_April​-2022​.pdf accessed May 2023. 65 ‘PPMI Technical Guidance’ (n 8) 46. 66 See industry perspectives: GARD, ‘The Poseidon Principles – What Will It Look Like in Practice?’ (June 2022) www​.gard​.no/​web/​articles​?documentId​=​33760933​#:​~:​text​=​First​%20of​%20all​ %2C​%20the​%20Poseidon​,for​%20their​%20fuel​%20consumption​%20data accessed 2 March 2023. 67 ‘PPMI Technical Guidance’ (n 8) 13. 63 64

The Poseidon Principles for Marine Insurance  345 these may support or hinder the implementation of the PPMI. This feeds into how the PPMI could be construed within the H&M contract, if not more loosely within the contractual relationship. As the PPFI have had a longer lifespan, a good point of comparison is provided with regard to the type of contractual terms that can be engaged by the PPFI and/or the PPMI. There is a notable caveat, though: ship finance agreements have a different contractual architecture to marine insurance contracts. In general contract law, contractual terms can be classified into conditions, condition precedents, warranties and innominate terms.68 The type of contractual term determines the obligations and the remedies on breach. Insurance contract law is aligned with general contract, but there are differences, as insurance is a type of aleatory contract, and marine insurance is a contract of indemnity. A condition goes to the root of the contract and the innocent party is entitled on breach of this term to treat the contract as repudiated and to refuse to proffer or accept future performance.69 This remedy is available in addition to a right to damages. A condition precedent is where the insurer’s obligation to pay is conditional on the insured’s contractual performance.70 The remedy is neither damages nor termination of the contract, but rather the insured is off-risk and its duty to pay the claim is removed. An innominate term is where the rights of the innocent party depend on the seriousness of the consequences of any breach.71 In general contract law, a warranty is a statement that a particular present or future condition is true, and is a relatively minor term of the contract with the remedy sounding in damages.72 Insurance warranties are a type of condition precedent and were a fundamental contractual term that functioned as a means for insurers to properly circumscribe the risk and to guard against an alteration of the risk that would render it materially different from the risk assumed by the insurer.73 Since the Insurance Act 2015 the remedy for breach of insurance warranties has been tempered, but the remedy remains distinct from general contract law. The PPMI require that shipowners disclose the climate alignment of their vessels to their H&M insurers and that the latter publicly disclose the climate alignment of their portfolios – the PPMI ‘bind’ the signatory and not its shipowner client. It is more difficult to mandate disclosure by shipowners than public disclosure by insurers, and therefore it is more significant to focus on the disclosure 'obligations' of the insured shipowner. The status of the PPMI raises questions about the nature of the legal duties, and for that discussion a comparison between ship finance agreements under the PPFI and H&M polices under the PPMI is useful. Ship finance agreements set out the terms of the loan and the obligations of parties, including repayment of the loan. Under the PPFI, signatory banks may incorporate contractual obligations of disclosure of climate alignment as part of finance agreements which could improve

This chapter will not focus on a complete exposition of insurance contract terms. Rob Merkin and Özlem Gürses, ‘Insurance Contracts after the Insurance Act 2015’ (2015) 132 Law Quarterly Review 445, 446. 70 Hugh Beale (ed), Chitty on Contracts (34th edn, Sweet & Maxwell 2021) Chap 4. 71 Hong Kong Fir Shipping Co. Ltd. v Kawasaki Kisen Kaisha Ltd. [1962] 2 Q.B. 26; Alfred McAlpine plc v BAI (Run-Off) Ltd [2000] 1 Lloyd’s Rep. 437; Friends Provident v Sirius International Insurance [2006] Lloyd’s Rep. IR 45. 72 Chitty on Contracts (n 70) Chap 44, section 7. 73 Bank of Nova Scotia v Hellenic Mutual War Risk Association (Bermuda) Ltd, The Good Luck [1992] 1 AC 233. See also HIH Casualty & General Ins Ltd v New Hampshire Ins Co [2001] 2 Lloyd’s Rep 161, [101]. 68 69

346  Research handbook on marine insurance law the availability of finance for green shipping and influence lending decisions based on climate alignment. Covenants have long been included in ship finance agreements whereby borrowers undertake to comply with certain environmental legislation or regulations, or to promise to act in a particular way (such as for the borrower to comply with certain obligations such as to register the ship in a ship register) or refrain from certain conduct (such as not to encumber its assets with liens or sell the vessel). Environmental or ‘green’ covenants operate in a similar manner but are related to environmental activities.74 Technically speaking, the reporting requirements under the PPMI relate to annual reporting requirements, and these could be captured by a continuing warranty.75 This is unlikely to happen and is ill advised for two reasons. First, by virtue of s 10 and s 11 of the Insurance Act 2015 an insurer can only rely on non-compliance with such a warranty to avoid a claim if the non-compliance with the warranty could not have increased the risk of loss which actually occurred in the circumstances in which it did (assuming that it is not a term which defines the risk as a whole, which is excluded by s 11(1)). Second, that enquiry should not even arise, as the PPMI are not about indemnification but climate alignment. Given the severity of the effect of breach in relation to insurance contracts (discussed below), it would be inappropriate for cover to be withdrawn based on insured shipowners failing to disclose their climate alignment. It will render the H&M market less competitive as these are voluntary principles and shipowners will take their business elsewhere. The obligations imposed by the PPMI do not clearly demarcate the boundaries of the insured’s obligation and how compliance is to be assessed. The PPMI use terms such as ‘best endeavours’ and ‘collaboration’ and are therefore better seen as part of the marine insurance relationship, which is conceptually broader than the insurance contract. Viewed in this way, could the reporting requirement on shipowners under the PPMI be seen as a good faith duty? In the seminal case of Carter v Boehm, Lord Mansfield described the duty of good faith: ‘Good faith forbids either party by concealing what he privately knows, to draw the other into a bargain, from his ignorance of that fact, and his believing the contrary.’76 Given the approaches and commentary on the role of utmost good faith in insurance law, it is contended that it would be inherently problematic to elevate an environmental duty or a reporting obligation under the PPMI as tantamount to the principle of good faith in insurance contracts.77 The Law Commissions have also substantially diminished the reach of the doctrine of utmost good faith in insurance law.78 See generally, Mohammed A. Bekhechi, ‘Some Observations regarding Environmental Covenants and Conditionalities in World Bank Lending Activities’ in A. von Bogdandy and R. Wolfrum (eds) Max Planck Yearbook of United Nations Law (Kluwer Law International Ltd 1999) 3, 287–314. 75 A present fact warranty regulates the past or present state of affairs whereby the insured states unequivocally that a certain state of affairs exists at the time of making the warranty (such a warranty would state, for example, that a vessel has been surveyed in the past 12 months and has complied with the recommendations of that survey); a continuing warranty regulates what an insured may or may not do during the currency of the policy so as not to increase the risk undertaken (for example, a warranty that a vessel shall not navigate in certain areas during the currency of the policy). 76 Carter v Boehm (1766) 3 Burr 1905; 97 Eng Rep 1162 at 1164. 77 See also R Hasson, ‘The Doctrine of Uberrima Fides in Insurance Law: A Critical Evaluation’ (1969) 32 Mod L Rev 615. See also Manifest Shipping Co Ltd v Uni-Polaris Ins Co Ltd, The Star Sea [2003] 1 AC 469, [45]; Banque Financiere de la Cite SA (formerly Banque Keyser Ullmann en Suisse SA) v Westgate Insurance Co (formerly Hodge General & Mercantile Co Ltd) [1990] 2 Lloyd’s Rep. 377. 78 The Insurance Act 2015, s 14. 74

The Poseidon Principles for Marine Insurance  347 5.3

Default and Enforcement

Allied to incorporation and the status of the PPMI is the issue of enforcement. Neither the PPMI or the PPFI stipulate the consequence of non-compliance if a shipowner fails to comply with its duties; therefore it is not clear what would constitute breach of the PPMI and what remedies would lie in contract law. The stated purpose of the PPMI is not enforcement but rather for signatories to guide and support their marine insurance clients to ‘adopt and align their business with responsible environmental impacts’.79 There is reticence when it comes to ESG risks due to the uncertainty inherent in understanding these risks. Shipowners are cautious about the PPMI and therefore enforcement is largely absent from the PPMI in these early stages. There is a notable difference between ‘enforcement’ under the PPMI and under the PPFI due to the contractual nature of ship finance contracts as opposed to H&M contracts. Enforcement is intended to come through the SCC but the SCC is silent on the effect of non-compliance in both the PPFI and PPMI. The SCC for ship finance contracts under the PPFI mandates compliance with Regulation 22A under Annex VI of Marpol for collection and reporting of ship fuel oil consumption data for a Ship’s Energy Efficiency Management Plan.80 The SCC in the PPMI is a shorter clause and has no such reference. A key difference between the PPFI and the PPMI turns on ‘default’ under ship finance agreements versus H&M policies. Under a ship finance agreement, a borrower’s failure to comply with the terms of the contract would amount to a default. In such agreements, it is also common to stipulate for ‘events of default’ which will trigger a remedy for the bank. Where there are express ‘terms of default’, the shipowner borrower would be bound by the express event of default, and categorising the term in question as a different type of contractual term to trigger a different remedy would not bear relevance.81 (Marine) insurance is different, as the issue of breach of insurance contract arises at the stage of an insurance claim – it is at that stage that insurers will seek to establish whether insureds have breached their contractual obligations, thereby entitling the insurer to repudiate the claim. There is no default of contractual provisions in the context of traditional contract law; rather, in (marine) insurance, ‘an insurer’s obligation to pay a claim would be contingent on prescribed contractual performance of the insured. The remedy would be that the insurer’s duties are not triggered because the insured has not met certain pre-conditions; the remedy is not damages and/or termination of the insurance contract.’82 Under the PPFI, ship finance agreements could provide for breach of the SCC Clause to be categorised as an event of default (as mentioned above), but, given the lack of clarity and certainty on the SCC and the PPFI more broadly, it is questionable what enforcement will look

‘PPMI Technical Guidance’ (n 8) 7. The IMO has implemented the Energy Efficiency Design Index (EEDI) the purpose is to monitor the amount of GHG emissions form vessels. The Ship Energy Efficiency Management Plan (SEEMP Plan) is developed and implemented by the shipowner to potentially reduce the operational cost of the ship which will eventually help in reducing the overall fuel consumption, including emissions and losses in the longer run. 81 Pia Rebelo, ‘Poseidon Principles: Legal Directions for Implementation and Enforcement’ (2020) 26(2) JIML 116. 82 James Davey, ‘Remedying the Remedies: The Shifting Shape of Insurance Contract Law’ [2013] LMCLQ 476. 79 80

348  Research handbook on marine insurance law like.83 It has been suggested that for the PPFI, contractual enforcement obligations should be gradually implemented, with clear expectations about the remedies for noncompliance.84 This is not the solution that is suggested for marine insurance; given the severity of the effect of breach in relation to insurance contracts (the loss of cover) and the fact that the PPMI is not concerned with the risk insured but with climate alignment, it would be inappropriate for cover to be withdrawn on the basis of insured shipowners failing to disclose their climate alignment. In furthering the net-zero agenda, insurers are developing various tools, including impact underwriting.85 Translating impact underwriting into contractual mechanisms would find expression where compliance with green terms would either trigger benefits or trigger an adverse effect for non-compliance. The PPMI do not (yet) translate into impact underwriting as they remain a collaborative commitment by insurers and insureds to work together to transition to net-zero through a reporting framework. An interesting example of how insurers are driving ESG considerations in marine insurance is found in the most recent 2023 version of the Nordic Marine Insurance Plan of 2013 (‘the Nordic Plan’).86 Generally, under H&M policies, claims to the insured will reflect the insured loss, as it is a policy of indemnification. But property policies, of which H&M is an ideal example, also allow for the property that has been damaged to be repaired (or replaced). The Nordic Plan has incorporated ESG considerations in the choice of tenders by repair yards where a repair yard may be disqualified if it has a poor track record in environmental and social governance.87 Under the Nordic Plan, if an insured ship is damaged and there has been no total loss then the insurer will be liable for the costs of repairs but can request that tenders be obtained from different repair yards. Although the insured can choose the repair year, this choice is limited to the most cost-efficient option for the insurer, and the insured can also reject any tender as long as there is a ‘justifiable reason’ for doing so. The new Nordic Plan has incorporated ESG considerations into the choice of repair yard and allows the insured to recover some of the increased costs if its choice implies reduced CO2 emissions compared to the most cost-efficient alternative. In the commentary to the Nordic Plan, ESG factors are listed as relevant issues for consideration, thereby providing insureds with the right to demand that the repair yard that is selected complies with ESG standards. The PPMI create a disclosure and reporting framework, not an enforcement one. The question is whether an enforcement framework is necessary to support the objectives of the PPMI in ensuring that good-quality emissions data is shared by shipowners (or consent is given for sharing that data) and that H&M insurers comply with their reporting obligations.

Rebelo (n 81) 119. Ibid 109. 85 EIOPA, ‘Impact Underwriting: Report on the Implementation of Climate-Related Adaptation Measures in Non-Life Underwriting Practices’ (6 February 2023) www​.eiopa​.europa​.eu/​system/​files/​ 2023​-02/​Impact​%20underwriting​%20​-​%20Report​%20on​%20the​%20implementation​%20of​%20climate​ -related​%20adaptation​%20measures​%20in​%20non​-life​%20underwriting​%20practices​.pdf​.pdf accessed 12 March 2023, 6. EIOPA introduced the concept of impact underwriting, which refers to the ability of insurance undertakings to contribute to the adaptation of the society to climate change by means of their underwriting practices. 86 The Nordic Marine Insurance Plan is a collaboration between the Nordic Association of Marine Insurers (Cefor), Danish Shipping, the Finnish Shipowners’ Association, the Norwegian Shipowners’ Association and the Swedish Shipowners’ Association which is updated every four years. 87 The Nordic Marine Insurance Plan 2023, version 2023, clause 12-12, sub-clause 3. 83 84

The Poseidon Principles for Marine Insurance  349 In other words, without a contractual enforcement framework, do the PPMI remain merely aspirational? This could raise concerns about the potential for greenwashing.88 The PPMI have adopted cross-cutting principles of transparency and disclosure as their normative framework, rather than contractual clarity and enforcement. The regulatory approach of the PPMI for climate alignment is given effect through public law regulatory norms rather than through contractual enforcement measures.

6.

CLIMATE ALIGNMENT AND THE (IR)RELEVANCE OF (MARINE) INSURANCE CONTRACT LAW?

There is a body of scholarship on ‘insurance as governance’ in which insurance is viewed as ‘governing’ policyholders and is therefore akin to government regulation.89 There are, of course, concerns about insurers as a surrogate for government regulation to control or incentivise the behaviour of insureds due to the risk of abuse where insurers may have self-interested motivations that conflict with the broader regulatory purpose.90 Part and parcel of insurance is a ‘control’ element to regulate policyholders’ behaviour within the limits of the policy through common contractual devices such as deductibles and warranties.91 While there may be general concerns about the limits of insurance as regulation, it is arguable that in some insurance contexts, insurers can act as effective regulators and insurance can be a tool of governance for social good.92 At this point in time, the PPMI operate through public regulatory norms of transparency rather than through contractual enforcement. The SCC is a clause that may be voluntarily adopted and, even where it is adopted, it creates soft obligations rather than binding duties. The value of transparency as a regulatory tool or an instrument of governance should, however, not be underestimated.93 Transparency is essential to the functioning of insurance markets and ‘can complement more aggressive regulatory tools […] by harness[ing] market discipline, and ensure regulatory accountability in ways that more aggressive regulatory tools often cannot’.94 The PPMI engage a quantitative assessment to further the decarbonisation agenda Financial Conduct Authority, ‘Sustainability Disclosure Requirements (SDR) and investment labels’ CP22/20 (October 2022). Green-washing refers to misleading claims around sustainability-related statements and products, and has become a growing issue for re/insurance alongside the increasing focus on ESG. 89 See Richard V. Ericson et al, Insurance as Governance (University of Toronto Press 2003); Carol A. Heimer, ‘Insuring More, Ensuring Less: The Costs and Benefits of Private Regulation through Insurance’ in T. Baker and J. Simon (eds), Embracing Risk (Chicago University Press 2002) 116–45. 90 Omri Ben-Shahar and Kyle D. Logue, ‘Outsourcing Regulation: How Insurance Reduces Moral Hazard’ [2012] Michigan Law Review 111; Kenneth S. Abraham and Daniel Schwarcz, ‘The Limits of Regulation by Insurance’ (2023) 98 Ind LJ 215. 91 Pat O’Malley, ‘Legal Networks and Domestic Security’ [1991] Politics and Society 171; Heimer (n 89); Ben-Shahar and Logue (n 90). 92 Shauhin A. Talesh, ‘Insurance Companies as Corporate Regulators: The Good, The Bad, and the Ugly’ (2017) 66 DePaul L Rev 463. 93 Financial Services Authority, ‘Transparency as a Regulatory Tool’ (Discussion Paper 08/3, May 2008) www​.fca​.org​.uk/​publication/​discussion/​fsa​-dp08​-03​.pdf accessed 6 March 2023. 94 Daniel Schwarcz, ‘Transparently Opaque: Understanding the Lack of Transparency in Insurance Consumer Protection’ (2014) 61 UCLA Law Review 394 https://​papers​.ssrn​.com/​sol3/​papers​.cfm​?abstract​ _id​=​2130908 accessed 2 February 2023. 88

350  Research handbook on marine insurance law and is intended to be as ‘unintrusive’ for their shipowner clients. Contractual enforcement, at this stage, might be too aggressive to align with the decarbonation agenda. While insurers may be viewed as ‘climate enablers’,95 how that occurs varies within the broad category of climate change. While impact underwriting and investment are tools to incentivise commitments to reduce GHG emission, the PPMI are not yet at that stage in marine insurance. A contrary viewpoint is that the PPMI as a regulatory framework and not an enforcement one does not go far enough; that focusing on the regulation paradox is short-sighted as marine insurance law should attract contractual solutions. It was previously mentioned that the PPMI render H&M policies as less than contract due to the absence of enforceable duties, thereby falling short of Abraham’s conception of ‘insurance as contract’.96 The view expounded in this chapter – that of ‘marine insurance as agreement’ – also does not align with Abraham’s conception of ‘insurance as a product’. The product conception is better suited to areas such as cyber risks, where the cyber insurance product has evolved from the typical product of risk transfer to one where insurers, through cyber liability insurance, actively provide value-added risk management services to manage and prevent cyber losses.97 In that case the risk is aligned with the value-added service. However, at this point in time, this is not the case in relation to the PPMI, where the risk (that is, to be indemnified for covered losses under the H&M policy) is distinct from the service (that is, to assist decarbonisation efforts of shipowners).98 Furthermore, with the ‘product conception’ the primary paradigm would be tort rather than contract law, implying that insurers should be liable for defective insurance products much like the position in other areas of law, such as sale of goods. This cannot be correct when one considers the objectives of the PPMI are rooted in compliance than enforcement and considering that such perceived liability would contradict the competitiveness of the market and undercut the public utility goal of the PPMI.99 What, then, is an appropriate conception of marine insurance when viewed through the PPMI? The author is of the opinion that the ‘insurance as agreement’ conception embodies aspects of both the public utility and the governance conceptions. However, the latter two conceptions are not strong enough in their own right to support the conception of marine insurance as either public utility or governance due to shortcomings with both. The public utility function is helpful in recognising the essential role of insurance in the economy and in social ordering.100 As Spencer Kimball put it, ‘the need for insurance became so great in twentieth-century society that insurance tended to become a kind of public utility’ so that ‘insurance companies

Arnold-Dwyer (n 45). Abraham (n 40). See also: C. French, ‘Understanding Insurance Policies as Noncontracts: An Alternative Approach to Drafting and Construing These Unique Financial Instruments’ (2017) 89 Temp L Rev 535. 97 Livashnee Naidoo, ‘Conceptualising the Cyber Insurance Product: How Insurance Can Enhance Risk Management Frameworks?’ in Livashnee Naidoo and James Davey, Dissent in Insurance Law (2022) Special Issue of the British Insurance Law Association Journal (Platinum Edition). 98 The insurance industry is taking steps to aligning decarbonisation with H&M policies. See n 44. 99 Abraham (n 39) 674, referring to Schwarcz, ‘A Products Liability Theory’ (n 60) 1405–6; Stempel, ‘The Insurance Policy as Thing’ (n 61) 818; Stempel, ‘The Insurance Policy as Social Instrument’ (n 61) 1495. 100 Spencer L. Kimball, Insurance and Public Policy (University of Wisconsin Press 1960) 7. The public utility concept views insurance as similar to an organisation that provides an important public service such as energy or water. It should be noted that Kimball has in mind the US insurance market. 95 96

The Poseidon Principles for Marine Insurance  351 might properly be required to supply their services to all who sought them’.101 The public utility conception therefore has greater importance in some insurance lines (such as health) than in others. The public utility concept on its own does not explain the conception under the PPMI, which remains peripheral to the core business of the marine insurer in a mixed market economy.102 However, the ‘governance conception’ can be explained when insurers use rating factors to influence human behaviour by penalising or rewarding more responsible behaviour, such as by imposing higher premiums on more risky drivers or reducing premiums for homeowners who install fire alarms on the insured property. The punish-or-incentivise function is not part of the PPMI tapestry. The most appropriate conception is that of insurance as agreement, where shipowners and insurers recognise the important public utility function of climate alignment and where the PPMI reflect an ‘organisational conception of governance’103 to promote collaboration and transparency to meet the decarbonisation trajectories.104 Internal disclosure among signatories allows for additional information to be shared at an aggregated level to build trust and improve reporting over time.105 Marine insurance in England historically reflects a responsive and collaborative nature.106 The history of the shipping and marine insurance sectors has been one driven by merchant tradition and private ordering.107 So too is that of the PPMI: a collaboration between insurers, shipowners, brokers and professional insurance bodies. How one conceptualises marine insurance under the PPMI leads to questions about how regulation and law should meet the climate challenge. With the PPMI (and climate mitigation), traditional marine insurance contract law has shifted to the periphery, so the focus is not on the marine insurance contract but rather on the marine insurance contractual relationship. The latter is broader and implies a ‘relational’ approach to the insurance relationship – and in these early days of the PPMI, marine insurance contract law has a small role to play as legally binding rules.108 This may well change in the future as insurers seek to meet decarbonisation efforts through amendments and extensions to existing H&M policies.

Ibid. This view may change if insurers include decarbonisation as an extension to the hull policy (n 44). 103 Abraham (n 40) 684. 104 François Ewald (translated by Stephen Utz), ‘The Return of Descartes’s Malicious Demon: An Outline of a Philosophy of Precaution’ in Tom Baker and Jonathan Simon (eds), The Changing Culture of Insurance and Responsibility (University of Chicago Press 2002) 273. 105 ‘The Poseidon Principles Technical Guidance’ (n 8) 45. 106 Jeffrey Thomson, ‘A History of English Marine Insurance Law: Merchants, Their Practices, the Courts and the Law’ in Andrew Hutchison and Franziska Myburgh (eds), Research Handbook on International Commercial Contracts (Edward Elgar Publishing 2020) 198. 107 Private ordering is a socio-legal concept that has been a key feature of commercial contracts both historically and to varying extents in modern times. The notion is one of norms created from ‘the bottom up’ through market participants than through the apparatus of the state. 108 ‘Relational’ is a complex term in contract law and theory. In this concept the term is used to signify a collaborative ethos that focuses on the contract relationship rather than on formal legal rules. This contention by no means implies that marine insurance contracts are relational. This is a topic for further research in relation to climate-related risks and relational contract. See also J. Feinman, ‘The Insurance Relationship as Relational Contract and the Fairly Debatable Rule for First-Party Bad Faith’ (2009) 46 San Diego L Rev 553, 556–9. 101 102

352  Research handbook on marine insurance law

7. CONCLUSION Climate change is a global concern that requires a global mandate and a coordinated response across sectors. The shipping and insurance sectors have recognised the important role which they play in mitigating climate change and the PPMI is a significant first step for marine insurance markets. The PPMI provide a framework for H&M insurers to quantitatively assess and report the climate alignment of their portfolios, thereby supporting their shipowner clients to transition their shipping fleets to meet the IMO decarbonisation trajectory. The PPMI are grounded in enhancing commitments and responsibility for climate mitigation; they are a principles-based, private regulatory tool developed by market participants. They are built on other insurance-sector sustainability initiatives that are constructed around public regulatory values of collaboration, reporting and disclosure and transparency. These regulatory techniques of the PPMI are aimed at compliance rather than contractual enforcement. This chapter examined the PPMI from a legal and regulatory perspective and analysed their impact in marine insurance law and practice. To that extent, it explained the role of the PPMI in marine insurance law and markets, but when we explain a concept, ‘the description is likely to have normative implications’.109 That has been the case here, as it has not only explained the PPMI but has situated it within broader sustainability initiatives and contract law and regulation. In doing so, it has not simply examined what the PPMI are but considered whether they could be an effective blueprint for the marine insurance industry to mitigate climate change. Climate change is an evolving threat and ESG risks continue to present challenges for the marine insurance sector due to a lack of understanding about ESG strategies and a lack of knowledge and expertise on ESG risks, which may lead to inadequate coverage of these risks through insurance due to exclusions or poor coverage under policies. The broader drive to net-zero of which the global insurance industry is seen as a key player has necessitated sectoral approaches to address climate alignment targets. The IMO’s decarbonisation agenda is a significant step as it requires that all industries work together to set tangible standards for implementation. The PPMI are a tool of governance that aims to make emissions monitoring and reporting an industry norm. In this respect, private governance frameworks in insurance law can play a greater role to address the effects of climate change. The PPMI are set to evolve as commitments to address climate change become more concrete. The PPMI are not at the stage where they can directly incentivise commitments to climate change, as it is important to first understand the key performance indicators before insurers can consider whether to link pricing to a vessel’s GHG emissions or to create enforceable green obligations. Marine insurers have recognised that their role in climate change should not be confined to paying claims for climate-related losses; rather, they should be proactive in mitigating the very causes of climate change, namely, in this case, in the reduction of GHG emissions from vessels.

Abraham (n 40) 655.

109

17. The German Pandemic Exclusion Clause and its write-back clauses Dieter Schwampe

1.

THE PANDEMIC EXCLUSION CLAUSE

The Pandemic Exclusion Clause uses the term pandemic both in its name and in its text, linking it in para. 3 to a determination by the World Health Organization (WHO) based on the International Health Regulations 2005 (IHR 2005).1 The IHR 2005 themselves do not use the term pandemic. Instead, in Art. 6 (1) they speak of a public health emergency of international concern, which is defined in their Art. 1.2 1.1

Structure of the Clause

The clause has a complex structure. In para. 1, two different types of exclusion are stipulated. 1.1.1 The disease itself after public determination As the first alternative exclusion, para. 1.1 addresses loss, damage, liability, costs or expenses caused by a dangerous communicable disease, or its pathogens or the toxins they produce, which is classified either as a pandemic or as an epidemic. The definition of a dangerous communicable disease is contained in para. 2, which does not refer to but essentially combines definitions contained in ss. 2 nos 33 and 3a4 of the German Infection Protection Act (Infektionsschutzgesetz/IfSG). According to this definition, a dangerous communicable disease means any disease caused by pathogens or the toxins they produce that are communicated to humans directly or indirectly and that may, due to its severe clinical course or its way of transmission, pose a grave danger for the general public. The German originals of the clauses use the language of the IfSG in this respect. As there does not exist an official translation of the IfSG into the English language, for any particular items the German original of the clause should be considered.

Available at www​.who​.int/​publications/​i/​item/​9789241580496. [P]ublic health emergency of international concern’ means an extraordinary event which is determined, as provided in these Regulations: (i) to constitute a public health risk to other States through the international spread of disease and (ii) to potentially require a coordinated international response. 3 See: übertragbare Krankheit: eine durch Krankheitserreger oder deren toxische Produkte, die unmittelbar oder mittelbar auf den Menschen übertragen werden, verursachte Krankheit. 4 See: [B]edrohliche übertragbare Krankheit: eine übertragbare Krankheit, die auf Grund klinisch schwerer Verlaufsformen oder ihrer Ausbreitungsweise eine schwerwiegende Gefahr für die Allgemeinheit verursachen kann. 1 2

353

354  Research handbook on marine insurance law For the clause to operate, it is necessary but not sufficient that a disease qualifies medically as a dangerous communicable disease as defined. Rather, a certain public determination is required. According to para. 3 of the clause, this may be a determination by the WHO that the requirements for a public health emergency of international concern pursuant to IHR 2005 are met; alternatively, according to para. 4 of the clause, it may either be a determination of the event as an epidemic by the German Parliament on basis of the IfSG (para. 4.1)5 or a determination by any other state, according to the legislation applicable to its territory, that the requirements for an epidemic of national concern are met (para. 4.2). These additional criteria serve the purpose of avoiding practical problems in determining when a pandemic exclusion shall start to operate.6 1.1.2 Precautionary measures by governments or third parties As a second alternative exclusion, para. 1.2 excludes loss, damage, liability, costs, or expenses caused by, resulting from or in connection with a precautionary measure to prevent the (further) spread of the dangerous communicable disease. In this alternative, thus, it is not the dangerous communicable disease itself which has caused the loss, but measures taken to prevent or fight such a disease. This second alternative contains two sub-alternatives. First, it addresses measures imposed by a government authority, in particular the closing of borders, quarantine measures, inbound or outbound travel restrictions, plant/business closures, export bans, prohibition from practising certain professions, disinfection of corporate premises/equipment, making available for alternative utilisation or destruction of inventories or goods. Second, it relates to meas-

See: Der Deutsche Bundestag kann eine epidemische Lage von nationaler Tragweite feststellen, wenn die Voraussetzungen nach Satz 6 vorliegen. Der Deutsche Bundestag hebt die Feststellung der epidemischen Lage von nationaler Tragweite wieder auf, wenn die Voraussetzungen nach Satz 6 nicht mehr vorliegen. Die Feststellung nach Satz 1 gilt als nach Satz 2 aufgehoben, sofern der Deutsche Bundestag nicht spätestens drei Monate nach der Feststellung nach Satz 1 das Fortbestehen der epidemischen Lage von nationaler Tragweite feststellt; dies gilt entsprechend, sofern der Deutsche Bundestag nicht spätestens drei Monate nach der Feststellung des Fortbestehens der epidemischen Lage von nationaler Tragweite das Fortbestehen erneut feststellt. Die Feststellung des Fortbestehens nach Satz 3 gilt als Feststellung im Sinne des Satzes 1. Die Feststellung und die Aufhebung sind im Bundesgesetzblatt bekannt zu machen. Eine epidemische Lage von nationaler Tragweite liegt vor, wenn eine ernsthafte Gefahr für die öffentliche Gesundheit in der gesamten Bundesrepublik Deutschland besteht, weil 1. die Weltgesundheitsorganisation eine gesundheitliche Notlage von internationaler Tragweite ausgerufen hat und die Einschleppung einer bedrohlichen übertragbaren Krankheit in die Bundesrepublik Deutschland droht oder 2. eine dynamische Ausbreitung einer bedrohlichen übertragbaren Krankheit über mehrere Länder in der Bundesrepublik Deutschland droht oder stattfindet. Solange eine epidemische Lage von nationaler Tragweite festgestellt ist, unterrichtet die Bundesregierung den Deutschen Bundestag regelmäßig mündlich über die Entwicklung der epidemischen Lage von nationaler Tragweite. 6 See in this respect the discussion by Rixecker in Schmidt, COVID-19 – Rechtsfragen zur Corona-Frise [COVID-19 – Legal Questions of the Corona Crisis], 3rd ed. 2021, § 12 marginal note 8 et seq; see also Günther/Piontek, Die Auswirkungen der “Corona-Krise” auf das Versicherungsrecht [The Impact of the “Corona Crisis” on Insurance Law], r+s [Recht und Schaden] 2020, p.242 et seq. 5

The German Pandemic Exclusion Clause and its write-back clauses  355 ures imposed by a third party involved in the legal or economic interest of the insured, in particular the closure of port, handling or storing facilities. Most notably, none of those two sub-alternatives requires that any governmental body has established that an epidemic or pandemic is occurring. The exclusion thus operates at a time at which such governmental bodies may have yet to become active. 1.2

Priority and Relationship with Other Clauses

The clause claims priority over all other agreements in the insurance contract. The clause does not contain any reference to individual agreements. This does not mean, however, that the clause overrides individual agreements. Under German law, the provisions of the clause qualify as standard business conditions. As per § 305 b of the German Civil Code (Bürgerliches Gesetzbuch/BGB), individual agreements always prevail over standard business conditions.7 If the policy contains an individual agreement on cover for a pandemic risk, thus, such individual agreement will not be affected by the exclusion clause claiming priority.8 This is different if the policy does not contain individual agreements but only standard insurance conditions. Where such conditions do not expressly address pandemic risks, but impliedly include them, the exclusion clause will have priority and exclude such risks from cover. In particular this is the case for all-risks policies, which do not expressly exclude pandemic risks and therefore include them as ‘silent pandemic’. However, it is accepted that there are different types of standard conditions, namely ‘printed’ and ‘written’ conditions.9 ‘Printed’ conditions, for example, are all conditions drafted by the GDV or a similar association, while ‘written’ conditions may be clauses developed by brokers and used not only for one client and risk but for all the broker’s clients. It is settled that in case of conflicts between written and printed conditions, the former will prevail over the latter.10 If a policy contains written provisions and those conditions expressly provide cover for pandemic risks, such cover will prevail over the exclusion by the clause. As far as measures by public authorities are concerned, the Pandemic Exclusion Clause partly overlaps with the exclusion of sovereign acts in section 2.4.1.3 DTV Cargo (‘confiscation, deprivation of possession or other acts of authorities’),11 because protective measures by public authorities always qualify as other acts of authorities.12 As regards causation, German 7 For details see Fornasier in Münchener Kommentar zum BGB [Munich Commentary on the Civil Code], 9th ed. 2022, § 305b marginal notes 5 et seq. 8 Schimikowski, Versicherungsvertragsrecht [Insurance Contract Law], 6th ed. 2017, Part 1, Ch. IV marginal note 18. 9 Armbrüster in Prölss/Martin, Versicherungsvertragsgesetz [Insurance Contract Act], 31st ed. 2021, Introduction marginal notes 291 et seq. 10 Bundesgerichtshof [Federal Court of Justice], dec. of 25.10.1960, court file number II ZR 35/60, VersR [Versicherungsrecht] 1963, p.32; Court of Appeal of Hamm, dec. 15.01.1960, court file number 7 U 150/59, VersR 1960, 687; Court of Appeal of Düsseldorf, dec. of 01.02.1966, court file number 4 U 287/65, VersR 1957, 1189; Court of Appeal of Cologne, dec., of 05.07.73, court file number 166/72, VersR 1974, 877; Court of Appeal Koblenz, dec. of 07.06.73, court file number 1079/72, VersR 1975, 29. 11 See www​.tis​-gdv​.de/​wp​-content/​uploads/​tis/​bedingungen/​avb/​ware/​2011​_W1​_Volle​_Deckung​ .pdf 12 Cf. for this exclusion Ehlers in Thume/de la Motte/Ehlers, Transportversicherungsrecht [Transport Insurance Law], 2nd ed. 2011, Ch. 5 marginal note 133 et seq.

356  Research handbook on marine insurance law marine insurance law customarily applies the doctrine of proximate causation when it comes to determining which of various insured and uninsured perils caused the event insured.13 If an act of authorities, as an excluded peril under section 2.4.1.3 DTV Cargo, has not been the proximate cause, cover will remain intact. Under the Pandemic Exclusion Clause, in contrast, cover is lost if the excluded peril only contributed, even though not necessarily as proximate cause. For details on causation see below at 1.3.3. Under DTV Cargo, the exclusion of cover for acts of authorities can be written back by the GDV Confiscation Clause.14 However, such re-inclusion of the risks of sovereign acts in the policy does not affect or invalidate the pandemic exclusion. On the one hand, the Confiscation Clause itself states that it only fills the gap caused by the exclusion in section 2.4.1.3 DTV Cargo. The scope of the Confiscation Clause is, thus, limited from the outset. In addition, para. 5 of the Pandemic Exclusion Clause expressly states that it applies to the entire insurance contract, including all extensions of cover. However, the reader only discovers this at the end of the clause – unlike the cyber/blackout clause, also published in 2021, which makes this clear in its introductory words. 1.3

Elements of the Clause

1.3.1

Dangerous communicable disease

1.3.1.1 Terminology The term originates from the IfSG. The IfSG distinguishes between transmissible diseases, that is, diseases caused by pathogens or their toxic products that are directly or indirectly transmitted to humans, and – as their qualified form – the dangerous communicable disease. The latter is defined in s. 2 no. 3a IfSG as a transmissible disease which, due to its clinically severe course or the way it spreads, can cause a serious danger to the general public.15 Para. 1.1 and para. 2 combine both, but address not only causation by disease but also causation by the causative agent of the disease. Insofar as toxic products are mentioned, linguistically they are not toxic products of the pathogen (but see s. 2 no. 3 IfSG16) but toxic products of the disease. From the point of view of infection medicine, this is a questionable approach, because pathogens can develop toxic products, but not diseases. However, causation by a dangerous communicable disease only leads to exclusion if it has been classified by the WHO as a pandemic or by the German Bundestag or another state as an epidemic of national significance according to para. 3. The classification must already exist at the time of causation. A subsequent classification does not affect cover for damage that has already occurred.

Schwampe, Die Bestimmung der Causa Proxima in der Transportversicherung durch die Rechtsprechung [The Determination of the Proximate Cause in Transport Insurance by the Courts], RdTW [Recht der Transportwirtschaft] 2020, 85. 14 See www​.tis​-gdv​.de/​wp​-content/​uploads/​tis/​bedingungen/​avb/​ware/​2011​_W6​_confiscation​.pdf 15 For details see Gerhardt, Infektionsschutzgesetz [Infectious Disease Act], 6th ed. 2022, § 2 marginal notes 20 et seq. 16 Ibid. at marginal note 23. 13

The German Pandemic Exclusion Clause and its write-back clauses  357 1.3.1.2 Classification by the WHO As regards the classification by the WHO, para. 4 refers to Art. 1 and Annex 2 IHR. However, neither Art. 1 nor Annex 2 provide the basis for such a classification. Art. 1 only contains definitions and defines a ‘public health emergency of international concern’ as an extraordinary event which is determined to constitute a public health risk to other states through the international spread of disease and to potentially require a coordinated international response. Annex 2 of the IHR is a flow chart, not for WHO decisions, but for ‘national surveillance and response systems’. If certain criteria are met, state parties should – as the final stage of the flow chart – report the event to the WHO. However, such a report does not constitute a classification by the WHO. The actual determination – the IHR do not speak of classification but of determination – follows the procedure laid down in Arts 12, 48 and 49 IHR. The procedure begins with a preliminary determination alone by the WHO Director-General (Art. 12(2) IHR).17 This is sufficient for the purposes of para. 4. Para. 4 does not address the consequences if the WHO through its Director-General determines, in accordance with Art. 12(5) IHR,18 that a public health emergency of international concern has ended. In view of the close link between the cover exclusion and the WHO determinations, however, the clause must be interpreted in such a way that the exclusion no longer has any effect from that moment on.19 1.3.1.3 Classification by the German Parliament According to s. 5 (1) sentence 1 IfSG, the German Parliament may (not must) resolve the existence of a dangerous communicable disease if the conditions according to s. 5(1) sentence 6 IfSG are met. This requires serious danger to public health existing throughout the Federal Republic of Germany.20 According to s. 5 (1) sentence 6 no. 1, this prerequisite is always met if the WHO has determined a public health emergency of international concern, and if the introduction of a dangerous communicable disease into the Federal Republic of Germany is imminent21 – in this case, however, the prerequisites of para. 4 of the clause are already met.

‘If the Director-General considers, based on an assessment under these Regulations, that a public health emergency of international concern is occurring, the Director-General shall consult with the State Party in whose territory the event arises regarding this preliminary determination. If the Director-General and the State Party are in agreement regarding this determination, the Director-General shall, in accordance with the procedure set forth in Article 49, seek the views of the Committee established under Article 48 (hereinafter the “Emergency Committee”) on appropriate temporary recommendations.’ 18 ‘If the Director-General, following consultations with the State Party within whose territory the public health emergency of international concern has occurred, considers that a public health emergency of international concern has ended, the Director-General shall take a decision in accordance with the procedure set out in Article 49.’ 19 See in respect of the temporary effect of sanctions clauses Sigl in Münchener Kommentar zum VVG [Munich Commentary on the Insurance Contract Act], 2nd ed., Ch. 500 marginal note 511 et seq.; Wandt, Versicherungsverbote im Rahmen von Embargomaßnahmen [Prohibition of Insurance in the context of Embargo Measures], VersR 2013, p.257; Heinisch, Die praktische Umsetzung von Sanktionen in der (Rück-) Versicherungswirtschaft [Practical Implementation of Sanctions in the (Re-) Insurance Industry], CCZ [Zeitschrift für Corporate Compliance] 2012, p.136. 20 Gerhardt, Infektionsschutzgesetz [Infectious Disease Act], 6th ed. 2022, § 5 marginal note 5. 21 See Gausing in BeckOK Infektionsschutzrecht [Beck Online Commentary Infectious Disease Law], 16th ed. of 10.01.2023, § 5 IfSG marginal note 18 et seq. 17

358  Research handbook on marine insurance law In such scenarios, the requirements of para. 3 of the clause are met and so it is no longer necessary to refer to para. 4. However, according to s. 5 (1) sentence 6 no. 2 IfSG, the German Parliament may also become active if a dynamic spread of a dangerous communicable disease is imminent or takes place in several Länder, that is, the states of the Federal Republic of Germany.22 This is the actual scope of application of para. 4. As with para. 3 of the clause, para. 4 does not address the question of whether the exclusion is only effective as long as Parliament’s determination of the situation has legal effect. In this respect, the mechanisms under the IfSG are different from those under the IHR. According to s. 5 (1) sentence 3 IfSG, the determination of an epidemic situation of national importance is repealed if the German Bundestag does not determine its continued existence no later than three months after the determination of such a situation.23 As is the case with WHO determinations, in view of the purpose of the regulation, the exclusion loses its effect as soon as the effects of the determination either are lifted or cease to exist. For foreign states, it is not determined which state institution has made determinations. In view of the many different forms of states and different state organisations, the clause cannot be specific. The competence of the foreign body to make such a determination is to be determined on the basis of the foreign legal provisions. 1.3.1.4 The disease as the cause of the loss As to the cause of the loss, this must be the disease or its agents or toxic products. This includes cases in which pathogens or toxic products impinge the surface of things, which under German law constitutes damage if it impairs the ability to use the product.24 As causation by disease is given ‘without regard to other causes’, which qualifies as a deviation from the proximate cause rule,25 this exclusion can apply in many cases. For details of causation see A.III.3. However, the exclusion also operates if a person falls ill and causes damage as a result of the illness. In fact, due to the broad causation aspect, such damage caused by the disease would also fall under para. 1.1. However, they are regulated separately and conclusively in para. 1.2. The distinction is significant, because para. 1.1 requires the disease to be classified as a pandemic or epidemic whereas para. 1.2 does not. 1.3.2 Protective measures The clause distinguishes between measures by authorities in para. 1.2.1 and measures by third parties in para. 1.2.2. Both require a chain of causation. A dangerous communicable disease must have given rise to the protective measure, which then causes the damage and so on.

See marginal note 21. Gerhardt, Infektionsschutzgesetz [Infectious Disease Act], 6th ed. 2022, § 5 marginal note 10 et

22 23

seq.

Cf. Segger in Martin/Reusch/Schimikowski/Wandt, Sachversicherung [Property Insurance], 4th ed. 2022, § 2 marginal notes 38 et seq. 25 See insofar in the identical aspect in respect of the nuclear and biochem weapons exclusions clause in the German H&M Conditions, cl. 36 and 39 DTV-ADS: Schwampe, Seeschiffsversicherung [Ocean Vessel Insurance], cl. 36 marginal note 11 and cl 39 marginal note 15. 24

The German Pandemic Exclusion Clause and its write-back clauses  359 The clause does not specify what qualifies as protective measures. It is therefore irrelevant whether they were necessary and appropriate, useless or excessive, legal or illegal.26 The prerequisite, however, is that there has been a dangerous communicable disease, because the protective measure must have been applied in order to prevent the (further) spread of the disease. Since para. 1.2 only refers to the dangerous communicable disease, but not to its classification as a pandemic or epidemic, the exclusion also applies to measures taken before such classifications. 1.3.2.1 Protective measures by public authorities Para. 1.2.1 merely refers to public authorities. This includes both German public authorities and any foreign public authority. While the clause refers to certain special activity, namely, the closing of borders, quarantine measures, inbound or outbound travel restrictions, plant/ business closures, export bans, prohibition from practising certain professions, disinfection of corporate premises/equipment, making available for alternative utilisation or destruction of inventories or goods, this is not an exhaustive list. The express statement that these activities are addresses in particular shows that they are meant to be examples only.27 1.3.2.2 Protective measures by third parties The third party addressees in para. 1.2.2 may, but not must, be appointed by the policyholder or insured person. It is sufficient if the third party was engaged in their interest. This includes all third parties that, for example, a freight forwarder engages for the performance of a transport contract. In their case, too, the measure taken must be aimed at preventing the (further) spread of a dangerous communicable disease. Thus, if a third party closes its business because it can no longer obtain operating resources or its employees cannot reach the workplace, this is not a protective measure. 1.3.3 Causation Under ss. 130 et seq. VVG, marine insurance and transport insurance are subject to the causa proxima doctrine. Where insured and uninsured causes have contributed to the occurrence of a loss, there is cover only if the insured peril was the proximate cause of the loss. In German law, the causa proxima doctrine is regarded as customary law, but is not mandatory and can be modified by contract.28 Para. 1 represents such a variation. Para. 1 starts by stating that the exclusion shall operate irrespective of contributory causes and then continues with slightly different wording for paras 1.1 and 1.2. Para. 1.1 merely requires that the loss must have been caused by a dangerous communicable disease. Para. 1.2, in contrast, addresses that the loss was caused by, resulted from or occurred in connection with a precautionary measure. However, the broader language used in para. 1.2 does not imply any 26 See insofar the identical aspect in respect of the exclusion of act of public authorities, Oberster Gerichtshof Österreichs [Supreme Court of Austria], dec. of 10.07.1986, court file number 7 Ob 7/86, VersR 1988, 198; Schwampe, Seeschiffsversicherung [Ocean Vessel Insurance], cl. 38 marginal note 4. 27 See the same aspect for enumerated perils in 3 28 ADS and cl. 27 DTV-ADS Schwampe, Seeschiffsversicherung [Ocean Vessel Insurance], cl. 27 marginal note 36. 28 For details to the causa proxima doctrine as applied in German law see Schwampe in Bruck/ Möller, Versicherungsvertragsgesetz [Insurance Contract Act], 9th ed. 2016, § 130 marginal note 58; for the various definitions developed by the Courts see Schwampe, Die Bestimmung der Causa Proxima in der Transportversicherung durch die Rechtsprechung [The Determination of the proximate cause in Transport Insurance by the Courts], RdTW [Recht der Transportwirtschaft] 2020, 85.

360  Research handbook on marine insurance law different effects. The decisive words are already contained in the opening part of para 1 (‘irrespective of contributory causes’) and are sufficient to also cover for para. 1.1 what para 1.2 expresses. The background of the wording used in para 1.2 is the use of clauses frequently used in reinsurance contracts. Nevertheless, it is questionable that GDV used different language without the intention of creating different effects. In any event, the effect is that risks excluded by the Pandemic Exclusion Clause prevent coverage even if they are mere remote causes, causae remotae, in the sense of the causa proxima doctrine.29 It is therefore sufficient that the circumstances specified in detail in the clause are merely contributory causes. In this respect, the adequacy doctrine of the general German civil law applies. This means that only such circumstances are not causative for a loss which lead to the loss only under extremely unusual circumstances.30 Where the causa proxima rule does not apply, as in transport liability insurance, the adequacy doctrine applies in any case, whereby an adequately contributing excluded cause prevails and excludes insurance. 1.3.4 Burden of proof Under German insurance, the burden of proof for exclusions rests with the insurer.31 As the Pandemic Exclusion Clause is of this type, the burden of proof lies with the insurer on all points. This means that he must also prove the existence of a dangerous communicable disease. Classifications by the WHO according to clause 3 or by the German Parliament according to para. 4.1 should be easy to prove, but the insurer must prove the existence of the disease at the time when the peril struck,32 which is not always the time when the loss occurs. The burden of proof may be more difficult to discharge in case of determinations by other states according to para. 4.2 of the clause, because the provisions applicable there must also be proven. However, the classifications according to paras 3 and 4 do not make superfluous the proof that a dangerous communicable disease actually exists. It can be even more difficult to prove if none of the above-mentioned findings have been made, but measures have nevertheless been initiated by the authorities or – certainly even more problematic – by third parties. The insurer must also prove that the measures taken were intended to prevent the (further) spread of the disease. As far as third parties are concerned, the insurer must prove that they were involved in the legal or economic interest of the policyholder or insured person.

29 Ritter/Abraham, Das Recht der Seeversicherung [The Law of Marine Insurance], 2nd ed. 1967, § 28 marginal note 22; Enge/Schwampe, Transportversicherung [Transport Insurance], 4th ed. 2012, p.86. 30 Bundesgerichtshof [Federal Court of Justice], dec. of 27.01.1981 court file number VI ZR 204/79, NJW [Neue Juristische Wochenschrift] 1981, p.983; dec. of 04.07.1994 court file number II ZR 126/93, NJW 1995, p.126. 31 Bundesgerichtshof [Federal Court of Justice], dec. of 23.6.2004 court file number IV ZR 130/03, VersR 2004, 1039; dec. of 17.12.2014, court file number IV ZR 90/13, VersR 2015, 181. 32 Hanseatic Court of Appeal oh Hamburg, dec. of 03.11.2011, court file number 6 U 181/08, TranspR [Transportrecht] 2012, 152.

The German Pandemic Exclusion Clause and its write-back clauses  361

2.

THE WRITE-BACKS OF THE EXCLUDED PERILS

GDV has developed three different clauses for the partial reinsurance of the pandemic risk: for insurance of ships, for cargo insurance and for transport liability insurance. All three write-back clauses follow the same structure. Para. 1 always contains the actual write-back, which is different for all three classes of insurance. Para. 2 (for cargo insurance: para. 2.2), determines limits per event and insurance year. Para. 3 deals with termination of the write-back, and para. 4 stipulates that the write-back does not otherwise extend the insurance cover beyond its original scope, that is, the scope it had without the effects of the Pandemic Exclusion Clause. The write-back for transport liability insurance and insurance of ships contain options. Only the write-back for insurances of ships affects the causation rule and re-introduces the causa proxima rule. 2.1

Write-back for Insurances of Ships

2.1.1 Two options The write-back clause contains two options: Option 1 in para. 1 and Option 2 in para. 2. Already in its header, para. 1 stipulates that Option 1 applies if para. 2 has not been agreed on. The effect is that the plain inclusion of the write-back as such always leads to the application of Option 1.33 If Option 2 is not chosen, however, this does not mean that automatically everything stipulated in para. 1 is included in the contract. Rather, the extension of cover in para. 1.2.1 requires an additional agreement between the parties. 2.1.2

Option 1 (para. 1)

2.1.2.1 Reintroduction of the proximate cause doctrine (para. 1) As mentioned above, the Pandemic Exclusion Clause has set aside the causa proxima rule otherwise applicable in German marine insurance law, with the effect that mere causative contribution of an excluded peril, be it as a causa remota only, excludes the insurance cover. This effect is repealed in Option 1. Thus, under the Pandemic Exclusion Clause as amended by Option 1 of the write-back, the causa proxima rule remains in force, so that cover is excluded only if one of the perils named in the Pandemic Exclusion Clause was the proximate cause of the loss. Only the write-back for insurance of ships provides for a reinstallation of the causa proxima rule. The write-back for cargo insurance does not contain this feature. In transport liability insurance, the causa proxima rule does not apply anyway. 2.1.2.2

Crew conduct (para. 1.2.1)

2.1.2.2.1 Special agreement on application As mentioned above, Option 1 applies automatically if the parties have not agreed on the application of Option 2, contained in para. 2. However, this does not mean that cover for crew conduct also applies automatically. Rather, the parties have to agree on the application of para. 33 See for the similar effect of a preferred option in cl. 33 DTV-ADS: Schwampe, Seeschiffsversicherung [Ocean Vessel Insurance], cl. 33 marginal note 5.

362  Research handbook on marine insurance law 1.2.1 separately. In the absence of such a special agreement, Option 1 of the write-back is agreed, but not para. 1.2.1. 2.1.2.2.2 Cover provided and scope of application Para. 1.2.1 stipulates that, in deviation from para. 1 of the Pandemic Exclusion Clause and only within the scope of the provisions of the insurance contract, loss or damage caused by the conduct, which includes acts and omissions, of a crew member infected with a dangerous communicable disease as per para. 1.1 of the Pandemic Exclusion Clause is insured while carrying out such crew member’s professional or operational tasks. Although the provision addresses para. 1 of the Pandemic Exclusion Clause generally, and thus also the protective measures regulated in para. 1.2 of the Pandemic Exclusions Clause, para 1.2.1 of the write-back only deals with damage caused by crew members and is, thus, limited to para. 1.1 of the Pandemic Exclusion Clause. ‘Crew’ within the meaning of the write-back is the ship’s company (term used in the translation of the law by the German Transport Law Association)34 as defined in s. 478 of the German Commercial Code: the master, the officers, the crew as well as all other persons involved in the operation of the ship, who have been employed by the shipowner or by the bareboat charterer, or who have been put at the disposal of the shipowner or bareboat charterer by a third party for the purpose of performing work relating to the ship’s operation, and who are subject to the orders of the master. In order for the write-back to be effective, the loss must have been caused by the crew in the course of its occupational or operational activity. This means that the loss caused during leisure time will not be covered. To build an example: a crew member who is ill, smokes in bed and drops the cigarette due to illness is not carrying out his professional activity at that moment. There will be no cover for resulting damage to the vessel. 2.1.2.2.3 Limits The write-back refers to limits per loss event and insurance year, but, for reasons of German antitrust legislation, does not fix the limits itself. If the parties do not agree on limits in the individual case, the insurer is liable up to the amount insured as agreed in the policy.35 Ship insurance in Germany usually only is on basis of sums insured, without any sub-limits. Limits for loss events and insurance years are generally alien to marine insurance. The write-back, thus, introduces a new feature into insurance of ships. 2.1.2.3

Loss of hire

2.1.2.3.1 No requirement of a special agreement While para. 1.2.1 of the write-back requires a special agreement that this write-back shall apply, para. 1.2.2 does not contain such requirement. Cover provided by this provision, thus, is available simply upon inclusion of the write-back in the policy.

See https://​dgtr​.de/​wp​-content/​uploads/​HGB​_5​_Buch​_Uebersetzung​_DGTR​_2015​_05​_11​.pdf For the similar effect of a lack of agreement on a deductible provided for in insurance conditions, see Schwampe, Seeschiffsversicherung [Ocean Vessel Insurance], cl. 56 marginal notes 24 and 31. 34 35

The German Pandemic Exclusion Clause and its write-back clauses  363 2.1.2.3.2 Cover provided and scope of application The provision is applicable if the write-back as such is agreed and the parties have not agreed that para. 2 shall apply. In contrast to para. 1.2.1, there is no need to agree on the applicability of para. 1.2.2. The provision only applies to loss of insurance. There do exist loss of hire insurance conditions from GDV being Section Four of the DTV – German Standard Terms and Conditions of Insurance for Ocean-Going Vessels 2009 (DTV-ADS 2009).36 However, the Pandemic Exclusion Clause and the write-back for insurance of ships can also be agreed if insurance cover is provided on the basis of different terms, such as the Nordic Plan. It must be observed, though, that the insurance conditions in use in the German market, be it the Nordic Plan37 or DTV-ADS,38 require that the loss of hire was triggered by damage to the ship which is covered under the ship’s hull and machinery insurance. If hull and machinery and loss of hire are insured under separate policies, as is usually the case, it must first be considered whether the underlying damage to the ship was caused by a pandemic risk and whether the hull insurance contains a pandemic exclusion. If this is the case, there is no damage covered under the hull and machinery policy, which could trigger cover under the loss of hire insurance. If the loss of hire policy then includes the write-back for insurance of ships, this does not lead to coverage, because there is no underlying hull damage covered under the hull and machinery policy. Moreover, para. 1.2.2 of the write-back does not extend the cover of the underlying hull policy, but only extends loss of hire caused by the effects of a pandemic. The write-back provides cover for loss of income in certain enumerated scenarios. Loss of income is covered if the ship is unable to earn the full freight or hire as a consequence of a hull damage covered under the ship’s hull and machinery insurance; as a consequence of the inaccessibility, limitation or exclusion of the operational availability or prevention from leaving a shipyard or other place of repair; as a consequence of the delayed presence or unavailability of required service technicians or specialists for repairs; or as a consequence of delayed delivery or unavailability of spare parts required for the repair, which was caused by a dangerous communicable disease as per para. 1.1 of the Pandemic Exclusion Clause (inexplicably, the clause here uses the term perilous communicable disease), or which was caused by a precautionary measure as per para. 1.2 of the Pandemic Exclusion Clause. It is explicitly stated that the risks excluded by para. 1.1 and para. 1.2 of the Pandemic Exclusion Clause are written back. In the case of para. 1.1 of the Pandemic Exclusion Clause, this means that a determination as a pandemic or epidemic must have been made in accordance with paras 3 and 4 of the Pandemic Exclusion Clause. If this was not the case the exclusion does not apply, so there is no need for a write-back. Cover is then available not only within the limits of the write-back, but already under the terms of the loss of hire insurance.

English translation available at www​.tis​-gdv​.de/​wp​-content/​uploads/​tis/​bedingungen/​avb/​see/​DTV​ -ADS2009​_EN​.pdf. 37 Cl. 16-1 Nordic Plan: ‘The insurance covers the assured’s loss of income due to the vessel being wholly or partially deprived of income-earning activity as a consequence of damage to the vessel which is recoverable under the conditions of the Plan, or which would have been recoverable if no deductible had been agreed, see Cl. 12-18.’ 38 Cl. 70.1 DTV-ADS: ‘The Insurer will indemnify for loss of income from the insured vessel for the period during which the vessel is unable to earn the full freight or hire due to an indemnifiable hull damage.’ 36

364  Research handbook on marine insurance law The write-back is only relevant where there is no loss of hire cover available because of the pandemic exclusion. All other conditions for a claim under the applicable insurance conditions of the loss of hire insurance must, therefore, be met. The write-back is, therefore, especially important where no pandemic or epidemic situation has (yet) been established in accordance with paras 3 and 4 of the Pandemic Exclusion Clause. 2.1.2.3.3 Limits The provision stipulates that both the daily rate and the duration of cover (number of days) shall be agreed separately. The provision is silent as to what happens when there is no such agreement but merely the inclusion of the write-back in the loss of hire policy. In such case, the effect is not that there is no effective write-back. Rather, there will be cover within the limits agreed in the loss of hire policy.39 In contrast to paras 1.2.1 and para. 2, an annual maximum is not provided for. 2.1.3

Option 2 (para. 2)

2.1.3.1 Application Option 2 provided for in this provision does not already apply with the agreement of the re-inclusion clause, but only if the parties have additionally agreed on it expressly. 2.1.3.2 Cover provided While Option 1 only changes the causation rule (para. 1.1) and makes partial reinstatements of cover for occupational crew behaviour and certain causes of loss of hire, Option 2 fully reinstates the cover in terms of perils, but limits it in para. 2.2. Option 2 is, therefore, intended for cases where insurers have no fundamental problem with covering pandemic risks but only want to limit the accumulation risk. 2.1.3.3 Limits The clause provides for limitation per any one loss or occurrence and for all occurrences within the policy year. In contrast to para. 1.2.1 of the write-back, the limits here are not limited to crew behaviour, but apply to all losses caused by a dangerous communicable disease or a protective measure, etc. This also includes loss of hire dealt with in para. 2.2.2 of the write-back. 2.1.3.4 Special limits for loss of hire insurance In addition to the limitation by the limits of para. 2.2.1, the limitations of para. 2.2.2 shall apply for the loss of hire insurance. These additional limits apply to loss of hire caused by inaccessibility, limitation or exclusion of the operational availability or prevention from leaving a shipyard or other place of repair; delayed presence or unavailability of required service technicians or specialists for repairs; and delayed delivery or unavailability of spare parts required for the repair. For these cases, the provision requires the agreement of a daily rate and a maximum period of time. This leads to the question of the relationship of the two limits. This is unproblematic insofar as para. 2.2.2 introduces limitations which are not also found in para. 2.2.1. This is the case with the daily rate to be agreed. The provision has no effect if the parties leave it to agreeing Option 2 without agreeing a special daily rate. In that case, loss of hire arising

See footnote 35.

39

The German Pandemic Exclusion Clause and its write-back clauses  365 is subject to the daily rate otherwise agreed in the insurance contract.40 If, however, a special daily rate is agreed, then the insurance benefit is limited to this amount per day, irrespective of the limits of para. 2.2.1 and of the main policy. However, the parties shall also agree on a maximum number of days covered. If a specific number of days is agreed, the combination of daily rate and days covered potentially competes with the loss event maximum of para. 2.2.1. If the product of daily rate and days is below the event maximum agreed under para. 2.2.1, the lower amount under para. 2.2.2 will prevail, as this lower limit will apply ‘in addition’. If it is higher, the question arises whether the lower event maximum of para. 2.2.1 can be decisive. In the end, it cannot. While the limitations of para. 2.2.2 only apply to ‘additional’, which means that the limits of para. 2.2.1 shall remain valid, para. 2.2.2 and the agreements made in this respect then represent the more specific regulations for loss of hire. This speciality has conclusive priority. 2.1.3.5 Termination 2.1.3.5.1 Right of termination The purpose of the right to terminate is to provide, in addition to the agreed limits, further protection against claims accumulation. The provision requires the parties to agree on a notice period. If the parties fail to agree on such period, the write-back is not terminable without a notice. Rather, as the termination becomes effective only after expiry of the notice period, without an agreed period such period cannot expire with the consequence that the termination cannot become effective. The parties can certainly agree on immediate effect by setting the notice period at zero days. However, the absence of any agreement is not to be equated with the agreement of immediate effect of termination. In the absence of an agreement, a notice of termination is simply not effective. 2.1.3.5.2 Party entitled The provision does not restrict the right of termination to the insurer, but the policyholder will not have cause to cancel a write-back. Under German law, if the policy is a subscription policy involving more than one insurer, there exist as many legally independent insurance contracts as there are co- insurers.41 The power of attorney of the co-insurers in favour of the leading insurer contained in clause 9.2 DTV-Hull Clauses42 and clause 19.2 DTV-ADS43 also covers the declarations of termination.

See footnote 35. Court of Appeal of Hamm, dec. of 10.06.1983, court file number 20 U 232/82, VersR 1984, 149; Court of Appeal Bremen, dec. of 13.01.1994, court file number 2 U 104/93, VersR 1994, 710; Hanseatic Court of Appeal Hamburg, dec. of 19.02.2008, court file number 6 U 119/07, VersR 2008, 1249; Schwampe, Seeschiffsversicherung [Ocean Vessel Insurance], cl. 19 marginal note 13. 42 Schwampe, Seekaskoversicherung [Marine Hull Insurance], cl. 9 marginal note 9. 43 Schwampe, Seeschiffsversicherung [Ocean Vessel Insurance], cl. 19 marginal note 21 et seq. 40 41

366  Research handbook on marine insurance law 2.2

Write-back for Cargo Insurance

Standard market conditions in Germany are the DTV Cargo 2000 conditions, last updated in May 2020.44 They provide for all-risk insurance. The Pandemic Exclusion Clause excludes the perils addressed above. The cargo write-back does not simply re-include the perils excluded (as, for example, para. 2 of the write-back for insurance of ships does). What is excluded by the Pandemic Exclusion Clause is not even addressed here. Rather, para. 1 lists various perils which shall be covered if they have caused the loss or damage. 2.2.1 Cover provided The cargo write-back obviously requires that a dangerous communicable disease is causative. However, it must be causative together with other perils enumerated in para. 1 of the write-back, namely theft, robbery, embezzlement or other disappearance; accidents involving the means of transport carrying the goods; collapse of warehouse buildings; fire, lightning, explosion, earthquakes, seaquakes, volcanic eruptions or other natural disasters; collision with or crashing of a flying object or parts thereof, including its cargo; general average sacrifice; jettison, washing overboard or otherwise being lost overboard as a result of heavy weather; discharging, interim storage, loading of cargo at a port or airport of distress entered as a result of an insured event or following an emergency landing; and total loss of entire packages during loading onto or unloading from a means of transport, or during transhipment to or from a means of transport. This changes the usual all-risk policy to named perils cover where a dangerous communicable disease is involved. 2.2.2 Limits Like the write-back for insurance of ships, the cargo write-back contains, in para. 2, limits for occurrence and for the insurance year. Also here the amounts must be agreed, and, in the case of absence of an agreement, the limits agreed in the policy as such apply. 2.2.3 Termination Para. 3 provides for a termination right. It is in exactly the same wording as the termination clause in the write-back for insurance of ships. 2.3

Write-back for Liability Insurance for Carriers, Freight Forwarder and Warehouse Operators

2.3.1 Two options The underlying insurance conditions for the liability insurance are those of the DTV Terms & Conditions of Liability Insurance for Open-Cover Policies of Carriers, Freight Forwarders and Warehouse Operators 2003/2011 (‘DTV-VHV’).45 According to cl. 1.1 DTV-VHV, there is cover for liability arising under service contracts, defined in brackets as carriage, forwarding and warehousing contracts, concluded during the term of the insurance policy by the poli44 Available at www​.tis​-gdv​.de/​wp​-content/​uploads/​tis/​bedingungen/​avb/​ware/​2011​_W1​_all​_risks​ .pdf. 45 Available at www​.tis​-gdv​.de/​wp​-content/​uploads/​tis/​bedingungen/​avb/​vhv/​DTV​_VHV​_laufende​ _Versicherung​_en​.pdf.

The German Pandemic Exclusion Clause and its write-back clauses  367 cyholder in his capacity as a carrier in road transport operations, as a freight forwarder or as a warehouse operator and placed in accordance with cl. 11, insofar as the services associated with these contracts are documented explicitly in the ‘business description’ section of the policy. Like the write-back for insurance of ships, this liability write-back comes with two options, Option 1 (para. 1.1) being applicable unless the parties agree expressly on the application of Option 2 (para. 1.2). The write-back is for liability under the contracts referred to in cl. 1.1 DTV-VHV. 2.3.2 Option 1 (para. 1.1) Option 1 applies, if the write-back as such is agreed between the parties, and if the parties did not agree on the application of Option 2, dealt with in para. 1.2. Option 1 fully re-includes the pandemic risks excluded by the Pandemic Exclusion Clause, but only within the limits agreed. The write-back expects the parties to agree on limits for the three different kinds of contracts addressed in para. 1, namely carriage, forwarding or warehousing contracts, and for each of those in respect of loss of and damage to cargo as well as pure financial loss. As a fourth category, the parties shall also agree on a limit for claims based on tort law, irrespective of the type of service contract or the nature of the loss/damage. Option 1, thus, is designed for insurers who are, as a matter of principle, prepared to take over pandemic risks and only want to control their exposures by way of cover limits. 2.3.3 Option 2 (para. 1.2) Option 2 only applies if the parties have not only agreed on the write-back as such, but also on the application of Option 2. The cover continues to be determined in principle according to cl. 1 and. 3 DTV-VHV. However, it is now limited to the risks specified in Option 2, namely exclusively for loss of or damage to cargo during transport or handling and for delays, non-delivery or perishing caused by faulty scheduling. This list can be extended by the parties, but the mentioned two classes are part of the printed conditions and apply to any cases in which Option 2 is agreed. 2.3.4 Limits and termination Paras 2 and 3 contain limits for per occurrence and per insurance year as well as a termination right. The wording is identical to that of the cargo write-back.

3. WORDINGS All following conditions are non-binding recommendations by the German Insurance Association (GDV) for optional use. Other conditions may be agreed. In case of deviations the German wording shall prevail.

368  Research handbook on marine insurance law

CLAUSE FOR THE EXCLUSION OF LOSS/DAMAGE DUE TO A DANGEROUS COMMUNICABLE DISEASE FOR THE USE IN MARINE INSURANCE (‘PANDEMIC EXCLUSION CLAUSE’) Standard policy conditions of the GDV Version July 2021 1

Notwithstanding other provisions in the insurance contract and irrespective of contributory causes, the cover does not include any loss/damage, liability, costs, or expenses 1.1 caused by a dangerous communicable disease (or its pathogens or the toxins they produce) within the meaning set out in paragraph 2 that is classified as a pandemic or epidemic as per paragraphs 3 or 4 or 1.2 caused by, resulting from, or in connection with a precautionary measure to prevent the (further) spread of the dangerous communicable disease within the meaning set out in paragraph 2, 1.2.1 imposed by a government authority, in particular the closing of borders, quarantine measures, inbound or outbound travel restrictions, plant/business closures, export bans, prohibition from practising certain professions, disinfection of corporate premises/equipment, making available for alternative utilisation, or destruction of inventories or goods or 1.2.2 imposed by a third party involved in the legal or economic interest of the Insured, in particular the closure of port, handling or storing facilities. 2 A dangerous communicable disease means any disease caused by pathogens or the toxins they produce that are communicated to humans directly or indirectly and that may, due to its severe clinical course or its way of transmission, pose a grave danger for the general public. 3 A dangerous communicable disease is classified as a pandemic if the World Health Organization finds that the requirements for a public health emergency of international concern pursuant to Article 1 in conjunction with Annex 2 of the International Health Regulations 2005 of the World Health Organization, third edition, or pursuant to similar successor regulations are met. 4 A dangerous communicable disease is classified as an epidemic if 4.1 the German Bundestag finds, pursuant to Section 5 of the Act on the Prevention and Control of Infectious Diseases in Man (Protection against Infection Act – IfSG) or pursuant to similar successor regulations, and/or 4.2 any other state finds, according to the legislation applicable to its territory, that the requirements for an epidemic of national concern are met. 5 Concluding provisions 5.1 This Clause applies to the entire insurance contract including all cover extensions. 5.2 The provisions of this Clause do not extend the existing cover. 5.3 This Clause shall only apply to the extent that this is not precluded by mandatory legal requirements on compulsory insurance.

WRITE-BACK CLAUSE ‘Dangerous Communicable Disease for the Use in Marine Insurance for Vessels’ Standard policy conditions of the GDV Version July 2021 1

Option 1 (applies unless paragraph 2 was agreed)

The German Pandemic Exclusion Clause and its write-back clauses  369 1.1 Causation By way of amendment to the ‘Clause for the Exclusion of Loss/Damage due to a Dangerous Communicable Disease for the Use in Marine Insurance’ (hereinafter: Pandemic Exclusion Clause), the following shall apply: • The wording ‘and irrespective of contributory causes’ in paragraph 1 of the Pandemic Exclusion Clause is deemed cancelled. 1.2 Write-back 1.2.1 Crew conduct To the extent agreed by the parties, in deviation from paragraph 1 of the Pandemic Exclusion Clause and only within the scope of the provisions of the insurance contract, loss/damage caused by the conduct (act or omission) of a crew member infected with a dangerous communicable disease as per subparagraph 1.1 of the Pandemic Exclusion Clause is insured while carrying out such crew member’s professional or operational tasks, the maximum indemnity being limited to EUR […] per loss occurrence and to EUR […] for all loss occurrences within an insurance year. 1.2.2 Loss of hire In deviation from paragraph 1 of the Pandemic Exclusion Clause and only within the scope of the provisions of the insurance contract, loss of income from the insured vessel for the period during which the vessel is unable to earn the full freight or hire as a consequence of a hull damage to be indemnified under the insurance contract due to • inaccessibility, limitation or exclusion of the operational availability or prevention from leaving the shipyard or other places of repair, • delayed presence or unavailability of required service technicians or specialists for repairs or • delayed delivery or unavailability of spare parts required for the repair caused by • a perilous communicable disease as per subparagraph 1.1 of the Pandemic Exclusion Clause, or • a precautionary measure as per subparagraph 1.2 of the Pandemic Exclusion Clause is insured, the amount to be indemnified by the insurer being limited to one daily rate (agreed value) of EUR […] for a maximum period of […] days per any one loss occurrence. 2 Option 2 (applies only, if agreed) 2.1 Write-back In deviation from paragraph 1 of the Pandemic Exclusion Clause and only within the scope of the provisions of the insurance contract and to the extent insured under the insurance contract, loss/ damage, liability, costs, expenses caused by • a dangerous communicable disease as per subparagraph 1.1 of the Pandemic Exclusion Clause, or • a precautionary measure to prevent the (further) spreading of a dangerous communicable disease as per subparagraph 1.2 of the Pandemic Exclusion Clause is insured. 2.2 Limits of indemnity 2.2.1 The maximum indemnity is limited to EUR […] per any one loss occurrence and to EUR […] for all loss occurrences within an insurance year. 2.2.2 In addition, the insurer’s indemnity for loss of income (to the extent that such is insured) from the insured vessel for the period during which the vessel is unable to earn the full freight or hire as a consequence of a hull damage to be indemnified under the insurance contract due to

370  Research handbook on marine insurance law • inaccessibility, limitation or exclusion of the operational availability or prevention from leaving the shipyard or other places of repair, • delayed presence or unavailability of required service technicians or specialists for repairs or • delayed delivery or unavailability of spare parts required for the repair is limited to the amount of one daily rate (agreed value) of EUR […] for a maximum period of […] days per any one loss occurrence. 3 The write-back pursuant to paragraph 1 or item may be terminated at any time in text form. The termination is effective […] days after receipt. 4 The write-back pursuant to paragraph 1 or paragraph 2 does not grant any additional insurance cover beyond the other provisions of the insurance contract.

WRITE-BACK CLAUSE ‘Dangerous Communicable Disease for the Use in Cargo Insurance’ Standard policy conditions of the GDV Version July 2021 1 In deviation from paragraph 1 of the ‘Clause for the Exclusion of Loss/Damage due to a Dangerous Communicable Disease for the Use in Marine Insurance’ and only within the scope of the provisions of the insurance contract, any damage, cost, or expenses caused by • theft, robbery, embezzlement, or other disappearance, • an accident involving the means of transport carrying the goods, • collapse of warehouse buildings, • fire, lightning, explosion, earthquakes, seaquakes, volcanic eruptions or other natural disasters, collision with or crashing of a flying object or parts thereof including its cargo, • general average sacrifice, • jettison, washing overboard or otherwise being lost overboard as a result of heavy weather, • discharging, interim storage, loading of cargo at a port or airport of distress entered as a result of an insured event or following an emergency landing, • total loss of entire packages during loading onto or unloading from a means of transport, or during transhipment to or from a means of transport • […] are insured. 2 Limits of indemnity per loss occurrence/insurance year 2.1 The insurer indemnifies a maximum amount of EUR .............. per any one loss occurrence. 2.2 The Insurer’s aggregate limit for all loss occurrences of a given insurance year is EUR ..................... 3 The write-back pursuant to paragraph 1 may be terminated at any time in text form. The termination is effective … days after receipt. 4 The write-back pursuant to paragraph 1 does not grant any additional insurance cover beyond the other provisions of the insurance contract.

The German Pandemic Exclusion Clause and its write-back clauses  371

WRITE-BACK CLAUSE ‘Dangerous Communicable Disease for the Use in Liability Insurance for Carriers, Freight Forwarder and Warehouse Operators’ Standard policy conditions of the GDV Version July 2021 1 In deviation from paragraph 1 of the ‘Clause for the Exclusion of Loss/Damage due to a Dangerous Communicable Disease for the Use in Marine Insurance’ and only within the scope of the provisions of the insurance contract, the liability under a service contract (carriage, forwarding, or warehousing contracts) of the Insured as the contractor in a service contract is included in the cover either as per 1.1 Option 1 (applies unless the insurance contract states that subparagraph 1.2 applies) with the following insurance sums per claim that deviate from the limits of indemnity amounts in the insurance contract: • for contracts of carriage: in the case of loss/damage to cargo EUR ...................... in the case of pure financial loss EUR ........................... • for forwarding contracts: in the case of loss/damage to cargo and consequential losses EUR ........................... in the case of pure financial loss EUR ........................... • for warehousing contracts: in the case of loss/damage to cargo and consequential losses EUR ........................... in the case of pure financial loss EUR .......................... • for claims based on tort law – irrespective of the type of service contract or the nature of the loss/ damage – EUR ......................... or as per 1.2 Option 2 (only applies if expressly agreed in the insurance contract) exclusively for: • loss of or damage to cargo during transport or handling, • delays, non-delivery, or perishing caused by faulty scheduling • […]. 2 Limits of indemnity per loss occurrence/insurance year 2.1 The insurer indemnifies a maximum amount of EUR .............. per any one loss occurrence. 2.2 The Insurer’s aggregate limit for all loss occurrences under the insured service contracts of a given insurance year is EUR ..................... 3 The write-back pursuant to paragraph 1 may be terminated at any time in text form. The termination is effective … days after receipt.

Index

abandonment 19–21, 142–5, 155–9, 162 Abraham, Kenneth 340 absolute total loss 133, 135 actual total loss (ATL) definition of 20, 131–3 deprivation of possession 137–8 destruction/damage 136–7 no hope of recovery 133–6 ad hoc forwarding agreements 179–80 agency theory 69 American certificates 262 American Law Institute (ALI) 238–9 Amis, Kingsley 181 annual efficiency ratio (AER) 337 anti-suit injunction 185–7 anti-suit relief 192–3 arbitration clause 183–203 artificial general average 167 artificial general intelligence (AGI) 292 artificial intelligence (AI) 291–3 Ashton, Kevin 289 assured’s election 154–5 Athens Convention 312 ‘attributable to’ concept 90 BARECON 89 clauses 12 and 13 70–71 Big Data 291–3 Bigham clause 177–9 blockchain 285–7 British Insurance Law Association (BILA) 264 Brussels Regulation Recast (2012) 187–9, 213, 216 Budapest Convention 313 burden of proof 360 business contingency cover (BCC) Clause 123, 126 cargo insurance 269–71, 366 cargo owners 103–4 causa proxima doctrine 359–60 causa proxima rule 360–61 causation 359–60 CEND Clause 123, 126 certificate assignment document 264–6 Chalmers, Mackenzie Sir 56 Cheshire Cat issue see utmost good faith principle choice-of-law principles 223–43

admiralty and maritime jurisdiction 224–5 in context 224–6 horizontal 223, 233–5 vertical 223, 229–33 Wilburn Boat decision 226–9, 235–42 comprehensive federal statute 237–8 individual level solutions 239–42 law of marine insurance restatement 238–9 single-sentence federal statute overruling 237 Supreme Court reconsideration 235–6 Civil Jurisdiction and Judgments Act (1982) 189, 200 CL 380 clause 323 climate alignment 334, 336, 349–51 climate change 333, 335, 341 climate enablers 341, 350 co-insurance hull policies case studies 63–7 demise charter 67–77 BARECON 89 clauses 12 and 13 70–71 as insurance contract 70 legal relationship with owner 67 safe port guarantees impact 71–7 shipowner as insurance contract 68–70 Gard Marine case study 82–3 Haberdashers case study 83 output analysis 84–6 coalition against insurance fraud 275, 298 Comité Maritime International (CMI) 221 common benefit allowances 166 constructive total loss (CTL) 138–59 abandonment 142–5 assured’s election 154–5 damage 149–53 cost of repair 149–53 relevant value 153 definition of 20 deprivation of possession 145–9 cost of recovery 148–9 unlikelihood of recovery 145–8 Japanese Commercial Code provisions 19–21 marine cargo insurance 111 Marine Insurance Act, section 60 of 138–40 notice of abandonment 155–9 prudent assured 141–2

372

Index  373 Consumer Insurance Act 283 Consumer Rights Directive 249 contra proferentem rule 75, 80–82 contractual enforcement 350 cost of recovery 148–9 cost of repair 149–53 credit risks 110 crew conduct, write-back clause 361–2 cover provided 362 limitations 362 scope of application 362 special agreement on application 361–2 cyber insurance 306–32 attacks 306–7 risk see cyber risk standardisation of policies 329–30 cyber risk 307–21 definition of 308 forms of 308–9 future research 328–31 incidents 307 normative force 318 post-contractual regulation 318–21 pre-contractual regulation 318–21 private governance 310–12 public-made instruments 312–17 as regulatory tool 317–18, 322–8 abuse of power 325 exclusion clauses 321–4 lack of information 322 legal instruments 326–8 vulnerabilities 324–5 cyber threat 308 dangerous communicable disease 353–4, 356–8, 368–71 cause of the loss 358 German Parliament classification 357–8 pandemic exclusion clause 368 transmissible diseases vs. 356 WHO classification 357 write-back clause 368–71 demise charter 67–77 BARECON 89 clauses 12 and 13 70–71 as insurance contract 70 legal relationship with owner 67 safe port guarantees impact 71–7 breach of insurance warranty 75–6 charterer on vessel 76–7 intent of charterer as co-insured 74–5 legal interests of owner 76–7 prevailing characteristics 72 wilful misconduct 74–5 shipowner as insurance contract 68–70 agency theory 69

standing offer theory 68–9 waiver of subrogation clause 69–70 deprivation of possession actual total loss 137–8 constructive total loss 145–9 cost of recovery 148–9 unlikelihood of recovery 145–8 destruction/damage actual total loss 136–7 constructive total loss 149–53 cost of repair 149–53 relevant value 153 discretionary cover 36 distributed blockchains 285 doctrine of abandonment 162 doctrine of frustration law of contracts of carriage 164–6 rule G(3) 177 double-counting, rule G(3) 176–7 duty of disclosure 11–12, 24 Electronic Trade Documents Act, UK (ETDA) 3, 261, 271–3 emerging technologies 284–93 artificial intelligence 291–3 Big Data 291–3 blockchain 285–7 fraud vs. 293–304 law to technology 302–4 non-existent cargo 300–302 overvaluation 298–300 scuttling 294–8 Internet of Things 289–91 overview of 284–5 smart contract 287–9 energy efficiency operational indicator (EEOI) 337 environmental, social and governance (ESG) risks 333 European Convention on Human Rights 195 excluded loss 88 exclusion clauses 321–4 Financial Conduct Authority 28 Financial Services Act (2012) 27 Financial Services and Markets Act (2000) 27, 29 foreign judicial proceedings 185–93 contractual claims (shipowner) 185–9 anti-suit injunction 185–7 EU law intervention 187–9 quasi-contractual claims (third party) 189–93 anti-suit relief 192–3 direct claims 189–91 fortuity 91–2 forwarding expenses 164

374  Research handbook on marine insurance law Fraudulent Bill of Lading (FBOL) Clause 115–18 fraudulent claim rule 52–8 Fraudulent Documents clause 118–24, 127 general average (GA) artificial 167 forwarding expenses 164 founding principle 163 General Data Protection Regulation (GDPR) 314–15 German Infection Protection Act 353 Germany’s Commercial Code 7, 9 ghost broking 277 governance conception 351 Hague Convention 246 Hague–Visby Rules 312 horizontal choice-of-law principles 223, 233–5 hull and machinery (H&M) insurance 111, 164–5, 208, 261, 333–5, 340–52, 363 hypothetical alternative expense (rule F) 173–4 included loss 88 Incoterms® 2010 Rules 245–59 analysis 257–9 insurance requirements 255–7 International Chamber of Commerce 249–54 passing of risk 245–9 legal concept 245–6 Vienna Convention 246–9 innocent co-assured 96–8 insolvency 110 Institute Cargo Clause (ICC) 8, 110, 112, 114–16, 118, 120, 123, 140, 148, 245, 255–9 Institute Time Clause (ITC) 8, 148–9, 153 insurable interest 13, 266–7 Insurance Act (2015) 1–2, 24, 40, 42, 44–5, 75, 90, 276, 307, 326–7, 331, 345–6 Insurance Business Act (IBA) 7–8 insurance claim and payment assessment 25 Insurance Contract Act (2008) 7, 9 insurance documents 260–73 cargo insurance 269–71 certificate assignment document 264–6 electronic trade document 271–3 insurable interest 266–7 international trade 267–9 legal reforms 271–3 marine 261–2 new technologies 269–71 policy assignment document 263–4 transfer of title to goods 266–7 Insurance Fraud Register (IFR) 276 insurance period 14–15 insurances of ships (write-back clause) 361–5

crew conduct 361–2 cover provided 362 limitations 362 scope of application 362 special agreement on application 361–2 loss of hire 362–4 cover provided 363–4 limitations 364–5 no requirement of special agreement 362 scope of application 363–4 party entitled termination 365 proximate cause doctrine 361 right of termination 365 insurtech 3, 284, 324 International Chamber of Commerce (ICC) 249–54 International Convention on Civil Liability for Oil Pollution Damage (1992) 204 International Health Regulations (IHR) 2005 353 International Maritime Bureau 276 International Maritime Organization (IMO) 222, 334 Internet of Things (IoT) 289–91 Japanese Commercial Code provisions 11–21 abandonment 19–21 constructive total loss 19–21 duty of disclosure 11–12 exclusions 16–17 insurable interest 13 insurance period 14–15 principles 16 provisional insurance 13–14 risk alteration 17–19 warranty alteration 17–19 Japanese insurance law revisions 9–11 Japanese marine insurance law 7–9 Kingsley Amis’ Jake’s Thing 181 large-risk insurance see non-mass-risk insurance law of armed conflict (LOAC) 315 law of contracts of carriage 164–6 law of general average non-separation agreements 175–6 port of refuge expenses 166–8 substituted expenses under rule F 168–70 ledger blockchains 285 legal principles duty of disclosure 24 exclusions 24 insurance claim and payment assessment 25 interpretation principle 23–4 policy provisions 23 risk alteration 24

Index  375 liability insurance 366–7 liability risk 333–4 loss of hire, write-back clause 362–4 cover provided 363–4 limitations 364–5 no requirement of special agreement 362 scope of application 363–4 malicious acts of third parties 87 marine cargo insurance 108–29 constructive total loss 111 credit risks 110 financial risks 123–8 Fraudulent Bill of Lading Clause 115–18 Fraudulent Documents clause 118–24, 127 insolvency 110 issues/importance 108 physical loss of or damage 109, 123, 125–7 policy characteristics 128–9 property insurance 112–15 English marine cargo cases 112–14 physical damage to goods 114 physical loss approach 114–15 rejection risks cover 111 risks of physical loss or damage 111–12 traditional cargo insurance cover 109 Unites States cases 115–18 marine cargo underwriters 110 Marine Insurance Act 1906 (MIA 1906) 1, 13, 29, 42, 74–6, 79–80, 87, 111, 130–40, 142–3, 145–6, 148, 154–7, 159–62, 166, 237, 261, 277, 311, 326–7 marine insurance contract law paradigm 338–41 marine insurance documents 261–2 marine insurance fraud 275–305 burden of proving 280–82 consequences of 282–4 definition of 276–7 emerging technologies vs. 293–304 law to technology 302–4 non-existent cargo 300–302 overvaluation 298–300 scuttling 294–8 post-contractual fraud 278–80 pre-contractual fraud 277 marinetech 284 maritime autonomous surface ships (MASSs) 290–91, 293 Maritime Law Association of the United States (US MLA) 237–9 mass-risk insurance definition of 22–3 legal principles 23–5 mortgagees’ interest insurance (MII) attributable to 90

cargo owners 103–4 claims 102–3 double-edged sword 98–101 fortuity 91–2 included/excluded losses 88 innocent co-assured 96–8 negligence 91–2 overview of 87–8 proximate cause of loss 89 scuttling 94–6 subrogation 105 wilful meanings 92–4 witness statements 95–6 mutual blockchains 285 negligence 91–2 net-zero insurance alliance (NZIA) 335, 341 no hope of recovery 133–6 non-mass-risk insurance definition of 23 legal principles 23–5 non-separation agreements (NSAs) 175–6 notice of abandonment (NOA) 155–9 Omnibus Rule 36–7 organisational conception of governance 351 pandemic exclusion clause 353–60 burden of proof 360 causation 359–60 dangerous communicable disease 353–4, 356–8 cause of the loss 358 German Parliament classification 357–8 transmissible diseases vs. 356 WHO classification 357 precautionary measures, governments/third parties 354–5 priority and relationship 355–6 protective measures 358–9 public authorities 359 third parties 359 partial loss definition of 131 measure of indemnity 130 party entitled termination 365 passing of risk 245–9 legal concept 245–6 Vienna Convention 246–9 physical loss of or damage (PLOD) 109, 123, 125–7 physical risk 333 policy assignment document 263–4 policy provisions 23

376  Research handbook on marine insurance law Poseidon Principles for Marine Insurance (PPMI) 3, 333–8 accountability 336 climate alignment 334, 336, 349–51 contractual architecture 343–8 contract terms 344–6 default 347–9 enforcement 347–9 incorporation 343–4 legal obligations 344–6 enforcement 336 insurance contract law 349–51 objectives of 336–7 principles-based, private regulatory tool 334 quantitative methodology 338 transparency 337 post-contractual cyber risk regulation 318–21 post-contractual fraud 278–80 pre-contractual cyber risk regulation 318–21 pre-contractual fraud 277 principle of salvage 162 Principles of European Insurance Contract Law (PEICL) 22 private governance 310–12 property insurance 112–15 English marine cargo cases 112–14 physical damage to goods 114 physical loss approach 114–15 Protection and Indemnity (P&I) Clubs 2, 26–7, 33–40, 183–203 foreign judicial proceedings 185–93 contractual claims (shipowner) 185–9 quasi-contractual claims (third party) 189–93 Prestige case study 196–202 binding effect, CJEU judgment 201–2 Court of Appeal judgment 198–9 English proceedings 197–8 EU sequel 199–201 factual background 196–7 rules 183–4 sovereign immunity 194–6 protection and indemnity (P&I) insurance 26–40 contracts of insurance 29–34 definition of insurance 29 judicial and commercial approaches 30–31 right vs. discretion 31–3 cover 34–8 discretionary 36 exclusions 37–8 Omnibus Rule 36–7 provisional insurance 13–14 proximate cause doctrine 361 proximate cause of loss 89

prudent assured 141–2 Prudential Regulatory Authority 27–8 public-made cyber instruments 312–17 quasi-contractual claims 189–93 anti-suit relief 192–3 direct claims 189–91 reasonableness conundrum (rule F) 170–72 rejection risks cover 111 right of subrogation 77–82 contra proferentem rule 80–82 definition of 77 Gard Marine case study 82–3 Haberdashers case study 83 legal ground 77–8 legal nature 77–9 output analysis 84–6 safe port warranty impact 79–82 right of termination 365 risk alteration 17–19, 24 risk-based insurance policies mass-risk insurance 22–5 non-mass-risk insurance 23–5 rule F hypothetical alternative expense 173–4 operation of CAP 174–5 reasonableness conundrum 170–72 substituted expenses 168–70 rule G(3) double-counting 176–7 frustration 177 rule G(4), Bigham clause 177–9 salvage, principle of 162 Santarém, Pedro de 305 saving-to-suitors clause 225 scuttling emerging technologies vs. fraud 294–8 insurance claims 96 standard of proof 94–5 silo-specific approach 341–2 smart contract 287–9 smart contract code 288 smart legal contracts 288 sovereign immunity 194–6 standard covenant clause (SCC) 338, 343–4 standing offer theory 68–9 State Immunity Act (1978) 194, 197, 202, 210 subrogation 105 right of 77–82 substituted expenses (rule F) 168–70 successive total loss 160–61 sustainable marine insurance 341–3

Index  377 Third Parties (Rights against Insurers) Act (2010) 205 third-party rights of direct action (TRDAs) 3, 204–22 advantages 205 definition of 204–5 EU approach 213–16 evaluation 219–21 Japanese law 218–19 New York law 216–18 private international law issues 205 UK approach 207–13 total loss actual see actual total loss (ATL) constructive see constructive total loss (CTL) definition of 131–3 rights upon payment 161–2 successive 160–61 timing of 159–60 traditional cargo insurance cover 109 Transaction Premium Clause (TPC) 123–4 transfer of title to goods 266–7 transition risk 333 Uniform Law on the International Sale of Goods 246 Uniform Rules on Digital Trade Transactions (URDTT) 268–9 United Nations Convention on Contracts for the International Sale of Goods see Vienna Convention

unlikelihood of recovery 145–8 utmost good faith principle 42–62 breach of the duty remedies 60–61 controversies 44–5 issues 42–4 legal duties 51–2 post-formation of contract fraudulent claim rule 52–8 general 52–4 insured 54–8 insurer 58–60 section 17 of the MIA 45–51 vertical choice-of-law principles 223, 229–33 Vienna Convention 246–9 vulnerabilities 324–5 waiver of subrogation clause 69–70 warranty alteration 17–19 wilful meanings 92–4 witness statements 95–6 write-back clause 361–7 cargo insurance 366, 370 dangerous communicable disease 368–71 insurances of ships 361–5 liability insurance 366–7, 371 limits and termination 367 options 366–7 York–Antwerp Rules 163–4, 166–71, 175, 178–9